University of Colorado, Boulder CU Scholar Political Science Graduate Theses & Dissertations Political Science Spring 1-1-2011 Treaty Content and Costs: Explaining State Commitment to the International Criminal Court Yvonne Marie Dutton University of Colorado at Boulder, yvonne.dutton@colorado.edu Follow this and additional works at: https://scholar.colorado.edu/psci_gradetds Part of the International Law Commons, and the International Relations Commons Recommended Citation Dutton, Yvonne Marie, "Treaty Content and Costs: Explaining State Commitment to the International Criminal Court" (2011). Political Science Graduate Theses & Dissertations. 13. https://scholar.colorado.edu/psci_gradetds/13 This Dissertation is brought to you for free and open access by Political Science at CU Scholar. It has been accepted for inclusion in Political Science Graduate Theses & Dissertations by an authorized administrator of CU Scholar. For more information, please contact cuscholaradmin@colorado.edu.
TREATY CONTENT AND COSTS: EXPLAINING STATE COMMITMENT TO THE INTERNATIONAL CRIMINAL COURT by YVONNE M. DUTTON B.A., Columbia University, 1989 J.D., University of Columbia School of Law, 1991 M.A., University of Colorado at Boulder, 2008 A thesis submitted to the Faculty of the Graduate School of the University of Colorado at Boulder in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Political Science 2011
This thesis is entitled: Treaty Content and Costs: Explaining State Commitment to the International Criminal Court Written by Yvonne Dutton Has been approved for the Department of Political Science Dr. Moonhawk Kim, Committee Chair Dr. Joseph Jupille, Committee Member Date: May 5, 2011 The final copy of this thesis has been examined by the signatories, and we find that both the content and the form meet acceptable presentation standards of scholarly work in the above mentioned discipline.
ABSTRACT Yvonne M. Dutton (Ph.D., Department of Political Science) Treaty Content and Costs: Explaining State Commitment to the International Criminal Court Thesis directed by Dr. Moonhawk Kim The International Criminal Court is the first permanent, treaty-based international criminal court established to help end impunity for perpetrators of genocide, crimes against humanity, and war crimes. More than 100 countries have ratified the treaty creating the court. By doing so, they have agreed to cede to an independent prosecutor the power to prosecute the state s own nationals for mass atrocities when the prosecutor and the ICC court determine the state is unwilling or unable to do so domestically. But, why have states committed to an institution like the ICC which has serious enforcement mechanisms to punish noncompliant behavior? States do regularly ratify the many toothless treaties designed to hold states accountable to respecting individual human rights. However, because enforcement mechanisms in those treaties are weak or non-existent, states can show their good will by ratifying, yet ignore treaty terms when compliance becomes inconvenient. This dissertation examines the puzzle of ICC commitment. I theorize that states will view the ICC s enforcement mechanisms as a credible threat and only commit if their retrospective cost calculations about their ability to comply with treaty terms show that commitment will not lead to a significant sovereignty loss. In this case, states should consider (1) the institutional design of the treaty specifically, the level of enforcement mechanisms to punish noncompliance and (2) the state s domestic characteristics relating to its ability to comply iii
with treaty terms. Empirical findings here provide support for the credible threat theory. In contrast to prior studies empirically examining state commitment to international human rights treaties, I find that states with poorer human rights practices are less likely than states with good practices to commit to the ICC. Although this means that member states tend to have relatively good human rights practices, it does not imply that the ICC will not positively influence state behavior. Indeed, the ICC is uniquely situated to improve international cooperation on human rights matters since it has been designed so that commitment requires compliance. All states that have joined the court including those with poor practices will have to comply or face sovereignty losses. iv
Contents ABSTRACT... iii List of Tables... xi List of Figures... xiii CHAPTER ONE INTRODUCTION... 1 The Puzzle of State Commitment to the International Criminal Court... 1 The Argument: Strong Enforcement Mechanisms as a Credible Threat... 7 Research Design... 12 Organization of the Dissertation... 15 CHAPTER TWO THEORIES OF COMMITMENT... 25 Existing Literature... 25 The Rationalist View... 25 Retrospective Calculations... 26 Prospective Calculations... 31 The Normative View... 36 Synthesis and Analysis: Literature Addressing State Decisions to Join the ICC... 38 Strong Enforcement Mechanisms as a Credible Threat... 42 CHAPTER THREE TESTING STATE COMMITMENT TO THE INTERNATIONAL CRIMINAL COURT... 50 Methodology... 50 Dependent Variable... 50 Independent Variables... 51 v
The Main Explanatory Variables -- Level of Human Rights Practices and Level of Domestic Law Enforcement Institutions... 51 Control Variables: The Rationalist View... 53 Control Variables: The Normative View... 56 Empirical Analyses and Discussion of Results... 59 CHAPTER FOUR... 73 TESTING STATE COMMITMENT IN THE CONTEXT OF OTHER INTERNATIONAL HUMAN RIGHTS TREATIES... 73 International Human Rights Treaties and their Enforcement Mechanisms... 74 Methodology... 79 Dependent Variables... 80 Independent Variables... 80 The Main Explanatory Variable: Level of Human Rights Practices... 80 Control Variables: The Rationalist View... 82 Control Variables: The Normative View... 83 Empirical Analyses and Discussion of Results... 84 CHAPTER FIVE AN INTRODUCTION TO THE CASE STUDIES... 98 The Goals of the Qualitative Analyses... 98 The Logic of Case Selection... 101 States with Very Good Human Rights Practices and Domestic Law Enforcement Institutions... 102 States with Very Good Human Rights Practices, but Worse Domestic Law Enforcement Institutions... 105 States with Very Bad Human Rights Practices and Domestic Law Enforcement Institutions... 107 vi
The Uniqueness of the United States Regarding the ICC... 111 Conclusion... 115 CHAPTER SIX GERMANY: AN INTERNATIONAL CRIMINAL LAW CONVERT... 118 Background: The Shameful Past... 120 Germany s Human Rights Violations during World War I and the Failed Attempts to Hold Violators Accountable... 121 Germany s Human Rights Violations during World War II and the Nuremberg Trials... 124 After World War II: Germany and the Protection of Human Rights... 126 The Reaction to the Nuremberg Trials... 128 German Confrontation of its Nazi Past: An Evolution... 131 Germany and the International Human Rights Regime: International Commitment Follows Domestic Reforms... 135 Strong Leadership in Negotiations and Support for a Strong Court... 137 Germany s Initial Position on the ICC... 140 The Issue of the Court s Jurisdiction... 141 The Role of the ICC Prosecutor... 144 The Issue of State Surrender of Nationals to the ICC... 146 Germany s Record on ICC Commitment and Compliance... 148 Signing and Ratification... 148 National Implementing Legislation... 150 Germany s Leadership Role in Advancing Commitment and Compliance Post-Ratification... 155 Germany and the ICC: Assessing the Explanatory Power of the Credible Threat Theory... 156 vii
CHAPTER SEVEN TRINIDAD AND TOBAGO: COMPLIANCE BEFORE NORMS... 161 The Proposal to Establish an International Criminal Court that Could Meet Trinidad and Tobago s Domestic Needs... 165 The Impetus: The Narcotics Trafficking Problem in Trinidad and Tobago... 165 Trinidad s Support for a Court with Jurisdiction over Narcotics Trafficking Offenses and with Authority to Impose the Death Penalty... 167 Ratification of the ICC Treaty... 170 Commitment to the International Human Rights Regime: A Focus on Compliance Costs. 175 Trinidad and Tobago s Death Penalty and Corporal Punishment Practices... 177 Trinidad and Tobago s Refusal to Ratify the Convention Against Torture... 178 Commitment to, and Denunciation of, the ICCPR Optional Protocol and the American Convention... 180 Trinidad and Tobago and the ICC: Assessing the Explanatory Power of the Credible Threat Theory... 186 CHAPTER EIGHT RWANDA: CREDIBLE THREAT, NOT CREDIBLE COMMITMENT 193 Background: The Genocide of the 1990s... 200 Colonial Rule and the Ethnic Divide... 200 The 1994 Genocide... 201 The Aftermath of the Genocide: Kagame Closes the Refugee Camps... 203 Violence in Closing the Kibeho Camp... 204 Kagame s 1996 Attack on the Paramilitary Camps in Zaire... 206 Rwanda and the ICTR... 208 Establishment of the ICTR and Rwanda s Lone Dissenting Vote... 208 Rwanda and the ICTR in Practice: The Issue of RPF Indictments... 213 Rwanda s Participation in the International Human Rights Regime: Avoiding Costly Commitment... 218 viii
Rwanda and the ICC: Assessing the Explanatory Power of the Credible Threat Theory... 223 CHAPTER NINE... 231 KENYA: HOPE BECOMES REGRET?... 231 Background: Kenya s Transition from the Moi to the Kibaki Presidency... 238 Kenya s Participation in the International Human Rights Regime: Avoiding Costly Commitment... 240 Kenya s Commitment to the ICC: Rationality Succumbs to Pressure... 241 Kenya s Commitment to the ICC Post-Ratification: Mainly a Record of Non-Compliance. 246 The Refusal to Sign a Bilateral Immunity Agreement... 246 The 2007 Post-Election Violence and the ICC Investigation... 248 Kenya Invites Sudan s President Bashir to Celebrate its New Constitution... 252 Kenya and the ICC: Assessing the Explanatory Power of the Credible Threat Theory... 254 CHAPTER TEN CONCLUSION... 261 Credible Threat and the Case for Stronger Enforcement Mechanisms... 261 BIBLIOGRAPHY... 277 Appendix A... 287 States Parties to the ICC Treaty and Ratification Dates (as of December 31, 2008)... 287 Appendix B... 288 States Parties to the 14 Different Treaties, Articles, and/or Protocols (as of 2010).... 288 Appendix C... 293 Countries by World Bank Regions... 293 Sub-Saharan Africa... 293 East Asia/Pacific... 293 ix
Eastern Europe/Central Asia... 294 Latin America/Caribbean... 294 Middle East/North Africa... 295 South Asia... 295 The West... 295 x
List of Tables Table 1: Comparison of Empirical Works Testing Commitment to the ICC... 39 Table 2: State Commitment Decisions Expectations... 49 Table 3: Summary Statistics... 58 Table 4: ICC Treaty Ratification Patterns Based on Likelihood of Human Rights Violations... 59 Table 5: ICC Treaty Ratification Patterns Based on Level of Domestic Law Enforcement Institutions... 62 Table 6: ICC Treaty Ratification Patterns for States with Poor Human Rights Practices Based on Whether Democracy or Not... 64 Table 7: Cox Proportional Hazards Models Explaining Ratification of the ICC Treaty... 69 Table 8: The Six Primary International Human Rights Treaties... 75 Table 9: 14 Human Rights Treaties and Levels of Enforcement Mechanisms... 79 Table 10: Summary Statistics for Common Independent Variables... 84 Table 11: Ratification of the 14 Different Treaties Based on Human Rights Ratings... 85 Table 12: Poorest Human Rights Countries and Ratifications... 87 Table 13: Weakest Enforcement Mechanisms... 90 Table 14: Stronger Enforcement Mechanisms... 95 Table 15: States Selected for Case Studies Based on ICC State Commitment Decisions Expectations... 101 Table 16: States with Very Good Human Rights Practices and Domestic Law Enforcement Institutions... 102 Table 17: Very Good Human Rights Practices, but Worse Domestic Law Enforcement Institutions... 105 Table 18: Very Bad Human Rights Practices and Domestic Law Enforcement Institutions... 108 Table 19: Commitment to the Six Primary International Human Rights Treaties... 114 xi
Table 20: Commitment to the Six Primary International Human Rights Treaties... 136 Table 21: Commitment to the Six Primary International Human Rights Treaties... 177 Table 22: Commitment to the Six Primary International Human Rights Treaties... 219 Table 23: Commitment to the Six Primary International Human Rights Treaties... 241 xii
List of Figures Figure 1: States Ratifying the Rome Statute over Time (as of 2008)... 51 Figure 2: ICC Treaty Ratification by Region (as of October 2010)... 57 Figure 3: ICC Ratification by Region over Time... 57 Figure 4: ICC Ratification by Average CIRI Human Rights Ratings (percentage within each rating category)... 61 Figure 5: Ratification of the 14 Different Treaties Based on Average Human Rights Ratings... 86 Figure 6: Case Study States Human Rights Ratings by Year... 116 Figure 7: Case Study States Rule of Law Ratings from 1996 to 2008... 117 xiii
CHAPTER ONE INTRODUCTION The Puzzle of State Commitment to the International Criminal Court The creation of the International Criminal Court (the ICC ) on July 1, 2002 is a remarkable event for many reasons. The existence of the court is the result of a journey that commenced with the Nuremberg trials after the conclusion of World War II and during which nations have searched for ways to ensure that states and individuals protect against, and are deterred from committing, human rights abuses. The idea of a permanent international criminal court dates from at least 1948 when the Genocide Convention referenced the possibility of individuals being tried by such international penal tribunal as may have jurisdiction. 1 The International Law Commission (the ILC ) was tasked with preparing draft statutes for such a permanent court in 1951. However, the Cold War intervened, and it was not until another four decades had passed that the global community again took up the idea of an international criminal court. 2 But even after the idea of a permanent international criminal court again became a reality, the 1994 ILC draft statute envisaged a relatively weak institution which would allow states to guard much of their sovereignty. For example, regarding the court s jurisdiction, the 1994 ILC draft provided that by virtue of ratifying the statute creating the court, states would only confer automatic jurisdiction over the crime of genocide. 3 For other crimes, such as crimes against humanity and war crimes, all states that could otherwise assert jurisdiction over the matter (for example, the state where the acts were committed or the state with custody over the accused) 1 Genocide Convention, Art. VI. 2 Antonio Cassese, International Criminal Law, 2 nd Edition (Oxford: Oxford University Press, 2008), 323-28. 3 1994 ILC Draft, Art. 21. The ILC s 1994 Draft is available at Report of the International Law Commission on the work of its forty-sixth session, U.N. GAOR, 4 9th Sess., Supp. 10, at 44, U.N. Doc. A/49/10 (1994). 1
would have to consent to conferring jurisdiction upon the international criminal court. As to the initiation of investigations and prosecutions, only states or the Security Council as opposed to an independent prosecutor could commence proceedings. 4 Finally, according to the ILC draft, any permanent member of the Security Council would be able to use its veto power to prevent the ICC from exercising jurisdiction over a matter since no prosecution could be commenced without Security Council approval. 5 Despite the opposition of some major powers, however, the majority of states rejected this conservative institutional design that preserved state autonomy and sovereignty and instead opted to create an entirely new type of international human rights institution. By the Rome Statute which was adopted in July 1998, 6 states created an ICC with a strong and independent prosecutor and court with significant and legally-binding enforcement powers to encourage state compliance with the goal of ending impunity for perpetrators of the most serious crimes of genocide, crimes against humanity, and war crimes. 7 By committing to the Rome Statute, states grant the court automatic jurisdiction over those core crimes. 8 Moreover, states agree that an independent ICC prosecutor may initiate investigations against their nationals for the covered 4 1994 ILC Draft, Arts. 23 and 25. 5 Adriaan Bos, From the International Law Commission to the Rome Conference (1994-1998), in Antonio Cassese, Paola Gaeta, John R.W.D. Jones, eds., The Rome Statute Of The International Criminal Court: A Commentary, Volume 1 (Oxford University Press: Oxford, 2002), 49-50. 6 The Rome Statute was finally adopted in July 1998 during a United Nations conference in Rome. Attending the conference were 160 states, 33 international governmental coalitions, and a coalition of more than 200 nongovernmental organizations ( NGOs ). Caroline Fehl, Explaining the International Criminal Court: A Practice Test for Rationalist and Constructivist Approaches, European Journal of International Relations 10 (2004): 362. Of those states in attendance, 120 voted in favor of adopting the statute, seven voted against, and 21 abstained. In July 2002 after the required 60 states had ratified the statute, the ICC came into existence. 7 Rome Statute of the International Criminal Court, July 17, 1998, U.N. Doc. A/CONF 183/9 (1998), Preamble, 4 & 5. At the present time, the crimes over which the ICC does have jurisdiction are genocide, crimes against humanity, and war crimes. Art. 5. The parties to the Rome Statute also declared that the ICC will have jurisdiction over the crime of aggression once a provision is adopted defining that crime and setting out the conditions under which the court can exercise jurisdiction over it. Arts. 5-8. The parties agreed to that definition at the 2010 Rome Statute Review Conference in Kampala. However, the ICC will not be able to exercise jurisdiction over the crime of aggression until after January 1, 2017, and after the parties vote to amend the Rome Statute accordingly. See Resolution RC/Res. 6 at Annex 1, 2 & 3(3). http://www.icc-cpi.int/iccdocs/asp_docs/resolutions/rc-res.6- ENG.pdf. 8 Rome Statute, Arts. 5-8, 11, and 12(2). 2
crimes on his own with the approval of the court or based on referrals from a State Party or the United Nations Security Council. 9 The prosecutor and court operate without direct United Nations Security Council oversight, with the Council having no veto power over what situations are investigated or prosecuted. 10 In addition, the treaty does not recognize any immunity that states may otherwise grant to heads of state who engage in criminal activity. 11 Finally, although it is true that under the complementarity provision of the Rome Statute the ICC operates as a court of last resort, the ICC will obtain jurisdiction over the nationals of States Parties where the state is unwilling or unable genuinely to proceed with a case. 12 Unwillingness includes instances where national proceedings are a sham or are inconsistent with an intention to bring the person to justice, either because such proceedings are unjustifiably delayed or are not being conducted independently or impartially. 13 The idea behind including the unwillingness provision was to preclude the possibility of sham prosecutions aimed at shielding perpetrators due to, for example, government participation in, or complicity with, the offense. 14 A nation s inability to prosecute includes instances where, because of the collapse or unavailability of its national judicial system, the nation cannot obtain the accused or the necessary evidence, or is otherwise incapable of carrying out the proceedings. 15 It bears noting that the ICC not States Parties will determine whether the unwilling or unable bases for proceeding before the court have been met. Accordingly, although states initially envisioned creating a weak institution or one 9 Rome Statute, Arts. 13-15. 10 Christopher Rudolph, Constructing an Atrocities Regime: The Politics of War Crimes Tribunals International Organization 55 (2001): 679-80; Lionel Yee, The International Criminal Court and The Security Council: Articles 13(b) and 16, in The International Criminal Court: The Making of the Rome Statute, ed. Roy S. Lee (Springer, 1999), 143-52; Jack Goldsmith and Stephen J. Krasner, The limits of idealism, Daedelus (2003): 47. 11 Rome Statute, Art. 27. 12 Rome Statute, Preamble 10 & Art. 17(1)(a). 13 Rome Statute, Art. 17(2). 14 John T. Holmes, The Principle of Complementarity, in The Making of the Rome Statute, 50. 15 Rome Statute, Art. 17(3). 3
controlled by powerful nations when the idea of the ICC again resurfaced in the early 1990s, during the course of negotiations, the majority of states ultimately determined instead to create an institution with stronger enforcement mechanisms to enforce compliance with treaty terms and punish noncompliant behavior. That the 1998 ICC negotiations produced a stronger institution than what was originally anticipated is interesting. Indeed, as discussed in more detail below, the ICC is different from other human rights treaties that have gone before it since most of those treaties contain weak or non-existent enforcement mechanisms. The ICC is also different from the ad hoc criminal tribunals which are specially created and which are generally imposed by powerful states upon the perpetrators of particular atrocities. Nevertheless, while questions about why and how states decided in July 1998 to create this unique institution with its unique enforcement mechanisms is certainly an interesting question, once states enacted the Rome Statute, its existence became exogenous to each state. The purpose of this study is to better understand the ratification decisions states made given the ICC treaty s particular institutional design. Thus, the question this dissertation asks is why states would commit to an institution like the ICC inasmuch as commitment can have such profound effects on their sovereign right to mete out justice within their own borders? It is true that the ICC treaty is not the first treaty which purports to bind states to protect individuals against human rights abuses occurring within the state s own territory. And, it is true that states regularly commit to such international human rights treaties. But, those treaties typically contain weak enforcement mechanisms that pose little risk for the state that fails to comply with treaty terms. As such, states can join them almost indiscriminately and without any intention of complying. 16 In the case of the ICC, however, 16 Oona A. Hathaway, The Cost of Commitment, Stanford Law Review 55 (2003): 1856-57; Oona A. Hathaway, Do Human Rights Treaties Make a Difference?, Yale Law Journal 111 (2002): 1982-87; Emilie M. Hafner-Burton 4
states run the risk that the ICC prosecutor will choose to investigate the state s own citizens and haul those citizens to stand trial at an international criminal court situated in The Hague. Therefore, while it may be reasonable for states to commit to treaties with weak enforcement mechanisms, the fact that more than 100 states have now committed to the ICC and its strong enforcement mechanisms poses a puzzle. 17 That more than 100 states have ratified the ICC treaty is puzzling because states typically guard their sovereignty and are reluctant to join international treaties with strong enforcement mechanisms particularly if they cannot comply with treaty terms. Can we expect that the more than 100 states that have ratified the ICC treaty all will abide by treaty terms and protect against human rights abuses and/or domestically prosecute any of their citizens who perpetrate mass atrocities? Does the fact that these states willingly committed to an international human rights treaty with relatively strong enforcement mechanisms mean that they are also committed to the goal of ending impunity for perpetrators of mass atrocities? After all, the intent of strong enforcement mechanisms must be to enforce compliance with treaty terms in this case by ensuring that perpetrators or mass atrocities are brought to justice and other potential perpetrators are deterred as a result. On the other hand, approximately 90 states are still not parties to the ICC treaty, and some states that do belong to the court ratified the treaty less swiftly than others. Why did these states fail to ratify the Rome Statute or ratify more slowly than others? Given the treaty s & Kiyoteru Tsutsui, Human Rights in a Globalizing World: The Paradox of Empty Promises, American Journal of Sociology 110 (2005): 1374. 17 As of October 2010, some 139 countries had signed the Rome Statute, and 114 had actually become States Parties to it. Of the States Parties, about 20 are from Western Europe, 17 are from Eastern Europe, 31 are from Africa, 14 from Asia, and 25 are from Latin America and the Caribbean. The United States, Israel, China, Russia, Indonesia, and India are notable powerful states that have declined to ratify the treaty. Also not parties to the treaty are a number of Islamic and African states, including Bahrain, Iran, Iraq, Kuwait, Pakistan, Qatar, Syria, Turkey, United Arab Emirates, Yemen, Algeria, Angola, Cameroon, Cape Verde, Cote d Ivoire, Egypt, Morocco, Sudan, Tunisia, and Zimbabwe. In June 2010, Bangladesh became the first country from southern Asia to join the court. 5
relatively strong enforcement mechanisms, should we expect that states with the worst human rights practices are among the states that have not ratified? For these states, joining international human rights treaties with weak enforcement mechanisms may be in their rational self-interest. Joining the ICC, however, is a different matter. But, if the majority of states joining the court are also those that already have the best human rights practices, can the ICC really have a significant impact on improving universal respect for human rights and deterring mass atrocities? This dissertation explores these questions and seeks to understand the puzzle of state commitment to the ICC. By exploring this puzzle, this dissertation will contribute to our understanding of why states joined or refused to join the ICC. In addition, the dissertation should also contribute to our understanding of how institutional design and enforcement mechanisms in particular affects state commitment and compliance in the context of the international human rights regime more broadly. Understanding commitment and also why states refuse to commit is important for evaluating the role of international human rights treaties and state ratification of those treaties in bettering state human rights practices and ending impunity for those who abuse individual human rights. Moreover, this study is unique in that it examines the puzzle of ICC commitment using both quantitative methods and qualitative case studies. Indeed, few studies have empirically examined the puzzle posed by state commitment to the ICC treaty, and none have specifically examined whether and how the apparently strong enforcement mechanisms associated with the treaty creating the court influence state commitment decisions. 18 18 Moreover, the few studies that do empirically examine state commitment to the ICC posit different theories, employ different dependent variables in different empirical models, and reach different conclusions about what variables are and are not driving ICC state behavior. See Beth A. Simmons and Allison Danner, Credible Commitments and the International Criminal Court, International Organization 64 (2010): 240-46 (arguing that non-democracies with recent civil wars are most likely to commit to the ICC because they want to tie their own hands and limit their ability to commit mass atrocities); Jay Goodliffe and Darren Hawkins, A Funny Thing Happened on the Way to Rome: Explaining International Criminal Court Negotiations, The Journal of Politics 71 6
The Argument: Strong Enforcement Mechanisms as a Credible Threat Building on existing scholarship, 19 I theorize that commitment to international human rights treaties is a function of two considerations relative to the costs of noncompliance: (1) the institutional design of the treaty specifically, the level of enforcement mechanisms to punish noncompliance and (2) the state s domestic political characteristics relating to its ability to comply with treaty terms. Traditionally, international human rights have been designed with weak enforcement mechanisms. For example, most require only that states self-report compliance. 20 Even the additional optional enforcement mechanisms contained in some treaties whereby states agree to accept state or individual complaints alleging noncompliance are far from strong. 21 In fact, the committees who review the complaints are not empowered to issue decisions that are actually (2009): 977 (arguing that a state s dependence networks are a primary influence on ICC commitment); Judith Kelley, Who Keeps International Commitments and Why? The International Criminal Court and Bilateral Nonsurrender Agreements, American Political Science Review 101 (2007): 573 (testing ICC commitment preliminary to the study s main focus on determining why states that joined the ICC would also sign or refuse to sign bilateral immunity agreements with the United States ). 19 Oona Hathaway is generally credited with first examining empirically the relationship between state human rights ratings and their tendency to enter into international human rights treaties. However, in her 2003 study, the only variables she used to test treaty commitment to several human rights treaties were a state s human rights ratings and whether or not it was a democracy without accounting for the timing of ratifications. Although she acknowledged that other variables may influence commitment, she purposely limited her study. Hathaway, The Cost of Commitment, 1849. In a later study, Hathaway included some additional variables testing commitment to several human rights treaties using a Cox proportional hazards model, though her study did not include the ICC. Oona A. Hathaway, Why Do Countries Commit to Human Rights Treaties?, Journal of Conflict Resolution 51 (2007): 588. 20 For example, under the International Covenant on Civil and Political Rights (the ICCPR ), states agree to submit reports on the measures they have undertaken to give effect to the matters addressed in the Covenant. ICCPR, Art. 40. The main treaties of the five other international human rights treaties similarly provide only that states selfreport compliance. International Covenant on Economic and Social Rights (the ICESCR ), Arts. 16 & 17; International Convention on the Elimination of All Forms of Racial Discrimination (the CERD ), Art. 9; The Convention on the Elimination of All Forms of Discrimination Against Women (the CEDAW ), Arts. 18 & 21; the Convention Against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment (the CAT ), Art. 19; and the Convention on the Rights of the Child (the CRC ), Art. 44. 21 For example, under Articles 21 and 22 of the CAT, states recognize the competence of the Committee Against Torture to hear complaints by states and individuals, respectively. Under Article 41 of the ICCPR, states may agree to recognize the competence of the United Nations Human Rights Committee to consider complaints of one state against another claiming the party is not fulfilling its obligations under the treaty. 7
binding on the states. 22 At least some scholars have suggested that these traditionally weak enforcement mechanisms may explain why states with poor domestic human rights practices nevertheless ratify these treaties designed to protect and promote human rights. 23 Even states with poor domestic practices may view commitment to these treaties as essentially costless from a sovereignty standpoint simply because the treaty enforcement and monitoring mechanisms are weak. 24 In short, the indirect and sometimes intangible benefits associated with appearing to be a state that embraces international human rights norms may outweigh the costs of commitment where treaty enforcement mechanisms to punish noncompliant behavior are weak or non-existent. However, as compared to the international human rights treaties that have preceded it, the ICC treaty has a unique institutional design. It contains relatively strong enforcement mechanisms in that it delegates enforcement authority to an independent decision-making body with resources that can be used to prevent abuses or punish offenders. If states view the enforcement mechanisms associated with the ICC treaty as strong enough to pose a credible threat to their sovereignty, they should only commit if their retrospective calculations about the costs of complying with treaty terms suggest that the risks to their sovereignty by joining the court are minimal. First and foremost, compliance with ICC treaty terms requires that the state have good human rights practices since the state can best avoid the specter of an ICC prosecution if its leaders and citizens do not commit the kinds of mass atrocities within the court s jurisdictional purview. As a secondary matter, states can also comply with ICC treaty terms and avoid having their citizens tried in The Hague if they have independent domestic law 22 In the instances described above, the Committee s enforcement mechanism is limited to attempting to facilitate a resolution or providing a report of its findings. See CAT, Arts. 21 & 22(7); ICCPR, Art. 42. 23 Hathaway, The Cost of Commitment, 1825; Hafner-Burton and Tsutsui, Human Rights in a Globalizing World, 1378. 24 Hathaway, The Cost of Commitment, 1838-40. 8
enforcement institutions that are also capable of prosecuting any human rights violations within their own states. As such, good human rights practices and independent and capable domestic law enforcement institutions are each individually sufficient for states to conclude that ratifying the ICC treaty will not be overly costly. Thus, if a country has either good human rights practices or independent and capable domestic law enforcement institutions, it should conclude that ICC ratification is relatively costless and should commit to the court. In addition, for a state to conclude that ratification of the ICC is essentially costless because the state can comply with treaty terms, either good human rights practices or independent and capable domestic law enforcement institutions are a necessary condition to ratification. However, because the ICC is able to scrutinize whether member states domestic prosecutions are adequate to ward off an ICC investigation, I suggest that good human rights practices become almost a necessary condition to ratification. Indeed, states concerned about compliance costs may not want to rely only on their own assessment of the independence and capability of their domestic institutions and would be wise to insure that their human rights practices are sufficiently good before committing to the court. On the other hand, a state with weak domestic institutions could nevertheless conclude that commitment to the ICC treaty would be relatively costless as long as the state s human rights practices are good. Of course, even states that meet these conditions may decline to commit to the ICC for other reasons. Accordingly, I predict that states with better human rights practices and independent and capable domestic law enforcement institutions will view the costs of complying with the ICC treaty s terms as relatively minimal and more readily commit to the court. In fact, because the primary way in which states can avoid an ICC investigation and prosecution is by having 9
relatively good human rights practices, even with weak domestic law enforcement institutions, those states should conclude that the costs of ICC commitment will be relatively minimal. States with poor human rights practices and biased or incapable domestic law enforcement institutions, however, should view commitment to the ICC treaty as costly even if previous studies suggest they regularly commit to other international human rights treaties. Indeed, even with independent and capable institutions (and there may be few states with such characteristics, in any event), a state with poor human rights practices may view the costs of compliance as being significantly high that it will refuse to commit to the court. This study focuses on the costs associated with ICC commitment, rather than any benefits states may receive as a result of joining the ICC, precisely because the ICC s relatively strong enforcement mechanisms can result in a significant loss of state sovereignty. As a preliminary matter, it is worth noting that on the whole, international human rights treaties are different from other treaties like arms control agreements and trade agreements which by their very terms provide tangible reciprocal benefits to states in exchange for their pledge to act in particular ways. 25 But, the ICC treaty is different even from other international human rights treaties because it has relatively strong enforcement mechanisms that can punish noncompliant behavior. Thus, while states might still behave rationally by ratifying toothless international human rights treaties solely because ratification can provide them some intangible benefits (such as the feelings associated with being a legitimate member of the world community) or indirect benefits (such as increased aid or trade), states should be less likely to approach ICC ratification by focusing primarily on such potential benefits. In short, although states might still hope to gain some benefits from joining the ICC and signaling their legitimacy as a state that respects and protects human rights, for a state that cannot comply with the ICC treaty s terms, any of those 25 Hathaway, The Cost of Commitment, 1823. 10
uncertain benefits will be overwhelmed by the sovereignty costs associated with noncompliance. Thus, I expect states will first calculate the potential costs of commitment and forgo those potential intangible and indirect benefits unless they can and will comply with treaty terms. I particularly focus on the compliance costs derived from the specific language of the text of the ICC treaty rather than any costs of commitment that may be unrelated to the treaty s precise terms (for example, the costs associated with a state s domestic ratification processes) for several additional reasons. First, in all cases of treaty ratification, one primary guide of a state s obligations and the risks associated with failing to comply with those obligations is the terms and provisions of the treaty. Second, in the case of the ICC, the institutional design of the treaty and its enforcement mechanisms are unique. This is not a case where states can look to other similar treaties or the actions of treaty bodies that oversee compliance with other similar treaties to help them interpret the actual strength and meaning of a treaty s enforcement mechanisms. In addition, by contrast to the previous international human rights treaties which contain only weak enforcement mechanisms, states should inherently have something to fear from an independent prosecutor and court. And, while states that wait to ratify the ICC may be able to look at the actions of the ICC prosecutor and the court to help them determine whether the treaty s enforcement mechanisms are actually as strong as they appear to be on paper, states that ratified promptly had only the treaty text on which to rely when making their commitment decisions. Finally, I focus specifically on the costs associated with complying with treaty terms because in the international human rights context (and others as well), a treaty s institutional design and its enforcement mechanisms can have implications for understanding state behavior and also the likelihood that the treaty s purposes and goals will be realized. The ICC treaty has 11
an institutional design and enforcement mechanisms that set it apart from other prior international human rights treaties. Presumably, states structured the ICC treaty in this way because they wanted to ensure compliance with its terms and deter human rights abuses by ending the culture of impunity whereby domestic governments either commit such abuses or fail to bring to justice those within their jurisdiction who perpetrate atrocities. Knowing why states commit to such a regime, and the kinds of states that commit to such a regime, should provide insights about whether structuring international human rights treaties with stronger enforcement mechanisms can ensure greater compliance with international human rights norms. Research Design I use a mixed methods research design to test the credible threat theory. Using event history analysis and case studies, I examine whether states view the ICC s enforcement mechanisms as a credible threat and actually engage in retrospective cost calculations about their ability to comply with treaty terms and provisions and, as such, their level of human rights practices and the independence and capability of their domestic law enforcement institutions when making commitment decisions. The quantitative analyses test state commitment to the ICC drawing on a database that includes more than 190 countries over the time period between 1998 and 2008. Although it is not, nor could it be, conclusive, the empirical evidence provides support for the idea that states view the ICC s enforcement mechanisms as a credible threat and more readily join the court when their retrospective calculations about their ability to comply with treaty terms indicates that commitment will not impose significant sovereignty costs. The results indicate that states with better human rights practices are more likely than states with poorer human rights practices to commit to the court. In addition, although the evidence is less conclusive about the role domestic law enforcement institutions play in state ratification 12
decisions, there is evidence showing that democracies are more likely to commit to the court even the relatively few democracies with poorer human rights practices. This finding provides support for the idea that states that already have checks on government power, such as through independent judicial institutions, view commitment to the ICC as less costly. Thus, these states have less to fear from the additional threat of an ICC prosecution since they should expect that they or their citizens would be prosecuted in any event if they committed the kinds of mass atrocities within the ICC s jurisdiction. The quantitative tests comparing state decisions to commit to the ICC with state decisions to commit to 13 other international human rights treaties, articles, and/or optional protocols which I arrange according to their level of enforcement mechanisms provide further support for the credible threat theory. The results of event history analysis for the years between 1966 and 2008 show that states commit to treaties with the weakest enforcement mechanisms (state reporting) willingly and regularly, but without regard to their level of human rights practices. Rather, the empirical examination of state commitment to six different human rights treaties with the same weak enforcement mechanism suggests that states view treaties with weak enforcement mechanisms as costless and commit for the purpose of window dressing only and without regard for compliance concerns. 26 The results of tests of ICC commitment are in stark contrast to these findings. A state s human rights ratings are a highly significant and positive predictor of whether a state will join the ICC treaty and risk running afoul of its relatively strong enforcement mechanisms. Thus, where enforcement mechanisms are relatively strong, the empirical evidence indicates that calculations about the costs of commitment are significantly influencing states ratification behavior. I use case studies to add depth and understanding to the empirical findings by tracing the 26 Hafner-Burton and Tsutsui, Human Rights in a Globalizing World, 1374. 13
historical and political processes underlying state ratification behavior, particularly as they relate to a state s ability to comply with treaty terms and the calculations states make in that regard. The four cases selected for study demonstrate variance on the dependent variable of ICC commitment and also on the main independent variables of interest: a state s level of human rights practices and the quality of its domestic law enforcement institutions. In addition, the cases for individual study both fit and do not fit my theoretical expectations. For example, both Germany and Trinidad and Tobago have had consistently good human rights practices since the ICC treaty was available for ratification in 1998. Germany also has independent and capable domestic law enforcement institutions; Trinidad and Tobago s institutions, however, are somewhat weak. Both countries promptly ratified the ICC treaty, and the evidence suggests that they did so based on rational and backward-looking calculations about the costs of compliance which for both countries were minimal given their relatively good human rights practices. On the other hand, both Rwanda and Kenya have had, and continue to have, poor human rights ratings and biased and/or weak domestic law enforcement institutions. Rwanda did not ratify the ICC which is behavior consistent with theoretical expectations because the country s past and present human rights records and its relatively weak domestic law enforcement institutions indicate it may have difficulty complying with treaty terms. Furthermore, the case study analysis indicates that the costs associated with ICC compliance best explain Rwanda s refusal to commit to the court. By contrast, Kenya did join the ICC in 2005 a decision which is inconsistent with the expectations of the credible threat theory since even at that time, Kenya s prospects for compliance with ICC treaty terms were at least uncertain. And, Kenya has not complied with treaty terms. Rather, Kenya has become the subject of the ICC s most recent case, and the prosecutor has charged six suspects with instigating violence following the 14
December 2007 election of Kenya s current president, Mwai Kibaki, which violence resulted in the deaths of more than 1000 people. Organization of the Dissertation In Chapter Two, I review the existing literature on commitment to international human rights treaties generally, and the international criminal court in particular. In this chapter, I expand on the credible threat theory, and suggest that even though prior studies have shown that states often ratify international human rights treaties without regard to their ability to comply, states making ICC ratification decisions should be guided primarily by retrospective calculations about the costs of complying with treaty terms given the strong and legally-binding nature of the ICC treaty s enforcement mechanisms. The next two chapters are quantitative tests of the credible threat theory. In Chapter Three, I empirically examine state commitment to the ICC using event history analysis. Employing a database of over 190 states for the period from 1998 to 2008, I examine the extent to which both constant and time-varying factors influence the probability that a state will ratify the ICC treaty in a given time period. As noted above, those results provide support for the credible threat theory and the idea that compliance costs influence ICC ratification behavior in that the human rights practices measure is a highly significant and positive predictor of whether or not a state will join the ICC. In Chapter Four, and to provide additional context for the findings regarding state commitment to the ICC, I quantitatively examine state commitment to the ICC in the comparative context of 13 other international human rights treaties, articles, and/or protocols. Again, as noted above, the results of event history analysis for the period between 1966 and 2008 provide additional support for the credible threat theory. I find that states with poor human 15
rights ratings regularly ratify treaties with the weakest enforcement mechanisms. Those states, however, are less likely to readily ratify the ICC treaty. Chapter Five serves as an introduction to the case study chapters and explains the logic of case study selection. More particularly, it explains the additional implications of the credible threat theory I examine through an in-depth analysis of a country s ratification behavior. For example, unlike the quantitative chapters which study the state s ratification behavior at a particular point in time, through the qualitative analyses, ratification behavior can be studied over time. Through the case studies, we can see whether state ratification decisions are influenced by any changes in domestic behavior, attitudes, or institutions. In addition, although the results of the empirical analyses indicated that cost of compliance calculations did influence ICC ratification behavior, the qualitative analyses will permit a more complete exploration of those calculations and any trade-offs states make when considering commitment. Finally, and among other things, the case studies allow us to focus on states post-ratification behavior. If states are truly concerned with the ICC s strong enforcement mechanisms and the costs of compliance, then we should see post-ratification efforts to comply. In sum, the case study chapters should allow for a better and deeper understanding of the inferences derived from the quantitative analysis since they will go beyond correlation arguments. The four chapters that follow are case studies of Germany, Trinidad and Tobago, Rwanda, and Kenya. Chapter Six examines Germany s decision to ratify the ICC in historical and political context which historical context is particularly interesting in that it suggests that Germany s ratification of the ICC should not be taken for granted. Not long ago, Germany was a country with terrible human rights practices and domestic institutions that were complicit in furthering government policies designed to abuse rather than protect human rights. During 16