INTRODUCTION
In 2012, the peace negotiations between the Colombian government and the Revolutionary Armed Forces of Colombia (FARC) became public knowledge, as the administration of President Juan Manuel Santos began seeking to alter the Colombian constitution in order to pave the way for a peace agreement. The constitutional amendment, the Legal Framework for Peace, provided the executive with the authority to prioritize and even suspend cases and sentences against members of armed groups. In response, the International Criminal Court’s (ICC) Office of the Prosecutor (OTP) released an unprecedented report outlining the status of its involvement in Colombia and providing a preliminary analysis of existing domestic proceedings. The report responded directly to the Santos administration’s efforts to alter the constitution with a thinly veiled threat: “[the Prosecutor’s Office] would view with concern any measures that appear designed to shield or hinder the establishment of criminal responsibility of individuals for crimes within the jurisdiction of the Court” (Office of the Prosecutor 2012, 63). It became the first of many public statements by the Prosecutor’s Office during the peace process, all with the same implication: peace without accountability could lead to ICC intervention.
The Santos administration knew the country faced the prospect of ICC intervention—Colombia had been publicly under preliminary examination by the Prosecutor since 2006. As early as 2010, administration officials had begun preparing to address the preliminary examination in the event that peace negotiations were successful, choosing to comply with its international obligations by showing it was able and willing to hold those most responsible for atrocities accountable. Why choose compliance with international obligations when negotiating a domestic peace process?
While designed to be a world court with “jurisdiction over the most serious crimes of concern to the international community as a whole,” the ICC has become an increasingly controversial institution (Rome Statute of the International Criminal Court 1998). This controversy is due in large part to claims about bias in court decision-making (Bates and Zvogbo Reference Bates and Zvogbo2021; Boehme Reference Boehme2022; Krcmaric Reference Krcmaric2023), and concerns about its effects. Existing research on the effects of the ICC has focused on the court’s impacts on democracy and political competition, atrocity prevention, and conflict duration and termination. This work has found that the existence of the ICC can limit peaceful transitions to democracy (Nalepa and Powell Reference Nalepa and Powell2016), and that the court’s reliance on state cooperation may even entrench autocrats, who can use the ICC to target domestic regime opponents (Ba Reference Ba2020; Hillebrecht and Straus Reference Hillebrecht and Straus2017).
In conflict, the effects of the ICC have been even more controversial. Recent work has found that ICC involvement can deter atrocities, especially by rebel groups and those who believe they can bargain with the prosecutor to reduce the risk of punishment (Jo and Simmons Reference Jo and Simmons2016; Miller Reference Miller2023). But research on the ICC has also found that court involvement can prolong civil conflict by making the benefits of victory more attractive (Krcmaric Reference Krcmaric2020; Prorok Reference Prorok2017) and limiting the range of options available during peace negotiations (Gissel Reference Gissel2015; Prorok Reference Prorok2017). Nevertheless, states and their opponents negotiate peace agreements in the era of a world court: 832 agreements related to civil conflict have been negotiated since the ICC was established in 2002, and nearly 30% of these agreements were negotiated in states facing ICC involvement. Despite the prevalence of these agreements in the era of international justice, research has paid little attention to their content.
In this article, I argue that involvement by the ICC places a unique pressure on elites negotiating peace agreements in the wake of political violence. To avoid going to trial in the international arena, elites facing intervention have an incentive to make use of the complementarity principle, which gives states the first opportunity to ensure accountability before intervention occurs. This unique design feature of the court provides those facing the prospect of intervention the opportunity to trade international prosecution for domestic accountability processes. By agreeing to hold domestic trials, elites can signal to the international community that they can handle their own business. When intervention does occur, I argue it can reinforce rather than offset this dynamic, pushing elites to commit to domestic trials.
I evaluate my argument using data on the content of the peace agreements negotiated between 2002 and 2019, as well as original data on the varying levels of ICC involvement in each country over time. I find that ICC involvement, via the concrete threat of intervention and intervention itself, is associated with an increase in the likelihood that a peace agreement contains a commitment to implement domestic trials. I present two case studies to explore mechanisms in more detail, paying careful attention to the role of potential alternative explanations. The first case study focuses on the concrete threat of intervention in Colombia. The second uses the case of Sudan to highlight the role that intervention can play in raising the stakes of negotiations. I end by evaluating the genuineness of these commitments in Colombia and Sudan, briefly discussing the implementation of the agreements negotiated. I find that while ICC involvement created incentives to commit to holding domestic trials, elites in Colombia and Sudan were varied in their willingness to actually implement the commitments. However, despite efforts to see these commitments as simply “window dressing,” elites subsequently found themselves bound by them.
The article makes two contributions to political science research. First, it contributes to research on the role credible commitments can play in ensuring state compliance with international law. Prior research demonstrates that treaty commitments contribute to compliance with international law by tying the hands of states that might otherwise have an incentive to deviate from approved practice in the areas of democracy (Moravcsik Reference Moravcsik2000), trade (Mansfield, Milner, and Rosendorff Reference Mansfield, Milner and Rosendorff2002), and human rights and accountability (Simmons Reference Simmons2009; Simmons and Danner Reference Simmons and Danner2010). In the area of human rights, these commitments give domestic civil society groups (Simmons Reference Simmons2009) and state institutions (Hillebrecht Reference Hillebrecht2014) the necessary tools to promote compliance from the bottom up. This article shows that institutions like the ICC can increase credible commitments to justice from the top down. By incentivizing domestic elites to formally commit to implementing accountability at home, the ICC can help local advocates by broadening the set of tools they can use to promote compliance. Moreover, it shows one way that institutional design features like the complementarity principle can create incentives for states to make credible commitments to justice.
Second, this article contributes to the literature on advocacy and norm diffusion in domestic and international politics. Existing work highlights the role that norm diffusion plays in advancing human rights goals globally (Lutz and Sikkink Reference Lutz and Sikkink2001; Sikkink Reference Sikkink2011). It focuses in particular on the role of transnational advocacy networks, linking these groups to the spread of norms across space and time (Keck and Sikkink Reference Keck and Sikkink1998). This article shows that international institutions can directly contribute to this work by highlighting how the ICC can enforce norm compliance even at lower levels of involvement.
JUSTICE AND PEACE
Since at least the beginning of the third wave of democratization, states recovering from periods of repression or political violence have prosecuted those responsible for atrocities, including international crimes like war crimes, crimes against humanity, and genocide. These efforts fall within a broader set of policies called transitional justice (TJ), the “formal and informal procedures implemented
$ \dots $
for rendering justice to perpetrators and their collaborators, as well as to their victims” (Kaminski, Nalepa, and O’Neill Reference Kaminski, Nalepa and O’Neill2006, 295). In practice, TJ can constitute a variety of policies, including criminal trials against perpetrators, amnesties, truth-telling processes, vetting, reparations, and memorialization projects. TJ can be implemented at the international, state, or local level, depending on the nature of the abuses and the context of TJ’s development. While all forms of TJ can play an important role in addressing past atrocities and ensuring stable transitions to peace, criminal trials raise the stakes of implementing TJ by directly threatening those responsible for past harm.
An important body of literature points to the role that bottom-up demands for TJ can play in getting trials on the agenda. During political transitions, demands for trials are often driven by domestic human rights advocates and victims’ communities, with varying degrees of support and opposition from domestic elites (Lutz and Sikkink Reference Lutz and Sikkink2001). In peace processes, the presence of these activists can signal broader public support for trials, pushing elites to commit to implementing them. However, these organizations’ success is contingent upon whether or not there are barriers to effective advocacy (Bell and Kitagawa Reference Bell and Kitagawa2023). The literature highlights two interrelated mechanisms for promoting effective advocacy: first, human rights advocates receive support and technical assistance from transnational NGOs, who work to spread norms across borders (Keck and Sikkink Reference Keck and Sikkink1998). For trial advocates, the anti-impunity norm underpins much of this advocacy, linking domestic and international actors seeking to punish those responsible for atrocities (Sikkink Reference Sikkink2011).
The second mechanism centers on the use of international law as a catalyst for mobilization (Simmons Reference Simmons2009). This literature argues that law and treaties can lower barriers for human rights advocates by giving them the information and tools they need to pressure governments to comply (Dancy and Michel Reference Dancy and Michel2016; Hathaway Reference Hathaway2001). In peace processes, these advocates can use international law to push for accountability, and may be particularly successful when states have ratified relevant human rights treaties (Bell and Kitagawa Reference Bell and Kitagawa2023; Simmons and Danner Reference Simmons and Danner2010). While this work shows the role of international law in supporting bottom-up mobilization, in this article, I highlight the direct role that top-down demands from a key international institution, the ICC, can play in promoting domestic TJ.
Another body of research focuses on international courts and tribunals, showing how they can advance human rights practice in general. For example, research on regional human rights courts explains how courts can issue important rulings that influence state policy (Hillebrecht Reference Hillebrecht2014; Reference Hillebrecht2021). More directly, since at least the Nuremberg trials against the Nazis, international criminal tribunals have been charged with holding perpetrators accountable for international crimes (Bass Reference Bass2000). These courts are given the authority to investigate and prosecute individuals, with varying formal and informal jurisdictional features that interact with the interests of domestic elites in ways that can support or run counter to the interests of justice (Loken, Lake, and Cronin-Furman Reference Loken, Lake and Cronin-Furman2018; Loyle and Davenport Reference Loyle and Davenport2016) and the broader anti-impunity norm (Cronin-Furman Reference Cronin-Furman2022; Labuda Reference Labuda2023; Subotić Reference Subotić2009).
The ICC has been a particularly controversial institution in the literature on international criminal tribunals. At the core of criticisms about the impacts of the ICC is the acknowledgment that the court relies on state cooperation to be able to investigate and successfully prosecute perpetrators. This reliance on state cooperation opens the door to political instrumentalization, often by governments that are themselves directly responsible for atrocities. Some argue that states leverage cooperation to dictate the timing and sequencing of ICC investigations to minimize risk for state agents (Hillebrecht and Straus Reference Hillebrecht and Straus2017). They may also cooperate in some issue areas while actively undermining investigations (Boehme Reference Boehme2022), especially by posing technical challenges (Hillebrecht Reference Hillebrecht2021; Hillebrecht and Straus Reference Hillebrecht and Straus2017). This body of literature most often accuses states of instrumentalizing the ICC by using the court to target domestic regime opponents (Ba Reference Ba2020; Hillebrecht and Straus Reference Hillebrecht and Straus2017). It suggests that international justice is primarily a threat to opposition groups: states can limit the court’s impact by simply not cooperating while using investigations to target enemies. This article contributes by showing that state actors do fear the prospect of court intervention and that they use a variety of strategies—including promoting domestic accountability—to assuage concerns about international justice.
A growing body of research also highlights the ICC’s controversial role in civil conflict. This literature has largely coalesced around the finding that ICC involvement can limit peaceful transitions from autocracy to democracy (Nalepa and Powell Reference Nalepa and Powell2016) and prolong conflict (Krcmaric Reference Krcmaric2020; Prorok Reference Prorok2017), though it may simultaneously reduce the level of atrocities (Jo and Simmons Reference Jo and Simmons2016; Miller Reference Miller2023).Footnote 1 Many of these arguments rest on the idea that the threat of international justice can alter the underlying cost–benefit calculations of both state and rebel leaders. First, by intervening during conflict, the ICC makes it logistically difficult to negotiate peace agreements because leaders avoid capture and prosecution by remaining in strongholds (Prorok Reference Prorok2017). By centering authority with international rather than local actors, ICC intervention may also limit conflicting parties’ ability to overcome sticking points in peace negotiations, reducing the likelihood of an agreement even when negotiations occur (Gissel Reference Gissel2015). Finally, by taking amnesty off the table, the ICC makes the benefits of victory more attractive: the court’s reliance on state cooperation means that gaining or maintaining control of the state may be the only mechanism to avoid accountability (Krcmaric Reference Krcmaric2020; Prorok Reference Prorok2017). In this body of research, international justice is also primarily a threat for opposition groups or those who otherwise lose a conflict.
But parties in conflict can and do seek to mitigate the risk of international punishment in other ways. Miller (Reference Miller2023) finds that the threat of international punishment can lead to a temporary reduction in the commission of atrocities. This is because the parties in conflict seek assurances: they believe they can negotiate with international prosecutors to avoid accountability for past crimes at the ICC if they agree not to commit future crimes. But the ICC Prosecutor does not give assurances and has never indicated a willingness to negotiate over the removal of arrest warrants. Instead, those facing the prospect of trial in the Hague are left to thwart intervention by other means. In this article, I identify one mechanism parties use to thwart intervention: agreeing to implement domestic trials.
Finally, numerous case studies show how the threat of international justice affects the process of peace negotiations. Beyond determining the likelihood of negotiations or their success or failure, ICC intervention can shape who participates in negotiations and the agenda (Hayner Reference Hayner2018; Kersten Reference Kersten2016). This can happen directly: Kersten (Reference Kersten2016) argues that when investigations and warrants from institutions like the ICC are central concerns of the parties negotiating, they can become agenda items during negotiations. But this may also happen indirectly: negotiating parties can put the issue of alternate accountability measures on the agenda precisely to address concerns about international investigations (Hayner Reference Hayner2018). By taking a broader view of the ICC’s influence on the various stages of peace processes, these works make an important contribution to the literature by moving beyond the question of peace versus justice. However, they understate the extent to which the ICC can influence the agenda during negotiations.
In sum, the literature remains divided on the extent to which institutions like the ICC matter for peace processes, especially for state agents. I contribute to this debate by showing that in some contexts, both state agents and opposition groups are concerned about ICC intervention for a variety of reasons. While negotiating parties use other tools to avoid accountability, including leveraging the need for state cooperation to influence investigations, I show that they also try to thwart intervention by committing to implement accountability domestically. I also contribute by empirically evaluating the extent to which concerns about intervention directly influence the agenda and ultimately the agreements made during peace negotiations.
ARGUMENT
Like the ad hoc tribunals before it, the ICC is seen as an enforcer of the anti-impunity norm, stepping in to hold perpetrators accountable when states do not. But there are limits to what the court can do, some of which are driven by the jurisdictional and admissibility constraints enshrined in the Rome Statute. I argue that elites know this and try to avoid trials in the Hague by using the principle of complementarity, a key constraint on the ICC, to their advantage when negotiating peace agreements.
The ICC as Enforcer
Powerful actors with international reach and justice-oriented goals can pose a threat to those seeking to avoid accountability for atrocities in the wake of conflict. Indeed, intervention (or the concrete threat of it) by actors like the ICC’s OTP is the perfect context to consider this argument given the court’s features. For example, unlike many international courts, the ICC is charged with holding individuals rather than states accountable for atrocities committed.
There are three mechanisms by which situations are referred to the Court: (1) a referral by a state party to the Rome Statute, (2) a referral by the UN Security Council (UNSC), and (3) a referral by the OTP itself using its propio motu authority (Rome Statute of the International Criminal Court 1998). The OTP is thus authorized to initiate proceedings regarding situations in the states party to the Rome Statute without relying exclusively on referrals from the states themselves. Referrals by the UNSC can explicitly address situations in states not party to the Rome Statute, though the increasingly internationalized nature of ICC investigations has raised questions about the scope of the Court’s jurisdiction, even for the OTP itself.Footnote 2
Additionally, the Rome Statute “overrides any immunities that states may grant to presidential, parliamentary, or legislative officials in their domestic systems” (Simmons and Danner Reference Simmons and Danner2010, 230). Only the UNSC has the authority to suspend an ICC investigation, and only when investigations are believed to pose a threat to international peace and security.
While its independent prosecutor and expansive authority over international crimes make it a potentially effective stick against elites seeking to avoid accountability for atrocities, the jurisdictional and admissibility considerations outlined in the Rome Statute nonetheless constrain the ICC. The most important constraint is the complementarity principle, which states that countries facing ICC intervention are given the first opportunity to investigate crimes and hold those most responsible for offenses accountable through domestic proceedings. The ICC can find a case admissible only when the state in question is either “unwilling or unable genuinely to carry out the investigation or prosecution” (Rome Statute of the International Criminal Court 1998, Art 17(1)(a)). The complementarity principle is thus an escape clause so that countries are not subject to the stick of intervention.Footnote 3
The ICC’s role in addressing complementarity concerns has been controversial due to claims surrounding both the substance of complementarity decisions and the (un)intended effects of the complementarity system more generally (Clark Reference Clark2018; Labuda Reference Labuda2023). Particularly controversial has been the concept of positive complementarity, the idea that “[r]ather than competing with national systems for jurisdiction, [the OTP] will encourage national proceedings whenever possible.”Footnote 4 The court itself, especially the OTP, often invokes complementarity when attempting to persuade the public to support its actions (Zvobgo and Chaudoin Reference Zvobgo and Chaudoin2025). And political elites are aware of and take into consideration questions of complementarity when assessing the prospect of being subject to a court investigation. Both states and individual defendants make complementarity challenges—arguments that cases are not admissible before the court because of the existence of or potential for local prosecutions—in efforts to avoid ICC intervention.
The Threat of International Justice
I argue that involvement by an institution like the ICC can affect the likelihood that those negotiating peace agreements will commit to some form of domestic justice. This is because when negotiating, those most responsible for core international crimes want to avoid being tried internationally.
Those sitting at the negotiating table are usually the direct participants in armed conflict and their representatives. State representatives often include government officials and potentially representatives from the military. While the interests of these two groups may not always converge on all matters during negotiations, it is reasonable to assume they do about issues related to accountability for atrocities. This is because government officials may themselves be current or former members of the military, especially in authoritarian regimes. In states with strong civilian oversight of the military, government representatives may have even more reason to be concerned about the attribution of blame for atrocities and thus seek to avoid accountability. On the other side of negotiations sit the leaders and members of non-state armed groups and their representatives, who also prefer to avoid accountability.
The groups negotiating may have these preferences for many reasons, both individual and political. State officials may be concerned about the geopolitical ramifications of international accountability. That is, concerns about sovereignty and authority may lead state officials to prefer noninterference from justice-oriented international institutions. U.S. presidents from Clinton onward have, for example, expressed concerns about the implications of an independent ICC prosecutor and a lack of complete UNSC oversight as justifications for U.S. opposition to the Court (Bosco Reference Bosco2013). Similar claims were taken up by African Union (AU) member states after years of ICC involvement in the region (Mills Reference Mills2012).
Individuals may also personally prefer nonintervention for several reasons. First, international trials are visible public trials that put the worst crimes on display for the world to see. Being labeled a war criminal can ostracize elites within the international arena, as Serbian leaders facing arrest warrants from the International Criminal Tribunal for the former Yugoslavia discovered in the early 2000s (Bass Reference Bass2000). The recent issuing of ICC arrest warrants for Russia’s Vladimir Putin and Israel’s Benjamin Netanyahu has raised questions about where these leaders are safe to travel and which states might ultimately comply with the warrants.Footnote 5 ICC trials are generally open to the public and are actively publicized. The court streams hearings online, and has an active social media presence across all of the major platforms.Footnote 6 Even in the early stages of ICC investigations, court involvement attracts media attention, which can put a spotlight on those accused of atrocities and shift public opinion against them and their actions (Chaudoin Reference Chaudoin2023).
Individuals may prefer nonintervention because of the potential physical separation from loved ones caused by being held for trial abroad. For example, notorious Ugandan warlord Joseph Kony was allegedly deeply concerned about being held in the Hague for trial, in part because it would remove him from his community (Clark Reference Clark2018; Hayner Reference Hayner2018).
Perhaps the most important personal reason to seek to avoid intervention is that when trials are held in the international arena, individuals may be unable to leverage their institutionalized authority to engage in damage control and affect a finding of guilt or innocence. Those who had been supportive of ICC investigations in the wake of the 2007 post-election violence in Kenya later sought to deter intervention precisely because they discovered they were targets of the investigation (Chaudoin Reference Chaudoin2016). Individuals can intimidate witnesses, bribe judges, and otherwise remove access to evidence to ensure a favorable outcome at trial. Some of these options are available in the international arena—witness intimidation contributed to the collapse of the Kenya cases at the ICC—but there are a wider range of options to affect the outcome of a trial in the domestic arena than internationally.Footnote 7
When international actors like the ICC are able and willing to hold culpable individuals accountable directly, they can alter the political environment by undercutting any gains or losses the negotiating parties face as a result of the agreement. This provides an incentive to those negotiating about justice to attempt to thwart this kind of international intervention. They do so by committing to hold perpetrators accountable at home.
Hypothesis 1. ICC involvement increases the probability that a peace agreement will contain commitments to hold domestic trials.
Mechanisms
Precisely because of personal and strategic considerations, ICC involvement, which poses a concrete rather than abstract threat of intervention, is likely to increase negotiators’ willingness to agree to implement domestic accountability during peace negotiations. Prior to intervention, ICC involvement poses a threat that is concrete but still avoidable—negotiators can stave off intervention by proving they can handle their own business. Once intervention occurs by, for example, the issuance of an international arrest warrant, those most immediately impacted may seek other recourse, inhibiting investigations by promoting non-cooperation (Ba Reference Ba2020; Hillebrecht Reference Hillebrecht2021; Mills Reference Mills2012).
However, intervention itself can also act as a concrete threat. International arrest warrants for atrocities are issued against individuals, not groups, political parties, or state institutions. When intervention occurs in this manner, it serves as a realization of the threat for some while signaling the credibility of the threat for those not immediately affected. That is, even when intervention occurs, it can increase the likelihood that groups negotiating peace agreements will commit to implementing some form of domestic TJ, especially prosecutions.
In sum, I argue that ICC involvement can increase the likelihood that the parties negotiating will agree to implement domestic accountability mechanisms. It does this via two pathways: first, by getting involved but not intervening, the ICC creates incentives to commit to domestic accountability because establishing domestic trials can create the conditions for a successful complementarity challenge. Second, those who see the ICC intervene know they may also face accountability and are more likely to seek out the promises of complementarity. By actually intervening and issuing arrest warrants, the ICC further incentivizes commitments by showing its resolve.
While these commitments may not be genuine—many of the benefits of avoiding international accountability are seen as benefits precisely because they allow perpetrators to avoid true accountability—perpetrators’ attempts to subvert accountability norms nonetheless result in commitments on paper. These window-dressing commitments can have potential downstream effects, even if they are designed to sidestep the possibility of facing trial.
DATA AND RESEARCH DESIGN
To evaluate this argument, I first turn to quantitative data. I test my argument using data on the peace agreements negotiated between 2002, when the Rome Statute establishing the ICC went into effect, and 2019. These data are drawn from the PA-X Peace Agreement Database, which collects and codes data on the content of all peace agreements negotiated since 1990 (Bell and Badanjak Reference Bell and Badanjak2019; PA-X 2020). PA-X collects data on all agreements related to any conflict, but I subset the agreements in this analysis to those related to civil conflict. Each observation in the dataset is a civil conflict-related peace agreement. Since 2002, 217 of the 832 peace agreements related to civil conflict have included TJ provisions of some kind. While the creation of the ICC may be perceived as a threat to all countries, in this article, I limit my analysis to the post-2002 time frame because I am focused on the concrete threat of prosecution posed by ICC involvement.Footnote 8 However, I also take seriously the possibility that the effect of the ICC varies with its level of involvement. I evaluate this by measuring the level of involvement in the quantitative analysis below.
Dependent Variable
Because I am interested in how the threat of intervention by the ICC affects commitments to holding domestic trials against human rights violators during peace negotiations, I use as a dependent variable PA-X’s measure of TJ related to national courts. This measure is a binary variable, which takes the value of 1 if the peace agreement includes a provision dealing with substantive domestic judicial accountability (a trial), and 0 otherwise.Footnote 9 I supplement this measure by including any agreement that contains an amnesty provision that expressly prohibits the ability to amnesty relevant crimes like genocide, war crimes, and crimes against humanity. In 39 peace agreements between 2002 and 2019, the parties to the conflict made commitments to hold trials or limit amnesties at the domestic level. This is a rare outcome: trial commitments occur in just under 5% of peace agreements.Footnote 10 However, the rarity of trial commitments is unsurprising because, as noted above, codifying a commitment to prosecute is a high bar for negotiators to pass. Because elites do not want to face accountability, it is unsurprising that they avoid promising to do so when possible.
Explanatory Variables
In order to measure the key explanatory variables, I construct a measure of ICC involvement in the country participating in the peace negotiations. The stages of ICC involvement include: (1) the opening of a preliminary examination, (2) the initiation of a formal investigation, (3) the issuing of arrest warrants, (4) holding trials, and (5) securing convictions. Preliminary examinations are a formal process that precede active investigations in a situation; they primarily involve determining if crimes subject to the ICC’s jurisdiction have occurred and if cases involving the prosecution of those crimes are admissible before the court. A situation may remain in the preliminary examination stage for years, it may be closed, or it may result in a formal investigation. A formal investigation involves collecting evidence and building cases against individuals for committing crimes covered by the court’s jurisdiction, often but not exclusively those outlined in the preliminary examination analysis. While the OTP is primarily responsible for conducting preliminary examination activities, it must request and receive authorization from the court’s Pre-Trial Chamber to begin a formal investigation. Once a formal investigation begins, the Prosecutor can request that the court issue arrest warrants for suspects who, if arrested and the charges are confirmed, subsequently appear at trial. After trials are held and convictions or acquittals are secured, the Prosecutor or defense may file any necessary appeals that, once exhausted, result in a final judgment.
I collected information about the court’s involvement around the world directly from the ICC website, as well as from annual reports issued by the OTP on the status of its preliminary examination activities. Figure 1 provides an overview of the data, showing the extent of ICC involvement over time in every country with a relevant peace agreement.Footnote 11 Each shade represents a stage of involvement, moving from preliminary examinations (the lowest level of involvement) to the securing of convictions (the highest level of involvement). The darker shades of gray indicate higher levels of ICC involvement.
Extent of ICC Involvement (2002–2019)
Figure 1. Long description
A horizontal bar chart with the Y-axis listing 42 countries and the X-axis representing years from 2002 to 2019. A legend on the right defines six stages of involvement using a grayscale gradient: None (lightest gray), Preliminary Examination, Formal Investigation, Arrest Warrant, Trial, and Conviction (black).
Key data trends include:
* Central African Republic and Democratic Republic of Congo show the most advanced progression, reaching the Conviction stage (black) starting around 2016.
* Mali also reaches the Conviction stage by late 2016.
* Cote d'Ivoire, Sudan, and Uganda reach the Trial stage (dark gray).
* Libya and Kenya (listed as Kenya in context, though labels vary) show progression to the Arrest Warrant stage.
* Many countries, such as Afghanistan, Burundi, Georgia, Guinea, Iraq, Nigeria, Palestine, and Ukraine, remain in the Preliminary Examination or Formal Investigation stages for extended periods.
* A significant number of countries, including Algeria, Angola, Chad, Ethiopia, India, Indonesia, and Morocco, show no involvement (None) throughout the entire period.
Using data presented in Figure 1, I construct a measure which breaks down the stages of involvement into threat or intervention, with the residual category being no involvement. Footnote 12 By threat, I mean concrete ICC involvement prior to intervention. I therefore include only the commencement of a preliminary examination and the initiation of a formal investigation. Everything prior to the initiation of a preliminary examination is not considered to be involvement.Footnote 13 Everything after an arrest warrant is issued is considered intervention.
I code the level of involvement for all agreements in a country because while prosecutions target individuals, investigations focus on entire situations, making the threat of prosecution diffuse across an entire place. This coding is also justified in the case of peace agreements because many agreements are not dyadic—they can involve multiple actors from different parties to various conflicts. Even if the OTP were to focus on particular individuals at the initiation of its involvement, the participation of more than two groups in peace negotiations makes it more likely that at least some of those groups contain individuals who are facing the threat of international prosecution.
Controls
I include several control variables to account for arguments in the existing literature and to address the potential for omitted variable bias. Using data from Smith and Wiest (Reference Smith and Wiest2005) and Murdie and Davis (Reference Murdie and Davis2012), I control for human rights organization member/volunteer presence to account for arguments that human rights organizations and networks help explain justice outcomes (Bell and Kitagawa Reference Bell and Kitagawa2023; Simmons Reference Simmons2009). I also control for whether or not the state involved in negotiating the peace agreement has ratified the Rome Statute. I do this for two reasons: first, because ratifying the Rome Statute means accepting the Court’s jurisdiction, ratification may increase the likelihood of ICC involvement. Additionally, because ratification has been associated with increased human rights protections generally (Simmons and Danner Reference Simmons and Danner2010), it may also be associated with an increase in commitments to punish those who have committed atrocities.
Because the OTP takes into consideration the gravity of the atrocities committed when deciding to get involved, I also explicitly measure the level of atrocities committed during the recent years of the conflict. I use data on civilian killings from the UCDP Georeferenced Event Dataset (Sundberg and Melander Reference Sundberg and Melander2013). I calculate the cumulative number of civilian deaths (logged) caused by the state and/or armed groups to the conflict after 2002, and I lag the variable by one year throughout the analysis.Footnote 14 I also include a measure of the intensity of the conflict in question by controlling for overall combatant casualties (logged).
I control for the involvement of other parts of the international community in the negotiation process because international attention may draw the attention of the ICC, and it may make certain kinds of peace agreement provisions more likely. The PA-X data include two different measures to account for international involvement in the negotiation process. The first is a binary variable that accounts for whether or not the UN is a signatory to the agreement, and the second is a binary variable that accounts for whether or not any other international organization is a signatory. IO Signature is the sum of these two measures. I also include a measure of development assistance as a percent of GNI to account for the role that foreign aid can play in shaping peace processes (Campbell and Spilker Reference Campbell and Spilker2022). The measure comes from the World Bank’s World Development Indicators Dataset (Arel-Bundock Reference Arel-Bundock2025).
Finally, I include a series of controls to capture other domestic and regional factors. To account for the fact that commitments to hold domestic trials may result from demands from below, I include a measure for regime type using the participatory democracy index from the V-Dem expert survey.Footnote 15 I also use V-Dem’s access to justice variable to measure domestic judicial capacity because justice-sector capacity may affect whether the negotiating parties are willing to commit to holding domestic trials. While commitments to trials in peace agreements do not necessarily lead to trials actually occurring, because trials are financially costly for states to implement, I also use data from the World Development Indicators Dataset to control for GDP per capita (logged) (Arel-Bundock Reference Arel-Bundock2025). I include a dummy variable for whether or not the conflict occurred in Africa, as well as a linear time trend. The Africa dummy addresses claims that ICC intervention is more likely in Africa.
Summary statistics for all of these measures are shown in Table A1 in the Supplementary Material and replication materials are available in Bates (Reference Bates2025). In the next section, I estimate a series of rare events logit models that regress the dependent variable on my key explanatory variables and relevant controls.Footnote 16 A rare events logit model is appropriate because of the rareness of the dependent variable. The main models use robust standard errors clustered at the country level to account for nonindependence across observations within a country.
RESULTS
The results in Table 1 provide suggestive evidence in support of my argument. Model 1 includes only the primary explanatory variables, model 2 includes variables accounting for alternative explanations and conflict dynamics, and model 3 includes the full battery of controls. Across all models, both threat and intervention make the odds of a peace agreement containing a domestic trial commitment more likely than when there is no ICC involvement. More specifically, the threat of being held accountable by the ICC, as measured by the involvement of the OTP prior to intervention (the preliminary examination or formal investigation phase), makes the odds of a peace agreement containing a domestic trial commitment seven times greater than under no threat. Similarly, intervention itself (arrest warrants, trials, and convictions) makes the odds of a peace agreement containing a domestic trial commitment seven times greater than under no threat.
ICC Involvement and Domestic Trial Commitments in Peace Agreements
Table 1. Long description
The table presents coefficients, confidence intervals in brackets, and p-values in parentheses for three models of domestic trial commitment.
Key variables and their results across Models 1, 2, and 3:
* Threat: 1.377, 2.157, and 1.921. All are significant at the p < 0.01 level.
* Intervention: 1.273, 2.362, and 1.910. All are significant at the p < 0.01 level.
* H R organizations: Not included in Model 1; 0.036 in Models 2 and 3, both significant at p < 0.01.
* Rome statute ratification: Not included in Model 1; -0.650 (p < 0.1) in Model 2 and 0.032 (not significant) in Model 3.
* Foreign aid (lag): Not included in Model 1; 0.020 (p < 0.05) in Model 2 and 0.025 (p < 0.05) in Model 3.
* Other variables including Combatant casualties (lag), Atrocities, I O signature, Participatory democracy, Access to justice, G D P per capita (lag), Africa, and Time show varying non-significant results.
Summary statistics for all three models:
* Number of observations: 719.
* Log likelihood: -133.115 for Model 1, -127.818 for Model 2, and -125.893 for Model 3.
* Standard errors: Clustered by country.
Note:
$ {}^{*}p< $
0.1;
$ {}^{**}p< $
0.05;
$ {}^{***}p< $
0.01.
The involvement of human rights organizations also appears to consistently increase the odds that an agreement will contain a domestic trial commitment, as does the presence of foreign aid. All other controls are insignificant or inconsistent.
I conduct several robustness checks in Section B of the Supplementary Material, including an analysis that accounts for the possibility that clustering the standard errors by country, as presented in Table 1, may be insufficient given the relatively small number (43) of countries in the data. Table B1 in the Supplementary Material presents the results of the models with bootstrapped standard errors. I also present results of analyses that explicitly model how situations are referred to the ICC, and where ICC involvement is broken down by each separate stage of involvement. Tables B2 and B3 in the Supplementary Material provide the results of these analyses, while Tables B4 and B5 in the Supplementary Material replicate the main models with lagged explanatory variables. The results are largely consistent: both the threat of intervention and intervention itself are associated with increased odds of agreements containing domestic trial commitments.
I also include several sensitivity analyses, exploring how sensitive my results are to dropping any particular agreement with a domestic trial commitment or dropping all agreements from any particular country. The results of these analyses, which use the full model specification (model 3 of Table 1), are presented in Figures B1 and B2 in the Supplementary Material. While some of the results from these sensitivity analyses are consistent with the results presented in this section, Figure B2(a) in the Supplementary Material in particular highlights an important limitation of this analysis: when agreements in Colombia are dropped from the dataset, the primary results are no longer significant for the threat of ICC intervention, though dropping Colombian agreements from the dataset does not affect the results for intervention. There are several potential explanations for this finding, which I explore more in Section B.5 of the Supplementary Material. First, it may simply be the case that the threat of intervention does not matter outside of Colombia. The consistent and robust finding that intervention itself matters would then suggest that the second mechanism I propose—the OTP showing resolve by acting on its threat—still leads those negotiating agreements to commit to holding domestic trials. The importance of Colombia to the findings presented above may instead be a result of its particularly long preliminary examination. As I discuss in the next section, the results of the sensitivity analysis provide further motivation for case selection.
CASE STUDIES
To evaluate how negotiators think about questions of accountability during peace processes, I present two case studies of Colombia and Sudan.Footnote 17 Each case study is a critical case, both because these particular peace negotiations have laid the foundation for post-conflict accountability in each country and because of the role each has played for the ICC in international politics. The negotiations and subsequent peace agreement between the Colombian government and the FARC produced a comprehensive agreement that is being used in Colombia as the basis for agreements with other armed groups. In Sudan, the negotiations and subsequent agreement between the Sudanese government and the Darfuri rebel groups became the foundation for subsequent negotiations and agreements, surviving even the regime upheaval that began in 2019. Internationally, these cases have also been critical for the ICC, albeit in different ways. The OTP’s contentious investigation in Sudan was seen as a critical turning point in the court’s relationship with the AU, giving rise to claims about bias and contributing to subsequent backlash (Boehme Reference Boehme2022; Hillebrecht Reference Hillebrecht2021). The OTP’s preliminary examination in Colombia was its longest running preliminary examination to date, and the prosecutor’s decision to close it in late 2021 was seen as an important signal to other states facing ICC intervention (Bates and Zvogbo Reference Bates and Zvogbo2021). The cases are thus influential for understanding both the ICC and the dynamics of contemporary peace processes.
The cases are also important analytically. As the results of the sensitivity analysis suggest, agreements in Colombia may be driving the results about the threat of intervention presented in Table 1: Colombia is likely an extreme case, both because of the long preliminary examination and the number of agreements with justice commitments (Seawright and Gerring Reference Seawright and Gerring2008). Sudan is influential because existing research on the case suggests it should be a hard case for my argument: the backlash against the ICC in the wake of the investigation could have provided the Sudanese government with the cover to ignore the threat posed while negotiating a peace agreement. However, that is not what occurred.
Below, I evaluate if and how the threat of intervention shaped the peace negotiations between the Colombian government and the FARC. Over the course of 2.5 years (2017–2020), I conducted semi-structured interviews with 27 Colombian elites who participated in or had expert knowledge of the peace negotiations.Footnote 18 My interlocutors included members connected to both negotiating teams, as well as civil society leaders, academics, and bureaucrats.
I then evaluate if ICC intervention can lead those previously opposed to implementing domestic accountability to put it on the agenda during peace negotiations. I draw on existing research on ICC intervention in Sudan and the peace negotiations between the Government of Sudan and the Darfuri rebel movements. I focus on the incentives and actions of the Sudanese government, which was a key stakeholder in both the negotiations and the ICC investigation. I then turn to the question of alternative explanations in both cases, evaluating the role bottom-up demands for accountability played. Finally, I briefly explore the genuineness of these agreements by outlining the implementation process in both countries.
ICC Involvement in Colombia
The OTP made clear its position on the Colombian peace process from the very beginning. The situation in Colombia had been under preliminary examination at the court since 2004, when the OTP—using death count as its metric—determined Colombia to be one of the most grave situations (Bosco Reference Bosco2013). By the time peace negotiations between the government and the FARC became public, the OTP had begun to focus on admissibility considerations.
I argue that the threat posed by ICC involvement can incentivize those negotiating peace to put the matter of accountability on the table during negotiations. While those I spoke to pointed to several factors, including the strength of domestic legal institutions and concerns about centering victims in the peace process, many pointed to the involvement of the ICC as an important feature of the negotiations. For example, when I spoke with one member of the government’s negotiating team, this person explained to me that “[i]n all the discussions there was an argument that would always surface which is, if we don’t do this, then the ICC might intervene. It was there [in discussions] about the importance of investigating certain crimes,
$ \dots $
and [in discussions] about what types of sanctions could be compatible with the Rome Statute” (Interview G9, 2018). Other government bureaucrats who advised during the negotiation and implementation of the agreement echoed these sentiments, explaining that the negotiating teams “always thought that the Court was a real concrete risk” (Interview G2, 2017), and that “when [the ICC] are hanging around you and they are watching you, you walk straighter” (Interview G9, 2018). A civil society member and expert on the peace process told me, “I think [Rome Statute obligations] were completely at the center of the negotiations with the FARC. They were looking at how to find alternatives to comply with those obligations, but at the same time give the softest criminal scenario for former combatants.
$ \dots $
That was the whole exercise, how to do that without violating international obligations” (Interview N3, 2017). Another person, an activist and legal academic said, “the argument of the [ICC] was very important during the negotiations to say to the actors: we have to negotiate a liability scheme, it cannot simply be an amnesty scheme because if not the [ICC] is going to come” (Interview A2, 2018).
This threat was particularly relevant for members of the Colombian military, who were facing scrutiny over the false positives scandal, a widespread practice where the military killed innocent civilians and dressed them as rebel combatants to boost body-counts for rewards (Kraul Reference Kraul2012). Some estimates put the total victim count near ten thousand (Daniels Reference Daniels2018). In its 2012 special report, the OTP singled out the military and the false positives killings as a particular point of concern (Office of the Prosecutor 2012). Several individuals I spoke to explained that members of the military involved in false positive killings and other potential areas of interest for the OTP were especially interested in facing a domestic accountability mechanism so they could “get their piece of paper” from the domestic system to avoid prosecution at the ICC (Interview G5, 2017).
But even for FARC combatants, the role the ICC could play in demanding accountability was well understood. One member of the legal team representing the former rebel group explained that the issue of amnesty was taken off the table by considering Colombia’s Rome Statute obligations: “what are serious crimes? War crimes, genocides, crimes against humanity. They are not amnestiable, and that was the first debate in which we all agreed” (Interview F1, 2019).Footnote 19 And while initially, the FARC was resistant to the idea of establishing any TJ system built around formal accountability for atrocities, the ICC threat helped push the FARC negotiating team to agree to establish a new system (Interviews G3, 2017; G5, 2017, 2018; G13, 2019; F1, 2019).
The outcome of those negotiations, the 2016 peace agreement, is widely recognized as one of the most comprehensive peace agreements in history. Notably, the agreement included provisions outlining the establishment of a multi-faceted TJ system. This included the creation of the Special Jurisdiction for Peace (JEP), a tribunal-like institution that allows those accused of atrocities to voluntarily appear and testify about their involvement in order to receive a reduced sentence. If they do not appear, they face a traditional criminal trial with a potential 20-year prison sentence if found guilty.
Intervention as Resolve in Sudan
While the OTP had been formally monitoring the situation in Colombia for years, because the situation in Darfur was referred via UNSC resolution, OTP involvement increased rapidly in a short period of time. While initial ICC involvement raised concerns within the Sudanese government, it was not until the OTP requested the first round of arrest warrants that government negotiators began to seriously consider the possibility of committing to domestic trials in their peace negotiations with Darfuri rebels.
In 2004, during the first round of Darfur peace talks, the rebel movements called for an international inquiry with the goal of bringing perpetrators of atrocities to justice (Nouwen Reference Nouwen2013). A UN Commission of Inquiry was established, and in January 2005, it released its report. The report claimed that government forces had been involved in committing atrocities in the region, including both war crimes and crimes against humanity, and called for a UNSC referral of the situation in Darfur to the ICC (Bosco Reference Bosco2013; Peskin Reference Peskin2009). In March 2005, the Security Council referred Sudan to the ICC, and the OTP opened a formal investigation just three months later.
The government of Sudan responded by publicly welcoming members of the OTP, who arrived in Khartoum to assess Sudan’s internal judicial processes and evaluate its capability of investigating atrocities—they even invited OTP officials on a Nile boat cruise (Bosco Reference Bosco2013). At the same time, Sudanese officials were trying to assess the governments’ own prospects under the complementarity principle. For example, Nouwen (Reference Nouwen2013) describes a senior legal official within the Sudanese Ministry of Justice pointedly asking for training in international criminal law, stating that “[n]ow that the situation in Darfur has been referred to the [ICC], my staff must learn more about this. Especially the principle of complementarity” (2). It even created the Special Criminal Court for Events in Darfur, presumably to address complementarity concerns (de Waal Reference de Waal, Waddell and Clark2008).
Despite government officials’ private concerns about the investigation, when it came to the negotiations with the Darfuri rebel movements, its approach to questions of domestic justice was far more cautious. During the fifth round of peace talks in July 2005, the mediation team removed all discussion of domestic accountability from the agenda due to concerns that it would only harm the negotiations. In fact, the 2005 Declaration of Principles for the Resolution of the Sudanese Conflict in Darfur makes commitments to compensate Darfuri communities to “address grievances for lives lost, assets destroyed or stolen, and suffering caused” and to promote reconciliation, but it never discusses criminal accountability (Government of Sudan, Sudan Liberation Movement/Army and Justice and Equality Movement 2005).
It was not until 2007, when the OTP formally intervened and announced that it was seeking arrest warrants for two government officials, that the government’s approach to both the OTP and the prospect of domestic accountability changed. After the arrest warrants were issued, the Sudanese government took a two-pronged approach to dealing with the ICC. First, its relationship with the OTP became openly hostile: the government stopped responding to requests for information and shut down its limited cooperation with the investigation (Peskin Reference Peskin2009). Second, the government outwardly snubbed the Court, going as far as to elevate Ahmed Haroun, a senior government minister and the subject of an ICC arrest warrant, to a ministerial post for humanitarian affairs (Bosco Reference Bosco2013; de Waal Reference de Waal, Waddell and Clark2008).
When the OTP announced in 2008 that it would seek an arrest warrant for Sudanese President Omar al Bashir, the government responded in a similar fashion: it kicked all major humanitarian aid organizations out of Darfur and rallied its fellow AU and Arab League member states behind an anti-ICC agenda (Mills Reference Mills2012; Peskin Reference Peskin2009). More subtly, it pushed for the UNSC to vote to suspend the ICC investigation, considering the surrender of other ICC fugitives, including Haroun, in exchange for a Security Council suspension (Bosco Reference Bosco2013; Duursma Reference Duursma2020).
Much of the research and media coverage of the ICC investigation in Sudan has focused on how the Sudanese government engaged with the international community in the wake of the arrest warrant for Omar al Bashir. The government’s behavior in the domestic sphere, however, has been largely overlooked. At the same time that it was publicly defying the ICC, the Sudanese government was also trying to address the question of complementarity head-on. When the initial arrest warrants were issued in 2007, there was a proliferation of domestic criminal processes and institutions, all focused on creating local accountability for those individuals wanted by the court: the Sudanese Prosecutor-General initiated proceedings against Haroun, and the Special Court for West Darfur began its own investigations. The government also established the Special Prosecutor for Darfur (Nouwen Reference Nouwen2013). A similar array of institutions was suggested in the wake of the al Bashir warrant.
The warrant against Omar al Bashir also put the question of domestic TJ commitments squarely back on the agenda at the negotiating table. While AU and UN mediators were reluctant to do so, it became clear that any agreement with Darfuri rebels would need to address the question of accountability (Nouwen Reference Nouwen2013). Initial debates at the negotiations were explicitly about the question of cooperation with the ICC investigation: the Darfuri rebel groups, who presented themselves as friends of mankind and had every incentive to center the investigations, called for full cooperation with the OTP, while government representatives made clear that it was non-negotiable (Nouwen Reference Nouwen2013; Nouwen and Werner Reference Nouwen and Werner2010).Footnote 20 These debates were ultimately tabled in favor of discussions about a domestic accountability package, where the negotiating parties all pulled information from the final report of the AU High Level Panel on Darfur.Footnote 21 The 2011 final agreement, the Doha Document for Peace in Darfur, includes sweeping TJ provisions and provides for the creation of a Special Court for Darfur with a mandate of applying Sudanese and international law—including criminal, human rights, and humanitarian law—to bring perpetrators to justice. It provided amnesty for ordinary crimes and explicitly prohibited amnesty for international crimes, including those covered by the Rome Statute.Footnote 22
Unlike in Colombia, initial OTP involvement had limited influence on negotiations between the Sudanese government and the Darfuri rebels. While the early-stage investigation led the government to express interest in the topic of complementarity, it had little effect on the ongoing peace negotiations. Only after the ICC intervened and issued arrest warrants did the government address the threat directly. Their two-pronged approach—openly defying the court while simultaneously seeking to invoke the promises of complementarity—led them to emphasize domestic accountability independently and to accept the idea of trial commitments as part of the peace process.
Assessing Alternative Explanations: Bottom-Up Demands
The role of victims, activists, and bottom-up demands for accountability have frequently been cited as an explanation for why full amnesties have been increasingly rejected as part of TJ (Bell and Kitagawa Reference Bell and Kitagawa2023; Simmons Reference Simmons2009), suggesting that the trial commitments may have been the result of bottom-up demands rather than top down pressures from the ICC in Colombia and Sudan. I explore this alternative explanation in more detail below, evaluating it across the two case studies.
In Sudan, there were few effective bottom-up demands for TJ for several reasons. First, prior to the ICC’s involvement in the country, there was little discussion of TJ. While human rights organizations were present in Sudan, these groups did not advocate for TJ in the negotiations between the government and the Darfuri rebel groups any more than they had in previous peace processes. To the extent that any concern for justice existed, it took the form of individual cases driven by a victim’s family. Prior agreements committed to compensating victims or pursuing reconciliation, but formal or de facto amnesties were common (Nouwen Reference Nouwen2013).
Even where there was a desire to advocate for accountability, the authoritarian nature of the regime limited the ability of justice advocates to push for it. The level of repression was evident in how the Sudanese government responded to local media reporting on the ICC after arrest warrants were issued: officials arrested journalists covering the investigation and implemented widespread formal censorship policies (Sudan Tribune 2008). Even after the formal policies were relaxed, the government encouraged journalists not to cover the cases, with then President al Bashir noting that editors should “avoid mixing what is patriotic and what is destructive to the nation” (Sudan Tribune 2009). While victims and civil society may have had preferences for accountability, the overall political environment in Sudan limited their ability to influence the negotiations.
Bottom-up demands are more evident in Colombia, where there was an active civil society and the idea of a full amnesty was unpopular. Members of both domestic and international civil society advocated for ensuring that those who committed atrocities would face criminal accountability (BBC 2016; Grandin Reference Grandin2016). Those on the political right were particularly vocal in their demands for punishment of FARC members (Meernik, DeMeritt, and Uribe-López Reference Meernik, DeMeritt and Uribe-López2019). And these demands clearly mattered for those negotiating the peace agreement. One person, who negotiated on behalf of the government, explained to me that “all the emphasis so far has been on the victims. And in my view, that was the critical move. That’s what made it possible to come to those agreements with the FARC
$ \dots $
” (Interview G13, 2019). A member of civil society echoed this sentiment, saying that “[it] is clear that social mobilization has been a key strategy [for combating impunity], which has had very specific tools, some more political, others more legal” (Interview N5, 2018).
However, though these groups made independent demands of those negotiating, they also sought to exert influence and ensure accountability by working with the OTP. For example, a coalition of human rights groups provided evidence related to the false positives case directly to the OTP in 2012 after deciding the Colombian government was acting too slowly on the cases (Kraul Reference Kraul2012). And when victim’s organizations learned that the OTP was particularly interested in crimes related to sexual and gender-based violence, several groups called for intervention due to what they described as near total impunity under the existing system (El Espectador 2014).
While the threat of intervention from the OTP was not the only incentive for the creation of the JEP, my interviews with those involved with advising and negotiating the agreement suggest that pressure from the OTP was both present and an important consideration for those negotiating. Personal considerations for members of the military and the FARC, as well as concerns about international legal obligations and reputation, drove much of the discussion about working within the constraints imposed by the court. One person I spoke to summarized this balance, noting that “the two elements just mixed up in the perfect way: the OTP had the eternal preliminary examination, yet there it is, it’s a credible threat, at least in Colombia, and there are strong institutions that say we need to deal with this and strong civil society organizations that are willing to fight for it as well” (Interview G3, 2017).
To What End? Implementing Justice Commitments
This article focuses primarily on the role pressure from above can play in pushing negotiators to commit to holding domestic trials for perpetrators of atrocities. However, commitments must actually be implemented. Implementation can serve as a key metric for evaluating when these commitments are genuine and when they are simply window dressing. While a systematic analysis of trial commitment implementation is beyond the scope of this article, below I briefly discuss implementation efforts in Colombia and Sudan, showing how even attempts at window dressing can lay important foundations for the pursuit of accountability in the wake of changing political circumstances.
In Colombia, the implementation of the agreement had a rocky start, first by failing to pass a national plebiscite, which required a hasty renegotiation and passage through congress. In 2018, Ivan Duque was elected as president after running on what was widely seen as an anti peace process agenda, promising to modify the peace agreement with a particular focus on issues of accountability (Amat Reference Amat2018). And the Duque administration did attempt to derail the implementation of the accords by, for example, refusing to sign the key final piece of legislation formalizing the establishment of the JEP (Murphy and Vargas Reference Murphy and Vargas2019). The JEP began holding its first hearings in July of 2018, and despite Duque’s efforts, the final implementing legislation became law in June 2019. The JEP’s work is ongoing, but in the years since its establishment, it has issued rulings against high-level members of the military and FARC as part of numerous macro-cases, including in cases related to the false positives (Trucco Reference Trucco2023).
In Sudan, the commitment to establish a Special Court for Darfur was always intended as window dressing, at least for government officials. Nevertheless, the government did take steps toward implementation early on, appointing a public prosecutor for the tribunal in 2012 and making perfunctory efforts to hold trials, focusing especially on rebel group members (International Federation for Human Rights 2015; Sudan Tribune 2012). By 2015, further efforts to implement the agreement stalled, and advocates highlighted the government’s “lack of political will to successfully complete investigations and prosecutions for international crimes” (International Federation for Human Rights 2015).
However, the 2019 ouster of Omar al Bashir began to change the political dynamics in Sudan, creating a window of opportunity for those seeking to implement previous trial commitments. The joint military–civilian transitional government quickly initiated new peace talks in Darfur, and the resulting agreement renewed commitments to the creation of a war crimes tribunal, with the promise that al Bashir would face trial for atrocities in Sudan, just as he had for corruption and the coup that brought him to power (BBC 2020; Magdy Reference Magdy2020). But some members of the transitional government wanted to turn al Bashir over to the ICC pursuant to the 2009 warrants (Dahir Reference Dahir2021; Magdy Reference Magdy2020). The debates continued for the following three years, until they were derailed by the onset of civil war in Khartoum. However, for a brief time, elites within Sudan appeared to take genuine steps to implement accountability at home.
CONCLUSION
Since the Rome Statute went into effect in 2002, scholars and policymakers alike have debated what the actual impact of the Court may be. Many have suggested that the ICC can be biased (Ba Reference Ba2020; Boehme Reference Boehme2022), and that it harms prospects for peace (Prorok Reference Prorok2017). In this article, I contribute to a growing literature that addresses these claims directly, providing evidence of the ICC’s impact on a critical component of contemporary conflicts: commitments made during peace processes. I show that ICC involvement can lead negotiating parties to make domestic trial commitments during peace negotiations. I argue that at least one underlying motivation for making these commitments is the threat posed by the prospect of facing accountability in the international arena. Governments and their opponents codify their commitments to domestic accountability so they can signal to the ICC that they are able and willing to handle their own business. I explore the mechanisms in more detail using case studies of peace processes in Colombia and Sudan. In each peace process, the negotiating parties had incentives to avoid committing to accountability, but they nonetheless agreed to establish domestic trials for those accused of committing atrocities. In both cases, the negotiators were aware of ICC involvement and they saw this involvement as a threat. While in Colombia, the threat posed by early ICC involvement was enough, in Sudan, the OTP had to signal resolve by actually intervening to alter the incentives of regime elites.
My findings build on existing work by showing first that institutions like the ICC can impact peace processes in a way that contributes to domestic justice efforts. I show that in some contexts, both state and opposition elites face concerns about ICC involvement and that they see domestic trial commitments as a viable option to avoid intervention. They codify these trial commitments in peace agreements, which can then tie their hands during agreement implementation, regardless of the genuineness of the initial commitment. I also show that the OTP’s influence goes beyond the effects of its direct intervention: at times, ICC involvement that makes concrete the mere threat of future intervention can lead negotiating parties to commit to domestic accountability.
Additional research exploring the effects of tribunal involvement on the negotiation and especially the implementation of peace agreements is necessary. While this article suggests that ICC involvement increases the likelihood of domestic trial commitments, it may be the case that this is more likely to occur under certain conditions. For example, when state actors have committed particularly egregious atrocities, or when rebel groups are especially strong, the ICC’s role as enforcer of the accountability norm may be strongest, and the threat of international trials most present. These contexts include places like Sudan, where accusations of government-sponsored genocide rose all the way to the UNSC, but they also include Colombia, where state agents are accused of the false positive killings.
A systematic evaluation of justice commitment implementation across contexts can also shed important light on the implications of ICC involvement for peace consolidation. The case studies I present show that implementation efforts in Colombia and Sudan both faced setbacks to varying degrees over time, driven in large part by recalcitrant elites opposed to following through on the agreements. Concerns about setbacks in implementation can play a critical role in the breakdown of peace processes. However, the case studies also show that recalcitrant elites may nonetheless find themselves bound by the commitments they make. Understanding more about these dynamics, and how institutions like the ICC can influence their development across a variety of contexts, can make an important contribution to research on norm subversion (Cronin-Furman Reference Cronin-Furman2022; Loken, Lake, and Cronin-Furman Reference Loken, Lake and Cronin-Furman2018; Subotić Reference Subotić2009) and transitional injustice (Loyle and Davenport Reference Loyle and Davenport2016).
My findings also have implications for the study of backlash against international courts and institutions in contemporary politics (Alter, Gathii, and Helfer Reference Alter, Gathii and Helfer2016; Hillebrecht Reference Hillebrecht2021; Madsen, Cebulak, and Wiebusch Reference Madsen, Cebulak and Wiebusch2018; Voeten Reference Voeten2020; Walter Reference Walter2021). Backlash against these institutions has largely been driven by the rise of populist leaders whose regimes have little to gain from liberal internationalism. The leaders have, among other things, argued that international law and institutions are ineffective and fail to reflect the will of the people (Voeten Reference Voeten2020), and of the growing number of states in the international system (Boehme Reference Boehme2022; Brett and Gissel Reference Brett and Gissel2020). My findings suggest a potential counter to at least one of these claims: arguments about effectiveness. At his swearing in as the first ICC Prosecutor, Luis Moreno Ocampo famously stated that the court’s effectiveness “should not be measured by the number of cases that reach it. On the contrary, complementarity implies that the absence of trials before this Court
$ \dots $
would be a major success.”Footnote
23 I find that despite challenges in navigating the delicate balance between peace-building and justice promotion, the ICC can at times deliver on this promise. For scholars of other international institutions seeking to counter claims made in the context of backlash, this article highlights the need to look for effectiveness in different ways.
SUPPLEMENTARY MATERIAL
To view supplementary material for this article, please visit https://doi.org/10.1017/S0003055425101317.
DATA AVAILABILITY STATEMENT
Research documentation and data that support the findings of this study are openly available at the American Political Science Review Dataverse: https://doi.org/10.7910/DVN/2NCI2E.
ACKNOWLEDGEMENTS
For helpful advice and comments, I thank Jessica Weeks, M.P. Broache, Oumar Ba, Lisa Sundstrom, Courtney Hillebrecht, Geoff Wallace, Andres Uribe, Geneva Cole, Alexandra Chinchilla, the Dissertation Improvement Group (DIG) and Political Violence Working Group (PVWG) at the University of Chicago, and audiences at Harvard, UC Berkeley, University of Minnesota, University of Washington, the University of Nebraska–Lincoln Human Rights and Act in Courts Virtual Paper Workshop, and the APSA and ISA annual meetings. I would also like to thank all of my contacts and interlocutors in Colombia and beyond who shared insights, experiences, and much of their time with me.
FUNDING STATEMENT
Portions of this project were funded by fellowships from the Pearson Institute for the Study and Resolution of Global Conflict, the Center for Latin American Studies (CLAS), the Center for International Social Science Research (CISSR), and the Pozen Family Center for Human Rights at the University of Chicago.
CONFLICT OF INTEREST
The author declares no ethical issues or conflicts of interest in this research.
ETHICAL STANDARDS
The author declares that the human subjects research in this article was reviewed and approved by the University of Chicago Institutional Review Board (IRB 17-1246). The author affirms that this article adheres to the principles concerning research with human participants laid out in APSA’s Principles and Guidance on Human Subject Research (2020).
Open access
Comments
No Comments have been published for this article.