Constructing an AtrocitiesRegime: The Politicsof War CrimesTribunals Christopher Rudolph

Itis the bidding of mercy, if not of justice, that, except for reasons that are weightyand will affect the safety of many, no action should be attempted wherebyinnocent persons may be threatened with destruction. —HugoGrotius (1583– 1645) 1 Imagesof destruction are often incomprehensible. Describing the aftermath of ethnicviolence in Rwanda in the spring of 1994, one reporter wrote,

Blood oweddown the aisles of churches where many sought refuge; Ž ve priestsand twelve women hiding out in aJesuitcenter were slaughtered.A Red Crossambulance was stoppedat a check-point,the six wounded patients draggedout and bayoneted to death. Toddlers lay sliced in half, and mothers withbabies strapped to their backs sprawled dead on thestreets of Kigali.The Žghtingwas handto hand, intimate and unspeakable, a kindof bloodlustthat leftthose who managed to escape it hollow-eyed and mute. 2

Unfortunately,Rwanda is but one of manycases where con ict has been marked byatrocities.The linkage between ethnic con ict and unspeakable brutality, as well astheir propensity to spreadacross borders, has caused some analysts to warn that “theever-increasing number of identity-based con icts suggests that we maybe facingan epidemic of potentiallycatastrophic proportions— anew world disorder.” 3 Surveyingthe evidence of Serbia’ s programof “ ethniccleansing” in Kosovo, UN war crimesprosecutor Louise Arbour remarked, “ thescale of the destruction is

IthankArthur Stein; Kenneth Abbott; Gary JonathanBass; theeditors of IO,Peter Gourevitchand DavidLake; and two anonymous referees fortheir extremely helpfuland constructive comments. The generousŽ nancialsupport of the Institute for the Study of World Politics, Washington, D.C., and the UCLA Graduate Divisionis gratefullyacknowledged. 1.Grotius [1925] 1962, 733– 34. 2. Time, 16May1994, 58. 3.Lake andRothchild 1998b, 339.

InternationalOrganization 55,3, Summer 2001,pp. 655– 691 © 2001by TheIO Foundationand the Massachusetts Instituteof Technology 656 InternationalOrganization reallyvery overwhelming.” 4 From thenotorious “ killingŽ elds”of Cambodia to recentevidence of brutality in Sierra Leone,the grizzly nature of ethnicand other identity-orientedcon ict incites horror, outrage, and a humandesire for justice. Inresponse to reports of atrocities in Bosnia, Kosovo, and Rwanda the interna- tionalcommunity established ad hoc international war crimestribunals to investi- gatecrimes and prosecute perpetrators. 5 Successiveefforts havebeen made to expandthe atrocities regime by forming a permanenttribunal, the International CriminalCourt (ICC), aprocessguided by the International Law Commission (ILC). Thisprocess culminated in the Rome Statute of the International Criminal Court(Rome Statute) created in Juneand July 1998. 6 WhenratiŽ ed byatleastsixty nations,the statute will establish the ICC atThe Hague. Proponents support internationaltribunals not only as a meansof holding perpetrators of atrocities accountablebut also as amechanismof peace by establishingjustice and promoting reconciliationin war-torn regions. Former U.S. Secretaryof State Madeline Al- brightproposed that, “ Inthe end, it isvery difŽ cult to havepeace and reconciliation withoutjustice.” 7 Thisanalysis has two primary purposes— the Ž rst isconstitutive (regime as dependentvariable), the other is causal (regime as independent variable). 8 I seek to identifyand analyze the myriad political and procedural obstacles to establishingan effectiveatrocities regime by examininghumanitarian norms, the strategic interests ofpowerful states, and bureaucratic factors. I alsoseek to determine how the emergentregime may or may not be effective in achieving its primary goal (individualconvictions) as well as itssecondary and perhaps more salient goal: to manageviolent con ict and reduce the likelihood of future transgressions. I argue thatalthough liberal humanitarian ideas have created the demand for political action,the process of dealing with brutality in war hasbeen dominated by realpolitik —thatis, furthering the strategic interests of the most powerful states. However,by understandingthe political interests and procedural obstacles involved, theinternational community can make institutional adjustments in the design and implementationof an atrocities regime to bridge the gap between idealpolitik and realpolitik. Thisanalysis is a Žrst steptoward explaining both how a newinterna- tionallegal regime is constitutedand how it mayfunction in managing a formidable problemin world politics. Althoughevents in areas such as the Balkans, Rwanda, Sierra Leone,and Cambodia(among others) are truly horriŽ c, they are not unique in the history of warfare; however,the current drive to establish humanitarian norms of con ict is novelin humanhistory. Historically, warfare hasbeen viewed as consistentwith the

4.Interview with Louise Arbour by FrancinePelletier, TheNational Magazine, CanadianBroadcast- ingCorporation, 29 July1999. 5.Another is beingestablished for Sierra Leonepursuant to S.C. Res. 1315,UN SCOR,UN Doc. S/RES/1315(2000). 6.Rome Statuteavailable at ^http://www.un.org/law/icc/statute/romefra.htm &. 7. LosAngeles Times, 19October 1999, A6. 8.On constitutive versus causal analyses,see Wendt1999. Constructingan Atrocities Regime 657 lawsof nature.Hugo Grotius, in his seminal work DeJureBelli ac PacisLibri Tres (TheLaw of War and Peace), providesvivid accounts of wartime brutality consistentwith norms of the time, citing Hellenic, Roman, and Biblical texts. Moreover,though Grotius includes some limitations on what was permissiblein war, theywould certainly be considered barbaric by modern liberal sensibilities. Thesenorms permitted, for example,the killing or injuringof all who were inthe territoryof the enemy, including women, children, captives, and those whose surrenderhad not been accepted. 9 Ratherthan focusing on the jusin bello, Grotius isin fact more concerned with notions of the jus ad bello.10 Inmatters involving acts of war andtreatment of a nation’s citizenry,the dominantnorm in the modern period is deference to national sovereignty. In fact, “priorto 1945,no principleof internationallaw was morewidely revered in practice thanthe idea of ‘domesticjurisdiction’ on matters relating to humanrights.” 11 Since theHolocaust, however, there has been tremendous interest in promoting human rightsand creating more stringent standards of international conduct, including duringarmed con icts, that is consistentwith these evolving ideas. 12 Thisnormative andinstitutional development prompts several related questions: What explains the dramaticturn in the 1990s toward legalization? What drives the process of forming andapplying the regime in given cases? Once established, what makes tribunals moreor less effective? How usefulis legalizationin curbing abhorrent behavior and promotingnational reconciliation?

Ideas, Interests, andInstitutions

Scholarsof international law (IL) focusessentially on precedent and procedure, whereasscholars of international relations (IR) focuson explicating the political dynamicsthat shape outcomes, the approach used here. By introducing an IR perspectiveto the process of legalization,Kenneth Abbott suggests that, “ IRhelps usdescribelegal institutions richly, incorporating the political factors that shape the law;the interests, power, and governance structures of states and other actors; the information,ideas, and understandings on whichthey operate; [and] the institutions withinwhich they interact.” 13 Althoughthe movement to establish a universal atrocitiesregime (through an international criminal court) is predicated on the internationalcommunity’ s desireto strengthen norms of human rights and justice, itis fraught with political obstacles and differing views on how to negotiate this complicatednormative and strategic terrain. In this analysis I synthesizeapproaches

9.Grotius [1925] 1962, 641– 62. 10. Jus in bello focuses onconduct in war andthe protection of civilians during armed conict (crimes ofwar), whereas jus ad bello refers toacceptable justiŽcations for the resort to armed force (thejust war). See Christopher1994. 11.Beres 1988,124. 12.See Sikkink1993 and 1998; and Finnemore 1996. 13.Abbott 1999, 362. 658 InternationalOrganization inIL andIR toexplain the linkages between law and politics in the international arena. Withinthe domain of IRtheory, realists have traditionally said little about issues inworld politics that do not involve questions of relative power and material interest.Realists generally argue that in aworldof asymmetricalpower distribution withno international body to exert pressure, “ logicsof consequences dominate logicsof appropriateness.”14 However,although power and material interests do not explainthe establishment of a humanitarian-basedregime, they do explain state responsesto domestic pressures for regimeformation. 15 Realistspredict that powerfulstates will not accept a regimethat signiŽ cantly undermines its ability to respondto perceived security threats. Moreover, they would predict that both the forms suchinstitutions take and the application of their jurisdictions in particular caseswill thus re ect the interests and relative power of thestates involved. Thus, “humanitarian”policies may in fact be manifestations of what Stephen Krasner refers toas “ organizedhypocrisy.” 16 Incontrast to realists, constructivists reject this notionthat state interests are static and centered only on material factors; they suggestthat such factors explain neither state behavior regarding human rights nor humanitarianintervention. 17 Regardingthe creation of war crimestribunals, con- structivistswould argue that evolving liberal ideas and concern for humanrights explainoutcomes and that analysis should focus on these variables in explaining regimeformation. Ideas and norms produce outcomes either through “ pathdepen- dence”or internationalsocialization and gain strength as theybecome increasingly embedded,producing an idealpolitik tocomplement realpolitik .18 Bridgingthe gap between these two points of view, liberal institutionalism suggeststhat the proclivity for conict in the anarchic international system can be overcomethrough carefully designed institutions whose purpose is international cooperation. 19 Statesengage in international regimes and abide by international treatiesto realizegains contingent on cooperation,and states may forgo short-term gainsto obtain long-term objectives. In the case of theemerging atrocities regime, thesegoals clearly are attempts to alleviate political and identity-based con ict, both inspace and time (by containing the spread of conict and by preventingcon ict in thefuture). These goals have both normative (human rights) and instrumental (internationalstability) dimensions. 20 Obviously,establishing an atrocities regime

14.See Krasner 1999,51; see alsoMorgenthau 1985; and Carr 1961. 15.In thecase ofthe atrocities regime, thispressure comes primarilyfrom international lawyers and nongovernmentalhuman rights organizations. 16.Krasner 1999. 17.See Finnemore1996; Goldstein and Keohane 1993; Katzenstein 1996;Sikkink 1993; and Wendt 1992and 1999. 18.On “ pathdependence,” see Weber 1920;Goldstein and Keohane 1993; and Meyer, Boli, and Thomas1987. On international socialization, see Bull1977; and Watson 1992. 19.See Abbottand Snidal 1998; Axelrod 1984; Axelrod and Keohane 1985; Keohane and Martin 1995;Oye 1986;and Stein 1990. On regime theory,see Hasenclever, Mayer,and Rittberger 1997; and Krasner 1983. 20.Lake andRothchild 1998c. Constructingan Atrocities Regime 659 goesbeyond problems of collective action and coordination. Legalization along theselines is intended to produce compliance (that is, deterrence). 21 TheIL variant ofinstitutionalism suggests that codiŽ ed norms of conduct coupled with speciŽ c sanctionswill affect state behavior directly, a distinctbroadening of the scope of institutionalismwithin IR. 22 Appliedto the case of war crimestribunals, this perspectivesuggests that success hinges on regime design and the strength of the resultinginstitution. 23 Thecentral tension here is between “ hard”and “ soft”law. 24 Thosewho favor hard law in international legal regimes argue that it enhances deterrenceand enforcement by signaling credible commitments, constraining self- servingauto-interpretation of rules, and maximizing “ compliancepull” through increasedlegitimacy. 25 Thosewho favor soft law argue that it facilitates compro- mise,reduces contracting costs, and allows for learningand change in the process ofinstitutional development. 26 Otherissues of institutional design and procedure crucialto the creation and functioning of an effective regime include such key factorsas jurisdiction, evidence collection, witness protection, and coordination withnational governments and their domestic legal systems. Institutionalists would predictthat a well-structuredatrocities regime will not only hold orchestrators of genocideand crimes against humanity accountable but also deter future atrocities andhelp to alleviate tensions in sensitive regions prone to egregious acts of violence. Ibeginmy analysis with three cases where tribunals were successfullyestab- lished:Bosnia, Rwanda, and Kosovo. These cases show the strong link between politicalchallenges and legal (and procedural) challenges, especially when strategic interestsof powerfulstates are not at stake.Whereas the case of Bosniareveals the politicalobstacles to initially establishing an internationallegal regime, the cases of Rwandaand Kosovo illustrate both the dynamic process of legalization and the effectsof institutional learning; they also reveal the limited deterrent capability of theatrocities regime— atleast in the early stages of itsdevelopment. I thenexamine twocases where tribunals were notsuccessfully established: Cambodia and East Timor.I alsoexamine the case of the ICC, which continues to be marked by difŽculties in achievinggreat power support. These difŽ culties show how powerand strategicinterests dominate regime formation; they also point to the need for a

21.On “ legalization,”see Abbottet al.2000. 22.Goldstein et al.2000. 23.Keohane 1997, 501. Oran YoungidentiŽ es three typesof regimes: spontaneous,negotiated, and imposed.While constructivists might focus on “ spontaneous”orders, liberal institutionalists would examine thefactors at playas theelements ofanew regime are negotiated,as Idohere. Young 1983, 98–101. 24.Kenneth Abbott and Duncan Snidal deŽ ne “ hard”legalization as legallybinding obligations characterized byhighdegrees ofobligation,precision, and delegation, and deŽ ne “ soft”legalization as amore exiblemanifestation characterized byvarying degrees alongone or more ofthese same dimensions.Abbott and Snidal 2000. 25.See Abbottand Snidal 2000; and Franck 1990. 26.Abbott and Snidal 2000; on  exibilityand learning in international agreements, see Koremenos 1999. 660 InternationalOrganization

“softening”of the legalization process if politicalobstacles are to be successfully overcome.Finally, I presentan analytical section that addresses my primary questions:why and when are tribunals established, how successful are they in addressingand preventing atrocities, and what factors shape the development of the atrocitiesregime?

The ICTYin Bosnia

Thecase of theInternational Criminal Tribunal for theformer Yugoslavia(ICTY) illustratesthe political difŽ culties associated with establishing an internationallegal regimewhere the strategic interests of powerful states are not directly at stake, thoughsuch challenges are not insurmountable. This case is especiallysalient given internationallawyers’ initial desires to form aregimebased on hardlaw, that is, one thatcould transcend realpolitik byeliminatingdistinctions between powerful states andweak states (equality under the law) andcould challenge long-held notions of sovereignty.There are legal obstacles to creatinghard law in an institution built on internationalismand attempting to bring together states with very disparate legal foundations.The case of the ICTY revealsthe relevance of realism in explaining tribunalaction and the process of institutionalization.Although norms and ideas of humanrights prompt calls for stateaction in casesof genocideand war crimes,the caseof theICTY illustrateshow the strategic interests of powerfulstates (through theUN SecurityCouncil) shape the process of institutionalizati onand its use. Callsfor aninternational tribunal were Žrst madein spring 1992, and the ICTY was establishedon 25May1993 by theUN SecurityCouncil. 27 Theresolution was approvedby a unanimousvote and established the Ž rst tribunalof its kind since WorldWar II. However,even though the resolution passed unanimously, the tribunalsince its inception has been fraught with political con ict and doubt. As one representativeof theUN OfŽce of theLegal Counsel lamented, “ Inmy twenty years ofexperiencein the United Nations, I havenever encountered as much skepticism ashas surrounded the establishment of this tribunal.” 28 Thatwar brokeout in Bosnia on6April1992 certainly should not have come as atremendoussurprise given the longhistory of ethnic violence in the region. 29 At theend of the Cold War the disintegrationof Sovietbloc nations proceeded with considerable speed, a process thataugmented majority-minority relations and exacerbated ethnic tensions in Yugoslavia.Whereas the ethnic demographics in Croatia were stronglydominated byCroats, with Serbians representing only a smallminority, the ethnic mix in Bosniawas morediverse and so the situation was muchmore problematic. In

27.S.C. Res. 827,UN SCOR,UN Doc.S/ RES/827(1993). 28.Quoted in Cohen 1997, 149. 29.War brokeout ten months earlier inCroatia,coinciding with its claim ofindependence(along with Slovenia)on 25 June 1991. Bosnia was ofŽcially recognized as anindependent nation by the United States andthe European Community on 7 April1992. See Bennett1995; and Silber and Little 1996. Constructingan Atrocities Regime 661

Bosnia,Serbs represented roughly one-third of theprewar populationof 4.3million; theremaining population was composedof Slavic Muslims (43.7 percent), Croats (17.3percent), and a smallminority with mixed heritage. 30 Serbiannationalism, spearheadedby the ascension of Slobodan Milosevic to the position of Serbian president,was metwith apprehension and a revivalof Croatand Slovene separatism thatled, after claims of independence, to con ict with the Serb-dominated Yugo- slavianarmy (JNA). Talesof wartimeatrocities surfaced soon after the outbreak of civil war. Inthe early stages of the war inYugoslavia (1990 – 91),the international communityseemed intent on preserving the territorial integrity of thecountry and was hesitantto becomeentangled in a turbulentregion that had ignited World War I.31 TheBrioni Accords, created with the help of negotiators from theEuropean Community,established a cease-Žre on8July1991; however, violence soon broke outagain. Unrest spread quickly, and although Western nations desired stability, theywere generallyunwilling to intervenemilitarily to establish and maintain order inthe region. One of theŽ rst eventsto promptdecisive international action was the discoveryof atrocities at the Omarska detentioncamp near Prijedor. On 2 August 1992 NewYork Newsday reportedthat Bosnian Muslims held at the camp were beingslaughtered by their Serbian guards. Moreover, subsequent reports likened conditionsin the camp to Nazi concentration camps. 32 Similarconditions were allegedat another camp at Trnopolje. Television coverage worldwide showed strikingimages of men with protruding rib cages, recalling for viewersimages of inmatesfreed from concentrationcamps at the close of World War II. 33 The similaritybetween events in Nazi Germany and contemporary Bosnia served to cultivateclose associations with World War IIandits lessons. Considerations of the “Munichanalogy” necessitated some kind of intervention. 34 Furtherprompting analogies to Nazi-era crimes against humanity was thepro- gramof “ ethniccleansing” being undertaken in Bosnia. Before this program was initiated,the population in the Prijedor municipality of northwestern Bosnia, for example,was 112,470,of which 44 percent were Muslim,42.5 percent were Serbian,5.6 percent were Croat,5.7 percent were ofmixed ethnicity, and 2.2 percentwere “other.”35 ByJune1993, Ž guresreleased by the Serbian media showed thatthe number of Muslims living in Prijedor had declined from 49,454to 6,124; andthe number of Croats from 6,300to 3,169; but the number of Serbs had increasedfrom 47,745to 53,637. 36 Aninternational consensus developed that Serbs were theprincipal instigators of wartime atrocities; however, those who were to

30.Neier 1998,112. 31.Germany’ s early recognitionof Croatia and Slovenia conspicuously went against the European consensusregarding the Balkan con ict. See Crawford1994. 32.Gutman 1993. 33.Neier 1998,135. 34.Ibid., 136 – 37. 35.Ibid., 138. 36.Ibid., 139. 662 InternationalOrganization investigatethe situation would Ž ndit more complex than it appeared at the time. CedricThornberry noted that “ allthree sides were responsiblefor appalling developmentsin Bosnia.The actions of someof the Croats of western Herzogovina rivaledin barbarity those of Serbchieftains of eastern Bosnia, and what was done tothe Muslims of Mostarby Croatswas perhapsas badas theSerb shelling of the mainlyMuslim parts of Sarajevo.”37 Whiledocumented atrocities demanded inter- nationalhumanitarian intervention, the political and strategic complexities involved providedan unappealing scenario for theinternational community. Some observers drew ananalogy between Bosnia and the Vietnam War, andpundits considered the Balkancrisis a conict that presented a “slipperyslope” for allwho dared to involve themselves. Althougha UN-sanctionedarms embargowas leviedon the combatants, direct militaryintervention was eschewed,even amidst striking reports of the growing humanitariancrisis. Europeans generally favored sending ground troops to thearea, butthis policy option was unpopularin theUnited States given the fate of eighteen U.S. soldierswho had been brutally killed during the intervention in Somalia. Policymakerswere leftwith a difŽcult political choice: On the one hand were domesticdemands that the well-publicized atrocities be dealt with. On the other hand,domestic support for interventionplummets as the propensity for casualties increases,as the events in Mogadishu well evidenced. Torn between the ethical desireto promote human rights and the tactical and political challenges of inter- vention,the creation of aUNtribunalrepresented a palatablecompromise. As one analystnoted, “ Itwas awayto do something aboutBosnia that would have no politicalcost domestically.” 38 Usingits authority under Chapter VII ofthe UN Charter,the Security Council passedthe resolution to createthe ICTY for thepurpose of prosecutingfour clusters ofoffenses: (1) Gravebreaches of the 1949 Geneva Conventions (Art. 2),(2) violationsof the laws or customs of war (Art. 3),(3) genocide(Art. 4),and (4) crimesagainst humanity (Art. 5). 39 Thesubstantive and procedural norms crucial to thesuccess of the ICTY arebased on those established at Nuremberg. Yet the challengesfaced by the ICTY areunique; they not only speak to the probability of successin adjudicating war crimesin an ad hoc, case-speciŽ c settingbut also provideuseful insights into the process of establishing a workinginternational criminalcourt whose ultimate goal is deterring atrocities and alleviating tensions thatmay lead to violence. TheŽ rst challengeof theICTY was toestablishguidelines for fairnesswithin its institutionalstructure, considered by internationallawyers to be akeycomponent of itslegitimacy. As oneICTY prosecutorremarked, “ If thetribunal is necessary . .. tobring a senseof justiceto the victims, and thereby undercut the hopeless cycle of

37.Thornberry 1996, 79. 38.Neier 1998,129. 39.ICTY Fact Sheet,16 September 1999, available at ^http://www.un.org/icty/glance/fact.htm &. Constructingan Atrocities Regime 663 revenge,then it is imperativethat everything the tribunal does be fair tothe accused andconducted according to the highest standards of due process.” 40 Hence,there hasbeen a strongpush to make the body truly “ international,”though the in uence oftheUN SecurityCouncil is omnipresent. Judges are nominated and elected by the memberstates of theUN GeneralAssembly, but the list of nomineesmust Ž rst be approvedby the UN SecurityCouncil. 41 Moreover,the chief prosecutor— akey Žgurein theadjudication process— isappointed exclusively by theSecurity Council ontherecommendation of theSecretary General, rather than being nominated by the GeneralAssembly, as is the case for judges.To avoid drawing criticism that the chiefprosecutor merely serves the interests of thepowerful, council members have carefullyselected “ neutral”parties for thisposition. 42 Todate,there have been three chiefprosecutors for theICTY: Richard Goldstone, justice of the South African ConstitutionalCourt; Louise Arbour, justice of the Canadian Court of Appeal in Ontario,and current chief prosecutor Carla Del Pointe, the Swiss attorneygeneral. Whilethe international character of the institution may promote a senseof fairness,it has also presented legal and procedural challenges. Processes for resolvingsubstantive legal issues are further complicated by the appointment of judgeswith differing legal backgrounds. This lack of a uniŽed conceptual frame- workand common operating procedures has presented considerable institutional inertiathat initially affected the functioning of the ICTY. OneICTY prosecutor noted,“ Ina contextwhere there is no particularauthority and where no jurisdiction necessarilycarries more weight than any other, we havefound no better way than toconsider each issue as it comes up, debate it, and try to Ž nda sensible consensus.”43 Needlessto say, such endeavors can result in bureaucratic and organizationalgridlock. Thetribunal’ s legaljurisdiction poses another challenge. According to currently acceptednotions of international humanitarian law, war crimesare limited to situationsof internationalarmed con ict. 44 Moreover,while the ICTY mayprose- cutebreaches of the 1949 Geneva Convention, its jurisdiction is limited to “ grave breaches.”As onelegal analyst noted, “ A‘gravebreach’ can only be committed againsta personprotected by theConvention; that is, only a personof anationality differentfrom thatof theperpetrator.” 45 Therefore,the grave breach clause does not cover,for example,the slaughter or rapeof aBosnianMuslim by a BosnianSerb. Whileinternational legal sovereignty was grantedto Croatia, facilitating adjudica- tionby making the domestic/ internationalline more distinct, less clear are cases involvingKosovo and Rwanda because the con ict was betweenrival ethnic groups

40.Shrag 1995, 194. 41.The roster of judges is diverse,though nationals of the permanent members ofthe Security Councilcomprise nearly30 percent of the presiding judges. 42.However, the selection of Goldstonewas, infact, fraughtwith political maneuvering. See Bass 2000,215– 19. 43.See Schrag1995, 191; and Van Schaack 1999. 44.Morris and Scharf 1995, 391. 45.Scharf and Epps 1996, 651. 664 InternationalOrganization andno such sovereignty has been granted. These crucial issues of jurisdictionwere broughtup by the defense in the case of Dusko Tadic, a former ofŽcial at the Omarska prisoncamp. However, the court ruled that although Article 2 ofthe GenevaConvention applies only to international con icts, Article 3 appliesto war crimeswhether or not combatants are from differentcountries, adding that “ the distinctionbetween interstate wars andcivil wars islosingits value as far ashuman beingsare concerned.” 46 Otherprocedural challenges facing the ICTY concerncollection of evidence.To documentwar crimesand other atrocities the prosecutors at Nuremberg beneŽ ted from themeticulous records kept by the Nazis. 47 Nosuch paper trail was evident inBosnia. Prosecutors must rely primarily on eyewitness accounts and forensic evidenceuncovered by investigators.This is problematicfor severalreasons. First, accessto witnesses is limited where either domestic security has not been estab- lishedor domestic governments are uncooperative. Second, witnesses may fear retaliationand so are wont to stepforward . Thisis compounded by theabsence of comprehensiveprograms to protect witnesses, a problemowing to budget con- straintsof the ICTY. Althoughthe ICTY’ s Victimsand Witnesses Unit protects witnesseswhile they are testifying, no such protection is afforded once they return home,leaving them vulnerable. Moreover, critics have challenged the tribunal’ s practiceof concealing witnesses’ identity during their testimony, charging that it underminesthe due process rights of the accused and the overall fairness of the proceedings—aprimary legal goal of theICTY. 48 Conversely,some witnesses may providefalse testimony in order to further their own aims of retribution, forcing investigatorsto Ž ndcorroborating evidence to support witnesses’ claims. This increasesthe burden of proofon the prosecution, adds to thecost of investigations, andsigniŽ cantly slows the adjudication process. Drawn outadjudication processes aredoubly damaging. First, the sooner a witnesstestiŽ es after an event, the more accurateand credible the testimony; slowing down the process may undermine the qualityof witness testimony prosecutors depend on. Second, prolonging the legal proceedingsseparates the crime from itspunishment, weakening both the deterrent effectof the tribunal and its ability to foster reconciliation among former combat- ants. Themost pressing challenge the ICTY facesis apprehending and detaining defendants.At Nurembergmost surviving instigators of Hitler’ s “Žnalsolution” were apprehendedby the Allies and detained for trial.The ICTY beganwith no defendantsin custody.The problem this presents is clear:“ Thead hoc tribunal for former Yugoslaviahas itself to arrange the capture of those it is to try. For this crucialelement in the procedure it will be totally dependent on the assistance of belligerentand third states.” 49 Thischallenge is further compounded by the

46. Prosecutorv. Tadic ,IT-94-1-AR72,P77 (1995). 47.See Taylor1992; and Conot 1983, 38. 48.See Leigh1996. 49.Fox 1993, 194. Constructingan Atrocities Regime 665 tribunal’s prohibitionon trialsin absentia,a componentof the institutional structure intendedto bolster fairness of theproceedings; however, as TheodorMeron noted, “without in absentia trials,the tribunal is left with few options.The international communityhas given the tribunal strong rhetorical support, but little aid in enforcement.” 50 Consequently,the tribunal initially tried those having little political poweror signiŽcance, since those masterminding wartime atrocities were betterable toelude apprehension. Thus, as one analyst remarked, “ thesecuring of the attendanceof theaccusedwar criminalmay be random,ineffectual, and arbitrary.” 51 Suchobstacles, though initially daunting, have not been insurmountable. As of 1March2001, thirty-Ž ve defendantswere awaitingtrial in the ICTY detentionunit, andtwelve cases had been concluded through the appeals stage. On 26 August1999, Austrianpolice arrested General Momir Talic, a keycommander during the war in Bosnia,who was then ownto The Hague and handed over to the tribunal. 52 In April2000 French authorities arrested Momcilo Krajisnik, president of theBosnian Serbassembly from 1990to 1995. In January 2001 Biljana Plavsic, former deputy toRadovan Karadic, was arrested. 53 Onthe capture of Krajisnik,NATO Secretary- GeneralGeorge Robertson optimistically remarked, “ Thenet is closing in.” 54 The recentarrest of Slobodan Milosevic by Yugoslav police on 1 April2001 and subsequentextradition to The Hague on 28 June 2001 certainly represents a milestonefor thetribunal regime. Milosevic is theŽ rst headof state to facetrial at thetribunal. TheICTY’ s experiencein theBalkans reveals not only the legal and procedural difŽculties in designing a regimeto combat atrocities but also the in uence of powerfulstates during the process of institutionalizati on.While vivid images from Balkanprison camps recalled memories of the Holocaust and engendered public callsfor action,powerful states used the ICTY asa meansto respond to such calls ina politicallyinexpensive way. Moreover, once the international community decidedto establishan adhoc tribunal,the in uence of the UN SecurityCouncil was omnipresentin key aspects of its design, in particular, its jurisdiction and the appointmentof judgesand the chief prosecutor. These same factors are also evident inthe application of theatrocities regime in Rwanda.

Genocide in Rwanda

In1994 the atrocities regime was extendedto Rwanda. 55 Thiscase is instructivefor tworeasons: it illustrates how the interests of thegreat powers affect the process of

50.Meron 1997, 4. 51.Fox 1993, 196. 52. New YorkTimes, 26August1999, A10. 53. See LosAngeles Times, 4April2000, A1, A6; and LosAngelesTimes ,11January2001, A1, A12. 54. Ibid. 55.An excellent historical account of the tragedy in Rwandacan befoundin Des Forges1999. 666 InternationalOrganization regimeformation, and, perhaps more importantly, it demonstrates that negotiating thepolitical terrain between “ hard”and “ soft”law is adynamicprocess involving degreesof institutional learning. Given the scope and magnitude of the atrocities committedin Rwanda and the procedural, bureaucratic, and budgetary obstacles involvedin developing an effective tribunal, this case illustrates the need for institutional exibility.Moreover, because the tribunal in Rwanda followed the precedentset by theICTY, thiscase allows us toassessthe regime’ s broadergoals: deterrenceand national reconciliation. Violenceplagued Rwanda for mostof the late 1980s and early 1990s, and on 6April1994 the plane carrying Juve ´nalHabyarimana, president of Rwanda, and CyprienNtaryamira, president of Burundi, was shotdown over Kigali, Rwanda. EthnicHutus immediately blamed Tutsi rebels of theRwandan Patriotic Front, and withinminutes after the crash soldiers of thepresidential guard began hunting down Tutsisand indiscriminately killing all they encountered. Aid workers estimated that asmany as 500,000Tutsis were killedin themonth after the assassination. 56 More thanthree-quarters of the Tutsi population in Rwanda are estimated to have been killed.57 Anotherestimate suggests that in April,May, and June 1994 more than half ofRwanda’s populationof 7.5million people were eitherkilled or displaced. 58 As was thecase in the early stages of ethniccon ict in theformer Yugoslavia,Western governmentswere reluctantto intervene for fear ofcasualties. Some observers suggestedthat Rwanda simply was notin the strategic interests of Western nations.59 Evenmore caustic were chargesthat the failure to intervene was yet anotherexample of Westernracism. 60 Thebloodshed continued until the Rwandan PatrioticFront secured victory in September 1994. Whilemilitary intervention was notforthcoming after the events of April and May 1994,61 theUN SecurityCouncil created the International Criminal Tribunal for Rwanda(ICTR) on8 November1994. Its jurisdictionis timespeciŽ c— that is, itcovers only the period from 1January1994 to 31 December 1994; its scope is limitedto those events temporally proximate to the assassination of President Habyarimana.To promote consistency between the two ad hoc tribunals— consid- eredcrucial to establishinga clearprecedent and consistent legal norms — Article12 ofthestatute speciŽ es thatthe appeals chamber of theICTY willalso serve as the appealschamber for casesbrought before the ICTR. Moreover, to encourage consistencyin investigationsand prosecutorial strategy Article 15 speciŽes thatthe chiefprosecutor of the ICTY willalso serve as the chief prosecutor of theICTR. Thatthe basic structure of the ICTY was implementedin another atrocities scenariospeaks to itssuccess in being perceived as an appropriate policy option in

56. Time, 16May 1994, 57. 57.Kuperman 2000, 101. 58. Time, 13June 1994, 36. 59.Lawyers’ Committee forHuman Rights 1997. 60. Time, 6June1994, 34. 61.See Des Forges1999; and Kuperman 2000. Constructingan Atrocities Regime 667 caseswhere massive human rights violations have occurred. By 22 February 2001 forty-foursuspects were beingheld at the UN detentionfacility in Arusha, Tanzania.62 TheICTR was initiallymore successful than the ICTY indetaining high-proŽle defendants, including former militarycommanders and political lead- ers,whereas those detained in The Hague have tended to be low-level Ž gures withoutcommand responsibility. 63 Yetthe ICTR faces many of thesame political andprocedural challenges as the ICTY. Althoughit canbe arguedthat the war inBosnia and Croatia was aninternational conict stemming from internationallegal recognition granted to the separatist republics,this was clearlynot the case in Rwanda. However, by rulingthat Article 3ofthe Geneva Convention applies to both interstate and intrastate con ict, the ICTY openedthe door to internationaladjudication of internalcon icts, such as that in Rwanda.64 Thenormative importance of thisprecedent cannot be overstated,for itclearly expands the jurisdiction of the tribunal and applies international law to issuesthat traditionally have deferred to national sovereignty. 65 Whilethis prece- dentcertainly aids the ICTR in trying suspected war criminalsin Rwanda, this expansionof jurisdiction may become a signiŽcant obstacle to a workinginterna- tionalcriminal court, since powerful states have expressed concern about an internationalcourt that seeks to expand its authority. Thelimited temporal jurisdiction applied to the ICTR is alsoa pointof contention thatinitially threatened cooperation between the tribunal and the Rwandan govern- ment.In fact the Rwandan government opposed the establishment of thetribunal as articulatedin the Security Council’ s resolution,even though it initially solicited SecurityCouncil action. 66 TheRwandan ambassador to the United States explained, “thegovernment of Rwanda regarded the dates set for the rationetemporis competenceof the international tribunal for Rwanda. ..asinadequate. The genocidewhich the world witnessed in April 1994 had been the result of a long periodof planning during which pilot projects for exterminationhad been success- fullytested before this date.” 67 Reportsof massacres and ethnic violence taking placein 1991– 93 were documentedby several agencies, including the Special Rapporteurof the UN (May1993). Because of this, the Rwandan government proposedthat the tribunal’ s jurisdictionbe extended back to 1 October1990, a proposalultimately rejected by the Security Council. While the Security Council’ s decisionclearly helps to expedite the adjudication process by limiting its investi- gation,Rwandan representatives have countered that this will severely curtail its abilityto achievedomestic reconciliation: “ Aninternational tribunal which refused toconsider the causes of thegenocide . ..cannotbe ofany use to Rwandabecause

62. ^http://www.ictr.org &. 63.LCHR 1997,2. 64. Prosecutorv. Tadic ,IT-94-1-AR72(1995). 65.Meron 1995. 66. New YorkTimes, 29December 1994,A1. 67.Bakuramutsa 1995, 645. 668 InternationalOrganization itwill not contribute to eradicating the culture of impunity or creating a climate conduciveto national reconciliation.” 68 Here we seeacute political tension between theneed for expediencyin the adjudication process and the need for domestic cooperationand holistic efforts todeal with the causes of the con ict. Anotherissue related to jurisdiction is theinteraction between the tribunal and the nationalcourts. While the intent of thetribunal statute was for internationallaw to workin concertwith the national courts when dealing with such widespread human rightsviolations as those in Rwanda, it also intended the tribunal to take a more activerole when either the national legal system was inadequateor unjust rulings were likely.Rwanda began its own genocide trials in the national courts in December1996; many analysts predicted, however, that “ thesheer magnitude of the genocideand violence of 1994 would overwhelm the system.” 69 Whileearly trials were criticizedon several counts— accused had not been granted legal representa- tion,suspects had been tortured during interrogation, requests for defensewitnesses hadbeen denied— more recent trials show improvement in due process and legal safeguardsfor defendants.One analysis conŽ rmed that “ evenwith enormous logisticalweaknesses and shortages of trainedpersonnel, Rwanda can provide trials thatmeet or approach minimum guarantees of fairness and due process.” 70 Yet exactlyhow the tribunal and the domestic courts may share a “mixedjurisdiction” remainsunclear. Accordingto the tribunal statute, the ICTR’ s jurisdictionhas primacy over nationalcourts, and it may request national courts to defer to it at any stage of ongoingproceedings. 71 Clearly,for suchtransfers to take place, cooperation with stateauthorities is imperative. In addition Article 9 ofthe statute conforms to the principleof nonbis in idem. 72 Thesetwo principles are clearly at odds when nationalcourt proceedings are underway or havebeen completed. In cases where an ongoingnational trial is not impartialor independent,jurisdiction is to betransferred tothe ICTR; however, the ICTR’ s rulesof procedure and evidence offer noclear guidelinesfor doingthis, nor do they specify who is to make such decisions. Moreover,the primacy of the ICTR’ s jurisdictionover that of the national courts alsopays little heed to the cultural elements of local legal norms, an element that maybe crucial to the tribunal’ s goalof achieving national reconciliation and alleviatingethnic tensions. The ICTR is authorized to imposea maximumsentence oflifeimprisonment, whereas the Rwandan national courts may impose the death penaltyfor thosefound guilty of capitalcrimes. Rwandan diplomats have expressed thecommon belief that those tried by the tribunal “ wouldget off morelightly than

68.Ibid., 646. 69.LCHR 1997,23. 70.Ibid., 29. 71.ICTR statute, Art. 8(2). Similar jurisdictional primacy is codiŽed in Art. 9. 72. Nonbis in idem refers toprohibitions against trying defendants twice forthe same crime(s), often referred toas “doublejeopardy.” Constructingan Atrocities Regime 669 ordinaryRwandans who faced the death penalty in local courts.” 73 Theprovisions prohibitingdouble jeopardy leave the national courts no recourse when the tribu- nal’s decisionsare seen as unjust. According to Rwandan legal sensibilities, the ICTRdoes not offer anadequate range of sentencing options to distinguishtop-level plannersfrom thosewho carried out the plans. Because it ispossiblethat those who devisedand organized the genocide may escape capital punishment (if triedby the tribunal)but those who simply carried out the orders may not (if triedby domestic courts),such incongruities may not be conducive to national reconciliation in Rwanda.74 Thisperceived incongruity was alsocited by the Rwandan government asa reasonthey could not support the tribunal; instead, Rwanda established the OrganicLaw onthe Organization of Prosecutions for Offenses Constitutingthe Crimeof Genocide or Crimes Against Humanity Committed Since October 1, 1990.75 Thesenew national laws, to be adjudicated in the national courts, classify suspectsinto four categories according to degree of culpability— leaders and organizersare subject to the death penalty, whereas those accused of lesser crimes maybe eligiblefor reducedpenalties in exchangefor acompleteconfession, a guilty plea,and an apology to victims. 76 Otherprocedural and bureaucratic problems impede the ICTR’ s likelihoodof success.As isthe case with the ICTY, intheabsence of materialevidence tribunal investigatorsmust rely primarily on witness testimony. This presents several prescientchallenges: (1) obtainingtestimony amid mass refugee ows,(2) witness protection,(3) validityof witness testimony and the ICTR’ s attemptat “ ethnic neutrality,”and (4) theslow pace of proceedings. Findingwitnesses that were apartof themass exodusfrom Rwandawill not only betimeconsuming for investigatorsbut also costly. The need for witnessprotection isat odds with the tribunal’ s goalof maintainingfair trialproceedings. Pursuant to Rule69 of theICTR statute,the prosecution may keep conŽ dential the identity of victimsor witnesses who may be at risk of retaliation; however, once under the protectionof thetribunal, their identities must be disclosedto thedefense. Given the meagerresources available for aneffective witness protection program, the “ pro- tectionof the tribunal” may offer littleassurance to witnesses reluctant to testify. During1996, the UN HumanRights Field Operation in Rwanda documented more thansixty-four separate incidents in which genocide survivors and their family and friendswere killed(227 people) or were injured(56 people) as a consequenceof theirtestimony before the tribunal. The report also contains numerous accounts of threatsand intimidation directed at potential witnesses and their associates. Estab- lishingthe validity of witnesstestimony is especially challenging in casesof ethnic hatecrimes. The tribunal is committed to impartialityin thebelief that the divisive

73.Bakuramutsa 1995, 648. 74.Bakuramutsa 1995. See also New YorkTimes, 2November1994; and InternationalHerald- Tribune, 9November1994. 75.Available online at ^http://www.rwandemb.org/prosecution/law.htm &. 76.As speciŽed in Art. 2, 5,and14 –16. 670 InternationalOrganization issueof ethnicitymay be transcended “ whenconfronted with the irrefutable truths attrial by judgeswho share none of thesebiases.” 77 Suchtactics did not garner the desiredeffects when applied in the Balkans, where reactions to tribunal decisions remainstrongly divided along ethnic lines. 78 Adearthof material evidence and a strongreliance on witnesstestimony that may or may not be tainted by ethnicbiases couldactually spur ethnonationalist tensions and violence; one ethnic group may feelthat the evidence used to convict their kinsmen was fraudulentand that the tribunalwas usedby the enemy to furtherpersecute their ethnic group, aggravating thegroup’ s senseof threat.Finally, proceedings moved at a snail’s pacebecause of thechallenge of investigating cases dependent on witness testimony (especially whenwitnesses are geographically dispersed) and bureaucratic mismanagement by theICTR that led to forensic investigations at mass gravesbeing thwarted due to lackof funds and necessary equipment. 79 If preservingan accurate collective memoryfacilitates not only accurate witness testimony but also the process of nationalreconciliation, the pace of investigations and proceedings is crucial. Furtherhindering the ICTR’ s abilityto foster national reconciliation is the tribunal’s lackof relevanceto the Rwandan population. While the statute identiŽ es neutralityand independence as institutional imperatives— largely because Security Councilmembers believed the tribunal’ s neutralitywas essentialfor reconcilia- tion—neutrality may in factwork against reconciliation. “ Thestructural distance of theICTR from theRwandan social process makes it very difŽ cult for theICTR’ s workto berelevantand even more unlikely that its work will address the root causes ofthegenocide.” 80 This“ socialdistance” takes place at several levels. The tribunal convenesin Arusha, Tanzania, not in Rwanda. This location, though chosen to promoteneutrality, instead separates the proceedings from thepeople they were intendedto help. Moreover, “ thereis a disconnectionbetween the ICTR trials and theinternal social process. Not only the physical distance, but the way in whichthe ICTRhas operated and publicized its efforts doesnot involve the population in any realsense.” 81 Whilethe government-operated radio station provides limited cover- ageof trialproceedings, there are no television broadcasts outside the capital city, andfew Rwandansunderstand the legal procedures and proceedings. From theoutset, the relationship between the largely Tutsi government of Rwandaand the ICTR has been, in the words of one analyst, “ frosty.”82 While simplelogistics give the ICTR a strongincentive to limit the duration of its legal jurisdiction—in August 1999 Rwandan detention facilities held over 124,000 prisonersawaiting legal procedures 83—thislimitation may profoundly affect the tribunal’s successin establishing reconciliation among the Rwandan population.

77.Alvarez 1999,437. 78.Ibid., 437. 79. TheBoston Globe, 11December 1996,A1. 80.Howland and Calathes 1998,161. 81.Ibid., 155. 82. New YorkTimes, 21November1997, A10. 83. Deutsche Presse-Agentur, 13August 1999. Constructingan Atrocities Regime 671

Otheranalysts counter that other forces are at play: “ Those[temporal] limits were theproduct of a highlypolitical process within the Security Council and re ect diplomaticconcerns. Broader jurisdiction for theICTR could well have led to inquiriesthat would have embarrassed either the UN asa wholeor particular permanentmembers of the Security Council.” 84 Yetdealing with a war crimes scenarioas vast as that encountered by the ICTR often poses a dilemma:Limiting thescope of theinvestigation and trials may impede justice by notholding all of the guiltyaccountable for theiractions and reduce the tribunal’ s successin achieving reconciliationin Rwanda(and elsewhere); however, a moreexpansive role burdens analready over-extended institution and may signiŽ cantly affect its ability to quicklyresolve cases. Recently, the slow pace of legal proceedings in the case of Jean-BoscoBarayagwiza, former Rwandandirector of PoliticalAffairs andminister ofForeign Affairs, resultedin the ICTR’ s decisionto release him because he had beenheld too long without trial. 85 Inresponse the Rwandan government immedi- atelyrefused to further cooperate with the tribunal, stating that “ Barayagwiza’s releaseserves as a pretextfor allother perpetrators of the genocide to live with impunityall over the world.” 86 Withoutcooperation of the Rwandan government, thetribunal’ s abilityto conducttrials will certainly be curtailed.How to“soften”the legalizationof the atrocities regime to promote effective “ mixedjurisdiction” and cooperationremains an important challenge to institutional design.

The ICTYin Kosovo

Furtherapplication of the tribunal system became necessary in 1999 as ethno- nationalistwarfare brokeout between ethnic Albanian nationalist forces and the Serbianarmy. While initial casualties were lightby international standards, num- beringsome 2,500, accusations of renewed “ ethniccleansing” by Serbian forces surfacedafter the failure of theRambouillet talks and subsequent NATO airstrikes. Reportsof mass graves,torture, rape, and executions of ethnicAlbanians poured out ofKosovo as quickly as the thousands of refugees who left their homeland under duress;calls for war crimesinvestigations were nearlyconcurrent with NATO action.On 29 September1999 it was announcedthat the ICTY’ s jurisdictionunder itsoriginal statute would be extended to Kosovo. Like the case of Rwanda, this case shedslight on whether the tribunal’ s actionsin Bosniahad any effect on deterrence andnational reconciliation. It not only addresses a conict that occurred after a tribunalaction elsewhere but also allows us toassess whether fear ofadjudication affectsthe decisions of political and military leaders. In this case many of those accusedof atrocities had already been named as perpetratorsin the Bosnian con ict.

84.Alvarez 1999,397. 85. LosAngeles Times, 7November1999, A12. The decision to release Barayagwiza was later reversed inresponse to public outrage. See LosAngeles Times, 1April2000, A11. 86. LosAngeles Times, 7November1999, A12. 672 InternationalOrganization

There-application of the atrocities regime to the volatile situation in the Balkans alsobrings to the surface the public’ s perceptionof tribunal action, that is, whether decollectivizationof guilt can promote national reconciliation and peace. On both accounts,the case of ICTY actionin Kosovo is not encouraging. Togain “ institutionalmomentum” during the Bosnia investigations, the ICTY activelypursued investigations against defendants at all levels of culpability.Most ofthe defendants and detainees in the Bosnia trials were atthe lower rungs of commandand control, yet they were consideredimportant for establishingproce- duralnorms and precedent. Functional considerations have prompted the tribunal to pursueexceptionalism, focusing investigations on successful prosecution of the signiŽcant players. One court ofŽ cial noted, “ As far asI’ mconcerned,[the tribunal] simplycan’ t tryevery Tom, Dick, and Harry.” 87 Thetribunal prosecutor added, “ It isclear that the OTP [OfŽce of theProsecutor] ICTY hasneither the mandate, nor theresources, to functionas the primary investigative and prosecutorial agency for allcriminal acts committed on the territory of Kosovo.” 88 Whilethere are tactical beneŽts to prosecuting low-level perpetrators, most analysts have stressed that the ICTY’s deterrencevalue hinges ultimately on its ability to successfully prosecute thoseat the highest levels. 89 TheKosovo case is also very useful in analyzing the tribunal’ s abilityto shape stateaction, since the con ict in Kosovo followed two tribunals that successfully triedwar crimescases. Clearly, the evidence emerging from Kosovo—mass graves, witnessaccounts of summary executions of civilians, torture— suggest that Serbs were clearlyundeterred by the presence of the ICTY. Thiswas documentedby CedricThornberry, who in hisdealings with those involved in war crimesatrocities observedthat

Our interlocutorsplainly were skepticalthat the “ internationalcommunity” woulddo anything. In Belgrade and Zagreb, they usually preserved the diplomaticniceties and kept straight faces, but often the sneer around the table was nearlyaudible. In less sophisticated circles, when we spokedirectly with thosewe knewhad been the instigators and warned them that justice would someday come, the local establishment and its forces of lawand order often snickeredaloud. 90

Similarattitudes were alsoevident during hearings on Rule 61 held in The Hague inJuly 1996 (discussed later); a witnesstestiŽ ed that Ratko Mladic scoffed openly atNATO’ s inabilityto protectthe Muslims in Srebenicain July1995, an eventthat occurredtwo years after the ICTY was created. 91 Clearly,the desired effect of

87. LosAngeles Times, 27August 1999, A5. 88.Statement byCarla delPonte, 29 September 1999. Available online at ^http://www.un.org/icty/ pressreal/p437-e.htm &. 89.See Scharf1997, 219, 225; Alvarez 1999;and Morris 1997. 90.Thornberry 1996, 77. 91.Guest 1996,80. Constructingan Atrocities Regime 673 adjudication,to deterwar crimes,has been signiŽ cantly hampered by thedifŽ culty ofarrestingsuspects, especially during the tribunal’ s earlyperiod. Even prosecutors inThe Hague agree that “ theonly true deterrents . ..arenot investigations but arrests.”92 Yetmembers of theinternational community seem to havelittle desire to takethe tactical risks involved in apprehendingthe high-level perpetrators currently indicted.93 After theDayton Accords brought the con ict in Bosnia to a close,the NATO implementationforce (IFOR) was givenextremely cautious instructions for dealingwith indicted war criminals. 94 Initialoperating procedures authorized IFOR troopsto arrestthose they encountered but did not permit IFOR toseekthem out. 95 Rumorssurfaced that because of this “ handsoff” policy, indicted war criminal RadovanKaradzic was ableto pass through several NATO checkpointsunmolested, evenafter the Dayton Accords were signed,though these rumors were never substantiated. 96 After NATO bombedSerbia during the con ict in Kosovo, the UnitedStates opposed any operation to seizeSlobodan Milosevic, opting instead to offer a$5million reward for hisarrest. 97 PresidentClinton remarked, “ Idonot believethat the NATO alliescan invade Belgrade to try to deliver the indict- ment.. ..Our heaviestresponsibility . ..isto get the Kosovars back home in safety.”98 Policymakerswere clearlysensitive to “ missioncreep,” a painfullesson from theVietnam era. Inan effortto increasethe deterrence value of the tribunal, given the difŽ culties involvedin arresting indicted war criminals,the ICTY establishedRule 61 (under Art. 15ofthe statute), which provides for a“super-indictment”in certain instances. Thepurpose of this rule is to broaden world awareness of perpetrators’ actions withoutviolating the mandate forbidding trials in absentia. It allows the indictment andall supporting evidence to be submittedto the tribunal in an opencourt session. Thismay include examination of witnesses whose testimony becomes part of the record.Under the provisions of Rule61, the prosecution may present highlights of thecase in the absence of the accused, essentially for themedia. 99 Whilethe line betweenexecuting Rule 61 and prohibiting absentia trials is rather thin, the aim of thesuper-indictment is unmistakable: to ensure that those indicted will be consid- eredinternational pariahs, even if they manage to elude arrest. 100 Some legal

92. New YorkTimes, 15September1999, A3. 93.A detailedplan to capture Radovan Karadzic, called “OperationAmber Star,”was completedin April1997 and involved several hundredFrench and U.S. commandos. When advised of the plan, PresidentClinton wanted French forces tospearhead theraid; however, Jacques Chirac was reluctantto assume sucha “highrisk” position for fear ofreprisals againstFrench troops in the region. Likewise, Clintonand British Prime MinisterTony Blair were reluctantbecause ofconcerns over potential casualties, andso the plan was neverexecuted. Time, 10August1998, 68 –70. 94.IFOR was later renamed “StabilizationForce” (SFOR). 95.See Meron1997, 5; and Bass 2000. 96. Time, 10August 1998, 68 –69. 97. New YorkTimes, 26June1999, A6. 98. SanDiego Union Tribune, 18June 1999, A21. 99.Thornberry 1996, 83. 100.Scharf and Epps 1996, 649. 674 InternationalOrganization analystsare not optimistic: “ Thecontribution of publicizing evil acts, although worthy,alone falls far shortof meetinga comprehensivepunishment objective for acourtsystem. An increasein internationalawareness . ..probablydoes not equate togenerally deterring others.” 101 Ratherthan establishing closure through justice, thesemeasures seem to be stop-gapattempts to providesome sense of “justice”to victimsof war crimesuntil the guilty parties can be brought to trial. The question is,will such stop-gap measures provide the necessary deterrence and reconciliation tomitigate future transgressions of the jus belli,especiallyas time between transgressionand adjudication becomes ever greater? Althoughthe slow pace of proceedings is understandable given the legal and logisticalhurdles facing the ICTY, itmay hinder both deterrence and national reconciliation. 102 Ina recentnews report, interviews revealed that many Serbs are avoidingresponsibility for theethnic hatred that drove the program of ethnic cleansing,and many deny that atrocities, such as those committed at Srebenica, ever reallyoccurred. 103 Inan opinionpoll published in June 1999 by thenewsmagazine Nin, almosttwo-thirds of Serbs do not believe that the atrocities alleged in the tribunalproceedings occurred; instead they “ emphasizethe high price that Serbs are nowpaying.” 104 Thissense of “reversal”was wellarticulated by a Serblawyer: “ I didn’t killanyone, but an Albanian neighbor told me I wouldnever be safe in Kosovo.I ama victimof their ethnic cleansing.” 105 Othersconsidered tribunal reportsas nothing less than anti-Serb propaganda. Ethnic Albanians seem particu- larlysensitive to what they perceive as awhitewashingby theSerbian government. PajazitNushi, member of theCouncil for Defenseof HumanRights and Freedoms inPristina, notes, “ Still,now, there is no single Serbian political voice that has condemnedthe crimes.” 106 Moreover,the withdrawal of Serbian troops from Kosovohas been accompanied by acts of violent retribution by ethnic Albanians. Onenews account noted, “ Inthe early days of NATO occupation,many Serbs who stayed[in Kosovo] were optimisticthat they could forge a futurewith their Albanian neighbors.But a waveof retaliatory killings of Serbs by Albanians enraged by wartimeatrocities has calciŽ ed emotions.” 107 Timeis certainlynot assisting efforts tocreate a peaceful,multiethnic Kosovo, as new justiŽ cations for animosity betweenethnic groups are kindled and old hatreds reinforced. Clearly,the deterrence value of theemergent regime has been, to thispoint, quite weak,owing largely to the reluctance of the international community to aggres-

101.Howland and Calathes 1998,150. 102.One tribunaljudge remarked thatif Milosevic were turnedover to the ICTY fortrial, it would bethree years beforehis case wouldŽ ndaplace onthedocket. See LosAngeles Times, 27August1999, A5.In a June2000 report to the UN SecurityCouncil, it was estimated thatit wouldtake sixteenyears tocomplete theICTY’ s currentcaseload. See LosAngeles Times, 6July2000, A10. 103. LosAngeles Times, 6July2000, A1, A10. 104. LosAngeles Times, 2July1999, A1. 105. Ibid. 106. LosAngeles Times, 10October1999, A1. 107.Ibid., A30. Constructingan Atrocities Regime 675 sivelypursue high-level perpetrators; however, the arrest of Milosevic and the possibilityof hisextradition for trialat theHague tribunal leaves considerable room for optimismthat the regime’ s deterrencepower may dramatically increase. The caseof ICTY actionin Kosovoalso illustrates the limitations of theatrocities regime inpromoting national reconciliation in ethnically torn states. It remains to be seen whetherthe arrest of Milosevic will serve to disclose the truth of events that occurredduring the Balkan con ict and promote national healing, or whether his arrestand extradition in response to Western pressure will further calcify animos- itiesbetween ethnic groups in the region. The ability of the ICTY toobtain Milosevic’s extraditionis a crucialpoint in the development of a moreviable atrocitiesregime.

Justice in SoutheastAsia?

Thatthe ICTY hasnot only survived but has served as a modelfor otherad hoc tribunals,including a permanentinternational criminal court, could indicate that war crimesadjudication is a successfulpolicy tool. However, although the regime has overcomeconsiderable procedural and structural obstacles in the Balkans and Rwanda,these obstacles remain formidable in othercases. In regions dominated by powerpolitics, regime/ normdevelopment remains in the formative stage, especially insituations where powerful states have strong incentives not tobecomeinvolved. Withoutthe direct intervention by strongstates and cooperation by governments in stateswhere atrocities are alleged to have occurred, the atrocities regime lacks strength.

Cambodia Ithas been estimated that more than a millionCambodians died from execution, torture,disease, or hungerfrom 1975to 1979 under the Khmer Rougeregime; some estimatesgo ashighas 2million.Although it isunclear why a war crimestribunal was notestablished earlier in the wake of such a profoundhuman tragedy, the institutionalmomentum of the atrocities regime has prompted the UN toseek to establisha judicialmechanism for Cambodia.The failure to establish a tribunal earliercan be attributed to the interests of severalSecurity Council member states andto the recalcitrance of the current Cambodian government. At thetime atrocities were committeda tribunalwas notin the strategic interests oftheUnited States; in theaftermath of theVietnam War therewas littleincentive onceagain to becomeentangled in SoutheastAsia’ s politicalquagmire. Moreover, inadjudicating charges of war crimes,information about U.S. secretbombings of Cambodiaand other sensitive information could become part of thepublic record. WilliamDowell, UN correspondentfor Time,suggeststhat many countries, includ- ingthe United States, “ haveused the Khmer Rougeto pursue their own political interestsin the region at onetime or another,and all are reluctant to talkabout their 676 InternationalOrganization relationshipwith the Khmer Rouge.”108 Thisfear maybe particularly acute for China,already dealing with image problems that complicate its bid for membership inthe World Trade Organization. Given the current political climate, Beijing is understandablyhesitant to have its role in supporting the Khmer Rougeregime exposedto theinternational community it wishesto engage. 109 Whilesuch reasons maydiscourage powerful governments from becominginvolved, public demands for actionin Cambodia have also been less acute than was thecase for theBalkans orRwanda.In the United States public desire for justiceand accountability has been temperedby anequallycompelling desire to “ closethe book” on theVietnam era, reducingdomestic demands for stateaction. Domesticresistance is alsoan importantfactor in Cambodia.Initial UN attempts toestablish a tribunalfor Cambodiawere metwith little cooperation from the Cambodiangovernment, especially Prime MinisterHun Sen. The UN hasproposed severalpossible tribunal conŽ gurations, all of which display institutional adjust- mentsstemming from thelessons learned in the Balkans and Rwanda. First, the UN wishesto try in a singletrial only twelve former politicaland military leaders of the Khmer Rouge,thereby avoiding the protracted proceedings that plague other ad hoc tribunalscurrently in operation;however, the Cambodian government has expressed littleinterest. “ We haveno conŽ dence in aninternational court of law,”noted Hun Sen,showing concern that a trialmay upset his fragile hold on power in Cambo- dia.110 HunSen has been concerned that the scope of criminalculpability may make reconciliationthrough justice problematic in Cambodia. As oneobserver remarked, “justiceitself seems a rustychain that will only bloody anyone who tries to touch it.To try Khmer Rougechieftains would be, in a sense,to prosecute the whole country.”111 Attemptsby tribunaladvocates to bypassthe Cambodian government by extend- ingthe jurisdiction of existingad hoc tribunals have been vetoed by China,forcing UNofŽcials to consider concessions in tribunaldesign in order to garnerCambodian governmentsupport. 112 Sinceits Ž rst proposals,the UN modiŽed its rigid stance on “internationalimpartiality” by suggesting an institutional structure that allows Cambodianrepresentation among the justices. 113 TheUN wouldselect an indepen- dentprosecutor and international judges, and the Cambodian government could selectnational judges, who would be in the minority, to serveon thebench. Again, theCambodians were notresponsive. The UN hassince agreed to allow majority representationof Cambodian judges should a tribunalbe established. 114 This “mixedtribunal” re ects both logistical interests and an increased sensitivity to connectinglegal proceedings with the Cambodian people. On 2 January2001, after

108. Time, 22January 1999 . 109. SouthChina Morning Post, 25August 1999, 14. 110. Time, 22March 1999, 56. 111. Time, 16August 1999 . 112. New YorkTimes, 18April 1998, A5. See also BostonGlobe, 1May1998, A10. 113. New YorkTimes, 12August 1999, A8. 114. LosAngeles Times, 3January2001, A1, A11. Constructingan Atrocities Regime 677 considerableinternational pressure, Cambodia’ s NationalAssembly approved the newUN guidelines;however, signiŽ cant political obstacles remain, including approvalby the Cambodian Senate, legal determination of the guidelines’ consti- tutionality,and King Norodom Sihanouk’ s signature. Thecase of Cambodia also illustrates the problem time poses when relying on adjudicationto promote peace and reconciliation. Although there is no statute of limitationson tribunal indictments, human rights groups argue that because of the advancedage and poor health of manysuspects, quick action to createa tribunalis imperativelest Cambodia lose its chance to bring Khmer Rougeleaders to jus- tice.115

Indonesiaand East Timor Inresponse to a successfulreferendum in September 1999 declaring East Timor’ s independencefrom Indonesia,pro-Indonesia militias mounted a campaignof vio- lenceand intimidation throughout East Timor. The acts of violence were inmany casesgruesome. A Reutersnews team documented victims found in a truckwho appearedto have been tied up, hacked with machetes, and set ablaze after being soakedwith gasoline. 116 Eyewitnessesclaimed that Indonesian police had joined the militiamenin thisatrocity. The UN HighCommissioner for HumanRights declared, “therewas overwhelmingevidence that there had been a deliberate,vicious, and systematiccampaign of gross violations of human rights in East Timor, including mass killings,forcible expulsions, violence against women, and a breakdownof law andorder. The extreme violence . ..was initiatedby different militia groups, in whichelements of security forces were alsoinvolved.” 117 Inlight of evidence of humanrights abuses, the UN Commissionfor HumanRights (UNCHR) openeda specialsession that resulted in a resolutioncalling for apreliminaryinvestigation intowar crimesin East Timor, seen by manyas the Ž rst steptoward establishing a war crimestribunal. 118 Theresolution speciŽ cally refers toSecurity Council Resolution1264, in which the Council “ demandedthat those responsible for such actsbe brought to justice.” 119 However,the government of Indonesia quickly rejectedthe UNCHR resolution,a movethat denied UN investigatorsaccess to Jakarta’s militaryŽ les.During the special session of the UNCHR, theIndonesian representativedismissed the need for internationalintervention: “ TheGovernment lastnight had established a fact-Žnding commission to compile information on human-rightsviolations and bring the perpetrators to justice. It was importantto ensurethat this august body not do anything that would open old wounds and

115. New YorkTimes, 12August 1999, A8. 116. TheIndependent (London),30 September 1999, 15. 117.UNCHR press release HR/CN/99/67,23 September 1999. 118.UNCHR Res. 1999/S-4/1. 119.S.C. Res. 1264,UN Doc.S/ RES/1264(1999). See alsoS.C. Res. 1272,Art. 16, UN Doc. S/RES/1272(1999), available online at ^http://www.un.org/Docs/scres/1999/99sc1272.htm &. 678 InternationalOrganization exacerbateproblems in the territory.” 120 Indeed,the Indonesian government’ s lack ofcooperationmakes the creation of a tribunalquite unlikely. Thattribunals were notestablished in Cambodia and Indonesia re ects two weaknessesin relying on international law to provide peace and reconciliation in war-tornregions: the need for cooperation,both internationally and in war-torn regions,and the hesitancy of the international community to intervene militarily. Whilead hoc tribunals may be formedby Ž atof theSecurityCouncil, the difŽ culties encounteredby the ICTY showhow lack of cooperation may sti e institutional effectivenessand regime development. Proponents of aninternationalcriminal court pointto Cambodiaand East Timor, where the atrocities regime appears beholden to theinterests of the powerful, as evidence that such a permanentinstitution is necessaryif a trulyeffective regime is to be established.

The International Criminal Court

KoŽAnnan, speaking at the Rome conference to establish an International Criminal Court(ICC), declared,“ Victimsof pastcrimes and potential victims are watching us.They will not forgive us if we fail.”121 Whilethe ICC is not a speciŽc caseof theapplication of a legalregime to an instance of genocide or crimes against humanity,examining its development is crucial to understanding the political challengesof expanding the existing ad hoc tribunal system to a moreuniversal atrocitiesregime. This case illustrates the tension between the need for greatpower supportand the desire to establish a hardlaw regime that transcends power and politicalinterests (that is, holds strong and weak states equally accountable). The difŽculty in achieving great power support for theICC was anticipatedby some legalanalysts: “ If apermanentinternational criminal court were created,it is extremelyunlikely, at this juncture, that any of the more powerful nations would allowit mandatory jurisdiction or would opt to submitto itsjurisdiction.” 122 The ad hocsystem employed in the existing atrocities regime is appealing to powerfulstates becauseit facilitates adjudication, yet control over its application in a givencase remainswith the Security Council. It is understandable, therefore, that powerful statesare reluctant to delegate authority to an independent body. While the statute tocreate the ICC is an established fact, its power as part of the atrocities regime remainscontested and indeŽ nite, and its development is marked by concessions madeto great power interests. This case suggests that if theatrocities regime is to gainwidespread acceptance, the process of legalization will likely undergo “ soft- ening”in order to mitigate the political contracting costs of the new regime. As notedby Kenneth W. Abbottand Duncan Snidal, hardening the legal foundations of

120.UN press release, HR/CN/99/67,23 September 1999, 6. 121.Quoted in Time International, 27July 1998, 46. 122.Scharf and Epps 1996, 639. Constructingan Atrocities Regime 679 theatrocities regime is a sensitiveand protracted process that may involve initially takingsofter positions. 123 AlthoughPresident Clinton signed the Rome Statute on 31 December 1999 that createdthe ICC, theUnited States has long opposed several key components of the RomeStatute, opposition still expressed by the Bush administration. 124 The Ž rst involvesthe universal jurisdiction provisions as articulatedin the statute that subject anystate, signatory to the statute or not, to the court’ s jurisdiction. 125 While agreeingthat such expansive jurisdictional provisions are reasonable in cases of genocide,U.S. representativessought a clauseto allowa stateto optout oftheICC’s jurisdictionfor aperiodof ten years in cases of war crimesand crimes against humanity.In compromise, the Rome Statute now allows states to optout for seven yearsbut only in cases of allegedwar crimes. 126 TheUnited States was alsoconcerned that the scope of crimescovered under the court’s jurisdictionwas overlybroad. “ Crimesof aggression,” for example,is included,though no precise deŽ nition of “ aggression”was agreedon during the draftingof thestatute. James Rubin, U.S. StateDepartment representative, lamented thedrafters’ unwillingness to address reservations about the treaty and expressed concernthat use of nuclear weapons may be includedunder the current deŽ nition of crimesof aggression. 127 Anotherconcern was theprosecutor’ s authorityto investigate crimes even in caseswhere no state party had issued a complaint.Under Articles 13 and 15, the prosecutormay investigate crimes propriomotu basedon information provided by partieswithin the court’ s jurisdiction. 128 U.S. negotiatorswanted to limitthe power tobring cases to thecourt to the Security Council, consistent with the precedent set bythe ad hoc tribunals. Without this limitation, U.S. negotiatorsargued, members ofthe U.S. armedforces “ wouldbe subject to frivolous, politically motivated charges”that may hinder crucial peacekeeping missions in thefuture if there was a possibilityof “ maliciousprosecution.” 129 Inyet another attempt at political com- promise(“ softening” the process of legalization), Article 16 affords the Security Councilthe power to halt investigations and prosecutions for twelvemonths, and thismoratorium may be renewed with the same conditions. Some delegates protested,arguing that “ ifstatescan simply opt in or outwhen they want, the court willbe unworkable.” 130 Yetthis provision does not give single members of the SecurityCouncil veto power; instead, prosecution may be deferred only by unan-

123.Abbott and Snidal 2000. 124. LosAngeles Times ,15February2001, A4. 125.Rome Statute,Article 4(2). 126.Rome Statute,Article 124. 127.U.S. Department ofState,Daily Press BrieŽng, 20 July 1998. 128.Rome Statute,Article 13(c);15(1) 129.David 1999, 357. 130. Time, 27July 1998, 46. 680 InternationalOrganization imousvote of theSecurity Council. And if deferral is to be extendedfor additional periods,the political consensus favoring delay must be maintained. 131 Finally,the Clinton administration insisted on an exception for personnelin- volvedin ofŽ cial military action. David Scheffer, U.S. ambassador-at-largefor war crimesissues, stated that the United States wanted “ aclearrecognition that states sometimesengage in verylegitimate uses of militaryforce to advanceinternational peaceand security.” 132 Theargument is thatquestions of the jus ad bello supersede questionsof the jusin bello. Duringthe NATO airstrikes in Kosovo,Serbia insisted thatNATO was violatinginternational law. Additional claims were madeby China afterthe errantbombing of theChineseembassy in Belgrade. Similarly, given recent evidenceof civilian deaths at the hands of U.S. troopsat No Gun Ri during the KoreanWar, thosehigh in the military’ s chainof command could be investigated andprosecuted by the ICC. 133 Skepticismis also on the rise among other key membersof the Security Council, including Russia, which faced considerable internationalcriticism over its handling of the con ict in Chechnya. Critics, however,argue that exceptions would render the ICC anempty vessel. Richard Dicker,associate counsel for HumanRights Watch, argued that the exceptions favoredby the United States represent “ aloopholethe size of theGrand Canyon that anyrogue state would drive right through.” 134 Although139 nations had signed the Rome Statute as of 12 February 2001, whetherit will be ratiŽed by the60 nationsnecessary for itsestablishment remains to be seen.135 Moreover,even if it does succeed in garnering the necessary signatories,how effective it will be without the support of the major powers, especiallythe United States, is alsonot clear. One U.S. ofŽcial remarked, “ We have shownthat the only way to get war criminalsto trial is for theU.S. totake a prominentrole. If theU.S. isnota leadplayer in thecreation of thiscourt, it doesn’ t happen.”136 WhileClinton’ s signingof the Rome Statute was laudedby ICC proponentsand human rights organizations, it may be more symbolic than instru- mental.Articulating the Bush administration’ s stanceat the UN, Secretaryof State ColinPowell declared, “ As youknow, the United States . ..doesnot support the InternationalCriminal Court. President Clinton signed the treaty, but we haveno plansto send it forward toour Senate for ratiŽcation.” 137 As normativeconsider- ationspress for harderlegalization in theemergent atrocities regime, 138 negotiating thepolitical dimensions necessary to building institutional strength seems predi-

131.David 1999, 368. 132.Quoted in AssociatedPress, 14August 1999, PM Cycle. 133. LosAngeles Times, 14October1999, A1. 134.Quoted in AssociatedPress, 14August 1999, PM Cycle. 135. See ^http://www.un.org/law/icc/statute/status.htm &. 136. Time, 27July 1998, 46. 137. LosAngeles Times ,2February2001, A4. 138.For example, holding perpetrators of genocide, war crimes, andcrimes againsthumanity accountableindependent of politicalpower and interests involved. Constructingan Atrocities Regime 681 catedon softeningsome aspects to gainthe necessary international consensus. The evidencesuggests that such softening measures have already taken place.

Evaluatingthe Atrocities Regime Formation Theevidence suggests that expanding liberal norms of stateconduct and protecting humanrights certainly explain the existence of tribunals in locales with little strategicor materialimportance. The proliferation of humanrights norms is evident incurrent legal trends in both the United States and Europe. 139 Inthe United States the term humanrights was articulatedin only 19 federalcourt casesprior to 1900; this number grew to34 from 1900to 1944,191 from 1945to 1969,803 in the 1970s, 2000 times in the 1980s, and over 4000 times in the 1990s. InEurope the case load of the European Court of HumanRights jumped from 11 casesduring 1959 –73to 395 cases during 1974 –92. 140 Theincreasing frequency of callsfor investigationsinto war crimesand crimes against humanity is a strong indicatorof changingnorms and sensibilities. Moreover, evolving notions of human rightsare reconŽ guring the norms of sovereignty that have limited international interventionin cases of internal atrocities. 141 Exponentialgrowth in the articulation of human rights norms is not only a functionof whatOran Youngtermed “ spontaneousregime development” ; itisalso beingcultivated by nongovernmental human rights organizations and aided by growingmedia coverage, often generated by such groups as HumanRights Watch andAmnesty International. 142 Inaddition the emergent atrocities regime itself may beseen as a normentrepreneur. 143 Onceestablished, the tribunal articulates and reinforcesnorms of state conduct and may also apply direct pressure to states throughcalls for investigationsor by releasing information to the media. Such pressuresmay be manifest at the systemic level, through states’ desiring to avoid beinglabeled “ pariahs”or “rogues”or simply through emulation. 144 In a world of interdependence,reputation is a valuableasset in maintaining positive relations with keypartners. 145 Pressures mayalso follow a “bottom-up”path, especially in liberal democracieswhere public exposure can generate policy demands. Certainly, addi- tionalresearch is necessaryto tracesuch demand-side questions and to identifythe roleof the tribunals themselves in generating demands for politicalaction. However,though these developments signal the evolution of norms to protect civiliansduring armed con ict, they may also be building norms that preclude

139.Henkin 1990. 140.See Jacobson1996; and Lutz and Sikkink 2000. 141.Abbott 1999, 372. See alsoFinnemore 1996; and Risse, Ropp,and Sikkink 1999. 142.Young 1983, 98 –99. 143.I thankan anonymous IO reviewer forthis important observation. 144.Rosecrance 1999. 145.See Chayesand Chayes 1995, 230; and Keohane 1997, 501. 682 InternationalOrganization militaryintervention at earlystages of crises.The danger of relyingon mechanisms thatonly respond expost facto toatrocities is clearly evident in both Bosnia and Rwanda.Though cognizant of atrocities in Bosnia, “ themajor powers . ..backed awayfrom signiŽcant armed intervention. Facing domestic criticism for allowing theslaughter to continue unchecked, some governments seemed to feel obliged to showthat they were doing something. Itwas inthis vacuum that the proposal for a tribunaladvanced.” 146 Ithas been suggested that while intervention may not have preventedgenocide in Rwanda, armed intervention may have been able to save as manyas 125,000 Tutsi lives— about 25 percentof thevictims. 147 Althoughhuman rightsnorms may be strengthening,norms of militaryintervention (often necessary for successfulatrocities adjudication) make action increasingly difŽ cult to initiate. Thesame groups that lobby for adjudicationand accountability are often the most vocalopponents of militaryintervention. Moreover, norms of interventionincreas- inglyrequire multilateral rather than unilateral action for bothoperational (cost- sharing)and political (legitimacy) reasons. 148 Clearly,this has troubling implica- tionsfor enforcement,for asthe evidence presented here suggests, military interventionmay be necessary in many cases for successfuladjudication.

Application

Realistvariables of power and interest best explain why tribunals may be estab- lishedin somecases but not in others.Power andinterest strongly in uence a state’s reluctanceto establish a givenad-hoc tribunal or besignatory to a comprehensive internationallegal regime. In the cases of Cambodia, East Timor, Chechnya, and Korea,great power nations were obviouslyreluctant to exposesensitive issues in a publicarena, especially past or present collusion with despotic regimes (in the Cambodiancase). In addition, strategic interests Ž gureprominently in thereluctance ofstrong states to ratify the Rome Statute. Modern warfare oftennecessitates destroying“ civilian”targets for militaryvictory, and in general “ collateraldamage” from bonaŽ demilitary missions has rarely been considered a violationof human rights,even by critics. 149 Thesemilitary actions may further the overall good, even whenthe human cost is high;in other words, the “ justwar” may sometimes involve regrettablehuman costs that should not be prosecutableoffenses under international law.The evidence presented here suggests that powerful states are reluctant to engageany regime that may signiŽcantly impedemeasures deemed necessary to achievingsecurity. The dominance of theSecurity Council in decisions to establish adhoc tribunals has been, to date, driven by stateinterests. While it canbe argued thatthe Balkans and Rwanda offer noparticularly salient security incentives,

146.Neier 1998,112. 147.Kuperman 2000, 108. See alsoDes Forges1999. 148.Finnemore 1986, 180 – 85. 149.Donnelly 1998, 531. See alsoMorgenthau 1985, 253– 60. Constructingan Atrocities Regime 683 establishingtribunals was certainlynot seen as threateningor compromisingto great powerinterests. Finally,intelligence problems Ž gureprominently in decisions to intervene in conicts and adjudicate war crimes;and this may also be afunctionof greatpower interests.In the case of Rwanda, Alan Kuperman noted that “ U.S. intelligence agenciescommitted virtually no in-countryresources to whatwas considereda tiny statein a regionwith little strategic value.” 150 WithinsufŽ cient information on the magnitudeand nature of the con ict, it was initiallydepicted as a two-sidedcivil war ratherthan one-sided genocide. Moreover, intelligence failures resulted in reportsof decliningviolence when violence was, infact, increasing. 151 Thus,realist variablesof great power interests can generate outcomes directly, by declining to initiateinvestigations in politicallysensitive areas, and indirectly, by not allocating adequateresources for intelligencegathering in areas perceived to have little strategicimportance.

ExpandedGoals and Institutional Adjustments

Whatfactors drive the manner in which an atrocities regime is constituted? Once established,what makes the regime more or lesseffective? How is“success”to be deŽned? While evolving norms of humanrights may initiate the construction of the atrocitiesregime in the Ž rst place,differentials in powerand the interestsof the most powerfulstates clearly shape the process of institutionalization.E. H.Carr suggested that,“ Thelaw is . ..theweapon of the stronger. . ..Law reects not any Ž xed ethicalstandard, but the policy and interests of thedominant group in agivenstate ata givenperiod.” As such,“ Politicsand law are indissolubly intertwined.” 152 This certainlyapplies to the case of war crimesadjudication. Iain Guest suggests that suspicionsran high, especially early in the tribunal’ s development,that the tribunal was servingas “ asubstitute,an alternative, to the kind of tough political action whichwould put an end to theethnic cleansing that was takingplace.” 153 States Ž nd establishinga tribunalsystem appealing because it provides an economically and politicallyinexpensive means of respondingto demands for internationalaction; it enablesstates to commit at alevelcommensurate with their strategic interest in the regioninvolved. From thestandpoint of realpolitik, theregime is a successwhether ornot it succeeds in bringing justice or alleviating ethnic con ict. From the standpointof idealpolitik, themeasures of success— reducing human suffering, protectinghuman rights, and promoting regional stability— are certainly left want- ing.Here we mustassess the tribunal’ s successfrom anotherdimension— as a componentof conict management.

150.Kuperman 2000, 101. 151.Ibid., 102. 152.Carr 1961,176 –77. 153.Quoted in Commissionon Securityand Cooperation in Europe 1996, 12. 684 InternationalOrganization

TheodorMeron offers thebest articulation of the regime’ s moreexpansive and idealisticaims: “ Thegreat hope of tribunaladvocates was thatthe individualizatio n anddecollectivizatio nofguilt . ..wouldhelp bring about peace and reconcilia- tion.. ..Anotherof thetribunal’ s objectiveswas deterrenceof continuedand future violationsof the law.” 154 For internationallawyers the connection between a functioninglegal regime and political order is clear:“ Therecan be nopeacewithout justice,no justice without law, and no meaningful law without a courtto decide whatis just and lawful under any given circumstance.” 155 If peaceis a functionof lawand justice, is anatrocitiesregime the panacea for theproblem of ethnonation- alistviolence? Here, thecurrent evidence is certainly not compelling. Effective deterrencerequires three elements— commitment, capability, and credibility. 156 The existenceof war crimestribunals and the successful prosecution of initial cases did littleto curb actions in any of thecases examined. The record of U.S. andNATO interventionin ethnic con icts over the past thirty years has been marked by very limitedcommitments, especially in cases where threats to U.S. interestswere limited.157 Becauseof the rather spotty record of the West regarding intervention andthe formidable institutional obstacles facing the  edglingtribunal system, perpetratorsof brutalityhave had little reason to takeUN commitmentseriously. In termsof capability, the United States has certainly possessed the power to appre- hendwar criminalsand political despots indicted by the tribunal. However, the difŽculty of apprehendingsuch people came at an unacceptably high logistical and politicalcost, considering that a large-scalemilitary commitment would be neces- saryand that to ensure stability such forces would need to remain for prolonged periods.158 Therelationship between a functioningatrocities regime and other elementsof amorecomprehensive strategy for managingcon ict— primarily armed intervention—isabundantly clear when assessing the regime’ s deterrencevalue. As oneanalyst noted, in the Bosnian case “ U.S. andEuropean (NATO) ofŽcials failed tosatisfy even the most basic strategic requirements of deterrence. These conditions includedeŽ nition of unacceptable behavior, clear communication of a commitment topunish transgressors, and demonstration of intent (that is, resolve) to carry out retaliation.”159 Preliminaryevidence does not seem to support notions that decollectivizatio nof guiltthrough war crimesadjudication is, on itsown, an effective means to achieving nationalreconciliation— seen as essential in dealing with ethnic or religious vio-

154.Meron 1997, 6. See alsoPejic 1998. 155.Ferencz 1980,1. 156.See Morgan1977; George and Smoke 1974; Lebow and Stein 1990; and Spiegel and Wheling 1999,497– 500. 157.See Callahan1997, 187– 199; and Harvey 1998. 158.Chaim Kaufmannremarked that“ suchpeaces last onlyas longas theenforcers remain.”Once peacekeepers are removedfrom the situation, the artiŽ cially established balance ofpower shifts, an “ethnicsecurity dilemma” arises, andthe credibility of majority commitment notto exploit minority ethnicgroups falters, threateningto renew thecycle ofviolence. See Kaufmann1996, 137; Posen 1993; andFearon 1998. 159.Harvey 1998, 230. Constructingan Atrocities Regime 685 lence(identity-based con ict). In the former Yugoslavia,ethnic tensions remain highand are accompanied by sporadic violence and acts of retaliation on both sides.160 Whileinstrumentalists may argue that ethnic tensions are manipulated by actorsto further material or politicalinterests, the ability to generate group solidarity andethnic blood-lust is certainlyfacilitated by a historicalcycle of violence. 161 In thissense, ethnic violence is congruent with other forms ofidentity con ict, includingreligious wars, andgroups have long endured cycles of violence and reprisal.162 Decollectivizingguilt is a curativemeasure taken by the state to break thishistorical cycle. However, the effectiveness of sucha strategyis contingent on detaininghigh-level perpetrators and, presumably, giving amnesty to thoseat lower levels(perhaps in return for admittingguilt, fully disclosing events, and testifying attrials of political and military leaders, as hasoccurred in truthand reconciliation proceedingselsewhere). Yet early precedent set by the tribunals runs an opposite course. Intheory, decollectivizing guilt may provide one way to stop the impulse for retributionand violence, but other approaches may be necessary as well. Chaim Kaufmannsuggests that spatial separation and “ ethnicunmixing” is important, at leastin the short run, to defuse the ethnic security dilemma. 163 However,such an approachto multiethnicsocieties is anathemato Westernliberal sensibilities, which regardforced population transfers as promoting rather than alleviating human suffering. Decollectivizingguilt also does not provide a meansof promoting tolerance by shapingethnic and national identities. Social constructivists argue that ethnic identitiesare malleable and shaped by continually changing social contexts, yet noneof thecurrently debated elements of ethniccon ict management incorporate a mechanismfor “re-imagining”the sociopolitical community. 164 Itwould seem that somemechanism of socialeducation should accompany decollectivization of guilt ifthe atrocities regime is to succeed within these more expansive agendas. Overcomingpolitical obstacles and achieving these higher aspirations are largely functionsof bothregime design and effective operating procedures. Institutionalists wouldpredict that a well-functioningregime in uences state behavior, but the evidencehere does not support this, at least where atrocities and ethnic con ict are involved.However, it may be premature to assess results, for similarto the constructivistview of newly emergent norms, the atrocities regime is a “workin progress.”In the process of establishinga well-functioninginstitution, the atrocities regimemust continue to reŽ ne its legal foundations, moderate the scope of its

160. See LosAngeles Times, 25March 2000, A5; and LosAngeles Times ,4March2001, A1, A9. 161.Instrumentalist accounts also do notexplain why ethnic and religious con ict tendto be somuch more barbaricthan other forms ofcon ict. Targeting of women and children and organized programs created toterrorize apopulationcertainly carry nospeciŽ c advantagesto conventional con ict in attainingmaterial gains.See Lake andRothchild 1998a, 5– 7; and Brown et al.1997. 162.Girard 1977, 24. 163.See Kaufmann1996; and Posen 1993. 164.Anderson 1983. 686 InternationalOrganization jurisdiction,and streamline its operating procedures. If we examinethe evolution of theatrocities regime, there is considerable evidence of “ institutionallearning.” Threeadjustments made between the Ž rst internationaltribunal and the Rome Statutemay prove signiŽ cant in making the regime more effective: (1) focusingon majortransgressions and pursuing a policyof exceptionalism toward those being prosecuted,(2) deŽning and reŽ ning key legal issues (deŽ nitional and procedural), and(3) recognizinga needto coordinate with national courts in the adjudication process.The fact that in several cases adjudication proceedings have been com- pletedin multiple jurisdictions must be seen as a successgiven the dearth of precedentand the desire to establish consensus among jurists with diverse and disparatelegal backgrounds. Both the ICTY andthe ICTR have successfully establishedinitial procedural guidelines and have articulated speciŽ c legalrules and principlesthat may strengthen norms of compliance regarding the jusin bello. Indeed,it can be arguedthat concretely articulating and clarifying norms of conduct areprerequisites to garnering widespread compliance. Since both ad hoc tribunals aidedin constituting such prerequisites, compliance/ deterrencemust be measured aftersuch norms have been articulated and disseminated in the international community.Important challenges remain, however, such as (1) allayingfears of greatpowers about expanding jurisdiction, (2) avoidingineffectiveness in the absenceof internationalcooperation or whena despoticregime remains in power, (3) addressingEuropean legal/ culturalbias, (4) establishinga balancebetween consistencyof legalnorms and  exibilityin speciŽc cases,(5) acceleratingthe pace ofproceedings,and (6) mitigatingthe cost of investigations.

Conclusion

Whatlessons can be drawnfrom theseinitial developments in theatrocities regime? Realistfactors have dominated the politics of war crimesadjudication, but the atrocitiesregime is in itsinfancy. To dismissthe efŽ cacy of theatrocities regime at thisstage is premature, and the evidence here suggests that its development is proceedingrapidly. From aninstitutionalist perspective, we canask how the regime canbe strengthened, and what lessons can be learned from theexisting ad hoc tribunalsystem. IL analystssuggest that the strength of legal regimes centers on consistency(precedent) and legitimacy, on hard law. 165 Conversely,regime ana- lysts,most notably in the Ž eldof international political economy, suggest that exibility,rather than rigidity, increases regime strength. 166 RobertKeohane argues that“ Institutionsbased on substantive rules have proven to be fragile entities,” adding“ exibilityand openness . ..mayincrease the usefulness of aninternational

165.See Franck1990; Jackson 1984; and Trimble 1990. 166.Krasner 1983. Constructingan Atrocities Regime 687 institution.”167 Flexibilityis also important when the long-term impacts of the institutionare uncertain, especially when state sovereignty and/ ornational security areinvolved. 168 Thekey to establishing an effective regime lies in squaring the circlebetween hard legalization and political  exibilityand locating the regime withina comprehensiveprogram of ethniccon ict management. On the Ž rst point, examiningthe cases as partof a dynamicpolitical development suggests that steps arebeing taken to “ soften”the legalization process— atleast in the short run— in orderto attain exibilityand minimize concerns about sovereignty and security. On thesecond point, the regime must be linked with other policy tools applicable to ethnicviolence, including preventive diplomacy, foreign aid, international interven- tion,spatial separation and reconŽ guring political spaces, and social education programs.169 War crimesadjudication also presents analytical challenges. A purelylegalistic (IL) viewcannot accurately explain many of the political dimensions involved in formingan atrocitiesregime nor can the highly macroscopic, analytical view of IR. Theissues presented here suggest the need for awar crimesvocabulary and more mid-leveltheories for understandingwar crimestribunals and their use in estab- lishingjustice and promoting peace. 170 Clearly,to understand and inform the developmentof theatrocities regime, we needresearch that incorporates the overlap betweenIL andIR. 171 Whileresearchers remain at the forefront of this agenda, promotingpeace and ameliorating human suffering provide strong incentives for furtheranalysis.

References

Abbott,Kenneth W. 1999.International Relations Theory, International Law, andthe Regime Governing Atrocitiesin Internal Con icts. TheAmerican Journal of International Law 93(2):361– 79. Abbott,Kenneth W., RobertO. Keohane,Andrew Moravcsik, Anne-Marie Slaughter, and Duncan Snidal.2000. The Concept of Legalization. InternationalOrganization 54(3):401– 19. Abbott,Kenneth W., andDuncan Snidal. 1998. Why States Act ThroughFormal International Organi- zations. TheJournal of Con ict Resolution 42(1):3– 32. ———.2000.Hard andSoft Law inInternational Governance. InternationalOrganization 54 (3):421– 56. Alvarez, Jose E.1999.Crimes ofStates/ Crimes ofHate: Lessonsfrom Rwanda. YaleJournal of InternationalLaw 24(2):365– 483. Anderson,Benedict. 1983. ImaginedCommunities: Re ections on theOrigin and Spread of Nationalism. London:Verso. Axelrod,Robert. 1984. TheEvolution of Cooperation. New York:Basic Books. Axelrod,Robert, and Robert O. Keohane.1985. Achieving Cooperation Under Anarchy: Strategies and Institutions. WorldPolitics 38:226–54.

167.Kahler 1995,137. See alsoGoldstein et al.2000, 392. 168.Abbott and Snidal 2000. 169.See Jentleson1998; Kaufmann 1996; Lake andRothchild 1998b; and Walter andSnyder 1999. 170.I owe thisimportant insight to an anonymous IO reviewer. 171.See Goldsteinet al.2000; Keohane 1997; and Slaughter 1993. 688 InternationalOrganization

Bakuramutsa,Manzi. 1995. Panel III: Identifying and Prosecuting War Crimes: TwoCase Studies–The FormerYugoslavia and Rwanda. New YorkUniversity LawSchool Journal of Human Rights 12 (3):631–88. Bass, Gary Jonathan.2000. Staythe Hand of Vengeance: The Politics of War Crimes Tribunals. Princeton,N.J.: Princeton University Press. Bennett,Christopher. 1995. Yugoslavia’s BloodyCollapse: Causes, Course, and Consequences. New York:New YorkUniversity Press. Beres, LouisRene ´.1988.Justice andRealpolitik: International Law andthe Prevention of Genocide. AmericanJournal of Jurisprudence 33:123–59. Brown,Michael E., Owen R.Cote´,Jr.,Sean M. Lynn-Jones,and Steven E. Miller,eds. 1997. Nationalismand Ethnic Violence. Cambridge,Mass.: MIT Press. Bull,Hedley. 1977. TheAnarchical Society. New York:Columbia University Press. Callahan,David. 1997. UnwinnableWars: American Power and Ethnic Con ict. New York:Hill and Wang. Carr, E.H.1961. TheTwenty Years’ Crisis, 1919 – 1939. London:Macmillan andCo. Chayes,Abram, andAntonia Handler Chayes. 1995. TheNew Sovereignty:Compliance with Interna- tionalRegulatory Agreements. Cambridge,Mass.: Harvard University Press. Christopher,Paul. 1994. TheEthics of War and Peace: AnIntroduction to Legal and Moral Issues. EnglewoodCliffs, N.J.: Prentice Hall. Cohen,Laurie A. 1997.Application of theRealist andLiberal Perspectives totheImplementation of War Crimes Trials:Case Studiesof Nuremberg and Bosnia. UCLAJournal of International Law and ForeignAffairs 2(1):113–70. Commissionon Security and Cooperation in Europe. 1996. TheWar Crimes Trialsfor the Former Yugoslavia:Prospects and Problems. Washington,D.C.: Commission on Securityand Cooperation in Europe. Conot,Robert E. 1983. Justice atNuremberg. New York:Harper andRow. Crawford,Beverly. 1994. Germany’ s Unilateral Recognitionof Croatia and Slovenia: A Case of Defectionfrom International Cooperation. Working Paper Series, 2.21.Berkeley: Center forGerman andEuropean Studies, University of California. David,Marcella. 1999.Grotius Repudiated: The American Objectionsto the International Criminal Court andthe Commitment to International Law. MichiganJournal of InternationalLaw 20(2):337– 412. Des Forges,Alison. 1999. Leave Noneto Tell the Story: Genocide in Rwanda. New York:Human Rights Watch. Donnelly,Jack. 1998.UnŽ nished Business. PS:PoliticalScience andPolitics 31(3):530 – 34. Fearon,James D.1998.Commitment Problems and the Spread of Ethnic Con ict. In TheInternational Spreadof Ethnic Con ict, editedby David A. Lake andDonald Rothchild, 107– 26. Princeton, N.J.: PrincetonUniversity Press. Ferencz, BenjaminB. 1980. AnInternational Criminal Court, A StepToward World Peace: A DocumentaryHistory andAnalysis. Vol.1. Dobbs Ferry, N.J.: Oceana Publications. Finnemore,Martha. 1996. Constructing Norms ofHumanitarian Intervention. In TheCulture of National Security, editedby Peter J.Katzenstein,153– 85. New York:Columbia University Press. Fox,Hazel. 1993.An InternationalTribunal for War Crimes: Willthe UN Succeed Where Nuremberg Failed? TheWorld Today 49(10):194 –97. Franck,Thomas M. 1990. ThePower of LegitimacyAmong Nations. New York:Oxford University Press. George,Alexander L., and Richard Smoke. 1974. Deterrence inAmericanForeign Policy: Theory and Practice. New York:Columbia University Press. Girard,Rene ´.1977. Violenceand the Sacred. Baltimore,Md.: Johns Hopkins University Press. Goldstein,Judith, and Robert O. Keohane,eds. 1993. Ideasand Foreign Policy. Ithaca, N.Y.:Cornell UniversityPress. Goldstein,Judith, Miles Kahler, Robert O. Keohane,and Anne-Marie Slaughter. 2000. Introduction: Legalizationand World Politics. InternationalOrganization 54(3):385– 99. Constructingan Atrocities Regime 689

Goldstein,Judith, Miles Kahler, Robert O. Keohane,and Anne-Marie Slaughter, eds. 2000. Legalization andWorld Politics. InternationalOrganization 54(3).Special issue. Grotius,Hugo. [1925] 1962. TheLaw of War and Peace [De Jure Belliac Pacis Libri Tres]. Translated byFrancis W. Kelsey. Reprint,Indianapolis, Ind.: The Bobbs-Merrill Co. Guest,Iain. 1996. The International Criminal Tribunal for the Former Yugoslavia: A Preliminary Assessment. In TheWar Crimes Trialsfor the Former Yugoslavia:Prospects and Problems, 75–84. Washington,D.C.: Commission on Security and Cooperation in Europe. Gutman,Roy. 1993. AWitnessto Genocide. New York:Macmillan. Harvey,Frank. 1998. Deterrence Failureand Prolonged Ethnic Con ict intheCase ofBosnia.In Peace inthe Midst of Wars: Preventing and Managing International Ethnic Con icts, editedby David Carment andPatrick James, 230– 64.Columbia: University of South Carolina Press. Hasenclever, Andreas,Peter Mayer,and Volker Rittberger. 1997. Theoriesof International Regimes. Cambridge:Cambridge University Press. Henkin,Louis. 1990. TheAge of Rights. New York:Columbia University Press. Howland,Todd, and William Calathes. 1998.The UN’ s InternationalCriminal Tribunal: Is ItJustice or Jingoismfor Rwanda? ACallfor Transformation. VirginiaJournal of InternationalLaw 39 (1):135– 67. Jackson,John H. 1984.Perspectives onthe Jurisprudence of InternationalTrade: Costs and BeneŽ ts of LegalProcedures in theUnited States. MichiganLaw Review 82(5/ 6):1570– 87. Jacobson,David. 1996. RightsAcross Borders:Immigration and the Decline ofCitizenship. Baltimore, Md.:Johns Hopkins University Press. Jentleson,Bruce W.1998.Preventive Diplomacy and Ethnic Con ict: Possible, DifŽ cult, Necessary. In TheInternational Spread of Ethnic Con ict, editedby DavidA. Lake andDonald Rothchild, 293– 316. Princeton,N.J.: Princeton University Press. Kahler,Miles. 1995. InternationalInstitutions and the Political Economy of Integration. Washington, D.C.:Brookings Institution. Katzenstein,Peter J.,ed.1996. TheCulture of National Security: Norms andIdentity in World Politics. New York:Columbia University Press. Kaufmann,Chaim. 1996. Possible and Impossible Solutions to Ethnic Civil Wars. InternationalSecurity 20(4):136 – 75. Keohane,Robert O. 1984. After Hegemony:Cooperation and Discord in the World Political Economy. Princeton,N.J.: Princeton University Press. ———.1997.International Relations and International Law: TwoOptics. HarvardInternational Law Journal 38(2):487–502. Keohane,Robert O., and Lisa L.Martin.1995. The Promise of Institutionalist Theory. International Security 20(1):39 –51. Koremenos,Barbara. 1999.On the Duration and Renegotiation of InternationalAgreements. Ph.D.diss., Universityof Chicago. Krasner, StephenD. 1999. Sovereignty:Organized Hypocrisy. Princeton,N.J.: Princeton University Press. Krasner, StephenD., ed.1983. InternationalRegimes. Ithaca, N.Y.:Cornell University Press. Kuperman,Alan J. 2000.Rwanda in Retrospect. ForeignAffairs 79(1):94– 118. Lake,David A., and Donald Rothchild. 1998a. Spreading Fear: TheGenesis ofTransnational Ethnic Con ict. In TheInternational Spread of Ethnic Con ict ,editedby David A. Lake andDonald Rothchild,3– 32. Princeton, N.J.: Princeton University Press. ———.1998b.Ethnic Fears andGlobal Engagement. In TheInternational Spread of Ethnic Con ict, editedby DavidA. Lake andDonald Rothchild, 339 –50.Princeton, N.J.: Princeton University Press. Lake,David A., andDonald Rothchild, eds. 1998c. TheInternational Spread of Ethnic Con ict. Princeton,N.J.: Princeton University Press. Lawyers Committee forHuman Rights. 1997. Prosecuting Genocide in Rwanda: A Lawyers Committee Reporton the ICTR and National Trials. Available online at ^http://www.lchr.org/pubs/rwanda.htm & (accessed May2001). 690 InternationalOrganization

Lebow,Richard Ned, and Janice Gross Stein.1990. When Does Deterrence Succeed andHow Do We Know?Occasional Papers Series, 8.Ottowa: Canadian Institute for International Peace andSecurity. Leigh,Monroe. 1996. The Yugoslav Tribunal: Use ofUnnamed Witnesses AgainstAccused. American Journalof InternationalLaw 90(2):235– 38. Lutz,Ellen L., andKathryn Sikkink. 2000. International Human Rights Law andPractice inLatin America. InternationalOrganization 54(3):633– 59. Meron,Theodor. 1995. International Criminalization of Internal Atrocities. AmericanJournal of InternationalLaw 89(3):554 –77. ———.1997.Answering for War Crimes: Lessonsfrom the Balkans. ForeignAffairs 76 (1): 2–8. Meyer,John W., JohnBoli, and George M. Thomas.1987. Ontology and Rationalization in the Western CulturalAccount. In InstitutionalStructure: Constituting State, Society, and the Individual, edited by GeorgeM. Thomas,John W. Meyer,Francisco O. Ramirez, andJohn Boli, 12– 37. Newbury Park, Calif.:Sage. Morgan,Patrick M. 1977. Deterrence: AConceptualAnalysis. BeverlyHills, Calif.: Sage. Morgenthau,Hans J.1985. PoliticsAmong Nations: The Struggle for Power and Peace. 6th ed. New York:Knopf. Morris,Madeline H. 1997.The Trials ofConcurrentJurisdiction: The Case ofRwanda. Duke Journalof Comparativeand International Law 7(2):349– 74. Morris,Virginia, and Michael P. Scharf.1995. AnInsider’ s Guideto the International Criminal Tribunal forthe Former Yugoslavia:A DocumentaryHistory andAnalysis. Irvington-on-Hudson,N.Y.: TransnationalPublishers. Neier, Aryeh.1998. War Crimes. New York:Times Books. Oye,Kenneth A., ed. 1986. CooperationUnder Anarchy. Princeton,N.J.: Princeton University Press. Pejic,Jelena. 1998.Creating a PermanentInternational Criminal Court: The Obstacles toIndependence andEffectiveness. ColumbiaHuman Rights Law Review 29(spring):291– 354. Posen,Barry R. 1993.The Security Dilemma andEthnic Con ict. In EthnicCon ict andInternational Security, editedby Michael E. Brown,103– 24. Princeton, N.J.: Princeton University Press. Risse, Thomas,Stephen C. Ropp,and Kathryn Sikkink, eds. 1999. ThePower of Human Rights: InternationalNorms andDomestic Change. New York:Cambridge University Press. Rosecrance, RichardN. 1999.Emulation in International Politics. Paper presented at the95th Annual Meetingof theAmerican PoliticalScience Association,2– 5 September,Atlanta, Georgia. Scharf,Michael P. 1997. BalkanJustice. Durham,N.C.: Carolina Academic Press. Scharf,Michael P., andValerie Epps.1996. The International Trial of the Century? A “Cross-Fire” Exchangeon theFirst Case Before theYugoslavia War Crimes Tribunal. CornellInternational Law Journal 29(3):635–63. Schrag,Minna. 1995. The Yugoslav Crimes Tribunal:A Prosecutor’s View. Duke Journalof Compar- ativeand International Law 6(1):187–95. Sikkink,Kathryn. 1993. The Power of PrincipledIdeas: HumanRights Policies in the United States and Western Europe.In Ideasand Foreign Policy, editedby Judith Goldstein and Robert O. Keohane, 139– 70.Ithaca, N.Y.:Cornell University Press. ———.1998.Transnational Politics, International Relations Theory, and Human Rights. PS:Political Science andPolitics 31(3):517–21. Silber,Laura, and Allan Little. 1996. Yugoslavia:Death of aNation. New York:TV Books. Slaughter,Anne-Marie. 1993. International Law andInternational Relations Theory: A DualAgenda. AmericanJournal of International Law 87(2):205– 39. Stein,Arthur A. 1990. WhyNations Cooperate. Ithaca, N.Y.:Cornell University Press. Taylor,Telford. 1992. TheAnatomy of theNuremberg Trials: A PersonalMemoir. New York:Knopf. Thornberry,Cedric. 1996. Saving the War Crimes Tribunal. ForeignPolicy 104(fall):72– 85. Trimble,Phillip R. 1990.International Law, WorldOrder, and Critical LegalStudies. StanfordLaw Review 42(3):811– 45. Van Schaack,Beth. 1999. The DeŽ nition of Crimes AgainstHumanity: Resolving the Incoherence. ColumbiaJournal of TransnationalLaw 37(3):787–850. Constructingan Atrocities Regime 691

Walter,Barbara F.andJack Snyder,eds. 1999. CivilWars, Insecurity, and Intervention. New York: ColumbiaUniversity Press. Watson,Adam. 1992. TheEvolution of International Society. London:Routledge. Weber,Max. 1920. Gesammelte Aufsa¨tze zurReligionssoziologie. Tu¨bingen:J. C.B.Mohr. Wendt,Alexander. 1992. Anarchy Is WhatStates Makeof It:The Social Construction of PowerPolitics. InternationalOrganization 46(2):391–425. ———. 1999. SocialTheory of InternationalPolitics. Cambridge:Cambridge University Press. Young,Oran R.Regime Dynamics: TheRise andFall of International Regimes. In International Regimes, editedby StephenD. Krasner, 93–113. Ithaca, N.Y.:Cornell University Press.