the fujimori judgment - department-ambos.uni-goettingen.de · gent groups or opposed to the...

The Fujimori Judgment

A President’s Responsibility for Crimes AgainstHumanity as Indirect Perpetrator byVirtue of anOrganized PowerApparatus

Kai Ambos*

AbstractIn 2009, former Peruvian president Alberto Fujimori was convicted by the PeruvianSupreme Court to 25 years’ imprisonment as an indirect perpetrator of serioushuman rights violations amounting to crimes against humanity committed underhis presidency in 1991 and 1992. The conviction is based on the theory of control/domination of the act by virtue of an organized power apparatus (autor|¤ a mediatapor dominio de la voluntad en aparatos de poder organizados/mittelbareTa« terschaft kraft Willensherrschaft in organisatorischen Machtapparaten)demonstrating once again that this theory is a serious option to hold criminal lead-ers to account. The article gives a brief overview of the background of the case andthen moves on to analyse Fujimori’s responsibility from a national and internationalcriminal law perspective, focusing in particular on the evidentiary issues andthe five requirements of Organisationsherrschaft set out by the Peruvian SupremeCourt.

1. System Criminality and Selectivity in theFujimori Case

A. Historical and Political Background

In 1990, Alberto Fujimori was elected president of Peru in the midst of both ageneral economic and political crisis.1 Immediately following his electionFujimori concentrated his governmental politics on economic reforms and thefight against the insurgent groups Sendero Luminoso (SL) and Movimiento

* Kai Ambos is Professor at the Georg-August-Universita« t Go« ttingen and Judge at the Landgericht.I am indebted to Florian Huber, PhD candidate and research assistant at my Chair, for hisinvaluable support. [[email protected]]

1 Truth and Reconciliation Commission of Peru (‘TRC’), General Conclusions, x 98, available athttp://www.cverdad.org.pe/ingles/ifinal/conclusiones.php (visited 5 June 2010).

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Journal of International Criminal Justice 9 (2011), 137^158 doi:10.1093/jicj/mqq059

Advance Access publication 5 October 2010

� Oxford University Press, 2010, All rights reserved. For permissions, please email: [email protected]

by Antonio C

assese on February 16, 2011

jicj.oxfordjournals.orgD

ownloaded from

http://jicj.oxfordjournals.org/

Revolucionario Tupac Amaru (MRTA), which had emerged in 1980 and were in-tensifying their fight in the late 80s and early 90s.2 Between 1980 and 2000,the armed conflict caused the death of approximately 69,200 persons.3

Shortly after his election, Fujimori started to restructure and centralize theactivities of the intelligence agencies, the armed forces and the police, therebyplacing such organs under his direct command.4 At the same time, he reliedon a sophisticated system of corruption, embezzlement and bribery to financesecret military operations and to secure political support for his government.Since the beginning of his presidency, Fujimori created a clandestine and

well equipped special unit, later called the Colina group, which was integratedinto the intelligence structure of the armed forces and was designated in ac-cordance with state policy to undertake secret operations consisting in theidentification, control and elimination of those suspected of belonging to insur-gent groups or opposed to the government.5 The Colina group formed part ofthe Army Intelligence Service (Servicio de Inteligencia del Eje¤ rcito, SIE), whichwas itself a part of the General Staff’s Intelligence Directorate (Direccio¤ n deInteligencia del Eje¤ rcito, DINTE). The Intelligence Directorate was formally sub-ordinated to the Joint Command of the Armed Forces (Comando Conjunto de lasFuerzas Armadas), but in practice received orders from the highest intelligenceagency, the National Intelligence Service (Servicio de Inteligencia Nacional,SIN), led by Fujimori’s ally and closest adviser, Vladimiro Montesinos.6 TheColina group had its origin in a special intelligence group (grupo de ana¤ lisis)created in 1990 to conduct strategic intelligence operations, but was endowedin 1991 with operative tasks to realize special operations (operaciones especialesde inteligencia) consisting in the physical execution of alleged members of ter-rorist organizations and political opponents. Between 1991 and 1992, theColina group committed at least 11 operations involving extrajudicial execu-tions and forced disappearances, before it was dissolve in late 1992.7

One of the first operations, in November 1991, was the extrajudicial killing of15 persons erroneously suspected to be SL members. The killings took placein the Barrios Altos district of Lima. Another operation of the Colina groupconsisted of the kidnapping, execution and forced disappearance of nine stu-dents and a professor of the La Cantuta National University of Lima in July1992, as a reaction to a bomb attack of Shining Path. These two incidents

2 See B. Kay, ‘‘Fujipopulism’and the Liberal State in Peru, 1990^1995’, 38 Journal of InteramericanStudies and World Affairs (1996) 55^68; S. Stokes, ‘Democratic Accountability and PolicyChange: Economic Policy in Fujimori’s Peru’, 29 Comparative Politics (1997) 209^226.

3 TRC, supra note 1, x 2.4 D. Caro Coria, ‘Sobre la punicio¤ n del ex presidente Alberto Fujimori Fujimori como autor med-

iato de una organizacio¤ n criminal estatal’, 4 Zeitschrift fu« r internationale Strafrechtsdogmatik(‘ZIS’, www.zis-online.com) (2009) 581^596, at 590^591.

5 See TRC, Final Report, Vol. VI, x 154, available at http://www.cverdad.org.pe/ifinal/index.php(visited 5 June 2010).

6 See Corte Suprema de Justicia (‘CSJ’), infra note 33, x 275 et seq., 324 et seq. and 301 et seq.7 Ibid., x576.

138 JICJ 9 (2011), 137^158

by Antonio C



ownloaded from


were part of a number of crimes committed by the Colina group as part ofFujimori’s secret anti-terrorism policy to execute alleged members of terroristgroups.8

On 5 April 1992, in response to the parliament’s previous rejection of thegovernments new anti-terrorist decrees, Fujimori ç in alliance with the mili-tary and the assent of most of the population ç carried out a ‘self-coup’(auto-golpe) and dissolved the Congress, suspended the Constitution and tookover the judiciary in order to transfer legislative control to the executivebranch and weaken judicial independence.9 Due to international pressure, anew Constitution was adopted by popular referendum in 1993 and a newCongress dominated by supporters of Fujimori was elected.10 In 1995, thanksto his success in the fight against hyperinflation and the elimination of terror-ism, Fujimori was re-elected. Finally, on 28 July 2000, Fujimori assumedhis third term in office, but shortly after, in September 2000, the eruption ofan enormous corruption scandal, implicating Fujimori’s adviser and closestally, Vladimiro Montesinos, in the bribery of Congressmen in exchange fortheir political support of Fujimori, caused Fujimori to leave Peru. On14 November 2000, he fled to Japan and sent his resignation via fax to theCongress, which declared him morally incapacitated to carry out thepresidency.

B. Political and Legal Framework for Fujimori’s Prosecution

Several factors have paved the way for the investigation, prosecution and con-viction of Fujimori in the five cases of corruption and two cases of humanrights violations, which formed part of the extradition request of Peru andwere approved by the Chilean Supreme Court of Justice.

1. First Attempts of National Prosecution

Initial investigations of the Barrios Altos killings of 1991 by a parliamentaryinvestigating committee were interrupted by Fujimori’s self-coup and the clos-ure of the Congress. In 1995, as a response to new judicial investigations ofthe killings in the Barrios Altos case and the disappearances in the La Cantutacase, the government pushed the Congress to pass two amnesty laws inorder to avoid further investigations.11 On 15 June 1995, Congress passed

8 R. Gamarra, ‘A Leader Takes Flight: The Indictment of Albert Fujimori’, in E. Lutz and C. Reiger(eds), Prosecuting Heads of State (Oxford: Oxford University Press, 2009) 95^110, at 100^101.

9 See J.M. Burt, ‘‘Quien habla es terrorista’: The Political Use of Fear in Fujimori’s Peru’, 41 LatinAmerican Research Review (2006) 32^58, at 32.

10 M. McFarland Sanchez-Moreno, ‘When a Constitution is a Constitution: Focus on Peru’, 33 NYUJournal of International Law and Politics (2000^2001) 561^616.

11 See L. LaPlante, ‘Outlawing Amnesty: The Return of Criminal Justice in Transitional JusticeSchemes’, 49 Virginia Journal of International Law (2008^2009) 915^984, at 947, 952.

A President’s Responsibility for Crimes Against Humanity 139

by Antonio C



ownloaded from


Law No. 2647912 granting a blanket self-amnesty to all members of the securityforces and civilians for crimes committed during the period of 1980 to 1995in the fight against terrorism. The few convictions of members of the securityforces for human rights violations were immediately annulled. Consequently,eight members of the Colina Group who had been convicted and imprisonedfor the La Cantuta case, some of whom were also being prosecuted inthe Barrios Altos case, were set free. When a judge in the Barrios Altos casequestioned the constitutionality of Law No. 26479 and intended to con-tinue with the investigations, a second amnesty law, Law No. 26492 of 2 July1995,13 was adopted with the aim to prevent the judiciary from determiningthe legality or applicability of the first. This law extended the amnesty tooffences not yet investigated and decreed that the amnesty could not be‘revised’ by the judiciary instance and that its application was obligatory.14

These measures made further investigations impossible until the end ofFujimori’s government.

2. Overcoming Peru’s Amnesty Laws: The Role of the Inter-American Court ofHuman Rights

An important role in overcoming Peru’s amnesty laws and in setting the stand-ards with regard to the duty to prosecute serious human rights violations wasplayed by the Inter-American Court of Human Rights (IACtHR).15 At the endof Fujimori’s government, more than 300 cases were pending before theInter-American Court and the Inter-American Commission of Human Rights.Between 1995 and 2007, the IACtHR found the Peruvian state to be in breachof its international obligations in 22 cases, 18 of them dealing mainly with ar-bitrary killings, disappearances, torture and inhumane treatment, arbitrarydetention and violations of the right to fair trial under the government ofFujimori since 1990.16 In its landmark decision on the killings in the BarriosAltos case, the IACtHR ruled on 14 March 2001 ç only four months afterFujimori’s fall ç that all amnesty and other provisions ‘designed to eliminateresponsibility’ are ‘inadmissible’ because they violate, inter alia, the right toan effective remedy and the state’s obligation to investigate and punish

12 See ‘Normas Legales’, Diario Oficial El Peruano, Lima, 15 June 1995.13 See ‘Normas Legales’, Diario Oficial El Peruano, Lima, 2 July 1995.14 D. Kuwali and J.P. Pe¤ rez-Leon Acevedo, ‘Smokescreen ^ A Survey of the Evolving Trends in

Amnesty Laws in African and Latin America’, 2 Malawi Law Journal (2008) 115^134, at 128.15 See C. Sandoval, ‘The Challenge of Impunity in Peru: The Significance of the Inter-American

Court of Human Rights’, 5 Essex Human Rights Review (2008) 1^20, available at http://projects.essex.ac.uk/ehrr/V5N1/Sandoval.pdf (visited 5 June 2010).

16 In 1999, an attempt by the government to withdraw from the contentious jurisdiction of theCourt was rejected by the Court arguing that it is only possible to denounce the Conventionas a whole (Ivcher-Bronstein v. Peru Case, Judgment of 24 September 1999, IACtHR (Ser. C)No. 54 (1999), at x 56; Constitutional Court v. Peru Case, Judgment of 24 September 1999(Ser. C) No. 55 (1999), at x 55.)

140 JICJ 9 (2011), 137^158

by Antonio C



ownloaded from


those responsible.17 Furthermore, the Court made it clear that the ‘laws lacklegal effect and may not continue to obstruct’ the investigation and punish-ment of the responsible.18

3. The Peruvian TRC, Reopening Criminal Investigations Against Fujimori,and the Extradition Proceedings in Chile

The collapse of ‘Fujimorismo’and the weakness of the insurgency at the end ofthe 1990s facilitated an expeditious transition from Fujimori’s authoritarianregime to the reestablishment of democracy without having to deal with a ser-ious armed challenge against the new political rulers or to rely on peace talkswith the rebel groups.19 Thus, with the Peruvian Supreme Court of Justiceendorsing the Barrios Altos judgment of the IACtHR,20 new investigationsagainst members of the armed forces were quickly (re-) opened. Under the in-terim president Valent|¤ n Paniagua, a Truth and Reconciliation Commission(TRC) was established on 2 June 2001 and entrusted with ‘clarifying the pro-cess, the facts and the responsibilities of the terrorist violence and humanrights violations produced from May 1980 to November 2000’.21 The TRC wasentrusted with providing an interpretation of the historical period and the in-vestigation of serious crimes like murder, kidnapping, enforced disappearancestorture and other serious injuries.22 On 28 August 2003, it presented its finalreport containing 73 representative cases of human rights violations and rec-ommending criminal proceedings in 43 cases.23

However, with Fujimori’s voluntary exile to Japan in 2000, the chance tobring him back to Peru for trial seemed slim as his Japanese nationality pro-tected him from extradition.24 The situation changed with Fujimori’s surpris-ing trip to Chile in September 200525 where he was detained by the Chilean

17 Barrios Altos Case (Chumbipuma Aguirre et al. v. Peru), Judgment of 14 March 2001, IACtHR (Ser.C) No. 75 (2001), at xx 41^42; see also, La Cantuta v. Peru Case, Judgment of 19 November 2006,IACtHR (Ser. C) No. 162 (2006), at x152.

18 Barrios Altos, ibid., at 17, x 44.19 E. Gonza¤ lez Cueva, ‘The Peruvian Truth and Reconciliation Commission and the challenge of

impunity’, in N. Roht-Arriaza and J. Mariezcurrena (eds), Transitional Justice in the Twenty-FirstCentury: Beyond Truth versus Justice (Cambridge: Cambridge University Press, 2006) 70^93,at 74.

20 E.Vega Luna,‘La responsabilidad penal de los agentes del Estado’, in L. Magarrell and L. Filippini(eds), El legado de la verdad. La justicia penal en la transicio¤ n peruana (New York: ICTJ, 2006)105^133, at 111.

21 Supreme Decree No. 065-2001-PCM, Art. 1, 2 June 2001; translation available at: http://www.mississippitruth.org/documents/PERU.pdf (visited 5 June 2010).

22 Ibid., Art. 3; see also J. Ciurlizza and E. Gonza¤ lez, ‘Verdad y justicia desde la o¤ ptica de laComisio¤ n de laVerdad y Reconciliacio¤ n’, in Magarrell and Filippini, supra note 20, 85^105.

23 TRC, supra note 5,Vol.VII, Chapter 2. See also Vega Luna, supra note 20, at 105, 107.24 See K. Anderson, ‘An Asian Pinochet?’ 38 Stanford Journal of International Law (2002) 177^206.25 For the reasons of Fujimori‘s return, see P. Noboa, ‘Former Peruvian President Albert Fujimori’s

Extradition Process’, 14 Law and Business Review of the Americas (2008) 621^630, at 625.


by Antonio C



ownloaded from


authorities on grounds of an (internationalized) arrest warrant issued byInterpol on behalf of Peru. Peru then requested his extradition26 with regardto 13 cases, three of them dealing with human rights violations (includingBarrios Altos and La Cantuta, arbitrary detention of five persons by the SIE,and the forced disappearances of several persons) and 10 dealing withcorruption-related crimes.27

On11 July 2007, the extradition request was rejected by a first instance judgeof the Chilean Supreme Court invoking lack of evidence concerning Fujimori’sresponsibility and statutory limitations.28 Yet, two months later, on21 September 2007, a Chamber of the Chilean Supreme Court overruled thefirst instance ruling, rejecting the arguments of Fujimori’s defence team andgranting the extradition in seven cases. This included the case of the extra-judicial killings of Barrios Altos and the forced disappearances of La Cantuta,the arbitrary detentions of a journalist and a businessman in April andJuly 1992, and five of the seven corruption cases.29 The Chamber heldthat the evidence provides a probable cause for an indictment of Fujimorias an indirect perpetrator.30 However, due to the double criminality require-ment and the absense of forced disappearance within the Chilean CriminalCode,31 the Chamber limited the extradition to the offences of aggravated homi-cide and bodily injury in La Cantuta, thereby precluding Fujimori’s later convic-tion for the crime of forced disappearance under Peruvian criminal law.

26 The extradition proceedings in the case were governed by the 1932 Extradition Treaty betweenPeru and Chile, Art. 644^656 of the Chilean Code of Criminal Procedure and international in-struments and principles of law, like the Montevideo Convention on Extradition (26 December1932, O.A.S.T.S. No. 34) and the Convention on Private International Law, known as theBustamante Code (20 February 1928, O.A.S.T.S. No. 34); see M. Haas, ‘Fujimori Extraditable!.Chilean Supreme Court Sets International Precedent for Human Rights Violations’, 39University of Miami Inter-American Law Review (2007^2008) 373^408, at 387.

27 The corruption cases referred to personal enrichment, secret intelligence operations as part ofhis anti-terrorism politics, the spying on political opponents, the bribing of politicians, see alsoinfra note 32.

28 CSJ Chile, Primera Instancia, decision of 11 July 2007, Rol. N8 5646^2005, at x 112, 115, 122,available at http://www.emol.com/noticias/documentos/pdfs/FalloMinalvarezFujimori.pdf (vis-ited 5 June 2010).

29 CSJ Chile, Segunda Instancia, decision of 21 September 2007, Rol. N8 3744-07, availableat http://www.latinreporters.com/perouFujimoriFalloExtradicionChile21092007.pdf (visited5 June 2010).

30 Ibid., at xx 4^5.31 So far, Chile has not ratified the Inter-American Convention on Forced Disappearance of

Persons which stipulates in its Art. 5 that the forced disappearance shall be deemed to beincluded among the extraditable offences in every extradition treaty concluded between StatesParties. See generally J-L. Guzma¤ n, ‘Cap|¤ tulo III-Chile’, in K. Ambos (ed.), Desaparicio¤ n Forzadade Personas, Ana¤ lisis Comparado e Internacional (Bogota¤ : GTZ/Temis, 2009) 55^73.

142 JICJ 9 (2011), 137^158

by Antonio C



ownloaded from


2. Fujimori’s Criminal Responsibility as a SeniorPolitical and Military Commander

Since his extradition to Peru in 2007, Fujimori has been convicted in all sevencases for which the extradition request was granted. As to the corruption cases,Fujimori avoided public trials by accepting the charges and was convicted threetimes as a direct perpetrator.32 However, in the two human rights cases (the‘Barrios Altos/La Cantuta case’ and the case of arbitrary detentions, called‘So¤ tanos SIE case’) Fujimori denied any responsibility for the extrajudicial execu-tions, forced disappearances and arbitrary detentions. On 7 April 2009, one anda half years after the opening of the public trial and after more than 60 trial ses-sions, the Special Criminal Chamber of the Supreme Court of Justice of Peru con-victed Fujimori to the maximum prison sentence of 25 years for aggravatedhomicide/murder (homicidio agravado/asesinato) in 25 cases and serious bodilyinjury in four cases concerning the events of Barrios Altos and La Cantuta, andaggravated kidnapping in the two cases of arbitrary detentions.33 On30 December 2009, the First Transitory Criminal Chamber of the Supreme Courtconfirmed the conviction, essentially following the Special Chamber.34

A. The Imputed Conduct From an International and National Perspective

The Supreme Court explicitly qualifies the crimes of Barrios Altos, La Cantuta andSo¤ tanos SIE as state crimes and ç in so far following the IACtHR35 ç crimesagainst humanity.36 In the appeals judgment, the Chamber makes clear thatdespite the missing incorporation of crimes against humanity in the PeruvianCriminal Code, this classification neither violates the principle of legality (espe-cially in its variant of non-retroactivity) nor the principles of double criminal-ity and the specialty of extradition law since Fujimori was prosecutedpursuant to the Peruvian Criminal Code, i.e. for aggravated homicide/murder,serious bodily injury and aggravated kidnapping, as set out in extradition

32 CSJ, Sala Penal Permanente, judgment of 11 December 2007, Exp. No. AV 13-2003 (CaseAllanamiento); CSJ, Segunda Sala Penal Especial, judgment of 10 April 2008, Exp. No. AV13-2003 (Case Allanamiento: prison sentence of 6 years); CSJ, Sala Penal Especial, judgmentof 20 July 2009, Ex. No. AV 23-2001 (Case 15 Millones:7,5 years); CSJ, Sala Penal Especial, judg-ment of 30 September 2009, Ex. No. AV 33-2003 (Cases Medios de Comunicacio¤ n,Congresistas Tra¤ nsfugas, Interceptacio¤ n Telefo¤ nica: 6 years), all available at www.pj.gob.pe.See also supra note 27.

33 CSJ, Sala Penal Especial, judgment of 7 April 2009, Exp. No. AV 19-2001, available athttp://www.pj.gob.pe/CorteSuprema/spe/index.asp?codigo¼10409&opcion¼detalle_noticia (vis-ited 5 June 2010); a German translation of Part III of the judgment was published in 4 ZIS(2009) 622^657, available at www.zis-online.com; an English translation of the Peruvian gov-ernment’s synopsis of the trial and conviction is announced for the 6. Issue (Special Issue) of25 American University International Law Review (2010).

34 CSJ, Primera Sala Penal Transitoria, Exp. No. AV 19-2001, judgment of 30 December 2009,available at http://www.pj.gob.pe/CorteSuprema/SalasSupremas/SPT/documentos/R.N.%20N%C2%BA%2019-01-2009-A.V..pdf (visited 5 June 2010).

35 La Cantuta, supra note 17, x 225.36 CSJ, supra note 33, x653, 675, 717.


by Antonio C



ownloaded from


decision of the Chilean Supreme Court.37 With the recourse to the term ‘crimesagainst humanity’ the Court did not, as alleged by Fujimori’s defence team, pre-tend to introduce the International Criminal Court (ICC) Statute into thePeruvian domestic legal order in a ‘self-executing’ manner, but only character-ize the international nature and dimension of the crimes committed by theFujimori regime.38 While thereby a violation of the above-mentioned principleswas avoided, the crime of forced disappearance had to be excluded from theproceedings due to its nonexistence in Chilean domestic law39 pursuant tothe double criminality and specialty principles.

B. Evidentiary Issues in the Context of System Criminality

One of the key issues of the proceedings against Fujimori was how to deal withevidence regarding Fujimori’s individual criminal responsibility in the crimescommitted by the Colina group. The events of the Barrios Altos and LaCantuta cases had obtained broad media coverage, were part of the investiga-tions of the TRC,40 had been dealt with before by the IACtHR (demonstratingthe systematic and widespread character of the crimes committed by theFujimori regime and suggesting, at least, negligent ignorance on the part ofFujimori), and prepared the ground for the convictions of several low rankmembers of the Colina group.To prove Fujimori’s criminal responsibility, the Supreme Court relied on indir-

ect or circumstantial evidence which it found as reliable as direct evidence and,in addition, considered as the more appropriate form of evidence to demonstratecriminal responsibility of high ranking officials who use a clandestine powerstructure to commit such crimes.41 Indeed, it is true that circumstantial evi-dence is commonly used in national and international criminal proceedings.42

It requires a multiplicity of circumstances which ç based on criteria of

37 CSJ, supra note 34, at 111^113, 124.38 CSJ, supra note 34, at 124.39 See Guzma¤ n, supra note 31, 55 et seq. Ironically, the crime existed in Peruvian domestic law

since 1991 (now Art. 320 Criminal Code) with a little interruption (abolition through DecreeLaw No. 25475 of 5 May 1992 and re-enactment through Decree Law N8. 25592 of 2 July1992), see I. Meini, ‘Cap|¤ tulo V ^ Peru¤ ’, in Ambos, supra note 31, 107^134.

40 The TRC considered Fujimori and his close allies responsible for the crimes committed by theColina group, see TRC, supra note 1, x100.

41 CSJ, supra note 33, xx658^659, CSJ, supra note 34, at 137.42 For the international level see, Judgment, Delalic¤ (IT-96-21-T),Trial Chamber,16 November1998,x 386; Judgment, Blas› kic¤ (IT-95-14-T), Trial Chamber, 3 March 2000, x 307; Judgment, Halilovic¤(IT-01-48-T), Trial Chamber, 16 November 2005, x 66; Judgment, Oric¤ (IT-03-68-T), TrialChamber, 30 June 2006, xx 319^320; see alsoVela¤ squez-Rodr|¤ guez v. Honduras Case, Judgment of29 July 1988, IACtHR (Ser. C) No. 4 (1988), at xx 130^131 and God|¤ nez-Cruz v. Honduras Case,Judgment of 20 January 1989, (Ser. C) No. 5 (1989), at xx 136^137; for the correct standard ofproof see A. Zahar and G. Sluiter, International Criminal Law (Oxford: Oxford University Press,2008), at 417 et seq. See for the Peruvian criminal procedure C. San Mart|¤ n, Derecho ProcesalPenal Vol. II (2nd edn., Lima: Grijley, 2003), at 851^865; for the German criminal procedureU. Eisenberg, Beweisrecht der Strafprozessordnung (6th edn., Mu« nchen: C.H. Beck, 2008), x9.

144 JICJ 9 (2011), 137^158

by Antonio C



ownloaded from


experience, logic and rationality and the absence of contradictions or possiblediffering conclusions ç allow to consider a certain fact as existing, even if nodirect evidence pointing to that fact can be established.43 In other words, thefact to be proven must be the only possible conclusion to be drawn on thebasis of the existing circumstantial evidence.44 Due to the lack of direct evi-dence regarding Fujimori’s role in the establishment and supervision of themilitary operations of the Colina group,45 the Chambers relied on the followingproven facts to infer Fujimori’s responsibility:46

(i) the crimes occurred during the presidency and were directed by mem-bers of the armed forces against political opponents or supposed mem-bers of terrorist organizations;

(ii) the operations were planned by the central intelligence agency SINwhich was run de facto by Vladimiro Montesinos, the closest ally andadviser of Fujimori during the 1990s, who regularly informed Fujimoriabout the operations of the SIN;

(iii) like in the case of Montesinos, persons close to Fujimori were assignedas high ranking state and army officials;

(iv) the Colina group had its origin in a special intelligence group estab-lished by Fujimori and received since 1991 a broad logistical, materialand financial support;

(v) any intent of denouncing or investigating the crimes attributed to theColina group was opposed by Fujimori denying the responsibility ofstate forces in the events, defending the officials suspected of thecrimes, impeding the work of the judiciary and issuing amnesty lawsto avoid further criminal investigations.

These circumstances allowed the Chambers to conclude that the crimes com-mitted by the Colina group did not occur behind Fujimori’s back, but werepart of his anti-terrorism strategy to establish a clandestine group of armymembers to execute civilians suspected of terrorist activities.

C. Indirect Perpetration byVirtue of an Organized Power Apparatus

1. Theoretical Starting Point

The conviction of Fujimori as indirect perpetrator47 is based on the theory ofcontrol/domination of the act by virtue of a hierarchical power apparatus.48

The ‘man in the background’ dominates the direct perpetrator’s will and acts

43 CSJ, supra note 33, x659, CSJ, supra note 34, at 136.44 See CSJ, supra note 33, x658, CSJ, supra note 34, at 81 et seq., 132 et seq.45 CSJ, supra note 33, x662.46 CSJ, supra note 33, x660, CSJ, supra note 34, at 81^87.47 Art. 23 of the Peruvian Criminal Code reads in the relevant part: ‘El que realiza por medio de

otro el hecho punible ::: ’.48 CSJ, supra note 33, xx718^748, CSJ, supra note 34, at 39^52.


by Antonio C



ownloaded from


by means of an organizational apparatus of hierarchical power, i.e. an orga-nized power structure (Organisationsherrschaft).49 This theory, in turn, isbased on a concept of control or domination of the act (Tatherrschaft)50 which isused in differentiated systems of participation51 and now also by the ICC,52

to delimitate forms of perpetration (commission as a principal) from forms ofsecondary participation (instigation, inducement, aiding, assisting).53 In fact,the control over the act is the key structural difference between indirect per-petration and joint criminal enterprise, the latter resting in contrast on theshared intent or common purpose of the members of a criminal enterprise.54

49 See C. Roxin, Ta« terschaft und Tatherrschaft (8th edn., Berlin: De Gruyter, 2006), at242^252,704^717; see also K. Ambos, DerAllgemeine Teil desVo« lkerstrafrechts (Berlin: Dunker &Humblot, reprint 2004), at 557 et seq.; idem, La Parte General del Derecho Penal Internacional(Bogota¤ : Temis, 2006), at 216 et seq.; H. Ola¤ solo, The criminal responsibility of senior political andmilitary leaders as principals to international crimes (Oxford and Portland: Hart Publishing,2009), at 116 et seq.

50 K. Ambos, ‘Command responsibility and Organisationsherrschaft: ways of attributinginternational crimes to the most responsible’, in A. Nollkaemper and H. van der Wilt (eds),System Criminality in International Law (Cambridge: Cambridge University Press, 2009)127^157, at 142.

51 While the unitarian system does not distinguish between different forms of participation atthe level of attribution, with the consequence that all persons causing the criminal resultare considered as principals/perpetrators (equivalency theory), the differentiated system(participation model) distinguishes between principals/perpetrators and accessories/secondary participants according to his or her relative contribution to the crime. Under thelater model, secondary participants (encouraging/abetting or assisting/aiding) can only beheld liable if the principal at least attempts the (unlawful) act; the relationship principal/accessory participant is legally/normatively determined. For the distinction between the twosystems see G. Fletcher, Rethinking Criminal Law (Oxford: OUP reprint 2002), at 634 et seq.;idem, Basic Concepts of Criminal Law (Oxford: OUP 1998), at 190 et seq.; see also E. vanSliedregt, The Criminal Responsibility of Individuals for Violations of International HumanitarianLaw (The Hague: TMC Asser Press, 2003), at 61^64; K. Ambos, Der Allgemeine Teil, supra note49, at 543 et seq.; idem, La Parte General, supra note 49, at 169 et seq.; idem, ‘Article 25:Individual criminal responsibility’, in O. Triffterer (ed.), Commentary on the Rome Statute of theICC (2nd edn., Mu« nchen: C.H. Beck, special print 2008) 743^770c, at 745^746. On the unitar-ian system recently T. Rotsch, ‘Einheitsta« terschaft’ statt Tatherrschaft (Tu« bingen: Mohr Siebeck2009), at 9 et seq.

52 Warrant of Arrest, Al Bashir (ICC-02/05-01/09), 4 March 2009; Decision on the Confirmation ofCharges, Lubanga (ICC-01/04-01/06(-803)), Pre-Trial Chamber I, 29 January 2007, x 342;Decision on the confirmation of charges, Katanga and Ngudjolo Chui (ICC-011 04-01107),Pre-Trial Chamber I, 30 September 2008, xx 480^486.

53 See for many C. Roxin, Strafrecht Allgemeiner Teil. Band II. (Mu« nchen: C.H. Beck, 2003), at 9et seq.

54 See among other cases from the ICTY: Judgment, Tadic¤ (IT-94-1-A), Appeals Chamber, 15 July1999, xx 195^229; Judgment, Kvoc› ka and others (IT-98-301/1), Appeals Chamber, 28 February2005, xx 79^119; Judgment, Krnojelac (IT-97-25-A), Appeals Chamber, 17 September 2003,xx 28^33, 84; Judgment,Vasiljevic¤ (IT-98-32-A), Appeals Chamber, 25 February 2004, xx 94^102;Judgment, Stakic¤ (IT-97-24-A), Appeals Chamber, 22 March 2006, xx 58^65; Judgment, Brd~anin(IT-99-36-A), Appeals Chamber, 3 April 2007, xx 410^432; Judgment, Krajis› nik (IT-00-39-T),Trial Chamber, 27 September 2006, xx 883^885. For the ICTR see Judgment, Ntakirutimana andNtakirutimana (ICTR-96-10-A), Appeals Chamber, 13 December 2004, xx 461^468. See also re-cently C. Farhang, ‘Point of No Return: JCE in Brdanin’, 23 Leiden Journal of International Law(2010), 137^164.

146 JICJ 9 (2011), 137^158

by Antonio C



ownloaded from


It is important to understand that the concept of Organisationsherrschaft, asoriginally developed by Claus Roxin,55 is intimately linked to the differentiatedsystem’s distinction between (primary) perpetrators and (secondary) partici-pants and, as to the former, direct, indirect and co-perpetrators consideringthe head of the power apparatus as an indirect perpetrator.56 While thisdoes not, as a matter of principle, exclude the establishment of anautonomous doctrine of Organisationsherrschaft with a view to the specialneeds of International Criminal Law ç I will return to this in a moment çany further development must not lose sight of this theoretical starting point.Organisationsherrschaft constitutes a unique form of indirect perpetration inthat it does not require, as the classical form of indirect perpetration, that thedirect perpetrator lacks (full) responsibility and is thus employed as a mere in-strument by the indirect perpetrator.57 This is typically the case when theperson who physically carries out the objective elements of the crimes lacksthe necessary intent to commit the crime, acts under mistake, duress or hasno capacity of culpability and the indirect perpetrator knowingly makes useof this circumstance. In so far it makes sense to speak of perpetration ‘bymeans’since the direct perpetrator is ‘a means’ in the hands of the indirect per-petrator, dominated and controlled by the latter’s superior will or knowledge.Yet, in the area of macro-criminality, i.e. systematic or mass criminalityplanned and organized by high level officials of the state or leaders of nonstate actors, the direct perpetrators usually act fully responsible, i.e. the trad-itional ‘perpetration by means’ theory cannot satisfactorily explain the indirectperpetrator’s control over the direct perpetrator. Instead, in this situation of a‘(indirect) perpetrator behind the (direct) perpetrator’ (Ta« ter hinter dem Ta« ter)the control of the former over the latter does not depend on his responsibilitybut on the power structure in which they operate.Despite the recent triumphal arrival of the Organisationsherrschaft at The

Hague58 one must not overlook that it is not at all ç even and especially inits mother country Germany ç uncontested. Several scholars reject its appli-cation in the Fujimori type cases and invoke instead the concepts ofco-perpetration59 or instigation60 as forms of attribution for the leaders of

55 See supra note 49.56 Crit. in this regard, however, Rotsch, supra note 51, at 316 et seq. (331) for whom the doctrine

breaks with the traditional structures of a differentiated system.57 Ambos, Art. 25, supra note 51, at 752.58 Katanga, supra note 52, x470 et seq. See also F. Jessberger and J. Geneuss,‘On the Application of a

Theory of Indirect Perpetration in Al Bashir: German Doctrine at The Hague?’ 6 Journal ofInternational Criminal Justice (‘JICJ’) (2008) 853^869.

59 H.H. Jescheck and T. Weigend, Lehrbuch des Strafrechts: Allgemeiner Teil (5th edn., Berlin:Duncker & Humblot, 1996), at 670; G. Jakobs, Strafrecht Allgemeiner Teil (2nd edn., Berlin: deGruyter, 1991), at 649; idem, ‘Zur Ta« terschaft des Angeklagten Alberto Fujimori Fujimori’,4 ZIS (2009) 572^575, at 573.

60 See R. Herzberg, ‘Mittelbare Ta« terschaft und Anstiftung informalen Organisationen’, inK. Amelung (ed.), Individuelle Verantwortung und Beteiligungsverha« ltnisse bei Strafta« tern inbu« rokratischen Organisationen des Staates, der Wirtschaft und der Gesellschaft (Sinzheim:Pro-Universitate, 2000) 33, at 39; with regard to Fujimori see Herzberg,‘Das Fujimori-Urteil: Zur


by Antonio C



ownloaded from


organized power structures. The traditional argument against indirect perpet-ration in cases of a fully responsible direct perpetrator (Ta« ter hinter dem Ta« ter)is a naturalistic one: it is inconceivable that the indirect perpetrator controlsthe will of a fully responsible direct perpetrator.61 Yet, from a normative per-spective, it can well be argued, as will be shown below (infra 2(d)) that thedomination of the organization prevails over the individual freedom of thedirect perpetrator.62 Be that as it may, the traditional system of individualresponsibility, as applied for ordinary criminality characterized by the indi-vidual commission of single crimes, must be adapted to the needs of interna-tional criminal law aiming at the development of a mixed system ofindividual-collective responsibility in which the criminal enterprise or an or-ganization as a whole serves as the entity upon which attribution of criminalresponsibility is based. In such a system the indirect perpetrator controls notso much the subordinates, but the apparatus of power63 and it is through thisorganizational control that he indirectly controls the conduct of the direct per-petrators. In other words, it is not so much the individual but the global act(the criminal enterprise) that constitutes the central object of attribution(Zurechnungsprinzip Gesamttat).64 While the indirect perpetrator as a perpetra-tor behind the direct perpetrator bears the responsibility primarily for the or-ganizational wrongdoing (Organisationsunrecht), the direct perpetrator isprimarily responsible for the individual wrongdoing (Individualunrecht).65

The doctrine of Organisationsherrschaft thus illustrates quite vividly the aim ofattributing the individual crimes committed within the framework of thesystem, organization or enterprise to the leaders and masterminds who bearthe greatest responsibility for controlling the organization or having set inmotion and supervised the criminal enterprise. Although these persons aregenerally far away from the actual execution of the criminal acts andare therefore normally considered indirect perpetrators or even accessoriesbefore the fact, they are, from a normative perspective, the main perpet-rators while the executors (the direct perpetrators) are merely secondaryparticipants in the implementation of the overall criminal enterprise. Theindividual criminal contributions of the participants must be assessed inthe light of their effect on the criminal plan or purpose pursued by the

Beteiligung des Befehlsgebers an denVerbrechen seines Machtapparats’, 4 ZIS (2009) 576^580,at 580; see also T. Rotsch, ‘Tatherrschaft kraft Organisationsherrschaft?‘, 112 Zeitschrift fu« r diegesamte Strafrechtswissenschaft (‘ZStW’) (2000) 518^562, at 562; crit. also Rotsch, supra note51, at 316 et seq., 372 et seq.

61 See for example Herzberg, Fujimori, supra note 60, at 578.62 See for a further development of the argument Ambos, supra note 50, at 144 et seq.63 See R. Bloy, ‘Grenzen der Ta« terschaft bei fremdha« ndiger Tatausfu« hrung’, 143 Goltdammers

Archiv fu« r Strafrecht (‘GA’) (1996) 424^442, at 441.64 On this new concept of attribution for collective criminality, see the fundamental work of

F. Dencker, Kausalita« t und Gesamttat (Berlin: Duncker & Humblot, 1996), at 125 et seq., 152et seq., 229, 253 et seq. and passim.

65 R. Bloy, ‘Besprechung von Wolfgang Schild, Ta« terschaft als Tatherrschaft’, 143 GA (1996)239^243, at 241.

148 JICJ 9 (2011), 137^158

by Antonio C



ownloaded from


criminal apparatus or organization. One can speak of a system of ‘organiza-tional domination of stages’ (stufenweise Organisationsherrschaft), where domin-ation requires, however, at least some form of control over part of theorganization. Thus, taking up the distinction between main perpetrators andsecondary participants, there are in fact three levels of participation: thefirst and highest level is composed of those (main) perpetrators who planand organize the criminal events as a whole and as such belong to the leader-ship level (Fu« hrungsta« ter); at the second level one can find those (still main)(co-) perpetrators of at least the mid-level of the hierarchy who exercisesome form of control over part of the organization (Organisationsta« ter);the third and last level consists of the low-rank direct perpetrators whomerely execute the crimes physically (Ausfu« hrungsta« ter).66 This three-prongedapproach is also reflected in the criminal organisation led by Fujimori. He im-plemented and supervised, together with his closest aides, an apparatusof power directed against supposed enemies and opponents of his governmentand supported by mid-level commanders who coordinated the crimes com-mitted by wilful direct executors. Thus, the doctrine of Organisationsherrschaftreflects adequately the ‘beaurocratic nature of mass atrocity, without goingso far as to ground liability on mere organizational membership.’67 Thatguarantees the respect for the principle of culpability as a cornerstone of crim-inal law.68

2. Requirements of Indirect Perpetration byVirtue of an Organized Power Structure

Following essentially the great German criminal law scholar Claus Roxin,69 theChambers of the Supreme Court set out five requirements for indirect perpetra-tion by virtue of an organized power structure:

(1) the existence of a hierarchical organization (as a general requirement);and ^ as specific requirements ^

(2) a responsible command (poder de mando, Befehlsgewalt) of the indirectperpetrator;

(3) the detachment of the organization from the law (desvinculacio¤ n del derecho,Rechtsgelo« stheit);

66 See also H. Vest, Genozid durch organisierte Machtapparate (Baden-Baden: Nomos, 2002), at239^242.

67 M. Osiel, Making Sense of Mass Atrocity (Cambridge: Cambridge University Press, 2009), at 95.68 On this principle in this context see K. Ambos, ‘Remarks on the general part of International

Criminal Law’, 4 JICJ (2006) 660^673, at 671.69 The Chambers further quoted the following authors: Kai Ambos, Christoph Grammer, Gerhard

Werle (from Germany), Javier Villa Stein, Percy Garc|¤ a Cavero, Jose¤ Urquizo Olaechea, Jose¤Hurtado Pozo, Felipe Villavicencio Terreros, Ivan Meini Me¤ ndez, Luis Alberto Arias Torres, Luisy Bramont Bramont Arias, Rau¤ l Pariona Arana,Yva¤ n Montoya Vivanco, Jose¤ Luis Castillo Alva(from Peru), Carlos Julio Lascano, Eduardo Andre¤ s Bertoni, Matilde Bruera (from Argentina),Carolina Bolea Bardo¤ n, Patricia Faraldo Cabana, Eva Ferna¤ ndez Iba¤ nez (from Spain) andEnrique Bacigalupo Zapater (Argentina/Spain).


by Antonio C



ownloaded from


(4) the interchangeability or replaceability (fungibilidad, Fungibilita« t) of thedirect perpetrator;

(5) the direct perpetrator’s (pre-)disposition (disposicio¤ n,Tatgeneigtheit) tocommit the act.70

(a) Hierarchical structure of the organization

Indirect perpetration by virtue of a hierarchical apparatus presupposes anorganized power structure with different roles assigned to its members thatallow the organization to function autonomously notwithstanding individuals’contributions.71 The assignment of roles through the superior level of the or-ganization occurs in a vertical direction by giving orders and therefore differsfrom the coordinated attribution of tasks, typical for co-perpetration.72 Theautomatic functioning of the apparatus accounts for the domination of the im-mediate perpetration and over the immediate perpetrators by indirect perpetra-tors at the leadership level. Yet, explicit orders by the leader are not necessary,as long as the acts of the direct perpetrators are committed within the contextof the objectives established and followed by the organization.73 Thus, it is ir-relevant ‘how’ and by whom the orders will be executed if only the master-mind(s) can be sure that someone will execute it somehow.74 Thishierarchical character of the organization is typical for state-sponsored vio-lence, but can also be found among non-state armed groups. In the Peruviancase this has been confirmed in the judgment against Abimael Guzma¤ n, theShining Path leader, who has been convicted to a life sentence for indirect per-petration of murder by virtue of SL.75 The ICC accepted it with regard toAfrican paramilitary groups in the Katanga/Ngudjolo Chui proceedings.76

Thus, it seems as if the theory is, contrary to Mark Osiel’s assertion,77 not onlyapplicable to rigidly formal state bureaucracies but also to less formalizednon-state groups.

(b) Responsible command and levels of hierarchy

Indirect perpetrators generally belong to the leadership level of the respectiveorganization. President Fujimori, who was not only vested with political

70 CSJ, supra note 33, xx726^727, CSJ, supra note 34, at 42^43.71 CSJ, supra note 33, x726.72 Ibid., at x 726; for a contrary view, see Jakobs, supra note 59, at 573.73 CSJ, supra note 33, x726.74 See Ambos, supra note 50, at 145.75 See CSJ, Segunda Sala Penal Transitoria, judgment of 26 November 2007, R.N. N8 5385^2006.76 Katanga, supra note 52, xx543^544.77 Osiel, supra note 67, at 100 (‘Roxin’s analysis assumes the existence of a rigidly formal bureau-

cracy of the sort contemplated byWebers’s famous ideal-type, developed from his understand-ing of the authoritarian Prussian army, which in the organizational chart perfectly mirrorsthe behavior of the people occupying positions within it’).

150 JICJ 9 (2011), 137^158

by Antonio C



ownloaded from


powers as President, but exercised at the same time command over the secur-ity forces, including the armed forces, the intelligence service and the police,is a paradigmatic example in this regard. The mastermind’s orders must not ne-cessarily, as already said before, be issued formally through instructions; theymay also be given, as in the case of SL leader Abimael Guzma¤ n, informally inan encoded manner or they may be taken for granted.78

Another issue dealt with by the Special Chamber is the delimitation of indir-ect perpetration from co-perpetration along levels of hierarchy with regard tomid rank commanders.79 In fact, the issue of up to what level of hierarchyone can assume an indirect perpetrator’s control over the act by virtue of ahierarchical organization has not been thoroughly analyzed, let alone besolved, by the doctrine.80 The judgment follows a broader understanding of in-direct perpetration ç similar to Roxin’s concept and the German case law çapplying it to mid rank officials who, apart from receiving orders themselvesfrom the leadership level, also possess certain powers to issue orders to theirsubordinates.81 In such cases, the mid-level commander cannot exclude his re-sponsibility with the argument that if he had not acted, someone else wouldhave forwarded the order since, first of all, such a hypothetical course of caus-ality does not do away with the in fact causal order in the first place and, sec-ondly, such a mid-ranking commander ç think of Adolf Eichmann ç is noteasily replaceable.82 According to the Supreme Court, indirect perpetration istherefore not limited to the leadership level with the possible consequence ofa whole chain of indirect perpetrators in a hierarchical organization.83

However, while the possibility of an Organisationsherrschaft at a hierarchicallevel below the leadership level has been recognized with the Eichmanncase,84 one can hardly deny the fact that absolute control through and/orover an organized power apparatus can only be exercised at the leadershiplevel, i.e. at the government level in the case of state violence.85 In addition,the leadership level represents the state in a particular way and as such bearsthe overall responsibility for possible violations of fundamental human rights;all other power is derived from this highest authority and thus in its exerciseattributable to it.86 Only the leader’s power and authority can neither beblocked nor disturbed in any way from above. As a consequence, control overthe act by virtue of Organisationsherrschaft can only vest in those persons,

78 CSJ, supra note 33, xx730, 732.79 Ibid., xx730^731.80 See Ambos, supra note 50, at 151 et seq.81 CSJ, supra note 33, x73182 Cf. CSJ, supra note 33, x731.83 Ibid.; see also C. Roxin, ‘Probleme von Ta« terschaft und Teilnahme bei der organisierten

Kriminalita« t’, in E. Samson et al. (eds), Festschrift fu« r Gerald Gru« nwald (Baden-Baden: Nomos,1999) 549^561, at 556.

84 See Ambos, supra note 50, at 152^153.85 See H.Vest,‘Humanita« tsverbrechen ^ Herausforderung fu« r das individuelle Strafrecht?’ 13 ZStW

(2001) 457^498, at 493 et seq.; Ambos, supra note 50, at 153.86 Ambos, supra note 50, at 153.


by Antonio C



ownloaded from


whose orders and instructions cannot be revoked or cancelled without any fur-ther ado, i.e. those, who, in this sense, can rule and control without any inter-ference whatsoever.87 This only applies to the leadership level of the formallyestablished government and, in exceptional cases, to the top hierarchy of themilitary, police and intelligence forces. Obviously, their ability to exerciseOrganisationsherrschaft is self-evident when they are ruling themselves orform part of the government. As to the Fujimori regime we can say that theleadership level was composed of Fujimori, as President and commander ofthe armed forces, Montesinos and Salazar Monroe, as de facto and de jurechiefs of the national intelligence service SIN, and Bari Hermoza Rios as com-mander of the general staff of the armed forces.In contrast, perpetrators who do not belong to the leadership but only to the

mid-level of the organization exercise at most control over their own subordin-ates within the apparatus. In other words, they do not exercise control overthe apparatus as a whole but only over some members of it.While one may con-sider them as indirect perpetrators at least with regard to the events takingplace under their (partial) control, their dependence on and their integrationinto the organization as a whole rather militates for their responsibility asco-perpetrators on the basis of a functional division of labor.88 Roxin, however,rejects co-perpetration for principled reasons since he neither sees a jointlyagreed plan nor a division of labour between the mid-level commander andthe direct perpetrator(s) but rather, due to their superior-subordinate relation-ship, a unilateral imposition of orders by the superior.89 This is not convincingfor the following three lines of reasoning. First, as to the joint decision orcommon plan, an informal consensus or agreement of the persons involvedwould suffice.While Roxin is right, in that the superior giving orders and thesubordinate executing them within the framework of a criminal apparatus, asa rule, neither know each other nor take joint decisions, this is not the decisiveissue. For an informal agreement it is enough that the direct perpetratormakes clear, through his belonging to the criminal organization, that heagrees with the organization’s policies as set out by the leaders.90 Secondly,co-perpetration through functional division of labour is not to be interpretedtoo strictly as to excluding any act of preparation of the mid-level command-er.91 The famous gang leader, to take a classical example form ordinary crimin-ality, does not dirty his hands but lets his gang members do the dirty work.Functional control over the acts means nothing else than a division of labourbetween the persons involved. In the case of a mid-level commander and thedirect perpetrator(s) this division consists of the mid-level commander’s order-ing, preparing or planning of the act and the subordinate(s) executing it. Boththe contribution of the mid-level commander and of the direct perpetrator(s)

87 Ibid., at 154. See also Jessberger and Geneuss, supra note 58, at 861.88 Ambos, DerAllgemeine Teil, supra note 49, at 604; idem, La Parte General, supra note 49, at 232.

The Lubanga PTC follows this concept of co-perpetration, see Lubanga, supra note 52, x342.89 For these common arguments against co-perpetration, see Roxin, supra note 83, at 552 et seq.90 Ibid., at 605.91 Ibid.

152 JICJ 9 (2011), 137^158

by Antonio C



ownloaded from


is indispensable for the commission of the crime, so that both can be con-sidered to control the act equally. Thirdly, the ç in principle correct çstructural difference between vertical indirect perpetration and horizontalco-perpetration does not per se entail a reliable delimitation in the borderlinecases of macro-criminality and complex organizations. In fact, the argumentof a structural difference is only valid in cases in which the vertical relation-ship between the indirect and the direct perpetrator is not disturbed, as in thecase of mid-level perpetrators, by the existence of a further relationship ofthis indirect perpetrator to his superior.92 It is worthwhile mentioning in thiscontext that ç prior to the conviction of Fujimori ç the Lima Superior Courthad convicted Salazar Monroe as an indirect perpetrator of the forced dis-appearances in the La Cantuta case arguing that he formed part of the leader-ship level at the top of the governmental apparatus.93 Yet, in the samejudgment several mid level commanders of the armed forces and the membersof the Colina group were convicted as co-perpetrators.94 In sum, the gist ofthe difficult delimitation between indirect perpetration and co-perpetration inthe case of mid level members of an organizational power structure comesdown to the question whether one is rather prepared to accept a deficiency ofleadership on the part of the indirect perpetrator or an unequal ranking ofthe co-perpetrators. In my view, the latter position is more convincing since in-direct perpetration rests on absolute control over the act and as such cannotbe reconciled with only partial or even a lack of control because of a lack of un-disturbed leadership.95

(c) Detachment of the organization from the law

The second specific prerequisite recognized by the Supreme Court is the de-tachment of the organization from the law.96 Accordingly the organization, beit part of the state or a non-state actor, must operate outside the national andinternational law. The detachment must be of a structural nature, i.e., isolatedviolations of human rights ç not forming part of a policy ç do not suffice.In the case of state-sponsored mass violence it is not necessary that thewhole state apparatus as such functions beyond the limits of national or inter-national law; it suffices that part of the institutions, e.g. the security forces,function like a ‘State within the State’and implement a policy of human rightsviolations under the leadership of high-rank officials. The detachment fromthe rule of law can occur instantaneously through the decision of the leader-ship level to abandon the rule of law, or gradually through the continuous

92 Ibid., at 606.93 Corte Superior de Justicia de Lima, Primera Sala Penal Especial, judgment of 8 April 2008, Exp.

No. Av. 03-2003-18 SPE/CSJLI, pp. 98^110, available at http://www.unifr.ch/ddp1/derechopenal/jurisprudencia/j_20080616_38.pdf (visited 5 June 2010).

94 Ibid., at x154.95 Ambos, DerAllgemeine Teil, supra note 49, at 606.96 CSJ, supra note 33, xx733^736, CSJ, supra note 34, at 48^49.


by Antonio C



ownloaded from


deterioration of the democratic structures of the state.97 The degree of culpabil-ity is higher when the organization acts under the pretext of formal legalityand hides its real intentions behind a fa� ade of legality to avoid that legal stepsare being taken against it.98 The Supreme Court convicted Fujimori as indirectperpetrator for crimes committed by the Colina group before the self-coup(Barrios Altos case) and after the coup (La Cantuta case). Thus, the self-coupas such is not considered as a decisive circumstance to assume that Fujimoriled an extralegal hierarchical organization.The underlying premise of this requirement is that in a normal situation, i.e.

where an organized power structure acts in accordance with the law, an iso-lated criminal order from a superior may usually not be sufficient to cause asubordinate to commit a crime for it can be expected that this person willresist or at least hesitate to comply with an unlawful order, not least becauseof the possible consequences he may have to face. If notwithstanding this, thesubordinates executes the unlawful order, the crime will only be committed‘against the organized power structure’, but not ‘through the organized powerstructure’.99 It can therefore not be attributed to the organization as such butonly to particular superiors within the organization who would thus be con-sidered as mere instigators. In any case, it is not so decisive whether the organ-ization acts in not just exceptional situations beyond the law, but whether itsleaders can control it at their will so that its members look like anonymousand interchangeable cogs in the machinery of a criminal apparatus withoutthe capacity to influence the outcome of the events. For this very reason, thisrequirement cannot be considered a necessary precondition for indirect per-petration.100 The ICC seems to hold the same view since it has not (yet) men-tioned this requirement in its decisions on indirect perpetration.

(d) Interchangeability of the direct perpetrator

The third requirement consists of the easy interchangeability or replaceabilityof the direct perpetrator(s).101 This possibility compensates for the lack of con-trol over a fully responsible direct perpetrator who after all may, at any time,decide to abandon the criminal plan. Only if the organization produces a suffi-cient number of potential willing and interchangeable executors one mayspeak of a control by virtue of the organization, i.e., by way of the interchange-able executors. Thus, the indirect perpetrator does not so much or, from a nat-uralistic perspective, not at all dominate the direct perpetrator (as he is fully

97 CSJ, supra note 33, x735.98 Ibid., x731.99 Ola¤ solo, supra note 69, at 120.100 See for more details Ambos, Der Allgemeine Teil, supra note 49, at 606^611; idem, La Parte

General, supra note 49, at 234^238; against the relevance of this requirement with regard tothe direct perpetrator, see also I. Meini, ‘La autor|¤ a mediata por dominio de la organizacio¤ nen el caso Fujimori’, 4 ZIS (2009) 603^608, at 606.

101 CSJ, supra note 33, xx737^739; CSJ, supra note 34, at 49^51.

154 JICJ 9 (2011), 137^158

by Antonio C



ownloaded from


responsible!) but rather the collective of direct perpetrators part of the criminalorganization. These direct perpetrators are nothing more than interchangeablecogs in the machine of the organizational power apparatus, which makestheir apparent freedom a naturalistic date which from a normative perspectiveis of no importance.102

With a view to the precise understanding of the concept of interchangeabil-ity, the Supreme Court’s Trial Chamber distinguishes between negative andpositive interchangeability.103 The first one is understood as the classical formand refers to situations in which a potential denial or abstention of the directperpetrator to commit the crime could be immediately compensated by an-other direct perpetrator, so that the commission of the crime cannot be fru-strated. The positive interchangeability refers to the situation in which theleaders select from a plurality of potential perpetrators the most qualified onesfor the execution of the criminal act, so that any failure can be excluded. Asto the Fujimori regime this means that the fact that the Colina group wasonly composed of about 30 persons ç compared to the thousands of directperpetrators of the Holocaust or the Rwanda genocide ç does not exclude theinterchangeability of the direct perpetrators since the regime’s leadership previ-ously selected the most qualified persons and trained them for the specificacts. For this reason, a failure in the commission of the crimes could practic-ally be excluded.In any case, the criterion of interchangeability cannot satisfactorily compen-

sate the lack of factual control over fully responsible direct perpetrators andthus cannot satisfactorily explain the theory of Organisationsherrschaft. Froma purely factual, naturalistic or empirical perspective there are always caseswhere direct perpetrators cannot be replaced without frustrating the comple-tion of the crime as a whole. If, for example, a border guard or a whole borderpatrol at the East German Border had refused to shoot at an East-German citi-zen who wanted to flee to the West, there would not have been other guardsimmediately available to hinder this particular citizen from jumping over thewall. Thus, the doctrine can only be explained through a normative theorywhich transfers the structure of offences requiring a certain position of duty(Garantenstellung) on the part of the perpetrator (the so called duty-offences,Pflichtdelikte) to the doctrine of control over the act and operates with a mater-ial concept of freedom.104 Accordingly, the notion of Pflichtdelikte, which re-quire from a person the protection of others or the prevention of therealization of a risk, can be used to attribute to the Head of State criminal re-sponsibility as a principal if he wilfully does not comply with his special duty

102 Ambos, supra note 50, at 144.103 CSJ, supra note 33, x738.104 See for a more detailed explanation Ambos, supra note 50, at 148^151; idem, Der Allgemeine

Teil, supra note 49, at 599^600, U. Murmann, Die Nebenta« terschaft im Strafrecht: ein Beitrag zueiner personalen Tatherrschaftslehre (Berlin: Dunker & Humblot, 1993), at 60 et seq.; idem,‘Tatherrschaft und Weisungsmacht’, GA (1996), 275 et seq.; idem, ‘x 25’, in H. Satzger,B. Schmitt and G. Widmaier (eds), Strafgesetzbuch Kommentar (Ko« ln: Carl Heymanns Verlag,2009), marginal number 28.


by Antonio C



ownloaded from


to protect his people, be it by action, acquiescence or omission.105 In this per-spective the state appears ç as guarantor of basic rights with a resultingduty to protect its citizens ç in a specific position of responsibility vis-a' -visits citizens and thus possesses a specific power of violation (Verletzungsmacht)towards these. The state breaches its duty to protect by illegally ordering thedirect perpetrator to attack and injure a particular person. The relationship be-tween the state and its citizens is therefore comparable with the situation of aguarantor position in crimes of omission. The guarantor is in this case liablesince he does not comply with his particular duty to protect flowing from hisposition as guarantor. In the case of illegal state orders, the ç at least from anormative perspective ç dependence of the citizen on the state establishesthe state’s control over the act. The state exercises this control by ordering themediator of the act, a citizen, to injure the victim, another citizen. The attribu-tion of the victim’s injury to the leaders of the state organization cannot onlybe considered from the perspective of relationship between the leaders andthe mediator of the act, but must also take into account the victim and the spe-cial state’s duty vis-a' -vis its citizen, which provides in all those cases a norma-tive foundation of the control over the act where the notion ofinterchangeability fails for empirical reasons. With this approach theOrganisationsherrschaft is not abandoned, but reinforced by normative,value-based considerations. This normative explanation does not substitute,but complement the naturalistic, empirical perspective.The normative considerations also explain why the group of indirect perpet-

rators should be limited to the leadership-level. For only the leaders ç as rep-resentatives of the state ç bear the special duty to protect the state’s citizens.Clearly, though, this normative explanation cannot apply in the case ofnon-state groups, where there is no comparable relationship with the citizens;therefore, in this case, naturalistic considerations must prevail. Indeed, alongthese lines, it is compelling that the ICC held that with regard to non-statearmed groups ‘attributes of the organization ç other than the replaceabilityof subordinates ç may also enable automatic compliance with the senior au-thority’s orders. An alternative means by which a leader secures automaticcompliance via his control of the apparatus may be through intensive, strict,and violent training regimes’ using child soldiers.106 Notwithstanding, in thisspecific case of the forced recruitment of child soldiers, it would also be pos-sible to consider the commanders as indirect perpetrators since the child sol-diers cannot act criminally responsible due to their age and thus are to beconsidered as mere tools for the commission of the crimes of their adultsuperiors.

105 See Jakobs, supra note 59, at 574.106 Katanga, supra note 52, x518.

156 JICJ 9 (2011), 137^158

by Antonio C



ownloaded from


(e) Disposition of the direct perpetrator to commit the act

This is an additional requirement which goes beyond the traditional require-ments ((a)-(d)) which thus far have been considered sufficient to establish anOrganisationsherrschaft. It has its origin in Schro« ders theory about indirect per-petration107 and has previously been referred to in the German judgments con-cerning the killings at the German wall (Mauerschu« tzenfa« lle).108 The PeruvianSupreme Court also adopts this requirement,109 but it is controversial whetherit is indispensable or not. Roxin himself is not completely clear on the matter.While he recently admitted it as an independent criterion ofOrganisationsherrschaft,110 thus adding to his original criterion of control overthe result (Erfolgsherrschaft) one of control by virtue of superior knowledge(Wissensherrschaft),111he then seems to paddle back arguing that the direct per-petrator’s disposition can normally be deduced from the other three specific re-quirements, since the superior’s control over the hierarchically structuredorganization and the direct perpetrator’s replaceability limit the later one’s cap-acity for independent manoeuvre.112 In a similar sense this requirement is as-signed only a symbolic value considering it as mere characteristic ofOrganisationsherrschaft.113 In any case, a typical manifestation of the direct per-petrator’s disposition regularly consists in his motivation to commit thecrimes not for personal reasons, but for and as part of the organization whichexercises control over him.

3. ConclusionsThe various trials against former President Fujimori reflect the strong links be-tween corruption in an authoritarian regime and serious human rights viola-tions, especially where a state is dealing with the threat of terrorism and thestate’s democratic and legal framework is unstable and underdeveloped.114

Future prosecutions against Fujimori for other crimes are not excluded, but

107 F.-C. Schoeder, Der Ta« ter hinter dem Ta« ter: ein Beitrag zur Lehre von der mittelbaren Ta« terschaft(Berlin: Dunker & Humblot, 1965), at 221^222.

108 BGHSt 40, 218; see also F.C. Schroeder, ‘Tatbereitschaft gegen Fungibilita« t’, 4 ZIS (2009)569^571, at 569, 570.

109 CSJ, supra note 33, xx740^741; CSJ, supra note 34, at 51^52.110 C. Roxin, ‘Organisationsherrschaft und Tatentschlossenheit’, in A. Hoyer et al. (eds), Festschrift

fu« r Friedrich-Christian Schroeder (Heidelberg: C.F. Mu« ller, 2006) 387^400, at 397^398.111 See crit. Rotsch, supra note 51, at 390 and passim.112 C. Roxin, ‘Bemerkungen zum Fujimori-Urteil des Obersten Gerichtshofs in Peru’, 4 ZIS (2009),

565^571, at 567 (‘ :::nicht ::: selbsta« ndige Voraussetzung ::: ’).113 See for example R. Pariona, ‘La autor|¤ a mediata por organizacio¤ n en la Sentencia contra

Fujimori’, 4 ZIS (2009) 609^614, at 614 (‘so¤ lo caracter|¤ sticas:::’; emphasis in the original).114 See generally N. Roht-Arriaza, ‘Prosecutions of Heads of State in Latin America’, in Lutz and

Reiger (eds), supra note 8, at 46^76; V. Carranza, ‘Plunder and Pain: Should TransitionalJustice Engage with Corruption and Economic Crimes?’ 2 International Journal of TransitionalJustice (2008) 310^330.


by Antonio C



ownloaded from


would have to be approved by the Chilean Supreme Court because of the spe-cialty principle governing extradition law. In any case, Fujimori’s convictionon the basis of the ‘control over the act’ theory and the theory ofOrganisationsherrschaft shows once more that this theory must seriously beconsidered as a tool in the prosecution of high-level criminals. On the otherhand, Vladimiro Montesinos and General Hermoza Rio¤ s, commander of theArmed Forces, have not been convicted so far for serious human rights viola-tions committed during the Fujimori regime. This shows that criminal prosecu-tions remain highly selective. Last but not least, the judgments also showhow basic principles of criminal law and the exigencies of internationalhuman rights law regarding the prosecution of international core crimes canbe reconciled and convincing results can be achieved even if internationalcrimes have not been fully incorporated into domestic criminal law.

158 JICJ 9 (2011), 137^158

by Antonio C



ownloaded from


the fujimori judgment - department-ambos.uni-goettingen.de · gent groups or opposed to the...

Documents