![Sample our Politics & International Relations journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5947/TOC-Subject-Sample-2021-Politics-International-Relations.jpg"})
ABSTRACT
Why do some states comply with their legal obligations to arrest suspects indicted by international criminal tribunals (ICTs) while others do not? Research on this question has mostly focused on “target” states, like the former Yugoslav republics, where ICTs have intervened. In contrast, this article offers the first test of theories regarding ICT arrest-warrant compliance and noncompliance by third-party states. I examine the International Criminal Tribunal for Rwanda (ICTR) and 26 third-party states implicated in the pursuit of the court's 91 indicted suspects. Using fuzzy-set qualitative comparative analysis, I find support for the procompliance influence of liberal democratic norms and foreign aid dependency on third-party states. I also find that noncompliance — something existing studies tend to leave untheorized — can be explained by the presence of either non- compliance constituencies or high official corruption. By testing several theories of compliance and noncompliance on a so far understudied class of cases, these findings provide support for the generalizability of a number of explanations in the broader literature on compliance with human rights obligations. The analysis also shows that problematizing noncompliance — and not merely reducing it to an absence of procompliance factors — can help us develop fuller explanations of compliance behavior.
Acknowledgments
I am grateful to Diana Kapiszewski, Wayne Sandholtz, and Charles Anthony Smith for their helpful comments on earlier drafts of this article.
Notes
1. The North Atlantic Treaty Organization (NATO) forces in the former Yugoslavia did eventually pursue and arrest some indicted ICTY suspects, but these were stabilization forces already stationed in these territories and not deployed specifically for this purpose. See Kerr (Citation2004: Chap. 7). Thus, the involvement of foreign or international forces in the arrest of suspects is the exception, not the norm.
2. For the purpose of this study, the concept of “compliance with ICTs” refers to states’ legal obligations to pursue, to arrest, and to transfer suspects indicted by ICTs and believed to be in their territories.
3. For a list of ICTY indictees and the locations where they were apprehended, see data compiled by the American University Washington College of Law War Crimes Research Office: War Crimes Research Office (2006).
4. It may seem intuitive to attribute states’ failures to pursue indicted suspects to weak state capacity. Nevertheless, I follow other authors (e.g., McClendon Citation2009: 350) in assuming that state capacity is usually not a dominant factor in ICT noncompliance. This is not to say that capacity is irrelevant, but it is often the case that once previously noncompliant states decide to pursue suspects, they arrest them relatively quickly.
5. For details regarding the negotiations that preceded the establishment of the ICTR, see Peskin (Citation2008: 156–169).
6. Two individuals were indicted for crimes unrelated to participation in the actual conflict. Léonidas Nshogoza, a defense investigator, was indicted for contempt and surrendered to the court in Tanzania. See The Prosecutor v. Léonidas Nshogoza (2008). A witness, known by the pseudonym “GAA,” was indicted for providing false testimony to the court and arrested in Rwanda. See The Prosecutor v. GAA (2007). I exclude these two indictees from my analysis.
7. The text of Article 28 (UN Security Council 1994) reads: “States shall comply without undue delay with any request for assistance or an order issued by a Trial Chamber, including but not limited to: a) The identification and location of persons;b) The taking of testimony and the production of evidence;c) The service of documents;d) The arrest or detention of persons;e) The surrender or the transfer of the accused to the International Tribunal for Rwanda.”
8. Not all formal democracies — states with popularly elected governments — are liberal democracies. Following Slaughter, I define the latter as also requiring the presence of and a commitment to “separation of powers, constitutional guarantees of civil and political rights, juridical equality, and a functioning judicial system dedicated to the rule of law” (Citation1995: 511).
9. For a detailed treatment of QCA procedure, see Schneider and Wagemann (Citation2012).
10. It may sometimes be possible for regression analyses to approximate causal asymmetry using interaction terms, but this approach is problematic. First, regression-based interaction terms and set-theoretic causal configurations rest on different conceptual foundations, as the former models the multiplicative effect of one variable on another, while the latter refers to the logical intersection of particular sets. Second, causal configurations in QCA solutions often contain three or more conditions, and such high-order interactions are highly problematic — both technically and conceptually — for regression analysis. See Schneider and Wagemann (Citation2012: 87–88).
11. Data are taken from the ICTR website: (ICTR n.d.). The highest number of indictees arrested by any one state was 15 arrested by Kenya over five years. Several states arrested only a single indictee.
12. I also removed one case from my sample, Angola, because of conflicting reports regarding its connection to one arrest. Though the ICTR officially reported that one indictee was arrested in Angola, there were conflicting reports that the individual was actually residing and arrested in DRC. See Prunier (Citation2009: 44n97).
13. This lack of empirical diversity also helps explain why despite the presence of democracy in the solution for Model 1, the noncompliance constituencies solution in Model 2 contains no terms referring to democracy — whether its presence or absence. Nevertheless, the fact that noncompliant Italy scores positive on both democracy and noncompliance constituencies is not inconsistent with the solution in Model 1 that found an effect of democracy on compliance. As I discussed earlier, the set-theoretic foundation of QCA allows for the possibility that the same condition contributes to both the occurrence and nonoccurrence of an outcome depending on its combination with other conditions.
14. For discussion of the problem of endogeneity in compliance research, see Downs et al. (Citation1996).
Additional information
Notes on contributors
Mark S. Berlin
Mark S. Berlin is an Assistant Professor in the Department of Political Science at Marquette University. He was also formerly a guest researcher at the Peace Research Institute Frankfurt. His research focuses on international law and human rights with an emphasis on criminal accountability for atrocities.