final project

TheLegalLegacyofPinochet.pdf

Home >Political Science homework help >final project

Chapter 7

The Legal Legacy of Pinochet: Universal Jurisdiction and Its Discontents

The spectacle of Chile's once-powerful General Pinochet under arrest, and the favorable initial decisions from the House of Lords in Britain and the Audiencia Nacional in Spain, kindled new hopes in human rights advocates around the world. The courts of many countries were closed to investigations or lawsuits involving abuses by the local military or police, due to formal amnesty laws or informal threats, bribes, or other pressures. Maybe the Pinochet path was a viable alternative. Trans- national prosecutions in the courts of other states, hitherto considered legally possible but more than a little fanciful, started looking much more interesting.

Cases began to arrive before the Spanish and Belgian courts because they had the most well-known laws allowing extraterritorial jurisdiction. Some were easy to dispose of. A group of Cubans and Chilean Pinochet supporters called onJudge Garzon to investigate Cuban president Fidel Castro, but the court quickly found that a head of state was immune from suit in Spanish courts while in office. Cases against Morocco's King Hassan II and Theodoro Nguema of Equatorial Guinea met the same fate. A case against former Communist Party head Santiago Carrillo was dismissed as well.

Other allegations were more promising. Four cases in particular before European courts seemed to be natural extensions of the Pino- chet precedent. One involved the massacre of hundreds of thousands of Guatemalans, mostly Mayan indigenous people. Two others involved Africans: Hisscne Habre, former dictator of Chad, and Ab- dulaye Yerodia Ndombasi, a high-ranking official in the Congolese government. A fourth involved Desire Bouterse, former strongman of Suriname. Each would, in its own way, come up against the political and legal limits of prosecution based on universal jurisdiction. Together, they mark at least a temporary retreat from the initial euphoria of 1998.

Roht-Arriaza, Naomi. The Pinochet Effect : Transnational Justice in the Age of Human Rights, University of Pennsylvania Press, 2010. ProQuest Ebook Central, http://ebookcentral.proquest.com/lib/apus/detail.action?docID=3441496. Created from apus on 2018-07-26 20:51:08.

C op

yr ig

ht ©

2 01

0. U

ni ve

rs ity

o f P

en ns

yl va

ni a

P re

ss . A

ll rig

ht s

re se

rv ed

The Legal Legacy of Pinochet 171

Guatemala: Subsidiary Jurisdiction and the Narrowing of the Spanish Forum

Rigoberta Menchu, a Qu'iche Indian woman, won the Nobel Peace Prize in 1992 for her advocacy of indigenous peoples' rights. Menchu saw her own family decimated by Guatemala's counterinsurgency war, which left over 200,000 dead or missing from 1960 to 1996. According to the UN- sponsored Commission on Historical Clarification (CEH, after the Span- ish acronym) that investigated the atrocities, the army and associated paramilitary groups targeted indigenous communities as well as student, labor and religious figures considered to support leftist guerrillas. More than 600 villages were completely destroyed, and some 40,000 people disappeared. 1 The figures made the Chilean and even the Argentine cases pale in comparison. Menchu began exploring the options in Spain. A number of people counseled patience. "Let's not overload the boat," Joan Garces advised, at least while the outcome of the Pinochet extradition was pending. Others insisted that, on the contrary, the time was ripe to push the doors of the Spanish courts open even further.

In December 1999, Rigoberta Menchu, together with Guatemalan groups of family members of those killed, Spanish labor unions, and sol- idarity groups, filed a complaint before the Audiencia Nacional charg- ing eight people with genocide, terrorism, and torture." The suspects include five generals, two police chiefs, and a colonel, among them for- mer presidents and defense and interior ministers. The most notorious is General Efrain Rios Montt, who took power in a coup in March 1982 and over the next eighteen months implemented a scorched earth pol- icy. Rios Montt continued to dominate Guatemalan politics for years, as president of Congress and a losing candidate in the 2003 presidential elections. Like Pinochet, he remains defiantly unapologetic about wag- ing what he considers to be a successful war on communist subversion. Another suspect, Donaldo Alvarez Ruiz, former interior minister and head of the police, is in hiding outside the country (many say in the U. S.), while a third lives in Venezuela. The rest are free in Guatemala; one has parleyed his expertise into a private security business, while another served as mayor of a small town.

The complaint focused on a number of incidents as exemplars of genocide. Rigoberta Menchus own family history was one. Her mother and brother were publicly tortured and killed by the army. Her father was burned alive when he, together with other peasant leaders, sat in at the Spanish Embassy in 1980 demanding an end to repression in the Mayan highland communities. Over the ambassador's protests, police shot incendiary bombs into the embassy, starting a fire that killed 39

C op

yr ig

ht ©

2 01

0. U

ni ve

rs ity

o f P

en ns

yl va

ni a

P re

ss . A

ll rig

ht s

re se

rv ed

172 Chapter 7

people, including eight members of the embassy staff. The Spanish Embassy fire provided one tie to Spain, but there were others. Three Spanish priests, Faustino Villanueva, Jose Marla Gran Cuerra, and Juan Alonso Fernandez, were assassinated in Guatemala, and a fourth, Carlos Perez Alonso, disappeared. Additional complainants referenced several army massacres and individual murder and disappearance cases.

The Guatemalan complaint had several strengths. The magnitude and drama of the massacres in Guatemala was undeniable. The underly- ing acts, at least in part, concerned Spain, and family members of the slain Spanish priests were complainants. A wide variety of Spanish orga- nizations, especially the leftist union confederation Comisiones Obreras, had signed on along with Guatemalan groups-nineteen com- plainants in all. As a matter of legal theory, it was much simpler to charge genocide in the Guatemalan case than in the Southern Cone cases, as the facts showed a clear pattern of targeting ethnically distinct people and the UN Commission's findings on genocide and on the mili- tary's responsibility had already paved the way. The existing precedents on Chile and Argentina also seemed to indicate that the Spanish courts would be receptive to the claims.

Indeed, on March 27, 2000, Judge Guillermo Ruiz Polanco accepted the complaint and agreed to open an investigation." In his ruling, the judge noted the connections to Spain and found that the accused were directly involved in the crimes charged. "The events reported clearly show the appearance of genocide. And that is sufficient for now," said the judge's ruling. The genocide charges included both the targeting of ethnic Mayans and the intended elimination of a part of the "national" group due to its perceived ideology. The judge, like Judge Garzon before him, found that Spanish jurisdiction was appropriate because the local courts had not acted. While Guatemala had "original territorial jurisdiction," it was not exclusive.

In the absence of the honorable and effective exercise of [territorial] jurisdic- tion, it must be replaced by courts-such as Spain's-that uphold the universal prosecution of crimes against human rights.... There is no reason to presume that the petition for justice before the Spanish courts was made with caprice or frivolity.... Despite the big words, the reticence of states in the issues under consideration forces the victims of crimes against humanity, their heirs, their families and their legal representatives to bear a costly international legal pil- grimage due to the passivity-if not the complicity-of the territorial judges who should assume these cases in the first place.

Once Ruiz Polanco found jurisdiction, evidence gathering began. Documentary evidence began to pour in-newspaper articles, reports of exhumations of mass graves, death certificates. But by mid-2000 the

C op

yr ig

ht ©

2 01

0. U

ni ve

rs ity

o f P

en ns

yl va

ni a

P re

ss . A

ll rig

ht s

re se

rv ed

The Legal Legacy of Pinochet 173

political and legal climate had chilled noticeably in Spain. Pinochet had gone home and the Argentine and Chilean cases, while still open, assumed a lower profile. Bombings and assassinations carried out by the Basque separatist group ETA had nerves on edge. Furthermore, the International Criminal Court (ICC) now looked as if it might become a reality in the near future. Even though under the Court's statute it would not be able to consider pre-2000 crimes and thus would never consider the Guatemalan case, still, public opinion attributed more power to the nascent court than it actually had. The local press and pun- dits worried publicly about a deluge of cases turning the Spanish courts into a "mini-ICC" that would solve all the world's problems except Spain's. Ruiz Polanco began to reflect some of these hesitations. The judge demanded a hefty bond to admit a Guatemalan co-complainant. He refused to add more potential defendants and held off on hearing more than a few live witnesses.

Despite its strengths, Guatemala was a harder case in many ways than the earlier Southern Cone ones. Far less was known internationally about the atrocities in Guatemala, and Rios Montt had never assumed the iconic status of a Pinochet. Europeans had protested u.S. interven- tion in Central America, but never in the numbers nor with the intensity that greeted the 1973 coup in Chile. There were far fewer Guatemalans than Argentines and Chileans living in exile in Spain and throughout Europe. Guatemalans who did live in Spain tended to be less educated and far less integrated into Spanish society, with less ability to move the necessary levers of politics and public opinion. In particular, there were no Guatemalan-related lawyers and activists living in Spain who could serve the "bridging" roles thatJoan Garces, Gregorio Dionis, and Carlos Slepoy had played in the Southern Cone cases. There was no one who had at once the intimate knowledge of recent Guatemalan history, the personal passion to pursue justice, and the ability to navigate Spanish law, politics, and public relations. Instead, the legal team had Guatema- lan human rights lawyers based in Guatemala, Spanish and Argentine labor and criminal law specialists with limited knowledge of Guatemala (including eventually Slepoy and Galan from the Argentine legal team), and a non-lawyer liaison from the Menchu Foundation trying to knit it all together. The case got much less publicity, and while a sizable num- ber of long-standing Spanish solidarity groups signed on as acci6n popu- lar, the widespread support from Parliament, the press, and intellectuals that characterized the earlier cases was missing.

Guatemalan realities counseled caution as well. The transition from military rule there was far from complete and human rights activists and lawyers were continually threatened and harassed. The posture of the Guatemalan courts was murky. The 1996 Guatemalan Law of National

C op

yr ig

ht ©

2 01

0. U

ni ve

rs ity

o f P

en ns

yl va

ni a

P re

ss . A

ll rig

ht s

re se

rv ed

174 Chapter 7

Reconciliation granted amnesty to those involved in the war, but, in response to UN urging, it excluded cases of genocide, torture, and dis- appearances (but not necessarily massacres)." The local courts up to that time had refused to apply the law in cases involving human rights viola- tions, so there was no formal impediment to prosecution. Rather, courts had found ways to fail to investigate, to leave cases in legal limbo and to release defendants on obscure technicalities. The few convictions for human rights-related crimes were of the rank and file, not the higher- ups. Judges, prosecutors, and victims' lawyers were routinely threatened, and the system was notoriously corrupt. It was a case of de facto rather than de jure impunity.

Local groups pursued cases before the domestic courts despite the obstacles, including a case charging genocide against the same set of defendants. There was at first little coordination between the groups spearheading local actions and those focusing on the Spanish case, although it improved over time.

According to some observers, Menchu had launched her complaint precipitously, both to influence upcoming elections in Guatemala and because she feared that the Spanish forum might close. The complaint itself included vastly disparate incidents with no tight common thread among them, and relied heavily on the CEH's findings and the church- authored REHMI report on human rights violations. Any investigating magistrate would have been daunted by the thought of managing this case. Despite these difficulties, the Guatemalan complainants were con- fident that the Spanish courts would support them just as they had sup- ported their Southern Cone counterparts.

Just as he had in the Southern Cone cases, the public prosecutor objected to the Spanish court'sjurisdiction. Some of his arguments were identical to the earlier ones: the facts didn't support charges of geno- cide or terrorism; Spanish courts had no extraterritorial jurisdiction; this was retroactive application of law; a domestic amnesty law precluded prosecution. Others were specific to the Guatemalan situation: a diplo- matic settlement between Spain and Guatemala meant that the Spanish Embassy massacre was not actionable; the 1996 peace accords provided for sufficient domestic truth seeking. On May 4, 2000, prosecutor Pedro Rubira appealed Ruiz Polanco's acceptance oljurisdiction."

Over the summer of 2000, the Audiencia Nacional, acting as an appeals court, agreed that given the importance of the case they would hear it in full rather than sit as a panel. The public hearing, set for November 30, carried over to the next day. Thirteen days later, just before the Christmas holidays, the court ruled against the Guatemalans." It held that "for the moment" Spanish courts had no jurisdiction over the alleged crimes, and that the case should be closed. The judges gave

C op

yr ig

ht ©

2 01

0. U

ni ve

rs ity

o f P

en ns

yl va

ni a

P re

ss . A

ll rig

ht s

re se

rv ed

The Legal Legacy of Pinochet 175

two reasons. First, while the Spanish courts could consider genocide and terrorism committed elsewhere by non-Spaniards, any such inquiry had to be subsidiary to the state where the crimes took place. Other national courts could act only if there were clear legal impediments to prosecu- tion, or ifjudges were "subject to pressure from official or de facto pow- ers that create a climate of intimidation or fear making it impossible to carry out the judicial function with the serenity and impartiality required." Second, since the CEH had only published its report in 1999 and the Law of National Reconciliation permitted genocide prosecu- tions, there was insufficient evidence in the record that the Guatemalan courts were not able or willing to do the job themselves. Therefore, Spanish courts should stay out of the case. It was not difficult to see the decision as a signal that the courts were uninterested in broadening the opening provided by the Southern Cone cases. Others alleging geno- cide need not apply.

The decision sparked a vigorous debate among the lawyers and com- plainants groups. Now what? There were two options: appeal or wait and refile the case. Appeal was to the Spanish Supreme Court, widely consid- ered more conservative than the Audiencia Nacional. Pursuing an appeal was risky. Not only could the court close off the possibility of ever refiling the Guatemalan case in Spain, but it could overturn the existing jurisprudence of the Argentine and Chilean cases, destroy the pending cases against Scilingo and Cavallo, and close off many if not all future possibilities of access. For some people, an appeal would give the public prosecutor, and by extension the Aznar government, just the opportu- nity they had been waiting for. Far better, these people said, to docu- ment better both the inactivity of the Guatemalan courts and the exact contours of the case, and then go back to the investigating judge, who was obviously sympathetic, and start again. Others argued that in Guate- mala this would be seen as an admission of defeat, and would discourage those trying to break through the wall of impunity at home. Moreover, it would leave intact a crabbed interpretation of the Spanish law on extraterritorial prosecution of genocide, one that ignored the text of Spanish law. An appeal would also open up the possibility of further appeals to the Constitutional Court or, eventually, the European Court of Human Rights, where a more favorable decision might be possible. It was important to push as far as possible, and not give up so easily, said these advocates. After a series of personal and e-mail consultations among the lawyers, experts, and complainant groups, the latter position prevailed and an appeal was filed.

The Supreme Court, recognizing the importance of the issue, agreed to hear the case en bane (as a whole), and heard oral argument in July 2002. A decision expected in September 2002 was put off indefinitely as

C op

yr ig

ht ©

2 01

0. U

ni ve

rs ity

o f P

en ns

yl va

ni a

P re

ss . A

ll rig

ht s

re se

rv ed

176 Chapter 7

rumors of internal disagreement swirled. On February 25, 2003, over two years after the Audiencia Nacional decision, the fears of those who doubted the wisdom of an appeal came partially truc.? The Spanish Supreme Court, by a vote of 8-7, overturned in part the Audiencia Nac- ional decision and gutted Spain's universal jurisdiction law. The major- ity held, in short, that only cases with a clear tie to Spain could proceed. The case was reopened and remanded to pursue investigations into the possible torture of Spanish citizens in the 1980 Embassy massacre, and the torture of the four Spanish priests killed by the military in 1980 through 1982. All the genocide and terrorism charges and the torture charges against non-Spaniards were dismissed. While the Spanish courts remained open for cases involving victims of Spanish ancestry (and per- haps for refugees residing in Spain), as a situs for universal justice they were no longer an option.

The majority opinion largely follows the arguments of the Spanish government's prosecutors' office. The opinion first quickly disposes of the Audiencia Nacional's inquiry into the availability of an alternative forum which, the Court found, was inappropriate. "Basing subsidiarity on the real or apparent inactivity of local courts implies a judgment of one state's courts about the ability to administer justice of the similarly situated organs of another sovereign state." While such an "unable or unwilling" inquiry might be appropriate for an ICC, national courts should not be making these kinds of judgments, which could have an important effect on foreign relations and should be left to the political branches.

Next, the majority construed the 1948 Genocide Convention. The court found, as the Southern Cone cases had held, that article 6 of the Convention was not limited to territorial and international criminal jurisdictions. However, article 8 directs states to respond to genocide occurring outside their borders by going to the UN, not by exercising universal jurisdiction. The presence of a UN mission showed that the UN knew about conditions in Guatemala, yet had failed to create an ad hoc tribunal along the lines of those in Yugoslavia or Rwanda.

The heart of the matter, for the majority, was that article 23.4 of the LOPD, despite its apparent clarity, could not be so open-ended as to allow criminal investigations based on news of crimes being committed anywhere in the world. Spanish law had to conform to other principles of international law, including respect for other states' sovereignty and the principle of non-intervention in the internal affairs of other states. Extraterritorial jurisdiction, when not authorized by the UN or specifi- cally regulated by treaty, required a point of contact with national inter- ests. The majority cited cases from the German and Belgian courts, and the ICJ's Arrest Warrant (Congo v. Belgium) case (discussed later), in sup-

C op

yr ig

ht ©

2 01

0. U

ni ve

rs ity

o f P

en ns

yl va

ni a

P re

ss . A

ll rig

ht s

re se

rv ed

The Legal Legacy of Pinochet 177

port of these propositions. The court pointed to the "extradite or prose- cute" provisions of a number of treaties, including the Torture and Terrorism Conventions, as requiring the presence of the defendant to proceed when there is no other type of national interest (like the protec- tive principle or active or passive personality principles) involved. A con- nection to a state interest, the majority opined, creates legitimacy and rationality in international relations and respect for the non-interven- tion principle. What is more, this connection to a national interest should exist in the principal charges against the defendant, not just in related or ancillary ones. For that reason, all the genocide charges were bad-none of the defendants were present and there was no allegation that genocide had been aimed at Spaniards as a group. Only the torture charges, to the extent they involved Spanish citizens, could stand, because the Convention Against Torture allows for passive personality jurisdiction.

The seven dissenters started from the position that universal jurisdic- tion in cases of genocide was necessary to avoid impunity, and that in such cases the state acted in representation of the international commu- nity. The majority's view was too restrictive and, therefore, "incompati- ble with the treatment of this grave crime in our internal law and in international law." It ignored the legislative intent and the language of article 23.4, as well as international law. It confused passive personality and universal jurisdiction, left any extraterritorial prosecution of geno- cide practically impossible, confused the treaty-based section (g) of the statute with the rest, and converted the minimum requirements of trea- ties into a maximum. On international law, the majority ignored a later, and contrary, German Supreme Court decision, misconstrued the ICJ's Arrest Warrant case (which had only dealt with immunity), ignored the Belgian court's subsequent upholding of a non-presence-based universal jurisdiction law in Sharon, and failed to cite a number of other relevant cases (including the British House of Lords decision in Pinochet) in sup- port of Europe's widespread acceptance of a broad view of universal jurisdiction.

The dissent answered the majority's concern about chaos in the inter- national system by superimposing a necessity criterion, such that no extraterritorial intervention was needed where the domestic courts were doing the job. That criterion, however, could not be based on the court's inactivity, which was too hard to measure and too politically charged at an early stage in the proceedings. Rather, any limits had to come from a flexible, prudential rule of reason aimed merely at practi- cal concerns like the potential effectiveness of an investigation and extradition request or a potentially high burden on the Spanish courts. For the dissenters, a tie to Spain was merely an aid in applying this rule

C op

yr ig

ht ©

2 01

0. U

ni ve

rs ity

o f P

en ns

yl va

ni a

P re

ss . A

ll rig

ht s

re se

rv ed

178 Chapter 7

of reason, not ajurisdictional prerequisite. In this case, they found more than enough links to Spain to justify the Spanish courts' intervention. The countries were linked by historic, social, linguistic andjurispruden- tial ties. The crimes at issue involved Spanish citizens, not as victims themselves of genocide, but as victims targeted because they were defending others from genocide. This case was a paradigmatic example of those where Spain should exercise its jurisdiction: there would never be a more compelling case. "If there is no nexus in this case, then a nexus requirement becomes a mere pretext to exclude or suppress uni- versal jurisdiction in all genocide cases." That, the dissenters argued, should not be done.

A second case reaffirmed the need for a tight nexus requirement. On March 8, 2004, a panel of the Spanish Supreme Court reaffirmed its Guatemala holding in a case involving Chilean General Hernan Brady. 8

The Court allowed the case to move forward because the victim, Car- melo Soria, was of Spanish nationality. The repetition means that the Court's limited view of universal jurisdiction now has jurisprudential value." It will be followed when it decides pending challenges to Spanish jurisdiction in the cases of Argentine torturers Adolfo Scilingo and Ricardo Cavallo.!" If so, trial should be allowed to proceed, as there were numerous victims of Spanish descent killed or disappeared in the Naval Mechanics School (ESMA) where the two men worked.

The Court also grappled with the argument that the Spanish courts were "subsidiary" to the territorial courts or an international criminal court. On May 20, 2003, a three-judge panel of the Spanish Supreme Court affirmed the dismissal of a case alleging genocide, terrorism, tor- ture, and arbitrary detention against ex-presidents Alan Garcia and Alberto Fujimori, ex-General Vladimiro Montecinos, and other high- ranking government officials in Peru.!' The Peruvian complaint, like the Guatemalan, was based on the universal jurisdiction provisions of Span- ish law and argued that the Peruvian courts were not investigating the crimes. The panel's decision backtracked slightly from the harshness of the majority's ruling in the Guatemala case. It restated the Spanish courts' ability to judge genocide cases based on universal jurisdiction, and reaffirmed the Audiencia Nacional's rulings in the Argentine and Chilean cases.

The panel in the Peru case characterized the issue as one not of sub- sidiarity but of the "principle of necessity ofjurisdictional intervention." To evaluate necessity, the Court considered whether the territorial courts were exercising effective jurisdiction. It looked to whether the events at issue were in fact the subject of prosecution, without attempting to analyze why or why not or to evaluate the existence of a state of de facto impunity. In the Peruvian case, investigations against several of the

C op

yr ig

ht ©

2 01

0. U

ni ve

rs ity

o f P

en ns

yl va

ni a

P re

ss . A

ll rig

ht s

re se

rv ed

The Legal Legacy of Pinochet 179

defendants were proceeding in local courts, and suspects were injail or had fled the jurisdiction (the case with ex-president Fujimori). Thus, for the time being, Spanish prosecution was inappropriate.

Bouterse and the Conservatism of National Law

Not all the countries in Latin America were once colonies of Spain or Portugal. Suriname, for instance, was a Dutch colony. Many Surinamese over the years migrated to the Netherlands, and Surinamese residents often had dual nationality. Suriname was governed during the early 1980s by a military regime that brooked no dissent. On the night of December 8, 1982, fifteen people, including one Dutch national, were arrested, tortured, and summarily executed by the military government, under the command of Desire Bouterse. The victims were prominent opposition figures. The official story was that they had been arrested for plotting a coup and shot while trying to escape, but witnesses contra- dicted the "attempted escape" theory.

Relatives of the victims, unable to get a hearing in Suriname, asked the Amsterdam Court of Appeals to order prosecution of Bouterse in the Netherlands." Unlike Spain, the Netherlands has no procedure for victims to initiate criminal proceedings directly, although they can join in to ask for compensation once an investigation starts (and can appeal a prosecutor's decision not to prosecute). The Court, facing a novel issue, asked international law expert John Dugard to advise whether the customary law on torture or crimes against humanity would allow the court to prosecute these acts, especially given the fact that the Nether- lands only became a party to the Convention Against Torture (and passed its implementing legislation) after the acts in question. This was, of course, the same issue that had troubled the House of Lords in Pino- chetin the extradition context. The Court of Appeals, based on Dugard's reply, backed the position of Lord Millet in Pinochet l1L13 The court allowed the prosecution because torture violated Dutch law on assault, as well as customary international law, long before 1982, and all the Tor- ture Convention did was codify the existing rules and create a new pro- cedure. Therefore, retrospective application of the Convention's rules on extraterritorial jurisdiction did not violate the principle of legality or nonretroactivity of law. Furthermore, the Court added, there were good reasons to allow prosecution of this case in the Netherlands: the case would not be prosecuted at home, the events in question had shocked public opinion in the Netherlands, some of the victims had Dutch nationality (Netherlands law, like Spanish, does not contemplate juris- diction based simply on the victims' nationality), and the complainants

C op

yr ig

ht ©

2 01

0. U

ni ve

rs ity

o f P

en ns

yl va

ni a

P re

ss . A

ll rig

ht s

re se

rv ed

180 Chapter 7

live in the Netherlands. The Dutch courts had a good claim to jurisdic- tion.!'

On September 18, 2001, the Dutch Supreme Court reversed." The court found that under the Netherlands legal system, retrospective application of the 1988 Torture Act would violate the principle of legal- ity. The court found that, while torture may have been a crime under Dutch law and customary international law before that time, the courts could do nothing about it because they had no extraterritorialjurisdic- tion before the Torture Convention implementing legislation was passed. This exactly paralleled the majority's view in Pinochet III. Even though Dutch law theoretically allowed the direct incorporation of international law into national law, customary (unwritten) international law did not override contrary statutory provisions. The Court (mis)read the customary law at issue to be the possibility of retroactive application of the prohibition on torture. The court looked to the legislative history of the implementing law to support its holding that it could not apply before 1989. The court also chose to apply the statute of limitations in force for domestic assault, which had long since passed. The more extended statute of limitations applicable to torture, like the substantive offense, was only codified after the events in question and so could not be applied retroactively. Thus the statute of limitations barred the prose- cution.

The "retroactivity" problem has two variants: when the conduct inves- tigated took place before the investigating state ratified the treaties that allow for extraterritorial jurisdiction, or when it happened before the domestic laws changed to authorize extraterritorial prosecutions, codify crimes, or extend statutes of limitations. The Dutch courts have not been alone in grappling with the question of whether either of those two cases constitutes "retroactive" or ex post facto application of law, which would violate the basic principles of legality. There has been no unanimous response.

Those courts that have not found retroactivity an insurmountable obstacle follow the reasoning of the Dutch Court of Appeals, distinguish- ing between the underlying criminal conduct and the provisions allow- ing for extraterritorial jurisdiction over that conduct. The reason for insisting on nonretroactive justice is that defendants must have ade- quate notice of what conduct is prohibited, and governments must not be allowed to come along after the fact and criminalize politically incon- venient acts or actors. But in cases of torture, disappearance, and mass murder, it is difficult to imagine that notice of criminality is lacking, and indeed, the underlying acts have been prohibited in national and inter- national law for quite some time. As a matter of international law, at least the prohibitions on genocide, war crimes, and crimes against humanity

C op

yr ig

ht ©

2 01

0. U

ni ve

rs ity

o f P

en ns

yl va

ni a

P re

ss . A

ll rig

ht s

re se

rv ed

The Legal Legacy of Pinochet 181

(including systematic torture) date back over fifty years. Therefore, this is not retroactive but retrospective law. The Spanish courts followed this approach in the Chilean and Argentine cases.

The Supreme Court also limited the reach of the Dutch Implement- ing Act of the Torture Convention. Article 5 extends the courts' jurisdic- tion to "any person who commits outside the Netherlands one of the offenses described," but the Bouterse court read the section to require a "point of contact" with the Netherlands, either presence or nationality/ residence. Meanwhile, as with other transnational investigations, this one too triggered a preliminary judicial proceeding in Suriname. The Court of Appeals noted that if such a domestic proceeding were to pros- per, it would suspend the Dutch proceedings. To date, however, the Suri- namese case seems not to have progressed.

The Hissene Habre case and the Politics of Prosecution

Chileans and Argentines prosecuted in Spain, Rwandans in Belgium- the pattern seemed obvious to those opposed to transnational prosecu- tions. "It's an act of colonialism," thundered the right-wing National Renovation Party in Chile.!" Indeed, it did seem like a group of devel- oped country, often ex-colonial country, courts were judging a group of developing country defendants. This imbalance may necessarily be a fact of life, at least for now-after all, it is easier to keep track of dictators who travel to rich countries, where the exile and activist networks that sustain investigations tend to be strongest-but it is an uncomfortable fact. A truly universal system ofjustice would be more balanced.

So it was with great interest that Reed Brody, advocacy director of the New York-based organization Human Rights Watch, listened to Del- phine Djiraibe. Djiraibe, president of the Chadian Association for the Promotion and Defense of Human Rights, wanted help in bringing Chad's former dictator Hisscne Habrc to justice for massive violations of human rights."? From 1982 to 1990, Habrc periodically targeted various ethnic groups, killed political prisoners, and ran a political police accused of torture and murder. A subsequent Truth Commission accused his regime of committing crimes against humanity, although it was unable to investigate fully. The new, post-1990 government was headed by Habrcs former chief of staff and defense minister, and it was naturally disinclined to investigate or prosecute the crimes of its prede- cessor. A Chadian Association of Victims of Political Repression and Crime had compiled dossiers on 792 cases, but had been unable to advance domestic prosecutions. Most interestingly, Habrc had fled, not to a European country, but to Senegal. This would be the first attempt to prosecute a former African head of state in his place of refuge. If it

C op

yr ig

ht ©

2 01

0. U

ni ve

rs ity

o f P

en ns

yl va

ni a

P re

ss . A

ll rig

ht s

re se

rv ed

182 Chapter 7

worked, there was a long line of other African dictators and their hench- men who were then living in nearby states, including Idi Amin of Uganda (living in Saudi Arabia) and Mengistu Haile Mariam of Ethiopia (living in Zimbabwe), who might begin to feel unwanted in their new homes.

Brody and the Chadians agreed that Senegal seemed a promising place to bring a case based on universal jurisdiction. It had a relatively independent judiciary, a stable democracy, and had demonstrated lead- ership in human rights. It was the first country to ratify the Statute of the ICC, and had been a party to the Torture Convention since 1986. A successful prosecution could serve as an example to other African states that Africa could deal with its own, and dispel the notion that only ex- colonial powers were interested in bringing dictators to justice.

Reed Brody, in a 2001 article, tells what happened:

International researchers visited Chad twice, where they met the remaining faithful of the moribund Association of Victims who provided them with the thousands of pages of documentation they had prepared in 1991, and had kept in hiding when the government changed its mind on digging up the past. Work- ing in secret, because of fears that someone, including Chadian officials, might tip off Habre who could then flee Senegal for a more protective shelter, the researchers were introduced to victims and potential witnesses and sought docu- mentation of Habrcs crimes. Meanwhile, Human Rights Watch quietly orga- nized a coalition of Chadian, Senegalese and international NGOs to support the complaint, as well as a group of Senegalese lawyers to represent the victims. The coalition decided to bring the case as a private prosecution rather than present- ing the evidence to Senegalese authorities and requesting a state prosecution;"

Seven Chadians and the Victims Association filed a complaint on Janu- ary 25,2000 in Dakar Regional Court charging Habrc with torture, bar- barous acts (a crime under Senegalese law), and crimes against humanity, including forced disappearances. By chance, the case fell to investigating magistrate Demba Kandji, who had advanced education and training in international human rights law. Kandji moved quickly to get the state prosecutor's advice, to hear testimony, and to summon Habrc and indict him on initial charges of being an accomplice to tor- ture, leaving open the possibility of further charges to follow. Judge Kan- dji sent letters rogatory to Chad, asking a Chadian judge to interview witnesses there, and heard testimony from ex-political prisoners who told of Habrc's visits to prisons and presence during torture sessions.

The victims' initial euphoria was short-lived, however. In February, Habrcs lawyers filed a motion to dismiss the case before the Court of Appeals, alleging lack of jurisdiction. More ominously, the Senegalese press began portraying the ex-dictator as the victim of a French-Ameri- can plot and not such a villain after all. There were indications that

C op

yr ig

ht ©

2 01

0. U

ni ve

rs ity

o f P

en ns

yl va

ni a

P re

ss . A

ll rig

ht s

re se

rv ed

The Legal Legacy of Pinochet 183

Habre was spending some of the millions he was accused of stealing from Chad in the public relations effort. "We learned that you have to build the case in public opinion before building it at court. We started doing this too late, and Habrc 's lawyers and friends got there first," recounted Pascal Kambale, a lawyer for the complainants, at a confer- ence in March 2001. 19

The political climate soon got worse. In March 2000, Senegal elected a new president, Abdoulaye Wade, whose advisor on justice issues was none other than Habrc's defense counsel. The government's attitude toward the prosecution quickly worsened. Government attorneys, pre- viously supportive of the complainants, now supported dismissal. The Superior Council of Magistrates, headed by the president, transferred Judge Kandji off the investigation to a new post and promoted the appeals court judge hearing the case. The signs were clear and accurate. On July 4,2000, the appeals court dismissed the charges against Habre.

The court found that Senegalese courts had no jurisidiction over extra-territorial torture. Senegal had ratified the Torture Convention, but it had never made the prosecution of torture part of its domestic criminal law, and therefore the Convention's obligations to extradite or prosecute suspected torturers were not effective in local courts. The court rejected the argument that Senegal's constitution automatically made international treaties part of domestic law, finding that although that may generally be true, penal law, because it can affect individual liberties, requires full translation into the penal code. The court also rejected the use of the customary international law prohibiting crimes against humanity, because, like torture, these crimes had not been codi- fied as such in the local penal code. Months later, on March 20, 2001, the Cour de Cassation affirmed. 20

The case, like Bouterse, was a clear reminder of the unfinished state of affairs in international law. True, Senegal was a leader in signing and ratifying treaties, but without domestic implementing laws, it was hard to make use of them in the local courts. One of the lessons of Habre in Senegal is that even civil law judges will be much more comfortable, especially in the criminal setting, if substantive treaty provisions have been translated to the penal and procedural codes, with appropriate penalties attached. Even where judges could, under domestic law, apply customary international law principles, most (although not all) will be reluctant to do so in a case involving individual criminal defendants. They want codification.

Moreover, law was uncomfortably subject to political decisions to quash cases: the New York TimesJuly 21 editorial on the case denounced what "looks suspiciously like interference from the country's new presi-

C op

yr ig

ht ©

2 01

0. U

ni ve

rs ity

o f P

en ns

yl va

ni a

P re

ss . A

ll rig

ht s

re se

rv ed

184 Chapter 7

dent." The Executive Branch in Senegal had achieved what Spain's con- servative government could not: sink a judicial inquiry.

While to date governments seem to be much more willing to inter- vene to quash than to promote investigations, the Senegalese experi- ence in the Habrc case highlighted one of the key worries of skeptics on universaljurisdiction: governments could equally push courts to investi- gate where such investigation was not warranted. After all, not all courts are equally independent or equally free of pressure or equally solicitous of the rights of the accused and the victims. In the hands of the wrong court, a prosecution could turn into either a witchhunt or a whitewash. On the other hand, the same is true for prosecutions based on any other ground of jurisdiction. Most politicized or unfair trials are run-of-the- mill territorial affairs, and in these no one argues that the courts have no jurisdiction. Instead, rules to determine what constitutes a fair trial, basic but still meaningful, can be found in the core global and regional treaties on human rights and the dozens of Guidelines, Principles and other such documents crafted by international lawyers and judges over the years." The rules specify such bedrock principles as an independent judiciary, the accused's right to a lawyer and to not incriminate himself, the rule against ex post facto charges, and the like. If states implement- ing universal jurisdiction fail to respect those rules, they can be chal- lenged and criticized just as they would be for any other type of prosecution.

The victims and their supporters, horribly disappointed by the Sene- galese dismissal, took solace in the increased visibility of the issue within Chad. As in Chile and Argentina, the transnational prosecution galva- nized local efforts. In September, the president of Chad met with the Association of Victims, expressed his disappointment with the decision, and promised support for their cases within Chad and access to pre- viously inaccessible evidence. In October 2000, seventeen victims lodged criminal complaints for torture, murder, and disappearances against members of the former political police. Dozens of cases filed against individual torturers have followed. Initially dismissed on grounds that a 1993 law ousted the regular courts of jurisdiction, in April 2001 the Constitutional Council reinstated the complaints and found that the 1993 law should be withdrawn;" In May 2001, the Chadian government authorized NGO researchers to search the former secret police prem- ises. They found thousands of archived pages documenting the repres- sion, which are now being analyzed for future prosecutions. The continuing limits of domestic prosecution became clear early on, how- ever, when the office of the prosecutor handling the file was ransacked, and victims' lawyer Jacqueline Moudeina was attacked and badly injured by a potential defendant.

C op

yr ig

ht ©

2 01

0. U

ni ve

rs ity

o f P

en ns

yl va

ni a

P re

ss . A

ll rig

ht s

re se

rv ed

The Legal Legacy of Pinochet 185

With the help of the international committee of NGO supporters (including Human Rights Watch and the International Federation for Human Rights), the victims' association decided to continue with efforts to investigate and try Habrc outside Chad. The Belgian courts seemed the best possibility: Belgium had excellent legislation on universal juris- diction and French-speaking magistrates. The only problem was that any Belgian investigation was likely to take some time. Meanwhile, with no outstanding warrant against him, Habrc was free to go, and likely to quickly find a more sheltered haven.

The human rights lawyers sought ways to pressure the Senegalese authorities to keep Habrc in his adopted home until a Belgian warrant could be issued. They lobbied the Committee Against Torture, a United Nations expert body in charge of monitoring compliance with the Tor- ture Convention, to find that Senegal's compliance with the treaty depended on not letting Habre slip away;" Eventually, UN Secretary General Kofi Annan issued a personal plea to the Senegalese president. With international scrutiny and criticism of the Senegalese courts grow- ing, President Wade agreed that Senegal would hold Habrc pending an extradition request from a Belgian judge.

The Habre case in Belgium was assigned to investigating magistrate Daniel Fransen. In the spring of 2002, he traveled to Chad to investigate the charges against Habre. Survivors of torture and secret detention came to the worn courthouse in N'Djamena, the Chadian capital, to tes- tify before Fransen. He also traveled to the sites of mass graves and five hidden detention centers, including one on the grounds of the ex-presi- dential compound. He moved around the dusty capital with a prosecu- tor, four Belgian policemen, and a court clerk. The judge's visit again energized the victims. "Before, everyone thought we were just a bunch of crazy people. Now, they can see that Habrc could go on trial for his crimes," Souleymane Guengueng, now the head of the victims' associa- tion, told a Canadian reporter."

Reed Brody accompanied Judge Fransen in Chad. He recalls one par- ticular incident involving Ismael Hachim, an ethnic minority prisoner held for seventeen months in a covered-over swimming pool turned underground prison. When Hachim got out of jail upon Habrc 's over- throw, the officers who had arrested him told him that it was Touka Hal- iki, Habres director of intelligence, who had ordered his arrest and torture. It was only in 2001, when Hachim got access to his personal secret police file, that he could be sure. Haliki, still a police supervisor, was one of those Judge Fransen called in to testify, and when Haliki denied involvement in the persecution of the Zaghawa minority group, the judge called Hachim to encounter Haliki face-to-face. When Haliki still claimed his innocence, Hachim whipped out the document in

C op

yr ig

ht ©

2 01

0. U

ni ve

rs ity

o f P

en ns

yl va

ni a

P re

ss . A

ll rig

ht s

re se

rv ed

186 Chapter 7

which Haliki ordered his arrest and another in which Haliki signed off on Hachim's interrogation. "In that moment, he became very small, and I became very tall," explained Hachim to all who would listen. "Even if he is never prosecuted, I now feel like some justice has been done. That's what this is really about, isn't it?"25

The case against Habrc in the Belgian courts continues as of this writ- ing. Both Senegal and Chad have indicated that they have no objection to an extradition request from Belgian authorities. But the case is one of only three to survive the 2003 reform of Belgium's universaljurisdic- tion law.

Yerodia, Sharon, and the Reform of Belgium's Universal Jurisdiction Law

The 1999 Belgian law went farther than any other to date. It allowed "pure" universal jurisdiction, that is, jurisdiction that does not require the presence of the defendant to open an investigation. It also allowed anyone, even plaintiffs who did not live in Belgium, to bring charges, and it explicitly ruled out any immunity from prosecution, even for sit- ting heads of state. Predictably, it generated a great deal of business for the Belgian criminal courts. An early complaint, originally filed in November 1998 (at the same time as the Belgian complaint against Pino- chet) charged Abdulaye Yerodia Ndombasi with grave breaches of the Geneva Conventions and crimes against humanity. The complainants were mostly Congolese exiles living in Belgium; five were Belgian citi- zens. They charged that, as part of the Laurent Kabila government's efforts to expel an ethnically Tutsi rebel force in the Eastern Democratic Republic of the Congo (DRC) , top officials in that government, includ- ing Kabila himself, Yerodia Ndombasi, and two others, publicly called for acts of violence against the "invaders." Some of their incendiary speeches were captured on TV tape. Their calls were soon answered by a wave of lynchings, arrests, and persecution of Tutsis throughout the DRC.

After a year's investigation and with the approval of the state prosecu- tor, on April 11, 2000,Judge Vandermeersch issued an arrest warrant for Yerodia as author or co-author of war crimes and crimes against human- ity. There was only one small problem: at the time he issued the warrant (although not when the acts took place) Yerodia was foreign minister of the DRC. Judge Vandermeersch recognized the potential awkwardness of a foreign minister trying to do his job with an Interpol warrant out for his arrest. He noted that while under Belgian law there was no reason to preclude the ability of the courts to try the case, execution of the arrest warrant had to be stayed while the suspect was a state representa- tive on an official visit. To do otherwise would impermissibly gum up the

C op

yr ig

ht ©

2 01

0. U

ni ve

rs ity

o f P

en ns

yl va

ni a

P re

ss . A

ll rig

ht s

re se

rv ed

The Legal Legacy ofPinochet 187

works of international diplomacy and tread on the rights of the state, not just the individual. However, there was a point to issuing the warrant even if it could not be executed: it allowed the judge to compile a dos- sier of information about the alleged crime that would be available once the defendant no longer held a diplomatic post. Vandermeersch also notified the Congolese courts of the case, but they made no effort (not surprisingly) to open their own investigation.

The DRC did react, however. On October 17, 2000, the Congolese government filed a complaint with the International Court of Justice, the venerable tribunal for legal disputes between states. It accused Bel- gium of overstepping its authority, violating the customary law principle that a state may not exercise its authority on the territory of another state, and trampling on the rules about the immunity of foreign minis- ters from suit. It asked the court to tell Belgium to quash the arrest war- rant.

By the time the case came before the court, Yerodia was no longer part of the Congolese government and that government had dropped its objections to universal jurisdiction in general to focus on the immu- nity issue. Belgian lawyers argued that since Yerodia was no long entitled to immunity, the case was moot and should be dismissed. Not so, said the DRC. It was still an affront to them that their foreign minister had been so badly treated, and the arrest warrant was still pending against him.

The ICJ decision, announced on Valentine's Day, February 14, 2002, vindicated the DRC position.s" The Court ordered Belgium to rescind the arrest warrant against Yerodia. The Court, against the wishes of some of the judges, focused entirely on the immunity issue. A majority found that customary law protected sitting heads of state, heads of government and diplomats in their personal capacities. It also covered sitting foreign ministers who needed to travel freely in order to carry out their duties. In other words, they enjoyed a functional immunity from civil or crimi- nal suit in another country. The presence of specific provisions denying immunity in international instruments like the Statutes of the ad hoc tribunals and the ICC only covered those specific tribunals and did not apply to national courts.

This was arguably not true at all, since many of the immunity-denying instruments were aimed at national as well as international courts. How- ever, the Court then went much further, muddying up again the little bit that the House of Lords had left reasonably clear. First, the Court ignored the nature of the charged crimes and the customary law that certain international crimes had to be prosecuted no matter who the suspect was. Second, the ICJ found four situations under which immuni- ties under international law would not bar prosecution: when the

C op

yr ig

ht ©

2 01

0. U

ni ve

rs ity

o f P

en ns

yl va

ni a

P re

ss . A

ll rig

ht s

re se

rv ed

188 Chapter 7

national state institutes proceedings against its own officials; when it waives their immunity before another state; when the case is before an international criminal court; or, "in respect to acts committed prior or subsequent to his or her period of office, as well as in respect of acts committed during that period of office in a private capacity."

The first two were unlikely to arise: after all, if the home state were willing to prosecute, what would be the point of going elsewhere? The language about private acts was particularly problematic. How would it apply in the case of a Pinochet or other high-ranking official? Did the Court's language about private acts mean that they could be tried for murder committed during a fit of pique, but not for a deliberate policy of state-sponsored murder? Could torture or summary execution ever be "private" when their definition involves persons acting in an official capacity? Even international crimes that do not explicitly require an ele- ment of official action as part of the crime would usually be carried out as part of an ex-minister's public, not private, functions. If read to pre- clude prosecution in those cases, the Court's decision set the law back decades and contravened much national and international jurispru- dence (and state diplomatic efforts) on the subject.

A trio of judges on the Court, each with a long and distinguished career protecting human rights, saw the problem clearly. In a separate opinion,"? they tried to minimize the damage by suggesting that interna- tional crimes, by their very nature, could never constitute anything but private (unofficial) acts, and so no issue of immunity arose once the per- sonal immunity of a sitting minister or head of state no longer applied. Therefore, former heads of state, ministers and the like could be arrested and tried for such acts. If this is true, then the Court's decision creates no great impediment to current practice, but the terminology deeming large-scale ethnically or politically motivated killings "private" remains both unnecessary and a little strange.

The Arrest Warrant case left for another day the question of whether universal jurisdiction was an unacceptable infringement on sovereignty. That day was not long in coming. On December 9, 2002, the Congo (Brazzaville, next door to the DRC) filed a complaint in the IC] arising from the prosecution in France of its President, Minister of the Interior, Inspector General of the Army and and captain of the presidential guard, on charges of crimes against humanity and torture stemming from the killing and disappearance of over 350 refugees on the shores of the Congo River in May 1999 (known as the "disappeared of the Beach" case) .28 The army inspector general, Norbert Dabira, had a house in France that he visited regularly, which allowed a local French magistrate to open a criminal investigation. The case, as framed, directly chal- lenged the French effort to use universal jurisdiction to prosecute the

C op

yr ig

ht ©

2 01

0. U

ni ve

rs ity

o f P

en ns

yl va

ni a

P re

ss . A

ll rig

ht s

re se

rv ed

The Legal Legacy ofPinochet 189

defendants "for crimes allegedly committed in connection with the exercise of his powers for the maintenance of public order." The case is ongoing as this book goes to press.

While the Yerodia case returned to the Belgian indictment chamber to determine whether a new arrest warrant was in order, new cases took center stage. Several of them were explosive. Cumulatively they would determine the fate of the 1999 law. Rather than focus on obscure defen- dants in little-known conflicts, they forced the Belgian courts to meta- phorically stick their head into the lion's mouth of global politics, with predictable results. Of these, perhaps the most explosive was the case filed against Israeli Prime Minister Ariel Sharon.

The Sharon case began onJune 18,2001, when 23 Lebanese-Palestin- ian complainants, only one resident in Belgium, filed a complaint against Ariel Sharon, at the time defense minister and now prime minis- ter of Israel, and Amos Varon, then the commander of Israeli forces in Lebanon and now CEO of the Israeli Defense Forces, along with other Israeli military officials and members of the Phalange Lebanese militia. The complaint alleged that Sharon, Varon, and the others had commit- ted war crimes, crimes against humanity, and genocide by allowing Leb- anese militiamen to murder hundreds of innocent civilians in the Sabra and Shatila refugee camps during Israel's 1982 invasion of Lebanon. They pointed as evidence to an Israeli commission of inquiry that had found Sharon "indirectly responsible."

The Belgian investigating magistrates recognized full well the explo- sive political implications of investigating Israel's head of state, but felt constrained by the requirements of Belgian criminal procedure to open a file, after first getting the approval of the state prosecutor. The deci- sion set off an outcry. The Sharon case seemed the worst possible test case: a high-profile, powerful, controversial defendant, a current head of state, no ties to Belgium to speak of, no cooperation from the state where the defendant resided or where the crimes were committed. As Israel pointed out, the Israelis had delved extensively into the case, Sharon had lost his job, and there had been at least an implicit decision that criminal prosecution was not warranted. On the other hand, there was plenty of evidence tying the suspects to awful crimes. How could the complainants' lawyers tell their clients that their suffering and the loss of their family members was too politically sensitive and not worth trying every possible legal avenue to redress?

In a June 26, 2002 decision, the Brussels Court of Appeals first reaf- firmed that a law establishing Belgian jurisdiction over international crimes committed outside Belgium by non-Belgians (that is, under uni- versaljurisdiction) was not per se invalid. It was, however, subject to the requirement that the defendant be present on Belgian soil.>" Ariel

C op

yr ig

ht ©

2 01

0. U

ni ve

rs ity

o f P

en ns

yl va

ni a

P re

ss . A

ll rig

ht s

re se

rv ed

190 Chapter 7

Sharon, presumably, was not to be found in Belgium. In that case, the court found, the case could not be heard and should be dismissed. The Court of Cassation in February 2003 confirmed the dismissal, but now on immunity grounds, following the ICJ majority opinion. However, it allowed the case against other defendants, including Amos Varon, to go forward.

The Sharon case raised the profile of Belgium's universal jurisdiction law and its potential for discomfiting the powerful. More high-profile cases found their way into the courts. In the context of the buildup of war in Iraq and fierce European opposition to U.S. government actions there, a group of Iraqis in March 2003 sued former president George H. E. W. Bush, Richard Cheney, Colin Powell, and Norman Schwarzkopf for bombing an air raid shelter in Baghdad during the 1991 Gulf War. 30 In May, another complaint accused General Tommy Franks and other U.S. officials of war crimes committed during the 2003 invasion of Iraq."

This was too much. Even Foreign Minister Michel, a supporter of the Belgian law, called the Franks suit an abuse of the law. With the U.S. threatening to pull NATO meetings out of Belgium and Defense Secre- tary Donald Rumsfield publicly fuming at Belgian effrontery, the Parlia- ment acted. Since January 2003, the Belgian Senate had been working on a series of amendments to the law. Now the legislature quickly passed a sweeping package of changes designed to curb the use of the Belgian courts. In August 2003, in the face of continuing U.S. pressure the incoming government coalition agreed to amend the law still further, to make it even more difficult to use in cases not related to Belgium through nationality or the place of the crime.

One change harmonized Belgian law with the international law on immunity (that is, the ICJ decision in the Arrest Warrant case). The most important changes concern the issues of links to the forum, executive discretion, and the relationship to other jurisdictions." A case cannot be opened unless there is a link between the crime and Belgium: either the suspect must be Belgian, or the victim must be Belgian or resident in Belgium for at least three years, or a treaty (e.g., the Torture Conven- tion) must require prosecution, for example when the suspect is present in Belgium. Victims can now file suit directly (as partie civil) only if the accused is Belgian or lives in Belgium. Otherwise, the decision to investi- gate lies entirely with the state prosecutor or, in some cases, the office of public prosecutions. The prosecutor is obliged to proceed unless, among other reasons, in the interests ofjustice and in keeping with Bel- gium's international obligations, the case should instead be brought in another jurisdiction, where the administration ofjustice is independent and impartial. The decision not to proceed cannot be challenged. These last provisions seemed tailormade to get rid of the cases against Israelis

C op

yr ig

ht ©

2 01

0. U

ni ve

rs ity

o f P

en ns

yl va

ni a

P re

ss . A

ll rig

ht s

re se

rv ed

The Legal Legacy of Pinochet 191

and Americans and likely to produce negotiations over what the inter- ests ofjustice consist of in every new case.

Belgian and international lawyers and NGOs protested that the law was being gutted based simply on U.S. antipathy to any kind of interna- tional justice, and that the proposed reforms went too far. A few pending cases, including Habre, were "grandfathered in" because the investiga- tions were at an advanced stage, but most were not.

Tightening Up State Control: Links and Presence

Along a number of different axes and in a number of countries, the law has been reduced and tightened up. Bothjudges and legislators seem to be reacting to a sense that universal jurisdiction without specified limits is too unbounded, too subject to confusion when more than one juris- diction can prosecute the same course of conduct. In an effort to create an orderly process of prioritization faced with the theoretical possibility of multiple prosecutions for the same course of action, Spain and Bel- gium, through different means, have in effect superimposed a national- ity tie (or at least something close to it) on something they are still calling universal jurisdiction. The partial closing of the Spanish and Bel- gian forums leaves few states with universal jurisdiction provisions that do not require either the presence of the defendant or a nationality (or at least long-term residency) link. The German universal jurisdiction law, for example, on its face requires no link to Germany in cases of genocide, crimes against humanity and war crimes. However, section 153f of the Code of Crimes Against International Law directs the public prosecutor to excercise discretion to avoid cases where there is no tie to Germany (nationality, presence, or anticipated presence), and to defer to an international court or to a state with ties to the crime, the defen- dant, or the victim that intends to extradite. The law does, however, allow cases without a nationality or presence link in extraordinary cir- cumstances." South Africa and Canada allow universal jurisdiction where either the perpetrator or victim is a citizen or resident or the per- petrator is present. 34

The practical effect on existing cases may not be great. In the Argen- tine, Chilean and Guatemalan cases in Spain, for instance, there are Spanish victims; in Belgium, the ongoing Rwandan and Chadian (Habre) cases, among others, involve Belgian citizens and long-time res- idents. But the conceptual effect is much more serious. The re-imposi- tion of a nationality tie in effect negates the whole point of universal jurisdiction, reducing it to a simple variant on passive personality juris- diction. There are in general preferences in the law for judicial forums with close ties, of some sort, to the subject matter of a case, out of practi-

C op

yr ig

ht ©

2 01

0. U

ni ve

rs ity

o f P

en ns

yl va

ni a

P re

ss . A

ll rig

ht s

re se

rv ed

192 Chapter 7

cal concerns for the ability of a court to actually try the case. But if uni- versal jurisdiction exists at all because these crimes are of concern to all states, why should any additional tie be necessary as a jurisdictional prerequisite? Most international criminal law treaties simply require states either to extradite suspects to a country that will investigate and, if necessary, try them, or to do the job themselves. Either one is fine. Here it seems that the difference between the prudential "rule of rea- son" of the Spanish Supreme Court dissenters and the jurisdictional prerequisite of the Spanish majority and the Belgian legislator is the dif- ference between some universal jurisdiction and jurisdiction that is uni- versal in name only.

The question whether a defendant needs to be present in the jurisdic- tion of the court, and at what stage, goes to the heart of what universal jurisdiction is about. Completely unmoored prosecutions seem too open to potential abuse, too much a case of complainants shopping for a forum. But at what point does the defendant have to be present? Can a court investigate, issue a warrant, request extradition of the defendant, and then satisfy the presence requirement through extradition? If, as is likely, extradition is denied, the case is ended, but the further result is a legal obligation (known as aut dedere aut judicare, or extradite or prose- cute), at least under several relevant treaties like the Convention Against Torture, to pursue the case at home. After all, if the defendant has to be present for the judicial process to commence, it will be difficult to ever get to the arrest stage, as ex-dictators and torturers are unlikely to linger somewhere long enough for a conscientious judge to put a dossier together, at least once they get wind of an investigation. Universaljuris- diction, under that scenario, will still playa constructive role, but it will be very much to ensure that dictators stay home, that there is no foreign safe haven for such people, not to see that they are actually brought to justice. Nor can courts put together the evidence and testimony that might jump-start a domestic prosecution if the "investigate and extra- dite" route is closed. Under a "presence of the defendant" rule, the Pinochet case would never have happened. Recall that British human rights groups had tried four times to start a prosecution against him on various trips to the UK, but each time he had left before the slow machinery of the justice system (and the requisite political will) could be brought to bear. Without the Spanish extradition request, it would have been much more difficult to put that machinery into operation.

Many countries-more than 120, according to a 2001 Amnesty Inter- national study35-have some universal jurisdiction provisions, and most of these do not specify one way or the other whether the defendant needs to be present at the start. More recent universal jurisdiction provi- sions, enacted as states bring their law into conformity with the Rome

C op

yr ig

ht ©

2 01

0. U

ni ve

rs ity

o f P

en ns

yl va

ni a

P re

ss . A

ll rig

ht s

re se

rv ed

The Legal Legacy ofPinochet 193

Statute, show no clear pattern, although a majority require presence when there is no other tic.:" There seems to be no clear rule on presence under the little existing customary law that allows states to prosecute. Treaties are similarly split: while the Geneva Conventions grave breaches provisions have no presence requirement, treaties covering everything from torture to hijacking to attacks on diplomats start from the premise that the defendant is present. Other forms ofjurisdiction do not require the defendant's presence to investigate or charge. To take a current example, the Al-Qaeda bombings of U.S. embassies in Kenya and Tanza- nia in 1998 resulted in a series of convictions in U. S. courts, but the first named defendant is still at large: the case caption is United States v. Osama Bin Laden. It is hard to imagine precluding investigation until the defendant was present in the forum."? Yet why should these cases be dif- ferent?

There are valid concerns that judges will be overwhelmed with the expense and time involved in preparing arrest warrants and extradition requests that are unlikely to prosper. Also, complainants and their law- yers may be tempted to bring charges under these circumstances simply for the publicity value, or to make a timely political point (a charge that could be leveled easily against the Belgian cases against U. S. officials, for example). Where an extradition treaty leaves proof of a defendant's guilt to the requesting state, the burden will be considerably lessened. Perhaps a useful role for NGOs would be to compile lists of witnesses and evidence against likely suspects, files that can be submitted on short notice to investigating magistrates or prosecutors. Not knowing whether such a file exists and could be quickly exploited would at least provide some disincentives to Riviera retirements and shopping sprees for genoci- daires. But on balance, for transnational prosecutions to playa useful role either as deterrent or catalyst, courts' ability to do some pre-pres- ence investigation seems to be a prerequisite.

Territorial and Transnational Prosecutions

The relationship of the transnational forum to the domestic one is com- plex. First are the concerns, previously mentioned, about unfair trials and double standards between strong and weak countries. More funda- mentally, does universal jurisdiction only come into playas a backstop when national courts cannot or will not act? Even if you agree with the general premise that, where possible, domestic courts are better atjudg- ing questions of mass crimes on their own territory-they have more knowledge of the situation, access to the evidence, ability to change local perceptions and power balances-that still leaves open the thorny ques- tions of capacity. Everyone agrees that under some circumstances, the

C op

yr ig

ht ©

2 01

0. U

ni ve

rs ity

o f P

en ns

yl va

ni a

P re

ss . A

ll rig

ht s

re se

rv ed

194 Chapter 7

local courts cannot do the job, especially where the suspects are still powerful or the courts compromised. How should one court decide whether another country'sjudiciary is unable or unwilling to act, or ade- quately criminalizes the acts and provides due process?

One approach, adopted by the Spanish Audiencia Nacional in the Guatemalan case seems to apply a version of the "unable or unwilling" standard. First, see if there are legal impediments to prosecution, like amnesty laws. If not, see if fear and intimidation might keep judges from doing their job even without a formal law. The approach illustrates some of the potential problems and unknowns. What if the formal impedi- ments to prosecution come not from an amnesty law, but from other legal devices like statutes of limitation, immunities, or repeated proce- dural maneuvers by defendants? Will a foreign court be willing and able to wade through all the possible veiled impunity devices of someone else's law? Perhaps so, with assistance from the parties. But does such a rule merely ensure that new dictators will protect themselves from prose- cution not through the amnesty laws common in the 1970s and 1980s (and increasingly discredited now), but through less formal, more deadly devices that create a de facto rather than legal impunity?

The Guatemalan case shows how difficult it is for courts to judge de facto impunity. Why did the court pick 1999 (the year the Historical Clarification Commission issued its report) as the baseline, rather than the beginning of civilian government (1986) or the beginning of the "peace process"? (1996). If one of those dates had been the jumping- off point, the inactivity and hostility of the local courts would have been much more salient. How long should the transnational court (or the ICC, for that matter) wait to see if domestic action is forthcoming? How many cases need to be underway before the courts are given a pass? Will a few (how many?) "emblematic" cases do, or do most cases have to be investigated? What if the local courts are willing to condemn the lower- level triggermen but not the higher-ups who gave the orders? For this reason, the Supreme Court in the Peruvian case abandoned the effort, looking simply to whether the events at issue were in fact the subject of prosecution, without attempting to analyze why or why not or to evaluate the existence of a state of de facto impunity. This creates a more work- able bright-line rule, but also leaves open the possibility that all possible forums will be de facto closed to complainants.

On the other hand, requiring complainants to show that trial at home is impossible creates tensions among lawyers and NGOs who will need to work together to present the case. The tensions are, in a sense, the flip side of the dilemma that transnational prosecutions create for the governments of the "target" country: to argue that it is illegitimate to try the suspect abroad, they have to allow trial at home. Conversely, it is

C op

yr ig

ht ©

2 01

0. U

ni ve

rs ity

o f P

en ns

yl va

ni a

P re

ss . A

ll rig

ht s

re se

rv ed

The Legal Legacy of Pinochet 195

hard for complainants to show that trial at home is impossible while at the same time participating in or encouraging efforts to open up at least some space for domestic investigations-which is, after all, one of the aims of transnational prosecutions in the first place. This dilemma was present in the Chilean case as well. Experts were providing affidavits to Judge Garzon assuring him that it was impossible to bring a successful complaint against Pinochet in Chilean courts, even as day to day those complaints began to pile up, due in no small measure to the catalytic effect of Garzon's investigation. In an application of the Heisenberg uncertainty principle to law, the very success of the foreign investiga- tions changes the domestic conditions that give rise to them and the attitudes of the actors involved. Eventually, complainants must choose the most promising forum, but forcing them to do so prematurely can choke off useful pressure for change. It can also force victims who are often scattered around the world to concentrate their resources on a single place, which may prove much more difficult for them.

Multiple investigations in competing jurisdictions raise a number of difficulties. Extradition requests from a number of investigating states for the same defendants have to date been solved through a combina- tion of a "first in time, first in line" rule and a preference for states with which the extraditing state has a relevant treaty, and these rules seem by and large to be adequate. Extradition requests from the territorial state subsequent to a transnational indictment raise the question of the genu- ineness of the inquiry: in the Cavallo case, for example, Argentine courts opened investigations based on lesser charges in a transparent attempt to bring the defendant home. Even when both investigations and prose- cutions are carried out in good faith, at what point should the case be transferred back to the territorial forum, and what guarantees should the transnational forum request that a domestic court will see it through? What about the existing investigative file, the witness state- ments and the like? Some witnesses might be willing to testify abroad, but not be happy about having their statements transferred to a domes- tic court they are reluctant to trust. There is also the issue of judicial economy: at some point, the balance of equities shifts to finishing trial where the investigation has begun.

A final problem is that not all de jure amnesties are alike. The Chilean and Argentine decisions hold that an amnesty, especially one con- demned by regional and global human rights bodies, will not impede prosecution of beneficiaries who stray outside their borders. The French Cour de Cassation agreed in the Quid Dah case, finding that recogni tion of a domestic amnesty would undermine the principle of universal juris- diction.:"

But what about amnesties that make nonprosecution contingent on

C op

yr ig

ht ©

2 01

0. U

ni ve

rs ity

o f P

en ns

yl va

ni a

P re

ss . A

ll rig

ht s

re se

rv ed

196 Chapter 7

truth-telling, contrition, or community service? The South African example is the clearest case of a conditional amnesty law that was widely supported internationally and had significant domestic backing. If a beneficiary of that amnesty were to travel, say, to Belgium, would he be subject to potential prosecution for committing crimes against humanity?

A first answer might be to say that such scenarios should arise rarely, because the amount of support and resources it takes to mount a credi- ble transnational prosecution presupposes at least a significant amount of support from victims groups and human rights organizations at home, which would not be forthcoming if a conditional amnesty had widespread legitimacy. Post-South Africa, conditional amnesty-type laws in places like Rwanda and East Timor have excluded organizers, plan- ners, or leaders in genocide, crimes against humanity, and serious war crimes. They have, instead, turned to alternative forms of punishment and accountability for low-level offenses, while insisting on punishment for the worst crimes.

Nonetheless, these are not complete answers. In reality, as with the determination of whether the courts are capable of acting, both the ICC and transnational prosecutions do take decisions about the validity and legitimacy of amnesty and pardon laws out of the purely national ambit. International law provides some guidelines for judges: blanket amnes- ties, self-amnesties granted by the perpetrators, and amnesties that do not require disclosure of the facts require no deference. In closer cases, a fact-specific inquiry will be needed. As Garth Meintjes andJuan Men- dez suggest, the proportionality between the political aim sought and the means employed, combined with absolute prohibitions on amnesty for the most serious international crimes, may be the best we can do in terms of general rules.!" It will be up to judges in transnational cases (as well as at the ICC) to evaluate whether the application of a particular contingent amnesty law (or alternative punishment measure) to specific facts passes muster and therefore makes a transnational prosecution inappropriate.

Stepping Forward, Stepping Back

So should activists have restrained themselves, not pushed the envelope of universal jurisdiction quite so hard, in hopes of not triggering a back- lash by states? By targeting highly visible current heads of state and by bringing dozens of cases at once, mostly in the two venues that seemed most accessible (Belgium and Spain), activist lawyers did indeed "over- load the boat" while the process was still new, highly controversial, and fragile. A more prudent strategy would have waited until the courts had

C op

yr ig

ht ©

2 01

0. U

ni ve

rs ity

o f P

en ns

yl va

ni a

P re

ss . A

ll rig

ht s

re se

rv ed

The Legal Legacy of Pinochet 197

enough experience with low-level defendants in more straightforward cases to make the more controversial ones seem firmly rooted in normal judicial practice. The result, in the short run, is a loss of momentum from the heady days of 1998.

In the end, though, the effect may be salutary, as courts and legisla- tures have to iron out some of the stickier points of transnational investi- gations and prosecutions, making it easier for other legislatures and courts to follow in their wake. Futhermore, the ebbs and flows of trans- national prosecutions follow the natural evolution of any new regime. Linear progress is seldom the way of such things.

Indeed, by 2003 a number of universal jurisdiction cases involving rel- atively low-ranking offenders living in Europe were underway. In part, these cases were newly possible because the crimes took place after the date of implementation of the Torture Convention in most European states. For example, in September 2003, Dutch authorities arrested Sebastian Nzapali, a former colonel in the Zairian army under Mobutu known then as the "king of the beasts," and accused him of torture and rape committed in 1995 and 1996, using the 1988 implementing legisla- tion to the Torture Convention. Nzapali, who had been living in the Netherlands since 1998, was tried and convicted on the torture charges in Rotterdam in March 2004. He was acquitted on the rape charges and sentenced to thirty months in prison.

In addition to Ely Ould Dah, the Mauritanian alleged torturer, cases in France include two Algerians, the mayor of a small town and his brother, members of the Algerian Relizane Militias. The two were indicted on March 30, 2004, by a Nimes investigating judge for torture and crimes against humanity stemming from a terror campaign against civilians between 1994 and 1997. The two brothers had moved to France, and the prosecutor initiated an investigation in 2003, bringing witnesses from Algeria to testify before the magistrate. The brothers have been released on bail awaiting trial.

Some progress has indeed been made. The Pinochet cases established the legitimacy of transnational prosecutions based on both universal and passive personality jurisdiction, at least under some circumstances. They showed that the existing universal jurisdiction laws could actually be used, and touched off a new willingness by advocates and courts to use them. They made clear that there are some limits to the immunity of government officials when hauled before national courts accused of international crimes, even if we still debate exactly where those limits are. They strengthened the idea that proper accountability for such crimes is the business of justice everywhere, and that domestic laws enshrining unfair trials or shielding perpetrators are subject to outside scrutiny and cannot per se bind foreign courts. They yielded landmark

C op

yr ig

ht ©

2 01

0. U

ni ve

rs ity

o f P

en ns

yl va

ni a

P re

ss . A

ll rig

ht s

re se

rv ed

198 Chapter 7

jurisprudence in the highest national courts of a handful of countries, jurisprudence that both draws from international courts and ideas and feeds back into them.

Yet there has been perhaps less progress than might be hoped for. The outer limits of what law can do when confronted with power are not fixed, but they do take time to expand: for now, current heads of state and figures who retain influence in powerful countries may be beyond the pale.

The danger is not that politically motivated courts will run amok, but that complainants will overreach. It is true that advocates and victim groups may have complex motivations for launching complaints-it is hard to doubt that some complainants in the Sharon case wanted to embarrass the Israeli government and remind the world of the past per- fidies of its leader. But so what? Most human endeavor is marked by mul- tiple agendas and interests. All that can usefully be said is that advocates, prosecutors, and human rights groups should exercise care in recogniz- ing and understanding these multiple agendas. The fact that they exist is not reason enough to limit the courts to cases involving unknown, powerless or uncontroversial protagonists, although it may counsel greater strategic selectivity.

In the best of circumstances, a cohesive group of lawyers and activists would plot out, step by determined step, how and where to bring the cases that would systematically expand the outer limits of international law. They might look for models to the way the NAACP plotted the fall ofJim Crow laws in the U. S. South over at least two decades in the mid- twentieth century."? But although there is communication and coordina- tion among many of the lawyers involved in these cases, the days are long gone when any single group can define and enforce limitations on appropriate cases. Someone out there will always want to try something new and audacious, even if it makes bad law. The resulting ups and downs are the price to be paid for a decentralized, global process largely driven by victims' groups. The very nature of transnational prosecutions makes them opportunistic, supplemental, ad-hoc. They will never be the only mechanism for achieving justice. But they are one piece of the emerging architecture, an architecture with a number of pillars.

Enter the ICC

The U.S. pressure on Belgium to gut its universal jurisdiction law seemed particularly galling because it dovetailed with larger U.S. efforts to derail international justice, particularly the implacable U. S. opposi- tion to the nascent International Criminal Court, which had at the time barely appointed judges and a prosecutor. The advent of the ICC

C op

yr ig

ht ©

2 01

0. U

ni ve

rs ity

o f P

en ns

yl va

ni a

P re

ss . A

ll rig

ht s

re se

rv ed

The Legal Legacy of Pinochet 199

changed how transnational investigations and prosecutions fit into the larger picture.

The International Military Tribunal at Nuremberg was not the first international court to try war crimes: the first such trial was in the Mid- dle Ages. But in the wake of World War II the idea took hold that there should be a way to try individuals for the worst crimes even when their home state was not willing or able to do so. The 1948 Genocide Conven- tion contemplated an international criminal court, but the idea soon foundered on the rising Cold War. It wasn't until the heady days of the post-Communist, post-Gulf War New World Order that the idea was suc- cessfully refloated. Trinidad and Tobago proposed a court to try drug dealers and war criminals, but the drug dealer idea morphed over the course of the 1990s into a court that would deal with a very small group of very bad international crimes: genocide, crimes against humanity, and certain war crimes. The International Law Commission prepared a draft and the UN set up a Preparatory Commission, which worked for nearly three years. At the end, in a marathon five-week session in June andJuly 1998, the delegates of 120 countries and some 175 nongovern- mental organizations hashed out the Rome Statute creating an Interna- tional Criminal Court. Four years later (record time by international treaty standards), the required sixty countries had agreed to sign on to the Court, and it formally came into being."

The drafters of the ICC statute were not working completely from scratch. They had a wealth of experience to draw on from the two ad hoc international criminal tribunals. These tribunals had been one of the first innovations in a post-Cold War world. As conflict raged in the Balkans in the early 1990s and TV viewers recoiled to see emaciated con- centration camp inmates in the heart of a Europe that had in living memory vowed "never again," the UN Security Council created an International Criminal Tribunal (ICTY) to investigate andjudge individ- uals charged with war crimes, crimes against humanity, or genocide in the former Yugoslavia. A year later, the slaughter of more than three quarters of a million Rwandans, mostly of the Tutsi minority (and the rest majority ethnicity Hutus who opposed the killing), and most by their neighbors using hoes and machetes, led to calls for the UN to treat African genocide as seriously as European. The result was the Interna- tional Criminal Tribunal on Rwanda (ICTR), based in Arusha, Tanzania. Both ad hoc tribunals, as Security Council creations (under the Coun- cil's Chapter VII powers to act in the face of threats to peace), could compel the cooperation of any UN member state-assuming the Coun- cil and its members were willing to back up the tribunals.

It is not easy to create a new court from scratch and imbue it with the necessary law, procedure, and rules of evidence. These had to be a blend

C op

yr ig

ht ©

2 01

0. U

ni ve

rs ity

o f P

en ns

yl va

ni a

P re

ss . A

ll rig

ht s

re se

rv ed

200 Chapter 7

of the world's major legal systems, especially the civil law systems of Europe (and ex-European colonies like Rwanda) and the common law systems that were familiar to many of the judges, prosecutors, and financial backers of the tribunals The tribunals also had to have offices, judges, clerks and bookkeepers. The office of the prosecutor needed its own lawyers, investigators, vehicles, computers, translators, and the like. Defense counsel had to be recruited from the private bar. Cooperative arrangements had to be made for seeking and requesting the surrender of suspects and for a jail to hold them in. Appropriate penalties had to be devised. This is not the place to tell the story of how all those things happened, with their fits and starts, ups and downs, in the history of the two ad hoc tribunals. They did happen, and the cumulative experiences of setting up and running the tribunals gave a big boost to the architects of the ICC.42

Like the ad hoc tribunals, the ICC sets up a three-part structure, with a panel of eighteen judges who represent all regions, men and women, experts in criminal law and international law. A prosecutor's office along with the Security Council and a State Party can decide to initiate investi- gations into any set of events happening after (but not before) July 1, 2002. A Registry administers the Court. The Security Council can request investigation into any situation that it decides comes within the Court's mandate; if the Prosecutor or a State Party wants to start an investigation, either the state where the crimes allegedly took place or the state of nationality of the defendant has to be a party to the treaty (or accept the investigation on a one-time basis). The prosecutor, as a further check against any overly aggressive tendencies, must also get per- mission from a panel ofjudges before starting an investigation on his or her own. There are no official immunities and no statutes of limitations. Commanders are responsible for their subordinates' acts, and obeying orders is a defense in only very limited circumstances. The Rome Statute also sets out elaborate rules for the conduct of an investigation and trial, and for appeal.

Unlike the ad hoc tribunals, the ICC will not have first pick of the cases it wants to try. Instead, it will act as a backstop, stepping in only if national courts are "unable or unwilling" to investigate. The definition of "unwilling" in article 17 of the court's statute includes cases involving sham proceedings, excessive delays, a compromised judiciary, or other cases of apparent shielding of potential defendants. "Unable" is more straightforward: a state is unable to act if "due to a total or substantial collapse or unavailability of its national judicial system, it is unable to obtain the accused or the necessary evidence and testimony."

In the best possible scenario, the Court will have almost no business at all. The threat of an ICC investigation will be enough to push domes-

C op

yr ig

ht ©

2 01

0. U

ni ve

rs ity

o f P

en ns

yl va

ni a

P re

ss . A

ll rig

ht s

re se

rv ed

The Legal Legacy of Pinochet 201

tic courts to look into the alleged crimes and domestic governments to give the courts the leeway to do so, hoping to avoid the stigma of having its courts labeled inoperative. In a sense, the Court's objective is the same as that of a transnational investigation: to jumpstart stalled or obstructed domestic justice. Its main weakness lies in its jurisdictional requirements. The state of either nationality or territoriality (usually the same) must be a party to the treaty; violators who stay at home are only at risk if the Security Council wants them or their own government turns them in (or perhaps if their government in an earlier era became a party and the current regime neglected to back out). A Pinochet or a Saddam Hussein is likely to remain beyond the reach of the court, at least while still in power. The Court's main business will likely come from newly empowered "post-transition" governments wanting the internationals to bear the cost, and the political heat, of prosecutions of deposed and disgraced ex-leaders. The number of countries that have recently suf- fered from civil wars ratifying the ICC Statute indicates that at least some post-conflict governments see the Court's potential along these lines.

Does the advent of the ICC make transnational prosecutions obsolete? In a word, no. The significant limitations on the court's jurisdiction, in terms of both timing and state consent, mean that many cases will never even get near the ICC. Even for those cases that are technically within the court's purview, the Prosecutor will have to focus on a few crucial situations, leaving many cases untouched. Resources are likely to be inadequate. At the same time, the heightened expectations created by the advent of the Court are likely to vastly increase the demand for jus- tice, though it may take some time for national courts to take up the challenge. Transnational prosecutions have the potential to fill the gap, to take up the many cases the ICC will be unable to hear, and to act as an adjunct and multiplier of the emerging international criminal juris- prudence.

There is nothing in the Rome Statute that would keep the national courts of third-party states from acting, and some indication that their action takes precedence over the ICC. Article 17 of the ICC statute talks about "a State which has jurisdiction over" a case as having first crack under the complementarity provisions. So if a case were properly before a transnational court on the basis of universal (or victim-nationality- based) jurisdiction, that state could, by choosing to go forward, oust the ICC of jurisdiction. This makes sense in an overall scheme where the ICC remains the court of last resort. Transnational courts act as a second line of defense, allowing the ICC to focus on those cases where, for instance, head of state immunity allows only an international court to proceed. Only if the burden is thus distributed does the ICC have a

C op

yr ig

ht ©

2 01

0. U

ni ve

rs ity

o f P

en ns

yl va

ni a

P re

ss . A

ll rig

ht s

re se

rv ed

202 Chapter 7

chance of meeting the (unrealistically?) high expectations now heaped on it.

Institution-Building and Networking Approaches

A two-track approach to internationaljustice seems to be emerging. On the one hand are the new institutions of global justice, built from the ground up. On the other is the networking approach, featuring a (to date still small) group of existing national courts and webs of interna- tional and national lawyers and organizations that coalesce around spe- cific transnational prosecutions. The two approaches largely complement each other, but they do have different strengths and weaknesses. Along- side these approaches are an increasingly varied array of noncriminal mechanisms.

As the International Criminal Court, the pinnacle of the "institution- building" approach, comes on line, the Pinochet cases provide important lessons and insights that can help the new court be useful to the victims of terrible crimes. The "institution-building" approach to global justice is of a piece with other twentieth-century institution-building, including the United Nations itself, the International Court of Justice, the World Trade Organization, the World Bank and International Monetary Fund, and a host of specialized international and regional agencies and courts. Institution-building is integral to channeling and controlling the unbri- dled global marketplace, and to coordination and orderly relations among states. States voluntarily surrender a small (in reality, smaller for some countries than others) proportion of their sovereignty to these institutions in exchange for benefits and a say in their management, runs the theory. These institutions each require new rules (created by treaty), new personnel, new dispute resolution procedures, entire new international regimes. They are an integral part of the phenomenon we often refer to as globalization.

Globalization, of course, has its discontents. Critics complain that new global institutions are undemocratic, that they privilege rich countries and rich interests within rich countries, that they ignore the effects of their decisions on the poor and on the planet's health, and that the underlying social contract binding government to the governed at the national level does not exist internationally. Others excoriate the bureaucracy, enforced mediocrity, and sluggishness of global institu- tions. They demand a "globalization from below," a bubbling up of ideas and policy choices from civil society in every particular place, to balance and legitimize the top-down inaccessibility of many global insti- tutions.

Institutions of global justice, no matter how noble and well-inten-

C op

yr ig

ht ©

2 01

0. U

ni ve

rs ity

o f P

en ns

yl va

ni a

P re

ss . A

ll rig

ht s

re se

rv ed

The Legal Legacy of Pinochet 203

tioned, may well suffer from many of the same defects as other top-down global efforts. As a treaty-based institution, the ICC is a creature of states, subject to the least-common-denominator politics of state consensus and the power of large states to threaten and bribe where necessary. The formal independence of the judges and prosecutor insulates the Court to some extent, but cannot change the fact that it will be dependent on states for its budget and its muscle. U.S. opposition to the Court only exacerbates these problems.

Top-down institution-building efforts may also suffer from the lack of grounding in a specific local reality. The experience of the ad hoc tribu- nals provides some indication of the kinds of problems we might encounter. Both tribunals have suffered from being far, geographically and culturally, from their "target" populations. A lack of publicity (and in parts of the former Yugoslavia, heavily self-censored media during the 1990s) and of radio coverage of the trials meant that for the average Bosnian, Serb, Croat, or Rwandan the tribunals were no more than a far- off, unintelligible abstraction, if that. Add to these the problems inher- ent in any criminal proceeding: they are drawn-out, focused on detail, and hard to make accessible to the public. There may well have been a connection between the tribunals' activities and, say, the inability of cer- tain local thugs (who might worry about Tribunal indictments) to con- tinue to hold local office-but it is not clear that connection was apparent to the average townsperson.

The ad hoc tribunals have also been a mixed blessing for domestic efforts and local institutions. The local courts have not been the benefi- ciaries of anywhere near the resources and expertise they needed to become meaningful participants injustice at home. In the former Yugo- slavia, few domestic prosecutions occurred and those that did were prob- lematic. By 2002, it was clear that domestic prosecutions were necessary, but there were continuing squabbles about who should be in charge and how to ensure the safety and security of witnesses.

In Rwanda, these problems were even worse. The tribunal seat was in Arusha, Tanzania, a long, bumpy jeep ride (or, for those few who could afford or even imagine it, a short plane ride) away. Proceedings were in English and French, and translators to the local Kinyarwandan usually failed to make the context of motions and holdings intelligible to the local population. While the tribunal spent tens of millions on trials of most of the top ringleaders-as of the end of 2002, eleven cases had been completed and cases against sixty others were in progress or pend- ing-the local courts confronted the impossible task of sorting out over 100,000 prisoners detained in festering local jails.

For witnesses and the families of victims, the tribunals could prove frustrating. Exhumations and interviews of family members, asked to

C op

yr ig

ht ©

2 01

0. U

ni ve

rs ity

o f P

en ns

yl va

ni a

P re

ss . A

ll rig

ht s

re se

rv ed

204 Chapter 7

recall details of their loved ones' clothing and characteristics, for instance, were carried out with the sole purpose of evidence-gathering, leaving family members to complain that their need to reclaim the body for proper burial was decidedly secondary." Witness protection was offered, but only while the witness was testifying. In Rwanda, a number of potential witnesses were killed before investigative methods improved. The prosecutorial strategy has been largely a "top-down" one, where the victims and witnesses play little active part. Relations between the Rwanda tribunal and victims' groups in the country have been strained at best. A recent report found that women rape victims were especially upset that prosecutors had never prepared them ade- quately for the rigors of cross-examination, and had not explained how they fit into prosecutorial strategy. Judges and court personnel, they felt, were disrespectful and aloof. A 2003 study of witnesses before the Yugo- slav Tribunal also pointed out areas where prosecutors had been seri- ously neglectful of their witnesses."!

Victims at the Center

It is striking to compare the mixed impact of the ad hoc tribunals on victims and on local justice processes with the seemingly much greater and less ambiguous impact of the transnational investigations recounted in this book. Those transnational efforts, done at a fraction of the cost, relied for their effectiveness on existing national courts and on a net- work of complainants, NGOs and lawyers in a number of countries. Moreover, the impact seemed to occur even when the defendants were indicted or charged but unlikely to ever be extradited for trial.

Why? One answer has to do with the agency of victims and survivors. Rather than play passive roles in litigation driven by prosecutors, the vic- tims and witnesses, and their organizations and attorneys, were the driv- ing forces behind the cases. Once the cases were open, people spontaneously wrote to the investigating magistrate telling their story or asking if they could testify if they traveled to see him. There were limits to who could participate-a few would-be complainants were turned down on grounds that no one in the human rights community had ever heard of them and they were suspected of simply wanting access to the case file-but they were very broad limits. The cases stirred imaginations and opened possibilities precisely because they seemed decentralized, less controllable by state interests, more, if you will, acts of imagination.

Structural characteristics of these transnational prosecutions led to this sense of empowerment. The cases used existing courts and judges and needed no special political will on the part of states to proceed. Set- ting up international institutions requires resources and support that

C op

yr ig

ht ©

2 01

0. U

ni ve

rs ity

o f P

en ns

yl va

ni a

P re

ss . A

ll rig

ht s

re se

rv ed

The Legal Legacy of Pinochet 205

can only come from states, but that also makes the institutions (even with adequate safeguards for their formal independence) dependent on state will and appropriations. In contrast, at the national level investiga- tions can be undertaken as a matter of routine, where legal factors are more likely to predominate over the political ones, at least initially. The onus then is on the executive branch of a state, in particular the foreign affairs office, to step in and derail an investigation, not to put into action the machinery to start one. Although it can be done (as the Senegalese Habre case shows), there is a political price to pay for appearing to inter- fere with the independence of the courts. Eventually, the executive branch can weigh in, at the stage of extradition or the appeal of an indictment, but at that point the initial fact-finding, the outlines of the case, have already been laid down.

All the cases studied here involved continental systems that allow indi- viduals to initiate an investigation and to take part through the mecha- nism of the partie civile. In a 1997 report to the UN on principles to combat impunity, the Human Rights Sub-Commission reporter, Louis Joinet, endorsed this approach. "Although the decision to prosecute lies primarily within the competence of the State, supplementary procedural rules should be introduced to enable victims to institute proceedings, on either an individual or a collective basis, where the authorities fail to do so, particularly as civil plaintiffs. This option should be extended to nongovernmental organizations with recognized long-standing activities on behalf of the victims concerned."45 It is disturbing to note that the trend in Europe since 2002 has been in the opposite direction, to remove victims' initiative and place discretion to begin investigations entirely in the hands of a state prosecutor.

Common law countries do not have investigating magistrates who look into charges of crimes, but prosecutors who have full discretion to decide whether to press charges. Far fewer transnational prosecutions are likely to arise in these systems. Nonetheless, we allow some measure of the victim autonomy provided by the partie civile system by encourag- ing tort suits (for wrongful death or injuries suffered) in cases of crime. In the United States, Alien Tort Claims Act suits, like the one against Caravan of Death member Fernandez Larios, seem to provoke some- thing of the same positive response in plaintiffs. The use of these mecha- nisms and the development of similar ones in other common-law countries need to be protected and expanded.

In part in reaction to the experiences of the ad hoc tribunals, the ICC takes a more victim- and witness-friendly approach. There is a Victims and Witnesses Unit, in charge of witness security and protection and assistance to victims and families. Victims can be represented before the court and can ask to present their views at trial. They can also be present

C op

yr ig

ht ©

2 01

0. U

ni ve

rs ity

o f P

en ns

yl va

ni a

P re

ss . A

ll rig

ht s

re se

rv ed

206 Chapter 7

at preliminary hearings, make their views known, and be notified if the prosecutor decides not to go forward. While not quite the same as the partie civile procedure, it is also a long way from the prosecutorial model of the common law and the ad hoc tribunals.

The judges will playa large role in regulating how and to what extent victims participate during pre-trial procedures and at trial, including whether and how they can question witnesses and access documents. The Rules of Procedure establish the presumption that victims will be able to be heard, receive copies of the case file, and attend hearings." Unlike the ad hoc tribunals, the ICC rules reflect an increasingly-shared view that the interests of the victims may not always line up with those of the prosecution, and that victims need their own voice and defenders. Where there are many victims (with many possible lawyers) the victims can be asked to agree on joint representation, and if they cannot agree, the registrar can do it for them. During a subsequent reparations hear- ing, victims (and their attorneys) will be able to participate fully.

For the first time, the ICC envisions a central role for reparations to victims. Article 75 of the Rome Statute tells the Court to craft rules on reparations for victims. The Court can, upon request or on its own motion, decide on the amount of damages and the form of reparations. Reparations will be paid by individual perpetrators, but not by states. The statute also creates a trust fund for victims, which can be used to distribute awards, including collective reparations, when there are large numbers of victims. Judgments, however, will have to be collected from national courts, and it is not clear how that will work. It is also not clear where the money will come from (perhaps some perpetrators have findable secret bank accounts, but how many?), or whether the Court will have the capacity to identify victims and distribute funds.

It is much too early to tell how these provisions will be used in prac- tice. If taken seriously by judges and prosecutors, they could go quite some way toward bringing the international proceedings closer to the people most affected by the crimes. There is a large amount of discre- tion left to the judges, prosecutor, and registry to craft, in practice, the Court's relationship with victims and their representatives. There will be inevitable tensions between the needs of prosecutors and the needs of victims and witnesses. The transnational cases teach that the Court must take victim empowerment and agency seriously in everything it does, and especially during the pre-trial, investigative phases, if it is to truly aid victims and their communities.

Much will depend as well on whether the ICC sees transnational cases as allies, competitors, or even vestiges of a now unnecessary past. The Court should use the advantages of a decentralized approach. Decen- tralization allows a number of experienced investigative judges, each

C op

yr ig

ht ©

2 01

0. U

ni ve

rs ity

o f P

en ns

yl va

ni a

P re

ss . A

ll rig

ht s

re se

rv ed

The Legal Legacy of Pinochet 207

working with the same corpus of international law but with their own national procedures and implementing acts, to develop a jurisprudence of international criminal justice "from below." This jurisprudence will be, no doubt, heavily influenced by the work of the ad hoc and other international criminal tribunals and the ICC, but it will multiply the number of unanswered questions that can be tackled. The decentral- ized, organic development of jurisprudence is likely to generate the kinds of debate and even disagreement among courts, based on con- crete factual situations, that over time allows the working out of complex issues of law and policy. It also corresponds to the way other enforce- ment mechanisms have evolved in the human rights field: through mul- tiple, overlapping forums, where the number of possible routes of redress offsets the weaknesses of each individual mechanism and allows events to be viewed through multiple prisms. At the dawn of a new enforcement regime, this approach may better lead to bringing multiple talents to bear and devising solutions that have been tested in practice.

Admittedly, fostering a proliferation of transnational cases reduces the chances that like cases are treated alike (at least during some initial period), and runs the risk of creating bad national precedents. The trade-off, though, is a large number of domestic judges grappling with international criminal law and procedure, internalizing and domesticat- ing this law and the human rights and humanitarian law that comes with it. That domestication is likely to lead to wider and better application of human rights law and international criminal law more generally. It will better put perpetrators and would-be perpetrators on notice of where the limits are. It may also prove a far greater deterrent to crimes against humanity and their ilk if would-be genocidaires know that their acts are being watched, and may be judged, in many different places, subject to many different sets of political and legal constraints, not just one.

C op

yr ig

ht ©

2 01

0. U

ni ve

rs ity

o f P

en ns

yl va

ni a

P re

ss . A

ll rig

ht s

re se

rv ed