Emory International Law Review

Complexity and Efficiency at International Criminal Courts
Stuart Ford Assistant Professor of Law at the John Marshall Law School in Chicago, Illinois. This Article was greatly improved by the contributions of those who commented on earlier versions, including Professors Jenia Iontcheva Turner, Daryl Lim, Morse Tan, Anne Holthoefer, Lesley Wexler, Robert Knowles, Jonathan Hafetz, and David Wippman. Thanks are also due to those who provided feedback at the various conferences where this was presented. Fang Han, Gregory Cunningham, and Iveliz Orellano provided research assistance.

Abstract

One of the most persistent criticisms of international criminal tribunals has been that they cost too much and take too long. In response, this Article presents a new approach that utilizes two concepts: complexity and efficiency. The first half of this Article proposes a method for measuring the complexity of criminal trials and then uses that method to measure the complexity of the trials conducted at the International Criminal Tribunal for the former Yugoslavia (ICTY). The results are striking. Even the least complex ICTY trial is more complex than the average criminal trial in the United States, and the most complex ICTY trials are among the most complex trials that have ever taken place. This highlights why it is misleading to compare the cost and length of the ICTY’s trials to other trials, both domestic and international, without first accounting for their complexity.

The second half of the Article explores the efficiency of international criminal trials. Efficiency is defined as the complexity of a trial divided by its cost, and the Article calculates the overall efficiency of the ICTY and then compares that to the efficiency of the Special Court for Sierra Leone (SCSL) and a sample of criminal trials in the United States. The results show that the ICTY is more efficient than the SCSL and approximately as efficient as complex murder trials in the United States. The ICTY is less efficient than a typical domestic murder trial, but this appears to be because efficiency decreases as complexity increases, making such cases poor comparators. Although the data is sparse, the ICTY appears to be much more efficient than its closest domestic comparator—mass atrocity trials. Ultimately, the ICTY has been more efficient than cases of comparable gravity and complexity tried in domestic courts or at the SCSL.

Introduction

International criminal courts are established primarily to conduct trials, 1Shahram Dana, Turning Point for International Justice?, in XI Annotated Leading Cases of International Criminal Tribunals 962, 972 (Andre Klip & Goran Sluiter eds., 2007) (“The primary function of the international criminal tribunal is to determine the criminal responsibility and punishment of those individuals found guilty of the crimes under its jurisdiction.”); Adrian Fulford, The Reflections of a Trial Judge, 22 Criminal L.F. 215, 216 (2011) (“We are first, foremost and last a criminal court: our core business is to process criminal trials. All the rest, and I hasten to add some of the rest is very important indeed (such as our deterrent potential, reparations to victims and outreach), is secondary to the Court’s obligation to investigate, arrest and try alleged criminals.”); O-Gon Kwon, The Challenge of An International Criminal Trial as Seen from the Bench, 5 Journal of International Criminal Justice 360, 373 (2007) (“The task of determining guilt or innocence must take precedence over other, not strictly judicial, considerations”); U.N. Secretary-General, 20, U.N. Doc A/C.5/52/4 (Oct. 21, 1997) (“[Chambers] performs the fundamental core activity of the Tribunal, that is, the trial and determination of guilt or innocence of persons responsible for serious violations of international humanitarian law within the territory of the former Yugoslavia”). despite being given additional goals. 2See S.C. Res. 808, U.N. Doc. S/RES/808 (Feb. 22, 1993). The ICTY was given the additional goals of putting an end to the commission of crimes in the former Yugoslavia and restoring peace in the Balkans. Id. See also Stuart Ford, A Social Psychology Model of the Perceived Legitimacy of International Criminal Courts: Implications for the Success of Transitional Justice Mechanisms, 45 Vand. J. Transnat’l L. 405, 472 (2012) (noting that the goals of international criminal courts include establishing the historical record and fostering post-conflict reconciliation). Thus the trials at the International Criminal Tribunal for the former Yugoslavia (ICTY) ought to be one of its signature accomplishments. Yet, the trials have been one of its most criticized features. The ICTY will eventually cost more than $2.7 billion 3See Stuart Ford, How Leadership in International Criminal Law Is Shifting from the United States to Europe and Asia: An Analysis of Spending on and Contributions to International Criminal Courts, 55 St. Louis U. L.J. 953, 971 (2011) (estimating the cost as $2.3 billion). The estimate of $2.3 billion was based on the assumption that the ICTY would shut down in 2014. Id. Now that it is clear the ICTY will continue to operate into at least 2017, see infra note 4, it will certainly cost more than that. My current estimate is that it will ultimately cost $2.7 billion. See infra note 178. and take more than twenty years to finish. 4The ICTY was established in 1993. See S.C. Res. 808, U.N. Doc. S/RES/808/ (Feb. 22, 1993). According to its most recent report on its Completion Strategy, the ICTY will not finish its last trials and appeals until early 2017. See President of the International Criminal Tribunal for the Former Yugoslavia, Assessment and report of Judge Theodor Meron, President of the International Criminal Tribunal for the Former Yugoslavia, provided to the Security Council pursuant to paragraph 6 of Security Council resolution 1534 (2004) and covering the period from 24 May to 18 November 2013, transmitted by letter to President of the Security Council, U.N. Doc. S/2013/678, at 39 (Nov. 18, 2013) (by Judge Theodor Meron). As a result, many writers have argued that the ICTY has been too slow, too inefficient and has cost too much. 5See Alex Whiting, In International Criminal Prosecutions, Justice Delayed Can Be Justice Delivered, 50 Harv. Int’l L.J. 323, 324 & n.1 (2009). For example, there are seventy-five law review articles that use the word “slow” in the same sentence as “ICTY.” 6This data comes from a Westlaw search of the Law Reviews & Journals database using the search “ICTY /25 slow.” The search was conducted on June 24, 2014. The vast majority of the results are articles that either accuse the ICTY of being slow or note that the ICTY has been accused of being slow. Running the same search in the News database found 80 responsive newspaper articles. Similarly, eighty-five law review articles use either “expensive” or “costly” in the same sentence as “ICTY”, 7This data comes from a Westlaw search of the Law Reviews & Journals database using the search “ICTY /25 (expensive costly).” The search was conducted on June 24, 2014. The vast majority of the results are articles that either accuse the ICTY of being expensive or note that it has been accused of being expensive. Running the same search in the News database found 17 responsive newspaper articles. and sixteen use the word “inefficient” in the same sentence as “ICTY.” 8This data comes from a Westlaw search of the Law Reviews & Journals database using the search “ICTY /25 inefficient.” The search was conducted on June 24, 2014. The vast majority of the results are articles that either accuse the ICTY of being inefficient or note that it has been accused of being inefficient. Running the same search in the News database found 10 responsive newspaper articles. Professor Whiting describes this belief that the ICTY has been too slow and cost too much as the “consensus” position among academics. 9Whiting, supra note 5, at 323–24. Some have even argued that the slow pace of the trials “call[s] into question the efficacy of international criminal justice.” 10Etelle R. Higonnet, Restructuring Hybrid Courts: Local Empowerment and National Criminal Justice Reform, 23 Ariz. J. Int’l & Comp. L. 347, 347 (2006).

Nor are academics the only ones to criticize the ICTY’s pace and cost. States and policymakers have also been deeply critical. For example, Ralph Zacklin, who was the Assistant Secretary-General for Legal Affairs at the United Nations, described a sense of “dissatisfaction which has become pervasive among the Member States” that he called “donor fatigue.” 11Ralph Zacklin, The Failings of Ad Hoc International Tribunals, 2 J. Int’l Crim. Just. 541, 543 (2004). According to him, most states eventually decided that the cost of the ICTY was not justified and that the ICTY “exemplif[ied] an approach that is no longer politically or financially viable.” 12Id. at 545. Others have also diagnosed donor fatigue among states. 13See Steven D. Roper & Lilian A. Barria, Gatekeeping Versus Allocating in Foreign Assistance: Donor Motivations and Contributions to War Crimes Tribunals, 51 J. Conflict Resol. 285, 287 (2007); Higonnet, supra note 10, at 434; Elizabeth Nielsen, Hybrid International Criminal Tribunals: Political Interference and Judicial Independence, 15 UCLA J. Int’l L. & Foreign Aff. 289, 295 (2010); Jonathan O’Donohue, Financing the International Criminal Court, 13 Int’l Crim. L. Rev. 269, 273–74 (2013); Nancy Kaymar Stafford, A Model War Crimes Court: Sierra Leone, 10 ILSA J. Int’l & Comp. L. 117, 139 (2003).

One response to this criticism has been that critics are judging the cost and length of ICTY trials by comparing them to trials in domestic court systems and that this comparison is unfair. 14See, e.g., David Wippman, The Cost of International Justice, 100 Am. J. Int’l L. 861, 861–62 (2006). Thus, a supporter of the ICTY might say: Of course its trials took longer and cost more than the average domestic trial, they are vastly more complex than the average domestic trial, and we generally expect that more complex trials will take longer and cost more. 15See id. at 873 (“In general, however, complex cases take longer to try and generate higher overall costs.”). But, while it is widely acknowledged that the cases that come before the ICTY are more complex than the typical criminal case in a domestic jurisdiction, 16See id. (“By contrast, almost all ICTY trials, because of the nature of the prosecution undertaken, entail lengthy and expensive court proceedings and pretrial investigations.”); Patricia M. Wald, ICTY Judicial Proceedings—An Appraisal from Within, 2 J. Int’l Crim. Just. 466, 468 (2004). and some articles have cataloged the complexity of individual trials, 17See, e.g., Gillian Higgins, The Impact of the Size, Scope and Scale of the Milošević Trial and the Development of Rule 73bis before the ICTY, 7 Nw. J. Int’l Hum. Rts. 239, 246 (2009) (“The procedural and substantive demands of the [Milošević] trial upon all parties were excessive.”). the complexity of international criminal trials has not been studied quantitatively before. Thus, it has been impossible to directly compare the complexity of international and domestic trials.

As a result, studying the complexity of trials at the ICTY has a big payoff. By measuring and controlling for complexity, we can more fairly compare domestic and international criminal trials. Moreover, once we measure complexity, it is only a short step to measuring the efficiency of different courts. Efficiency can be thought of as the complexity of a court’s trials divided by their cost. Thus, with a complexity measure and cost data we can calculate the relative efficiency of different courts. In short, studying complexity tells us whether the ICTY’s trials have been too slow and expensive compared to other international courts or to domestic trials, given their complexity. This study sheds new light on one of the most persistent criticisms of the ICTY—that it has taken too long and cost too much.

Assessing complexity and efficiency quantitatively could also yield significant benefits for the International Criminal Court (ICC). Every year, the members of the Assembly of States Parties must decide on the budget of the ICC. In recent years, this has become a contentious process, with heated debates about how much money the ICC really needs. 18See generally Stuart Ford, How Much Money Does the ICC Need in The law and Practice of the International Criminal Court: A Critical Account of Challenges and Achievement (Carsten Stahn ed., forthcoming 2014). See also Jonathan O’Donohue, Financing the International Criminal Court, 13 Int’l Crim. L. Rev. 269, 279–81 (2013). One of the arguments made by those states that wish to rein in ICC spending is that the court is inefficient and that it could do more with the money it has if it became more efficient. The court disputes that it is inefficient and maintains that additional funding is necessary for it to comply with its mandate. Much of the debate about the ICC’s funding depends on an assessment of its efficiency, but current assessments are subjective and potentially unreliable. Thus, developing a quantitative measure of the complexity and efficiency of the court’s work could significantly improve the budgeting process at the ICC.

Of course, to assess complexity quantitatively, one needs to be able to measure it, and this Article proposes a measure of complexity that permits complexity to be measured and compared across international and domestic court systems. So, how complex are the ICTY’s trials? As far as I can tell, they are the most complex set of related criminal cases that has ever been tried by any court anywhere. They dwarf the complexity of domestic criminal prosecutions, and not even the prosecution of the senior Nazi leaders by the International Military Tribunal (IMT) can compare with the most complex ICTY cases. Collectively, the forty-one full trials completed by the ICTY so far are more than seventeen times as complex as the entire IMT.

This Article’s most important conclusion is that the ICTY’s trials have been efficient given their complexity. The ICTY has been more efficient than the Special Court for Sierra Leone (SCSL), despite the fact that the SCSL was specifically designed to be more efficient than the ICTY. It is less efficient than run of the mill domestic murder trials, but there are good reasons to believe that ordinary domestic murder trials are not a fair comparator. It has, however, been roughly as efficient as complex murder trials, like those of O.J. Simpson and Scott Peterson. Most importantly, the ICTY has been more efficient than comparable domestic mass atrocity trials, like the trial of Timothy McVeigh for the bombing of the federal building in Oklahoma City. The conventional wisdom—that the ICTY costs too much and takes too long—is simply not accurate. This Article shows that the ICTY has been more efficient than domestic courts would be, if faced with cases of comparable complexity and gravity.

I. About Complexity

The concept of complexity is central to this Article and requires some explanation. This Part discusses a number of questions about complexity, including why to use it, how to define it, and the limitations of a complexity-based approach. These theoretical discussions lay the groundwork for the rest of the Article.

A. The Utility of Complexity

It is common to allege that the ICTY is slow, expensive and inefficient. 19See supra text accompanying notes 5–13. But the obvious question is: Slow compared to what? Expensive compared to what? For critics, the answer appears to be slow and expensive compared to most domestic criminal trials. And while this is certainly true, a number of people have noted that comparison is unfair because the kinds of cases that come before domestic and international courts are very different. Trials at international criminal tribunals are much more complex than the average domestic criminal prosecution.

If we can measure and account for the differences in complexity between the two systems, we can make a much fairer comparison. The right comparison is not between the average domestic criminal trial and the average ICTY trial, but between trials of the same complexity in both systems. If the ICTY trial is still much slower and more expensive than a domestic trial of the same complexity, then it may be that the ICTY is too slow, too expensive and inefficient. At the least, it suggests that it would be cheaper and quicker to prosecute violations of international criminal law in domestic courts. On the other hand, if the ICTY trial costs less than a domestic trial of the same complexity, then the numbers suggests that the ICTY is not too slow, expensive or inefficient. Rather, the ICTY is doing better than a domestic court would do if it were faced with such a case. In effect, by measuring and controlling for complexity, we may be able to draw conclusions about the efficiency of the ICTY, with efficiency measured by the cost of trying a case of a given complexity.

This is not a perfect comparison because domestic criminal trials and international criminal trials serve slightly different purposes even though they engage in similar activities. Retribution is a key purpose of both kinds of trials, but international criminal trials have many purposes (post-conflict reconciliation, ending impunity, setting the historical record, ending conflict, etc.) that have no real analog in most domestic criminal trials. These additional purposes presumably have value and this suggests that we might be willing to pay more for international criminal trials, even after adjusting for complexity, than we would for domestic trials. So, even a complexity-adjusted comparison of domestic and international trials is not a perfect comparison. It is, nevertheless, a useful comparison.

Measuring complexity is not, of course, the solution to all of the difficult questions in international criminal law. It cannot, for example, tell us whether the level of complexity in a given trial is necessary or desirable. It is undisputed that the ICTY’s trials are exceedingly complex. Two trials have been more complex than the trials of the Nazi leaders at the IMT. 20See infra notes 159–60. But is this level of complexity required? The methodology described in this Article cannot answer that question. I will, however, return to it later, as it is an important question. 21See infra Conclusion.

Another thing we cannot do with complexity is decide whether we ought to prosecute violations of international criminal law at all. This is an important point as not all criticisms of the ICTY have focused on the question of its efficiency. Some have argued against the creation of international criminal courts for purely political reasons. 22See, e.g., Michael P. Scharf, The Politics Behind U.S. Opposition to the International Criminal Court, 6 Brown J. World Aff. 97, 98 (1999) (describing opposition to the creation of the International Criminal Court in the United States based on a fear that it would investigate or prosecute American citizens). Others have questioned whether international criminal courts can achieve their purposes, 23See Mirjan Damaska, What is the Point of International Criminal Justice?, 83 Chi.-Kent L. Rev. 329, 331–40 (2008) (questioning whether international courts can achieve their goals); David Wippman, Atrocities, Deterrence and the Limits of International Justice, 23 Fordham Int’l L.J. 473, 475 (1999) (questioning whether international courts can achieve deterrence). But see Payam Akhavan, Beyond Impunity: Can International Criminal Justice Prevent Future Atrocities?, 95 Am. J. Int’l L. 7 (2001) (arguing that prosecution of serious violations of international criminal law can have positive effects). which implicitly raises doubts about whether funding them is worthwhile, irrespective of the efficiency with which they carry out trials. Showing that the trials at international courts are no more expensive or time consuming than domestic trials given their complexity does not directly respond to these questions.

But the reality is that the international community is deeply committed to international criminal justice. This can be seen most visibly at the International Criminal Court (ICC), which currently has 122 members. Membership in the ICC represents a commitment to a permanent venue for the prosecution of serious violations of international criminal law, and the majority of states in the world have undertaken that commitment. In that sense, the question of whether we ought to prosecute violations is less relevant to most states than the questions of how long it will take and how much it will cost. Indeed, questions about the pace and cost of international criminal justice have taken center stage at the ICC, where recent meetings of the Assembly of States Parties have been dominated by debates about the ICC’s budget. 24See supra note 18. A better understanding of complexity can help states understand how much funding the ICC needs to complete its trials and how long those trials can be expected to last. This may not directly answer the question of whether international criminal justice is worthwhile, but it is valuable information.

The ideal way to justify spending on international criminal courts would be to directly measure and assign a value to the outcomes of international criminal justice and compare that to the cost of international criminal justice. Such an undertaking would resemble a cost-benefit analysis. 25See Ford, supra note 3, at 954–55. This analysis would allow us to directly measure the value of the ICTY rather than just comparing its cost to criminal trials in domestic systems, which we assume have roughly equivalent value. Professor Shany recently advocated a similar approach to measuring the effectiveness of international courts. 26See Yuval Shany, Assessing the Effectiveness of International Courts: A Goal-Based Approach, 106 Am. J. Int’l L. 225, 230 (2012). He stressed that courts should be evaluated by their ability to accomplish their goals, 27Id. at 230–31. and that their cost-effectiveness should also be measured by considering the resources expended in pursuit of those goals. 28Id. at 237–38. He uses the language of sociology, but the result would be quite similar to a cost-benefit analysis.

A cost-benefit analysis of international criminal justice would be exceedingly difficult, although probably not impossible in the long run. First, we would have to identify the purposes or goals of international criminal justice. Luckily, there is extensive literature on this subject. 29See Damaska, supra note 23, at 331; Minna Schrag, Lessons Learned from ICTY Experience, 2 J. Int’l Crim. Just. 427, 428 (2004); Jenia Iontcheva Turner, Defense Perspectives on Law and Politics in International Criminal Trials, 48 Va. J. Int’l L. 529, 537–43 (2008). The next step, however, is much harder. We would need to measure whether and how much courts’ outputs contribute to achieving their purposes. It may be that with careful thought and data collection it would be possible to measure the real world impact of international criminal courts, but it would be extremely difficult and time-consuming. 30Shany, supra note 26, at 239 (“The goals of public organizations, such as courts, tend to be ambiguous, and the public goods they generate, such as justice, peace, and legal certainty, are hard to quantify”). See also id. at 248–49 (“The key to assessing the effectiveness of international courts according to the rational-system or goal-based approach involves evaluation of judicial outcomes”). Without a way to directly measure a court’s ability to achieve its goals, there is no way to determine the absolute value of its work. Thus, studying complexity may not be the ideal way to assess the value of international criminal justice because it provides only a relative measure of value, but it is the best way that is currently possible.

B. Defining Complexity

Complexity in the law has been addressed extensively by Professors Tidmarsh, 31Jay Tidmarsh, Unattainable Justice: The Form of Complex Litigation and the Limits of Judicial Power, 60 Geo. Wash. L. Rev. 1684 (1992). Stempel, 32Jeffrey W. Stempel, A More Complete Look at Complexity, 40 Ariz. L. Rev. 781 (1998). and others. 33There are numerous scholarly articles that deal in some manner with the concept of complexity in the law. Some examples include Stephen B. Burbank, The Costs of Complexity, 85 Mich. L. Rev. 1463 (1987); Michael Heise, Criminal Case Complexity: An Empirical Perspective, 1. J. Empirical Legal Stud. 331 (2004); Ronit Dinovitzer & Jeffrey S. Leon, When Long Becomes too Long: Legal Culture and Litigators’ Views on Long Trials, 19 Windsor Y.B. Access Just. 106 (2001); Richard O. Lempert, Civil Juries and Complex Cases: Let’s Not Rush to Judgment, 80 Mich L. Rev. 68 (1982); Larry Heuer & Steven Penrod, Trial Complexity: A Field Investigation of Its Meaning and Effects, 18 Law & Hum. Behav. 29 (1994); Peter H. Schuck, Legal Complexity: Some Causes, Consequences and Cures, 42 Duke L.J. 1 (1992); Theodore Eisenberg, Paula L. Hannaford-Agor, Valerie P. Hans, & Nicole L. Waters, Judge-Jury Agreement in Criminal Cases: A Partial Replication of Kalven and Zeisel’s The American Jury, 2 J. Empirical Legal Stud. 171 (2005); Matthew A Reiber & Jill D Weinberg, The Complexity of Complexity: An Empirical Study of Juror Competence in Civil Cases, 78 U. Cin. L. Rev. 929 (2010). Much of the existing literature, however, has to do with civil trials 34Some studies have covered both civil and criminal trials. See Heuer & Penrod, supra note 33. Very few have exclusively studied exclusively criminal trials. The exception is Heise, supra note 33. Most of the literature cited below relates largely or exclusively to civil trials. in the United States 35One of the exceptions to the focus on the United States is the study by Dinovitzer and Leon, which examined civil trials in Canada. See Dinovitzer & Leon, supra note 33. and may not be appropriate for a study of international criminal trials. Moreover, agreement on a definition of complexity specific to the law has been elusive. 36Burbank, supra note 33, at 1463 (“Not even those charged with responsibility to devise procedures for complex cases in the federal courts have essayed a definition worth of the name.”); Heise, supra note 33, at 366–67 (noting that there is no accepted definition of complexity and that different groups may define it differently); Eisenberg et al., supra note 33, at 190 (noting that different people probably view complexity differently); Dinovitzer & Leon, supra note 33, at 117 (noting that courts have difficulty determining when to categorize cases as complex); Heuer & Penrod, supra note 33, at 30–31 (noting that there have been many proposed definitions of complexity but little agreement); Schuck, supra note 33, at 2–3 (noting that “legal complexity is hard to define” and that different people might conceive of it and define it in different ways); Stempel, supra note 32, at 785–86; Reiber & Weinberg, supra note 33, at 945 n.41 (noting that there have been attempts to create an all-encompassing definition of complexity, but that such attempts have not generally been successful). Professor Tidmarsh’s article highlights the difficulty of trying to reach an agreement on a definition of complexity through its extensive catalog of definitions that have been previously suggested, which covers more than 40 pages. See Tidmarsh, supra note 31, at 1692–1734. Professor Stempel explicitly declines to provide a definition of the term, choosing to focus instead on the factors that cause complexity. 37See Stempel, supra note 32, at 785–86 (noting that “a single definition of complexity is both unnecessary and perhaps misleading”). Professor Tidmarsh, on the other hand, argues that it is impossible to properly assess whether any given case is complex without a definition of complexity, 38See Tidmarsh, supra note 31, at 1690 (“a correct understanding of complex litigation might prevent well-intentioned reform from paving the path to a procedural hell”). and criticizes those who have tried to explain complexity without offering a definition. 39Id. at 1694–98.

Professor Tidmarsh’s attempt to define complexity is the most comprehensive, but it is a definition of complex litigation in civil trials in the United States and it is inextricably connected to assumptions about civil trials in the U.S. 40Id. at 1743–50 (basing his definition of complexity on seven assumptions about the nature of civil procedure in the United States). As such, it is not directly transferable to the context of the ICTY. Moreover, Professor Tidmarsh tried to distinguish between “routine” civil cases that could proceed under normal rules and “complex litigation” that would need special procedural rules, 41Id. at 1689–90. and his definition was tailored to that purpose. This Article does not treat complexity as a threshold between the routine and the complex. 42See infra notes 104–108. For these reasons, this Article will not adopt Professor Tidmarsh’s definition.

In contrast to the search for a definition of complexity among legal academics, participants in empirical studies of complexity are usually asked to measure it without being offered a definition. For example, in the Heise study, participants were simply asked to rate the complexity of a trial on a scale that ranged from “not at all complex” to “very complex.” 43Heise, supra note 33, at 346. In the Heuer and Penrod study, judges were asked to rate how complex a particular trial was compared to the average trial they heard. 44Heuer & Penrod, supra note 33, at 35. The assumption in the empirical literature is that people can adequately identify and measure complexity without a definition even as that literature acknowledges that different people might define and therefore report complexity differently. 45Heise, supra note 33, at 344–45, 350; Eisenberg et al., supra note 33, at 190. This suggests that we could proceed without a definition. It seems useful, however, to have a working definition of complexity.

A plain language definition will be used because it will help us to understand what most people mean when they say that a trial is “complex.” Complexity is usually defined by dictionaries as “[t]he quality or condition of being complex.” 46The Oxford English Dictionary 613 (J.A. Simpson & E.S.C. Weiner eds., Oxford University Press 2d ed. 1989). Other dictionaries define it similarly. See New Oxford American Dictionary (Angus Stevenson & Christine A. Lindberg eds., Oxford University Press 3d ed. 2010) (“the state or quality of being intricate or complicated”); The American Heritage Dictionary of the English Language 377–78 (Houghton Mifflin Co. 4th ed. 2000) (“the quality or condition of being complex”); Random House Webster’s College Dictionary 271 (2d ed. 1999) (“the state or quality of being complex”). If we look at definitions of “complex,” there are two distinct ideas that appear in most definitions. One idea is that multiple parts of a whole are related to one another in such a way that they can interact. Thus one dictionary defines complex as “consisting of or comprehending various parts united or connected together.” 47The Oxford English Dictionary, supra note 46, at 613. Other dictionaries offer similar definitions. 48See New Oxford American Dictionary, supra note 46 (“consisting of many different and connected parts”); The American Heritage Dictionary of the English Language, supra note 46 (“consisting of interconnected or interwoven parts”); Random House Webster’s College Dictionary, supra note 46 (“composed of many interconnected parts; compound; composite”). A second strand of the definition has to do with the difficulty of analyzing or understanding something complex. Thus one dictionary defines complex as being “not easily analysed or disentangled.” 49The Oxford English Dictionary, supra note 46, at 613. Other definitions also pick up on this aspect of the word. 50See New Oxford American Dictionary, supra note 46, at 271 (“not easy to analyze or understand; complicated or intricate”); Random House Webster’s College Dictionary, supra note 46 (“so complicated or intricate as to be hard to understand or deal with”). Combining these two strands, something is complex when it is composed of many interconnected parts such that their interaction makes the whole difficult to understand or analyze. Complexity then becomes the state or condition of having these attributes. Thus, for purposes of this Article, complexity occurs when multiple parts of a system interconnect and interact in such a way that the whole system becomes hard to understand or analyze.

Despite having rejected the definitions of complexity in the existing academic literature, there is much of value in that literature. For example, most legal scholars agree that there are different types of commonly occurring complexity. 51See Heise, supra note 33, at 358 (noting that judges in the study were asked to rate evidentiary and legal complexity separately and that the separate scores were then combined into an overall measure of trial complexity); Heuer & Penrod, supra note 33, at 35 (noting that judges in the study were asked to rate both evidentiary complexity and legal complexity); Tidmarsh, supra note 31, at 1702–12 (describing various types of complexity, including legal complexity, factual complexity, remedy complexity, and party complexity); Stempel, supra note 32, at 789–96 (discussing legal and factual complexity). The main types identified in the literature are: (1) legal complexity; (2) factual complexity; and (3) participant complexity. Some scholars identify other kinds of complexity, 52For example, Reiber and Weinberg treat “procedural complexity” as a separate form of complexity, although it looks most like what this Article calls legal complexity. Reiber and Weinberg, supra note 33, at 947. Tidmarsh identifies “remedy complexity” and “party complexity” as additional forms of complexity in civil trials. Tidmarsh, supra note 31, at 1707–12. These forms of complexity have less relevance in the context of criminal proceedings. but the three types listed above feature in the majority of complexity scholarship and are most applicable to a discussion of the ICTY.

Legal complexity arises from the law itself. Obviously, to reach a legal decision, a court must first identify the content of the law and then apply it to the facts of the case. 53Cf. id. at 1736 (arguing that legal adjudication requires “reasoned judgment, which is based on the consideration of the disputants’ circumstances, the relevant evidence, and the nature of the claimed obligation”). This process is sometimes complex, as the law in a particular field may be dense (i.e., the field is heavily regulated and there are many different overlapping legal rules to consider), difficult to ascertain (e.g., there are many different sources of the appropriate law, which makes it more difficult to decide which sources apply), extremely technical (i.e., they require technical expertise or sophistication to understand and apply), or uncertain (e.g., the law is standards rather than rules based, depends on multi-factor tests or balancing tests, is heavily fact dependent, or there are a multitude of conflicting prior precedents to analyze). 54Schuck, supra note 33, at 3–4; Stempel, supra note 32, at 789–92 (noting that legal complexity may stem from the technical nature of the law, the difficulty in ascertaining the law, or the indeterminacy of the law).

Factual complexity arises from the facts or information needed to reach a legal decision. Legal decisions require not just an understanding of the law, but an understanding of the facts and how the law should be applied to the facts to reach a legal decision. As a result, complexity can arise from the nature of the facts necessary to reach a decision. 55Heuer & Penrod, supra note 33, at 35–36 (noting that both the quantity of the evidence and the difficulty of the evidence caused perceived trial complexity to increase); Heise, supra note 33, at 352–54 (noting that the difficulty of the evidence increased perceived complexity of the trial). This can result in complexity if there is a need for a large numbers of facts, the facts are very technical, the facts are difficult to determine, or the facts are simply indeterminate. 56See Reiber & Weinberg, supra note 33, at 950–51 (testing a model in which factual complexity was represented by the need to reconcile the evidence of multiple witnesses who gave conflicting testimony and draw inferences from circumstantial evidence); Stempel, supra note 32, at 794–95; Tidmarsh, supra note 31, at 1766. There is some evidence that this form of complexity is the most important for determining overall complexity. 57See Heuer & Penrod, supra note 33, at 36 tbl.2 (reporting that the quantity of evidence and the complexity of the evidence had more impact on the reported complexity of the cases than the complexity of the law). But see Reiber & Weinberg, supra note 33, at 952–54 (suggesting that legal complexity makes juror decision-making more difficult than factual complexity). Of course, the ICTY does not use juries, so the Reiber and Weinberg finding may not be applicable at the ICTY.

The third kind of complexity is participant complexity. Legal decisions are made by people, not machines. Thus, the people who participate in the legal process, including the parties, judges, juries, lawyers, and witnesses, have an effect on that process, including its complexity. 58See Tidmarsh, supra note 31, at 1755–80; Dinovitzer & Leon, supra note 33, at 112–14 (noting studies that have indicated that the local legal culture of judges and lawyers affects the length of time it takes to decide a case); id. at 138 (noting that the vast majority of lawyers in the study believed that lawyer skill, knowledge and preparation would affect trial length). The lawyers for the parties may play a deliberate role that affects the overall complexity of the decision-making process. For example, some studies of criminal cases suggest that skilled prosecutors decrease perceived complexity, while skilled defense counsel increase perceived complexity. 59See Heise, supra note 33, at 355–56; cf. Heuer & Penrod, supra note 33, at 41–42. But even when individuals are not deliberately assuming roles that affect complexity, they may still have an effect on it. For example, an unskilled or overwhelmed participant might make a trial more complex because of their inability to fulfill their role in the process. 60See Stempel, supra note 32, at 796; Tidmarsh, supra note 31, at 1757–59. Indeed, there is some evidence that self-represented accused increase trial complexity, perhaps because they are unable or unwilling to fulfill their role. 61See Higgins, supra note 17, at 245, 248, 250; U.N. Secretary-General, Budget for the International Tribunal for the Prosecution of Persons Responsible for Serious Violations of International Humanitarian Law Committed in the Territory of the Former Yugoslavia since 1991, for the biennium 2012–2013, 55, U.N. Doc. A/66/386 (Sept. 29, 2011). Another concern is that humans’ inherent cognitive limitations 62Despite our beliefs otherwise, we are not fundamentally rational decision makers. Our decision-making is affected by both motivated reasoning and cognitive biases. See Ford, supra note 2, at 419–39. may increase perceived complexity by making legal decision-making more difficult. 63Schuck, supra note 33, at 25 (noting research in social psychology that shows that people have difficulty with truly complex decisions and tend to radically simplify their decision-making through the use of heuristics when confronted with complexity); Stempel, supra note 32, at 795–96.

Another insight from the legal literature is that the different forms of complexity arise from different causes and therefore can change independently of one another. Legal complexity can make decision-making difficult even in situations where the facts are straightforward and vice versa. 64Cf. Tidmarsh, supra note 31, at 1703 (noting that cases with relatively simple facts can generate “wonderfully difficult legal issues”). Participant complexity can move independently of factual or legal complexity depending on the role, abilities, and motivations of the participants. 65See supra notes 59–63. Thus, the different forms of complexity may not all move in lockstep. At the same time, in practice, the different forms of complexity will often be correlated. 66Heise, supra note 33, at 359 tbl.8 (noting that evidentiary and legal complexity were often related and cannot be assumed to be independent); Reiber & Weinberg, supra note 33, at 947 n.48 (noting that procedural and factual complexity may overlap with one another). Moreover, the most complex cases tend to be those that exhibit high legal, factual, and participant complexity. 67Tidmarsh, supra note 31 at 1702–03 (suggesting that cases are not truly complex unless they involve both factual and legal complexity).

Finally, a number of writers have noted that one form of complexity can sometimes cause or exacerbate other forms of complexity. For example, when both the law and the facts are complex, the total complexity of the resulting decision may be greater than the complexity of the individual components. In other words, legal complexity and factual complexity may interact in ways that generate additional complexity. 68Cf. id. at 1702–03 (suggesting that the most complex cases arise when there is a combination of factual and legal complexity). High legal or factual complexity may also cause participant complexity when it overwhelms the abilities of the participants to perform their roles. 69Stempel, supra note 32, at 796; Tidmarsh, supra note 31, at 1757–59. As a result, total complexity will sometimes be greater than the sum of the individual types of complexity, particularly in cases that exhibit high levels of factual and legal complexity.

C. Are ICTY Trials Complex?

Using the definition of complexity described above, 70See supra text accompanying notes 46–50. it is obvious that trials at international criminal courts exhibit a high degree of complexity. First, they are marked by many interconnected parts that together form the whole. Criminal trials are ultimately formed out of, inter alia, the interaction of the charges, the substantive law, the procedural rules, the evidence put before the court, and the personalities, motivations, and skills of the participants, including the prosecutors, defense lawyers, witnesses, judges, and the accused. Each of these parts interacts with the other parts to produce the whole that we call a trial, and there are many moving parts. While this is true at some level of all criminal trials, trials at international criminal courts tend to have many more moving parts than trials in most domestic criminal systems. For example, they include a larger number of charges, more individual crime sites, and a larger number of alleged victims. 71See Stuart Ford, Fairness and Politics at the ICTY: Evidence from the Indictments, 39 N.C. J. Int’l L. & Com. Reg. 45, 75–77, 84–85, 87 (2013). They also have larger numbers of witnesses and exhibits. 72See infra Table 3. The larger number of moving parts suggests a higher level of complexity than is present in most domestic trials.

Second, the parts interact with each other in ways that have made trials at international criminal courts extremely difficult to analyze. This is exemplified by the difficulty that courts have had predicting how long the trials will take and how much they will cost. For example, the ICTY initially predicted that all work, including appeals, could be completed by 2008. 73See Rep. of the ICTY, 57th Sess., Sept. 4, 2002, 6, U.N. Doc. A/57/379-S/2002/985 (Aug. 14, 2002). By 2004, the court was predicting that it would complete all of the first trials during 2008. 74See President of the International Tribunal for the Former Yugoslavia, Assessments and report of Judge Theodor Meron, President of the International Criminal Tribunal for the Former Yugoslavia, provided to the Security Council pursuant to paragraph 6 of Security Council resolution 1534 (2004), transmitted by letter to President of the Security Council, 3, 47–52, U.N. Doc. S/2004/420 (May 21, 2004) (by Judge Theodor Meron). Two years later, it was predicting that first trials would continue into 2009. 75See President of the International Tribunal for the Former Yugoslavia, Assessment and report of Judge Fausto Pocar, President of the International Tribunal for the Former Yugoslavia, provided to the Security Council pursuant to paragraph 6 of Council resolution 1534 (2004), 30, U.N. Doc. S/2006/898 (Nov. 16, 2006) (by Judge Fausto Pocar). Fast forward another two years, and the court was predicting that first trials would be finished in 2010. 76See President of the International Tribunal for the Former Yugoslavia, Assessment and report of Judge Patrick Robinson, President of the International Tribunal for the Former Yugoslavia, provided to the Security Council pursuant to paragraph 6 of Council resolution 1534 (2004), 42, U.N. Doc. S/2008/729 (Nov. 24, 2008) (by Judge Patrick Robinson). By 2010, the court was predicting that trials would finish in 2013. 77See President of the International Tribunal for the Former Yugoslavia, Assessment and report of Judge Patrick Robinson, President of the International Tribunal for the Former Yugoslavia, provided to the Security Council pursuant to paragraph 6 of Security Council resolution 1534 (2004), covering the period from 15 May 2010 to 15 November 2010, 4, U.N. Doc. S/2010/588 (Nov. 19, 2010) (by Judge Patrick Robinson). In 2013, the ICTY acknowledged that three trials would stretch into 2014 or beyond. 78See President of the International Tribunal for the Former Yugoslavia, Assessment and report of Judge Theodor Meron, President of the International Tribunal for the Former Yugoslavia, provided to the Security Council pursuant to paragraph 6 of Security Council resolution 1534 (2004), and covering the period from 16 November 2012 to 23 May 2013, 3, U.N. Doc. S/2013/308 (May 23, 2013). Now it seems likely that the court will be hearing appeals until 2017. 79See S.C. Res. 808, supra note 4; Letter from the President of the ICTY, supra note 4.

The court has even admitted that it cannot make accurate predictions: “[t]he estimation of the length of trial proceedings is more an art than a science, and the assessments that are always made prior to the commencement of a trial are by their very nature approximations.” 80See President of the International Tribunal for the Former Yugoslavia, Assessment and report of Judge Patrick Robinson, President of the International Tribunal for the Former Yugoslavia, provided to the Security Council pursuant to paragraph 6 of Security Council resolution 1534 (2004), covering the period from 15 May 2010 to 15 November 2010, 6, U.N. Doc. S/2010/588 (Nov. 15, 2010) (by Judge Patrick Robinson). Unsurprisingly, it attributes the difficulty of predicting the length of the trials to their complexity. 81Id. (“This is the nature of trials, particularly trials of the complexity that are heard at the Tribunal.”). Thus, trials at the ICTY exhibit not just a large number of moving parts, but interactions among these parts that make predicting or analyzing them difficult. In short, they are complex.

Trials at the ICTY also exhibit a relatively high degree of each form of complexity. Trials at the ICTY tend to be legally complex. 82See Kwon, supra note 1, at 364 (concluding that the elements of the crimes and modes of liability are “far more complex” in international criminal law than in domestic criminal law). For one thing, international criminal law is a specialized and technical area of the law, and those who practice it tend to be specialists. This stems in part from the fact that it has multiple sources, including treaties, 83Robert Cryer et al., An Introduction to International Criminal Law and Procedure 9–12 (2d ed. 2010) (discussing the sources of international criminal law). customary international law, 84Crimes against humanity are largely rooted in customary international law. See Stuart Ford, Crimes Against Humanity at the Extraordinary Chambers in the Courts of Cambodia: Is a Connection with Armed Conflict Required?, 24 UCLA Pac. Basin L.J. 125, 129–80 (2007) (chronicling the development of crimes against humanity over time). and the judicial decisions of a number of different courts. 85Cryer et al., supra note 83, at 109–80 (describing the various international courts). Moreover, these rules often overlap and the same act can be more than one crime, depending on how it is analyzed. For example, the distinction between murder as a crime against humanity and willful killing as a war crime depends on a careful evaluation of the contextual elements rather than any difference in the act of killing. 86Id. at 246.

The rules themselves are rarely bright-line rules, and they often incorporate something akin to a reasonableness standard. 87See Stuart Ford, Is the Failure to Respond Appropriately to a Natural Disaster a Crime Against Humanity? The Responsibility to Protect and Individual Responsibility in the Aftermath of Cyclone Nargis, 38 Denver J. Int’l Pol’y 227, 242, 245–46 (2010) (discussing the intent requirement in international criminal law). They are also highly fact dependent, and circumstantial evidence plays a large role. 88Robert Petit et al., Exploring Critical Issues in Religious Genocide: Case Studies of Violence in Tibet, Iraq and Gujarat, 40 Case W. Res. J. Int’l L. 163, 203–09 (2008) (discussing the use of circumstantial evidence to prove genocide). Moreover, the state of the law is often indeterminate. This is particularly true where the law is derived from international custom, as there will be no definitive writing that embodies the rule. Even when the overarching rule is treaty-based and thus presumably easier to determine, there is still often a large number of potentially inconsistent precedents to consider. 89Id. at 190–94 (discussing the various inconsistent and contradictory sources relating to the meaning of the word “destroy” in the crime of genocide). In short, international criminal trials demonstrate significant legal complexity.

International criminal trials also tend to be factually complex, at least compared to most domestic criminal cases. 90See Wippman, supra note 14, at 875 (“Relative to most national criminal court trials, ICTY trials are inherently complex.”); Kwon, supra note 1, at 364. The crimes themselves are often undertaken by hierarchical groups working together, rather than individuals, and proof of them usually requires detailed proof of the actions of individuals other than the accused. 91See Ford, supra note 71, at 67–68. This increases the factual complexity of all but the simplest ICTY cases compared to the majority of criminal cases in domestic jurisdictions, which usually involve individuals acting alone. 92Id. They often also take place across multiple locations and sometimes cover multiple countries or long periods of time. 93See id. at 85–87. For example, the trial of Slobodan Milošević covered acts that took place in three different countries at more than 400 crime sites. Id. at 87. See also Damaska, supra note 23, at 340–41 (describing the complexity of the Milošević case further). Some of the facts are also highly technical, particularly those relating to mass grave exhumations and the conduct of organized militaries, and require the use of expert witnesses to help the decision maker understand them. Finally, the facts themselves are often indeterminate. 94See generally Nancy A. Combs, Fact-Finding Without Facts: The Uncertain Evidentiary Foundations of International Criminal Convictions (2010) (documenting difficulties in fact-finding in international trials).

There is also evidence of participant complexity at the ICTY. For one thing, the ICTY has had a large number of multi-accused trials. 95 See infra Figure 1. Cases that include more than one defendant have more participants 96The additional participants include not just the extra accused, but the additional lawyers that represent those accused. than single accused trials, which are the norm in domestic criminal justice systems. 97See infra note 133. It seems likely that participant complexity increases as the number of participants increases. In addition, it appears that cases are more complex than they would otherwise be when accused proceed without legal counsel, 98See supra note 61. and several ICTY cases have included self-represented accused who appeared to go out of their way to increase the trial’s complexity. 99The most famous of these cases involved Slobodan Milošević and Vojislav Šešelj. Milošević and Šešelj both used their trials as a platform to attack the court, obstruct the proceedings, and try to influence public opinion in Serbia. See Turner, supra note 29, at 573–74. The high legal and factual complexity of the cases may also breed additional participant complexity as the participants reach their cognitive limitations. 100See supra notes 62–63.

D. Measuring Complexity

There are two approaches one can take to measure the complexity of trials. The approach taken by most empirical pieces is to measure complexity subjectively. 101See Eisenberg et al., supra note 33, at 175; Heise, supra note 33, at 343; Heuer & Penrod, supra note 34, at 34. Usually this is done by asking the participants or observers to rank the complexity of a trial on a scale. 102See supra notes 43–45. This makes sense as complexity is a subjective factor. Unlike physical attributes like mass, circumference, or height, complexity cannot be directly measured. It exists in the mind of the individual watching or participating in a trial and occurs when the interactions among the parts of the system make it hard for the individual to understand the proceedings.

The subjective method is not available here. So far, the Trial Chambers have sat for more than 7200 days. This results in more than 20 years’ worth of trial transcripts. It would be too expensive and time consuming to have a group of people read the trial transcripts and rate the complexity of the trials. Instead, this Article depends on the availability of an objective measure that can act as a proxy 103A proxy variable is “[a] measurable variable that is used in place of a variable that cannot be measured.” Graham Upton & Ian Cook, A Dictionary of Statistics 315 (2d ed. 2008). for complexity and that can be assessed from information that is publicly available. Moreover, I will try to use factors that are common to a broad range of legal systems, so that complexity can be measured and compared across different systems.

Another consideration is whether to treat complexity as a dichotomous or continuous variable. 104A dichotomous variable is a categorical variable with only two categories. Id. at 114. A continuous variable is one whose set of possible values is a continuous interval of real numbers. Id. at 88. Much of the literature about complexity in civil cases in the United States is concerned with whether any given case should be described as either complex or routine. This distinction matters in U.S. federal courts because cases that are labeled as complex can be subject to different rules than routine cases. 105See Fed. R. Civ. P. 16(c)(2)(L) (noting that cases that are identified as “complex” may be subject to “special procedures”). See also Manual for Complex Litigation (4th ed. 2004) (providing guidance to federal judges facing “complex” cases). However, it is obvious from Professors Stempel’s and Tidmarsh’s arguments that complexity is not a binary matter. 106See generally Stempel, supra note 32 (assessing criteria used to measure complexity); Tidmarsh, supra note 31 (developing a more formal and inclusive definition of complexity). Rather, complexity is a continuum. 107See Heuer & Penrod, supra note 33, at 33; Schuck, supra note 33, at 5. Some cases will be more complex, some will be less complex. While all cases at the ICTY are more complex than the average case in the United States, there is still a great deal of variation between the most complex and the least complex ICTY trials. 108See infra Part II.C. Therefore, complexity will be treated as a continuous variable.

With these considerations in mind, this Article measures trial complexity using a proxy variable that is a composite of: (1) the number of trial days the case takes to be heard; (2) the number of witnesses that testify; and (3) the number of exhibits that are entered into the record. There are a number of reasons for focusing on these three variables to create a proxy for trial complexity. First, in my personal experience, 109The author spent three years working in the Office of the Co-Prosecutors at the ECCC as an Assistant Prosecutor. He assisted in investigating and drafting the indictments and served on the trial team in the case of Kaing Keuk Eav alias Duch. Prior to working at the ECCC, he spent more than five years litigating complex commercial matters before federal district courts in the United States. these are the best outward indicators of the complexity of a trial. Second, experts on international criminal trials tend to describe trial complexity using these metrics. For example, attorneys who write about complex ICTY cases tend to describe them in terms of the number of trial days, witnesses and documents that are necessary to reach a decision. 110See Damaska, supra note 23, at 341; Higgins, supra note 17, at 246 (describing the complexity of the Milošević trial in terms of hearing days, witnesses, and trial exhibits). The Registry uses similar factors to determine the expected complexity of a case. 111See Defence Counsel–Pre-Trial Legal Aid Policy, U.N. Int’l Crim. Trib. For the Former Yugoslavia, 22 (May 1, 2006), available at http://www.icty.org/x/file/About/Defence/pretrial_payment_2006_en.doc (last visited Oct. 9, 2012) (using the number and type of witnesses and documents that are expected to be used at trial to determine the expected complexity of the case). Moreover, when international judges describe the cases they have decided, they often provide statistics about the number of trial days, trial witnesses, and trial exhibits, suggesting that they view these as indicators of complexity. 112See e.g., Prosecutor v. Krstic, Case No. IT-98-33-T, Judgment, 4 (Aug. 2, 2001) (“In all, the Trial Chamber heard more than 110 witnesses over 98 days of trial and viewed in the vicinity of 1,000 exhibits.”). See also Prosecutor v. Popović et al., Case No. IT-05-88-T, Judgment, Annex II, 1, 15–16, 19–26 (June 10, 2010); Prosecutor v. Thomas Lubanga Dyilo, Case No. ICC-01/04-01/06, Judgment pursuant to Article 74 of the Statute, 11 (Mar. 14, 2012); Prosecutor v. Charles Ghankay Taylor, Case No. SCSL-03-1-T, Judgment Summary, 8, (April 26, 2012); Judgment of the International Military Tribunal for the Trial of the German Major War Criminals, Introduction, (Sept. 30, 1946); Prosecutor v. Bagosora et al., Case No. ICTR-98-41-T, Judgment and Sentence 1 n.1 (Dec. 18, 2008). Sometimes the judges even make this connection explicitly. 113See, e.g., Prosecutor v. Jadranko Prlić, Bruno Stojić, Slobodan Praljak, Milivoj Petković, Valentin Čorić and Berislav Pušić, Judgment Summary, May 29, 2013, available at http://icty.org/x/cases/prlic/tjug/en/130529_summary_en.pdf (“During the trial, which closed on 2 March 2011, the Chamber admitted close to 10,000 exhibits into evidence and heard 145 Prosecution witnesses, six of whom were expert witnesses, as well as 61 Defence witnesses, seven of whom were expert witnesses. The Judgment is a 2,629-page-long document divided into six volumes, including four annexes and partially dissenting or separate opinions by Judge Trechsel and myself, all of which illustrates the complexity of this case.”).

Finally, the use of these variables is supported by the empirical evidence. For example, Professors Heuer and Penrod found that the length of the trial, the number of exhibits, and the number of witnesses were all good indicators of perceived complexity in their study. 114See Heuer & Penrod, supra note 33, at 36 tbl.2. Similarly, Professor Heise’s study found that trial length 115See Heise, supra note 33, at 352–53, 355–56, 359, 361–62. and number of witnesses 116Id. at tbl.7 & tbl.8 (finding that the number of expert witnesses affected attorney and judicial perceptions of trial complexity). correlated significantly with perceptions of trial complexity. Professor Heise did not have data on the number of exhibits and so could not draw any conclusions about the influence of exhibits on perceived complexity. In short, these three factors generally correlate with subjective evaluations of trial complexity. Therefore, they can act as a proxy for trial complexity. 117Cf. id. at 362 (noting that “[a]lthough it is possible to conclude that trial length itself may make a trial more complicated, a more likely explanation is that complicated criminal trials take longer to litigate”); Heuer & Penrod, supra note 33, at 36 tbl.2 (finding that the quantity of information presented at trial was the best predictor of subjective measures of trial complexity). Thus, these three variables were chosen to create a measure of the complexity of ICTY trials (hereafter called a Complexity Score).

There are potential criticisms of this approach. One is that two of the variables, the number of witnesses and exhibits, primarily measure factual complexity, and that as a result the Complexity Score does not do a good job of capturing legal or participant complexity. There are a number of reasons why I believe this criticism is misplaced. First, the three factors used in the Complexity Score are the factors that are most often identified as measures of complexity by those who deal with this issue regularly, 118See supra notes 109–112. and there is often value in a “wisdom of the crowd” approach. 119See Jan Lorenz et al., How Social Influence Can Undermine the Wisdom of the Crowd Effect, 108 Proc. Nat’l Acad. Sci. 9020, 9020 (2011) (describing the “wisdom of the crowd” effect, where the “aggregate of many people’s estimates tends to be closer to the true value than all of the separate individual or even expert guesses.”). Anyone who thinks that these are not the appropriate variables to measure complexity would have to explain why professionals who deal with these trials think they are appropriate. Second, our subjective assessment of trial complexity is driven largely by factual complexity. 120See Heuer & Penrod, supra note 33 (noting studies that show factual complexity is the largest driver of subjective evaluations of overall complexity). If factual complexity is the largest component of overall trial complexity, then it makes sense that a measure that tries to duplicate subjective measures of trial complexity would use variables that are associated with factual complexity.

Third, it is simply not true that the components of the Complexity Score only measure factual complexity. For one thing, empirical studies of complexity have shown that, particularly in complex cases, the various forms of complexity are often inter-related, and factual complexity cannot be neatly separated from the other forms of complexity. 121See Heise, supra note 33. For example, complex legal questions like whether an individual had the requisite specific intent for genocide or whether a widespread or systematic attack against a civilian population has occurred tend to require large numbers of witnesses and exhibits to prove. In other words, the number of witnesses and exhibits used in the ICTY’s trials is driven, in part, by their legal complexity. Thus, measuring the number of witnesses and exhibits does capture information about the legal complexity of the case.

The Complexity Score is also clearly affected by participant complexity. In multi-accused cases, each accused typically calls his or her own witnesses and offers his or her own exhibits. 122See Commc’n Serv. of the Int’l Criminal Tribunal for the former Yugoslavia, Case Information Sheet for Prlić et al., ICTY, available at http://icty.org/x/cases/prlic/cis/en/cis_prlic_al_en.pdf (last visited Sept. 27, 2013), at 4 (noting the number of witnesses called and exhibits offered by each individual defendant); Case Information Sheet for Popović et al., ICTY, available at http://icty.org/x/cases/popovic/cis/en/cis_popovic_al_en.pdf (last visited Sept. 27, 2013), at 5 (same). As a result, the number of exhibits, witnesses, and trial days in multi-accused trials is a function of participant complexity in addition to legal and factual complexity. But the Complexity Score also picks up other aspects of participant complexity. For example, to the extent that self-represented accused deliberately try to obstruct the court or draw out the hearings for political purposes, 123See Turner, supra note 29, at 530. this will show up as additional witnesses, exhibits, and trial days. In short, there is good reason to believe that the components of the Complexity Score do capture legal, participant, and factual complexity.

A different criticism might be that other variables should be added to the composite Complexity Score. For example, one might expect that increased factual and legal complexity would result in longer trial judgments as the judges wrestle with complex facts and difficult legal decisions. 124The Judgment Summary in Prlić explicitly claims that the length of the judgment is an indicator of the complexity of the case. See Prosecutor v. Jadranko Prlić, supra note 113. Indeed, there is a strong correlation between the Complexity Scores in this Article and the length of the trial judgments, suggesting that it too has some validity as a proxy for trial complexity. 125The correlation coefficient for the Complexity Scores and trial judgment length (measured in pages) was r = .72. This result was highly significant, with p ≤ .001. One could probably identify other variables that could be measured and would correlate with complexity. However, it is not clear that adding more factors would improve the measure.

Adding additional components has costs. For the Complexity Score to be useful as a means of comparing across courts it has to be based on components that can be readily measured across courts. Detailed statistical information about courts is hard to come by, so a simpler measure is better because it makes it easier to create Complexity Scores for other institutions. For this reason, the Complexity Score is composed only of the three components that the majority of experts usually cite as measures of complexity, even though one can find other measureable criteria that also correlate with trial complexity.

Indeed, for some purposes, it might be necessary to have an even simpler measure of complexity because not all courts report all three components. For example, it is extremely difficult to find data on all three components of the Complexity Score for domestic criminal trials. If complexity has to be measured using a single variable, the number of trial days is the best choice as it is the most commonly reported information, and at least one study has found it to have the strongest correlation with perceived trial complexity. 126See Heise, supra note 33. Moreover, measuring trial complexity with just the number of trial days should produce results quite similar to the Complexity Scores because of the high degree of correlation among the components of the Complexity Score. 127See infra Table 3. Nevertheless, this Article will use all three factors to create a Complexity Score where possible because most observers of international trials agree that trial complexity is best measured using all three components.

II. The Complexity of the ICTY’s Trials

This Section measures the complexity of the ICTY’s trials. Unless otherwise indicated, data comes from a database about the ICTY created by the Author. More information about the database and how it was created can be found in an earlier work about the ICTY’s indictments. 128See Ford, supra note 71, Section VII (describing the creation of the database). A copy of the data is available upon request from the author. For this Article, the database was updated to include data through June 17, 2013.

A. The Cases

The ICTY indicted 161 accused, and these were divided into 79 cases. However, 17 cases ended without a trial. 129Three cases were withdrawn by the prosecutor because the accused were not of sufficient importance, seven cases were transferred to domestic courts, and seven cases ended with the death of the accused prior to entry of a judgment. In 59 cases, first trials have now been completed. The trials of Ratko Mladić, Radovan Karadžić, and Goran Hadžić are still underway. These three trials represent the last cases the ICTY will try before it shuts down. As a result, the ICTY will eventually make a determination of guilt or innocence in 62 cases, involving 112 accused. This makes the ICTY the largest international criminal tribunal ever, at least by total number of accused. 130Its closest competitor, the International Criminal Tribunal for Rwanda (“ICTR”), has tried only 74 accused. See Status of Cases, Int’l Crim. Trib. For Rwanda, http://www.unictr.org/Cases/tabid/204/Default.aspx (last visited June 17, 2013).

Summary statistics about the ICTY’s cases are presented below in Table 1. The median ICTY case involved nine counts against the accused, including four war crimes charges, five charges of crimes against humanity (CAH), and zero genocide charges. 131Given that the averages for most of these figures are skewed by a few outliers, the median values are more representative of the cases that came before the ICTY. The number of counts, the kinds of counts, and the use of different modes of liability are consistent with what one would expect from an international criminal court that was principally interested in determining the guilt or innocence of the accused through fair trials. 132See generally Ford, supra note 71 (finding that charging and sentencing practices in the ICTY are generally comparable to the range of practices considered fair in domestic criminal justice systems).

Table 1: Summary Statistics of All ICTY Cases

Mean

Median

S.D.

Min.

Max.

Total Counts

13.3

9

11.9

1

66

War Crimes Counts

7.7

4

8.9

0

49

CAH Counts

5.2

5

4.7

0

24

Genocide Counts

.4

0

.78

0

2

JCE*

.4

-

.5

-

-

Superior Responsibility*

.7

-

.4

-

-

Planning*

.6

-

.5

-

-

Ordering*

.7

-

.5

-

-

Instigating*

.6

-

.5

-

-

Aiding and Abetting*

.7

-

.5

-

-

Direct Perpetration*

.4

-

.5

-

-

* represents a dummy variable

Forty of the ICTY’s trials were single-accused trials; the remaining 22 were multi-accused trials, but these multi-accused trials involved 71 accused. Thus while the majority of the trials have been single accused trials, the majority of the accused (63%) will be tried in multi-accused trials. Indeed, large multi-accused trials have been a hallmark of the ICTY’s practice. See Figure 1 below.

data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAqwAAAExCAYAAAC50vAwAAAAAXNSR0IArs4c6QAAAARnQU1BAACx jwv8YQUAAAAJcEhZcwAAFxEAABcRAcom8z8AAB0kSURBVHhe7d0/b+TI/SdgRUoU6Q0oG8DYUKHj TQUo2GSdeRMHzpQsNrEiewD7Fw3gUD7AwFhvYVNDMG6VO9bBBpwImzgQFCjgXUlduhKnyP5X7GKX ngf4AiM2h2QV2axPs9ndB3/4wx8+/r/qlFJKKaWUmmH9rxBYcw8opZRSSik1i3oNrAAAMCcCKwAA syawAgAwawIrAACzJrACADBrAisAALMmsAIAMGsCKwAAsyawAgAwawIrAACzJrACADBrAisAALMm sAIAMGsCKwAAsyawAgAwawIrAACzJrACADBrAisAALMmsAIAMGsCKwAAsyawUt3l5WV3cHCwVt3c 3Dz/34eHh+709PR52tXV1fO01oT+OT4+7u7u7hZTNnd+fv5FX56dnXVPT0+LOV5cX19/Md/FxcXi 0d3rb09um/fF/f19d3R09NqWdfdt/P+tHu+ruL297Q4PD98cE/GckJM77nNV4xjPtWVZrXvMbHue bOn5x/4SWJmNfnDtD0DpSTc+lp5IWzqJbhtqxqT9GGtooA4D/cnJSff4+LiYUk8L+zqGk7g/Y5Ba pz2xH957aOgfx2OBNUqDa3pc9wPZKssqJR4T6TrT7ek/98Nj654PSjx3SiwDtiGwMhurDhph0ImP tXiFNXfFpWRgDXJXtXP9F7blr3/96+KvuloYMGNg6gfWdY7d/jLeq00Ca3oM9V+Ipc+JXb5Iyz3H 0u3M7ef/+Z//WWvflzhPCqzUJrAyG6sG1jDfKoPTvkuvspYOJ2HA+9vf/rb0bVWBtZxNAlZf/8VM Ky/SNlE6sK56/tmFdFvm8sJk359/7D+BldkYGzDCQP3TTz8t/nrRn79/Eu0P7v0Kb4P333oPg9jP P//8ZiAMFbelP398Kz1enem/tb7NILhKYE0H7XUGkXiFpr99/fX0A+u2/RXm/9e//vVmWtzu/v7q 91V/wAzbFf8euyI2tg/W3Z9D+les+0Fy6FhcdflRfz2r7PNl+zgam2/d/R6lb8GHx4auDI7to6gf UH/zm9+8+Tv3f/rS9fSPmfSxXB+NbWOp4yhati3RKuvtb3f/mOn3a26eIF3OKscdlCawMhtjA0I4 AecGpKGTaDo9/L/0xN4fqNIwER8bu3qTzh8GhLBtv/zlL79Ydn+wWvft33SbVw0ZuT7KSYND3M5Y aT/2A2uwTX+l/ye33v/+97+vy+m3ud/Wfr/292uwyj5YdX8OicvsryP+HY310SrC8fDVV191f//7 39/059hyYvvjPo3b1u/bVeZbd7/HZcZpcZn97V1lH6XrifNtcuyn/6e/b+N2jD02to3bHkepdDvH AmuwynrT5eXOk7Fd6TknnS8YWgbsisDKbKQnxFBxAIrTcwNS7iSaDm7pQBEHmVDpQJMLs6sGsF/9 6lfP2xDnjyf+dJ5+O1YdvFYJrOl2rjOI9K90pX0TKm1HP7Cu21+5+YPcvgvSbUmXMzR/Oj1ud7Dq Plhlfw5J1x3XkVtvMNZHqwjrCtvTX87QNm7S/rH51tnv6fS0nSFMpX+vuu40TObWkU4fk+6vdPnp 9P5zbZN+XPc46hvbnr5V1psuL/dci+tI+7S/3qFlwK4IrMxGekLMVW5Ayp1E0xP40KCXDiDbBNZ0 +an+QBDkBr4xqwTWTfUDa7+9oUKoD9tcM7CmLyyG5k/7dWwwDnL7YJX9OWTZOtLjbKyPVhHWFf9P 2hfrHIM5q863aWANle7H1Cr7KF1WOt8m/bnsHJPbzlW2sT9t3eOoL93OZftllfUOPXfi9Nz+7K93 aBmwKwIrs9EfTOJAEINmbkDKnUSHTuBpYC11hTV34u7/31ytMrjuMrAGabti/elPf5p9YF1le3IV t3HZ/hwytH+Gjr+xPlomLPObb77Jtjm3rHRdY8fOqvMF6+739PkWqt+3q+6joX7epD/TYyjdN0NW 3cZg0+MoJ93OZftllfUOPXdS6XJipX26yjJgSgIrs5GeEPsnyzD45Qak3Ek0HWTSQSkNQumy1h2I lw0QQwPsukotJ2fVD7+E6l91Wre/cvMHQwNgicC6Tt9tGjSG1pEuL23vJgErCsd/fz+k/dR/y3nV 9q/TT7l+HmtT/7H0/wWbbOO2/ZkeQ+myhqzTP5seRznpdpZY79BzJ4rH0o8//jjYp8uWAVMTWJmN 9ITYP1mGk3L/WwKCoZNoOj0sJx14xgb3VQbiZQPENsEktc5gua6hwBr0r4zNPbDmAuI6+2DToDG0 f3LbE2x6XPT/X67S9QSrrmudbVp3v0fp/gwVn3+rrnuon9fZ9ig9hvp9lrPOOjY9jnLS7ZwysKZ9 G/bLWHuHlgG7IrAyG+kJsX+yHDJ2Eu2ffIeWue5AvMoAMRS61jE0UJcwFliDse1ft79y8wdD+26b wJq+GFl1H2wTNOI6hgJrurx1wk8qtLv/IitI+zVUv41p+3P/P1p1vnX3e2rdfZ1K17HrwBrs4jjq S7dzysDaP37H+nRoGbArAiuzkZ4Q+yfLIUMn0XDi/frrr0dP9NG6A/G6A0R/YAyPrdK2dLvGBq04 6KwziCwLrGn7+4P0uv2Vmz8Y2ndDAWFo/nT60AA7tg+2CRq5dQ9tz1gfDYn/Z2jetK/62z7W/n/+ 85/dv//97+d/rzrfOvs9TP/+++/fbE+8cj+078b2UXrVP13Huv05tr4hq27jvgXW3PklnRYq7dPc MmCXBFZmo/9W9NjVnmjoJBoG8mUn+lT/SkO63FBDJ+6hQa8/mMa2hAHhd7/73fO/l0kHolC5AXmV efrioLSsf+N8uatK6/RXuo3pPhnbd3H6UGCNy0n7ud+eVffBKvtzTNzeuPz+39EmASssa+w47u// /r5K+zL2cdiO3/72t69BNFh3vmX7PbY1t1/Tdq+6j+KxGOaJy+yfL0It69O0nWP9mlp1G7c9jlLL bstJrbLedJ50//aPx/BC9he/+MWbaVFuGbBLAivV5QaetIZO1v3BMlQ4kf7nP/95c5UgV/2BLR0Q Q/V/RSf+n/58ocYGp3SAXDZvlGtXWv1tj+tYZRDpb09ueakQiPrfEhBs219/+ctf3kwLFYJAbvvi /g/9MhRWxtowtg/W3Z9DlgWMfrCM1Q+1UW670v3bDxtD8wX9bRsKaqvMt+p+D9sXrrD+4x//WOkF 1SrPk/66P3369Lrusf0f5I6rWKsG1xrHUb/67VxlvUPnyXCM9B8Lx21/Wljn2DJgVwRWmjQ2QMVa NsgBAPMgsNKksStQsYaubgEA8yKw0pz4NtlQII1vz47dFwYAzIfASnNiIB26hyzcLrDJ/WUAQB0C K03KfRghliurALBfBFYAAGZNYAUAYNYEVgAAZk1gBQBg1gRWAABmTWAFAGDWBFYAAGZNYAUAYNYE VgAAZk1gBQBg1gRWAABmTWAFAGDWBFYAAGZNYAUAYNYEVgAAZk1gBQBg1gRWAABmTWAFAGDWBFYA AGateGC9v7/vjo6OuoODg9c6Pj7u7u7uFnO8uL297Q4PDwcfBwCAoHhgvby8fBNWQ11cXCwefXF9 ff0mpMbwenNz8/w3AABERQNruLr67bffdk9PT4spX4pXYK+urhZTXoSge3Jy0j0+Pi6mAABA4cB6 fn6+9C3+EEzHbhHoB1kAAN63YoE1vSc11tnZ2ZurrQ8PD93p6ekX04N45bV/+wAAAO9bscCaSu9j Td/mj6E0F1jHwiwAAO/XJIE1iAE0hNZ41TRehc1dRY3zr3sf6+fPn18b0a8//vGP2elKKaWUUqp8 hVw2hbj84oE1iFdUYwidIrCOKd0eAAB2b9LAGqQfshq7T3XsdoFNCawAAPtv8sAarqp++PDh+arp rj90JbACAOy/yQNr+JGANKAOfd9qvF2g5I8HCKwAAPtv8sD63XffvfnO1RhMd/HDAQIrAMD+KxZY w48G9N/qD9NyV0z7Px4QrsKGbxMo/dOsAisAwP4rFlhj6ExrLICm85e+shoJrAAA+69YYJ0jgRUA YP8JrAAAzJrACgDArAmsAADMmsAKAMCsCawAAMyawAoAwKwJrAAAzJrACgDArAmsAADMmsAKAMCs CawAAMyawAoAwKwJrAAAzJrACgDArAmsAADMmsAKAMCsCawAAMyawAoAwKwJrAAAzJrACgDArAms AADMmsAKAMCsTRpYLy8vu5OTk+7x8XEx5f+7vb3tDg8Pu4ODg+74+Li7u7tbPFKOwAoAsP8mC6wx kOYC6/X19ZuQGue9ubl5/rsUgRUAYP9NEljv7++7o6Oj56un/cAaH7u6ulpMeTF2NXZTAisAwP6b JLCen593nz596k5PT78IoSGY5m4BiFdZ+0F2GwIrAMD+Kx5Yw9v9FxcX3cPDwxeBNU47Ozvrnp6e nqdF8cpr+L+lCKwAAPuvaGANofOrr756Dqi5wBpDaS6wjoXZTQmsAAD7r2hg/e67717f6s8F1vi2 f+4qam7+bQmsAAD7r1hgDfempvef7iqwfv78+bUR/fr48WN2eun69a9/PZvKbZ9SSiml1C4q5LIp xOVvFVhDEP3hhx8Wf73IBdCx+1THbhfY1KbtWVcuONYqAIDWFAms4epq+AqrsQrB9eeff34Osa19 6CoXHGsVAEBrigTWnKG3+Ie+b3WKHw8QWAEA9t/OA+vQ963u8w8H5IJjrQIAaM3OA2vQ//GA8N2t 4baBff1p1lxwrFUAAK2pEliDGFLj/a0lr6xGAisAwP6bLLDOgcAKALD/BNYCcsGxVgEAtEZgLSAX HGsVAEBrBNYCcsGxVgEAtEZgLSAXHGsVAEBrBNYCcsGxVgEAtEZgLSAXHGsVAEBrBNYCcsGxVgEA tEZgLSAXHGsVAEBrBNYCcsGxVgEAtEZgLSAXHGsVAEBrBNYCcsGxVgEAtEZgLSAXHGsVAEBrBNYC csGxVgEAtEZgLSAXHGsVAEBrBNYCcsGxVgEAtEZgLSAXHGsVAEBrBNYCcsGxVgEAtEZgLSAXHGsV AEBrBNYCcsGxVgEAtEZgLSAXHGsVAEBrBNYCcsGxVgEAtEZgLSAXHGsVAEBrBNYCcsGxVgEAtKZo YL2/v++Ojo66g4OD5zo5OekeHx8Xj751e3vbHR4ePs93fHzc3d3dLR4pR2AFANh/xQJrCKAfPnx4 DagxvOZC6/X19ZuQGsPrzc3N89+lCKwAAPuvSGB9eHjovv/+++7p6Wkx5UUuiMYge3V1tZjy4vLy cvSK7CYEVgCA/VcksA7JBdYQTHO3AMR5+0F2GwIrAMD+mzSwnp+fdxcXF4u/Xq7Enp6edmdnZ19c jY1XXtP5tyWwAgDsv0kC61D4jNNzgXUszG5KYAUA2H9FA2sMnfFbAkKloTW+7Z+7ihr/77r3sX7+ /Pm1Ef36+PFjdnrpygXHWpXbPqWUUkqpXVTIZVOIyy9+S0D4JoAYWuN9qVME1jEl2zMmFxxrFQBA ayYLrEEMqDGEjt2nOna7wKYEVgCA/TdpYA3CB69iYPWhq+kLAKA1kwfW8DVWaUAd+r7V3FdgbUtg BQDYf5MG1nhFNf1u1aHvW/XDAWUKAKA1RQJr/17VqP89rFH/xwPiB7T8NOv2BQDQmiKBNfd1VssC aPotAqWvrEYCKwDA/isSWOdKYAUA2H8CawG54FirAABaI7AWkAuOtQoAoDUCawG54FirAABaI7AW kAuOtQoAoDUCawG54FirAABaI7AWkAuOtQoAoDUCawG54FirAABaI7AWkAuOtQoAoDUCawG54Fir AABaI7AWkAuOtQoAoDUCawG54FirAABaI7AWkAuOtQoAoDUCawG54FirAABaI7AWkAuOtQoAoDUC awG54FirAABaI7AWkAuOtQoAoDUCawG54FirAABaI7AWkAuOtQoAoDUCawG54FirAABaI7AWkAuO tQoAoDUCawG54FirAABaI7AWkAuOtQoAoDVFA+v19XV3cHDwWsfHx93d3d3i0bdub2+7w8PDpfNt Q2AFANh/xQLr5eVld3Fxsfir6+7v77ujo6PnQHpzc7OY+iIE2zSkxvDan29bAisAwP4rElhDOA0h tC8G0ZOTk+7x8fF5WgyyV1dXz39HIfCm85UgsAIA7L8igXXIw8NDd3p6+uZqagimuVsAYrjtB9lt CKwAAPtv0sAanJ+fv145jQH27Oyse3p6WszxIl55TW8r2JbACgCw/yYNrDGgxhAaQ2kusI6F2U0J rAAA+2/SwBre5v/w4cPrfanxbf/cVdQYWEvexyqwAgDsv0kDa7gdIP3k/xSB9fPnz6+N6NfHjx+z 00tXLjjWqtz2KaWUUkrtokIum0JcfvHAGj5c1f8A1dh9qmO3C2yqZHvG5IJjrQIAaM0kgTV8xVXu 0/4+dDV9AQC0pnhgDW/7//DDD4u/vjT0favxdoGSPx4gsAIA7L+igTVcWR26evrtt98+T4/B1A8H TFMAAK0pFlhDWA0/wzpUaUDt/3hA/L9+mnX7AgBoTZHAGgJoP6CmlftlqzTglr6yGgmsAAD7r0hg nSuBFQBg/wmsBeSCY60CAGiNwFpALjjWKgCA1gisBeSCY60CAGiNwFpALjjWKgCA1gisBeSCY60C AGiNwFpALjjWKgCA1gisBeSCY60CAGiNwFpALjjWKgCA1gisBeSCY60CAGiNwFpALjjWKgCA1gis BeSCY60CAGiNwFpALjjWKgCA1gisBeSCY60CAGiNwFpALjjWKgCA1gisBeSCY60CAGiNwFpALjjW KgCA1gisBeSCY60CAGiNwFpALjjWKgCA1gisBeSCY60CAGiNwFpALjjWKgCA1gisBeSCY60CAGiN wFpALjjWKgCA1gisBeSCY60CAGjNJIH1/Py8Ozg46K6urhZTvnR7e9sdHh4+z3d8fNzd3d0tHilH YAUA2H9FA+v19fVzAI01FFjDfGlIjeH15ubm+e9SBFYAgP1XNLBGMbjmAuv9/X13dHT0xWOXl5fd yclJ9/j4uJiyPYEVAGD/7TywhmCauwUgXmUdu41gXQIrAMD+22lgfXh46E5PT7uzs7Pu6elpMfVF vPJ6cXGxmLI9gRUAYP/tNLDGUJoLrGNhdlMCKwDA/ttpYI1v++euosbAuu59rJ8/f35tRL8+fvyY nV66csGxVuW2TymllFJqFxVy2RTi8vc2sI4p1Z5lcsGxVgEAtGangXXsPtWx2wU2JbACAOy/nQZW H7qavgAAWrPTwBoMfd/qFD8eILACAOy/nQfWoe9b9cMBZQoAoDWTBtaht/f7Px4Q5/fTrNsXAEBr igbWePU0hM+0cldaY0gNVfrKaiSwAgDsv6KBdW4E1v2Xa2OtAgDqEFgLyIWbWtWaXBtrFQBQh8Ba QC7c1KrW5NpYqwCAOgTWAnLhpla1JtfGWgUA1CGwFpALN7WqNbk21qrW5NpYowBgGYG1gNwgXKta k2tjrWpNro01CgCWEVgLyA3Ctao1uTbWqtbk2lijAGAZgbWA3CBcq1qTa2Otak2ujTUKAJYRWAvI DcK1qjW5Ntaq1uTaWKMAYBmBtYDcIFyrWpNrY61qTa6NNQoAlhFYC8gNwrWqNbk21qrW5NpYowBg GYG1gNwgXKtak2tjrWpNro01CgCWEVgLyA3Ctao1uTbWqtbk2lijAGAZgbWA3CBcq1qTa2Otak2u jTUKAJYRWAvIDcK1qjW5Ntaq1uTaWKMAYBmBtYDcIFyrWpNrY61qTa6NNQoAlhFYC8gNwrWqNbk2 1qrW5NpYowBgGYG1gNwgXKtak2tjrWpNro01CgCWEVgLyA3Ctao1uTbWqtbk2lijAGAZgbWA3CBc q1qTa2Otak2ujTUKqC/33KxVkCOwFpB7wtWq1uTaWKtak2tjjQLqyz03axXkCKwF5J5wtao1uTbW qtbk2lijgPpyz81aBTkCawG5J1ytak2ujbWqNbk21iigvtxzs1ZBTrXAent72x0eHnYHBwfd8fFx d3d3t3ikHIF1/+XaWKtak2tjjQLqyz03axXkVAms19fXb0JqDK83NzfPf5cisO6/XBtrVWtybaxR QH2552atgpydB9b7+/vu6Oiou7q6Wkx5cXl52Z2cnHSPj4+LKdsTWPdfro21qjW5NtYooL7cc7NW Qc7OA2sIprlbAOJV1n6Q3YbAuv9ybaxVrcm1sUYB9eWem7UKcnYaWB8eHrrT09Pu7Oyse3p6Wkx9 Ea+8XlxcLKZsT2Ddf7k21qrW5NpYo1qTa2Otak2ujbWqNbk21qrW5NpYq/bZTgNrDKW5wDoWZjcl sO6/XBtrVWtybaxRrcm1sVa1JtfGWtWaXBtrVWtybaxV+2yngTW+7Z+7ihoDa8n7WAXW/ZdrY61q Ta6NNao1uTbWqtbk2lirWpNrY61qTa6NtWqf7X1g/fz582sj+vXx48fsdKWUUkopVb4+ffq0SGhl xeXv9JaAXGAdu11gU1O3573Sr9PRt9PRt9PRt9PRt9PQr9OZqm/DckP50BUr06/T0bfT0bfT0bfT 0bfT0K/Tmapvw3JD7SSwBkPftzrFjwfsoj3vkX6djr6djr6djr6djr6dhn6dzlR9G5YbameBdej7 Vvf5hwPeG/06HX07HX07HX07HX07Df06nan6Niw31M4Ca9D/8YDwU60HBwd7+9Os741+nY6+nY6+ nY6+nY6+nYZ+nc5UfRuWG2qngTWIITVU6Sur0Z///OfFvyhJv05H305H305H305H305Dv05nqr6t FlgBAGAVAisAALMmsAIAMGsCKwAAsyawAgAwawIrAACzJrACADBrAisAALMmsAIAMGsCKwAAsyaw AgAwawLrDJyfn3cHBwfd1dXVYgrbur6+fu7TWMfHx93d3d3iUbZxf3/fHR0dvfbtyclJ9/j4uHiU ki4vL/VvIf3j1nlhGuGYjf17cXGxmMq6csdrWu/xvCCwVtQPVQJrGeGEmZ4o0yf+zc3NYiqbuL29 7T58+PB6oox9K1SVF/r68PBQ3xaSBqlYAlU5cTzzIqCMfj7o13s8dgXWGYgHpsC6vRCgQn/2Gfy3 9/Dw0H3//ffd09PTYsqL2LdeDJSTvshyzG4v9Oe33377xbFLGfHFgDGsnO+++y4b/MN5+Ouvv36X LwoE1hkQWKcXnuSnp6de/U9AYC0v3Cb06dOn52NWYN1evO3K878841d54Zz6008/Lf56Kzz2zTff vMsXXwLrDHjC70YYtAz+5YV+9dZqOeF8EPozvshyzG4nvqAK59hYZ2dnrrYWEN8J0J+7E65mv9es ILDOgMA6vTj4C1blxMFKn5YT+vSrr756DqgCa3npfaz6dXuxP3/88cfnY1XfTus93w4QCKwzILBO L1xlST8sxOZikIqDUyihtYz0vjWBdRrp8eu43Vzaj+mtFvGKttsvynvPtwMEAusMCKzTC29bu8ey vHjsOn6313+rT2CdTnx3QN9ubux2gHjl1QuCst7z7QCBwDoDAuu03vuTfGrxiorBf3OhD3/44YfF Xy8E1mmF84KrgJuLz/tcKPWCoLxwPnjPtwMEAusMCKzTCX2rX6fnA23biVekxkr/lhUCl9uENjd2 D7sXW+W999sBAoF1BgTWaeSuWjGNELh8Urgsg/60wnnXMbu5seMzPqZ/y/FOocA6CwJreUODUbgq 4AvEy4qDk+O3LIF1WkNfzM7q4tjV/3xAvF3A5wbKCOeC9347QCCwzkB80rtBvYzYn0MlWG0mDkL9 AOV7WKchsJYRjs/+i1cfwiwn9GV6L/DYrQJsxruFLwTWimIAEKjKWXYvoA9ZbC4GqH6fGvinIbCW kXsB65gtq3/eNYaV5QXWC4EVAIBZE1gBAJg1gRUAgFkTWAEAmDWBFQCAWRNYAQCYNYEVAIBZE1gB AJg1gRUAgFkTWAEAmDWBFQCAWRNYAQCYNYEVAIBZE1gBAJg1gRUAgFkTWAEAmLUvAqtSSimllFJz rBBY/3d/olJKKaWUUnOo3//+9//n/wJtY4/kzfDKaQAAAABJRU5ErkJggk== One thing this data shows is that the cases that have come before the ICTY are not like the cases that come before domestic criminal courts in the United States. For one thing, there are significantly more multi-accused trials at the ICTY. In the U.S., most accused are tried alone. 133For example, in federal district courts in the United States, the average number of accused per criminal case is only 1.3. Thomas F. Hogan, Admin. Office of the United States Cts., Annual Report of the Director: Judicial Business of the United States Courts, at tbl.D (2012). If one compares the number of criminal cases opened in 2011 (78,440) with the number of new criminal defendants in that same period (102,931), one can surmise that there were approximately 1.3 defendants per case. Id. at 192, 195. The number of counts in the average ICTY case (13.3) is also significantly higher than in the United States. According to one study, nearly half of all felony defendants in the U.S. are charged with just a single crime. 134See Thomas H. Cohen & Tracey Kyckelhahn, Bureau of Justice Statistics, U.S. Dep’t of Justice, Felony Defendants in Large Urban Counties, 2006, at App. tbl.1 (2010) (noting that 48% of felony defendants had no other charges pending against them).

The cases heard by international criminal courts also involve much more serious allegations of criminality than the vast majority of cases heard by domestic criminal courts. For example, the average case at the ICTY involves allegations of the unlawful killing of more than 100 individuals, often in conjunction with the forced transfer or unlawful imprisonment of thousands of people. 135See Ford, supra note 71, at 104. By contrast, only 0.6% of felony defendants in U.S. courts are charged with even a single murder, and only 23% of felony defendants are charged with any sort of violent offense. 136See Cohen & Kyckelhahn supra note 134, at tbl.1. Only the most serious mass atrocity trials that take place in domestic systems can compare with the seriousness of even the average ICTY case. 137See Ford, supra note 71, at 104.

B. Computing Complexity Scores

Summary statistics for the number of trial days, trial witnesses, and trial exhibits needed for the ICTY’s trials are shown below in Table 2. These figures are based on the 41 cases that have had full trials so far. It excludes those cases that were dropped prior to trial, 138See supra note 129. the 18 trials that were resolved with guilty pleas, 139In cases where the accused pled guilty, either no evidence was presented or only limited evidence was presented. and the three trials that are still underway.

Table 2: Summary Statistics of Complexity Components

Mean

Median

S.D.

Min.

Max.

Trial Days

176

150

102

10

465

Trial Witnesses

121

104

63

14

315

Trial Exhibits

2,022

1351

2015

37

9,876

To construct a Complexity Score for each trial, the value of each of the three complexity components in that trial was converted to a ratio of that value to the highest value of that variable in any case that had a full trial. The three ratios were then added together to produce a Complexity Score. The scores have a theoretical range of 0 to 3, but the actual scores range from .07 to 2.63. 140See infra Appendix. The three components of the Complexity Score are highly correlated, as indicated below in Table 3.

Table 3: Correlations Between Components of Complexity Score

Trial Days

Trial Witnesses

Trial Exhibits

Trial Days

1

Trial Witnesses

.7877***

1

Trial Exhibits

.8819***

.6553***

1

* p < 0.05, ** p < 0.01, *** p < 0.001

C. Conclusion

The Complexity Scores allow us to make comparisons between ICTY cases. A list of the ICTY’s completed trials, ordered from least to most complex, is attached as an Appendix at the end of this Article. This list shows that there has been significant variability in complexity across the ICTY’s trials. Furundžija, the simplest ICTY trial, involved only 14 witnesses and 37 exhibits and lasted only 10 trial days. It had a Complexity Score of 0.07. The most complex full trial, Prlić et al., involved 199 witnesses, 9876 exhibits, and took 465 trial days to complete. It had a Complexity Score of 2.63. Thus, the Prlić et al. case was nearly 40 times more complex than the Furundžija case.

The average ICTY trial took 176 trial days, and involved 121 witnesses and 2022 exhibits. It would produce a Complexity Score of 0.97. By comparison, the median ICTY trial took 150 days, and involved 104 witnesses, and 1,351 exhibits. It would produce a Complexity Score of only 0.77. The median score is lower because outliers (like Prlić et al. and Popović et al.) push up the average values.

Collectively, in these 41 cases, more than 4900 witnesses have testified and more than 80,000 exhibits have been entered into the record over the course of more than 7200 trial days. The trial judgments, if added together, cover nearly 18,000 pages, while the appeal judgments cover more than 6000 pages. In addition to the judgments, there have been more than 15,000 other judicial decisions, and 22,000 filings by the parties, covering a dizzying array of topics. 141This information was obtained by searching in the ICTY Court Records database located at www.ICR.ICTY.org. Under “Select Name of Accused,” the option for “All Accused” was chosen. The results were then narrowed down to only those in English to avoid picking up duplicates of the same document in multiple languages. Judicial decisions are those documents that appear under the category of “Decisions and Orders.” Filings by the parties are a combination of the results that appear under the categories of “Briefs,” “Responses,” and “Motions.” The searches were conducted on July 7, 2013. The searches were then repeated in French and Bosnian/Serbian/Croatian. The results were substantially lower for both other languages, suggesting that significant numbers of documents are available only in English. There are also more than 20,000 official documents of various other sorts in the ICTY Court Records database. 142This figure combines the English language documents that were found under the categories of “Notices,” “Transcripts,” “Witness-Related Materials,” “Correspondence,” “Warrants and Subpoenas,” and “Other” on the ICTY Court Records Database. More than 10,000 of these documents are transcripts of court proceedings. The search was conducted on July 7, 2013. Along the way, the prosecution has cataloged, analyzed and disclosed millions of pages of documents to the defense teams. 143See Higgins, supra note 17, at 246 (noting that the prosecution disclosed more than 1.2 million pages of documents to the defense just during the trial of Slobodan Milošević).

III. Comparative Complexity

This section compares the ICTY to both other international criminal courts and to domestic criminal courts. While each of the modern international criminal courts (ICTY, ICTR, SCSL, ECCC, and ICC) is unique in certain ways, they are all also quite similar. They have jurisdiction over very similar substantive crimes, have similar structures, and often have similar procedural rules. Accordingly, it is quite common to make comparisons between international tribunals, 144See generally Catherine Cisse, The International Tribunals for the Former Yugoslavia and Rwanda: Some Elements of Comparison, 7 Transnat’l L. & Contemp. Probs. 103 (1997) (comparing the ICTY and ICTR); Gideon Boas, Comparing the ICTY and the ICC: Some Procedural and Substantive Issues, 47 Netherlands Int’l L. Rev. 267 (2000) (comparing the ICTY and ICC); Donna E. Arzt, Views on the Ground: The Local Perception of International Criminal Tribunals in the Former Yugoslavia and Sierra Leone, 603 Annals Am. Acad. Pol. & Soc. Sci. 226 (2006) (comparing the ICTY and SCSL). and most writers seem to accept that the courts are similar enough for such comparisons to be useful.

But, can the ICTY be compared to domestic courts? There are greater differences between the ICTY and domestic courts than between the ICTY and other international criminal courts. For example, international criminal law is substantively different from the vast majority of domestic criminal law; 145While there are similarities between domestic and international criminal law, the key international crimes (war crimes, crimes against humanity, and genocide) were created by international law and have no direct analogs in most domestic legal systems. See Cryer et al., supra note 83, at 8–12. trials at international criminal tribunals generally deal with crimes that are much graver than the average domestic trial; 146See supra notes 135–137 and accompanying text. the purposes of international criminal trials differ from those of domestic criminal trials; 147See supra Part I.A. and international criminal tribunals have their own procedural and evidentiary rules, 148See, e.g., ICTY, Rules of Procedure and Evidence, U.N. Doc. IT/32/Rev.49, (May 22, 2013), http://www.icty.org/x/file/Legal%20Library/Rules_procedure_evidence/IT032Rev49_en.pdf [hereinafter ICTY Rules]. although these may be quite similar to the rules that govern trials in the United States. 149See infra Part IV.A.

There are differences, but the differences are not so dramatic that domestic and international trials are fundamentally different undertakings. Indeed, trials proceed quite similarly before international tribunals and domestic courts in the United States. The prosecution has to prove the accused’s guilt beyond a reasonable doubt, and this is done by presenting evidence in the form of witness testimony and exhibits to the trier of fact. 150ICTY Rules, supra note 148, at R. 87(A). The trier of fact is often different, however. At the ICTY, the trier of fact is a Trial Chamber, composed of three judges. In many criminal trials in the United States the trier of fact is a jury. See Kwon, supra note 1, at 363. The accused, in turn, has the right to examine the evidence presented against him and to present his own evidence. 151ICTY Rules, supra note 148, at R. 85. The overwhelming majority of accused at international criminal courts are male. On the whole, domestic criminal trials and international criminal trials are more similar than different.

This similarity is unsurprising given that fair trial guarantees are a key component of international human rights law. 152See, e.g., International Covenant on Civil and Political Rights art. 14, Dec. 19, 1966, 999 U.N.T.S. 14668. Thus trials in both domestic and international courts must comply with the same minimum standards. There is still room for differences across legal systems, but the fundamentals should be quite similar. Indeed, the fundamentals are similar enough that, once the trials have been adjusted for their relative complexity, a comparison between the two systems tells us something useful about international criminal justice. 153Cf. Shany, supra note 26, at 249 (suggesting that it is appropriate to compare international and national courts that share “comparable functions”); Wippman, supra note 14, at 862.

Having said that, the trials at the ICTY look most like trials in common law countries like the United States, Canada, Britain, or Australia. 154While the overall ICTY system is a blend of common and civil law features, the common law features are more prominent, particularly at the trial phase. See Megan Fairlie, The Marriage of Common and Continental Law at the ICTY and Its Progeny, Due Process Deficit, 4 Int’l Crim. L. Rev. 243 (2004); Kwon, supra note 1, at 363. It may be that the differences between the ICTY’s trials and trials in jurisdictions that do not use a common law system would stretch the comparison too far. For example, it is not clear how one should account for the work done during the investigation phase in countries that use an investigating judge. 155The work of the investigative judge may well shorten the resulting trials by establishing much of the evidence that the parties would otherwise be required to present to the court, but would not show up in the variables used to compute the Complexity Score. Cf. Fairlie, supra note 154, at 248–51 (comparing the pre-trial phases in the common law and civil law systems). Thus, this Article will focus primarily on a comparison between the ICTY’s trials and trials that take place at other international tribunals or in courts in the United States.

A. Complexity of Trials at Other International Courts

This Part uses the methodology described above to compute Complexity Scores for a selection of trials at other international tribunals. For example, the ICTR’s first case, the trial of Jean-Paul Akayesu, took place over 60 trial days and involved the use of 41 witnesses and 155 exhibits. 156Prosecutor v. Jean-Paul Akayesu, Case No. ICTR-96-4-T, Judgment, 24, 28 (Sept. 2, 1998), http://www.unictr.org/Portals/0/Case/English/Akayesu/judgement/akay001.pdf. This results in a Complexity Score of only 0.27. Thus, the Akayesu trial would rank somewhere between the simplest and the second simplest trial the ICTY undertook. The case against Kaing Guek Eav alias Duch at the Extraordinary Chambers in the Courts of Cambodia (ECCC) involved 55 witnesses, approximately 1000 exhibits, and took 77 trial days. 157Kaing Guek Eav, Extraordinary Chambers in the Courts of Cambodia (ECCC), http://www.eccc.gov.kh/en/indicted-person/kaing-guek-eav (last visited Oct. 10, 2012) (identifying the number of trial days and witnesses); Prosecutor v. Kaing Guek Eav alias Duch), Case No. 001/18-07-2007/ECCC/TC, para 56 (2010), available at http://www.eccc.gov.kh/sites/default/files/documents/courtdoc/20100726_Judgement_Case_001_ENG_PUBLIC.pdf (identifying the number of trial exhibits). This results in a Complexity Score of 0.44. If the Duch trial had taken place at the ICTY, it would have been one of the simpler cases, falling somewhere between the case against Pavle Strugar and the case against Milorad Krnojelac. It would have been significantly less complex than the median ICTY case.

The Lubanga trial at the International Criminal Court (ICC) involved 204 trial days, 67 witnesses, and 1373 exhibits. 158See Prosecutor v. Thomas Lubanga Dyilo, Case No. ICC-01/04-01/06, Judgment pursuant to Article 74 of the Statute, 11 (Mar. 14, 2012), http://www.icc-cpi.int/iccdocs/doc/doc1379838.pdf. This implies a Complexity Score of 0.79. This is almost exactly the same as the complexity of the median ICTY trial. In contrast, the trial of Charles Taylor, the former head of state of Liberia, at the Special Court for Sierra Leone (SCSL) took 420 trial days and resulted in the calling of 115 witnesses, and the entry into the record of 1522 exhibits. 159See Prosecutor v. Charles Ghankay Taylor, Case No. SCSL-03-1-T, Judgment Summary, 8 (Apr. 26, 2011), http://www.worldcourts.com/scsl/eng/decisions/2012.12.17_Prosecutor_v_Taylor.pdf. This equates to a Complexity Score of 1.37, which would place the Taylor trial in the top quarter of ICTY trials in terms of its complexity.

Not even the trial of the Nazi leaders, however, can compare to the most complicated ICTY trials. The trial of the major German war criminals by the IMT was a massive undertaking. 160See International Military Tribunal, Trial of the Major War Criminals before the International Military Tribunal, Nuremberg, 14 Nov. 1945–1 Oct. 1946 (1947). The introduction to the judgment contains a description of the trial, from which the figures necessary for calculating its complexity have been drawn. There were 403 “sessions” of the Tribunal. 259 witnesses gave evidence, including 143 who gave evidence in written form. Finally, approximately 5800 exhibits were entered into the record. 161While the Judgment of the IMT is somewhat vague about the number of exhibits (it refers only to “several thousands of exhibits”), other sources give more exact figures. See Lawrence Douglas, The Memory of Judgment 12 (2001) (noting that there were approximately 4000 documents and 1800 photographs that were entered into the record as exhibits). This indicates a Complexity Score for the IMT trial of 2.27. Yet, had the ICTY heard the case against the major German war criminals, it would have been only the third most complex case the ICTY has heard. If the complexity scores for all the ICTY cases to date are added up, the ICTY’s trials have been more than 17 times as complex as the case tried at the IMT. And this multiple will only get higher as the remaining ICTY cases are completed. Figure 2 shows the relative complexity of trials at a number of international criminal courts.

data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAArMAAAHJCAYAAACWtrEhAAAAAXNSR0IArs4c6QAAAARnQU1BAACx jwv8YQUAAAAJcEhZcwAAFxEAABcRAcom8z8AAEnISURBVHhe7d29i1zZtfdxRZ0oUubEygSDQoVO 7EDg5DZoQImcWcF1dBnoRAwDreBy75jL/AMO5AcEGkWOJlFqhHlG2YBzmRmYRChxIBQoqOvV6tV3 99Zvn5eqs6r3qvP9wAJ1vZyz9utZXaquuvbf//3fX/8rNgRBEARBEASRLP6fFbPqDoIgCIIgCILo Pi6KWQAAACALilkAAACkRTELAACAtChmAQAAkBbFLAAAANKimAUAAEBaFLMAAABIi2IWAAAAaVHM AgAAIC2KWQAAAKRFMQsAAIC0KGYBAACQFsUsAAAA0qKYBQAAQFoUswAAAEiLYhYAAABpUcwCAAAg LYpZAAAApEUxCwAAgLQoZgEAAJAWxeyWHj9+vLl27dqsePny5dlz3717t7lz587ZbU+ePDm77RCU 7bK4cePG5vXr1+f3bu/Vq1ebo6Oji+NOCT/3mzdvNtevX7+4/ebNm5v379+fH3lZ9+7du5RDK3bJ 4fnz55eOdXx8vPnw4cP5vfNZ/5yenp7/9FGP87OeW3X77f5Hjx7t1Bf7tPQ4jhkbUzUPrkq9ZtU+ Mrb/7jpve+qPbWVZ226bNbHLtWGqXfts32t9rShmd1Rvql6wunIh+H3l5D6UiV1fgMqo+2Qu37DK 45R9WG9Odl95W7nhRRazpu6Hk5OT83s+3dS27Zcl54/N3zJH09v8LH9JqHMtxzbbWtpnP4+dS82D q+Dj6evXx77VP/X+u1SB1kt/7CLD2q7NzW/Xa8MUS/RZ7/1+CChmd1ROUotWgWKbst/X82/H2xp6 VXLXAtI2rKdPn57/9NHQhmW++eabi9vKAjO6mC3H1qK+mJT9dNUbox+nzrGX+Vn3ZWtt+eOyXST2 eYEbGtPWPLgKvj7qYrY1D8f2gW301B/b6n1tt8xdE7teG6ZYos/mtgvzUczuqJykFq0Lrj2udV92 Viz+8pe/vNgg6iJkqYtMaWzDKvVUzJavJG3bL0tsjOUxer1ol4X/WI52Ubt//z7F7Ew9zYN67UzZ L+fsA1NkWBdjMrdhiTWx9JxYwhLtwjCK2R2Vk9Si3IDtAvv999+f//RR/fh6Yvt/m5SPKcM2p/q/ sq1Ae/v27aULgYXn0vqvby+s6g1vqE3Kd99998mGUbaj3lDKi9Y+Nqy6mP3xxx8vfh567tx+MHOK 2VZhPXbe8v66/+rzq8e0XkW3fP785z9fus2e+/PPP1+aPx6e/1ibt+nHeh2MPcdyUO+ZLfvbon5l Ra2lcn5YeP8N5VS30fqg7Gc11kPjaIb6rW6XhRo/e059HD9Xax6o8PznjosbG4fWvlfPpVrZtnot q7H961//2sx/aF2UYzc0LlP22rl5uYi1vcR6WSp3M7YmphiaE6UpY9VaO26JdpVjZn0395VjfEQx u6Ohjc0WhVrYrYld3m7PKxdbvaGWm7/fVy+s8tzl422RWm6/+tWvPjl2uZCNL7S5/71S5l4v3qE+ m6o8xpxi1sLaWxYsdd+abfuhHgN/vis3rvo+M+W8Y/PHnzt1DOo81PHrdtXHM5Zr3T/b9mPZT2p8 pvBj1Oeu21sXUn4+z93D2vzPf/7zoh/qeVc/3ue1n7duR2sczdx5UOZizx0a7/K+1jxoHdv4XJi6 bqeOw9D+1TKUp6nH1tpe/nI2NCZ1fmbKuEzZa7fNy8+75Np2U8dpX7nX901VHmPs2jBlrFo5LdEu n0/1XrHNNXHtKGZ3VE7SchL67WpSqoldbuTlZuCT26LcMMuF448fuhiUi/Z3v/vdWQ7+eF+M5WPq dtQb1Jgyv7oPyjz3sWGN9VX9/F36oR6D8kJQ5qyKuannLY9T9p/PFW/PUBvLY5Q5mtbxy36sj2f3 PXjw4NJYbtuPdR9uM0fKNvi5VT5GzQ8z1s/1cfzCVD++PE6rGC0fP6ffylxsHO256u0WrXOVt9fz oD62s/GZ+skR5fHHxqEe9/K+lvL4ah+Yu/aH+mPquJSPa+21c/OKXtvl7WPjtK/cy/zmKI9RH782 ZaxaOe3arvLxZf/aPjJl7uMyitkdlZNUhZqUamKXi6rcGMsLZLkxjW0oFuW5W8cv1YvTtDa0Md7G bTekMWUfjm1Yc/tql36oj1tHfXEpTT2vmj/l7VMuMOUxpl7wjOdoURdmdYG+bT+W42WxzRwaO/fY WjKtfmj1QWutludV41XfPqff6r6qx9m1zlXeXs+DoX4ZWgOlOeMwtCZbyvxV2+eu/aH+mDou5W1l v5W2zUs9tm73UBvK++bMufI4+8q9zG+O8hhqTpSmjFUrp13bVfeZeoED01HM7qicpBa+iOv/Piip id1aVOUFspzsczeU8vhqk6ifq0K1pcU2x9bmsISyD8c2rDl9tWs/1M+3i0DZ963nzzmvmj+1+pzl 8015jKkXPNOap1988cWlsd6lH8vxspg7j8rnty7O5THV/DCtfvALv8XcYnbs+Nv0W+u8pVZbhuaB UW2tx7pl7jjUbW/Nj1KZv9oH5qx90+qPOeNStq+1PufmVSqPrx47NKZqHiyxXiJyb/XdmLE5UZoy VlNy2rZd5dqt78M8FLM7KiepRTmBbaKqRa0mdrkZlBtHeTEpjzV3QxlbtK0NbRvWvl2PMabsw7Fz zemrXfuhPq5fTMpxLMfXzTlva2N0fq4XL14050N5jCkXPFe3zwocm1tffvnl+SM+2qUf63Oo/hrS One5BspjqvlhWv0wt5idc/xt+m3Kc1ptGZoHpt437L2RU79MYO44DO1fLWX+qu2q74fO0+qPOeMy tteauXm5iLU9d5z2lXur78aMzYnSlLEay2mXdtX9ZlH2NaajmN1ROUktyglsC6X+NAPTmtjl7Xac ctOoN6W5G8rYop2yGU1heZUf0xWl7KuxDWtOX+3aD/XzfdzKHMrb3ZzztuZPPV+GjlkeY8oFr1QW bXb/H//4x0/6f9d+rF+xmPP8sh+2vTibVj9EFrPb9luZUz2eptWWoXlgynysL//zP/9zck5zx2Gb tpf5q31A9f0262JOblMKpLl5lY+3vLZpg1HzYO447Sv3Vt+NGZsTpSlj1cppyXaV69eiHjeMo5jd UTlJLcoJ3DI0sesF0Trm3A1lyqJtXaSnsvPfvXv3k83Dbl/6q0bnbFhz+2qXfqiPW25KY3Nl6nlb 88ef7/0x1MbyGFMueKWyP1uPMbv0Y32OuZt73RemtQbU/DBj/WwxpZhtnXfu8VvsOPbHK94GtR5a 5ypvb/Vx2a6yf6aYMw5D87WlzF+1e+7aH+qPqeMyZa/ddk+KWttzxmlfubf6bszYnChNGaupfbZr u5Zo+5pRzO6onIAW5QRuaU1aWwyqGFTmbihzF2190bL7xtpWbvZ1tC4M2y7aMtexDWuXC9rcfqiP O3RBVMceus/PWz7O+69so/dHeZtFq41TL3ilsh2t/pjanpby+WNjbAVXWWCUz1X9Vp5bzQ9TPr7s h7LtU4rZ8jitHMvjl7eP9Zut61u3bp09pjx/PW5TzlXPA1f2T+sxLeXxx8ZhaE22lMdSc2SXtT+0 LobGZelitnxs1Noub/djqdvMvnJv9d2Y8hhj+8a2xewS7bI+q1/k8TW8bdvXjGJ2R+UFxGLKZt9a sHaRHFt8Jb+o+nPK41q0FlO9Ebt6Q/K22CIde59c3Q91lLmUG0h931T1+aa+UuJ9NXRB26Uf6g2t 3pTqtpf3Tz1va2Os22Mfvv3ZZ59dus3Vm/jf/va3i6+FbM3Pkj+mNZfMLv3oxuaK97caf18fft76 Z6fmh2n1w5Ri1sJz9cerAsgfu808qNvuP/vzphTOQ/Og5G2o+3+KqeOg5vCYsX1Aje3QeYb6Y+q4 lH3dWh9z8lK3R6ztqeMUmftQflPNuTZMGSuV0xLt8mOU7dxlna0dxeyWxoq31gIqJ7WHTebWtyyV UU/w+uL1hz/84dIC8+fUj7MYKkJ8QU15rBnri7JAcH6OuRvW2LnG+sjC8ik3HY/6uXP7oX58GeV8 qNtQ98/QeVvzx/qwvs/OWd9WtlFtsEPHL/lGPHShcHP7USkvoHOONXZha62N+tuSLOyirsbYj1me q35sneeUfh7qt/q+v/zlL3L/sDzGzqXmQc36f5evCx4bh9b41oWUG9sH7Pjbrv2x/hgalyl77TZ5 1WNo7atvW2Jtb7Nelsp9Sn5DxuZE2T9mylgN5bRru+wLWOyVWftlo5z7dZ6YhmK2I+pCWQcTHehT XcweEis21Su2ANADitmO1P91oeLQLpLAoTjkYpbviwfQM4rZTvh/ebQugv7fcPV/+wDowyEVs94W a4f9kr30p5EAwJIoZjvhxWr9nh1nb0Fo3Qfg6h1SMetvebL39tl7+tTnZQNALyhmO6LekO7BK7JA v9QfedR/2JdJ+QdZc/4IBwCuAsUsAAAA0qKYBQAAQFoUswAAAEiLYhYAAABpUcwCAAAgLYpZAAAA pEUxC2Br9nFyp6en5z+1TfmqZouhj6ArvyFPPa7+aLvyo7GG7stgaj8rrW8W9I/c6vVLEXZpM4B1 oZgFsDX7ooCpXxBQF1Xll4CUn2tqoYrV8rNc688+9ed7kVp/6H/rviyfnzqnn0vlLxH183v/LNlt 2wxgfShmAWzFi8s5BUf5LVn1N9rVr57WBe3QK7NetNUFqz1u6L4Mtunn+heHly9fnt9zmT+ut2J2 mzYDWC+KWQCzla+SLlXMmvL+qW8FGCrcphZ1vdq2n4deka3ZK7T379/vppjdts0A1otiFsCFuviz qF+1KwulMlRxWhsrZltvNygLHAvPqX68hxVBQ/eV6mOXBW/9arE/19uxy7Gs/X/9618v5Vg+ftt+ rts9VsDbmKv3zJZjZVG/kq3a8/bt20/mj59/Sl9OaXP5GDv2N998M+mXHgCHi2IWwBkvxLzIKIuP uqDd9tWzsWK2LnjK85bnLG/f5ZXZuij1Qqks3Mri0B5nz/nVr3519nPZhrnHsrB2/Pzzzxdtrvtk m34uiz3Vx1P4Meq21DmU7fFzDfX5lL4carP3sR/T86rHFcC6UMwCOOOFgf/3flmU1P/lH1XM1oVQ +Zili9mysPLb/RzlecvH/e53vzt7jB/X2z71WGWx7rcv2c91e+tfQqYoz+ltUe0zY+2pHz+lL1tt Lo9bHtPmVPkzgPWhmAVwpi6+liyy3Nxitjxvec4litm6eDeqaCtvUzmbqceaW/zN7eehV7anGmtL mccuxWyrL6cUsxZZ/oAPQDyKWQCfKIsOj12KLNfLK7P17Sr8sWVfqOJwzrH2Xcy2CsaW8vmtYrY8 5i7FbKvQHmpzOX+GjgFgXShmAVzir8y9ePFisSLLzS1my2JlyWK2VbQpYwXYnGPNLf7m9nN9rEMr ZutjW8xtI4DDQzEL4ExZmFgRsWSR5caK2TKH+thLFrNDz6mNFWBzjhVdzJr61cuhfGq9F7POf+Ea exyAdaCYBXCmfq/k0kWWGStmy+PWr3IuWcyasiAaev/llAJs6rH2UcwO/UIwRT0PTKsPrqqYNa35 AGB9KGYByFfk6qIoupiti6C6KFy6mC2PV+di9/lj5xZgQ8faRzFryufVvxTUbEzKvi6f67mo28w+ i1k7bv15uD6fKGaBdaOYBSALEPsw+s8+++zSba4uSv72t79tnj59en6vVp+jLPqGCiBXFjlLFLP1 fV44WYF2enp69m9TnrcuVN3UY5X9NuUV8G362ZXPrY9rvBBVryT7q7PejvrnUv1KbtlfFuV5p/Rl q83eT+XY+7nrtgFYF4pZAGfqIsSKnKmFSVlgKF50jEXrv+jrPCy80CkLNg8ruupirryvVOdWFlnl K4/q/trcY1kBWP7C4LFtPyutfhhqh6nfe9sam7pdf/jDHy4V5xbWnjl9qdpsxay9MluPOYUsAIpZ AAAApEUxCwAAgLQoZgEAAJAWxSwAAADSopgFAABAWhSzAAAASItiFgAAAGlRzAIAACAtilkAAACk RTELAACAtChmAQAAkBbFLAAAANKimAUAAEBaFLMAAABIi2IWAAAAaVHMAgAAIC2KWQAAAKRFMQsA AIC0KGYBAACQFsUsAAAA0qKYBQAAQFoUswAAAEiLYhYAAABpUcxu4c2bN5vr169vrl27dhY3b97c vH///vxeAAAA7AvF7EyvXr3a3Lp166J49cKWghYAAGD/KGZnePfu3ebRo0ebDx8+nN/ykRW4R0dH m5cvX57fAgAAgH2gmF0AxSwAAMDVoJhdwL179zYnJyfnPwEAAGBfKGZ34O+XpZAFAAC4GhSzW7D3 zt65c+fi0wwsti1onz17djEIdfzP//yPvJ0gCIIgCCJbWM0TwY9PMbul58+fXxS0T548Ob91GYwH AADAMIrZBfgfgC398VyMBwAAwDCK2YXYH4FRzAIAAOwXxexCHj9+vDk+Pv7kM2h3wXgAAAAMo5hd gP9BGO+ZBQAA2C+K2Rla742N+pxZxgMAAGAYxewM6iO5LKK++YvxAAAAGEYx2zHGAwAAYBjFbMcY DwAAgGEUsx1jPAAAAIZRzHaM8QAAABhGMdsxxgMAAGAYxWzHGA8AAIBhFLMdYzwAAACGUcx2jPEA AAAYRjHbMcYDAABgGMVsxxgPAACAYRSzHWM8AAAAhlHMdozxAAAAGEYx2zHGAwAAYBjFbMcYDwAA gGEUsx1jPAAAAIZRzHaM8QAAABhGMdsxxgMAAGAYxWzHGA8AAIBhFLMdYzwAAACGUcx2jPEAAAAY RjHbMcYDAABgGMVsxxgPAACAYRSzHWM8AAAAhlHMdozxAAAAGEYx2zHGAwAAmN///vd7jUwoZjvG eAAAAKMKzsjIhGK2Y4wHAAAwquCMjEwoZjvGeAAAAKMKzsjIZJXF7PPnzzfXrl27iBs3bmxev359 fu80b9682Vy/fn3n4wyhmAUAAEYVnJGRyeqK2cePH29OTk7Of7pclL58+fL81nF2nLKQtSiPuwSK WQAAYFTBGRmZrKqYtcLVXpWtvXr1anN0dLS5efPm5v379+e3ttlxHjx4sPnw4cP5LTEoZgEAgFEF Z2RksqpituXdu3ebO3fuTH6bwL17985eiV36bQU1ilkAAGBUwRkZmVDMnrMCdcors/4qbvn2guPj 45BXaSlmAQCAUQVnZGRCMfsv/srs3Pe8lu+bnfoWhTkoZgEAgFEFZ2RkQjH7L/Zq661bt7YqRr0Q toKWPwADAAARVMEZGZlQzP6LvcVgzicZ1PwTEbZ5dfbZs2cXg1DH119/LW8nCIIgCGJdoQrOyFA5 7BpW80Tw46+2mLW3Cjx58uT8p+3ZcficWQAAEEEVnJGRyaqLWfuYriUKWbPLWxVaKGYBAIBRBWdk ZLLaYtaKzy+//PL8p91ZYbz0pxpQzAIAAKMKzsjIZJXFbKvw3OXLEB4+fLj4Z85SzAIAAKMKzsjI ZHXFrBWy/nFaKvxtB/6xW/XbEOyPxepCeNc/IGuhmAUAAEYVnJGRyaqK2fJzYVWUf8DVKmZVMRxR yBqKWQAAYFTBGRmZrKqYzYbxAAAARhWckZEJxWzHGA8AAGBUwRkZmVDMdozxAAAARhWckZEJxWzH GA8AAGBUwRkZmVDMdozxAAAARhWckZEJxWzHGA8AAGBUwRkZmVDMdozxAAAARhWckZEJxWzHGA8A AGBUwRkZmVDMdozxAAAARhWckZEJxWzHGA8AAGBUwRkZmVDMdozxAAAARhWckZEJxWzHGA8AAGBU wRkZmVDMdozxAACsiSqqIiMTlX9kZEIx2zHGAwCwJqqoioxMVP6RkQnFbMcYDwDAmqiiKjIyUflH RiYUsx1jPAAAa6KKqsjIROUfGZlQzHaM8QAArIkqqiIjE5V/ZGRCMdsxxgMAsCaqqIqMTFT+kZEJ xWzHGA8AwJqooioyMlH5R0YmFLMdYzwAAGuiiqrIyETlHxmZUMx2jPEAAKyJKqoiIxOVf2RkQjHb McYDALAmqqiKjExU/pGRCcVsxxgPAMCaqKIqMjJR+UdGJhSzHWM8AABrooqqyMhE5R8ZmVDMdozx AACsiSqqIiMTlX9kZEIx2zHGAwCwJqqoioxMVP6RkQnFbMcYDwDAmqiiKjIyUflHRiZdFrNv3rzZ nJ6env+0XhSzAIA1UUVVZGSi8o+MTLotZq9fv765du3a5saNG5vXr1+f39OH58+fn+XmEZUjxSwA YE1UURUZmaj8IyOTbotZKxjNu3fvNnfu3DkrGm/evLl5//792e1X5fHjx5uTk5Pzny4X3i9fvjy/ dRkUswCANVFFVWRkovKPjEy6LGZbvHA8Pj7efPjw4fzW/SmL7NKrV682R0dHixfbFLMAgDVRRVVk ZKLyj4xM0hSz9+7du/Rf+2Us/YroXP7q8dJvN6CYBQCsiSqqIiMTlX9kZNJlMWuvgNofgJVvMaij fHXWCt2rerXWWQ68MgsAwPZUURUZmaj8IyOTbotZfx9qHeX7VUv2X/1ffvnl+U/75UV3K7dtUcwC ANZEFVWRkYnKPzIySVPMPnny5PzeT/lbEK7qD8SskL5169bi56aYBQCsiSqqIiMTlX9kZNJ9MTv2 ftjysVdVzFoxve37dp89e3YxCHV8/fXX8naCIAiCOMRQRVVkqBx6DZV/ZKgcdg2reSL48bsrZh88 eDD5PbD+yuzQq7dR7KO6os7by3gAALAPqqiKjExU/pGRSZfF7Ji5xW4U+5iuyAKaYhYAsCaqqIqM TFT+kZFJl8WsFau3b98efMtA/eUF+7aPPzijmAUArIkqqiIjE5V/ZGTSTTH7j3/84+KV1inFbNQf XU1hr8iqjwJb+hVjilkAwJqooioyMlH5R0Ym3RSz/i1a9t7XqXEVf/BlhazKxWPJtx1QzAIA1kQV VZGRico/MjLppph1/sdcU2LfbzOwtzaoPDz4BjAAALaniqrIyETlHxmZdFfMmikF7VV9DNc+UcwC ANZEFVWRkYnKPzIy6bKYNV988cXBF6tjKGYBAGuiiqrIyETlHxmZdFvMTnHoBS/FLABgTVRRFRmZ qPwjI5Nuitnyiw/sUwHqr7NVcehvNaCYBQCsiSqqIiMTlX9kZNJFMVsWr1ag/vjjjxSz/0IxCwBY E1VURUYmKv/IyKSLYtbwyuynKGYBAGuiiqrIyETlHxmZdFPMlqyYPT09Pf+pjffMAgBwOFRRFRmZ qPwjI5Nui1n7coIxFLMAABwOVVRFRiYq/8jIpNtiduhrYd+9e7e5c+cObzMAAOCAqKIqMjJR+UdG Jt0Ws/aeWVWsll+oQDELAMDhUEVVZGSi8o+MTLotZu1tBl7U2h+FqW8Fo5gFAOBwqKIqMjJR+UdG Jl0Ws6XWV9tagct7ZgEAOByqqIqMTFT+kZFJl8WsvyKritiTk5PzRx0+ilkAwJqooioyMlH5R0Ym aYpZeyXW2NsPbty4sXn9+jWvzAIAcEBUURUZmaj8IyOT7otZ9Uosn2YAAMDhUUVVZGSi8o+MTLot Zm/fvj1YqFpB+/nnn1PMAgBwIFRRFRmZqPwjI5Nui9kpX5rwpz/9iWIWAIADoYqqyMhE5R8ZmXRZ zOIjxgMAsCaqqIqMTFT+kZFJ18Wsvze2/AOwKW9BOBQUswCANVFFVWRkovKPjEy6LWZfvXq1OTo6 +uTTDEz5iQaHjGIWALAmqqiKjExU/pGRSZfFbPmKrCpmjX2ZwvHx8ebDhw/ntxweilkAwJqooioy MlH5R0YmXRaz9laCBw8eXBSq9irt06dPz/5t+GguAAAOjyqqIiMTlX9kZNJlMVt/7FZZzJafQUsx CwDA4VBFVWRkovKPjEy6LGbN48ePL73NQMWhf7UtxSwAYE1UURUZmaj8IyOTbotZ9b7ZMvgDMAAA DosqqiIjE5V/ZGTSbTHr7JML6kL20F+RdRSzAIA1UUVVZGSi8o+MTLovZteM8QAArIkqqiIjE5V/ ZGSSupj94osv+AMwAAAOhCqqIiMTlX9kZNJFMVt+QsGc2PXTDOyzau049WfYTqFyXvp9vBSzAIA1 UUVVZGSi8o+MTFZZzNbvw92mmFWftrD0e3kpZgEAa6KKqsjIROUfGZmssph1XtTOLWYt3/JLHaJQ zAIA1kQVVZGRico/MjLpppg9PT09/2m6Xd8zu20x629PiP54MIpZAMCaqKIqMjJR+UdGJl0Us0PU f+dv87YAZZti1r6N7Ojo6FI+x8fHIa/SUswCANZEFVWRkYnKPzIy6baYHXvrwRJfZbvtK7OuLLQj vlqXYhYAsCaqqIqMTFT+kZFJt8Ws/1f+UOz6B1e7FrOm/KYy/gAMAIDtqaIqMjJR+UdGJl0Ws/6q bOs9qV5AXtUfgNU8323yefbs2cUg1PH111/L2wmCIAjiEEMVVZGhcug1VP6RoXLYNazmieDH766Y /eUvfzn4x1X2mNu3b3dRzBp7ywGfMwsAwPZUURUZmaj8IyOTLotZf+X15cuX57d8yv4Q6/79+zv9 4dWSxazlc+vWrUXfN0sxCwBYE1VURUYmKv/IyKTLYtZYodn6b3srdu/evfvJq6B/+tOfZhWTSxaz dqylP9WAYhYAsCaqqIqMTFT+kZFJl8Xs2CcZtGLue1aXLGYfPny4+GfOUswCANZEFVWRkYnKPzIy oZj91/PUpxD4x27Vha59ykL9CqzdNvSWiG1RzAIA1kQVVZGRico/MjJZZTGrvvjAoixcW8WsF8Bl RBSyhmIWALAmqqiKjExU/pGRSbfF7FV8vW1vKGYBAGuiiqrIyETlHxmZdFnMTnVoxWuNYhYAsCaq qIqMTFT+kZFJt8Vs+c1ardjmSwoyoZgFAKyJKqoiIxOVf2Rk0mUxO6WQtaCYBQDgcKiiKjIyUflH RiZdFrNT/wCMYhYAkI0qHKIjC5V7ZGSi8o+MTLouZsc+/5X3zAIAslGFQ3RkoXKPjExU/pGRSZfF rLGPxlriywwyo5gFgMOjCofoyELlHhmZqPwjI5Nui9nWV9aWeGUWAJCNKhyiIwuVe2RkovKPjEy6 Lmb5NAOKWQA4NKpwiI4sVO6RkYnKPzIy6baYta+IVQVsGRSzAIBsVOEQHVmo3CMjE5V/ZGTSZTHL pxl8RDELAIdHFQ7RkYXKPTIyUflHRiZdF7N8mgHFLAAcGlU4REcWKvfIyETlHxmZdFnMGnubwcuX L89/WieKWQA4PKpwiI4sVO6RkYnKPzIy6baYtT8A+/zzzwdfeeWVWQBANqpwiI4sVO6RkYnKPzIy 6bKY5T2zH1HMAsDhUYVDdGShco+MTFT+kZEJxWzHKGYB4PCowiE6slC5R0YmKv/IyIRitmMUswBw eFThEB1ZqNwjIxOVf2Rk0m0xe3p6ev5TG++ZBQBkowqH6MhC5R4Zmaj8IyOTLotZfMR4AMDhUYVD dGShco+MTFT+kZFJ98Xs48ePP3l7wdjnzx4KilkAODyqcIiOLFTukZGJyj8yMum2mLWP5rpz584n hazH8fHx5sOHD+ePPkwUswBweFThEB1ZqNwjIxOVf2Rk0m0xq16RrePQX6GlmAWAw6MKh+jIQuUe GZmo/CMjky6LWXtV9u7du5vXr1+f3/KpV69ebe7fv3/Qr85SzALA4VGFQ3RkoXKPjExU/pGRSZfF LJ9m8BHFLAAcHlU4REcWKvfIyETlHxmZdFnM8srsRxSzAHB4VOEQHVmo3CMjE5V/ZGTSZTFrprxn 9uTk5PzRh4liFgAOjyocoiMLlXtkZKLyj4xMui1m+TQDilkAOESqcIiOLFTukZGJyj8yMum2mHXP nz//pJB9+fLl+b2HjWIWAA6PKhyiIwuVe2RkovKPjEy6L2Z7du/evbPiOuojwhgPADg8qnCIjixU 7pGRico/MjKhmN1C/WoxxSwAYCpVOERHFir3yMhE5R8ZmXRRzNpHcV2/fv1Sgdh6T6wVkjdu3Bj8 pIN98aKWYhYAMJUqHKIjC5V7ZGSi8o+MTLooZo191NbR0dFZcTj2nlh77K1bt678M2YpZgEAc6nC ITqyULlHRiYq/8jIpKtids7nxtpHd13119lSzAIA5lKFQ3RkoXKPjExU/pGRSTfF7MOHD2e9daCH L02gmAUAzKUKh+jIQuUeGZmo/CMjky6KWftM2UePHs0qTO19trdv377StxpQzAIA5lKFQ3RkoXKP jExU/pGRSRfFrBWmp6en5z9NcyjF7LNnzy4GoY6vv/5a3k4QBEHkDVU4RIfKo8dQuUeGyqHXUPlH hsph17CaJ4If/8qL2QcPHsx6ZbaHPwLjlVkAwFyqcIiOLFTukZGJyj8yMumimLW3Gdy9e3fWe2bt D8Cu+ittKWYBAHOpwiE6slC5R0YmKv/IyKSLYtZYcXrz5s1Jr7T6x3hFFZFTUcwCgKYujtGRhco9 OrJQuUdGJir/yMikm2K2/OKEk5OT81svKx8ztfCNRDELAJq6OEZHFir36MhC5R4Zmaj8IyOTbopZ 48XhlLjqV2WN59sqvndFMQsgK3VxjI4sVO7RkYXKPTIyUflHRiZdFbPG30JQF69lXHUh28px6bwo ZgFkpS6O0ZGFyj06slC5R0YmKv/IyKS7YtapV2mv+g++9o1iFkBW6uIYHVmo3KMjC5V7ZGSi8o+M TLotZkExCyAvdXGMjixU7tGRhco9MjJR+UdGJhSzHWM8AGSlLo7RkYXKPTqyULlHRiYq/8jIhGK2 Y4wHgKzUxTE6slC5R0cWKvfIyETlHxmZUMx2jPEAkJW6OEZHFir36MhC5R4Zmaj8IyMTitmOMR4A slIXx+jIQuUeHVmo3CMjE5V/ZGRCMdsxxgNAVuriGB1ZqNyjIwuVe2RkovKPjEwoZjvGeADISl0c oyMLlXt0ZKFyj4xMVP6RkQnFbMcYDwBZqYtjdGShco+OLFTukZGJyj8yMqGY7RjjASArdXGMjixU 7tGRhco9MjJR+UdGJhSzHWM8AGSlLo7RkYXKPTqyULlHRiYq/8jIhGK2Y4wH0D91EYiMLFTu0ZGF yj06slC5R0YmKv/IyIRitmOMB9A/dRGIjCxU7tGRhco9OrJQuUdGJir/yMiEYrZjjAfQP3URiIws VO7RkYXKPTqyULlHRiYq/8jIhGK2Y4wH0D91EYiMLFTu0ZGFyj06slC5R0YmKv/IyIRitmOMB9A/ dRGIjCxU7tGRhco9OrJQuUdGJir/yMiEYrZjjAfQP3URiIwsVO7RkYXKPTqyULlHRiYq/8jIhGK2 Y4wH0D91EYiMLFTu0ZGFyj06slC5R0YmKv/IyIRitmOMB9A/dRGIjCxU7tGRhco9OrJQuUdGJir/ yMiEYrZjjAfQP3URiIwsVO7RkYXKPTqyULlHRiYq/8jIhGK2Y4wH0D91EYiMLFTu0ZGFyj06slC5 R0YmKv/IyIRitmOMB9A/dRGIjCxU7tGRhco9OrJQuUdGJir/yMiEYrZjjAfQP3URiIwsVO7RkYXK PTqyULlHRiYq/8jIhGK2Y4wH0D91EYiMLFTu0ZGFyj06slC5R0YmKv/IyIRitmOMB9A/dRGIjCxU 7tGRhco9OrJQuUdGJir/yMiEYrZjjAfQP3URiIwsVO7RkYXKPTqyULlHRiYq/8jIhGK2Y4wH0D91 EYiMLFTu0ZGFyj06slC5R0YmKv/IyIRitmOMB9A/dRGIjCxU7tGRhco9OrJQuUdGJir/yMiEYrZj jAfQP3URiIwsVO7RkYXKPTqyULlHRiYq/8jIZLXF7KtXrzZHR0eba9eubW7cuLF5/fr1+T3TvHnz ZnP9+vWz53tsc5whFLNA/9RFIDKyULlHRxYq9+jIQuUeGZmo/CMjk1UWs8+fP79UeHph+/Lly7Of p3j8+PGlQtbi5OTk/N5lUMwC/VMXgcjIQuUeHVmo3KMjC5V7ZGSi8o+MTFZXzPorqk+ePDm/5SMr Tm/evLl5//79+S1tdowHDx5sPnz4cH5LDIpZoH/qIhAZWajcoyMLlXt0ZKFyj4xMVP6Rkcnqilkr WtXbAfzV2brIVe7duxfytoIaxSzQP3URiIwsVO7RkYXKPTqyULlHRiYq/8jIZFXF7Lt37zZ37tzZ HB8ff/Kqqr9iO/ZWgfK9th7qeEugmAX6py4CkZGFyj06slC5R0cWKvfIyETlHxmZrKqY9YJVFZ9D hW5L+b7ZqW9RmINiFuifughERhYq9+jIQuUeHVmo3CMjE5V/ZGSyqmLWX1VVr756MTu3KPXnWUG7 zR+APXv27GIQ6vj666/l7QRB9BPqIhAZKoceQ+UeHSqPHkPlHh0qjx5D5R4ZKodeQ+UfGSqHXcNq ngh+fIrZLYtZ46/4Lv3q7KGPB3AI1EUgMrJQuUdHFir36MhC5R4Zmaj8IyOTVRWzQ++LHXoLwhSt PyzbBcUs0D91EYiMLFTu0ZGFyj06slC5R0YmKv/IyGRVxewSfwDWYq/63rp1i1dmgZVRF4HIyELl Hh1ZqNyjIwuVe2RkovKPjExWVcya1ufJ+lsQ5nxxQsm+iGHpTzWgmAX6py4CkZGFyj06slC5R0cW KvfIyETlHxmZrK6Y9aJ1ly9NUB4+fLj4Z85SzAL9UxeByMhC5R4dWajcoyMLlXtkZKLyj4xMVlfM mvr9rfaqqn0aQfmqrH/sVl302hcm1K/A2m3bvqI7hGIWvVAbXWRkovKPjCxU7tGRhco9OrJQuUdG Jir/yMhklcWs8QK29RmxrWK2fJ5HRCFrKGbRC7XRRUYmKv/IyELlHh1ZqNyjIwuVe2RkovKPjExW W8xmwHigF2qji4xMVP6RkYXKPTqyULlHRxYq98jIROUfGZlQzHaM8UAv1EYXGZmo/CMjC5V7dGSh co+OLFTukZGJyj8yMqGY7RjjgV6ojS4yMlH5R0YWKvfoyELlHh1ZqNwjIxOVf2RkQjHbMcYDvVAb XWRkovKPjCxU7tGRhco9OrJQuUdGJir/yMiEYrZjjAd6oTa6yMhE5R8ZWajcoyMLlXt0ZKFyj4xM VP6RkQnFbMcYD/RCbXSRkYnKPzKyULlHRxYq9+jIQuUeGZmo/CMjE4rZjjEe6IXa6CIjE5V/ZGSh co+OLFTu0ZGFyj0yMlH5R0YmFLMdYzzQC7XRRUYmKv/IyELlHh1ZqNyjIwuVe2RkovKPjEwoZjvG eKAXaqOLjExU/pGRhco9OrJQuUdHFir3yMhE5R8ZmVDMdozxQC/URhcZmaj8IyMLlXt0ZKFyj44s VO6RkYnKPzIyoZjtGOOBXqiNLjIyUflHRhYq9+jIQuUeHVmo3CMjE5V/ZGRCMdsxxgO9UBtdZGSi 8o+MLFTu0ZGFyj06slC5R0YmKv/IyIRitmOMx36pxRwZmaj8IyMTlX9kZKFyj44sVO7RkYXKPTIy UflHRiYUsx1jPPZLLebIyETlHxmZqPwjIwuVe3RkoXKPjixU7pGRico/MjKhmO0Y47FfajFHRiYq /8jIROUfGVmo3KMjC5V7dGShco+MTFT+kZEJxWzHGI/9Uos5MjJR+UdGJir/yMhC5R4dWajcoyML lXtkZKLyj4xMKGY7xnjsl1rMkZGJyj8yMlH5R0YWKvfoyELlHh1ZqNwjIxOVf2RkQjHbMcZjv9Ri joxMVP6RkYnKPzKyULlHRxYq9+jIQuUeGZmo/CMjE4rZjjEe+6UWc2RkovKPjExU/pGRhco9OrJQ uUdHFir3yMhE5R8ZmVDMdozx2C+1mCMjE5V/ZGSi8o+MLFTu0ZGFyj06slC5R0YmKv/IyIRitmOM x36pxRwZmaj8IyMTlX9kZKFyj44sVO7RkYXKPTIyUflHRiYUsx1jPPZLLebIyETlHxmZqPwjIwuV e3RkoXKPjixU7pGRico/MjKhmO0Y47FfajFHRiYq/8jIROUfGVmo3KMjC5V7dGShco+MTFT+kZEJ xWzHGI/9Uos5MjJR+UdGJir/yMhC5R4dWajcoyMLlXtkZKLyj4xMKGY7xnjsl1rMkZGJyj8yMlH5 R0YWKvfoyELlHh1ZqNwjIxOVf2RkQjHbMcZjv9RijoxMVP6RkYnKPzKyULlHRxYq9+jIQuUeGZmo /CMjE4rZjjEe+6UWc2RkovKPjExU/pGRhco9OrJQuUdHFir3yMhE5R8ZmVDMdozx2C+1mCMjE5V/ ZGSi8o+MLFTu0ZGFyj06slC5R0YmKv/IyIRidkuvXr3aHB0dba5du7a5cePG5vXr1+f3LIfx2C+1 mCMjE5V/ZGSi8o+MLFTu0ZGFyj06slC5R0YmKv/IyIRidgvPnz+/VMB6Yfvy5cuzn5cSMR5qwkZH Fir3yMhE5R8Zmaj8IyMLlXt0ZKFyj44sVO6RkYnKPzIyoZid6c2bN5vr169vnjx5cn7LR48fP97c vHlz8/79+/Nbdkcxu18q98jIROUfGZmo/CMjC5V7dGShco+OLFTukZGJyj8yMqGYncmKVvW2An91 ti5yd0Exu18q98jIROUfGZmo/CMjC5V7dGShco+OLFTukZGJyj8yMqGYneHdu3ebO3fubI6Pjzcf Pnw4v/Ujf8X25OTk/JbdUczul8o9MjJR+UdGJir/yMhC5R4dWajcoyMLlXtkZKLyj4xMKGZn8IJV FbNDhe62KGb3S+UeGZmo/CMjE5V/ZGShco+OLFTu0ZGFyj0yMlH5R0YmFLMz+FsJ1KuvXszOfd/s s2fPLgahjj/+8Y/ydoIgCIIgiGxhNU8EPz7F7AQRxSwAAAC2RzE7w9D7YofeggAAAIAYFLMz7PsP wAAAADCMYnam1ufJRn1xAgAAANooZmdqfZ5sxJcmAAAAYBjF7BbqL06wr7e9du0ar8oCAADsGcXs lryAteAVWQAAgKtBMQsAAIC0KGYBAACQFsUsAAAA0qKYBQAAQFoUswAAAEiLYhYAAABpUcwCAAAg LYpZAAAApEUxe4Dqbyjbl3v37l3Jebdl+fKFF/04hPFgTuEqZNt7gaVRzCbz6tWrzdHR0cW3jx0f H28+fPiw+eKLLy4uoPsuZt+9e7e5c+fOWT5Ln7ds11RjXy9c5kvh8dGbN282169fv5hXKqL6at/j UZ6vjCdPnpw/Yr6y/w51TvG13ePKb4YciyX6MXLv3Se7ZtX98+LFi0vr1K91LfW69sfve0yW0GpL y9zHz7XNdXjfKGYT8UK2vOj6bT1cQJcuoq1AuH379ux22asUtqBPTk7Ob9EsX4rZy9QcM+UFYaxf t7Xv8Wi1dReH/Mrs1HW1ZrZOykLCiwx125KF075fwIjgvxDWhVj5As5Qn/nj6n64qjFZwq5tX8K2 1+F9o5hNxC4m9UI3vUy2pTfUbYobW9T379/f/Nu//dtoLhSzn/JNURV45W//SxaA7hCK2UOdU3PW 1Zp99913l/pGFU7G+vP7778//2l3S++9V8H7qv5lyfrq97//vSx0Sw8fPtw8ePDgk364qjFZwq5t X0KWPY1iNhF/ZUT9hlb+N4A9Tk3q8lUjP5Y/rixU6omrnmdRbzqtDdVu9+fUxy7/e9biL3/5y+b0 9PTSedTzWux5Vpz4K4lDhUq9SFXb6vz88WV/WZQbjR+n3nzm9sO///u/X9zmj7fNzX9Tj3iFzI/f 6je/3/Mp856SY6s/TTke3ofl/Usba6uZ2z7VBovyHEN94Oy5U/qinFPqYjflXFNYDmPrquyXOh/V hqH14O1/+/bt2TrzfWVqe8b6xQydfymtwskNtWfOHmNtUXtvffyh60Ld1/vm7a3XlM2rL7/88mK8 1PXPHmO/bH311Vej+Y+NSU+WavvYPKjv3/U6fBUoZhMp/6tXLcRy8ysna3m7Tcbf/va3Z4vCJ/Bv fvObza9//euzx/tttqG0nmd8cZUbj9pQ7TZ/jB/bF4Qd/+7du5ceb4vHH2/PnbN47Pj+CrWfa2jD qo9vm8KtW7cufi77wtj9drGu21Ofo8zD7dIPfrFp5bUkb2OrwCvnRLmxTsnRjv2LX/ziop1+rvKC bMfweTaWy67mHH/qGFgbbA189tlncq3Uz/Mc/OextVqer56/lqM9b+q5prLjTF1XrfvrOd5aD15Q efvt1aZvv/32LG+7wE5pz1i/mNb5/TlL8fFU/eXnHWpPqz/t9nKPsfbUe68fz+dhvXbruVb2tT9n nzyfejytHVbQeXtUX1r7LWfVD7WhMenNEm2fMg+WvA5fFYrZZGxi+cZs0ZrcalL7c8uNyiZt/Vi7 rTyuep4viHKS1+etN1xTHss36vK4dpv9RmjssXMWkT2+3AhV20rl8W3B22+3ZV96fl7sqE3QfsGo z+EbkNu1H6wdZT94HvWmvwTf+IYKPMvHci8fM5aj/1wft3xe2Scmsp1mSlvd1DGwNqg5Z8+32+2/ McfmlGn1hT9Onb++aE2Zv1NYLuV5vC1D62poTYytB/WzmdKeqf0ydv6lDPX51PGZssfUfa76wfg5 y7kc1fa5WjmXbVVzz9pkhbj1Wd0PytCY9GbXtk+ZBz/++OPZv+u1tu11+KpQzCbkE9Q2II9ysrYW tLq9vkgbu61c6K3JbLeXm2B9/PKV5Dp8A7dz2c+t409dRNYn9W+Xfv56ITs//p///OfRjc0Xvx2v fKzaLOz9SyoPFVP6oR6j1ga1BL/we16K51o+ZizHuqBQ6vH2Y0RddKa01U0dg9ac9TlQXzDUnDLq OJaDP071zVB7hs41xM4zd135ucr7v/nmm1n7wtjaH1uPQ/0y5fxLmTqHh8ZHzbV6j7H+Kvfeobng 67fct4f6el9UO421xQs6b1f5GMu/nDfq2leaOiY92LXtU+eB/1vNg17mxxiK2eR8ElqMLWh1uz1/ 6KJpWpO5vkDXx7f7pywCe563oXx+67zK0AWqdYzWeUt+kbFj+H+B1ptg2W57vP+m7Hbth3qMWpv+ EoY2P+PnthzLwmwsRz9u+ZxaPd7RF52xtpamjkFrzpZrZeqcqo9Tr0u1/upxmXKuIdusK1P2l+Xg r/KYKeuh1Y9T2jPWL1POv5SxOTx1fKwNQ3tMeb/xPlBzu+6PVl/vW2tN2TotX4W2uVX2xdjbLWrR +8qSdm373HlgP1uUx+llfoyhmE2k9VlvflH2xdla0Or2+iJt6otmazLbQqknfflzff8Ye74tJD9X 67w135zKi7jzY7bus+O3LiJ+ofGNoLUJ+uNsE7Zj1hvHrv1Qj1Fr01/CWIHn99fjMpajP28o53q8 oy86Y20t/wp66hjUbXA+B+a8zUD1cf04u80vQPU8nzp/W/zxc9eVKfvWHls+bsp6UO2f056hfpm7 HncxlOOc9oztMXZb2SZro7Vbrbf6saqvr4K3v87Z5lJZ0JV7ibWz7Iu6bcpQP/dm17bPmQfObrfn +JzoZX6MoZhNxDZodfGoN4GhSVrfXl+kTX3RbB1v7HHlBlyyBfn06dOz+8tXbEx5oZm6iOx49ftd Xbn4a+Xx/XHl5lDmYoY2QTtWuQGUdu2HeowiN2PVD87bURcHZixH/9meWx/b2mrHq8c7sp1mqK12 7kePHl2cd+oYtOasrxUb7ylzSh1HrTe1H7g581ex/tlmXTnL187/H//xH5eOMbYejGr/1PaM9cuU 8y9lqM/njo+1q7XH2H3lsbyNQ3PRz6H6+ip4+9W4lAWdsTZYX5RtNnU/KHPXwVXate1T5sHPP/+8 yHX4qlHMJuKTuL741pO4taDV7XbMeqL6JC83OztvucnYZLfbyouGOn79XFuc/okBvtDqAtLzKY9X f1ag842pdfHy++tcjR2/bLvn6o/zNnp+/rP1zQ8//LD56aefzm43rQukW6ofjM8Di6U3GctLFXje dov6PjMlx/IYZZTtLNsTfdFptdVY/uVYTh2D+nGmbNfUOVX3hSnXZTmv6/DHTD2Xssu6cvX5S9Y+ u0+tB6PaP6U9U/rFjJ1/Kb62y3O7ueMztMdYe+p5Vx7Pz23tLD9RxKi+vgqtvrKc64LO127dF6of akNj0psl2j42D+Zcf1rX4R5QzCbibzOwyWWT06PciMqLbDmxy9st6q8KtMf+/e9/v3RbOZnt3/Yx QeXjy0ld51Re4Mr7ylxtEdmiaRUGvsjsdrWBl/db1JtTfb8/5p///OcnFzzLV+VZ5mY5+M9qI6z/ KKO2bT/UF2gfuyU3Y9VXdQydb2qOvhH741Q/W/jHMPnPS7Z1qOApo9zc54xB3ZZ67o7NqbG+8D7z i5QKz33sXMq266o+nvXZ559/fjGPa631ULe/3EumtGdKv5jW+ZfQmmNz50JN7TFlOyzK/mqtNzfU 1/tUt8GivkZZeH7Wv+UfJtbttKj7cOqY9EDlukvbh+bB2PWnXO899pWjmMUo22iW3uwPjS34+o8y gEjlJwSU7MJVv5qzJofaL+wxQBvFLEZRzI6zPipf9QEi2SsprVdJrGjr7Ws59+WQ+4U9BmijmMUo ilmt/O9M+gf7ZGvS5l3938Jrf/Xu0PqFPQaYhmIWTUPv28H/XWi4yOAqlIWOB3PxsPqFPQaYhmIW AAAAaVHMAgAAIC2KWQAAAKRFMQsAAIC0KGYBAACQFsUssAD7SCD1VYJD7BtX5j4nI/qmjb7BXMyZ NvpmOdn6hWL2ANmCrj+aRn2kVutx9pmMp6enZ49RH3MzJ8oP+a6P1frKxn2Z07axjySbs4mWH3nW 62ZB37TRN/+HvWYa5kzbmvpmyjq4ahnmjEIxe6Bscdj3KatN3L+nWU1U+23MJrF/i45tNOUxfKKr 23xR2nlbi8DPPbYp7cPcti1tzsa7b/RNG31zGXvNOOZM2xr6Zs466EXPc0ahmD1QvvjrBeKLqtwo auXm8t13312azGqjMXZc/6rIoXPYwu3h4mK2aduSet4s6Js2+uYy9ppxzJm2Q++bueugFz3PGYVi 9kC1LjD+W+DQJm/PffTokVxYrY2mZgvBzlP/119vv32WxtpmfWffxPP27duzx5UL3e6rF76/cmT9 YFF/i0+mzYK+aVt733j72WumYz21HVrfbLsO6rzrHL0frC1+Dn+M96Fqr3qeRb1eWv3i6608dnk+ v93Hp3xcJIrZA+WTq5ygvjh2mVhjG43zx/m57Ny3b98On9C7aLWtXKjWHvuO92+//fbst+0XL15c 3FcufO9r73//7bwcj0O4wNA39I23s8yRvWbY2ufMkEPqm23XgefpBXDZ9rKtdtzf/va3Z4/zc/3m N7/Z/PrXvz5rT9n+uv/8ecb6wG4f6xe7zR+j2ubH8eNaO+7fvz+4fpdCMXugfOKWk9MXyNjFYUhr o1Hs1RFfIPYb4NBvpj0Ya1u9UEv1wveF7q8WqWOrzaJX9E3b2vvGc2SvmW7tc2bIIfXNNuvAcyzX k/G2ePGo+sHmft0Wu61ub/08P2ddmNZ9Wf+SWB/Lj2PPs7eE2C8cU9u9K4rZA6UWhG/4cxZWTW0G Q/y/MeqF2aOxttnCLRd7qV74Jd+E6r4fek5v6Ju2tfcNe818rKe2Q+qbbdaBF8BegJd8jlvxqPK2 ++u+UcWs6j+73Y/tP5fH97aoKHP1fm6NURSK2QOlLjC+SHaZZHMvMH7O8rfAXi29iZaL2t8/tK9N dGn0Tdva+4a9Zj7WU9sh9c0268CLRlXMWq52X0Qx6+cdKmantmOoDVEoZg+UusD4beWEnYsLzLRN 1DfQq/rvraXRN21r7xvPkb1murXPmSGH1Dd+vjnrwAtB9T8MZa4q712L2fJ49fHr+1tsHX755Zdn 553y+KVQzB4oX0T1gvCF0too3MOHD+UkHNtoamu9wNQLf9+b6NLom7a1943nyF4z3drnzJBD65u5 68ALcNXGsjBVeU8tZlV7xx7nedXr3Nbd06dPz/5tj/H3yc5dv7uimD1QPvHURPLFpSa0P691QRg6 ruIXGP/NuGdjbbPFPWcTtT72dpcb2g8//LD56aefmptKj+ibtrX3zVD7PX+Vrz+PvYb1VDrEvpm7 Dso8vQ9sfv/iF7+4eL7Ke2oxa8cui1I/X5mDOn79XMvp1q1bZ+ez4vXu3buXHu/HrQvgCBSzB8gn XBn1BcN/a6of19okWo9vTVL1+OgNY1tT2mYbQnlf2Z/lfWUby9vtWP6zbSxfffXVxX0WrQv6VaNv 2ugb9pq5mDNth943c9eB/3KmHlf3Q/lxXRbW/r///e+XbvPn25q1f9tHc5WPL9dLva7LfinvUzn5 bV6k+2Otv6f8UrotilkAAIAV8GJWFdCZUcwCAACsAMUsAAAA0qKYBQAAQDrqPbu9vn96GxSzAAAA SItiFgAAAGlRzAIAACAtilkAAACkRTGLWewvIed+ILl9mPLc5yzJPnz6/v37oR/YvCR7o/7nn38e 9tem2fqjJbqfWnrov6tq+z6x1+zPGuYTDhvF7Ar4V8pNibG/bpxzgSn/evKqLjDlx5DYv6e0t/U4 +0aT09PTs8fM6dOxUF+/6d/+svRfm3p//Pjjj5e+naUVQ18NOqWfSnMfb8aeE9VPLd5/S8+n2pTn 7LvtU7DXXP7IIx8jb7N/C9IXX3zRLBynzpexY5d9UkZrTfc4n9aoNf61KfNk1+tUOVfqY0V/o9dc FLMrYJOwnHi+yanbIjayORelJdXtNrbQW9/77Zu5ytW/qs+/WnFun/p5y69mNH5clY8/Z6kxUf3h bVYXOHt8na+Z009m7uPNnOcs3U8tkfPJzX3Ovto+1dx1sbSe9hofy3Jt+W3qcz7njP2cY6vHtvQ2 n9ZqqX1l7nr087bWkJ+7x/lBMbsC33333aWJqSa0sYn6/fffn/+0nKu4wLQ2ZW97q4hQm4crN4a5 fer51Of159kGpDYIO6e68M3V6g9vt7rQ2XPqVw7n9tPcx5ttnrNUP7VEzyezzXNMdNvnYK/5P1ZY qLG0x9++fVsWnFPHfptjTylmTU/zaa2u6jrl/26dw+aduk71gGJ2hVoT2tmEtc3s7du3Z48rLw52 X32x8M3cCjKLeiO8iguMnVO1r7VJWLtaBaWz5z569Ej22Vifeh/V5zWWa+vcftypF6KWVn+0LnTl 5laa20/b9Os2z1mqn1r2MZ+2nYPRbd+F57bGvWZoPOu3Gcwd+znHbq3xlp7n01r4GOz7OuX8mlTO ASuS1fWrFxSzK9Sa0H67TWK7SDx48GDz7bffnm2EL168uLivvFjURZpvnOWk3/cFprURGHWft6G+ MM4xtkm0ill/3lD/2AY2tvkMGeqP1oXOxqzeMOf20zb9ustY7NpPLfuYT7u020S1fVetdeG3H/Je Yxd/a4PF0NhsM/ZTj23mFrOm1/m0FkvtK7XWeqz54/xcdu76Ff/eUMyu0NiE9t/K1G9/9cXCF5hv lOrY+77A+Oat8lebhD9+l817rE/rC7Hx26yvhy40u/bfUH/4fX5hLKN+/Nx+2qZfdxmLqHk21H9L zadd2m32vcamWvNeY7x9Hqofth37Kcc2fvw5xey++xGXXdV1quS/MFkO9stNa473gmJ2haZcYFq/ /Q1tcmVxVh573xujL8KpxYc/PnKTKPumjCkXmKH2TDH0/NaFzsasfvzcftqmX3cZi137qWXouEvN p13abaLavqs17zXO+8Ae57HU/jN2bLNNMdvrfFqLpfaV2th6rFkRq+ZUjyhmV2jpC4xfWOw5/t63 Xi8wapPwzb7V5inG+tT7yM/rj59yzl0vLEPPb13o7Pb6PbNz+2mbft1lLKIuwPuYT7u020S1fVdr 3msULw4sfM3tOvZOHdv48Slm87iq61TNz5lhHlDMrtCSFxi/uPhGqY7d0wVGbRJ+2y6b91if1sWs 8dvGNpZdLyxDz59zoZvbT9v06y5jEXUB3sd82qXdJqrtu/J2rXGvaX2WrK85z3ubsZ967PK21hqv /+Ld9Dqf1sLnxL6vUzWKWXRtbELPucDYplf+rI697wvM0OatNgnjm/fYIn/48KFsx1ifqmLWeK71 7aVd+2+oP8YudHb/06dPz3+a30/b9Ou2YxE1z/Y1n3aZg/teY1Otea9pvc9QzZm5Yz/n2GPzV/3l e6/zaS3UOJpd9gij1swQnzsUs+jS2CuCcy8wtrh8oywX2w8//LD56aef9r4xtjYCM9R2z13l6s9r LeqxPp1y3laf24Wrddwphvpj6EJn9926deuTnOb20zb9us1zdu2nln3Op22eY6Lavquh/jGHvNfY mJT5ulaOc8Z+zrHHCu5W7j3Op7VYel9xY+uxNjR3ekMxuyK+8dpCKKPczHyT9CgXRXlfuZDK2+1Y /rMtmK+++uriPovWIluaber1grXbylxUPq0+al1wx/q0db+6CPl95bn8+btuJnV/+KZW5qRCXejM Uv3UeryZ85yl+qllX/PJbNu3PV1wWm1Yy15j/K0A9TxZYuynHLt1rDrqedPjfFqTiH1lynosqcer 4rknFLM4SGO/oWZhv4UPXfymOpT+aFmqn1p67r/otmPYoa0t5hMyopjFwcq+KS99kTzUi9S+ioke ++/Qf0nJ4lDWFvMJWVHM4qDZRab3/x5Roi4qWfujZd8X3576j8KjL9nXFvMJmVHM4uDZJm1flznl De89sPcr3b17N+yimK0/WqL7qaWH/ruqtmNY1rXFfEJ2FLMAAABIi2IWAAAAaVHMAgAAIC2KWQAA AKRFMQsAAIC0KGYBAAAOmH2z2NyPjrNv2MvycXMUswAAAJ2xzy4uv1J2KMY+H3hOMVt+nS3FLAAA ALZixezx8fHF5xZ7kalui/iyi21ezb0qFLMAAACd+e677y4VkqqYNa9evdp8//335z8th2IWAAAA i2kVs87e43rz5s3N27dvzx5XFqLq/a/+Fcb+VgV77vv378/vpZgFAADAglrFrN/uBal9pfK33367 OTo62rx48eLivrIw9UL25OTk7Gd7ddce7z8bilkAAAAsZuyVWSs+rWhV75+tC1MvZp88eXL2szo2 xSwAAAAWM6WYrd8q4IYK0/LtBhSzAAAACLF0MetFbPk+W4pZAAAAhFiymOVtBgAAANirJYtZ+wzb 8meKWQAAAITyV1OXKmbtPbL+yqz/bMf+4YcfNj/99BPFLAAAAHbnr5pasVlG+TFa9jmy5X3lJxqU 95XFaXm7Hct/toL2q6++urjPIuIbxpZEMQsAAIC0KGYBAACQFsUsAAAA0qKYBQAAQFoUswAAAEiL YhYAAABpUcwCAAAgLYpZAAAApEUxCwAAgLQoZgEAAJAWxSwAAADSopgFAABAWhSzAAAASItiFgAA AGlRzAIAACAtilkAAACk9UkxSxAEQRAEQRDZworZ/1/fSBAEQRAEQRC9x3/913/9438BEIKnU6dL 1zwAAAAASUVORK5CYIJ=

B. Complexity of Domestic Criminal Trials

This Article’s methodology also allows for a comparison of the complexity of international trials and domestic trials in the United States. According to the Administrative Office of the United States Courts, there were 8453 criminal cases completed in the US federal district courts during the period September 2010 to September 2011. 162See Hogan, supra note 133, at tbl. T-2. Of these trials, 5779 (68%) lasted one day or less. Only 37—less than half of one percent—lasted more than 20 days. 163Id. No data was available on the average number of witnesses or exhibits introduced during those trials, making it impossible to calculate Complexity Scores for them. Nevertheless, it is clear that the average federal trial is nowhere near as complex as the cases that come before the ICTY.

A dataset does exist that contains data on all three components of the Complexity Score for a sample of domestic criminal prosecutions. It was compiled by Professors Larry Heuer and Steven Penrod in the early 1990s and consists of information on 160 random criminal trials conducted at both the federal and state level in 33 states. 164Heuer & Penrod, supra note 33, at 34–35. Among the information they collected was the number of witnesses who testified, the trial length, 165They collected the information on trial length in days and hours and then converted them to trial hours at the rate of four hours per day for criminal trials. Since trial days were used rather than trial hours in this study, trial hours were converted back to days using the same conversion rate. Id. at tbl. 1 (explanatory note). This may not be an entirely accurate comparison as the ICTY defaults to five courtroom hours per day and, in practice, often exceeds this. See U.N. Secretary-General, Budget for the International Tribunal for the Prosecution of Persons Responsible for Serious Violations of International Humanitarian Law Committed in the Territory of the Former Yugoslavia since 1991, for the biennium 2012–2013, 31, U.N. Doc. A/66/386 (Sept. 29, 2011) [hereinafter Budget for the International Tribunal]. and the number of exhibits entered into evidence. 166Heuer & Penrod, supra note 33, at tbl.1. Complexity Scores were calculated for the cases in their dataset.

The least complex case in Heuer and Penrod’s dataset has a Complexity Score of 0.007, making it ten times less complicated than the least complicated ICTY case. The median trial in their dataset had a Complexity Score of 0.028, while the average case had a Complexity Score of 0.043. Both of these scores are lower than the least complex ICTY case. Only the most complicated case in their dataset had a Complexity Score higher than the least complex ICTY case, with a score of 0.20. However, this case was still significantly less complex than even the second least complicated ICTY case, and nearly five times less complex than the average ICTY trial. Other datasets on domestic criminal trials yield similar results. 167According to one study, the average non-capital murder trial in North Carolina in the early 1990s took 3.8 days. See Philip J. Cook & Donna Slawson, The Cost of Processing Murder Cases in North Carolina tbl. 6 (1993). The study did not contain data on the number of exhibits or witnesses, but when comparing the average length of a murder trial in that study to the average length of an ICTY trial, the ICTY case takes 45 times longer to present than the murder trial. See also infra Part IV.C.2.e–g. As can be seen from Figure 3 below, the average trial in the Heuer and Penrod dataset barely even registers when placed on the same chart as the ICTY’s cases. 168The categories with an asterisk in Figure 3 are based on the data from the Heuer and Penrod dataset. See Heuer & Penrod, supra note 33.

data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAqgAAAG/CAYAAACUgyXiAAAAAXNSR0IArs4c6QAAAARnQU1BAACx jwv8YQUAAAAJcEhZcwAAFxEAABcRAcom8z8AAEGsSURBVHhe7d09qFVblv99o5MYmVWimVAYGt7Y sA4omFiZJh10IJxEpEED6bboqqizorEeuGAZFBXdpNLLoXg0brMGG4XbgdzkBmJgsP81jnucHmc6 5nrfY8+x1/cDA3S/rLXmXHPN9Ttrv136t3/7t+f/qA1FURRFURRFNVD/nwRU7w6KoiiKoiiK2kud B1QAAABgnwioAAAAaAoBFQAAAE0hoAIAAKApBFQAAAA0hYAKAACAphBQAQAA0BQCKgAAAJpCQAUA AEBTCKgAAABoCgEVAAAATSGgAgAAoCkEVAAAADSFgAoAAICmEFABAADQFAIqAAAAmkJABQAAQFMI qAAAAGgKARUAAABNIaACAACgKQTURj19+nRz6dKlUXV6enr23E+fPm1u3rx5dtuLFy/ObjsUt2/f vtDmk5OT7T3TvXnzZnN0dHRhuX115cqVzbt37zYfP37cXL58+fz2a9eubT5//rxd8rLKttdqzja8 evXqwrKOj483X7582d47nvTPkydPtv/7qsXxabfJa7/c/+jRo1l9EWnp/dinb59642BfymNWj2Vr 6LF2aPOrJ8sxPIY35+v5E+0goDauDKrlQWQnCr3Pnpx2fWKKUp5UbM1to05Wtm9tH5YnMLnP3mYn u10GVFH2gw3oZSiZOuEuOX5k/JZ/RLQ2Pm0YKbfV7ttsx1JkP/etyxsH+6D7U49f3ffeNpd/tNhj uww4hxxUMxzDU5T7l4DaHgJq4+xE0HUQyUSr92X/67Yk7bl169aFkGiDu3cFZAw52Xz//ffb/31l +91b/h/+8Ifz22xo3HVALSfV8sRhw9bUE8dSJx9dTrmNrYzPoScofVy2E3FkiOjap7VxsA96fJQB tTYO7TxTHtvlH4uHGFJbP4bnGHr8Y38IqI2zJ5mug0ged6gHmEwk//u//7v931f25DA3oHpsv/ct v6WAukRwt21fIuS2EEw8Nsz3baP8EXP37l0C6kgtjYMpgaQroIol/1BuTYZjeA4CavsIqI2zk0R5 EMlJ8/Xr19v/fVU+vjwxlS9NlSUTUXllQCbmn3/++cLBLKXbUj5eJzOdvMvJratNQ9l2lH/B24ln iYA1NqC+f/9+UHie0g9jAmotLPet195f9l+5fu8xNvjZku3505/+dOE2ee5PP/10Yfxo6fb3tXlK P5bHQd9zZBu896Da/pYqx6J3LNnxIaX917VNZRulD2w/e/u6az+Krn4r2yXl7T95TrkcXVdtHHil 2z92v6i+/VCb98qxVOo7nsrlelcTo8aINeSYsPtH7revCo09hvvG1tDwPmR+8frrxx9/7O2Xctn/ 9E//dOH/Q8ca4hBQG9d3EvEOKvsce3Db2+V5ZbCyk6+dCPW+8gC367aPl0lftu277777Ztk6WeuJ QSdCb2Kv0XXVJr0hk3Mfu4wxAVVK2mtPMGXfiqn90BfW7InFO/kOWW/f+NHn2naXJxG7jHI7vOWX 7SqXJ2Rby/6Z2o+2n7z9M4Quo1x32d4yVOj6dNu1pM2//PLLeT+U4658vI5rXW/Zjtp+FGPHgd0W eW7X/rb31cZBbdlCx8LQ43bofuiav2psn3vjpDz2y34Zum1LjRExZN/qY8oxZPuktu9EbX8LXbbe rsvum0d1mbqurvml7C+53/6hW+4ru+91+bYNUkPHG+IQUBtXO4j0du+g8iYPe4Dag1cnDyk7gXnh tWuCtxPGb3/727Nt0MfrhGAfU7bDm/w9dnu1ysnTbmc5sQ1l+7BvYu3rq/L5c/qh3Ae27XabvYA2 dL12Obb/yhNNVxvtMsr9U1u+7cdyeXLfvXv3qiepMf1Y9uGUMWLboOv2tkd440P09XO5HBtW7OPt cmoB0z5+TL/ZbZH9KM/13upQW5e9vRwH5bKV7J+h35hgl9+3H8r9bu+rsX3ujalymbX927dtS42R IfvWbrN9rrTV/r9r39W2a8zYKo2ZX8bOuXY/6nZNGQ+IRUBtnJ0IvPIOKm/ysBOHnSjsgWsnob4J QMquu7Z8q5yARG3C7mK3TcsLZHPYPiwnu9LYvprTD+VyyypPJNbQ9dZOPuWJpuuEYJcx9OQm7Im3 DFvlPp7aj+X4KbdhiL519x1LotYPtT6oHat2vd7+Km8f029lX5X7WdXWZW8vx0FXv3QdA9aY/TAl kIwNqHY79jFGhuzbcptrc2fXvuvbrto46aLLHDK/eP1V27+15dQej3YQUBtnJwJ7EOnE6R1U3uRh Jyk7AdoJ2E5UYyYAYZdvJyxVPtersROEnaS9dc5h+7Bvsp06Wdaqqx/K58uJw/Z97flj1ls7+Vjl Ou3zhV3G0JObqI3Thw8fnv9bzOlHu7+k7HqGsM+vBQG7TG98iFo/1MLHkIDat/wp/VZbr1VrS9c4 EF5by31dM3Y/lG2vjQ9rbEDVx+xjjIzZt7ZdUuVxKLr2Xd/Y6psz+9h+0rL7y+uv2v6t7Ysp4wGx CKiNsxNBeRDJJOMdVH2Th50A7UQ3dQIQdkLxJrvaJDFHrU1LsH3Yt71LTJZDlcvVE4fdj15fjFmv N34sXdff/va36niwyxhyclNl++TkK2Pr8ePH20d8Nacfy3WMHTu1de8yfIhaUByz/Cn9NuQ5tbZ0 jQNRzhvyPsKhX+g/dj/UjskuYwNqXz/vcoyM2bfldkuV7evad0uNLc+Q+cXrr9r+rfXtlPGAWATU xtmJoDyIZLIrP8UvapOavV2WYw/ccgIaMwGI8kSj61S7mgz0BOKtcw7bV32T7Zi+mtsP5fN1v9lt sLerMeutjZ9yvHQt0y5jyMnNsqFA7v/d7373Tf/P7Ue7jrHPt/0QFT7EEgF1ar/ZbSr3p6i1pWsc CLs90pfPnj0bvE1j98OUtvcFVLsNUtrGfYyRKe2zy5Gy+6hr3y05tpTtA1lf1/K8/qo9vrYv5m4v do+A2jg7EQw9iGqTmigPytoyx0wAoi+gitqJt4+2xzvB6QlkzPKGsH1oJzXP2L6a2g+iXG7thCJV 7teh662NH32+9kdXG+0yhpzcLNuftceIOf1YrsMbW13KvhC1Y8AbH6Kvn6Vsu2oBtbbescuvkeXI Bx+1Dd7xUFuXvb3Wx30hsMuY/dA1Xmv6ts22r+yXfYyRsftWTNl3Q7Zr7jHVtb/GzLn2drsvpowH xCKgNs5OBEMPotrkIQdk+YtMNWMmADEkoNrtKid7ua/WNjvp2efpOstlCX1ObVv62G0tTzylsX01 tR9EudzyJFDrKzF0vfZx2n+2jdof9japWhuHntws245afwxtT419ft8+lqBiT/j2uV6/2XV740PY x9t+sG0fElDtcmrbaJdvb+/rNznGrl+/fvYYu/5yvw1ZVy2w2P4ZG2rs8vv2w5RA0hVQy+WVgXDM ti01Ruzja/tWtrv8lgRt59B9V9uurvW/fft28+HDh+3/Lho7v3j91bV/7X7U26eMB8QioDbOHlhS Qybw2uQhk1rfidjSSVCfY5crZQ/orolJlROCtkUmm673nZV9YMtrjw3LUlMmnnKd5cnHsuvT7ema /Kb2gygnbbt/Rdl2e//Q9Xrjx2uPfLH3r3/96wu3Kbsdsoy///3v5z8nWxuflj6mNpbEnH5UfWNF +9vb/3p86HrL/ytvfIhaP9TCRzkmdVv18V4g0cdOGQdl2/X/+rxaMLLr6hoHlrah7P8hhu4Hbwx3 KR9v+3fosoZu21JjZMi+1cd4y7PtmHoM223T+2Sd//zP/1wNqF5/Dp1fhsy5duzqNnnnldp+xH4Q UBvVFcqkvBOmsBOHlhyQtV/rsVUenOUJqfzlDX1O+TiprmBhJ7C+xyqvP2p9IHQd5eTZp6/f+/pI SiZMO7Fqlc8d2w/l423ZvijbYE94omu9tfEjfVjeJ+ssb7Nt9E5iXcu39GTTtY/VlPFUsie8Mcsq +7rc3tqxUf4aj5SECW8f6zLtusrHlts5pJ+7+q28769//as7f8h29K3LGwcl6f85PyXbtx9q+7cM isrbD14NGZ+RY0R17Vs5tuQKqgRO2yfl/CSmHsN9c5CnXK60qbytdr4ZMueWz/uP//iP8/OZ13bs HwF1RYZMuhyoQJvKgHpIJEB6V1YBrBcBdUXKl0C8OrQTH3AoDjmg2t+BBwBBQF0JfXmjdmLTl8CG vGQFIN4hBVRti7RDX3Ke+vI+gMNEQF0JDaDle9WUvPxfuw/A/h1SQNW3G8l7F+W9kN73OQNYNwLq inhvLtfiyinQLu+DKUM+eNIq+6Gl2genAKwbARUAAABNIaACAACgKQRUAAAANIWACgAAgKYQUAEA ANAUAioAAACaQkAFDpz9/kyv5n7FmHx92ZMnT7b/y8lrg/3ltda+hq382qm+r2qq/RZ9V439Gqu5 /TW2TQAOGwEVWIkyqC4VumS5h/DLRmUbbGBqMSyN2T4NqKenp9tbLj6/DKNy39iAukR/td7nAOIQ UIGV6AokU+kyMwfUWhtavoIqxoQ5Cajff//99n9f9Y2Hsb+Pv0R/jWkTgMNGQAVWoi+QjGWXlzWg Zm7D3DC39HhYwtw2ATgcBFRgJboCSfkzuNeuXdv8+OOPF963aF8e1t9SL0ue9/nz5+2jLq5Tyi6j XKcGRH0rgvx/7HYpezVPqww8XW3405/+dOE2Lyz1vWVil9suogLqkP1U7udye5Zok91f0m9jr/AC yIWACqxEXyApP0gjAeGnn346Dydd4dO7+mgDjNCAYYOcXac8Tp7z3Xffnf1f1zd1u3S9NmCVoaer DUPCUtm2chlR217eN4R9ft8V1CH7qbY9S7RJx5IGeu1vL+ADOAwEVGAl+gKJDQ4aOuyVr/I5dnld wUxDhD7ehjL7uN/+9rdnj9F1eoFmyHZpeNHbp7bB3ucFLiltm9deEbXtkQG1tp9q2zO3Tfbxtm8l tBJQgcNFQAVWoi+Q9IWpMiDY5ZXhrgwlwgtx9jYbXK2p2zUkEHa1wQtLoq9tdjlR2x4ZUGv7qbY9 c9tU9leLH1gDsDwCKrASfYFkqYBaPscrL6DWQtbY7bLs8r3H1togvLBkt6UWUG2Ai9r2yIBaW9eQ 7ZnaJn2J37sPwGEioAIr0RdIlgqotRDnGRJ8poY8vdL5t7/9bXQbhBeWogLq2G2fEtj6xoO1RECd 06ayz6RsPwM4PARUYCX6AslSAXVIAFO7CKj28bJdU9ogvLC064A6ddtrfdelbzxYcwLqkm3SkKtV 7jMAh4OACqxEXyAZG6bs8sqgYINE13sGdxFQdd3axqltqIWlcvmi1o6oba/1XZe+8WDNCahLt2lu uwHkQEAFVqIvkCwZUO19uiwl9+lylg6o9rHaRnubfawY2ga7bfZ2XZZ3m4ja9ilBzT5/VwF1iTZJ fz169OjCOvU9qVPaDSAHAiqwEuUHTcormzaEaJjoCqhlaPn73/9+/nOa5fM0/Ek4efLkydm/hQ0k ZZBVY7bLu12+0P3Xv/71N48VXW3oCoB6VVDbVf5f7XLbu7ZviL7xYA3ZT7VwObdNugzbRu1v+1wA h4WAChy4MoiUJcGkvKolJYHKBgmtMSFJg4SWDTfeOsvwM2W77DZJSfvK2/raUD7e3qf6At4ut33I 9tX0jQfbN2LIfuranrlt+uWXX86uoMofDxr2pcrtBHBYCKgAAABoCgEVAAAATSGgAgAAoCkEVAAA ADSFgAoAAICmEFABAADQFAIqAAAAmkJABQAAQFMIqAAAAGgKARUAAABNIaACAACgKQRUAAAANIWA CgAAgKYQUAEAANAUAioAAACaQkAFAABAUwioAAAAaAoBFQAAAE0hoAIAAKApBFQAAAA0hYAKAACA phBQAQAA0BQCKgAAAJpCQJ3g48ePm8uXL28uXbp0VteuXdt8/vx5ey8AAADmIKCO9ObNm83169fP A6mGVUIqAADAMgioI3z69Gnz6NGjzZcvX7a3fCWh9ejoaHN6erq9BQAAAFMRUBdAQAUAAFgOAXUB t2/f3pycnGz/BwAAgDkIqDPo+08JpwAAAMshoE4g70W9efPm+af4paaG1JcvX57vhLL+/d//3b2d oiiKoiiqhZIcswu6fALqRK9evToPqS9evNjeugz2BwAAWCMC6gL0Q1JLf9UU+wMAAKwRAXUh8kEp AioAAMB8BNSFPH36dHN8fPzNd6TOwf4AAABrREBdgH5oivegAgAAzEdAHaH2XtNdfQ8q+wMAAKwR AXUE7+ulpHb1C1LsDwAAsEYE1IaxPwAAwBoRUBvG/gAAAGtEQG0Y+wMAAKwRAbVh7A8AALBGBNSG sT8AAMAaEVAbxv4AAABrREBtGPsDAACsEQG1YewPAACwRgTUhrE/AADAGhFQG8b+AAAAa0RAbRj7 AwAArBEBtWHsDwAAsEYE1IaxPwAAwBoRUBvG/gAAAGtEQG0Y+wMAAKwRAbVh7A8AALBGBNSGsT8A AMAaEVAbxv4AAABrREBtGPsDAACsEQG1YewPAACwRgTUhrE/AADAGhFQG8b+AAAAa0RAbRj7AwCA /O7fv7+3yoqA2jD2BwAA+XnBMaqyIqA2jP0BAEB+XnCMqqxWGVBfvXq1uXTp0nlduXJl8+7du+29 w3z8+HFz+fLl2cvpQkAFACA/LzhGVVarC6hPnz7dnJycbP93MWienp5ub+0ny7HhVMoudwkEVAAA 8vOCY1RltaqAKmFUrp6W3rx5szk6Otpcu3Zt8/nz5+2tdbKce/fubb58+bK9ZTcIqAAA5OcFx6jK alUBtebTp0+bmzdvDn6J/vbt22dXTJd+Sb9EQAUAID8vOEZVVgTULQmdQ66g6tVW+9L+8fHxTq6m ElABAMjPC45RlRUB9R/0CurY95Da96EOfXvAGARUAADy84JjVGVFQP0HuSp6/fr1SQFTw62E1Ckf knr58uX5Tijr+fPn7u0URVEUReUpLzhGlbc9S5bkmF3Q5a86oMrL+2M+wV/SbwJY+irqWvcHAACH xAuOUZXV6gOqvEz/4sWL7f+mk+XwPagAAKDkBceoymrVAVW+cmqJcCrmvE2ghoAKAEB+XnCMqqxW G1AlUD5+/Hj7v/kk7C79aX4CKgAA+XnBMaqyWmVArYXJOV/A/+DBg8W/E5WACgBAfl5wjKqsVhdQ y9/hL0tf8tevkCrfAiAfqCrD7dwPWdUQUAEAyM8LjlGV1aoCqvf7+bbsh5xqAdULuLsIp4KACgBA fl5wjKqsVhVQs2F/AACQnxccoyorAmrD2B8AAOTnBceoyoqA2jD2BwAA+XnBMaqyIqA2jP0BAEB+ XnCMqqwIqA1jfwAAkJ8XHKMqKwJqw9gfAADk5wXHqMqKgNow9gcAAPl5wTGqsiKgNoz9AQBAfl5w jKqsCKgNY38AAJCfFxyjKisCasPYHwAA5OcFx6jKioDaMPYHAAD5ecExqrIioDaM/QEAQH5ecIyq rAioDWN/AACQnxccoyorAmrD2B8AAOTnBceoyoqA2jD2BwAA+XnBMaqyIqA2jP0BAEB+XnCMqqwI qA1jfwAAkJ8XHKMqKwJqw9gfAADk5wXHqMqKgNow9gcAAPl5wTGqsiKgNoz9AQBAfl5wjKqsmgyo Hz9+3Dx58mT7v/UioAIAkJ8XHKMqq2YD6uXLlzeXLl3aXLlyZfPu3bvtPetCQAUAID8vOEZVVs0G 1FevXp39+9OnT5ubN2+ehdVr165tPn/+fHb7GhBQAQDIzwuOUZVVkwG1Rq+sHh8fb758+bK99XAR UAEAyM8LjlGVVZqAevv27bOrqF6dnp5uHxVDru7a9e/qbQgEVAAA8vOCY1Rl1WRA1Q9J2Zf3y7JX USW8Rl1Vffr06ebk5GT7v4vvl106KBNQAQDIzwuOUZVVswFVQ19ZNhxab9682Tx+/Hj7v92w7421 ZN1HR0eLv0eWgAoAQH5ecIyqrNIE1BcvXmzv/Za+/L+vD1Hpld6lX+onoAIAkJ8XHKMqq+YDat/L 5vax+/yUv4RkrqACAICSFxyjKqtmA+q9e/cGv6dUr6B2XWXdJb2CWnv7wVQEVAAA8vOCY1Rl1WRA 7TM2wO6avAf1+vXri1+9JaACAJCfFxyjKqsmA6oE0Bs3bnQGvvLT9PskV3CnfoL/5cuX5zuhrOfP n7u3UxRFURSVp7zgGFXe9ixZkmN2QZe/94D6P//zP+dXRIcE1F1dtRxLgvKu3lqwz/0BAACW4QXH qMqqmYCqX9Uk7yUdWvv+6VP5yqldvu+VgAoAQH5ecIyqrJoJqEo/8DSk9vkSf8T3rhJQAQDIzwuO UZVVcwFVDAmp+7x6KldOvV+uWvrDWwRUAADy84JjVGXVZEAVDx8+3Pv7Sz3l7/CXteRL/gRUAADy 84JjVGXVbEAdIjrEygeivFCqxS9JAQCAkhccoyqrZgKq/bJ9eam89lv8tvb9IaldI6ACAJCfFxyj KqsmAqoNpBI6379/T0D9BwIqAAD5ecExqrJqIqAKrqB+i4AKAEB+XnCMqqyaCaiWBNQnT55s/1fX 6geplkJABQAgPy84RlVWzQZU+bR8HwIqAABonRccoyqrZgNq1/eJfvr0aXPz5k1e4gcAAM3zgmNU ZdVsQJX3oHoB1H6JPwEVAAC0zguOUZVVswFVXuLXoCofnPJ+XYqACgAAWucFx6jKqsmAatV+9lRC K+9BBQAArfOCY1Rl1WRA1SunXjA9OTnZPurwEVABAMjPC45RlVWagKq/cS8v/etPinIFFQAAtM4L jlGVVfMB1btiyqf4AQBAFl5wjKqsmg2oN27c6AyfElLv3LlDQAUAAE3zgmNUZdVsQB3yRf1//OMf CagAAKBpXnCMqqyaDKj4iv0BAEB+XnCMqqyaDqj6XlP7IakhL/8fCgIqAAD5ecExqrJqNqC+efNm c3R09M2n+IX9JP8hI6ACAJCfFxyjKqsmA6q9cuoFVCFf4H98fFz9vf5DQEAFACA/LzhGVVZNBlR5 Gf/evXvn4VOupn7//fdn/xZ8zRQAAMjCC45RlVWTAbX8CikbUO13pBJQAQBA67zgGFVZNRlQxdOn Ty+8xO/Vof/sKQEVAID8vOAYVVk1G1C996Ha4kNSAAAgAy84RlVWzQZUJZ/YL8PpoV85VQRUAADy 84JjVGXVfEBdM/YHAAD5ecExqrJKHVAfPnzIh6QAAEDTvOAYVVk1EVDtJ/PH1NxP8ct3qcpyyu9Y HcLb5qXfF0tABQAgPy84RlVWqwyo5ftapwRU71sGln5vLAEVAID8vOAYVVmtMqAqDapjA6psr/0h gV0hoAIAkJ8XHKMqq2YC6pMnT7b/G27ue1CnBlR9a8Cuv+qKgAoAQH5ecIyqrJoIqF28l9KnvCTv mRJQ5Vetjo6OLmzP8fHxTq6mElABAMjPC45RlVWzAbXvZf8lfuZ06hVUZcPzLn52lYAKAEB+XnCM qqyaDaj6MnpXzf1Q0tyAKuwvXvEhKQAAUPKCY1Rl1WRA1auntfd4aijc14ekSrq9U7bn5cuX5zuh rOfPn7u3UxRFURSVp7zgGFXe9ixZkmN2QZffXEC9evVq5weQ5DE3btxoIqAKebmf70EFAAAlLzhG VVZNBlS9Qnp6erq95VvyYaW7d+/O+nDSkgFVtuf69euLvg+VgAoAQH5ecIyqrJoMqELCY+0lcwmw t27d+uZq5R//+MdRAXHJgCrLWvrT/ARUAADy84JjVGXVZEDt+wR/rca+B3TJgPrgwYPFvxOVgAoA QH5ecIyqrAio/3ie9+l7/QqpMrzKtwuUV0rltq63I0xFQAUAID8vOEZVVqsMqN6X7UvZMFoLqBpq be0inAoCKgAA+XnBMaqyajag7uOnT1tDQAUAID8vOEZVVk0G1KEOLZCWCKgAAOTnBceoyqrZgGp/ oalWU74YPxMCKgAA+XnBMaqyajKgDgmnUgRUAAB2zws+kZWd16aoyqrJgDr0Q1IEVAAAds8LPpGV ndemqMqq6YDa9/2kvAcVAIDd84JPZGXntSmqsmoyoAr5mqclvkA/MwIqAKAFXvCJrOy8NkVVVs0G 1NrPmVpcQQUAYPe84BNZ2Xltiqqsmg6ofIqfgAoA2D8v+ERWdl6boiqrZgOq/HyoF0ptEVABANg9 L/hEVnZem6IqqyYDKp/i/4qACgBogRd8Iis7r01RlVXTAZVP8RNQAQD75wWfyMrOa1NUZdVkQBXy Ev/p6en2f+tEQAUAtMALPpGVndemqMqq2YAqH5K6c+dO5xVSrqACALB7XvCJrOy8NkVVVk0GVN6D +hUBFQDQAi/4RFZ2XpuiKisCasMIqACAFnjBJ7Ky89oUVVkRUBtGQAUAtMALPpGVndemqMqq2YD6 5MmT7f/qeA8qAAC75wWfyMrOa1NUZdVkQMVX7A8AQAu84BNZ2Xltiqqsmg+oT58+/eal/b7vRz0U BFQAQAu84BNZ2Xltiqqsmg2ofb/Ff3x8vPny5cv20YeJgAoAaIEXfCIrO69NUZVVswHVu3Ja1qFf SSWgAgBa4AWfyMrOa1NUZdVkQJWrp7du3dq8e/due8u33rx5s7l79+5BX0UloAIAWuAFn8jKzmtT VGXVZEDlU/xfEVABAC3wgk9kZee1KaqyajKgcgX1KwIqAKAFXvCJrOy8NkVVVk0GVDHkPagnJyfb R+/H7du3z7ZjV++FJaACAFrgBZ/Iys5rU1Rl1WxAbflT/K9evbqwLQRUAMAh84JPZGXntSmqsmo2 oKoyDEqdnp5u790v3TYCKgDgkHnBJ7Ky89oUVVk1H1BbRkAFAKyBF3wiKzuvTVGVFQF1BgIqAGAN vOATWdl5bYqqrJoIqPK1UpcvXz5/CV+q9h5TCYVXrlzp/IR/FAIqAGANvOATWdl5bYqqrJoIqEK+ Nuro6Ogs8PW9x1Qee/369b1/B+oSAfXly5fnO6Gs58+fu7dTFEVRVGR5wSeyvG3KVF6bosrbniVL cswu6PKbCKhjvtdUvoZqV1cuh+IKKgBgDbzgE1nZeW2KqqyaCagPHjwY9bJ9C1/UT0AFAKyBF3wi KzuvTVGVVRMBVb7z9NGjR6PCprxv9caNG3t9mZ+ACgB5eCfvyMrMa09kZee1KaqyaiKgDv3tfYuA CgAYwzt5R1ZmXnsiKzuvTVGVVTMB9d69e6OuoLbwQSkCKgDk4Z28Iyszrz2RlZ3XpqjKqomAKi/x 37p1a9R7UOVDUvv8uVOhAfXk5GR7y7IIqACwHO/kHVmZee2JrOy8NkVVVk0EVCGB89q1a4OuiOpX Uu3qymUfXb+EU1tLbw8BFQCW4528Iyszrz2RlZ3XpqjKqpmAar+sv3ZF0j5maJjNjIAKAMvxTt6R lZnXnsjKzmtTVGXVTEAV+pL5kNrX1dNIBFQAWI538o6szLz2RFZ2XpuiKqumAqqovXxuaw3hVBBQ AWA53sk7sjLz2hNZ2Xltiqqsmguoyruauu8PRUUjoALAcryTd2Rl5rUnsrLz2hRVWTUbUEFABYAl eSfvyMrMa09kZee1KaqyIqA2jP0BAMvxTt6RlZnXnsjKzmtTVGVFQG0Y+wMAluOdvCMrM689kZWd 16aoyoqA2jD2BwAsxzt5R1ZmXnsiKzuvTVGVFQG1YewPAFiOd/KOrMy89kRWdl6boiorAmrD2B8A sBzv5B1ZmXntiazsvDZFVVYE1IaxPwBgOd7JO7Iy89oTWdl5bYqqrAioDWN/AMByvJN3ZGXmtSey svPaFFVZEVAbxv4AgOV4J+/IysxrT2Rl57UpqrIioDaM/QEAy/FO3pGVmdeeyMrOa1NUZUVAbRj7 AwCW4528Iyszrz2RlZ3XpqjKioDaMPYHACzHO3lHVmZeeyIrO69NUZUVAbVh7A8AWI538o6szLz2 RFZ2XpuiKisCasPYHwCwHO/kHVmZee2JrOy8NkVVVgTUhrE/AGA53sk7sjLz2hNZ2XltiqqsCKgN Y38AwHK8k3dkZea1J7Ky89oUVVkRUBvG/gCA5Xgn78jKzGtPZGXntSmqsiKgNoz9AQDL8U7ekZWZ 157Iys5rU1RlRUBtGPsDAJbjnbwjKzOvPZGVndemqMqKgNow9gcALMc7eUdWZl57Iis7r01RldVq A+qbN282R0dHm0uXLm2uXLmyeffu3faeYT5+/Li5fPny2fO1piynCwEVAJbjnbwjKzOvPZGVndem qMpqlQH11atXF8KkhtXT09Oz/w/x9OnTC+FU6uTkZHvvMgioALAc7+QdWZl57Yms7Lw2RVVWqwuo euXzxYsX21u+ksB57dq1zefPn7e31Mky7t27t/ny5cv2lt0goALAcryTd2Rl5rUnsrLz2hRVWa0u oEoQ9V6K16uoZXD13L59eycv6ZcIqACwHO/kHVmZee2JrOy8NkVVVqsKqJ8+fdrcvHlzc3x8/M3V T72y2vcyvX3vqpa3vCUQUAFgOd7JO7Iy89oTWdl5bYqqrFYVUDWEeoGyK7zW2PehDn17wBgEVABY jnfyjqzMvPZEVnZem6Iqq1UFVL366V0l1YA6Nmjq8ySk8iEpAGiXd/KOrMy89kRWdl6boiorAurW 1IAq9MrslOe+fPnyfCeU9fz5c/d2iqIoanx5J+/I8rYpS3ntiSxvmzKV16ao8rZnyZIcswu6/FW9 xO8F1K6X/4eoffhqjkPfHwAQyTt5R1ZmXnsiKzuvTVGV1aoC6hIfkqqRq7PXr19f9H2oBFQAWI53 8o6szLz2RFZ2XpuiKqtVBVRR+75Tffl/zJf1W/Ll/0t/mp+ACgDL8U7ekZWZ157Iys5rU1RltbqA qkF0zhf1ex48eLD4d6ISUAFgOd7JO7Iy89oTWdl5bYqqrFYXUEX5flG5+imfwrdXT/UrpMogK1/S X14pldumXnntQkAFgOV4J+/IysxrT2Rl57UpqrJaZUAVGkpr32FaC6j2eVq7CKeCgAoAy/FO3pGV mdeeyMrOa1NUZbXagJoB+wMAluOdvCMrM689kZWd16aoyoqA2jD2BwAsxzt5R1ZmXnsiKzuvTVGV FQG1YewPAFiOd/KOrMy89kRWdl6boiorAmrD2B8AsBzv5B1ZmXntiazsvDZFVVYE1IaxPwBgOd7J O7Iy89oTWdl5bYqqrAioDWN/AMByvJN3ZGXmtSeysvPaFFVZEVAbxv4AgOV4J+/IysxrT2Rl57Up qrIioDaM/QEAy/FO3pGVmdeeyMrOa1NUZUVAbRj7AwCW4528Iyszrz2RlZ3XpqjKioDaMPYHACzH O3lHVmZeeyIrO69NUZUVAbVh7A8AWI538o6szLz2RFZ2XpuiKisCasPYHwCwHO/kHVmZee2JrOy8 NkVVVgTUhrE/AGA53sk7sjLz2hNZ2XltiqqsCKgNY38AwHK8k3dkZea1J7Ky89oUVVkRUBvG/gCA 5Xgn78jKzGtPZGXntSmqsiKgNoz9AQDL8U7ekZWZ157Iys5rU1RlRUBtGPsDAJbjnbwjKzOvPZGV ndemqMqKgNow9geAkncCiqrsvDZFVmZeeyIrO69NUZUVAbVh7A8AJe8EFFXZeW2KrMy89kRWdl6b oiorAmrD2B8ASt4JKKqy89oUWZl57Yms7Lw2RVVWBNSGsT8AlLwTUFRl57UpsjLz2hNZ2Xltiqqs CKgNY38AKHknoKjKzmtTZGXmtSeysvPaFFVZEVAbxv4AUPJOQFGVndemyMrMa09kZee1KaqyIqA2 jP0BoOSdgKIqO69NkZWZ157Iys5rU1RlRUBtGPsDQMk7AUVVdl6bIiszrz2RlZ3XpqjKioA60Zs3 bzZHR0ebS5cuba5cubJ59+7d9p7lsD8AlLwTUFRl57UpsjLz2hNZ2XltiqqsCKgTvHr16kIo1bB6 enp69v+lsD8AlLwTUFRl57UpsjLz2hNZ2XltiqqsCKgjffz4cXP58uXNixcvtrd89fTp0821a9c2 nz9/3t4yH/sDh8ibQCMrO69NUZWd16bIysxrT2Rl57UpqrIioI4kQdR7SV+vopbBdQ72Bw6RN4FG VnZem6IqO69NkZWZ157Iys5rU1RlRUAd4dOnT5ubN29ujo+PN1++fNne+pVeWT05OdneMh/7o13e JBBZmXntiazsvDZFVXZemyIrM689kZWd16aoyoqAOoKGUC+gdoXXqdgf7fImgcjKzGtPZGXntSmq svPaFFmZee2JrOy8NkVVVgTUEfRlfO8qqQbUJd+Huuv94Q3kyMrMa09kZea1J7Ky89oUVdl5bYqs zLz2RFZ2XpuiKisC6gi7CKgvX7483wll/e53v3NvpyiKoiiKaqEkx+yCLp+AOkDX+0y7Xv4HAADA cATUEaI/JAUAALBGBNSRat93uqsv6wcAAFgbAupIte873cUX9QMAAKwRAXWC8sv65adP5Tf5uXoK AAAwHwF1Ig2lUlw5BQAAWA4BFQAAAE0hoAIAAKApBFQAAAA0hYAKAACAphBQAQAA0BQCKgAAAJpC QAUAAEBTCKgAAABoCgEVAAAATSGgAgAAoCkE1AP26dOnzc2bN/f2U6z7Xj/2h7EHoMvTp0/Pfir8 9PR0ewtwEQE1mY8fP24uX758dmBLHR8fb758+bJ5+PDhNyfifZ6k37x5szk6OjrbxtZDwu3bt8/7 EXWMvTjaf9rXUkNO5LbtUoc8rjluL/LGjFRrx0C5nYcUUDlul0VATUQH8cnJyfaWzebVq1dnA7rF E/GuQ4r0x/fff7/93zQauq5cubJ59+7d9laUGHsXLTH2hrB/FAw5aelVqV3vk6j213Dc1mnftB5y dP44xCuoHLfLIKAmIlcMvAEsB/oaQ4L0x4sXL7b/m0YnBqm5yzpkjL2Llhh7Q5RXZLpO5mNPinNE tb+G47aOgLp/HLfLIKAmoQPeu2Ig9925c2dVIUEntzkHm27fX//61xQT+r4w9i5aYuwNJe24devW 5tmzZ2frtFewSxLaZA7f9ViObL+H47YbAXX/OG6XobmUgJqA/PUjA8wbxN77AHWisidpvfIgk4L9 y817jJR3YMl2aFjpemxXSLDrlirbpM/V++Vk9OTJk7P77Dq1uiaAGnm54+7du2frtW0q6cvbdn06 qdb6sK99wk4Y+m9dRtl+Ka+N5bbVJp8h29OFsbfs2BtKtkVOdP/1X/91tk21MSqPkz8U3r9/f9a2 rv2r+7K2/S2138Nx233c6nOGPLbcXq9/yv4dsk1j+lHWafvD66++5XWNWTVkm5Yi28NxOx8BNZFy UtPJxKMHv5RMnj///POFwfv73/9+c/369bNJVZcrB8dvfvOb88lSB3Vt0pK/DvWxevDYiVYPmDIk yPp03UK31R6csjzbPvm/HlD//d//feEkMZUsU5+vy6sdtNp2b6KR+27cuHHenr72lRPJ/fv3N48f Pz7rb1m+Tmq238p9IXS/aRvK8aFtGdLffcplM/bmjb2hpB1yopMxp33ijVHZJrld+8nbt3qffb62 xfZTS+33yPbounVbvD4Razxuu8ZAjbedQvra9p1uqz5O/2/XNeQxQtsj84H+wVHrrznHrBiyjCVx 3C6DgJqMHuwy0MoBWvJO0t6BaSfecmKQdZUD2jvg9CCy6/LWL8oDSchtun5dln2M3Gb/Ip57sMny ypOTtLWrP2vrlNvtbX3tU96+ENpv9nZv3bJMr2/H9vdQjL2vauNgF6QdeqLTbSvDlvcY70TnjQ1h +7S19pdkWzhuu4/brjFQo9tZLlfGho43bZ899oTd9iGPUV4/esd9Xz8MGbN9y1gax+0yCKhJ6cGl 5Q06nSzs4K5NRN5B0BUSyoNNlAe8t349kOy225J16fPk/7XJY+7BJm2wk+iQddYmkQcPHpz3xZD2 qaFt0InIPtbrW6GP1TaM2Z6hGHtxE71sj57EhO5fu27pK70CVRujensZHITeJ31lr3a30P4Sx+3F 5Xlq7e1SOzbt8abHZK2fxZDHKG+d5XE/pB/6xsCcvpyK43YZBNTEygOvHJjeZFibiJYICeUk661f lvmrX/3qm+eWdN21ts052OyE5pU3GaiyjbKd8jKfGto+0dcGvV+2x3us7LNyP5T7Zsz2jMHYi5no pR32RGdPStou6T/dRr2/PNFpm7yxrX1VBpEW2m9x3A5TGwNddDvLfW23q/YYa8hjlPfYcpwO7Qd9 nizPW+Yu5sAuHLfLIKAm4n0YRchAl4FXDm4dwK2FBFlmuf4abZuUPUjnHGzyXG/y1m0r+8EqJxJp s23LmPZ1taE8iXmP1W3RfvEmubH9XcPYW2bsjSXtsCc6odsl65d9fu/evfO+r53odJu9E50ox5to of2WrJfjtl9tDHTR7SzXZ4+3Ift9yGOUt05tt/bNEsfsnL6ciuN2GQTUROxLUlZtcO8jJNjbvfV3 TZ6ybbJd8rwff/xxe+v/Padv4h9K2upNVrq9Xv9YOim8fv36wiQjhrRP1dpQTtKi9li9Xatc75jt 6cLYW2bsjSXbU57otG+kHf/yL/9yYTtqbdTnlP0sbF/JS4Uttd/iuP2q3J5S13PVH/7whwtjSren XK49rrrGkBryGOWts+zDIf3Qd8zO6cupOG6XQUBNRCZY7y8pHYzlfXYA6+CuTURzQ4Kuyx5g3vr1 tnJSljZcvXr1bJnymPK7NcvtnnqwSZvspzlLulyvn5X2i9ePQ9qnam3wbq/d1rWdYsz2dGHszR97 U8j2lCc6IX0m21D2m+4P2z6lzym3W58jt7fWfsVxO/y47RoDQu4vA3qtTTpmpL/sNpXbL88f+hil 67S36T7S5w7pB3lM15id05dTyTo5bucjoCaiA7U88OX28qQt9MD0QoI3EdVCQrk+CQmyDHswdQWH 2nJlGbZ0m/R55fLtcuy2yeP/8z//8+z2Lrrcsj1W10lM1dql+tqnavtCn6/L14lIHivb/vbt282H Dx/Ox4NXto1Dt6cLY+//lmO3TR4/ZOxNJcv3TnS1/tGxUrZb2HFkx7b0v7a3tfYL3aayrZbdr2s/ brvGgPZBuQx9jt3vsp3ffffd+XrlPvmZTLst9j59nvZP12OEPs7uLzu2VF8/jBmztWUsTbZp7cft Egioiej7AMsJoG9QS8kJXP5qts+TwfzTTz9deJyUHAQyuO1t5cAvl1cGhHL9UvbgKu+3E4UePHYi 99qo93sBqeT1R/kcb2ItJxIlj+2a3LraJ7r6V9j7pe36Rc72sbVJV2tofw/B2Js+9qbQE45uQ7kd cr89AXqPlyrHlfe4FtuvvLHEceurjQGv7DpU2Y/2WLTrLbe/7APR9xg7vqRkf3X17RLH7Ji+nMrb B2s8bpdCQMVoGhIyDPBDJicU70Qj+k7EWTH2kN0aj1tgCgIqRiMk7J9codDv0PP88MMPB7l/GHvI bK3HLTAFARWjERL2T/aBvFTjvZQpV2FqL3Fmx9hDZms9boEpCKgYrHwPT22iRYzae9kOcZ8w9nAo 1nTcAnMQUAEAANAUAioAAACaQkAFAABAUwioAAAAaAoBFQAAAE0hoGIy/XUL7xcr+sx57qGg/5AF 4w1LYM6Llb3PCKgrId+xV36tSa1qv3JSmjr47desZDlw6L/h9Lseh/SJ9/VR2iZp5+PHj88eN6X/ ve2w9wtv/d5PN841tJ0tynS8MvaWw5zHcbtvBNSVkMnG+21je5tOHkMnmzmy/WVH/43j9U9Jfxe6 /P5He1LQ+6b2v94my+rbll39AMCYdrYq0/HK2FvG2uc8jtv9I6CuRPkTerVJXP7qev369fZ/u5Pt wKH/xqn1j9LJv/a749o+PQHM6X9dlqzPO5HK/Xfu3NlJP45tZ6taH28WY28Za57zOG7bQEBdqb5J fNeyHzj0X7eu/pErMzL59/WdnPhqv1s+tv9lWfJyl9dfcjLaxRWgJdrZikzHK2NvN9Yy53HctoOA ulJDJxs9WOUvSf23DnZdhh38ekDI47S8vzKzHzj0X7da/9j29Z2Y5bGPHj1y+3do/1veVZFdnWSW aKdur1Y5DvR9jrJ87Q/5vx0T9r2Q3jiSdejLy12P7Rpvdt1Sdp+UbZC+18BWPnYpax97uzK03Znn vCXGCMftcgioK9U32ZSTxv3798/eEC4HgxwYz549O79PB793MNiD0eo6cDKg/7rV+kdvn/ueu77+ 95Trlj68devWrO2omdNOfa492dgT/c8//3xhbP3+97/fXL9+/Wwc6ElE+uU3v/nNeSAqx5GuQ27T 8aiP9cJUbbzJ+nTdQrfT7hddd3mbd+JdQm1szNknVm35Xcp173Ls7Upfu3WMyL6WyjjnzRkj+lyO 2+UQUFeqb7JR3sBV5eDX/9vH6vPtQSMiJptdov+61fpHJ+J9hAShk65MsvLvsl+XMqedcqLx9qvd duGNLR0Xcrs9wev2lO0tlym0b+021MabbGsZJPREaW+3t8m27PLK4drH3q4MbXfmOY/jdn/HrYeA ulJjJxtvMu2bMPQg8p4fMdnsEv3XrdY/OuHO3e6h/V/SfpM+HfvcMaa2U9tlTzxK79Nl6tjyTjTl ertOdN7JuDxZeeNNt0fHaFl2Xfaxc/d9n9rYWMvY25Wh7c4853Hc7u+49RBQV2qXk40+Rw7W2vMj Jptdov+61fpHJ9x9XcUSug3eyWQpU9vZtW26z3WZOjZ2caLToNB1opNl/upXvxrcvoh+F2sfe7uy hjlv6hjp2q8ct9MRUFdqV5ONHGT2wNnnZLNL9F+3Wv/odttJdIrWQ8LUdur+rm2bHR/62H2e6GSZ Q9un2zu2T8Za+9jblTXMeRy334o6bj0E1JXaxWTjTb77nGx2if7r1tU/OokOOUk/ePDgm0lYZAgJ U9qp2+bt13Kf69jY1YnO3u6Nt659INtmt0vWL+9f++WXX86W461zKYy93Rja7uxzHsftfo5bDwF1 pXYx2XiP3fdksyv0X7eu/tFtr/WLksm29hf70P73RIWEqe2UE5X3HG2z3q5jY+kTnW637VtvvNn2 2cfKdl69evXCSdJ+Gb1uy5R9NwRjbzeGtjv7nDd1jHDcLo+AulK7mGx0AOv/dR3yfJmQ3759u/nw 4YP73Gzov259/WPb5j1GJmuvz9TQ/vfoPokICVPaaZ9TngDtMmpjq+tEV7ZZTnTltnWd/GrLlWXY stsk21OejHW9u9gHjL3dGNruQ5jzpowR+xyO22UQUFdGB2w5ML0Bp4NRyx4M9mCU0gPDPkcOivfv 339zoJfPlSoPhFbRf/3Kdndtn07WQx8/pv9LQyblXVminXY7vbH1008/ueOiaxzKfTL27t27d35/ eZLrG2/l/eVJrnxO1/bMxdhb3prnPI7b/3tO1/bsCgEVAFZKT3T2xAagbWs5bgmoALBSBFQgHwIq AOCgEVCBfAioAICD5L0/bch7KQHsz9qOWwIqAAAAmkJABQAAQFMIqAAAAGgKARUAAABNIaACAACg KQTUlZCvpYj+1RJZ59Bf+5Bfrmj5V1Xov7boL7aUP+E3xJznHgL6bp4191/r8yAOCwH1wOmEKF9F of+2X1EhVf5kmfezfPYx5Vdd6H0PHz68MOnq+vp+Es1uY2vov/lqPxfo1dATkbZ57Ine7psMIYG+ m4f+W4adY/TfZf/tex7cJwnRtp1StfHkfVWUjgfps8ePH589bsrY9bbD3i+89bfatwTUA+ddWdMB 2jdJysFy9+5dd9KxYUgPpNryZBu88CS365cNy7/lINJlDT1Z7Br9N59sk3disrfpSShiu6cGjH2g 7+ah/5bR8jzYCm9slaQN0sayHfpce9/Usau3ybL6tqX1L/snoB4w+WvKG6A6gLsGr/AmFhnU3gQi B1NtYtH1lROcnZD+8pe/bP785z+fTVp9E14U+m8ZP/zww4VJsDaRS3+9fv16+7/d0f5srZ889N08 9N98rc+DraiNLaXhtLb92j4NqHPGri5L1uf94SX337lzp/kxSEA9UDqYa4NzysSiz/P+6uob8F0T T9+BvQ/03+7se3t1P2QKCYq+m4f+GyfTPLhvXWNLLyaM7Str7NiVZdUuWEhY9vZpawioB0r+6q0d yFMnFqF/BXrPLd87ZOnBYv961INWtrO1K4D03+4MnWi1fdJm21Zpmy7DtlX3izxOy3tJUB/Xej95 6Lt56L9xMsyDraiNLTs2+kKhPPbRo0dufw4du5Z31dbbH60ioB6gvoljzsSiE4ROwkP/CvPWKQeP Tn7yb1mWngz2+dcd/bdbfROttlX76P79+2cfHJCTpVx1efbs2fl92n59jj2ZyuO9vvAemwV9Nw/9 N5xua19f1e5Xu54HW1EbW3q7d8V4jL6x6ynXLf1369atWdsRiYB6gHRQen/BizkTiygnl6GTrQ1U lm5PbXuj0X+7NXSi1bDtPU7brP3h7RN7Fcwqn5sJfTcP/Tdctnlw32pjS9u5j4Aq9I8l2Y/y7xav PtcQUA+QHhC7mliUTBQ6uUj1DXx5vHeQ6va0cuDQf7s1NiR47dI2105UOil7z+97bsvou3nov+Gy zYP71hdQ5+7zqQFV95P07djn7hsB9QBFTSxCDxqdXLpeqml1YinRf7u1y5Cgz5F9V3t+7bkZ0Hfz 0H/DMQ+OUxtb2o9zt3no2PX07ctWEVAP0NCJpW+SlOXolwar2hvY9a/groPn0AIq/TfNrkJC2T+H GLLou3nov+GYB8epjS3tJ2nX0PfaegioBNSDoAO5azAOOchlki0n2AcPHrjPGXLwyDozTMz0324N nWjHhARvAj6EkFCi7+ah/4ZjHhyna9vt+0D7zOmbGgIqmqGTYNdA7pqAhRwMV69e/eZAkcnBG+R9 k9mQbWoF/bdbQyfaMSHBe2zt+eVzM6Hv5qH/htNtZR4cpmts6XZ39ZWQIFu7yjp07HoIqGiKDPS+ v2xlkvAOGB3M3oGkzykHutzetT49uLoOzpbQf7szdKIdExK0z/X/ug7t67dv324+fPjgPjcT+m4e +m8c5sHh+saWHRfeY6TtXe0aOnY9Op7L/m4dAfVADT2QdeDa6pog9L1D5fP6Jlx5fKaTGv23PD05 23ZLeZOmviSmZSdlO9FLaX/b50hfvX///psTQvlcqTnvC4tC381D/02j28w82K0cM1K1fev1Vdfj x4zdkv6RUD63xYDvIaAeMDlo7OS6L3qAZTkoFP0HYO2YB7EvBNQDJy8b7PuyfisT3BT0H4C1Yx7E PhBQD5z+1bmvA1smtsyTCv0HYO2YB7EPBNSVkPe9RL80In/xHsrLMfQfgLVjHkQkAioAAACaQkAF AABAUwioAAAAaAoBFQAAAE0hoAIAAKApBFQAaNQ+PsEs6xz6C0fy9T98whrYPf2qrym/pDXnuftE QAWAxugJRb4cXf9d/lxh+b2Q3s8a2seUP7Wp9+nPTqqh33lptxHA/6n9nKlXQ/8YnBoy7bxAQAUA zOJdmdSA2XeSkRPS3bt33fBqw6SeRGvLk23wwqfcrr+zLv+WE6wua+jJFjhkcjx4fxza2zRwRhwz U8PtvhFQAaAh8hK7d/Vy6JVNL6BKkPROTnIirZ20dH1lULbB9i9/+cvmz3/+81n4zXbyA3blhx9+ OPsDTnkBVcix+vr16+3/doeACgCYRU9k3lWVqQFVn6dXPS25786dO9WTVleArZ10AVy072OFgAoA mEWuntZOIlMDqpArqHLV03tu+R5US98aYK+icgUVGGdoQNVjS443e5zJcaXLsMeZzgnyOC3vbTkE VADAZH0BdE5A1aCpJ7GxH8yw67RvF+A9qEC/voBaBs379+9vHj9+fPYHq7zy8ezZs/P79NjzQqc8 3jsOvcdmQEAFgAboSaz2qfg5AVWUIXXoyar2/lXdHj7FD3QbewXVe1wZMr35wF6BtQioAIDJNEDu KqAqfblfqzyZleTxtfevyvb0PR9Yuykv8Zf6QqZePfWeT0AFAEwWFVCFnjD1hNb10nwtoAIYZpcB VZ8j8wZXUAEAixsaUPtOMrIcef+aVfsglF5N7TpxElCBeXYVUMtjk4AKAFicnsS63tM5JCzKSao8 QT148MB9zpATp6wz24kNaMkuAqr3By0BFQCwOD2JdJ3Euk5gQk6EV69e/SaMSsj0gm9fKB6yTQC6 7SKgeo8loAIAdkK/VqbrCqmETe8kpFdUvJObPqcMon1XZPXE6i0TwDC7vIKq/9d16HH+9u3bzYcP H9znZkFABYBGDA2EeiKz1RU09T2o5fP6Tljy+GwnNaAVGgztMSflvWJhP4UvZcOsDZ9Seqzb58hx +v79+/PH6fPL50pl+b5iAioANEROOi28pK4nV66eAtgHAioANKb2ntFIrQRlAOtEQAWAxujVy30F RAnIhFMA+0RABYBGyXtAo19ilyunvKwPYN8IqAAAAGgKARUAAABNIaACAACgKQRUAAAANIWACgAA gKYQUAEAANAUAioAAACaQkAFAABAUwioAAAAaAoBFQAAAE0hoAIAAKApBFQAAAA05ZuASlEURVEU RVEtlATU/7+8kaIoiqIoiqL2Uf/6r//6P/8P08HKOy7xXQEAAAAASUVORK5CYIJ=

Only the most complex U.S. criminal trials can compare to ICTY cases. For example, one of the longest and most complex criminal trials in U.S. history was the 1988 trial of the Lucchese crime family. 169Jesus Rangel, All 20 Acquitted in Jersey Mob Case, N.Y. Times, Aug. 27, 1988, at A111, 20. The article describes the Lucchese crime family trial as “[o]ne of the longest criminal trials in the United States.” Id. See also Paul Richter, Longest Mob Trial Ends in Acquittals, L.A. Times, Aug. 27, 1988, at 1, 20 (describing the trial as “the longest federal criminal trial in the nation’s history”). It lasted nearly 21 months and involved 90 witnesses and 850 exhibits; the 20 defendants were accused of multiple acts of selling and distributing cocaine, credit card fraud, gambling, and loansharking that took place over a nine year period. 170See Rangel, supra note 169. It rates a complexity score of 1.09, which makes it slightly more complex than the average ICTY trial, but there have been 15 cases at the ICTY that were more complex, including 2 that were more than twice as complex. While it is axiomatic that international criminal trials are long and complex, this is true in a way that virtually nobody who has not actually tried one understands.

C. Conclusion

The complexity of the cases that have come before the ICTY is mind boggling. The ICTY has completed 41 full trials so far, and will complete another three. The completed cases have a combined Complexity Score of more than 39, and this will only go up as the remaining trials are completed. Moreover, there are a number of reasons to believe that this measure undercounts the real complexity of the ICTY’s work substantially.

For example, the measure does not include the complexity of cases where the accused died prior to a judgment. That includes the case of Slobodan Milošević, who died just months before the close of his case. Mr. Milošević’s trial was arguably the most complex case the ICTY undertook and would have added significantly to its overall complexity, if it were included. 171See Higgins, supra note 17, at 241–49 (describing the complexity of the trial). At the time of his death, the judges had heard 295 witnesses, received about 5000 exhibits, and sat for 466 trial days, and the trial was not over yet. 172Damaska, supra note 16, at 340–41. This corresponds to a Complexity Score of 2.44, which is only slightly less complex than Prlić et al. and Popović et al., the ICTY’s two most complex trials. If the Milošević trial is included in the calculation, the total complexity of the ICTY’s cases rises to slightly more than 42.

Nor does it include the complexity of those cases that resulted in guilty pleas or cases where the accused were transferred back to domestic courts in the former Yugoslavia for trial. Eighteen cases have been resolved by guilty pleas, and some of them involved serious allegations against high-ranking officials. 173The most significant guilty pleas came from Biljana Plavšić, who had been the co-President of the Bosnian Serb Republic (later the Republika Srpska), and Milan Babić, the President of the Republic of Serbian Krajina. In addition, seven cases were transferred back to domestic courts in the former Yugoslavia after an investigation and issuance of an indictment by the ICTY. While none of those cases resulted in a full trial at the ICTY and thus did not contribute to the court’s overall Complexity Score, investigating the alleged crimes, preparing the indictments, and getting the cases ready for trial involved significant work.

The calculation also excludes the contempt cases the ICTY has litigated. It appears that, on average, they have been significantly less complex than the cases involving allegations of violations of international criminal law, but there have been a number of them. 174See, e.g., U.N. Secretary-General, Assessment and report of Judge Theodor Meron, President of the International Tribunal for the Former Yugoslavia, provided to the Security Council pursuant to paragraph 6 of Security Council resolution 1534 (2004), and covering the period from 15 November 2011 to 22 May 2012, 30–34, U.N. Doc. No. S/2012/354 (May 23, 2012) (by Judge Theodor Meron). In addition, 17 investigative dossiers, involving 43 individuals, were compiled and transferred to national courts. These dossiers contained the results of investigations by the Office of the Prosecutor (OTP) that produced evidence of crimes but for which an ICTY indictment was never confirmed. 175Budget for the International Tribunal, supra note 165, at 14. It took investigative resources to compile these dossiers, but that expenditure of resources is not captured in the court’s overall Complexity Score.

As a result of excluding these factors from the calculation, the estimate of the ICTY’s overall complexity used in this Article is inherently conservative and there are good reasons to believe that the true complexity of the ICTY’s work is significantly higher. Even just based on the cases that resulted in full trials, however, the complexity of the ICTY’s work is enormous. It will ultimately try the equivalent of more than 17 IMTs. To put it another way, the 41 trials the ICTY has already completed are roughly equivalent to 1400 of the median cases described in Heuer and Penrod’s dataset. 176See supra notes 164–168 and accompanying text. As far as I am aware, the trials completed by the ICTY thus far are the most complex set of related criminal cases that have ever been tried by any court anywhere. 177While I have not measured the complexity of all the cases at the ICTR, the ICTR has indicted fewer accused and they appear to have been, in general, less complex indictments and trials. The ICC may eventually overtake the ICTY in terms of the total complexity of the cases it tries. However, this prospect would require the equivalent of another 45 trials of the same complexity as the Lubanga trial. I am unaware of any domestic courts that have tried a comparatively complex set of related criminal cases.

IV. Efficiency

It is undeniable that the ICTY’s trials have been enormously complex, but the ICTY has also been very expensive. The tribunal will cost more than $2.7 billion over its lifetime. 178This was calculated using the methodology described in Ford, supra note 3, at 989–90. The figure reported in my earlier article was updated to include data from the 2010, 2011 and 2012 United Nations General Assembly resolutions authorizing appropriations for the ICTY’s budget. See G.A. Res. 65/253, 3, U.N. Doc. 65/253 (Dec. 24, 2010); G.A. Res. 66/239, 10, U.N. Doc. A/66/239 (Dec. 24, 2011); G.A. Res. 67/243, 10, U.N. Doc. A/RES/67/243 (Dec. 24, 2012). The calculation was also updated to reflect an estimated termination of the ICTY’s operations in 2017 rather than 2014. This reflects the fact that the timeline for the ICTY’s completion has slipped once again. See supra note 4. Once adjusted for inflation, that means the international community will ultimately pay the equivalent of $3.08 billion in 2012 U.S. dollars. 179Past inflation was accounted for using the Consumer Price Index produced by the U.S. Department of Labor, Bureau of Labor Statistics. Future inflation was assumed to be 2% per year. The question then becomes: did we receive a good return on our investment?

One way to answer that question is to compare the relative efficiency of different courts. 180See Shany, supra note 26, at 237 (suggesting the study of the relative efficiency of international courts); Wippman, supra note 14, at 862 (noting that one way to determine whether the ICTY provides “value for money” is to compare the cost of criminal trials in national legal systems with those at the ICTY). Efficiency is often defined as the ratio of work performed to energy supplied. 181See Oxford English Dictionary, supra note 46, at 84 (“[t]he ratio of useful work performed to the total energy expended”); Random House Webster’s College Dictionary, supra note 46, at 419 (“the ratio of the work done by a machine to the energy supplied to it”); The American Heritage Dictionary of the English Language, supra note 46, at 570 (“[t]he ratio of the effective or useful output to the total input in any system”). The principal work performed by international tribunals is to try those accused of having committed serious violations of international criminal law. 182See supra note 1. The energy supplied to them is the funding that makes those trials possible. Accordingly, one way to define the efficiency of courts is as the ratio of trial complexity to cost. Of course, this is not the only way one could think about the efficiency of international criminal courts, 183For example, one might also calculate efficiency as the total gravity of the crimes adjudicated divided by the cost of those trials. This would be a different but also valid way of thinking about efficiency. but it is a relevant measure of their efficiency, given that trials are supposed to be their principal output.

Using this model, the overall efficiency of the ICTY is 1.37 x 10-8 units of trial complexity per 2012 U.S. dollar. This ratio is based on the total complexity of the ICTY’s trials 184For this purpose, the complexity of the 41 full trials that have been completed plus the partial trial of Slobodan Milošević were included. This yields a total complexity of 42.13. That score does not include any estimation of the complexity of the three trials that are currently underway or any estimation of the complexity associated with guilty pleas, cases transferred to domestic systems, or the contempt proceedings. See supra notes 172–175. divided by the inflation adjusted total cost of the ICTY. Another way to think of this is that each point of trial complexity cost approximately $73 million to produce. If we can calculate a similar value for domestic criminal cases or for another international criminal court, we could compare their relative efficiency and draw some conclusions about which was more efficient.

A. Is the ICTY Efficient?

Before we begin comparing the ICTY to other courts, however, we must address a legitimate concern—that trial complexity at the ICTY is driven not by complex law and facts, but rather by loose procedural rules and judges that are unwilling to rein in the excesses of the parties. For example, the measures used to calculate complexity in the Article could be artificially inflated if the parties were permitted to introduce large amounts of irrelevant evidence. If this were the case, the complexity scores of ICTY trials would represent not the complexity of the underlying cases, but rather the inefficiency of the ICTY’s procedures and the propensity for the parties to waste time and judicial resources. This would, if true, undermine attempts to measure the court’s efficiency. After all, it makes no sense to think of a court’s efficiency as the complexity of its trials divided by their cost if you conclude that the trials could be completed in half the time and with fewer witnesses and exhibits but for time wasting by the parties.

ICTY complexity does not, however, appear to be driven by loose procedure or lax judges. First, the ICTY uses procedural and evidentiary rules to limit unnecessary evidence that are quite similar to those used in the United States. The Federal Rules of Evidence govern the admission of evidence before federal courts, 185See Fed. R. Evid. 101(a). and the vast majority of U.S. states follow very similar rules. 186See George Fisher, Evidence 2–3 (3d ed. 2013) (noting that 44 states “have adopted the Federal Rules [of Evidence] in whole or great part . . . .”). As a result, the Federal Rules of Evidence can be thought of as the basic rules that govern the admission of evidence before the majority of courts in the United States. Therefore, one way to evaluate the ICTY’s procedures is to compare them to the analogous provisions of the Federal Rules of Evidence.

The ICTY may only admit relevant evidence that it deems to have probative value, 187ICTY Rules, supra note 148, at R. 89(C). and the court may exclude otherwise relevant evidence if its probative value is substantially outweighed by the need to ensure a fair trial. 188 Id. at R. 89(D). These rules are functionally quite similar to Rules 401, 402, and 403 of the Federal Rules of Evidence. Moreover, the ICTY’s judges are admonished to “exercise control over the mode and order of interrogating witnesses and presenting evidence so as to . . . avoid needless consumption of time.” 189Id. at R. 90(F). This language is very similar to Federal Rule of Evidence 611. 190Fed. R. Evid. 611(a). The last three clauses of Rule 403 may serve a similar purpose. See Fed. R. Evid. 403 (giving the court discretion to exclude relevant evidence if it would result in “undue delay, wasting time, or needlessly presenting cumulative evidence.”) Cross-examination at the ICTY is “limited to the subject matter of the evidence-in-chief and matters affecting the credibility of the witness,” 191ICTY Rules, supra note 148, at R. 90(G) & 90(H). which is very similar to Rule 611(b). 192Fed. R. Evid. 611(b). Moreover, ICTY judges have the authority to limit the time given for the prosecution’s case in chief and “the number of witnesses the Prosecutor may call.” 193ICTY Rules, supra note 148, at R. 73 bis (C). Judges in the United States have similar authority. 194Who Does What, Witnesses: Qs & As, Federal Judicial Center (Aug. 23, 2013), available at http://www.fjc.gov/federal/courts.nsf/autoframe?OpenForm&nav=menu5a&page=/federal/courts.nsf/page/311?opendocument (last visited Aug. 23, 2013) (“[T]he judge often sets limits on the number of witnesses and the amount of time for testimony in a particular case. . . . The goal in setting limits is to ensure that each party has sufficient time to make his or her case, but without redundancy.”). In short, the ICTY has rules designed to control the trials and prevent the parties from presenting needless, irrelevant, or cumulative evidence that are very similar to those used in the United States.

Despite these tools, the ICTY has come under enormous and continuing pressure to reduce the time and cost of the trials. 195See Kwon, supra note 1, at 361 (“[T]he judges of the Tribunal have found themselves faced with the daunting task of determining how to change the existing procedures to be able to dispose of increasingly complex cases in a shorter period of time, while still respecting to the highest degree the rights of the accused to a fair trial.”). This pressure has forced the court to adopt some procedures that go considerably farther than simply overseeing the trial to ensure the parties do not waste time with needless or duplicative evidence. 196See Maximo Langer & Joseph W. Doherty, Managerial Judging Goes International, but Its Promise Remains Unfulfilled: An Empirical Assessment of the ICTY Reforms, 36 Yale J. Int’l L. 241, 245–51 (2011) (describing changes made to the Rules of Procedure and Evidence). For example, pursuant to Rule 73 bis, judges at the ICTY may, “in the interest of a fair and expeditious trial, invite the Prosecutor to reduce the number of counts charged in the indictment.” 197ICTY Rules, supra note 148, at R. 73 bis (D). If that does not work, the Court may simply tell the Prosecutor to limit the number of counts. 198See id. at R. 73 bis (E). In addition, the Trial Chamber can fix the number of crime sites or incidents about which the Prosecution may introduce evidence. 199Id. at R. 73 bis (D). The aim of these rules is to expedite pre-trial and trial proceedings. 200See Higgins, supra note 17, at 240. Judges do use Rule 73 bis to limit the scope of the prosecution’s case, a fact the Prosecutor has vehemently protested as an intrusion on her authority to decide which crimes and acts to charge. 201See id. at 251–57. In at least one case, the Prosecutor argued that the court’s use of Rule 73 bis denied the Prosecution a fair trial by forcing her to present a case that was not “reasonably representative” of the accused’s criminality. 202Id. at 254. But see Kwon, supra note 1, at 372–75 (arguing that Rule 73 bis gives the Trial Chamber much needed authority to “compel the prosecution to focus the evidence presented on the most important counts and charges in the indictment”).

The ICTY has also adopted a system, contained in Rules 92 bis and 92 ter, where written witness statements can be admitted in lieu of live testimony. These rules provide for the use of witness statements rather than live testimony when the evidence goes to “proof of a matter other than the acts and conduct of the accused as charged in the indictment,” and when “other witnesses will give or have given oral testimony of similar facts.” 203ICTY Rules, supra note 148, at R. 92 bis. This rule was adopted to reduce the length of the trials, but it has been the subject of considerable criticism. Patricia Wald, who was Chief Judge of the United States Court of Appeal for the District of Columbia Circuit before becoming a judge at the ICTY, considered this practice to be the “most troublesome” aspect of the ICTY’s proceedings. 204Wald, supra note 16, at 473. “To permit critical material to be admitted without the ability to directly view and question the witness goes to the heart of the process and threatens to squander the ICTY’s most precious asset—its reputation for fairness and truth seeking.” 205Id. Defense counsel at the ICTY share Judge Wald’s concern that the reliance on witness statements over live witnesses and cross-examination sacrifices fairness for expediency. 206See Turner, supra note 29, at 558–60. Not everyone agrees with these criticisms, of course. 207See Kwon, supra note 1, at 365 (arguing that the procedures on the use of witness statements “enhanced” the ability of the judges to manage the trials while simultaneously being “fully consistent with principles of justice and fair trial rights”).

Over time, the ICTY has adopted rules that permit judges to limit the size and scope of the prosecution’s case and permit the use of witness statements instead of live witnesses in certain circumstances. These rules were designed to reduce the length and cost of the trials, but they are in tension with the court’s duty to ensure that its trials are both “fair and expeditious.” 208International Tribunal for the Prosecution of Persons Responsible for Serious Violations of International Humanitarian Law Committed in the Territory of the Former Yugoslovia since 1991, Updated Statute of the International Criminal Tribunal for the Former Yugolslavia, Art. 20(1), (Sept. 2009), available at www.icty.org/sections/legallibrary/statuteofthetribunal. Indeed, at different times and on different issues, prosecutors, judges and defense counsel have all argued that in changing the rules in ways that promote expediency the ICTY has risked making the trials less fair. 209See supra notes 202 & 205. There is significant disagreement about this, however. See generally Kwon, supra note 1 (rejecting arguments that there are due process concerns arising out of the changes made to the Rules of Procedure and Evidence). Academics have also argued that, after years of pressure to speed up the trials, the ICTY now places too much emphasis on expediency at the expense of fairness. 210See Heidi L. Hansberry, Too Much of a Good Thing in Lubanga and Haradinaj: The Danger of Expediency in International Criminal Trials, 9 Atrocity Crimes Litigation Year-In-Review 357, 357 (2011) (arguing that “judges tend to consider any reason for delay as unjustified and to be avoided at all costs in order to promote trial expediency” and that this has undermined the protection of witnesses at the ICTY and ICC); Michele Caianiello, Law of Evidence at the International Criminal Court: Blending Accusatorial and Inquisitorial Models, 36 N.C. J. Int’l L. & Com. Reg. 287, 316–17 (2011) (expressing concern that the evidentiary rules at the ICTY were amended solely for purposes of expediency and not because of any legal deficiency in the previous rules); Charles Chernor Jalloh, Does Living by the Sword Mean Dying by the Sword?, 117 Penn. St. L. Rev. 707, 751–52 (2013) (arguing that over its lifetime the ICTY’s decision-making with respect to assignment of defense counsel shifted from an emphasis on the rights of defendants to an emphasis on expediency); Anna Petrig & Fausto Pocar, Case Referral to National Jurisdictions: A Key Component of the ICTY Completion Strategy, 45 Crim. L. Bull. 1, 1, 12 (2009) (worrying that referral decisions were driven largely by concerns of expediency in ways that undermined the legitimacy of those decisions); Daniel Tillay, The Non-Rules of Evidence in the Ad Hoc Tribunals, 45 Int’l Lawyer 695, 715–16 (2011) (worrying that rules about the presentation of evidence were changed largely for reasons of expediency without due regard for their effect on fairness).

In short, the ICTY has been under immense pressure to reduce its trial times. It has tried to achieve that goal by making various changes to the rules that favor expediency over other goals, including fairness. These far-reaching changes, however, have not reduced the length of the trials, 211See Langer & Doherty, supra note 196 (concluding that the procedural changes made at the ICTY failed to meaningfully effect the length of the trials). and it may not be possible to favor expediency much further and still have fair trials. Yet, the trials remain immensely complicated. The most reasonable way to interpret this evidence is that there is not a lot of time-wasting going on, and that there is little additional complexity that can be wrung out of the trials through procedural changes without profoundly undermining their fairness. In effect, dividing the total complexity of the ICTY’s trials by their cost does tell us something meaningful about their efficiency.

B. Comparison to the SCSL

Comparing the efficiency of the ICTY to the efficiency of other international criminal courts is possible because international criminal courts tend to be structured similarly, have similar purposes, and conduct similar trials. 212See supra note 144 and accompanying text. In addition, the total cost of an international criminal court is pretty close to the total cost of the justice they produce because their budgets cover the investigation phase, pre-trial phase (including costs of pre-trial detention), trial phase (including costs of prosecution and defense), and appeals. 213See generally Wippman, supra note 14. The only thing missing from their budgets is the cost of enforcing their sentences. 214All of the international criminal courts rely on states to enforce the sentences in their own criminal justice systems. See, e.g., Member States Cooperation, U.N. Int’l Crim. Trib. For the Former Yugoslavia, http://icty.org/sections/LegalLibrary/MemberStatesCooperation (indicating that 16 countries have entered into agreements for the enforcement of ICTY sentences). Thus, comparing the overall efficiency of one international criminal court to the overall efficiency of another international criminal court comes quite close to an apples-to-apples comparison.

This Article focuses on the SCSL as a comparator because the SCSL has had a small number of trials that have all been completed. Thus, most of the data necessary to calculate its efficiency is available and relatively easy to collect. Complexity Scores for the SCSL’s four trials are presented below in Table 4. The SCSL’s trials have had a total complexity of 4.27. Not surprisingly, the trial of Charles Taylor was the most complex trial, although the RUF trial was not far behind. Both of these trials were more complex than the average ICTY trial. The CDF and AFRC trials were roughly on par with the median ICTY trial. Unlike the ICTY, the SCSL did not have any relatively simple trials, probably because the SCSL focused on prosecuting leaders who bore the greatest responsibility for the conflict. 215See Statute of the Special Court for Sierra Leone art. 1(1), Jan. 16, 2002, 2178 U.N.T.S. 145, available at http://www.rscsl.org/Documents/scsl-statute.pdf (limiting the personal jurisdiction of the SCSL to those “who bear the greatest responsibility for serious violations of international humanitarian law” including “those leaders who, in committing such crimes, have threatened the establishment and implementation of the peace process in Sierra Leone”). In contrast, the ICTY initially charged a fairly large number of low-ranking perpetrators. See Ford, supra note 71, at 71–72.

Table 4: Complexity of SCSL Trials

Trial Days

Trial Witnesses

Trial Exhibits

Complexity Score

CDF Trial 216See Prosecutor v. Moinina Fofana and Allieu Kondewa, Case No. SCSL-04-14-T, Judgment on the Sentencing of Moinina Fofana and Allieu Kondewa, Annex F at 21, (Spec. Ct. for Sierra Leone, Trial Chambers I, Oct. 9, 2007), available at http://www.refworl.org/docid/484417252.html. The CDF trial judgment does not contain a direct statement of the number of exhibits entered into the record during the trial. However, a review of the evidence used to support the judgment indicates that the case was heavily skewed towards witness testimony. Exhibits are rarely mentioned. The highest numbered exhibit mentioned in the judgment is Exhibit 147. See id. 98.

162

119

147

0.74

RUF Trial 217See Prosecutor v. Issa Hassan Sesay et al., Case No. SCSL-04-15-T, Judgment Summary, Annex B at 32–36 (Mar. 2, 2009), available at http://www.refworld.org/docid/4ae9b31c2.html.

308

171

395

1.25

AFRC Trial 218See Prosecutor v. Alex Tamba Brima et al., Case No. SCSL-04-16-T, Sentencing Judgment, Annex A at 58–63, (Spec. Ct. for Sierra Leone, Trial Chambers II, June 20, 2007), available at http://www.refworld.org/docid/467fba742.

176

147

119

0.86

Taylor Trial 219See supra note 159.

420

115

1522

1.42

Total

4.27

The total cost of the SCSL over its lifetime is estimated to be $314 million. 220This cost was calculated by updating the methodology from Ford, supra note 3, at 993–95. It now appears that the SCSL will complete its operations at the end of 2013 with the issuance of the Appeal Judgment in the case against Charles Taylor. See Press Release, Special Court for Sierra Leone, Outreach and Public Affairs Office, Oral Hearings Conclude in Taylor Appeal, Judges Will Now Retire to Deliberate and Consider Judgment, (Jan. 23, 2013), available at http://www.sc-sl.org/LinkClick.aspx?fileticket=pTTabQ5Bm2U%3d&tabid=53. Once adjusted for inflation, that works out to $352 million in 2012 U.S. dollars. 221Past inflation was accounted for using the Consumer Price Index produced by the U.S. Department of Labor, Bureau of Labor Statistics. Future inflation was assumed to be 2% per year. This gives the SCSL an overall efficiency of 1.21 x 10-8 units of complexity per 2012 U.S. dollar, which means that it cost about $82.5 million to produce each point of trial complexity at the SCSL. In contrast, it costs only $73 million to produce each point of complexity at the ICTY, indicating that the SCSL has been 12% less efficient than the ICTY so far. Moreover, the ICTY’s efficiency is likely to increase over time.

All of the SCSL’s trials have been completed and are included in its overall efficiency rating, which is based on an estimate of its total lifetime cost. The ICTY’s efficiency rating is also based on an estimate of its total lifetime cost, but it excludes the complexity of the three trials that have not yet been completed. In other words, the SCSL’s efficiency rating is not likely to change between now and when the SCSL is formally shut down. The ICTY’s efficiency rating is likely to increase as the last three trials are completed. Ultimately, the ICTY will probably be more efficient than the current figures suggest.

In addition, the SCSL’s efficiency rating is a rating for all of the court’s work. In contrast, the ICTY’s efficiency rating excludes a significant part of the court’s work. For example, it does not include the significant resources expended in investigating and preparing for trial those cases that ended in a guilty plea. Nor does it include the trial complexity of the contempt cases or the investigative complexity of those cases and dossiers that were transferred to domestic courts in the former Yugoslavia. If there were a way to measure and account for this work, the ICTY’s efficiency rating would be even better.

The SCSL has been much cheaper than the ICTY, but that is because the Special Court has tried far fewer cases, not because it has been more efficient at trying cases. Ultimately, the SCSL has not lived up to the hopes of some of those who championed hybrid tribunals as an efficient alternative to the ad hoc tribunals. 222See Higonnet, supra note 10, at 348 (arguing that hybrid courts would be a “cost-efficient, high impact” alternative to the ICTY and ICTR); id. at 385–86 (arguing that the SCSL was “more efficient” than the ICTY); Mathias Holvoet & Paul de Hert, International Criminal Law as Global Law: An Assessment of the Hybrid Tribunals, 17 Tilburg L. Rev. 228, 232 (2012) (describing the SCSL as “effective” and “relatively cheap”); David M. Crane, Dancing with the Devil: Prosecuting West Africa’s Warlords: Building Initial Prosecution Strategy for an International Tribunal after Third World Armed Conflicts, 37 Case. W. Res. J. Int’l L. 1, 2 (2005) (arguing that the SCSL was “efficiently and effectively” delivering international criminal justice); Kelli Mundell, Capturing Women’s Experiences of Conflict: Transitional Justice in Sierra Leone, 15 Mich. St. J. Int’l L. 85, 94 (2007) (noting that the SCSL was supposed to be more cost effective and efficient than the ICTY); see generally J. Peter Pham, A Viable Model for International Criminal Justice: The Special Court for Sierra Leone, 19 N.Y. Int’l L. Rev. 37, 89–90 (2006). Of course, efficiency was not the only supposed benefit of hybrid tribunals like the SCSL, 223See, e.g., Laura A. Dickinson, The Promise of Hybrid Courts, 97 Am. J. Int’l L. 295, 306–08 (2003) (noting that hybrid tribunals may also enhance perceived legitimacy, perceived independence of the judiciary, accountability, and the creation of regional human rights norms, among other things). but it was certainly one of their key selling points. 224See Charles Chernor Jalloh, Special Court for Sierra Leone: Achieving Justice?, 32 Mich. J. Int’l L. 395, 446–51 (2011); id. at 446 (“The SCSL was specifically designed to improve on the trial efficiency of the ICTR and ICTY.”).

C. Comparison to Domestic Criminal Trials

Two issues make comparing the ICTY to domestic criminal justice systems more difficult than comparing the ICTY to the SCSL. First, very little systematic information is available on the cost and length of domestic criminal trials, either in the United States or elsewhere. 225See Wippman, supra note 14, at 863 (stating that neither federal nor state agencies routinely calculate the per trial cost and length of criminal trials). There are numerous newspaper reports, policy papers, and scholarly articles that contain information on the cost of individual parts of particular complex cases, particularly in the context of the death penalty, 226See Cook & Slawson, supra note 167, at 54 (reporting figures for the cost of murder trials in North Carolina in 1991 and 1992); The Cost of Justice, L.A. Times, Sept. 13, 1998, at 2 (containing cost figures for a number of famous trials in the Los Angeles area); Russell Gold, Counties Struggle with High Cost of Prosecuting Death-Penalty Cases Wall St. J., Jan. 9, 2002, at B1 (describing cost of death-penalty cases); Eli Braun, Ohio Just. & Pol’y Ctr., $42,000 for a Courthouse Hour: The Cost of Processing Adult Criminal Cases in Hamilton County, Ohio, (2010), available at http://www.ohiojpc.org/text/publications/court%20cost.pdf; Natasha Minsker, ACLU of N. Cal., The Hidden Death Tax: The Secret Costs of Seeking Execution in California (Claire Cooper & Elise Barducci eds., 2008), available at https://www.deathpenalty.org/downloads/The_Hidden_Death_Tax.pdf; John Roman, et al., Urb. Inst. Just. Pol’y Center, The Cost of the Death Penalty in Maryland, (2008), available at: http://www.urban.org/UploadedPDF/411625_md_death_penalty.pdf; John G. Morgan, Comptroller of the Treasury (Tenn), Tennessee’s Death Penalty: Costs and Consequences (2004); Arthur L. Alarcón & Paula M. Mitchell, Executing the Will of the Voters?: A Roadmap to Mend or End The California Legislature’s Multi-Billion Dollar Death Penalty Debacle, 44 Loy. L.A. L. Rev. S41, S72–75 (2011); Terance D. Miethe, Estimates of Time Spent in Capital and Non-Capital Murder Cases: A Statistical Analysis of Survey Data from Clark County Defense Attorneys (Feb. 21, 2012) available at http://www.deathpenaltyinfo.org/documents/ClarkNVCostReport.pdf; State of Kansas, Legislative Division of Post Audit, Costs Incurred for Death Penalty Cases: A K-GOAL Audit of the Department of Corrections (2003), available at http://www.kslpa.org/docs/reports/04pa03a.pdf; Aliza B. Kaplan, Oregon’s Death Penalty: The Practical Reality, 17 Lewis & Clark L. Rev. 1, 32–49 (2013); Jon B. Gould & Lisa Greenman, Report to the Committee on Defender Services—Judicial Conference of the United States—Update on the Cost and Quality of Defense Representation in Federal Death Penalty Cases (2010), available at http://www.uscourts.gov/uscourts/FederalCourts/AppointmentOfCounsel/FDPC2010.pdf; Margot Garey, The Cost of Taking a Life: Dollars and Sense of the Death Penalty, 18 U.C. Davis L. Rev. 1221 (1985). but the information in these sources is often incomplete. There have been few systematic studies of the cost and length of trials, and those that do exist tend to have small sample sizes. As a result, the information that is available is far from comprehensive.

The second problem stems from the fact that what cost data on domestic systems is available is usually not data on the total cost of the process from start to finish. Whereas the budgets of the international criminal courts contain data on all costs of a case except for enforcement of the sentences, data on domestic systems is often related to just one part of the total cost of the cases. The information most commonly reported relates to investigation costs, prosecution costs, defense costs and direct trial costs, but even then, it is relatively rare to find data on all four of these components for the same trial.

Moreover, the ICTY has paid for a number of services for which most domestic criminal courts do not pay. For example, a substantial portion of the ICTY’s budget was spent on translation and interpretation services. 227 See Translation and Interpretation, About the ICTY, http://www.icty.org/sid/165 (last visited Sept. 23, 2014). While domestic courts do sometimes need such services, it is extremely unlikely that they make up as large a component of costs in domestic systems as they did at the ICTY. Similarly, the ICTY has spent a lot of money bringing witnesses from the former Yugoslavia to testify in The Hague and on housing them during their stay in the Netherlands. 228See Victims and Witnesses Section, Information Booklet for ICTY Witnesses (2007) at 14, available at http://www.icty.org/x/file/About/Registry/Witnesses/witnesses_booklet_en.pdf. It has also in some cases paid to have witnesses and their families relocated to different countries. 229Id. at 22. Most domestic courts do not have to pay such high costs to obtain the testimony of witnesses. In addition, the ICTY has funded legacy costs, startup costs, and the costs of mentoring domestic criminal justice systems in the former Yugoslavia out of its budget. 230Richard Dickner & Elise Keppler, Beyond the Hague: The Challenges of International Justice, in Human Rights Watch World Report 2004, available at http://www.hrw.org/news/2004/01/26/beyond-hague-challenges-international-justice & http://www.hrw.org/legacy/wr2k4/download/10.pdf.) None of these costs have any obvious analog in most domestic court systems.

In effect, to achieve a fair comparison between the ICTY and domestic systems, any figures would have to include the same components for both systems (i.e., either both the domestic and the ICTY figures include costs of investigations or neither does, etc.). This means that we need to be able to calculate the cost of the various components of the ICTY’s trials.

1. Breaking Down the ICTY Budget by Functions

The ICTY’s budgets do not directly list the amount or percentage of the budget spent on particular functions. 231The Cost of Justice, About the ICTY, http://www.icty.org/sid/325 (last visited Sept. 23, 2014). Rather, the budgets indicate how much has been spent on different organs within the court (e.g., the Registry or the Office of the Prosecutor) and on different activities (e.g., travel costs of witnesses, rent on premises, and cost of expert witnesses). 232U.N. Secretary-General, Second Performance Report on the Budget of the International Tribunal for the Former Yugoslavia for the Biennium 2012–2013, 2–3 tbls. 1 & 2, U.N. Doc. A/68/582 (Nov. 11, 2013). To solve this problem, the Author reviewed the ICTY’s budget documents. For each year’s budget, each item in the budget was assigned to a particular function. Funding for the various organs was apportioned according to the number of personnel in that organ working on different functions. 233For example, the cost of the OTP was apportioned to investigation, prosecutions, appeals, and other tasks, depending on the number of personnel assigned to the Prosecution Division, Investigation Division, Appeals Division, or Immediate Office of the Prosecutor. The cost of the Registry was similarly apportioned between support tasks and tasks directly associated with trials and appeals depending on the number of personnel assigned to the various divisions within the Registry. In that fashion, it was possible to approximate the cost of each of the ICTY’s functions over time. 234This process was relatively easy for the ICTY budgets covering the years 1994 to 2005, as the publicly available budget documents for those years were very thorough. It was more difficult for the years 2006 to 2013 as the ICTY’s budgets have become less detailed over time. Nevertheless, it was possible to calculate the cost of the various functions of the ICTY with reasonable certainty, even for the latter years. Estimates for the years 2014–2017 are the result of educated guesses about the ways in which ICTY spending will change as the trials are completed and the focus shifts to appeals. The data that form the basis for Table 5 are available from the Author upon request. The result appears below as Table 5.

Table 5: Cost of Various ICTY Functions

Cost in 2012 dollars

Percentage

OTP Investigations

$ 350,268,000

11.37%

OTP Prosecutions

$203,761,000

6.62%

OTP Appeals

$146,206,000

4.75%

OTP Other

$81,186,000

2.64%

Direct Trial Costs

$359,698,000

11.68%

Direct Appeal Costs

$126,589,000

4.11%

Witness & Victim Costs

$100,205,000

3.25%

Interpretation & Translation

$339,603,000

11.03%

Legal Aid – Trials

$253,883,000

8.24%

Legal Aid – Appeals

$82,419,000

2.68%

Detention

$143,613,000

4.66%

Security

$212,305,000

6.89%

Registry Support Functions

$616,355,000

20.01%

Startup Costs

$42,501,000

1.38%

Legacy

$15,597,000

0.51%

Excluded

$5,537,000

0.18%

The largest component of the ICTY’s cost is for functions that support the judicial activities of the court. These are labeled “Registry Support Functions” in the Table. This category covers any non-judicial cost that is not covered by any other category. It includes the cost of support personnel within the Registry (e.g., personnel working on procurement, travel, human resources, finance, building maintenance, information technology, etc.). It also covers the cost of operating the court buildings, including the rent on the premises, as well as utilities, equipment, furniture and supplies. Startup costs, including construction work necessary to convert the main building into a court, equipment purchased to get the court off the ground, and some personnel costs in the court’s first year of existence, were calculated separately.

The security category includes the cost of the guards that ensure the safety of the court buildings, the staff, and the public. It does not include the cost of the detention center, which is calculated separately. The ICTY operates its own detention facility to house accused that are in pre-trial detention. The detention category includes the cost of rent and guards for the detention center, as well as medical care for the detainees.

In its early years, the ICTY Prosecutor also acted as the Prosecutor for the ICTR. As a result, some personnel within the ICTY’s OTP were working on ICTR cases. This cost has been separated and labeled “Excluded” in the table above. In its later years, the court also dedicated some resources to preparing its records and archives for the closure of the court, so that the information would not be lost when the court ceased to function. That cost has been labeled “Legacy.” All told, these functions, which are not directly related to judicial activities, account for about 34% of the court’s total budget over the years 1993-2017.

The remaining 66% of the court’s budget was spent on things that were directly related to the judicial process. The largest component of this is direct trial costs. This category includes the cost of the judges and their clerks, plus the cost of the personnel that staff the courtroom when it is in session (ushers, court reporters, etc.). This category also includes the majority of the judicial costs of pre-trial practice. 235There was no way to extract the pre-trial costs from the trial costs based on the information in the ICTY’s budgets. The direct appeals category includes the cost of the judges of the Appeals Chamber plus their clerks. 236Costs were apportioned between direct trial costs and direct appeal costs by taking total judicial costs (essentially the sum of the cost of the Chambers, the Judicial Support Division of the Registry plus any items related to judicial costs) and then prorating those costs based upon the number of judges who were assigned to trials versus appeals in any given year.

The witnesses and victims category includes the cost of arranging for the witnesses to appear before the tribunal. The court is located in The Hague, but most of the witnesses live in the former Yugoslavia. Thus, the ICTY bore considerable costs associated with bringing the witnesses before the court. 237See Wippman, supra note 14, at 877–78. The largest portion of this cost was travel costs, although it also included compensation for lost wages, medical costs, the cost of relocating some witnesses and their families, and the cost of the Victim and Witness Support Section within the Registry.

Interpretation and translation was a mammoth task at the ICTY. 238Id. at 877. The court worked in two official languages, English and French. 239Id. In addition, the majority of the witnesses and most of the documents were in a mix of languages spoken in the former Yugoslavia, including Bosnian, Serbian, Croat and Albanian. 240Id. The result was that interpreters and translators were crucial at all stages of the court’s work from the investigation stage, when translators produced English translations of documents for the investigators, to the courtroom, where interpreters provided English and French language interpretation of the testimony of the witnesses. Those costs are collected together in the interpretation and translation category.

The various costs of the Office of the Prosecutor (“OTP”) were put in one of four categories. The largest of these categories is the cost of investigations. This category includes the cost of the personnel involved in the investigations, their travel costs and various special costs associated with the investigations, like the cost of exhumations of mass graves. The OTP prosecutions cost includes pre-trial costs and the cost of the trials. It is largely the sum of the costs of the personnel in the Prosecution Division of the OTP plus miscellaneous costs associated with the trials, including travel costs, expert witness costs, and costs for temporary assistance provided during peak trial times. The OTP appeals cost represents the cost of those personnel assigned to the Appeals Division of the OTP. The “other” category includes a number of OTP functions that were not directly related to the investigations and trials, including personnel assigned to manage the OTP, train and mentor personnel in domestic criminal systems in the former Yugoslavia, and provide political analysis of events in the former Yugoslavia.

The last two categories relate to legal aid. They capture the cost of providing defense to indigent and partially indigent accused throughout the lifecycle of the court. In theory, accused are expected to pay for their own defense. In practice, the vast majority of accused have depended on legal aid, and a defense team costs approximately $360,000 per year per accused. 241See U.N. Secretary-General, Comprehensive Report on the Progress made by the International Criminal Tribunal for the Former Yugoslavia in Reforming its Legal Aid System, 31, U.N. Doc. A/58/288 (Aug. 12, 2003). Legal aid includes payments to the defense counsel as well as the cost of Office of Legal Aid within the Registry. The costs were divided between those associated with pre-trial and trial functions, and those associated with appeals. 242Id. 31, Annex I. As with direct trial and appeal costs, the total amount of legal aid was apportioned between trials and appeals based on the number of judges assigned to each of those functions in any given year.

2. Domestic Criminal Trials

With this information, it is now possible to make comparisons between the cost of the ICTY and the cost of criminal trials in the United States. Given the scarcity of data, comparisons will be made to both the small number of systematic studies that exist as well as a number of individual cases for which cost data is available. The latter category tends to feature high-profile cases because these are the type of cases for which there is sufficient public interest for journalists to investigate and report on trial costs.

a. Prosecuting Timothy McVeigh

On April 19, 1995, Timothy McVeigh and his accomplice, Terry Lynn Nichols, set off a bomb outside the Murrah Building in Oklahoma City, Oklahoma, that killed 168 people and injured numerous more. 243 See United States v. McVeigh, 153 F.3d 1166, 1176 (10th Cir. 1998). The cost of investigating and prosecuting Timothy McVeigh for the Oklahoma City bombing was $82.5 million between 1995 and 1998. 244See Wippman, supra note 14, at 862; Howard Pankratz, Bomb Trials Costs $82.5 Million, denverpost.com, Nov. 3, 1998, available at http://extras.denverpost.com/bomb/bomb1103.htm [hereinafter Bomb Trials]. After adjusting for inflation, the investigation and prosecution cost $124.3 million in 2012 U.S. dollars. 245See Bomb Trials, supra note 244 (noting that the FBI had 2,592 agents investigating the case in 1995 and that this number decreased rapidly over time). The 2012 cost was assumed to accrue in 1995 because it is clear that the vast majority of resources were expended during that year. (Author’s Calculation). Defense costs for McVeigh were more than $15 million. 246Bomb Trials, supra note 244. This translates into $21.5 million in 2012 U.S. dollars. 247Defense costs were assumed to accrue in 1997, the year of the trial. (Author’s Calculation). McVeigh’s trial took 23 days for the presentation of evidence during the guilt phase. 248McVeigh, 153 F.3d at 1177. The penalty phase took another 7 days. 249Id. at 1179. The presentation of evidence for the penalty phase began on June 4, 1997, and ended on June 12, 1997. Id. Assuming that the court did not hear evidence over the weekend, this means that the penalty phase took seven days. As a result, the entire trial took 30 trial days to present.

Using this information, we can calculate what the equivalent portion of an ICTY trial of the same length would cost. 250This calculation is based solely on trial days, rather than all of the components of the Complexity Score, because information is not readily available about the number of witnesses and exhibits presented during the McVeigh trial. If a complexity calculation is to be based on just one of the components, the number of trial days is the best component to use. See Heise, supra note 33, at 360. The investigation and prosecution portion of a 30-day trial at the ICTY would cost $2.2 million in 2012 U.S. dollars. 251This was calculated as follows: equivalent cost = (number of trial days in McVeigh trial)÷(total number of trial days in all ICTY trials)×(total cost of ICTY)×(percentage of cost attributable to prosecutions + percentage of cost attributable to investigations). With the actual figures it becomes: (30)÷(7691)×(3079726477)×(0.0662+0.1137). Subsequent calculations of the equivalent cost of an ICTY trial use an analogous methodology. The legal aid portion of a 30-day trial at the ICTY would cost $990,000 in 2012 U.S. dollars. Investigation and prosecution costs were more than fifty times higher for the McVeigh trial than an equivalent ICTY case. Defense costs were more than twenty times higher. No matter how one slices it, the cost of the McVeigh trial appears to have outweighed the cost of a similar trial at the ICTY by more than an order of magnitude. 252Professor Wippman suggests that the McVeigh trial was unusually short given its complexity due to tactical decisions made by the prosecution and defense to call a smaller number of witnesses and use fewer exhibits than either side had initially intended. See Wippman, supra note 14, at 873. Even if the prosecution and defense had chosen differently and thus doubled the length of the trial, the basic conclusion about the trial’s efficiency would remain the same.

The U.S. government was keenly aware of the high cost of the McVeigh trial. The Department of Justice acknowledged that it was the largest amount it had ever spent on an investigation and prosecution, but defended the expenditure on the grounds that bombing was “the largest terrorist act in the history of the United States” up until then. 253Bomb Trials, supra note 244. The spokesperson for the Department of Justice stated that “[w]e believe the cost is justified and that the taxpayers got their money’s worth.” 254Id.

b. Defending Jeffrey Skilling

In May 2006, former Enron CEO Jeffrey K. Skilling was convicted of conspiracy, securities fraud, making false representations to auditors, and insider trading in connection with the 2001 collapse of Enron Corporation. 255See United States v. Jeffrey K. Skilling, 638 F.3d 480, 481 (5th Cir. 2011). Total defense costs for Mr. Skilling were estimated at more than $65 million. 256See Carrie Johnson, After the Enron Trial, Defense Firm is Stuck With the Tab Wash. Post, June 16, 2006, at D1. This included the costs of his five trial attorneys, a team of associates and paralegals who helped prepare the defense, jury consultants, graphics specialists, and expert witnesses. 257Id. at D3. Adjusted for inflation, that means the defense spent $74 million in 2012 U.S. dollars. The trial took 60 trial days to complete 258See Brief for the United States as Appellee at 5, United States v. Jeffrey K. Skilling, 554 F.3d 529 (5th Cir. Nov. 13, 2007) (No. 06-20885), 2007 WL 4207556. and involved “some of the most sophisticated business deals ever to make their way into a courtroom.” 259See Johnson, supra note 248, at D1. In comparison, the defense costs of a 60-day trial at the ICTY would be approximately $1.98 million in 2012 U.S. dollars. Again, the ICTY costs are much smaller. In this case, they are more than thirty times smaller.

c. Prosecuting O.J. Simpson

On October 3, 1995, O.J. Simpson was acquitted of the murders of his ex-wife Nicole Brown Simpson and Ronald Lyle Goldman. 260See James Sterngold, Not Guilty: The Man; Simpson Walks Out of the Courtroom a Free Man and Into the Lucrative Free Market, N.Y. Times, Oct. 4, 1995, at A10. There is sufficient information to calculate a Complexity Score for the trial. 261According to USA Today, the trial lasted 141 trial days (including opening and closing statements), 126 witnesses testified, and 857 exhibits were entered into evidence. See The O.J. Simpson trial, by the numbers, USA Today (Oct. 18, 1996), http://usatoday30.usatoday.com/news/index/nns062.htm. The O.J. Simpson trial has a Complexity Score of 0.79. This means the Simpson trial was about as complex as the median ICTY trial. The trial reportedly cost Los Angeles County about $9 million for the direct trial costs and the prosecution costs. 262See id. This would be the equivalent of $13.6 million in 2012 U.S. dollars. By comparison, the direct trial costs and prosecution costs of a case with a Complexity Score of 0.79 would cost $10.6 million in 2012 U.S. dollars at the ICTY. Once again, it appears the ICTY trial would be cheaper, but the margin is much smaller.

d. Prosecuting Scott Peterson

In 2004, Scott Peterson was tried and eventually sentenced to death for the murder of his wife, Laci Peterson, and their unborn son; opening statements in the guilt phase began on June 1, 2004, and the state rested its case on October 5, 2004. 263See The State’s Case: Jun 1–Oct 15, 2004, PWC Consulting, http://pwc-sii.com/Timeline/State.htm (last visited Sept. 19, 2014) [hereinafter The State’s Case]. Note that this website is not an impartial source; it is devoted to the idea that Scott Peterson is innocent. Nevertheless, it appears to be reliable for the purpose of identifying the number of trial days. The defense began its case on October 18, 2004 and finished on October 26, 2004. 264See The Defendant’s Case, Jury Deliberations, & Verdict: Oct 6–Nov 12, 2004, PWC Consulting, http://pwc-sii.com/Timeline/Defense.htm (last visited Sept. 19, 2014). The penalty phase began on November 30, 2004, and ended on December 9, 2004. 265See Penalty Phase, Motions, & Sentencing: Nov 13, 2004–Mar 17, 2005, PWC Consulting, http://pwc-sii.com/Timeline/Penalty.htm (last visited Sept. 19, 2014). The court sat in session four days a week during that time. 266The State’s Case, supra note 263. They indicate that the court heard testimony Mondays through Thursdays but was not in session on Fridays. Id. There were also some additional days when the court was not in session. These days included July 1st, July 5th, July 8th, August 9th, August 19th, and September 6th. Id. All told, the Peterson trial took 81 trial days. According to one study, based on records kept by the State of California, the cost to prosecute the Peterson case was $1.4 million ($1.7 million in 2012 dollars). 267See Minsker, supra note 226, at 21. The equivalent cost to prosecute a case of the same length at the ICTY would be $2.1 million in 2012 dollars. Here, the ICTY costs are higher than the domestic costs, but not by a huge margin.

e. Murder Trials in North Carolina

In 1993, a group of researchers attempted to ascertain the cost of capital and non-capital murder trials in the state of North Carolina. 268See Cook & Slawson, supra note 167, at 47 (reporting figures for the cost of murder trials in North Carolina in 1991 and 1992). They calculated the average cost of murder cases including the cost of defense, prosecution and courtroom time. 269Id. at 45. Their calculation excludes investigative costs, as well as judicial costs not directly related to courtroom time (e.g., deliberating, researching, and writing opinions). 270Cf. id. In this way, their calculation does not include the total judicial cost of the trial, which the ICTY figures do, but there is no way to account for this discrepancy. They estimated trial length and costs of: (1) 3.8 trial days and $16,697 ($27,323 in 2012 dollars) for non-capital murder trials; (2) 10.6 days and $57,290 ($93,752 in 2012 dollars) for capital trials that ended after the guilt phase; and (3) 14.6 days and $84,099 ($137,623 in 2012 dollars) for a bifurcated capital trial. 271Id. at tbls 6.2 & 6.3.

Their calculations are not exactly comparable to the ICTY figures. First of all, their calculation of judicial costs excludes judicial time outside of the courtroom. Second, they included time spent on pre-trial motions as trial time, 272Id. at 96. whereas the ICTY figures do not count pre-trial hearings as trial time. These factors are likely to improve their estimated efficiency as compared to the ICTY. Nevertheless, comparable ICTY figures for the cost of prosecution, defense and direct trial costs are: (1) $403,000 for a 3.8 day trial; (2) $1,113,000 for a 10.6 day trial; and (3) $1,552,000 for a 14.6 day trial. The ICTY trials are an order of magnitude more expensive than the average murder trial in North Carolina in the early 1990s. Even given the differences in the way the numbers are calculated, it seems clear that ICTY trials are significantly less efficient.

f. Murder Trials in Maryland

A similar study was carried out in 2008 on the cost of death eligible murder trials in Maryland. 273See Roman, et al., supra note 226. In that study, the researchers concluded that the cost and trial length of an average murder trial was: (1) 3.1 days and $158,000 ($175,000 in 2012 dollars) for a non-capital murder trial; (2) 10.2 days and $672,000 ($743,000 in 2012 dollars) for a case where a death notice was filed; and (3) 13.2 days and $1,038,000 ($1,150,000 in 2012 dollars) for a case that resulted in a death sentence. 274Id. at tbls. 5 & 6. This calculation includes the cost of prosecution, defense and direct trial costs. 275Id. at app. B (describing how the cost data was calculated).

Equivalent costs at the ICTY would be: (1) $329,000 for a 3.1 day trial; (2) $1,047,000 for a 10.2 day trial; and (3) $1,355,000 for a 13.2 day trial. Here, the ICTY costs are higher than the equivalent costs in Maryland, but not by nearly as much as in the North Carolina study. Moreover, the difference is smaller for the cases where the prosecuting attorney gave notice that the death penalty would be sought (41%) than for cases where it was not sought (88%). Finally, the difference is smallest (31%) when comparing the ICTY cases to cases that resulted in a death penalty. To put it another way, the efficiency of the Maryland cases decreased as their complexity increased. This suggests that cost does not scale linearly with complexity.

g. Murder Trials in Kansas

In 2003 the Kansas state auditors estimated the cost of death penalty trials in Kansas. 276Kan. Legis. Division of Post Audit, Costs Incurred for Death Penalty Cases: A K-GOAL Audit of the Department of Corrections (2003), available at http://www.kslpa.org/docs/reports/04pa03a.pdf. Their report contains sufficient information to compare costs in Kansas to the ICTY. They estimated that the average cost and length of murder trials was: (1) $110,000 ($137,000 in 2012 dollars) and 7.4 trial days for a murder trial; (2) $291,000 ($363,000 in 2012 dollars) and 19 trial days for a case where the death penalty was requested but not imposed; and (3) $744,000 ($928,000 in 2012 dollars) and 22 trial days for cases in which the death penalty was imposed. 277Data used to calculate these figures was taken from id. at Chart I-2 and Appendix E. These figures include the direct trial costs, the defense costs, the prosecution costs, and investigation costs.

Equivalent costs for the ICTY would be: (1) $1,233,000 for a 7.4 day trial; (2) $2.9 million for a trial lasting 19 days; and (3) $3.3 million for a trial lasting 22 days. The ICTY is an order of magnitude more expensive than Kansas courts in prosecuting average murder trials. But, like the results for Maryland, as the cases get more complex, the differential gets smaller. When considering cases where the death penalty was imposed, the ICTY is only 3.5 times more expensive than the Kansas courts. Again, it seems that cost does not scale linearly with complexity.

h. Conclusion

One thing that is clear from the domestic data is that even after adjusting for complexity, there is no such thing as a general efficiency rating for domestic courts. Rather, there is great variability in the efficiency of courts in the United States. 278See Cook & Slawson, supra note 167, at 6; Roman et al., supra note 226, at tbl.1. Even if one considers just average murder trials, there are huge differences in cost and efficiency. North Carolina reported an average cost of $27,000, while Maryland reported an average cost of $175,000 (both figures in 2012 dollars). There are at least two factors that could partially explain the difference in cost: (1) time, and (2) location. In general, the cheapest figures are for the oldest studies. In this regard, the North Carolina study (1993) reports the lowest costs, the Kansas study (2003) reports higher costs, and the Maryland study (2008) reports the highest costs, even after adjusting for inflation. This may suggest that trial costs have been increasing faster than inflation. In addition, there are probably differences depending on where the trial takes place. Part of the reason that trial costs in Maryland are more expensive than in North Carolina is because the cost of living is higher in Maryland. 279For example, the cost of supporting a family consisting of two adults and two children in North Carolina is roughly $39,000 in North Carolina and $47,000 in Maryland. Amy K. Glasmeier, Living Wage Calculator, M.I.T., http://livingwage.mit.edu/ (last updated Mar. 24, 2014).

Complexity does not explain the differences in cost, as the average length of a murder trial in North Carolina (3.8 days) was very similar to the average length of a murder trial in Maryland (3.1 days). In the end, neither time, location or complexity seems capable of explaining the more than sixfold difference between the cost of a murder trial in North Carolina and Maryland. This suggests that efficiency is the explanation and that there are real differences in efficiency across different jurisdictions within the United States, even for average cases.

If we switch from looking at efficiency across states to looking at intra-state efficiency, the three studies that compare the costs of average murder trials to death penalty trials in the same location at the same time indicate that cost does not increase linearly with complexity. This can be seen in Table 6, where after compensating for trial complexity, death penalty cases are less efficient in Kansas, North Carolina, and Maryland than ordinary murder trials, sometimes by a large amount. The efficiency ratings for the high profile murder trials of O.J. Simpson and Scott Peterson are consistent with this finding. Both were significantly more complex than the average death penalty case in North Carolina, Kansas, or Maryland. Once adjusted for their complexity, the efficiency of these high-profile murder trials was fairly close to that of the ICTY, and generally less than that of the average death penalty case. Taken together, this data suggests that it is a general rule that efficiency will decrease as complexity increases.

Table 6: Relative Efficiency of Various Courts and Trials

Relative Efficiency

Murder Trial in North Carolina

14.9

Guilt Phase Death Penalty Trial in NC

12

Bifurcated Death Penalty Trial in NC

11.3

Murder Trial in Kansas

9

Death Penalty Sought in Kansas

7.99

Death Penalty Imposed in Kansas

3.56

Murder Trial in Maryland

1.88

Death Penalty Sought in Maryland

1.41

Scott Peterson Prosecution

1.24

Death Penalty Imposed in Maryland

1.18

ICTY

1

SCSL

0.89

O.J. Simpson Trial

0.78

Timothy McVeigh Defense

0.05

Jeffrey Skilling Defense

0.03

Timothy McVeigh Prosecution

0.02

Table 6 shows the relative efficiency of the various courts and trials discussed above. All results are comparisons to the ICTY’s efficiency, which was set at 1. Higher numbers mean greater efficiency, lower numbers mean less efficiency. Table 6 demonstrates the massive differences in efficiency across trials in the United States. The most efficient trials considered in this Article, murder trials in North Carolina, are 745 times more efficient than the Timothy McVeigh prosecution, even once you adjust for their relative complexity. The ICTY is in the bottom half of the list, but it is not last. Indeed, it is more efficient than trials at the SCSL, as well as high profile domestic trials like those of O.J. Simpson, Jeffrey Skilling, and Timothy McVeigh. It is also worth pointing out again that the efficiency rating for the ICTY underestimates its true efficiency for a number of reasons, including that it does not incorporate the complexity of the three remaining cases or the complexity associated with the guilty pleas obtained at the ICTY. 280See supra Part III.C.

The question then becomes not whether the ICTY is more efficient than domestic criminal trials, but rather which domestic criminal trials should the ICTY be compared to? If we compare the ICTY’s trials to the average murder trial, the ICTY will look grossly inefficient. If we compare it to the average death penalty trial, the ICTY will be slightly less efficient. If we compare it to high profile murder trials, the ICTY has roughly the same efficiency. On the other hand, if we compare the ICTY to extremely complex white collar crime cases like that of Jeffrey Skilling or to mass atrocity crimes like those of Timothy McVeigh, the ICTY appears to be a model of efficiency.

The ICTY’s trials should not be compared to average murder trials for a number of reasons. To begin with, ordinary murder trials usually involve a single perpetrator, 281See Ford, supra note 71, at 68 n.92. a relatively small number of counts, and are concerned with acts that take place in a relatively small period of time and at a small number of locations. Accused at the ICTY are much more likely to be tried in multi-accused trials, they are likely to be charged with many more counts than domestic accused, 282See supra Part II.A. and the alleged crimes are likely to have taken place at many more locations and over longer periods of time. 283See supra Part I.D. Moreover, such crimes are usually undertaken by hierarchical groups working together and proving them often requires proof of the criminal acts of individuals other than the accused. 284Id. In short, the trials at the ICTY look very little like the average murder trial before a domestic court.

In addition, cost seems to increase faster than complexity. One result is that efficiency decreases as complexity increases. This suggests that comparing the efficiency of cases that have vastly different complexities is not a fair comparison. The less complex cases will appear to be more efficient. A fair comparison requires the cases to be of roughly equivalent complexity. Thus, the average murder trial is not a fair comparator for the ICTY’s caseload because they are much less complex than ICTY cases. The average murder trial took 7.4 days in Kansas, 3.8 days in North Carolina, and 3.1 days in Maryland. 285See Cook & Slawson, supra note 167, at 49; Roman, et al., supra note 226, at 24; Kan. Legis. Division of Post Audit, supra note 276, at 18. Even the average death penalty case is not a fair comparator. The data from Kansas, North Carolina, and Maryland suggest that the average trial that resulted in a death penalty took between 13 and 22 trial days to complete. 286See Cook & Slawson, supra note 167; Roman, et al., supra note 233; Kan. Legis. Division of Post Audit, supra note 276. The average ICTY case takes 176 trial days to resolve.

Finally, the gravity of the crimes charged in ICTY cases is far greater than in the average murder trial. The average murder trial in a domestic system involves a single victim. The typical ICTY case involved allegations of the unlawful killing of more than 100 people, often combined with other unlawful acts like the torture, mistreatment, imprisonment, or forcible transfer of large numbers of people. 287See Ford, supra note 71, at 76, 104. Only the most serious mass atrocity crimes tried in domestic criminal justice systems approach the complexity and gravity of the average ICTY trial. 288See Cohen & Kyckelhahn supra note 134; id. at 91. This suggests that the most appropriate comparison is between the ICTY’s cases and domestic mass atrocity trials. And by this comparison, the ICTY appears to be vastly more efficient.

My sample includes cost data on only a single U.S. mass atrocity trial that is similar in gravity to the average ICTY case—the trial of Timothy McVeigh for the Oklahoma City bombing—but I suspect that other mass atrocity crimes investigated and prosecuted in the United States have equally large costs. For example, there is no data yet on the cost of the investigation of the bombing of the USS Cole in Yemen in 2000, although the government may eventually be required to disclose that information. 289See Carol Rosenberg, Lawyers Spar Over Funding in Guantanamo Court, Miami Herald, Apr. 12, 2012, http://www.miamiherald.com/760/index.html. Nevertheless, the investigation is described as a decade-long affair that involved personnel from the FBI, Naval Criminal Intelligence Service, CIA, and prosecutors from New York, the Justice Department, and the Pentagon. 290Id. There is also little concrete information on the cost of the government’s response to the bombing of U.S. embassies in East Africa in 1998, but what little information exists suggests a similarly massive investigation. 291The initial investigation was described as involving more than 1,000 FBI employees who were sent to East Africa, including more than 500 FBI agents. See Wippman, supra note 14, at n.80. Finally, although the U.S. response to the terrorist bombings on September 11, 2001, was not principally legal in nature, it can certainly be said that the cost of the response was not an overriding concern. 292See, e.g., Joseph E. Stiglitz & Linda J. Bilmes, The Three Trillion Dollar War: The True Cost of the Iraq Conflict 7–8 (2008) (“The tone of the administration was cavalier, as if the sums involved were minimal.”). This evidence is not definitive, but it suggests that when the victims of mass atrocities are Americans, we are willing to spend far more than the ICTY did to investigate and prosecute them.

Nor are we alone in this approach to mass atrocities. While cost data on trials in other countries’ domestic systems is also difficult to find, some data is available on the trial of Anders Breivik, who was accused of killing 77 people and wounding 42 more in Norway in July 2011. 293See Walter Gibbs & Elinor Schang, Norway Police To Track Killer’s Funds in Year-Long Probe, Reuters Norway (Aug. 3, 2011), http://in.reuters.com/article/2011/08/03/idINIndia-58607220110803. In addition, some information is available on the trial of the individuals accused of carrying out the bombing of trains in Madrid in 2004 that killed 192 and injured an additional 1,400. 294See Ford, supra note 71, at 79, 100.

The investigation of Anders Breivik’s crimes took more than one year and the crimes were investigated by a specially created police force comprised of more than 100 investigators with specialties in evidence, victims, witnesses and international cooperation. 295See Gibbs & Schang, supra note 284. The investigation was described as “the biggest in Norway’s history.” 296Id. The trial began on April 16, 2012, and concluded on June 22, 2012. 297See Timeline: Anders Breivik, Aljazeera, http://www.aljazeera.com/news/europe/2012/08/201282314116683446.html (last modified Aug. 24, 2012). Thus it lasted 50 days. It was described at the “most serious criminal trial” in Norway’s history by the chief justice of the court. 298See Alastair Reed, Kristin Myers & Josiane Kremer, Breivik Claims Self-Defense as Oslo Terror Trial Starts, Bloomberg (Apr. 16, 2013), http://www.bloomberg.com/news/2012-04-16/breivik-defends-massacre-as-norway-s-saddest-trial-starts.html. According to one source, the total cost of the legal proceedings against Breivik at the completion of the trial was 165 million Norwegian Kroners. 299See Breivik’s Attacks Cost Billions, newsinenglish.no (Aug. 24, 2012), http://www.newsinenglish.no/2012/08/24/breiviks-attacks-cost-billions/. This is the equivalent of approximately $27 million. 300This was calculated using an exchange rate of .1689 Kroner to the Dollar. See Historical Exchange Rates, http://www.oanda.com/currency/historical-rates/ (last visited Sept. 26, 2013). It appears that this figure includes the cost of the prosecutors, defense counsel, victim and witness participation, security, detention, and direct trial costs. 301See Breivik’s Attacks Cost Billions, supra note 299. Various other articles contain some cost data, but none seem as comprehensive. See Reed, Myers & Kremer, supra note 298 (noting that the court estimated before the trial began that the cost of court administration during the proceeding would be 76 million kroner). The cost of an equivalent trial at the ICTY including these components would be $8.3 million. Thus, the Breivik trial ended up being more than three times as expensive as an equivalent trial at the ICTY.

The trial of the Madrid train bombers took approximately four months and was described at the time as “Europe’s biggest terrorism court case.” 302Madrid Train Bombings Trial Wraps Up, L.A. Times, July 3, 2007, at 8, available at http://articles.latimes.com/2007/jul/03/world/fg-briefs3.2. A four-month trial equates to about 80 trial days. An 80-day trial at the ICTY would cost about $2.1 million to prosecute with another $3.7 million in direct trial costs. No data was available on the cost of the investigation, but the Spanish Ministry of Justice did release a partial accounting of the costs of the trial. Unfortunately, the figures do not include some very important costs, like the salaries of the judges or prosecutors. Nevertheless, the Spanish government reported that even these partial costs of the trial equated to €3.1 million with one month left to go in the trial. 303See Paloma Díaz Sotero, Desglose Económico del Ministerio: ¿Cuánto Cuesta el Macrojuicio del 11-M?, El Mundo, http://www.elmundo.es/elmundo/2007/06/01/espana/1180718228.html (last updated June 2, 2007). This is the equivalent of $4.7 million in 2012 dollars. 304The exchange rate for July 2007 was 1.3704 Euros to the Dollar. See Historical Exchange Rates, supra note 300. Inflation data came from the Bureau of Labor Statistics. See Consumer Price Index – All Urban Consumers, available at http://www.bls.gov/cpi/cpid07av.pdf. In short, with one month of the trial still to complete, the partial cost of the Madrid trial reported by the Spanish government excluding salaries ($4.7 million) was almost the same as the total cost of a similar trial at the ICTY including salaries ($5.8 million). Given this, it seems almost certain that the total cost of the Madrid trial exceeded the cost of an equivalent trial at the ICTY.

When the ICTY is compared to those domestic trials that most resemble its caseload—mass atrocity trials—the ICTY looks fairly efficient. In fact, the constant questions about the cost and efficiency of the ICTY seem to represent a subtle form of discrimination between “us” and “them.” The data on mass atrocity crimes in the United States suggests that had the atrocities in the former Yugoslavia been perpetrated against Americans, we would gladly have spent much more than $3 billion to investigate and prosecute them. In this regard, the Department of Justice was adamant that the money spent on the McVeigh trial was well spent because of the nature and gravity of his crime, despite the fact that it was approximately fifty times less efficient than trials at the ICTY. 305See Bomb Trials, supra note 244. The disparate expenditures on atrocity crimes when the victims are Americans versus non-Americans suggests that we believe the lives of others are worth less of an investment in justice. This is consistent with research showing that we generally tend to value foreign lives less than domestic lives. 306See, e.g., Arden Rowell & Lesley Wexler, Valuing Foreign Lives, 48 Ga. L. Rev. (2014) 499, 539 (describing how U.S. law consistently assigns lower values to foreign lives than domestic lives). While this Article focused on a comparison with domestic criminal trials in the United States, it seems likely that a similar dynamic would apply in other countries. 307See supra notes 295–304 and accompanying text.

Thus one may reverse the initial question and ask not why we spend so much on the ICTY, but rather why we spend so little on it compared to what we would spend if atrocities on a similar scale occurred here. The answer may be that humans have a tendency to distinguish between “us” and “them” and treat victims of our own group better than victims of other groups. 308See Naomi Ellemers & S. Alexander Haslam, Social Identity Theory, in 2 Handbook of Theories of Social Psychology 379, 380 (Paul A.M. Van Lange et al. eds., 2012). See also Rowell & Wexler, supra note 306, at Section II(B)(4). On the other hand, some argue that we have little or no obligation to those outside of our own borders. 309See Rowell & Wexler, supra note 306, at 519 (discussing differing philosophical views of whether states owe obligations to individuals outside their borders).

Conclusion

It is common to claim that the ICTY has cost too much and taken too long, 310See supra notes 5–10. but those who have made this claim have often not made a serious attempt to understand the complexity of the cases the ICTY tries. As this Article shows, trials at the ICTY are orders of magnitude more complex than most domestic criminal trials. Moreover, the gravity of the crimes alleged in even the average ICTY case would make them among the most serious crimes ever prosecuted in the United States. 311 See Ford, supra note 71, at 104 (noting that the average ICTY indictment contains allegations roughly equivalent to the bombing of the Alfred P. Murrah Federal Building in Oklahoma City). Finally, whereas almost all criminal prosecutions in the US are resolved through a plea deal, more than 80% of the accused at the ICTY receive a full trial. The result is a system of unparalleled complexity. In fact, the ICTY appears to have prosecuted the most complex set of related criminal cases ever attempted. Thus, it is hardly surprising that it takes a long time and costs a lot of money to try these cases.

Moreover, there is little evidence that the ICTY is inefficient. For one thing, the ICTY has been more efficient than the SCSL, even though the SCSL was partially designed to be a response to the alleged inefficiencies of the ICTY. The SCSL was certainly cheaper than the ICTY, but only because the court has tried fewer cases, not because it was more efficient. This has implications for a long-running debate about whether hybrid tribunals are a better model than the ad hoc tribunals. One of the principal arguments in favor of the hybrid tribunals has been that they will be more efficient than the ad hocs. 312See supra notes 222–224 and accompanying text. This turns out not be true for the SCSL. The arguments in favor of hybrid tribunals have never been solely about efficiency. Nevertheless, in light of my findings, it is harder to argue that the hybrid model is superior to the ad hoc model.

Comparing the efficiency of the ICTY to domestic criminal trials highlights the need to identify which kind of domestic trials to compare it to. The ICTY has been less efficient than the average murder trial, but the average murder trial is not a good basis for evaluating the efficiency of the ICTY because the average murder trial looks nothing like the average ICTY prosecution. Even complex, high-profile murder cases are not comparable to the ICTY’s cases, although they are closer. The only domestic cases that are comparable are those involving the prosecution of mass atrocities. Such trials are rare and published cost data is even rarer, which makes any conclusions somewhat tentative. Nevertheless, the ICTY appears to be much more efficient than the prosecution of comparable mass atrocity cases in the United States. It also appears to be more efficient than mass atrocity trials in Europe. Thus the conventional wisdom—that the ICTY’s trials have been too slow and cost too much—turns out not to be true.

We are still left, however, with the fact that states and policymakers view the ICTY (and international criminal justice more generally) as too expensive. 313See supra notes 5–13. My findings suggest that we are unlikely to significantly reduce the cost of international criminal trials by improving their efficiency because there is little evidence that they are inherently inefficient. Our alternatives are limited.

We could simply decide not to try serious violations of international criminal law before international courts, which would save several hundred million dollars a year. 314See Ford, supra note 3, at fig.2 (charting spending on international criminal tribunals over time). However, the international community has repeatedly committed itself to prosecuting these crimes, creating a host of situation-specific tribunals like the ICTY, ICTR, SCSL and ECCC. Moreover, it has created a permanent International Criminal Court and declared that “the most serious crimes of concern to the international community as a whole must not go unpunished and [] their effective prosecution must be ensured.” 315See United Nations Diplomatic Conference on the Establishment of an International Criminal Court, Rome, July 17, 1998, Rome Statute of the International Criminal Court, Preamble, U.N. Doc. A/CONF.183/9, available at http://www.icc-cpi.int/iccdocs/PIDS/publications/RomeStatutEng.pdf. In light of that commitment, it seems unlikely that we will stop prosecuting violations of international criminal law anytime soon.

A second alternative might be to further streamline the trials, but it is doubtful whether this would be successful. Previous attempts to modify the procedural rules to improve efficiency have raised questions about whether the court is favoring expediency at the expense of fairness while simultaneously failing to significantly shorten the trials. 316See supra Part IV.A. Additional changes probably would not save a huge amount of money, but may well jeopardize the fairness of the proceedings. Judge Wald has argued that the fairness of the ICTY’s trials was one of its greatest achievements, 317See Wald, supra note 16, at 466. and it seems unwise to sacrifice this to save a small amount of money. Fairness may also be central to the court’s legitimacy and its ability to persuade others to comply with its norms. 318See Tom Tyler, Legitimacy and Compliance: The Virtues of Self-Regulation, in Legitimacy and Compliance in Criminal Justice 8, 11 (Adam Crawford & Anthea Hucklesbury eds., 2013); Tom R. Tyler, Legitimacy and Rule Adherence: A Psychological Perspective on the Antecedents and Consequences of Legitimacy, in 11 The Psychology of Justice and Legitimacy 251 (D. Ramona Bobocel et al. eds., 2010). If this is true, then making the process less fair to save a relatively small amount of money would be counterproductive because it would undermine the long term goals of international criminal justice.

A third alternative might be to prosecute violations of international criminal law but have much simpler charges and trials by trying individuals only for one or two “symbolic” crimes. This would be a radical departure from the practice of the ICTY (and most other international courts), which has been to use representative charging. Of course, charging in international criminal courts is almost always incomplete. It is usually impossible to present evidence about every single criminal act that the prosecution believes an accused could be convicted of. Rather, the goal is to charge the accused with a representative selection of crimes that accurately conveys the scope of the accused’s criminality. 319See supra note 202. See also ICTY Rules, supra note 148, at R. 73bis(D) (suggesting that charges are “reasonably representative” if they take into account “all the relevant circumstances” including the classification and nature of the charges, the location of the alleged crimes, the scale of the crimes, and the victims of the crimes); Daryl A. Mundis, Book Review, 102 Am. J. Int’l L. 691, 693 (2008) (arguing that prosecutors are “duty bound to draft indictments that reflect the scale and scope of [the accused’s] criminality”). My goal as a prosecutor at the ECCC was to craft an indictment that was representative of the geographic scope of the crimes, the types of crimes that were committed, the temporal scope of the crimes, and the types of victims. This form of charging, called representative charging, is dramatically different from symbolic charging where only one or two of the most prominent or easily proved episodes are chosen.

Symbolic charging might make prosecutions cheaper and faster by dramatically reducing the number of counts and crime sites included in the indictment, which would presumably reduce trial complexity. But it is likely that even these streamlined cases would be very complex compared to average domestic prosecutions. Moreover, symbolic charging might cause more problems than it solves. Courts that have tried this approach, for example in the trials of Saddam Hussein and Thomas Lubanga, have been criticized for failing to accurately capture the accused’s criminality. 320See Robert Cryer, Prosecuting the Leaders: Promises, Politics and Practicalities, 1 Göttingen J Int’l L. 45, 72–74 (2009); Suzan M. Pritchett, Entrenched Hegemony, Efficient Procedure or Selective Justice?: An Inquiry into Charges for Gender-Based Violence at the International Criminal Court, 17 Transnat’l L. & Contemp. Probs. 265, 288 (2008); Mundis, supra note 319, at 693 (“Can anyone today claim that the trial of Saddam Hussein before the Iraqi Special Tribunal and his execution solely for the crimes in Dujail held him accountable for the totality of his criminal record?”). At least in the Lubanga case, it did not seem to shorten the trial either. 321See supra note 158 (noting that the Lubanga case took 204 trial days to hear). Moreover, my own research indicates that establishing the historical record should be one of the principal goals of international trials. 322See Ford, supra note 2, at Section IV. This would be much harder to achieve with symbolic charging.

A fourth alternative might be to charge only a very limited number of the most senior leaders. This is the strategy taken by the SCSL, ECCC, and ICC. It is a cheaper approach because it reduces the number of trials that take place, but there is no reason to believe that it decreases the length or cost of individual trials; it may even reduce overall efficiency. 323It seems likely that there are “economies of scale” at international criminal tribunals and that tribunals that try more cases will be more efficient than tribunals that try fewer cases. This may be one reason the SCSL’s efficiency was lower than the ICTY’s efficiency. Moreover, a focus on the most senior leaders may cause problems. There is no obvious correlation between seniority and number of counts the indictees were charged with or the length of the sentence they received. 324See Ford, supra note 71, at 73–74, 97. Indeed, the most serious and shocking crimes that came before the ICTY were committed largely by mid-level accused. 325See id. at 97. If, as the Preamble to the Rome Statute says, our goal is to prosecute the most serious crimes, then we have to be able to prosecute more than just one or two of the most senior leaders. 326See supra note 315.

Finally, there is no guarantee that a court that tries many fewer cases would be more effective than the ICTY (as opposed to more efficient). 327See Shany, supra note 26, at 230, 237 (contrasting effectiveness, the ability of an organization to accomplish its purposes, with efficiency, the relationship between an organization’s inputs and outputs). The question of how many trials are necessary for a court to achieve its purposes is beyond the scope of this Article, although it is certainly an important question. To answer it, however, you would need data on the effect of individual trials on the goals of international criminal justice. At the moment there is no way to measure how individual trials contribute to a court’s purposes. As a result, it is impossible to know the optimal number of trials necessary to maximize the court’s contributions to its purposes. A small number of trials could be better than a large number of trials, but it could be worse. The optimal number might vary depending on a host of factors specific to the particular situation. We simply do not know.

In short, the length, complexity, and cost of international criminal trials may well be largely unavoidable. Indeed, a number of judges who have worked at the ICTY eventually came to this conclusion. 328See Kwon, supra note 1, at 362 (concluding that the complexity of cases is driven by factual and legal complexity, as well as due process concerns); Wald, supra note 16, at 468 (concluding that international trials are fundamentally different from most trials in the United States and that they will inevitably be more complex). As the comparison of the complexity and efficiency of international and domestic trials shows, the cost and length of the ICTY’s trials are a function of the extraordinary seriousness and complexity of the crimes that the court has been tasked with adjudicating, rather than any inherent inefficiency arising from the international nature of the court.

While it may seem that this Article’s focus on the ICTY makes it largely retrospective in nature, adopting a quantitative approach to complexity and efficiency has prospective value as well. The ICTY will be shutting down soon, but these issues will not go away. Indeed, they have already become hotly contested questions at the ICC, where negotiations over the court’s budget have been marked by disputes about the court’s efficiency. 329See supra note 18 and accompanying text. Although it is too soon to draw any conclusions about the ICC’s efficiency using the methodology in this Article because the court has only completed one trial, the ICC’s efficiency can eventually be measured quantitatively. This would have substantial value as it would provide members of the Assembly of States Parties with objective information that they could use in deciding how much funding the ICC needs to carry out its mandate. This would be a decided improvement over the current debates about the ICC’s efficiency, which seem driven more by the desires of certain members to minimize their contributions than by an objective analysis of the court’s efficiency.

Appendix: Complexity of Completed ICTY Trials

Case Name

Case Number

Complexity Score

Furundžija

IT-95-17/1

0.07

Aleksovski

IT-95-14/1

0.32

Vasiljević

IT-98-32

0.34

Halilović

IT-01-48

0.35

Kunarac et al.

IT-96-23 & IT-96-23/1

0.35

Limaj et al.

IT-03-66

0.37

Strugar

IT-01-42

0.42

Krnojelac

IT-97-25

0.46

Lukić & Lukić

IT-98-32/1

0.51

Tadić

IT-94-1

0.62

Haradinaj et al.

IT-04-84

0.62

Martić

IT-95-11

0.62

Delić

IT-04-83

0.63

Mucić et al.

IT-96-21

0.69

Krstić

IT-98-33

0.69

Kvočka et al.

IT-98-30/1

0.73

Stakić

IT-97-24

0.75

Boškoski & Tarčulovski

IT-04-82

0.76

Blagojević & Jokić

IT-02-60

0.77

Mrkšić et al.

IT-95-13/1

0.77

Milosević, Dragomir

IT-98-29/1

0.77

Šešelj

IT-03-67

0.78

Kupreškić et al.

IT-95-16

0.81

Naletilić & Martinović

IT-98-34

0.82

Orić

IT-03-68

0.85

Simić et al.

IT-95-9

1.00

Đorđević

IT-05-87/1

1.11

Galić

IT-98-29

1.15

Perišić

IT-04-81

1.15

Blaškić

IT-95-14

1.17

Stanišić & Simatović

IT-03-69

1.28

Hadžihasanović & Kubura

IT-01-47

1.32

Tolimir

IT-05-88/2

1.47

Krajišnik

IT-00-39

1.51

Gotovina & Markač

IT-06-90

1.60

Brđanin

IT-99-36

1.63

Kordić & Čerkez

IT-95-14/2

1.72

Šainović et al.

IT-05-87

1.79

Stanišić & Župljanin

IT-08-91

1.84

Popović et al.

IT-05-88

2.46

Prlić et al.

IT-04-74

2.63

Footnotes

Assistant Professor of Law at the John Marshall Law School in Chicago, Illinois. This Article was greatly improved by the contributions of those who commented on earlier versions, including Professors Jenia Iontcheva Turner, Daryl Lim, Morse Tan, Anne Holthoefer, Lesley Wexler, Robert Knowles, Jonathan Hafetz, and David Wippman. Thanks are also due to those who provided feedback at the various conferences where this was presented. Fang Han, Gregory Cunningham, and Iveliz Orellano provided research assistance.

1Shahram Dana, Turning Point for International Justice?, in XI Annotated Leading Cases of International Criminal Tribunals 962, 972 (Andre Klip & Goran Sluiter eds., 2007) (“The primary function of the international criminal tribunal is to determine the criminal responsibility and punishment of those individuals found guilty of the crimes under its jurisdiction.”); Adrian Fulford, The Reflections of a Trial Judge, 22 Criminal L.F. 215, 216 (2011) (“We are first, foremost and last a criminal court: our core business is to process criminal trials. All the rest, and I hasten to add some of the rest is very important indeed (such as our deterrent potential, reparations to victims and outreach), is secondary to the Court’s obligation to investigate, arrest and try alleged criminals.”); O-Gon Kwon, The Challenge of An International Criminal Trial as Seen from the Bench, 5 Journal of International Criminal Justice 360, 373 (2007) (“The task of determining guilt or innocence must take precedence over other, not strictly judicial, considerations”); U.N. Secretary-General, 20, U.N. Doc A/C.5/52/4 (Oct. 21, 1997) (“[Chambers] performs the fundamental core activity of the Tribunal, that is, the trial and determination of guilt or innocence of persons responsible for serious violations of international humanitarian law within the territory of the former Yugoslavia”).

2See S.C. Res. 808, U.N. Doc. S/RES/808 (Feb. 22, 1993). The ICTY was given the additional goals of putting an end to the commission of crimes in the former Yugoslavia and restoring peace in the Balkans. Id. See also Stuart Ford, A Social Psychology Model of the Perceived Legitimacy of International Criminal Courts: Implications for the Success of Transitional Justice Mechanisms, 45 Vand. J. Transnat’l L. 405, 472 (2012) (noting that the goals of international criminal courts include establishing the historical record and fostering post-conflict reconciliation).

3See Stuart Ford, How Leadership in International Criminal Law Is Shifting from the United States to Europe and Asia: An Analysis of Spending on and Contributions to International Criminal Courts, 55 St. Louis U. L.J. 953, 971 (2011) (estimating the cost as $2.3 billion). The estimate of $2.3 billion was based on the assumption that the ICTY would shut down in 2014. Id. Now that it is clear the ICTY will continue to operate into at least 2017, see infra note 4, it will certainly cost more than that. My current estimate is that it will ultimately cost $2.7 billion. See infra note 178.

4The ICTY was established in 1993. See S.C. Res. 808, U.N. Doc. S/RES/808/ (Feb. 22, 1993). According to its most recent report on its Completion Strategy, the ICTY will not finish its last trials and appeals until early 2017. See President of the International Criminal Tribunal for the Former Yugoslavia, Assessment and report of Judge Theodor Meron, President of the International Criminal Tribunal for the Former Yugoslavia, provided to the Security Council pursuant to paragraph 6 of Security Council resolution 1534 (2004) and covering the period from 24 May to 18 November 2013, transmitted by letter to President of the Security Council, U.N. Doc. S/2013/678, at 39 (Nov. 18, 2013) (by Judge Theodor Meron).

5See Alex Whiting, In International Criminal Prosecutions, Justice Delayed Can Be Justice Delivered, 50 Harv. Int’l L.J. 323, 324 & n.1 (2009).

6This data comes from a Westlaw search of the Law Reviews & Journals database using the search “ICTY /25 slow.” The search was conducted on June 24, 2014. The vast majority of the results are articles that either accuse the ICTY of being slow or note that the ICTY has been accused of being slow. Running the same search in the News database found 80 responsive newspaper articles.

7This data comes from a Westlaw search of the Law Reviews & Journals database using the search “ICTY /25 (expensive costly).” The search was conducted on June 24, 2014. The vast majority of the results are articles that either accuse the ICTY of being expensive or note that it has been accused of being expensive. Running the same search in the News database found 17 responsive newspaper articles.

8This data comes from a Westlaw search of the Law Reviews & Journals database using the search “ICTY /25 inefficient.” The search was conducted on June 24, 2014. The vast majority of the results are articles that either accuse the ICTY of being inefficient or note that it has been accused of being inefficient. Running the same search in the News database found 10 responsive newspaper articles.

9Whiting, supra note 5, at 323–24.

10Etelle R. Higonnet, Restructuring Hybrid Courts: Local Empowerment and National Criminal Justice Reform, 23 Ariz. J. Int’l & Comp. L. 347, 347 (2006).

11Ralph Zacklin, The Failings of Ad Hoc International Tribunals, 2 J. Int’l Crim. Just. 541, 543 (2004).

12Id. at 545.

13See Steven D. Roper & Lilian A. Barria, Gatekeeping Versus Allocating in Foreign Assistance: Donor Motivations and Contributions to War Crimes Tribunals, 51 J. Conflict Resol. 285, 287 (2007); Higonnet, supra note 10, at 434; Elizabeth Nielsen, Hybrid International Criminal Tribunals: Political Interference and Judicial Independence, 15 UCLA J. Int’l L. & Foreign Aff. 289, 295 (2010); Jonathan O’Donohue, Financing the International Criminal Court, 13 Int’l Crim. L. Rev. 269, 273–74 (2013); Nancy Kaymar Stafford, A Model War Crimes Court: Sierra Leone, 10 ILSA J. Int’l & Comp. L. 117, 139 (2003).

14See, e.g., David Wippman, The Cost of International Justice, 100 Am. J. Int’l L. 861, 861–62 (2006).

15See id. at 873 (“In general, however, complex cases take longer to try and generate higher overall costs.”).

16See id. (“By contrast, almost all ICTY trials, because of the nature of the prosecution undertaken, entail lengthy and expensive court proceedings and pretrial investigations.”); Patricia M. Wald, ICTY Judicial Proceedings—An Appraisal from Within, 2 J. Int’l Crim. Just. 466, 468 (2004).

17See, e.g., Gillian Higgins, The Impact of the Size, Scope and Scale of the Milošević Trial and the Development of Rule 73bis before the ICTY, 7 Nw. J. Int’l Hum. Rts. 239, 246 (2009) (“The procedural and substantive demands of the [Milošević] trial upon all parties were excessive.”).

18See generally Stuart Ford, How Much Money Does the ICC Need in The law and Practice of the International Criminal Court: A Critical Account of Challenges and Achievement (Carsten Stahn ed., forthcoming 2014). See also Jonathan O’Donohue, Financing the International Criminal Court, 13 Int’l Crim. L. Rev. 269, 279–81 (2013).

19See supra text accompanying notes 5–13.

20See infra notes 159–60.

21See infra Conclusion.

22See, e.g., Michael P. Scharf, The Politics Behind U.S. Opposition to the International Criminal Court, 6 Brown J. World Aff. 97, 98 (1999) (describing opposition to the creation of the International Criminal Court in the United States based on a fear that it would investigate or prosecute American citizens).

23See Mirjan Damaska, What is the Point of International Criminal Justice?, 83 Chi.-Kent L. Rev. 329, 331–40 (2008) (questioning whether international courts can achieve their goals); David Wippman, Atrocities, Deterrence and the Limits of International Justice, 23 Fordham Int’l L.J. 473, 475 (1999) (questioning whether international courts can achieve deterrence). But see Payam Akhavan, Beyond Impunity: Can International Criminal Justice Prevent Future Atrocities?, 95 Am. J. Int’l L. 7 (2001) (arguing that prosecution of serious violations of international criminal law can have positive effects).

24See supra note 18.

25See Ford, supra note 3, at 954–55.

26See Yuval Shany, Assessing the Effectiveness of International Courts: A Goal-Based Approach, 106 Am. J. Int’l L. 225, 230 (2012).

27Id. at 230–31.

28Id. at 237–38.

29See Damaska, supra note 23, at 331; Minna Schrag, Lessons Learned from ICTY Experience, 2 J. Int’l Crim. Just. 427, 428 (2004); Jenia Iontcheva Turner, Defense Perspectives on Law and Politics in International Criminal Trials, 48 Va. J. Int’l L. 529, 537–43 (2008).

30Shany, supra note 26, at 239 (“The goals of public organizations, such as courts, tend to be ambiguous, and the public goods they generate, such as justice, peace, and legal certainty, are hard to quantify”). See also id. at 248–49 (“The key to assessing the effectiveness of international courts according to the rational-system or goal-based approach involves evaluation of judicial outcomes”).

31Jay Tidmarsh, Unattainable Justice: The Form of Complex Litigation and the Limits of Judicial Power, 60 Geo. Wash. L. Rev. 1684 (1992).

32Jeffrey W. Stempel, A More Complete Look at Complexity, 40 Ariz. L. Rev. 781 (1998).

33There are numerous scholarly articles that deal in some manner with the concept of complexity in the law. Some examples include Stephen B. Burbank, The Costs of Complexity, 85 Mich. L. Rev. 1463 (1987); Michael Heise, Criminal Case Complexity: An Empirical Perspective, 1. J. Empirical Legal Stud. 331 (2004); Ronit Dinovitzer & Jeffrey S. Leon, When Long Becomes too Long: Legal Culture and Litigators’ Views on Long Trials, 19 Windsor Y.B. Access Just. 106 (2001); Richard O. Lempert, Civil Juries and Complex Cases: Let’s Not Rush to Judgment, 80 Mich L. Rev. 68 (1982); Larry Heuer & Steven Penrod, Trial Complexity: A Field Investigation of Its Meaning and Effects, 18 Law & Hum. Behav. 29 (1994); Peter H. Schuck, Legal Complexity: Some Causes, Consequences and Cures, 42 Duke L.J. 1 (1992); Theodore Eisenberg, Paula L. Hannaford-Agor, Valerie P. Hans, & Nicole L. Waters, Judge-Jury Agreement in Criminal Cases: A Partial Replication of Kalven and Zeisel’s The American Jury, 2 J. Empirical Legal Stud. 171 (2005); Matthew A Reiber & Jill D Weinberg, The Complexity of Complexity: An Empirical Study of Juror Competence in Civil Cases, 78 U. Cin. L. Rev. 929 (2010).

34Some studies have covered both civil and criminal trials. See Heuer & Penrod, supra note 33. Very few have exclusively studied exclusively criminal trials. The exception is Heise, supra note 33. Most of the literature cited below relates largely or exclusively to civil trials.

35One of the exceptions to the focus on the United States is the study by Dinovitzer and Leon, which examined civil trials in Canada. See Dinovitzer & Leon, supra note 33.

36Burbank, supra note 33, at 1463 (“Not even those charged with responsibility to devise procedures for complex cases in the federal courts have essayed a definition worth of the name.”); Heise, supra note 33, at 366–67 (noting that there is no accepted definition of complexity and that different groups may define it differently); Eisenberg et al., supra note 33, at 190 (noting that different people probably view complexity differently); Dinovitzer & Leon, supra note 33, at 117 (noting that courts have difficulty determining when to categorize cases as complex); Heuer & Penrod, supra note 33, at 30–31 (noting that there have been many proposed definitions of complexity but little agreement); Schuck, supra note 33, at 2–3 (noting that “legal complexity is hard to define” and that different people might conceive of it and define it in different ways); Stempel, supra note 32, at 785–86; Reiber & Weinberg, supra note 33, at 945 n.41 (noting that there have been attempts to create an all-encompassing definition of complexity, but that such attempts have not generally been successful). Professor Tidmarsh’s article highlights the difficulty of trying to reach an agreement on a definition of complexity through its extensive catalog of definitions that have been previously suggested, which covers more than 40 pages. See Tidmarsh, supra note 31, at 1692–1734.

37See Stempel, supra note 32, at 785–86 (noting that “a single definition of complexity is both unnecessary and perhaps misleading”).

38See Tidmarsh, supra note 31, at 1690 (“a correct understanding of complex litigation might prevent well-intentioned reform from paving the path to a procedural hell”).

39Id. at 1694–98.

40Id. at 1743–50 (basing his definition of complexity on seven assumptions about the nature of civil procedure in the United States).

41Id. at 1689–90.

42See infra notes 104–108.

43Heise, supra note 33, at 346.

44Heuer & Penrod, supra note 33, at 35.

45Heise, supra note 33, at 344–45, 350; Eisenberg et al., supra note 33, at 190.

46The Oxford English Dictionary 613 (J.A. Simpson & E.S.C. Weiner eds., Oxford University Press 2d ed. 1989). Other dictionaries define it similarly. See New Oxford American Dictionary (Angus Stevenson & Christine A. Lindberg eds., Oxford University Press 3d ed. 2010) (“the state or quality of being intricate or complicated”); The American Heritage Dictionary of the English Language 377–78 (Houghton Mifflin Co. 4th ed. 2000) (“the quality or condition of being complex”); Random House Webster’s College Dictionary 271 (2d ed. 1999) (“the state or quality of being complex”).

47The Oxford English Dictionary, supra note 46, at 613.

48See New Oxford American Dictionary, supra note 46 (“consisting of many different and connected parts”); The American Heritage Dictionary of the English Language, supra note 46 (“consisting of interconnected or interwoven parts”); Random House Webster’s College Dictionary, supra note 46 (“composed of many interconnected parts; compound; composite”).

49The Oxford English Dictionary, supra note 46, at 613.

50See New Oxford American Dictionary, supra note 46, at 271 (“not easy to analyze or understand; complicated or intricate”); Random House Webster’s College Dictionary, supra note 46 (“so complicated or intricate as to be hard to understand or deal with”).

51See Heise, supra note 33, at 358 (noting that judges in the study were asked to rate evidentiary and legal complexity separately and that the separate scores were then combined into an overall measure of trial complexity); Heuer & Penrod, supra note 33, at 35 (noting that judges in the study were asked to rate both evidentiary complexity and legal complexity); Tidmarsh, supra note 31, at 1702–12 (describing various types of complexity, including legal complexity, factual complexity, remedy complexity, and party complexity); Stempel, supra note 32, at 789–96 (discussing legal and factual complexity).

52For example, Reiber and Weinberg treat “procedural complexity” as a separate form of complexity, although it looks most like what this Article calls legal complexity. Reiber and Weinberg, supra note 33, at 947. Tidmarsh identifies “remedy complexity” and “party complexity” as additional forms of complexity in civil trials. Tidmarsh, supra note 31, at 1707–12. These forms of complexity have less relevance in the context of criminal proceedings.

53Cf. id. at 1736 (arguing that legal adjudication requires “reasoned judgment, which is based on the consideration of the disputants’ circumstances, the relevant evidence, and the nature of the claimed obligation”).

54Schuck, supra note 33, at 3–4; Stempel, supra note 32, at 789–92 (noting that legal complexity may stem from the technical nature of the law, the difficulty in ascertaining the law, or the indeterminacy of the law).

55Heuer & Penrod, supra note 33, at 35–36 (noting that both the quantity of the evidence and the difficulty of the evidence caused perceived trial complexity to increase); Heise, supra note 33, at 352–54 (noting that the difficulty of the evidence increased perceived complexity of the trial).

56See Reiber & Weinberg, supra note 33, at 950–51 (testing a model in which factual complexity was represented by the need to reconcile the evidence of multiple witnesses who gave conflicting testimony and draw inferences from circumstantial evidence); Stempel, supra note 32, at 794–95; Tidmarsh, supra note 31, at 1766.

57See Heuer & Penrod, supra note 33, at 36 tbl.2 (reporting that the quantity of evidence and the complexity of the evidence had more impact on the reported complexity of the cases than the complexity of the law). But see Reiber & Weinberg, supra note 33, at 952–54 (suggesting that legal complexity makes juror decision-making more difficult than factual complexity). Of course, the ICTY does not use juries, so the Reiber and Weinberg finding may not be applicable at the ICTY.

58See Tidmarsh, supra note 31, at 1755–80; Dinovitzer & Leon, supra note 33, at 112–14 (noting studies that have indicated that the local legal culture of judges and lawyers affects the length of time it takes to decide a case); id. at 138 (noting that the vast majority of lawyers in the study believed that lawyer skill, knowledge and preparation would affect trial length).

59See Heise, supra note 33, at 355–56; cf. Heuer & Penrod, supra note 33, at 41–42.

60See Stempel, supra note 32, at 796; Tidmarsh, supra note 31, at 1757–59.

61See Higgins, supra note 17, at 245, 248, 250; U.N. Secretary-General, Budget for the International Tribunal for the Prosecution of Persons Responsible for Serious Violations of International Humanitarian Law Committed in the Territory of the Former Yugoslavia since 1991, for the biennium 2012–2013, 55, U.N. Doc. A/66/386 (Sept. 29, 2011).

62Despite our beliefs otherwise, we are not fundamentally rational decision makers. Our decision-making is affected by both motivated reasoning and cognitive biases. See Ford, supra note 2, at 419–39.

63Schuck, supra note 33, at 25 (noting research in social psychology that shows that people have difficulty with truly complex decisions and tend to radically simplify their decision-making through the use of heuristics when confronted with complexity); Stempel, supra note 32, at 795–96.

64Cf. Tidmarsh, supra note 31, at 1703 (noting that cases with relatively simple facts can generate “wonderfully difficult legal issues”).

65See supra notes 59–63.

66Heise, supra note 33, at 359 tbl.8 (noting that evidentiary and legal complexity were often related and cannot be assumed to be independent); Reiber & Weinberg, supra note 33, at 947 n.48 (noting that procedural and factual complexity may overlap with one another).

67Tidmarsh, supra note 31 at 1702–03 (suggesting that cases are not truly complex unless they involve both factual and legal complexity).

68Cf. id. at 1702–03 (suggesting that the most complex cases arise when there is a combination of factual and legal complexity).

69Stempel, supra note 32, at 796; Tidmarsh, supra note 31, at 1757–59.

70See supra text accompanying notes 46–50.

71See Stuart Ford, Fairness and Politics at the ICTY: Evidence from the Indictments, 39 N.C. J. Int’l L. & Com. Reg. 45, 75–77, 84–85, 87 (2013).

72See infra Table 3.

73See Rep. of the ICTY, 57th Sess., Sept. 4, 2002, 6, U.N. Doc. A/57/379-S/2002/985 (Aug. 14, 2002).

74See President of the International Tribunal for the Former Yugoslavia, Assessments and report of Judge Theodor Meron, President of the International Criminal Tribunal for the Former Yugoslavia, provided to the Security Council pursuant to paragraph 6 of Security Council resolution 1534 (2004), transmitted by letter to President of the Security Council, 3, 47–52, U.N. Doc. S/2004/420 (May 21, 2004) (by Judge Theodor Meron).

75See President of the International Tribunal for the Former Yugoslavia, Assessment and report of Judge Fausto Pocar, President of the International Tribunal for the Former Yugoslavia, provided to the Security Council pursuant to paragraph 6 of Council resolution 1534 (2004), 30, U.N. Doc. S/2006/898 (Nov. 16, 2006) (by Judge Fausto Pocar).

76See President of the International Tribunal for the Former Yugoslavia, Assessment and report of Judge Patrick Robinson, President of the International Tribunal for the Former Yugoslavia, provided to the Security Council pursuant to paragraph 6 of Council resolution 1534 (2004), 42, U.N. Doc. S/2008/729 (Nov. 24, 2008) (by Judge Patrick Robinson).

77See President of the International Tribunal for the Former Yugoslavia, Assessment and report of Judge Patrick Robinson, President of the International Tribunal for the Former Yugoslavia, provided to the Security Council pursuant to paragraph 6 of Security Council resolution 1534 (2004), covering the period from 15 May 2010 to 15 November 2010, 4, U.N. Doc. S/2010/588 (Nov. 19, 2010) (by Judge Patrick Robinson).

78See President of the International Tribunal for the Former Yugoslavia, Assessment and report of Judge Theodor Meron, President of the International Tribunal for the Former Yugoslavia, provided to the Security Council pursuant to paragraph 6 of Security Council resolution 1534 (2004), and covering the period from 16 November 2012 to 23 May 2013, 3, U.N. Doc. S/2013/308 (May 23, 2013).

79See S.C. Res. 808, supra note 4; Letter from the President of the ICTY, supra note 4.

80See President of the International Tribunal for the Former Yugoslavia, Assessment and report of Judge Patrick Robinson, President of the International Tribunal for the Former Yugoslavia, provided to the Security Council pursuant to paragraph 6 of Security Council resolution 1534 (2004), covering the period from 15 May 2010 to 15 November 2010, 6, U.N. Doc. S/2010/588 (Nov. 15, 2010) (by Judge Patrick Robinson).

81Id. (“This is the nature of trials, particularly trials of the complexity that are heard at the Tribunal.”).

82See Kwon, supra note 1, at 364 (concluding that the elements of the crimes and modes of liability are “far more complex” in international criminal law than in domestic criminal law).

83Robert Cryer et al., An Introduction to International Criminal Law and Procedure 9–12 (2d ed. 2010) (discussing the sources of international criminal law).

84Crimes against humanity are largely rooted in customary international law. See Stuart Ford, Crimes Against Humanity at the Extraordinary Chambers in the Courts of Cambodia: Is a Connection with Armed Conflict Required?, 24 UCLA Pac. Basin L.J. 125, 129–80 (2007) (chronicling the development of crimes against humanity over time).

85Cryer et al., supra note 83, at 109–80 (describing the various international courts).

86Id. at 246.

87See Stuart Ford, Is the Failure to Respond Appropriately to a Natural Disaster a Crime Against Humanity? The Responsibility to Protect and Individual Responsibility in the Aftermath of Cyclone Nargis, 38 Denver J. Int’l Pol’y 227, 242, 245–46 (2010) (discussing the intent requirement in international criminal law).

88Robert Petit et al., Exploring Critical Issues in Religious Genocide: Case Studies of Violence in Tibet, Iraq and Gujarat, 40 Case W. Res. J. Int’l L. 163, 203–09 (2008) (discussing the use of circumstantial evidence to prove genocide).

89Id. at 190–94 (discussing the various inconsistent and contradictory sources relating to the meaning of the word “destroy” in the crime of genocide).

90See Wippman, supra note 14, at 875 (“Relative to most national criminal court trials, ICTY trials are inherently complex.”); Kwon, supra note 1, at 364.

91See Ford, supra note 71, at 67–68.

92Id.

93See id. at 85–87. For example, the trial of Slobodan Milošević covered acts that took place in three different countries at more than 400 crime sites. Id. at 87. See also Damaska, supra note 23, at 340–41 (describing the complexity of the Milošević case further).

94See generally Nancy A. Combs, Fact-Finding Without Facts: The Uncertain Evidentiary Foundations of International Criminal Convictions (2010) (documenting difficulties in fact-finding in international trials).

95 See infra Figure 1.

96The additional participants include not just the extra accused, but the additional lawyers that represent those accused.

97See infra note 133.

98See supra note 61.

99The most famous of these cases involved Slobodan Milošević and Vojislav Šešelj. Milošević and Šešelj both used their trials as a platform to attack the court, obstruct the proceedings, and try to influence public opinion in Serbia. See Turner, supra note 29, at 573–74.

100See supra notes 62–63.

101See Eisenberg et al., supra note 33, at 175; Heise, supra note 33, at 343; Heuer & Penrod, supra note 34, at 34.

102See supra notes 43–45.

103A proxy variable is “[a] measurable variable that is used in place of a variable that cannot be measured.” Graham Upton & Ian Cook, A Dictionary of Statistics 315 (2d ed. 2008).

104A dichotomous variable is a categorical variable with only two categories. Id. at 114. A continuous variable is one whose set of possible values is a continuous interval of real numbers. Id. at 88.

105See Fed. R. Civ. P. 16(c)(2)(L) (noting that cases that are identified as “complex” may be subject to “special procedures”). See also Manual for Complex Litigation (4th ed. 2004) (providing guidance to federal judges facing “complex” cases).

106See generally Stempel, supra note 32 (assessing criteria used to measure complexity); Tidmarsh, supra note 31 (developing a more formal and inclusive definition of complexity).

107See Heuer & Penrod, supra note 33, at 33; Schuck, supra note 33, at 5.

108See infra Part II.C.

109The author spent three years working in the Office of the Co-Prosecutors at the ECCC as an Assistant Prosecutor. He assisted in investigating and drafting the indictments and served on the trial team in the case of Kaing Keuk Eav alias Duch. Prior to working at the ECCC, he spent more than five years litigating complex commercial matters before federal district courts in the United States.

110See Damaska, supra note 23, at 341; Higgins, supra note 17, at 246 (describing the complexity of the Milošević trial in terms of hearing days, witnesses, and trial exhibits).

111See Defence Counsel–Pre-Trial Legal Aid Policy, U.N. Int’l Crim. Trib. For the Former Yugoslavia, 22 (May 1, 2006), available at http://www.icty.org/x/file/About/Defence/pretrial_payment_2006_en.doc (last visited Oct. 9, 2012) (using the number and type of witnesses and documents that are expected to be used at trial to determine the expected complexity of the case).

112See e.g., Prosecutor v. Krstic, Case No. IT-98-33-T, Judgment, 4 (Aug. 2, 2001) (“In all, the Trial Chamber heard more than 110 witnesses over 98 days of trial and viewed in the vicinity of 1,000 exhibits.”). See also Prosecutor v. Popović et al., Case No. IT-05-88-T, Judgment, Annex II, 1, 15–16, 19–26 (June 10, 2010); Prosecutor v. Thomas Lubanga Dyilo, Case No. ICC-01/04-01/06, Judgment pursuant to Article 74 of the Statute, 11 (Mar. 14, 2012); Prosecutor v. Charles Ghankay Taylor, Case No. SCSL-03-1-T, Judgment Summary, 8, (April 26, 2012); Judgment of the International Military Tribunal for the Trial of the German Major War Criminals, Introduction, (Sept. 30, 1946); Prosecutor v. Bagosora et al., Case No. ICTR-98-41-T, Judgment and Sentence 1 n.1 (Dec. 18, 2008).

113See, e.g., Prosecutor v. Jadranko Prlić, Bruno Stojić, Slobodan Praljak, Milivoj Petković, Valentin Čorić and Berislav Pušić, Judgment Summary, May 29, 2013, available at http://icty.org/x/cases/prlic/tjug/en/130529_summary_en.pdf (“During the trial, which closed on 2 March 2011, the Chamber admitted close to 10,000 exhibits into evidence and heard 145 Prosecution witnesses, six of whom were expert witnesses, as well as 61 Defence witnesses, seven of whom were expert witnesses. The Judgment is a 2,629-page-long document divided into six volumes, including four annexes and partially dissenting or separate opinions by Judge Trechsel and myself, all of which illustrates the complexity of this case.”).

114See Heuer & Penrod, supra note 33, at 36 tbl.2.

115See Heise, supra note 33, at 352–53, 355–56, 359, 361–62.

116Id. at tbl.7 & tbl.8 (finding that the number of expert witnesses affected attorney and judicial perceptions of trial complexity).

117Cf. id. at 362 (noting that “[a]lthough it is possible to conclude that trial length itself may make a trial more complicated, a more likely explanation is that complicated criminal trials take longer to litigate”); Heuer & Penrod, supra note 33, at 36 tbl.2 (finding that the quantity of information presented at trial was the best predictor of subjective measures of trial complexity).

118See supra notes 109–112.

119See Jan Lorenz et al., How Social Influence Can Undermine the Wisdom of the Crowd Effect, 108 Proc. Nat’l Acad. Sci. 9020, 9020 (2011) (describing the “wisdom of the crowd” effect, where the “aggregate of many people’s estimates tends to be closer to the true value than all of the separate individual or even expert guesses.”).

120See Heuer & Penrod, supra note 33 (noting studies that show factual complexity is the largest driver of subjective evaluations of overall complexity).

121See Heise, supra note 33.

122See Commc’n Serv. of the Int’l Criminal Tribunal for the former Yugoslavia, Case Information Sheet for Prlić et al., ICTY, available at http://icty.org/x/cases/prlic/cis/en/cis_prlic_al_en.pdf (last visited Sept. 27, 2013), at 4 (noting the number of witnesses called and exhibits offered by each individual defendant); Case Information Sheet for Popović et al., ICTY, available at http://icty.org/x/cases/popovic/cis/en/cis_popovic_al_en.pdf (last visited Sept. 27, 2013), at 5 (same).

123See Turner, supra note 29, at 530.

124The Judgment Summary in Prlić explicitly claims that the length of the judgment is an indicator of the complexity of the case. See Prosecutor v. Jadranko Prlić, supra note 113.

125The correlation coefficient for the Complexity Scores and trial judgment length (measured in pages) was r = .72. This result was highly significant, with p ≤ .001.

126See Heise, supra note 33.

127See infra Table 3.

128See Ford, supra note 71, Section VII (describing the creation of the database). A copy of the data is available upon request from the author.

129Three cases were withdrawn by the prosecutor because the accused were not of sufficient importance, seven cases were transferred to domestic courts, and seven cases ended with the death of the accused prior to entry of a judgment.

130Its closest competitor, the International Criminal Tribunal for Rwanda (“ICTR”), has tried only 74 accused. See Status of Cases, Int’l Crim. Trib. For Rwanda, http://www.unictr.org/Cases/tabid/204/Default.aspx (last visited June 17, 2013).

131Given that the averages for most of these figures are skewed by a few outliers, the median values are more representative of the cases that came before the ICTY.

132See generally Ford, supra note 71 (finding that charging and sentencing practices in the ICTY are generally comparable to the range of practices considered fair in domestic criminal justice systems).

133For example, in federal district courts in the United States, the average number of accused per criminal case is only 1.3. Thomas F. Hogan, Admin. Office of the United States Cts., Annual Report of the Director: Judicial Business of the United States Courts, at tbl.D (2012). If one compares the number of criminal cases opened in 2011 (78,440) with the number of new criminal defendants in that same period (102,931), one can surmise that there were approximately 1.3 defendants per case. Id. at 192, 195.

134See Thomas H. Cohen & Tracey Kyckelhahn, Bureau of Justice Statistics, U.S. Dep’t of Justice, Felony Defendants in Large Urban Counties, 2006, at App. tbl.1 (2010) (noting that 48% of felony defendants had no other charges pending against them).

135See Ford, supra note 71, at 104.

136See Cohen & Kyckelhahn supra note 134, at tbl.1.

137See Ford, supra note 71, at 104.

138See supra note 129.

139In cases where the accused pled guilty, either no evidence was presented or only limited evidence was presented.

140See infra Appendix.

141This information was obtained by searching in the ICTY Court Records database located at www.ICR.ICTY.org. Under “Select Name of Accused,” the option for “All Accused” was chosen. The results were then narrowed down to only those in English to avoid picking up duplicates of the same document in multiple languages. Judicial decisions are those documents that appear under the category of “Decisions and Orders.” Filings by the parties are a combination of the results that appear under the categories of “Briefs,” “Responses,” and “Motions.” The searches were conducted on July 7, 2013. The searches were then repeated in French and Bosnian/Serbian/Croatian. The results were substantially lower for both other languages, suggesting that significant numbers of documents are available only in English.

142This figure combines the English language documents that were found under the categories of “Notices,” “Transcripts,” “Witness-Related Materials,” “Correspondence,” “Warrants and Subpoenas,” and “Other” on the ICTY Court Records Database. More than 10,000 of these documents are transcripts of court proceedings. The search was conducted on July 7, 2013.

143See Higgins, supra note 17, at 246 (noting that the prosecution disclosed more than 1.2 million pages of documents to the defense just during the trial of Slobodan Milošević).

144See generally Catherine Cisse, The International Tribunals for the Former Yugoslavia and Rwanda: Some Elements of Comparison, 7 Transnat’l L. & Contemp. Probs. 103 (1997) (comparing the ICTY and ICTR); Gideon Boas, Comparing the ICTY and the ICC: Some Procedural and Substantive Issues, 47 Netherlands Int’l L. Rev. 267 (2000) (comparing the ICTY and ICC); Donna E. Arzt, Views on the Ground: The Local Perception of International Criminal Tribunals in the Former Yugoslavia and Sierra Leone, 603 Annals Am. Acad. Pol. & Soc. Sci. 226 (2006) (comparing the ICTY and SCSL).

145While there are similarities between domestic and international criminal law, the key international crimes (war crimes, crimes against humanity, and genocide) were created by international law and have no direct analogs in most domestic legal systems. See Cryer et al., supra note 83, at 8–12.

146See supra notes 135–137 and accompanying text.

147See supra Part I.A.

148See, e.g., ICTY, Rules of Procedure and Evidence, U.N. Doc. IT/32/Rev.49, (May 22, 2013), http://www.icty.org/x/file/Legal%20Library/Rules_procedure_evidence/IT032Rev49_en.pdf [hereinafter ICTY Rules].

149See infra Part IV.A.

150ICTY Rules, supra note 148, at R. 87(A). The trier of fact is often different, however. At the ICTY, the trier of fact is a Trial Chamber, composed of three judges. In many criminal trials in the United States the trier of fact is a jury. See Kwon, supra note 1, at 363.

151ICTY Rules, supra note 148, at R. 85. The overwhelming majority of accused at international criminal courts are male.

152See, e.g., International Covenant on Civil and Political Rights art. 14, Dec. 19, 1966, 999 U.N.T.S. 14668.

153Cf. Shany, supra note 26, at 249 (suggesting that it is appropriate to compare international and national courts that share “comparable functions”); Wippman, supra note 14, at 862.

154While the overall ICTY system is a blend of common and civil law features, the common law features are more prominent, particularly at the trial phase. See Megan Fairlie, The Marriage of Common and Continental Law at the ICTY and Its Progeny, Due Process Deficit, 4 Int’l Crim. L. Rev. 243 (2004); Kwon, supra note 1, at 363.

155The work of the investigative judge may well shorten the resulting trials by establishing much of the evidence that the parties would otherwise be required to present to the court, but would not show up in the variables used to compute the Complexity Score. Cf. Fairlie, supra note 154, at 248–51 (comparing the pre-trial phases in the common law and civil law systems).

156Prosecutor v. Jean-Paul Akayesu, Case No. ICTR-96-4-T, Judgment, 24, 28 (Sept. 2, 1998), http://www.unictr.org/Portals/0/Case/English/Akayesu/judgement/akay001.pdf.

157Kaing Guek Eav, Extraordinary Chambers in the Courts of Cambodia (ECCC), http://www.eccc.gov.kh/en/indicted-person/kaing-guek-eav (last visited Oct. 10, 2012) (identifying the number of trial days and witnesses); Prosecutor v. Kaing Guek Eav alias Duch), Case No. 001/18-07-2007/ECCC/TC, para 56 (2010), available at http://www.eccc.gov.kh/sites/default/files/documents/courtdoc/20100726_Judgement_Case_001_ENG_PUBLIC.pdf (identifying the number of trial exhibits).

158See Prosecutor v. Thomas Lubanga Dyilo, Case No. ICC-01/04-01/06, Judgment pursuant to Article 74 of the Statute, 11 (Mar. 14, 2012), http://www.icc-cpi.int/iccdocs/doc/doc1379838.pdf.

159See Prosecutor v. Charles Ghankay Taylor, Case No. SCSL-03-1-T, Judgment Summary, 8 (Apr. 26, 2011), http://www.worldcourts.com/scsl/eng/decisions/2012.12.17_Prosecutor_v_Taylor.pdf.

160See International Military Tribunal, Trial of the Major War Criminals before the International Military Tribunal, Nuremberg, 14 Nov. 1945–1 Oct. 1946 (1947). The introduction to the judgment contains a description of the trial, from which the figures necessary for calculating its complexity have been drawn.

161While the Judgment of the IMT is somewhat vague about the number of exhibits (it refers only to “several thousands of exhibits”), other sources give more exact figures. See Lawrence Douglas, The Memory of Judgment 12 (2001) (noting that there were approximately 4000 documents and 1800 photographs that were entered into the record as exhibits).

162See Hogan, supra note 133, at tbl. T-2.

163Id.

164Heuer & Penrod, supra note 33, at 34–35.

165They collected the information on trial length in days and hours and then converted them to trial hours at the rate of four hours per day for criminal trials. Since trial days were used rather than trial hours in this study, trial hours were converted back to days using the same conversion rate. Id. at tbl. 1 (explanatory note). This may not be an entirely accurate comparison as the ICTY defaults to five courtroom hours per day and, in practice, often exceeds this. See U.N. Secretary-General, Budget for the International Tribunal for the Prosecution of Persons Responsible for Serious Violations of International Humanitarian Law Committed in the Territory of the Former Yugoslavia since 1991, for the biennium 2012–2013, 31, U.N. Doc. A/66/386 (Sept. 29, 2011) [hereinafter Budget for the International Tribunal].

166Heuer & Penrod, supra note 33, at tbl.1.

167According to one study, the average non-capital murder trial in North Carolina in the early 1990s took 3.8 days. See Philip J. Cook & Donna Slawson, The Cost of Processing Murder Cases in North Carolina tbl. 6 (1993). The study did not contain data on the number of exhibits or witnesses, but when comparing the average length of a murder trial in that study to the average length of an ICTY trial, the ICTY case takes 45 times longer to present than the murder trial. See also infra Part IV.C.2.e–g.

168The categories with an asterisk in Figure 3 are based on the data from the Heuer and Penrod dataset. See Heuer & Penrod, supra note 33.

169Jesus Rangel, All 20 Acquitted in Jersey Mob Case, N.Y. Times, Aug. 27, 1988, at A111, 20. The article describes the Lucchese crime family trial as “[o]ne of the longest criminal trials in the United States.” Id. See also Paul Richter, Longest Mob Trial Ends in Acquittals, L.A. Times, Aug. 27, 1988, at 1, 20 (describing the trial as “the longest federal criminal trial in the nation’s history”).

170See Rangel, supra note 169.

171See Higgins, supra note 17, at 241–49 (describing the complexity of the trial).

172Damaska, supra note 16, at 340–41.

173The most significant guilty pleas came from Biljana Plavšić, who had been the co-President of the Bosnian Serb Republic (later the Republika Srpska), and Milan Babić, the President of the Republic of Serbian Krajina.

174See, e.g., U.N. Secretary-General, Assessment and report of Judge Theodor Meron, President of the International Tribunal for the Former Yugoslavia, provided to the Security Council pursuant to paragraph 6 of Security Council resolution 1534 (2004), and covering the period from 15 November 2011 to 22 May 2012, 30–34, U.N. Doc. No. S/2012/354 (May 23, 2012) (by Judge Theodor Meron).

175Budget for the International Tribunal, supra note 165, at 14.

176See supra notes 164–168 and accompanying text.

177While I have not measured the complexity of all the cases at the ICTR, the ICTR has indicted fewer accused and they appear to have been, in general, less complex indictments and trials. The ICC may eventually overtake the ICTY in terms of the total complexity of the cases it tries. However, this prospect would require the equivalent of another 45 trials of the same complexity as the Lubanga trial. I am unaware of any domestic courts that have tried a comparatively complex set of related criminal cases.

178This was calculated using the methodology described in Ford, supra note 3, at 989–90. The figure reported in my earlier article was updated to include data from the 2010, 2011 and 2012 United Nations General Assembly resolutions authorizing appropriations for the ICTY’s budget. See G.A. Res. 65/253, 3, U.N. Doc. 65/253 (Dec. 24, 2010); G.A. Res. 66/239, 10, U.N. Doc. A/66/239 (Dec. 24, 2011); G.A. Res. 67/243, 10, U.N. Doc. A/RES/67/243 (Dec. 24, 2012). The calculation was also updated to reflect an estimated termination of the ICTY’s operations in 2017 rather than 2014. This reflects the fact that the timeline for the ICTY’s completion has slipped once again. See supra note 4.

179Past inflation was accounted for using the Consumer Price Index produced by the U.S. Department of Labor, Bureau of Labor Statistics. Future inflation was assumed to be 2% per year.

180See Shany, supra note 26, at 237 (suggesting the study of the relative efficiency of international courts); Wippman, supra note 14, at 862 (noting that one way to determine whether the ICTY provides “value for money” is to compare the cost of criminal trials in national legal systems with those at the ICTY).

181See Oxford English Dictionary, supra note 46, at 84 (“[t]he ratio of useful work performed to the total energy expended”); Random House Webster’s College Dictionary, supra note 46, at 419 (“the ratio of the work done by a machine to the energy supplied to it”); The American Heritage Dictionary of the English Language, supra note 46, at 570 (“[t]he ratio of the effective or useful output to the total input in any system”).

182See supra note 1.

183For example, one might also calculate efficiency as the total gravity of the crimes adjudicated divided by the cost of those trials. This would be a different but also valid way of thinking about efficiency.

184For this purpose, the complexity of the 41 full trials that have been completed plus the partial trial of Slobodan Milošević were included. This yields a total complexity of 42.13. That score does not include any estimation of the complexity of the three trials that are currently underway or any estimation of the complexity associated with guilty pleas, cases transferred to domestic systems, or the contempt proceedings. See supra notes 172–175.

185See Fed. R. Evid. 101(a).

186See George Fisher, Evidence 2–3 (3d ed. 2013) (noting that 44 states “have adopted the Federal Rules [of Evidence] in whole or great part . . . .”).

187ICTY Rules, supra note 148, at R. 89(C).

188 Id. at R. 89(D).

189Id. at R. 90(F).

190Fed. R. Evid. 611(a). The last three clauses of Rule 403 may serve a similar purpose. See Fed. R. Evid. 403 (giving the court discretion to exclude relevant evidence if it would result in “undue delay, wasting time, or needlessly presenting cumulative evidence.”)

191ICTY Rules, supra note 148, at R. 90(G) & 90(H).

192Fed. R. Evid. 611(b).

193ICTY Rules, supra note 148, at R. 73 bis (C).

194Who Does What, Witnesses: Qs & As, Federal Judicial Center (Aug. 23, 2013), available at http://www.fjc.gov/federal/courts.nsf/autoframe?OpenForm&nav=menu5a&page=/federal/courts.nsf/page/311?opendocument (last visited Aug. 23, 2013) (“[T]he judge often sets limits on the number of witnesses and the amount of time for testimony in a particular case. . . . The goal in setting limits is to ensure that each party has sufficient time to make his or her case, but without redundancy.”).

195See Kwon, supra note 1, at 361 (“[T]he judges of the Tribunal have found themselves faced with the daunting task of determining how to change the existing procedures to be able to dispose of increasingly complex cases in a shorter period of time, while still respecting to the highest degree the rights of the accused to a fair trial.”).

196See Maximo Langer & Joseph W. Doherty, Managerial Judging Goes International, but Its Promise Remains Unfulfilled: An Empirical Assessment of the ICTY Reforms, 36 Yale J. Int’l L. 241, 245–51 (2011) (describing changes made to the Rules of Procedure and Evidence).

197ICTY Rules, supra note 148, at R. 73 bis (D).

198See id. at R. 73 bis (E).

199Id. at R. 73 bis (D).

200See Higgins, supra note 17, at 240.

201See id. at 251–57.

202Id. at 254. But see Kwon, supra note 1, at 372–75 (arguing that Rule 73 bis gives the Trial Chamber much needed authority to “compel the prosecution to focus the evidence presented on the most important counts and charges in the indictment”).

203ICTY Rules, supra note 148, at R. 92 bis.

204Wald, supra note 16, at 473.

205Id.

206See Turner, supra note 29, at 558–60.

207See Kwon, supra note 1, at 365 (arguing that the procedures on the use of witness statements “enhanced” the ability of the judges to manage the trials while simultaneously being “fully consistent with principles of justice and fair trial rights”).

208International Tribunal for the Prosecution of Persons Responsible for Serious Violations of International Humanitarian Law Committed in the Territory of the Former Yugoslovia since 1991, Updated Statute of the International Criminal Tribunal for the Former Yugolslavia, Art. 20(1), (Sept. 2009), available at www.icty.org/sections/legallibrary/statuteofthetribunal.

209See supra notes 202 & 205. There is significant disagreement about this, however. See generally Kwon, supra note 1 (rejecting arguments that there are due process concerns arising out of the changes made to the Rules of Procedure and Evidence).

210See Heidi L. Hansberry, Too Much of a Good Thing in Lubanga and Haradinaj: The Danger of Expediency in International Criminal Trials, 9 Atrocity Crimes Litigation Year-In-Review 357, 357 (2011) (arguing that “judges tend to consider any reason for delay as unjustified and to be avoided at all costs in order to promote trial expediency” and that this has undermined the protection of witnesses at the ICTY and ICC); Michele Caianiello, Law of Evidence at the International Criminal Court: Blending Accusatorial and Inquisitorial Models, 36 N.C. J. Int’l L. & Com. Reg. 287, 316–17 (2011) (expressing concern that the evidentiary rules at the ICTY were amended solely for purposes of expediency and not because of any legal deficiency in the previous rules); Charles Chernor Jalloh, Does Living by the Sword Mean Dying by the Sword?, 117 Penn. St. L. Rev. 707, 751–52 (2013) (arguing that over its lifetime the ICTY’s decision-making with respect to assignment of defense counsel shifted from an emphasis on the rights of defendants to an emphasis on expediency); Anna Petrig & Fausto Pocar, Case Referral to National Jurisdictions: A Key Component of the ICTY Completion Strategy, 45 Crim. L. Bull. 1, 1, 12 (2009) (worrying that referral decisions were driven largely by concerns of expediency in ways that undermined the legitimacy of those decisions); Daniel Tillay, The Non-Rules of Evidence in the Ad Hoc Tribunals, 45 Int’l Lawyer 695, 715–16 (2011) (worrying that rules about the presentation of evidence were changed largely for reasons of expediency without due regard for their effect on fairness).

211See Langer & Doherty, supra note 196 (concluding that the procedural changes made at the ICTY failed to meaningfully effect the length of the trials).

212See supra note 144 and accompanying text.

213See generally Wippman, supra note 14.

214All of the international criminal courts rely on states to enforce the sentences in their own criminal justice systems. See, e.g., Member States Cooperation, U.N. Int’l Crim. Trib. For the Former Yugoslavia, http://icty.org/sections/LegalLibrary/MemberStatesCooperation (indicating that 16 countries have entered into agreements for the enforcement of ICTY sentences).

215See Statute of the Special Court for Sierra Leone art. 1(1), Jan. 16, 2002, 2178 U.N.T.S. 145, available at http://www.rscsl.org/Documents/scsl-statute.pdf (limiting the personal jurisdiction of the SCSL to those “who bear the greatest responsibility for serious violations of international humanitarian law” including “those leaders who, in committing such crimes, have threatened the establishment and implementation of the peace process in Sierra Leone”). In contrast, the ICTY initially charged a fairly large number of low-ranking perpetrators. See Ford, supra note 71, at 71–72.

216See Prosecutor v. Moinina Fofana and Allieu Kondewa, Case No. SCSL-04-14-T, Judgment on the Sentencing of Moinina Fofana and Allieu Kondewa, Annex F at 21, (Spec. Ct. for Sierra Leone, Trial Chambers I, Oct. 9, 2007), available at http://www.refworl.org/docid/484417252.html. The CDF trial judgment does not contain a direct statement of the number of exhibits entered into the record during the trial. However, a review of the evidence used to support the judgment indicates that the case was heavily skewed towards witness testimony. Exhibits are rarely mentioned. The highest numbered exhibit mentioned in the judgment is Exhibit 147. See id. 98.

217See Prosecutor v. Issa Hassan Sesay et al., Case No. SCSL-04-15-T, Judgment Summary, Annex B at 32–36 (Mar. 2, 2009), available at http://www.refworld.org/docid/4ae9b31c2.html.

218See Prosecutor v. Alex Tamba Brima et al., Case No. SCSL-04-16-T, Sentencing Judgment, Annex A at 58–63, (Spec. Ct. for Sierra Leone, Trial Chambers II, June 20, 2007), available at http://www.refworld.org/docid/467fba742.

219See supra note 159.

220This cost was calculated by updating the methodology from Ford, supra note 3, at 993–95. It now appears that the SCSL will complete its operations at the end of 2013 with the issuance of the Appeal Judgment in the case against Charles Taylor. See Press Release, Special Court for Sierra Leone, Outreach and Public Affairs Office, Oral Hearings Conclude in Taylor Appeal, Judges Will Now Retire to Deliberate and Consider Judgment, (Jan. 23, 2013), available at http://www.sc-sl.org/LinkClick.aspx?fileticket=pTTabQ5Bm2U%3d&tabid=53.

221Past inflation was accounted for using the Consumer Price Index produced by the U.S. Department of Labor, Bureau of Labor Statistics. Future inflation was assumed to be 2% per year.

222See Higonnet, supra note 10, at 348 (arguing that hybrid courts would be a “cost-efficient, high impact” alternative to the ICTY and ICTR); id. at 385–86 (arguing that the SCSL was “more efficient” than the ICTY); Mathias Holvoet & Paul de Hert, International Criminal Law as Global Law: An Assessment of the Hybrid Tribunals, 17 Tilburg L. Rev. 228, 232 (2012) (describing the SCSL as “effective” and “relatively cheap”); David M. Crane, Dancing with the Devil: Prosecuting West Africa’s Warlords: Building Initial Prosecution Strategy for an International Tribunal after Third World Armed Conflicts, 37 Case. W. Res. J. Int’l L. 1, 2 (2005) (arguing that the SCSL was “efficiently and effectively” delivering international criminal justice); Kelli Mundell, Capturing Women’s Experiences of Conflict: Transitional Justice in Sierra Leone, 15 Mich. St. J. Int’l L. 85, 94 (2007) (noting that the SCSL was supposed to be more cost effective and efficient than the ICTY); see generally J. Peter Pham, A Viable Model for International Criminal Justice: The Special Court for Sierra Leone, 19 N.Y. Int’l L. Rev. 37, 89–90 (2006).

223See, e.g., Laura A. Dickinson, The Promise of Hybrid Courts, 97 Am. J. Int’l L. 295, 306–08 (2003) (noting that hybrid tribunals may also enhance perceived legitimacy, perceived independence of the judiciary, accountability, and the creation of regional human rights norms, among other things).

224See Charles Chernor Jalloh, Special Court for Sierra Leone: Achieving Justice?, 32 Mich. J. Int’l L. 395, 446–51 (2011); id. at 446 (“The SCSL was specifically designed to improve on the trial efficiency of the ICTR and ICTY.”).

225See Wippman, supra note 14, at 863 (stating that neither federal nor state agencies routinely calculate the per trial cost and length of criminal trials).

226See Cook & Slawson, supra note 167, at 54 (reporting figures for the cost of murder trials in North Carolina in 1991 and 1992); The Cost of Justice, L.A. Times, Sept. 13, 1998, at 2 (containing cost figures for a number of famous trials in the Los Angeles area); Russell Gold, Counties Struggle with High Cost of Prosecuting Death-Penalty Cases Wall St. J., Jan. 9, 2002, at B1 (describing cost of death-penalty cases); Eli Braun, Ohio Just. & Pol’y Ctr., $42,000 for a Courthouse Hour: The Cost of Processing Adult Criminal Cases in Hamilton County, Ohio, (2010), available at http://www.ohiojpc.org/text/publications/court%20cost.pdf; Natasha Minsker, ACLU of N. Cal., The Hidden Death Tax: The Secret Costs of Seeking Execution in California (Claire Cooper & Elise Barducci eds., 2008), available at https://www.deathpenalty.org/downloads/The_Hidden_Death_Tax.pdf; John Roman, et al., Urb. Inst. Just. Pol’y Center, The Cost of the Death Penalty in Maryland, (2008), available at: http://www.urban.org/UploadedPDF/411625_md_death_penalty.pdf; John G. Morgan, Comptroller of the Treasury (Tenn), Tennessee’s Death Penalty: Costs and Consequences (2004); Arthur L. Alarcón & Paula M. Mitchell, Executing the Will of the Voters?: A Roadmap to Mend or End The California Legislature’s Multi-Billion Dollar Death Penalty Debacle, 44 Loy. L.A. L. Rev. S41, S72–75 (2011); Terance D. Miethe, Estimates of Time Spent in Capital and Non-Capital Murder Cases: A Statistical Analysis of Survey Data from Clark County Defense Attorneys (Feb. 21, 2012) available at http://www.deathpenaltyinfo.org/documents/ClarkNVCostReport.pdf; State of Kansas, Legislative Division of Post Audit, Costs Incurred for Death Penalty Cases: A K-GOAL Audit of the Department of Corrections (2003), available at http://www.kslpa.org/docs/reports/04pa03a.pdf; Aliza B. Kaplan, Oregon’s Death Penalty: The Practical Reality, 17 Lewis & Clark L. Rev. 1, 32–49 (2013); Jon B. Gould & Lisa Greenman, Report to the Committee on Defender Services—Judicial Conference of the United States—Update on the Cost and Quality of Defense Representation in Federal Death Penalty Cases (2010), available at http://www.uscourts.gov/uscourts/FederalCourts/AppointmentOfCounsel/FDPC2010.pdf; Margot Garey, The Cost of Taking a Life: Dollars and Sense of the Death Penalty, 18 U.C. Davis L. Rev. 1221 (1985).

227 See Translation and Interpretation, About the ICTY, http://www.icty.org/sid/165 (last visited Sept. 23, 2014).

228See Victims and Witnesses Section, Information Booklet for ICTY Witnesses (2007) at 14, available at http://www.icty.org/x/file/About/Registry/Witnesses/witnesses_booklet_en.pdf.

229Id. at 22.

230Richard Dickner & Elise Keppler, Beyond the Hague: The Challenges of International Justice, in Human Rights Watch World Report 2004, available at http://www.hrw.org/news/2004/01/26/beyond-hague-challenges-international-justice & http://www.hrw.org/legacy/wr2k4/download/10.pdf.)

231The Cost of Justice, About the ICTY, http://www.icty.org/sid/325 (last visited Sept. 23, 2014).

232U.N. Secretary-General, Second Performance Report on the Budget of the International Tribunal for the Former Yugoslavia for the Biennium 2012–2013, 2–3 tbls. 1 & 2, U.N. Doc. A/68/582 (Nov. 11, 2013).

233For example, the cost of the OTP was apportioned to investigation, prosecutions, appeals, and other tasks, depending on the number of personnel assigned to the Prosecution Division, Investigation Division, Appeals Division, or Immediate Office of the Prosecutor. The cost of the Registry was similarly apportioned between support tasks and tasks directly associated with trials and appeals depending on the number of personnel assigned to the various divisions within the Registry.

234This process was relatively easy for the ICTY budgets covering the years 1994 to 2005, as the publicly available budget documents for those years were very thorough. It was more difficult for the years 2006 to 2013 as the ICTY’s budgets have become less detailed over time. Nevertheless, it was possible to calculate the cost of the various functions of the ICTY with reasonable certainty, even for the latter years. Estimates for the years 2014–2017 are the result of educated guesses about the ways in which ICTY spending will change as the trials are completed and the focus shifts to appeals. The data that form the basis for Table 5 are available from the Author upon request.

235There was no way to extract the pre-trial costs from the trial costs based on the information in the ICTY’s budgets.

236Costs were apportioned between direct trial costs and direct appeal costs by taking total judicial costs (essentially the sum of the cost of the Chambers, the Judicial Support Division of the Registry plus any items related to judicial costs) and then prorating those costs based upon the number of judges who were assigned to trials versus appeals in any given year.

237See Wippman, supra note 14, at 877–78.

238Id. at 877.

239Id.

240Id.

241See U.N. Secretary-General, Comprehensive Report on the Progress made by the International Criminal Tribunal for the Former Yugoslavia in Reforming its Legal Aid System, 31, U.N. Doc. A/58/288 (Aug. 12, 2003).

242Id. 31, Annex I. As with direct trial and appeal costs, the total amount of legal aid was apportioned between trials and appeals based on the number of judges assigned to each of those functions in any given year.

243 See United States v. McVeigh, 153 F.3d 1166, 1176 (10th Cir. 1998).

244See Wippman, supra note 14, at 862; Howard Pankratz, Bomb Trials Costs $82.5 Million, denverpost.com, Nov. 3, 1998, available at http://extras.denverpost.com/bomb/bomb1103.htm [hereinafter Bomb Trials].

245See Bomb Trials, supra note 244 (noting that the FBI had 2,592 agents investigating the case in 1995 and that this number decreased rapidly over time). The 2012 cost was assumed to accrue in 1995 because it is clear that the vast majority of resources were expended during that year. (Author’s Calculation).

246Bomb Trials, supra note 244.

247Defense costs were assumed to accrue in 1997, the year of the trial. (Author’s Calculation).

248McVeigh, 153 F.3d at 1177.

249Id. at 1179. The presentation of evidence for the penalty phase began on June 4, 1997, and ended on June 12, 1997. Id. Assuming that the court did not hear evidence over the weekend, this means that the penalty phase took seven days.

250This calculation is based solely on trial days, rather than all of the components of the Complexity Score, because information is not readily available about the number of witnesses and exhibits presented during the McVeigh trial. If a complexity calculation is to be based on just one of the components, the number of trial days is the best component to use. See Heise, supra note 33, at 360.

251This was calculated as follows: equivalent cost = (number of trial days in McVeigh trial)÷(total number of trial days in all ICTY trials)×(total cost of ICTY)×(percentage of cost attributable to prosecutions + percentage of cost attributable to investigations). With the actual figures it becomes: (30)÷(7691)×(3079726477)×(0.0662+0.1137). Subsequent calculations of the equivalent cost of an ICTY trial use an analogous methodology.

252Professor Wippman suggests that the McVeigh trial was unusually short given its complexity due to tactical decisions made by the prosecution and defense to call a smaller number of witnesses and use fewer exhibits than either side had initially intended. See Wippman, supra note 14, at 873. Even if the prosecution and defense had chosen differently and thus doubled the length of the trial, the basic conclusion about the trial’s efficiency would remain the same.

253Bomb Trials, supra note 244.

254Id.

255See United States v. Jeffrey K. Skilling, 638 F.3d 480, 481 (5th Cir. 2011).

256See Carrie Johnson, After the Enron Trial, Defense Firm is Stuck With the Tab Wash. Post, June 16, 2006, at D1.

257Id. at D3.

258See Brief for the United States as Appellee at 5, United States v. Jeffrey K. Skilling, 554 F.3d 529 (5th Cir. Nov. 13, 2007) (No. 06-20885), 2007 WL 4207556.

259See Johnson, supra note 248, at D1.

260See James Sterngold, Not Guilty: The Man; Simpson Walks Out of the Courtroom a Free Man and Into the Lucrative Free Market, N.Y. Times, Oct. 4, 1995, at A10.

261According to USA Today, the trial lasted 141 trial days (including opening and closing statements), 126 witnesses testified, and 857 exhibits were entered into evidence. See The O.J. Simpson trial, by the numbers, USA Today (Oct. 18, 1996), http://usatoday30.usatoday.com/news/index/nns062.htm.

262See id.

263See The State’s Case: Jun 1–Oct 15, 2004, PWC Consulting, http://pwc-sii.com/Timeline/State.htm (last visited Sept. 19, 2014) [hereinafter The State’s Case]. Note that this website is not an impartial source; it is devoted to the idea that Scott Peterson is innocent. Nevertheless, it appears to be reliable for the purpose of identifying the number of trial days.

264See The Defendant’s Case, Jury Deliberations, & Verdict: Oct 6–Nov 12, 2004, PWC Consulting, http://pwc-sii.com/Timeline/Defense.htm (last visited Sept. 19, 2014).

265See Penalty Phase, Motions, & Sentencing: Nov 13, 2004–Mar 17, 2005, PWC Consulting, http://pwc-sii.com/Timeline/Penalty.htm (last visited Sept. 19, 2014).

266The State’s Case, supra note 263. They indicate that the court heard testimony Mondays through Thursdays but was not in session on Fridays. Id. There were also some additional days when the court was not in session. These days included July 1st, July 5th, July 8th, August 9th, August 19th, and September 6th. Id.

267See Minsker, supra note 226, at 21.

268See Cook & Slawson, supra note 167, at 47 (reporting figures for the cost of murder trials in North Carolina in 1991 and 1992).

269Id. at 45.

270Cf. id. In this way, their calculation does not include the total judicial cost of the trial, which the ICTY figures do, but there is no way to account for this discrepancy.

271Id. at tbls 6.2 & 6.3.

272Id. at 96.

273See Roman, et al., supra note 226.

274Id. at tbls. 5 & 6.

275Id. at app. B (describing how the cost data was calculated).

276Kan. Legis. Division of Post Audit, Costs Incurred for Death Penalty Cases: A K-GOAL Audit of the Department of Corrections (2003), available at http://www.kslpa.org/docs/reports/04pa03a.pdf.

277Data used to calculate these figures was taken from id. at Chart I-2 and Appendix E.

278See Cook & Slawson, supra note 167, at 6; Roman et al., supra note 226, at tbl.1.

279For example, the cost of supporting a family consisting of two adults and two children in North Carolina is roughly $39,000 in North Carolina and $47,000 in Maryland. Amy K. Glasmeier, Living Wage Calculator, M.I.T., http://livingwage.mit.edu/ (last updated Mar. 24, 2014).

280See supra Part III.C.

281See Ford, supra note 71, at 68 n.92.

282See supra Part II.A.

283See supra Part I.D.

284Id.

285See Cook & Slawson, supra note 167, at 49; Roman, et al., supra note 226, at 24; Kan. Legis. Division of Post Audit, supra note 276, at 18.

286See Cook & Slawson, supra note 167; Roman, et al., supra note 233; Kan. Legis. Division of Post Audit, supra note 276.

287See Ford, supra note 71, at 76, 104.

288See Cohen & Kyckelhahn supra note 134; id. at 91.

289See Carol Rosenberg, Lawyers Spar Over Funding in Guantanamo Court, Miami Herald, Apr. 12, 2012, http://www.miamiherald.com/760/index.html.

290Id.

291The initial investigation was described as involving more than 1,000 FBI employees who were sent to East Africa, including more than 500 FBI agents. See Wippman, supra note 14, at n.80.

292See, e.g., Joseph E. Stiglitz & Linda J. Bilmes, The Three Trillion Dollar War: The True Cost of the Iraq Conflict 7–8 (2008) (“The tone of the administration was cavalier, as if the sums involved were minimal.”).

293See Walter Gibbs & Elinor Schang, Norway Police To Track Killer’s Funds in Year-Long Probe, Reuters Norway (Aug. 3, 2011), http://in.reuters.com/article/2011/08/03/idINIndia-58607220110803.

294See Ford, supra note 71, at 79, 100.

295See Gibbs & Schang, supra note 284.

296Id.

297See Timeline: Anders Breivik, Aljazeera, http://www.aljazeera.com/news/europe/2012/08/201282314116683446.html (last modified Aug. 24, 2012).

298See Alastair Reed, Kristin Myers & Josiane Kremer, Breivik Claims Self-Defense as Oslo Terror Trial Starts, Bloomberg (Apr. 16, 2013), http://www.bloomberg.com/news/2012-04-16/breivik-defends-massacre-as-norway-s-saddest-trial-starts.html.

299See Breivik’s Attacks Cost Billions, newsinenglish.no (Aug. 24, 2012), http://www.newsinenglish.no/2012/08/24/breiviks-attacks-cost-billions/.

300This was calculated using an exchange rate of .1689 Kroner to the Dollar. See Historical Exchange Rates, http://www.oanda.com/currency/historical-rates/ (last visited Sept. 26, 2013).

301See Breivik’s Attacks Cost Billions, supra note 299. Various other articles contain some cost data, but none seem as comprehensive. See Reed, Myers & Kremer, supra note 298 (noting that the court estimated before the trial began that the cost of court administration during the proceeding would be 76 million kroner).

302Madrid Train Bombings Trial Wraps Up, L.A. Times, July 3, 2007, at 8, available at http://articles.latimes.com/2007/jul/03/world/fg-briefs3.2.

303See Paloma Díaz Sotero, Desglose Económico del Ministerio: ¿Cuánto Cuesta el Macrojuicio del 11-M?, El Mundo, http://www.elmundo.es/elmundo/2007/06/01/espana/1180718228.html (last updated June 2, 2007).

304The exchange rate for July 2007 was 1.3704 Euros to the Dollar. See Historical Exchange Rates, supra note 300. Inflation data came from the Bureau of Labor Statistics. See Consumer Price Index – All Urban Consumers, available at http://www.bls.gov/cpi/cpid07av.pdf.

305See Bomb Trials, supra note 244.

306See, e.g., Arden Rowell & Lesley Wexler, Valuing Foreign Lives, 48 Ga. L. Rev. (2014) 499, 539 (describing how U.S. law consistently assigns lower values to foreign lives than domestic lives).

307See supra notes 295–304 and accompanying text.

308See Naomi Ellemers & S. Alexander Haslam, Social Identity Theory, in 2 Handbook of Theories of Social Psychology 379, 380 (Paul A.M. Van Lange et al. eds., 2012). See also Rowell & Wexler, supra note 306, at Section II(B)(4).

309See Rowell & Wexler, supra note 306, at 519 (discussing differing philosophical views of whether states owe obligations to individuals outside their borders).

310See supra notes 5–10.

311 See Ford, supra note 71, at 104 (noting that the average ICTY indictment contains allegations roughly equivalent to the bombing of the Alfred P. Murrah Federal Building in Oklahoma City).

312See supra notes 222–224 and accompanying text.

313See supra notes 5–13.

314See Ford, supra note 3, at fig.2 (charting spending on international criminal tribunals over time).

315See United Nations Diplomatic Conference on the Establishment of an International Criminal Court, Rome, July 17, 1998, Rome Statute of the International Criminal Court, Preamble, U.N. Doc. A/CONF.183/9, available at http://www.icc-cpi.int/iccdocs/PIDS/publications/RomeStatutEng.pdf.

316See supra Part IV.A.

317See Wald, supra note 16, at 466.

318See Tom Tyler, Legitimacy and Compliance: The Virtues of Self-Regulation, in Legitimacy and Compliance in Criminal Justice 8, 11 (Adam Crawford & Anthea Hucklesbury eds., 2013); Tom R. Tyler, Legitimacy and Rule Adherence: A Psychological Perspective on the Antecedents and Consequences of Legitimacy, in 11 The Psychology of Justice and Legitimacy 251 (D. Ramona Bobocel et al. eds., 2010).

319See supra note 202. See also ICTY Rules, supra note 148, at R. 73bis(D) (suggesting that charges are “reasonably representative” if they take into account “all the relevant circumstances” including the classification and nature of the charges, the location of the alleged crimes, the scale of the crimes, and the victims of the crimes); Daryl A. Mundis, Book Review, 102 Am. J. Int’l L. 691, 693 (2008) (arguing that prosecutors are “duty bound to draft indictments that reflect the scale and scope of [the accused’s] criminality”).

320See Robert Cryer, Prosecuting the Leaders: Promises, Politics and Practicalities, 1 Göttingen J Int’l L. 45, 72–74 (2009); Suzan M. Pritchett, Entrenched Hegemony, Efficient Procedure or Selective Justice?: An Inquiry into Charges for Gender-Based Violence at the International Criminal Court, 17 Transnat’l L. & Contemp. Probs. 265, 288 (2008); Mundis, supra note 319, at 693 (“Can anyone today claim that the trial of Saddam Hussein before the Iraqi Special Tribunal and his execution solely for the crimes in Dujail held him accountable for the totality of his criminal record?”).

321See supra note 158 (noting that the Lubanga case took 204 trial days to hear).

322See Ford, supra note 2, at Section IV.

323It seems likely that there are “economies of scale” at international criminal tribunals and that tribunals that try more cases will be more efficient than tribunals that try fewer cases. This may be one reason the SCSL’s efficiency was lower than the ICTY’s efficiency.

324See Ford, supra note 71, at 73–74, 97.

325See id. at 97.

326See supra note 315.

327See Shany, supra note 26, at 230, 237 (contrasting effectiveness, the ability of an organization to accomplish its purposes, with efficiency, the relationship between an organization’s inputs and outputs).

328See Kwon, supra note 1, at 362 (concluding that the complexity of cases is driven by factual and legal complexity, as well as due process concerns); Wald, supra note 16, at 468 (concluding that international trials are fundamentally different from most trials in the United States and that they will inevitably be more complex).

329See supra note 18 and accompanying text.