1

The Justice Cascade: Human Rights Prosecutions and Change in World PoliticsChapter 1: Introduction

Kathryn Sikkink

I. Introduction: From the Unimaginable to the Commonplace

In the spring of 2009, when President Barack Obama, the Congress, and the

American public started debating whether or not any Bush administration officials should

be prosecuted for the use of torture, the United States joined a long list of countries that

have considered holding individual state officials criminally accountable for past human

rights violations. This represents a very new development in world politics. Only 30

years ago, it was virtually unheard of, almost unimaginable, for a national or international

tribunal to hold state officials criminally accountable for human rights violations.

The Nuremberg and Tokyo trials after WWII, and the “successor trials,” through

which countries tried the instigators and collaborators of Nazi atrocities during the war,

were in many ways the exception that proved the rule: only in cases of complete defeat

in war was it possible to hold perpetrators criminally accountable for human rights

violations.

In the mid-1970s when military coups proliferated in Latin America, the new

military leaders and their civilian opponents rarely mentioned or considered the

possibility that state officials could be held legally accountable for their actions. In

1981, on an official visit to Washington, military “president designate” of Argentina,

General Roberto Viola, argued, "A victorious army is not investigated. If the Reich's

troops had won the last World War, the Tribunal would have been held not in Nuremberg

but in Virginia." 1 Four years later in 1985, General Viola was prosecuted in Buenos

Aires in an unprecedented trial of top military leaders for human rights violations. Over

the next 25 years, he would be joined by other authoritarian leaders in the region. Ex-

President Alberto Fujimori of Peru was sentenced in 2009 to 25 years in prison for

human rights violations. Ex-President General Ríos Montt of Guatemala is the subject of

1 March 22, 1981, as cited in Iain Guest, Behind the Disappearances: Argentina's Dirty War against Human Rights and the United Nations (Philadelphia: University of Pennsylvania Press, 1990), p. 277.

2

an ongoing trial in Spain, while the former president of Uruguay, General Gregorio

Álvarez, was sentenced to 25 years for the murder of 37 people. Another former

authoritarian president of Uruguay, Juan María Bordaberry, was indicted in 2006 for the

murder of his political opponents. A brief discussion of Bordaberry’s case serves to

illustrate how unexpected such changes were for all the protagonists.

Bordaberry won the presidency of Uruguay in democratic elections in 1971, but

two years later, faced with economic crisis, an increasingly violent urban guerrilla

movement, and political stalemate, he carried out a "self-coup" with the support of the

Uruguayan military, closing the Congress and the Supreme Court. After the coup,

Bordaberry continued to serve as president, and his participation gave a veneer of

legitimacy to the new authoritarian regime, the first interruption of democracy in

Uruguay since the 1930s. After Congress was closed, three of the leading opponents to

Bordaberry’s coup went into exile in Argentina where a precarious democratic regime

was still in power: Senator Wilson Ferreira, the candidate for the Blanco Party who had

received the most votes in the 1971 elections; Héctor Gutiérrez Ruiz, the Speaker of the

Chamber of Deputies; and Senator Zelmar Michelini, one of the most eloquent

spokesmen of the leftist opposition coalition, the Broad Front. The exiles’ situation took

a grim turn with the Argentine military coup in March 1976. The relative safety of their

exile now turned dangerous, and their hope of a quick return to a democratic Uruguay

paled as they saw yet another country in the region revert to military rule. They started to

receive death threats. Michelini planned a trip to the United Stated where he intended to

meet with members of Congress and staff from Amnesty International to bring them up to

date on what was happening in Uruguay. The plane ticket was reserved, and the U.S.

government had already granted him the visa. But the Uruguayan government, following

explicit instructions from Foreign Minister Juan Carlos Blanco, refused to renew

Michelini’s passport, so he was unable to travel. Without current documentation, his

status in Argentina was ever more precarious. He wrote to a friend in the United States,

“Since I don’t have legal status in Argentina, I am now practically a man without a

3

country, a prisoner in this land…. You can’t imagine my anger, my grief, my

impotence….” 2

Around this time, the national intelligence agencies of Chile, Paraguay, Bolivia,

Uruguay and Argentina had begun to collaborate secretly in repression and assassination

of their political opponents, under the code name of “Operation Condor.” Because the

dictatorship of General Augusto Pinochet had taken the lead in forming Operation

Condor, it was named after the national bird of Chile. The designation is more evocative

than the founders intended: although the Condor is magnificent in flight, it is a vulture

and feeds almost exclusively on carrion. After the military coup in March of 1976,

Argentina stepped up its participation in Condor, collaborating with Chilean and

Uruguayan agencies to coordinate abductions, torture, disappearances, and murder,

especially of exiles from other South American countries who had who had sought refuge

in Argentina.3

In what we now know was a Condor joint operation, on May 18, 1976, groups of

armed men kidnapped Gutiérrez Ruiz and Michelini from their apartments in Buenos

Aires. Three days later, both men were found murdered, on their bodies signs of torture.

When the kidnappers bungled a similar attempt to capture Wilson Ferreira, he fled to

London. He left an open letter to the new military President of Argentina, General

Videla, describing at length the kidnapping and murder of Gutiérrez Ruiz and Michelini.

The letter ended, “When the hour arrives of your own exile, which will arrive, have no

doubt, General Videla, if you seek refuge in Uruguay, a Uruguay whose destiny will be

once again in the hands of its own people, we will receive you without cordiality and

affection, but we will grant you the protection that you did not give to those whose deaths

we are today grieving.”4

2 César di Candia, “Angustias y Tensiones de Tres Políticos Uruguayos Condenados a Muerte,” reprinted from “Qué Pasa,” supplement of El País, November 13, 2004, in Brecha, “Documentos: Para el Juicio de la Historia” 20 of October, 2006, p. 7-8, translation by author.

3 Kornbluh, Peter. The Pinochet File: A Declassified Dossier on Atrocity and Accountability . New York: The New Press, 2003.4 “Carta de Wilson Ferreira Aldunate a Jorge Rafael Videla, Buenos Aires, May 24, 1976, reprinted in: Brecha, “Documentos: Para el Juicio de la Historia” 20 of October, 2006, p. 31-39,translation by author.

4

Wilson Ferreira, in his hour of greatest despair, could imagine a distant future of

democracy in his country and Argentina when General Videla would be forced into exile

and Uruguay would be again in the hands of its people. But, at that time, he could not

imagine accountability for the human rights violations he had just witnessed. He did not

say that he awaited a time when Videla would be extradited or tried for his crimes.

Ferreira assumed that in the future, as in the past, leaders responsible for human rights

violations would go into an exile abroad where they would be given “protection” if not

affection. This had long been the political tradition in Latin America, where military

coups were a commonplace instrument of politics. Even the very viciousness of the

crimes he had just witnessed did not change Ferreira’s perception that leaders would not,

or could not, be judged for their crimes.

Less surprisingly, President Bordaberry also could not imagine the possibility of

accountability or judgment. In a speech in December 1974, he spoke of the military and

the coup of June 1973, “…[T]he Armed Forces must have a supreme tranquility,

knowing that their stance of having accompanied and supported the government in the

historic events of June 1973, can not be judged by the citizenry…. It would be like

assuming you could judge a man who broke the formal law to defend his mother, in this

case, his motherland. And this stance cannot be object of a judgment.”5

Thirty-two years after this speech, Bordaberry found to his shock and dismay that

he himself could in fact be the object of judgment. In November 2006, Uruguayan Judge

Roberto Timbal indicted Bordaberry and his Minster of Foreign Affairs Juan Carlos

Blanco, ordering them into preventive prison to await trial for the murder of Gutiérrez

Ruiz and Michelini. The human rights community and the general public in Uruguay had

never expected to witness such an outcome.6

The story of Jorge Bordaberry is just one of many stories of individuals for whom

the unimaginable had occurred: state officials accused of past human rights violations

were being held criminally accountable. A watershed moment came in 1998, when

5 From an end of year speech by Bordaberry, cited in Brecha, “Documentos: Para el Juicio de la Historia” 20 of October, 2006. p. 5. (Translation by author) 6 In interviews in Montevideo in 2004, just two years before the indictment of Bordaberry, members of the human rights community in Uruguay expressed doubt that legal accountability for past human rights violations could ever take place in Uruguay.

5

General Augusto Pinochet of Chile was arrested in London by British police executing a

Spanish extradition request. The Spanish court wanted Pinochet to stand trial in Spain

for crimes committed in Chile during his military dictatorship. No one predicted this

arrest, certainly not in London, where Pinochet had just weeks previously taken tea with

his old friend Margaret Thatcher. Even the most wild-eyed idealists of the human rights

movement did not foresee that Pinochet might be arrested in London for crimes he had

committed years earlier in Chile. They knew it was legally possible, but they still

believed it was politically unimaginable.

After Pinochet was brought into custody, many predicted that it would lead to the

demise of democracy in Chile because his supporters would carry out another coup. But

when Pinochet died in 2006 facing prosecutions in his home country for human rights

violations and for corruption, democracy was alive and well in Chile.

Some analysts offered similarly dire predictions in 1999, when President

Slobodan Milosevic of Yugoslavia was the first sitting head of state to be indicted for war

crimes. Critics argued that actions taken by the International Criminal Tribunal for the

Former Yugoslavia (ICTY) in the Hague would lead Milosevic to entrench himself in

power and prolong the agony in the Balkans. But Yugoslavs began to realize that

Milosevic was a liability and voted him out of office less than a year and a half after his

indictment. Facing intense pressure from the United States and the European Union, the

new Yugoslav government eventually extradited Milosevic to The Hague to stand trial,

where he died before he could be convicted for his crimes. In March 2003, a second

sitting president was charged for war crimes, when the Special Court for Sierra Leone

indicted Charles Taylor, the ex-warlord president of Liberia. Taylor went into exile in

Nigeria in August of 2003 after his forces were defeated by a rebel group. In 2006, he

was turned over to the Special Court and is currently on trial for his crimes. In 2009, the

International Criminal Court (ICC) indicted a third sitting president, Omar al-Bashir, for

war crimes and crimes against humanity in the Darfur region of Sudan.

As these examples make clear, there is a dramatic new trend in world politics

toward holding past state officials individually criminally accountable for past human

rights violations. Sometimes, as in the examples above, even heads of states are facing

prosecution for human rights violations carried out during their rule. More often lower-

6

level state officials are being prosecuted. These prosecutions are taking place in

international tribunals, like the trial of Milosevic in The Hague; in foreign courts, like the

trial of Pinochet in the United Kingdom and Spain; and mainly in the domestic courts of

the country where the human rights violation occurred, like the trial of Bordaberry in

Uruguay.7 “Domestic prosecutions” are those conducted in a single country for human

rights abuses committed in that country. “Foreign prosecutions” are conducted in a

single country for human rights abuses committed in another country. Finally,

“international prosecutions” take place when multiple states, typically acting on behalf of

the United Nations, set up tribunals such as the ICTY and the ICC. The international

prosecutions category also includes so-called “hybrid” criminal tribunals, such as those in

Cambodia, Sierra Leone, and Timor-Leste (formerly East Timor), which combine

international and national features. For example, in Cambodia in 2009, a hybrid tribunal,

the Extraordinary Chamber for Cambodia, put Kaing Guek Eav, or "Duch", on trial.

Duch was the head of the notorious prison, Tuol Sleng, where from its rise to power in

1975 to its fall in 1979, the Khmer Rouge government imprisoned, tortured and killed an

estimated 17,000 people.

II. The Justice Cascade

I argue that these international, foreign, and domestic human rights prosecutions

are all part of an interrelated trend in world politics towards greater accountability – a

trend that Ellen Lutz and I have called the “justice cascade.”8 The justice cascade is one

example of what Cass Sunstein has called a norm cascade: “a rapid, dramatic shift in the

legitimacy of norms and actions on behalf of those norms.”9 The justice cascade is a

dramatic shift in the legitimacy of the norms of individual criminal accountability for

7 Even when human rights trials are held in domestic courts, it may not be a purely domestic phenomenon. Sometimes countries rely on other countries to extradite former state officials back home to stand trial, as when Peru succeeded in getting Chile to extradite ex-President Fujimori to stand trial in Peru for both corruption and human rights violations. While domestic courts rely primarily on domestic criminal law, they may also use some aspects of international or regional human rights law to get a conviction. 8 Ellen Lutz and Kathryn Sikkink, “The Justice Cascade: The Evolution and Impact of Foreign Human Rights Trials in Latin America,” Chicago Journal of International Law Vol. 2, No. 1 (Spring 2001). 9 Cass Sunstein, Free Markets and Social Justice (NY: Oxford University Press, 1997).

7

human rights violations and an increase in actions (like prosecutions) on behalf of those

norms. It doesn’t mean that true justice will be done, just that the norm has new strength

and legitimacy, made apparent by how common it has become to put state officials on

trial. We call these phenomena cascades because, when they begin to catch on, they

move quickly to sweep along many actors in their wake.

Social life is full of these cascades, involving issues from the most trivial to the

most serious. Smoking in public in the United States, for example, was subject to a norm

cascade. In a relatively short time, smoking went from being a widely accepted practice

in every bar and working place of America, to being banned in most public places.

Initially, people had to work hard to persuade every legislator or City Council member

that a ban would drastically reduce health risks to their constituents. But at a certain

point, prohibiting smoking in public places reached a “tipping point,” and cities

everywhere began to ban smoking in this cascade fashion. Not smoking in public places

is now an accepted norm, “a standard of appropriate behavior,” in most of the United

States.

Cascade phenomena have been studied by scientists and popularized for the

reading public in books such as The Tipping Point, by Malcolm Gladwell.10 The tipping

point is the moment in the cascade where a critical mass of actors has adopted a norm or

practice, creating a strong momentum for change. Cascades with tipping points occur in

disparate phenomena from crime fighting to fashion; the book The Tipping Point, for

example, includes a long discussion of the rise in popularity and sales of Hush Puppy

shoes in the mid-1990s. In this book, I’m only interested in a subset of cascade

phenomena involving norms. Since a norm is defined as a standard of appropriate

behavior, it has a quality of “oughtness” that mere fashion does not. It is seen as the

“right thing” to do. Thus, the anti-smoking ban was indeed a norm cascade – it was a

dramatic shift in norms and practices about smoking – and it wasn’t only about science

or fashion, but about what cities and individuals ought to do.

10 Malcolm Gladwell, The Tipping Point: How Little Things Can Make a Big Difference (New York: Little Brown, 2000); for a discussion of “the life cycle of international norms” including norm “tipping points” see Martha Finnemore and Kathryn Sikkink, “Norm Dynamics and Political Change,” International Organization, 52, 4 (Autumn 1998).

8

Norms cascades also occur in international politics. An increasing number of

studies show that many international policies and actions diffuse: they are rapidly

adopted by many different countries for reasons that appear to have less to do with their

domestic politics or internal pressures, and more to do with the policies that other

countries are adopting. Increasingly, for example, the best explanation for the spread of

democracy to a country is how many other countries in the region are democratic, not

factors long associated with democracy like wealth of the country or its level of

education. There are similar explanations for the spread of policy innovations like neo-

liberal economic institutionalization and pension reform.11 It would seem that states, like

people, care about what other states are doing both as sources of new ideas and as

information about how to fit in with global trends.

Diffusion occurs when actions and choices in one country are “systematically

conditioned by prior policy choices” made elsewhere in the world.12 Some scholars even

talk about “virus” models of diffusion where one state “catches” a new policy just like

people catch a cold. Choices policy makers made about human rights trials were indeed

systematically conditioned by prior policy choices made elsewhere, but “diffuse” is all

too passive a verb to convey the concerted activity and struggle through which ideas

about justice have moved around the globe. One problem with the academic diffusion

literature is that real people are often absent from diffusion stories.13 At a tipping point,

disembodied ideas may travel rapidly around the globe, but at the beginning of any norms

cascade, specific people work hard to propose new ideas and policies, and they share

their ideas with others who carry them to new settings. The early adopters of new human

rights norms don’t get the ideas through the air like a virus, but always through struggle,

11 On the spread of democracy see Gleditsch, Kristian, Ward, Michael. "Diffusion and the International Context of Democratization." International Organization 60, no. 4 (2006): 911-33, and Mainwaring, forthcoming; on neo-liberal economic reforms, see Simmons, Beth and Elkins, Zachary. "The Globalization of Liberalization: Policy Diffusion in the International Political Economy." American Political Science Review 98 (2004): 171-89; on pension reforms see Weyland, Kurt. "Theories of Policy Diffusion: Lessons from Latin American Pension Reform." World Politics 57, no. 2 (2005): 262-95.12 Simmons, Beth, Dobbin, Frank, and Garrett, Geoff. "The International Diffusion of Liberalism." International Organization 60, no. 4 (2006), p. 787.13 {Graham, December 2008 #202}also argue that scholars need to pay more attention to who affects diffusion, including specific people who have varying preferences, goals and capabilities.

9

innovations, and often just plain luck or fortuitous circumstances. This book will tell the

whole diffusion story of human rights prosecutions from its beginnings with advocates in

Athens and Buenos Aires to the rapid spread of ideas and practices of individual criminal

accountability in the early 21st century.

The seeds of the justice cascade have been present for years – the legal

underpinnings were outlined in the Nuremberg Tribunals from 1945-1949. But the

Nuremberg precedents then lay dormant for decades. It was not until the mid 1970s,

around the time when Michelini and Gutiérrez Ruiz were murdered in Buenos Aires, that

the precedent was picked up, almost inadvertently, by the new democratic regime in

Greece when it put past state officials on trial for torture and murder. These

developments have provoked immense controversy in politics, in the academy, and even

in the human rights community itself. The controversy is about the very sources of

change, and whether these can be called “progressive” developments at all. For some

people, our whose society is already too litigious and places too much emphasis on

retribution and punishment. But others, and especially many victims, cannot get on with

their lives until they know that perpetrators of human rights crimes have been held

accountable. Most importantly, though, the debate is about the effects of such trials.

What difference does this all make? Can new efforts at accountability actually contribute

to improving human rights? Does they deliver justice? Or can trials actually undermine

democracy and threaten peace talks during civil wars? Do they offer such a thin and

compromised version of justice that the world would be better off without them?

III. Questions Animating the Book

In this book, I try to answer three main questions: (1) What are the origins or

sources of new ideas and practices about individual criminal accountability for human

rights? (2) How and why do these ideas diffuse regionally and internationally? (3) What

are the effects? After the introductory chapters, in Part 2, I take up the questions of the

origins of the justice cascade by looking at the first three cases of domestic trials in

Greece, Portugal, and Argentina. In Part 3, I explore diffusion by examining the spread

of these practices, first in domestic courts, and later in foreign courts and international

tribunals. The final section explores the effects of prosecutions on human rights practices

10

using both quantitative and qualitative evidence. Can human rights prosecutions help

prevent future human rights violations?

This is a book intimately related to a theme sometimes called “transitional

justice,” which involves efforts to address past human rights violations after a transition

to democracy. But this book is also about human rights more generally. Since WWII,

states and activists have devoted tremendous time and energy to creating new human

rights norms and law. In 1945, there were very few norms and rules about international

human rights; but by 2000, states had ratified dozens of diverse human rights treaties

from the Genocide Convention to the Convention on the Rights of the Child. But almost

all human rights treaties had relatively weak enforcement mechanisms: there were lots of

rules, but they didn’t have any teeth. There were few consequences for states and no

consequences for individuals who violated human rights. Many scholars and policy

makers have long argued that we wouldn’t expect these human rights treaties to make any

difference because they didn’t have any enforcement. They were aspirational, but many

repressive leaders were not swayed by statements of high principles, however legalized.

Now, in 2010, human rights prosecutions can be seen as a new kind of

enforcement of human rights law. They have some “teeth” because they can put

powerful people behind bars. If, as scholars claimed, human rights law didn’t work

because it lacked strength, this new form of enforcement should help improve

compliance with human rights law. In this sense, the book takes on bigger topics about

crime, punishment, and deterrence, and about the role of sanctions and enforcement in

compliance with rules and law.

Most broadly, this is a book about the explanations for major changes in

international politics. International Relations theory has been notoriously weak on

explaining change. For example, most international relations theorists failed to anticipate

the end of the Cold War, much less explain it satisfactorily.14 The justice cascade

represents an important change because it challenges two interrelated doctrines at the

core of world politics: sovereignty and sovereign immunity. One key element of

14 For a discussion of the inability of international relations theories to explain major changes in world politics, see Peter Katzenstein, ed. The Culture of National Security: Norms and Identity in World Politics (New York: Columbia University Press, 1996).

11

sovereignty is the idea that it is nobody’s business how a state leader treated his own

citizens. The doctrine of sovereign immunity in turn prevents a sovereign state or state

official from being prosecuted. Why would states undermine the very doctrines designed

to protect them from prosecution? Why wouldn’t they use their power to prevent this

trend? In this book, I will grapple with this important and puzzling change happening in

the world, and explore what it tells us about the nature and process of change in world

politics.

Historically, a huge gap existed between the treatment of crime in the domestic

and international realm.15 Domestically, there was a clear hegemony of the individual

criminal justice model, while this model was absent internationally. If an individual

killed one person, there was an expectation and an apparatus to permit that he would

stand trial for murder, and possibly be convicted and imprisoned. But if that individual

was a head of state and gave orders for thousands of individuals to be killed, the

expectation was that nothing would happen either domestically or internationally. When

their regimes were replaced by another, ex-dictators like Idi Amin of Uganda, Jean

Claude (Baby Doc) Duvalier of Haiti, or Alfredo Stroessner of Paraguay, traditionally

lived a comfortable exile without any expectation of facing criminal trials for human

rights violations committed during their regimes. Realist scholars of international politics

stress that it is difficult for international law to confront the power of states and their

leaders. How could a puny thing like law challenge the long practice that powerful

leaders could treat their citizens as they wished and still anticipate a secure retirement at

home or abroad? This is what General Pinochet believed when he traveled to Great

Britain for medical care. Slobodan Milosevic believed it too: even after he had been

indicted by the International Tribunal for the Former Yugoslavia, he initiated a campaign

of terror against the people of Kosovo, convinced the Tribunal could not touch him. And

this is also what George W. Bush, Dick Cheney, and Donald Rumsfeld apparently

believed when they ordered the use of kidnapping, indefinite detention without trial, and

torture against terrorism suspects after 9/11. Pinochet and Milosevic discovered that

15 This gap is a symptom of the broader “great divide” that saw domestic society as “rule bound” and the international system as anarchy. Ian Clark, Globalization and International Relations Theory (Oxford: Oxford University Press, 1999), p. 16.

12

human rights law was not as weak as they had believed, and both died facing prosecution

with a high likelihood of conviction.

The emergence of the individual criminal accountability model for basic human

rights violations means that the huge disjuncture between the treatment of crime in the

domestic and the international realms has started to narrow. Because we are in the midst

of change, it is sometimes difficult to stand back and see the process unfolding. I argue

that we are witnessing the initial emergence of a unified system of international

regulation of core political rights with often decentralized and fragmented enforcement,

primarily in domestic courts. Big differences continue to exist between the national and

the international criminal justice realms. In national systems, more or less effective

police forces arrest suspects and bring them in for questioning or prosecution. In the

international arena, no such international police force exists. As the world learned in the

case of the former Yugoslavia, even when international NATO forces were on the ground

keeping the peace, they were hesitant to expand their mandate to include arresting

indicted war criminal suspects to turn over to the international tribunal, the ICTY.

Despite these differences, important change is occurring, and for the most part, it

has been neither anticipated nor explained by political scientists. Many of my colleagues

are more skeptical. They would say that there have been all too many optimistic

assessments of this change in the world, and that we would be well advised instead to

“curb the enthusiasm” for transitional justice and provide reminders that sovereignty is

still alive and well, that power and self interest continue to govern international politics.

I argue that political scientists have erred on the side of caution and that, as a result, we

have been unable to foreshadow change or explain it. None of us expected Pinochet to be

arrested in London, Fujimori to be sentenced to 25 years in prison, Bashir to be indicted,

or, for that matter, an international criminal court to be up and running in the first place.

IV. A Little History: Three Models of Accountability

If we examine the history of the evolution of the human rights regime we can

think of three “models” of accountability that have been use for past human rights

violations: (1) the immunity, or “impunity” model; (2) the state accountability model;

and (3) the individual criminal accountability model. Under the impunity model, no one

13

is held accountable for human rights violations. Under a state accountability model, the

state is held accountable and it provides remedies and pays damages, while under a

criminal model, individuals are prosecuted, and if convicted, they go to prison.16

Prior to WWII, the “reigning orthodoxy” was the impunity model that said that

neither states nor state officials should or could be held accountable for past human rights

violations.17 There were isolated examples of accountability in ancient Greece and in

Revolutionary France, but no sustained attempts at domestic human rights prosecutions

until after the Second World War.18 At the international level, various pre-WWII

attempts at accountability for war crimes and mass atrocities failed to set up the necessary

institutions.19 The impunity model relied on a doctrine that said the state itself and

officials of the state should be immune from prosecution, both in domestic courts and

particularly in foreign courts. The intellectual history of the doctrine of sovereign

immunity traces it to various sources. Some say it stems from the ancient English

principle that the monarch can do no wrong, others from the inherent power of the state

to prevent prosecution. Some give functionalist explanations for immunity: governments

need to be protected from frivolous lawsuits so that they can concentrate on governing

and not be distracted from the tasks of office. Whatever the explanation for the doctrine

of immunity, prior to WWII it was almost unquestioned that state officials should be free

from prosecution for human rights violations, both in their own domestic courts and in

foreign courts or international tribunals.

The immunity model began to erode shortly after WWII. The Holocaust was the

shock that revealed the deep moral and political flaws of the reigning orthodoxy. States

and non-state actors realized that there was a complete lack of international standards and

accountability for massive human rights violations. To address this problem, they

16 Although I focus on individual criminal legal accountability, there is also an increase in individual civil legal accountability, especially in U.S. courts, where individuals found guilty of human rights violations are required to pay damages to their victims. These are cases brought mainly under the Alien Claims Tort Act, which permits tort claims for violations of international customary law.17 I borrow the term “reigning orthodoxy” from Jeffrey Legro to describe old ideas that are later replaced by new ones. {Legro, 2005 #212}18 See Jon Elster, Closing the Books: Transitional Justice in Historical Perspective. New York: Cambridge, 2004). 19 Gary Bass, Stay the Hand of Vengeance.

14

initiated action through the newly formed United Nations, first by drafting a set of

standards in the Universal Declaration of Human Rights in 1948, and later in a series of

more detailed human rights treaties. In these treaties, states proposed a new model of

accountability in which the state as a whole was held accountable for human rights

violations and was expected to take action to remedy the situation. But the state

accountability model went hand in hand with the idea that state officials themselves were

still immune from prosecution for human rights violations. So, for example, when a state

violated rights under the Convention on Civil and Political Rights, in some cases

individuals could bring petitions before the U.N. Human Rights Committee. But these

petitions were against the state itself, not a particular state official. States adopted a state

accountability model because it was the standard model used in other issues in

international law. But there was an uneasy fit between the state accountability model and

human rights that would be exposed in the coming years.

States negotiated and produced dozens of human rights treaties in the second half

of the twentieth century, most of which incorporate this state accountability model. It

began to be the “new orthodoxy” for accountability. It continues to be the model used by

virtually the entire human rights apparatus in the United Nations, including almost all of

the treaty bodies. It is also the model employed by the regional human rights

institutions--the European Court of Human Rights, the Inter-American Court of Human

Rights, and the new African Court of Human Rights. The European Court, for example,

found in 1978 the British government responsible for inhuman and degrading treatment

of Irish prisoners and called for it to change its interrogation policies.

Under this model, if the state refused to take action to change its policies or to

provide remedies to victims, there were few forms of enforcement available. Human

rights NGOs, international organizations, and other states mainly relied on reputational

accountability via moral stigmatization of state violators.20 This was the so-called “name

and shame strategy” of the human rights movement. Amnesty International, the U.N., or

20 I use Grant and Keohane’s definition of accountability that implies that “some actors have the right to hold other actors to a set of standards, to judge whether they have fulfilled their responsibility and to impose sanctions if they determine these responsibilities have not been met.” Legal and reputational accountability are two of the seven forms of accountability they discuss. Ruth Grant and Robert O. Keohane, “Accountability and Abuses of Power in World Politics,” American Political Science Review 99:1 (February 2005), pp. 29-43.

15

a foreign government would issue a report documenting human rights violations, and call

on the country to improve its record. Sometimes states would also cut military or

economic aid or bring other pressures to bear on recalcitrant human rights violators. In

the few cases where stronger enforcement mechanisms existed, especially the regional

human rights courts in Europe and the Americas, these courts could find that a state was

in violation of its obligations under the Convention, and ask it to provide some kind of

remedy, usually by way of changed policy. The actual individuals who carried out

human rights violations were not affected. In some cases, such pressures succeeded in

bringing about important modifications in human rights practices.21

In the 1980s and 1990s, after decades of drafting and ratifying human rights

treaties, it appeared to many that human rights violations were getting worse, not better.

In this context, some activists argued that as long as no individuals were even held

personally responsible for human rights violations, there would be no strong incentives

for changing behavior. They suggested that holding individual state officials criminally

accountable could help to supplement state accountability and add a new ways to enforce

human rights law. State accountability didn’t disappear. But instead of a model of state

accountability and individual immunity, a new model emerged of state accountability and

individual criminal accountability.22 The new criminal accountability model has emerged

over the last 20 years, alongside the state accountability model, and recently it has grown

more dramatically than the state accountability model. States are beginning to hold

individuals, including heads of state, accountable for past human rights violations.

It is this turn to and increasing use of individual criminal accountability that we

call “the justice cascade,” and is the focus of this book. 23 The justice cascade is “nested”

in a larger norms cascade around accountability for past human rights violations. Since

the 1980s, states are not just initiating prosecutions but also increasingly using multiple

alternative transitional justice mechanisms, including truth commissions, reparations,

lustration, museums and other ‘memory sites,’ archives, and oral history projects, to

21 Risse, et. al. 1999.22 Legal accountability is the requirement that “agents abide by formal rules and be prepared to justify their action in those terms in courts or quasi-judicial arenas.” Grant and Keohane, 2005. 23 Ellen Lutz and Kathryn Sikkink, “The Justice Cascade: The Evolution and Impact of Foreign Human Rights Trials in Latin America,” Chicago Journal of International Law Vol. 2, No. 1 (Spring 2001).

16

address past human rights violations.24 The spiking use of these practices all attest to a

broader norm cascade of accountability for past human rights violations, of which the

justice cascade is only one part. Likewise, practices of state accountability for human

rights violations have not diminished, but continue to exist beside the prosecutions for

individual criminal accountability, and these two forms of accountability can reinforce

one another.

This new individual criminal accountability model does not apply to the whole

range of civil and political rights but, rather, only to a small subset of rights sometimes

referred to as the “rights of the person” or “core crimes”, especially the prohibitions on

torture, summary execution, and genocide, as well as on war crimes and crimes against

humanity.25 This new regulatory model involves an important convergence of

international law (human rights, humanitarian, and international criminal law) and

domestic criminal law.26 A blurring distinction between international law and domestic

law is not unique to this issue area but characterizes many areas of global governance.27

In some human rights cases in domestic courts, international human rights law might be

absent from the reasoning of judges. Domestic criminal law prohibiting murder may be

perfectly adequate to prosecute individual government officials accused of carrying out

summary executions in their official capacity. But in the past, the idea that heads of state

and other state officials were immune from individual criminal prosecution kept the

model from being applied either in domestic courts or international courts. In other

cases, domestic amnesty laws and domestic statutes of limitations made prosecutions of

24 Jelin 2003.25 These include rights from only two or three of the 27 substantive articles of the International Covenant on Civil and Political Rights, those protecting the right to life and prohibiting torture. The new model also provides enforcement of the Genocide Convention, the Convention against Torture, and those parts of the Geneva Conventions prohibiting war crimes. 26 Rattner and Abrams (2001) refer to four interrelated bodies of law that underpin the move towards individual accountability for human rights violations: international human rights law, international humanitarian law, international criminal law, and domestic law, p. 9-14.. 27 Legal scholars talk about “global administrative law” or transnational legal processes, to describe what is going on in diverse fields including trade, finance, and the environment as well as human rights. See Nico Krisch and Benedict Kingsbury, “Introduction: Global Governance and Global Administrative Law in the International Legal Order,” European Journal of International Law Vol. 17, No. 1 (February 2006), p. 11. Add reference to Harold Koh, Why do States Obey International Law, and to forthcoming work by Geoff Shaffer – get cite.

17

public officials difficult. In these cases, international and regional human rights law was

often useful to help overcome domestic legal barriers to prosecution. The Torture

Convention, the Law Lords decision in the Pinochet case, and the Statute of the

International Criminal Court, for example, have been particularly important for limiting

absolute practices of immunity for state officials.28

V. The Scope and Dimensions of the Justice Cascade

What makes the justice cascade complicated is that it is occurring simultaneously

at three different levels: (1) domestic; (2) foreign or transnational; and (3) international.29

Our data reveal an unprecedented spike in state and international efforts to address past

human rights abuses by focusing on individual criminal responsibility since the mid-

1980s (see figure 1). 30

28See for example chapter VII on “immunities” in {Broomhall, 2003 #196}29To determine the actual dimensions of the global justice cascade, Carrie Booth Walling and I have created a new data set of domestic, foreign, and international judicial proceedings for individual criminal responsibility for past human rights violations. I am indebted to Carrie Booth Walling for her permission to use material from our joint data set, and for preparing some of the figures based on that data for this book.30We have created two data sets on human rights prosecutions, one for all human rights prosecutions and one for human right prosecutions in transitional countries. The data reported here is from the data set for all human rights prosecutions. For a full discussion of the data set for transitional countries only, see Kathryn Sikkink and Carrie Booth Walling, “The Impact of Human Rights Trials in Latin America,” Journal of Peace Research 44(4), July 2007.

18

Human Rights Trials 1979-2004

0

50

100

150

200

250

300

350

1979-1984 1985-1989 1990-1994 1995-1999 2000-2004

Total

Domestic

Foreign

International

As the figure shows, the bulk of enforcement in this new model is occurring in

domestic courts. The pie chart below shows that 88% of the prosecution years in the data

set occur in the country where the crime was committed, and fully 96% of trial activity

takes place either in domestic or foreign courts. Although the source of the bulk of

enforcement is mainly domestic courts, these courts often apply a combination of

domestic criminal law, international human rights law, and international humanitarian

law31

31 Because of this combination of forms of law, we could think of this as an example of “legal integration” of the kind discussed by Burley and Mattli, in reference the penetration of EC law into the domestic law of member states.Anne-Marie Burley and Walter Mattli, “Europe Before the Court: A Political Theory of Legal Integration,” 47(1) 1993, pp. 41-76.

19

Percentage of Total Human Rights Trials

Domestic88%

Foreign8%

International4%

Most previous discussions of these issues have only looked at parts of this trend,

examining just international trials, just specific international tribunals, just foreign trials,

or just domestic trials in certain countries.32 I argue that these different tribunals and

doctrines are all part of a related global phenomenon, the justice cascade. To understand

how this new model of global regulation of core political rights emerged and how it

functions, we need to look at the big picture of the entire system.

The Statute of the ICC embodies the new regulatory model of individual criminal

accountability, but because of the importance of domestic courts, the ICC is not the main

institution through which regulation of the new model is enforced. The ICC Statute

makes clear that the Court functions under a doctrine of “complementarity.” What this

means is that domestic courts have priority, and that the ICC can only exercise

jurisdiction if domestic courts are “unwilling” or “unable” to prosecute.33 The doctrine of

32 Universal Jurisdiction: National Courts and the Prosecution of Serious Crimes under International Law, edited by Stephen Macedo (Philadelphia: University of Pennsylvania Press, 2004); Naomi Roht-Arriaza, The Pinochet Effect (Philadelphia: University of Pennsylvania Press, 2005).33 Schabas 2001, p. 13, 67.

20

complementarity in the ICC can be seen as a broader expression of the new model in

which the primary institutions for enforcement are domestic criminal courts, and the ICC

and foreign courts are the backup institutions or the last resort when the main model of

domestic enforcement fails.34 Such backup institutions, however, are necessary to create

a fully functioning international model. If the model depended only on domestic courts,

perpetrators could always escape either by black-mail and veto in the domestic polities

(for example, the rattling of sabers and coup attempts that former military leaders in

Argentina and Chile tried each time they faced the possibility of domestic prosecution),

or by retirement abroad in a friendly third country. The back-up provided by foreign and

international prosecutions makes such options less possible than before. Where

enforcement was only done through domestic courts, it was more subject to “capture” by

domestic repressors, while the move to create a more transnational system of regulation

reduced the control that perpetrators in any single country had in preventing prosecution.

Many critics of the ICC or the specialized courts have not understood their role as

the back-up institutions in a global system of regulation. For example, Helena Cobban

argues that international tribunals “have squandered billions of dollars” and that domestic

solutions would be more cost-effective. 35 It would indeed be costly if international

tribunals or the ICC were designed to provide a comprehensive system of individual

criminal justice by themselves, but that is not how the model is currently working. The

use of international tribunals or foreign courts as a back-up is the exception, not the rule,

in the new model of regulation. For the most part, the new decentralized system of

enforcement depends primarily on enforcement through domestic courts. Because the

system is decentralized, however, the quality of the enforcement varies with the quality

of criminal justice systems in different countries. At the international level, there are also

concerns about the fairness of the focus of international tribunals. Such tribunals exist

for some countries, but not for others, and although many countries have ratified the

Rome Statute, to date the ICC has focused primarily on Africa.

34 Orentlicher calls this “domestic enforcement with an allowance for “fallback” international jurisdiction,” and Naomi Roht-Arriaza, refers to foreign trials as a “back-stop” for domestic justice. {Orentlicher, #20}Naomi Roht-Arriaza, The Pinochet Effect, p. 200. 35 Helena Cobban, “Think Again: International Courts” (March/April 2006), p.

21

There is significant variation in the frequency of human rights prosecutions in

different regions of the world. As the pie chart below indicates, the trend towards

domestic human rights prosecutions has been most pronounced in Latin America, which

accounts for 35% of total prosecutions, although it is home to approximately 8% of the

world’s population. This higher number of prosecutions is not due to Latin America’s

more severe forms of human rights violations in comparison to other developing regions

in the world. Indeed, the single genocidal episode in Rwanda in 1994 accounted for more

deaths than the combined number of deaths listed in all the truth commission reports in

Latin America. Not only do Latin American countries account for the majority of

domestic human rights prosecutions, but they are also the subject of the largest number of

foreign human rights. Most of the 101 foreign prosecutions in our data base were held in

the domestic courts of European countries for human rights violations committed largely

in the Americas. Increasingly, foreign human rights prosecutions are being held in one

Latin American country for human rights violations committed in another Latin

American country. The great bulk of these foreign prosecutions were brought to foreign

courts by human rights organizations or private lawyers using private prosecutor

provisions and acting on behalf of human rights victims or their relatives.

But, even in Latin America, there is significant variation among different

countries in the degree to which they have adopted the new regulatory model, and even

variation within a single country as to when it initiates criminal human rights

prosecutions. Some Latin American countries such as Argentina and Bolivia were

among the very first countries to start using human rights trials in the mid 1980s.

Argentina was both the leader in the region and also a global leader in the number of

human rights trials it has held. Argentina’s neighbors Brazil and Uruguay, which

experienced similar authoritarian regimes and transitions to democracy at roughly the

same time as Argentina, made different choices about prosecutions. Brazil has held no

human rights trials for violations during the authoritarian government, and Uruguay held

no prosecutions for the first 15 years after the transition, only to begin a handful of

prosecutions in the early 2000s.

22

Regional Distribution of Human Rights Trials

36%

20%

11%

10%

7%

6%

6%4%

Americas

Africa

W. Europe

E. Asia & Pacific

C. & E. Europe

S. Asia

M. East & N. AfricaEurasia

VI. Where Do Norms Come From?

The justice cascade was deeply motivated by shared principled ideas. Three key

ideas underpin the justice cascade: the first is the idea that the most basic violations of

human rights – summary execution, torture, and disappearance -- cannot be legitimate

acts of state and thus must be seen as crimes committed by individuals. A second and

related idea is that the individuals who commit these crimes can be, and should be,

prosecuted. These seem like simple ideas, even obvious ideas. But they run counter to

centuries of state practice. It took a major movement to put them forward, embed them in

law, and put them into practice. The third idea is that the accused, too, are bearers of

rights, and deserve to have those rights protected in a fair trial. This is now common

practice in legal systems around the globe, and yet it was often controversial when

applied to repressive former leaders or other violators of human rights.

It is this third set of practices about protecting the rights of the accused that

distinguishes a human rights trial most clearly from political trials or kangaroo trials. For

example, after the fall of the communist regime led by Nicolae Ceauşescu in Romania,

the leader and his wife and political partner Elena were put on trial for their crimes. The

trial was about human rights violations, and the individuals were being held criminally

23

accountable, but we would not consider it a human rights prosecution because the

accused were not given even a minimally fair trial. On the morning of December 25th,

1989, the Ceauşescus appeared before an ad-hoc military tribunal. Before the trial began,

the panel of judges knew that it was a foregone conclusion that the Ceauşescus would be

executed that afternoon. After a two-hour trial in which little hard evidence was

presented and the defendants were not allowed a strong defense—as even their defense

lawyer accused them of crimes—they were given a death sentence and executed

“commando style” in the same room where the “trial” occurred.36 This is not a human

rights prosecution but a political trial of the kind that has long been common during and

after violent regime changes.

The justice cascade was not spontaneous, nor was it the result of the natural

evolution of law or global culture in the countries where the prosecutions occurred.

These changes in ideas were fueled by the human rights movement. It initially started

due to the concerted efforts of small groups of activist lawyers, jurists, and other political

activists who pioneered strategies, developed legal arguments, often recruited plaintiffs

and/or witnesses, marshaled evidence, and persevered through years of legal challenges.

In the early cases of norm emergence considered here, Greece, Portugal, Argentina, and

Bolivia, these networks were primarily domestic groups with transnational linkages. In

the later cases discussed in this book, including Chile, Brazil, Uruguay, and South Africa,

transnational networks played a much more important role.

When we try to understand why governments chose to use the criminal

accountability model, there is a striking fact that comes to our attention. Once it became

possible to imagine accountability for individual leaders, the model of regulation chosen

(individual criminal accountability) was often an extension of the criminal model already

used for common crime in domestic courts. For hundreds of years, most societies have

regulated crimes like murder or kidnapping with domestic prosecutions for individual

criminal accountability. One reason why the criminal accountability model was adopted

internationally has to do with ingrained attitudes about the necessity for punishment in

the domestic criminal system.

36 {Stan, 2008 #200}

24

VII. What Difference Does it Make? The Effects of the Justice Cascade

There has been a lively debate in political science and international law about the

desirability and impact of human rights prosecutions. This debate has important policy

implications since governments, international organizations, and nongovernmental

organizations (NGOs) are engaged in ongoing decision making about whether they

should carry out or advocate human rights prosecutions, and, if so, what type and level of

trials are suitable. Many scholars and practitioners believe that such prosecutions are

both legally and ethically desirable and practically useful in preventing future human

rights violations.37 Both the literature on compliance with institutions in international

relations and the deterrence literature in sociology suggest that stronger enforcement

should lead individuals to reduce human rights violations.38 According to this logic,

movement towards more enforcement should lead to greater compliance with human

rights norms and deterrence of future human rights violations, the topic of the second half

of this book. In his review of the transitional justice literature, Mendeloff finds many

such claims about the positive effects of human rights trials but relatively little solid

evidence to support those claims.39

Other scholars are far more skeptical about the impact of trials. In the mid 1980s,

scholars of transitions to democracy such as Samuel Huntington and Guillermo

O’Donnell and Phillippe Schmitter generally concluded that trials for past human rights

violations were politically untenable and likely to undermine new democracies. 40 Even

many actors directly involved in transitions were equally pessimistic. Jose Zalaquett, a

Chilean human rights lawyer who later served on the Chilean truth commission, argued

that it was possible and desirable to seek truth, but that human rights prosecutions were

more difficult when dealing “with perpetrators who still wield considerable power.” He

37 {Roht-Arriaza, 1995 #12; Méndez, 1997 #13}. 38 See, for example, Downs, George W., David M. Rocke, and Peter N. Barsoom. 1996. “Is the Good News About Compliance Good News About Cooperation,” International Organization 50, 3: 379-406; and Daniel S. Nagin, “Criminal Deterrence Research at the Outset of the Twenty-First Century,” in Crime and Justice: A Review of Research ed. M. Tonry, 23. Chicago: University of Chicago Press: p. 1-42.39 { 2004 #142: 358, 36140 Huntington, 1991; O’Donnell and Schmitter, 1986.

25

was clearly thinking of General Pinochet who was at the time still commander in chief of

the Chilean armed forces. 41

Much of this literature dates from the late 1980s or early to mid 1990s, and since

much has happened empirically since then, it is useful to revisit these claims in light of

new developments. However, scholars of international relations and international law

are currently making very similar arguments to those made in the past. Some realist

scholars have argued that human rights trials will not deter future violations and that, in

some circumstances, they will actually lead to an increase in repression or to

humanitarian atrocities.42 For example, they contend that the threat of prosecution could

cause powerful dictators or insurgents to entrench themselves in power rather than

negotiate a transition from authoritarian regimes and/or civil war. For similar reasons,

Cobban concluded, “It’s time to abandon the false hope of international justice.”43

One of the most striking characteristics of the transitional justice scene and

literature is how heated and contested the debate has been over alternative practices and

policies. While we can understand why perpetrators and victims of human rights

violations might feel strongly about transitional justice, why have scholars and

practitioners also been so passionate and even vituperative in their positions?

One reason may be exactly because debates over trials invoke our ideals and

expectations for justice. Justice, like democracy, is one of those powerful concepts that

means many things to many people, but by almost any definition, in practice always falls

short of our ideals. These ideals, rarely achieved in the domestic arena, are all the more

problematic when transferred to the international. The justice cascade has resonated and

spread around the globe exactly because so many actors hold powerful ideas about

justice. But those powerfully held ideas then set up the very institutions of the justice

cascade to disappoint and disillusion. The justice cascade is caught in the paradox of its

own “success”. The very power of the ideals that gave impetus to the cascade also sets it

up for failure, when measured against those ideals. Victims, among the most ardent of

the advocates of transitional justice, are ultimately disillusioned with institutions that can

neither heal them nor return their loved ones to them. Perpetrators find them biased at all

41 1992: 1428-942 {Ku, 2007 #116}, Also add here Snyder and Vinjamuri and Krasner and Goldsmith43 2006: 22

26

times, even when courts lean over backwards to provide due process. State sponsors find

justice slow, expensive, and uncertain. And scholars point to the disconnect between the

glowing words used to justify international justice and the always disappointing reality.

And because people believe so firmly in justice, they are not only disillusioned, but often

feel betrayed. They reserve their venom not solely for the perpetrators who made such

institutions necessary in the first place, but for the institutions themselves that betray their

ideals.

Thus, to put it mildly, there is still little agreement about the impact of human

rights trials. In such a situation, how can we talk about the effects of the justice cascade?

Although the institutions of justice must always be judged against our ideals of justice, as

a social scientist, I advocate the use of empirical comparison and reasoned

counterfactuals to weigh such consequences. I try to move beyond some of these

passionate debates by using both quantitative and qualitative data to evaluate

systematically various claims about the impact of human rights trials. Enough time has

now passed since the initiation of the justice cascade that we can start to evaluate

empirically many of the negative and positive claims for human rights prosecutions.

Chapter 2 discusses the theoretical issues more deeply, and addresses both the

ethics and the methods of such an inquiry. In short, I argue that the justice cascade must

be judged by both the principles that motivate it and by the consequences of its actions.

Thus, this book will explore the impact of human rights prosecutions by comparing

countries which have experienced such trials to countries which have not experienced

trials to draw conclusions about what difference prosecutions make. I will focus on

whether prosecutions lead to improvements in human rights, democracy, and levels of

conflicts compared to countries that have not used prosecutions. I situate the debates

over the effects of trials within larger theoretical debates on compliance and deterrence in

political science, law, economics, and sociology.

The results of the quantitative analysis, presented in Chapter 5, show that

transitional countries in which human rights prosecutions have taken place are less

repressive than countries without prosecutions, holding other factors constant. Contrary

to the arguments made by the trial skeptics, transitional human rights prosecutions have

not tended to exacerbate human rights violations. The quantitative study shows that

27

countries with more accumulated years of prosecutions after transition are less repressive

than countries with fewer accumulated years of prosecutions. In addition, countries

surrounded by more neighbors with transitional prosecutions are less repressive, which

may suggest a deterrence impact of prosecutions across borders. Our findings refute the

arguments made by the trial skeptics that human rights trials lead to more repression and

demonstrate that prosecutions and truth commissions have a positive effect on human

rights practices.

The quantitative evidence, however, is mainly suggestive of the mechanisms

through which trials operate, so I turn to qualitative studies in Chapter 7 to examine in

more detail the ways in which human rights prosecutions can have an impact on human

rights practices, democracy, and conflict in specific countries. I explore the qualitative

evidence on the impact of prosecutions, particularly in relation to Latin America, but also

in Southern European countries that were the first countries to use domestic trials. I

focus on Latin America because cases in this region account for more of the country-trial-

years in the data set than any other region.44 Further, because many Latin American

countries were early innovators of human rights trials as well as truth commissions, the

most time has passed in order to be able to evaluate the impact of these transitional

justice mechanisms on future human rights practices, democratic consolidation, and

conflict, than in any other region. In summary, the research shows that holding human

rights prosecutions has not undermined democracy nor led to an increase in human rights

violations or conflict in Latin America. Twenty-five years after the first trials in the

region, there is not a single case of a country in Latin America where democracy has

been undermined because of the choice to use trials. Nor is there evidence that

prosecutions lead to worsening human rights situations.

While it is possible that Latin America is an exceptional region, equally possible

is that the trial skeptics have based their arguments on a few powerful but as yet

unresolved cases. Just as the frightening but ultimately unsuccessful coup attempts in

Argentina drove some of the early pessimism in early reflections on political transition,

44 We define country-trial years as the number of years during which a state is actively engaged in judicial proceedings for individual criminal responsibility for human rights abuse. This number does not reflect the number of trials underway within that state during those years which may be far greater.

28

the failure of international justice to dampen nationalism in Serbia or to help end conflicts

in Uganda or Sudan may fuel current trial skepticism. And, just as the transitions

literature was too hasty in its judgments about the impossibility and undesirability of

trials in Latin America, current trial skeptics might be well advised to monitor the

situations in the former Yugoslavia, Uganda, and Sudan longer before jumping to

conclusions about the pernicious effects of trials.

The research findings that human rights prosecutions and truth commissions can

contribute to lessening repression also have important policy implications. Human rights

research to date has not been very helpful about giving ideas for how to lessen repression.

In the past, major studies concluded that human rights violations were caused by big

structures over which governments had little short term control, including poverty,

authoritarianism, civil war, and population size.45 We continue to see the impact of these

big factors, but our study also suggests that more micro-level policy responses such as

prosecutions and truth commissions also can have an impact on human rights practices,

thus providing policy makers with more viable short term options.

Eventually, trials are one factor that can contribute to cultural change and the

internalization of human rights norms among state officials, but not all state officials

internalize norms to the same extent, as witnessed by the revival of debates about the

legitimacy of torture in the United States after 9/11. The final empirical chapter of the

book looks at the United States’ non-compliance with the prohibition on torture to probe

the most difficult question in the norm literature. If a law or norm truly has some teeth,

we would expect that eventually it would be able to have an impact even on the most

powerful states, in this case, the United States. The U.S. decision to use torture even

after the Pinochet case had established the legitimacy of universal jurisdiction for torture

is an indication that some state officials still believed only a short time ago that they were

above the law. U.S. officials simply did not see that the Pinochet precedent applied to

them, even though the United States committed torture after ratifying the Torture

Convention and implementing it in domestic law. Yet the chapter on the United States

shows that there is some indication that even U.S. state officials are not completely

45 {Poe, 1994 #131; Poe, 1999 #50}

29

immune from the possibilities of future prosecution. This is clearly the hard case, and too

early to fully predict.

In conclusion, the trend towards individual criminal accountability for past human rights

violations is well advanced, and has been associated with improvements in human rights

practices. There is not evidence that the use of prosecutions undermines democracy, exacerbates

conflict, or undermines the rule of law. Nor do we find, however, that countries must hold

prosecutions immediately in order for justice to be done. Indeed the most striking conclusion of

the book is the extreme persistence of the demand for justice. Human rights groups will and

should continue to press for human rights prosecutions. Politicians will be expected to weigh the

circumstances and conditions and select the best sequencing of transitional justice to fit the needs

of their society. Human rights prosecutions will not go away, but such prosecutions are not a

panacea for all the ills of society and will inevitably disappoint as they fall short of our ideals of

justice. They represent an advance, however, over the complete lack of accountability of the

past, and they have the potential to prevent future human rights violations through material

deterrence and symbolic communication.

For readers who wish to move directly to the story of the origins of the Justice Cascade in

trials in Greece and Portugal in the 1970s, skip Chapter 2 and go directly to Chapter 3. For

readers who would like to delve more deeply into the theoretical, ethical, and methodological

questions addressed in this text, continue to Chapter 2. If you are most interested in the question

of the effectiveness of human rights prosecutions, and want to know more about whether and

how human rights prosecutions help protect human rights, skip the first half of the book and read

chapters six and seven.

30

Chapter 2: Theory, Ethics, and Method in Studying the Justice Cascade

Kathryn Sikkink

In this chapter, I return to the three main questions of the book: (1) What are the

origins or sources of new ideas and practices about individual criminal accountability for

human rights? (2) How and why do these ideas diffuse regionally and internationally?

(3) What are the effects of human rights prosecutions? I try to lay out more clearly the

theoretical arguments I make to answer these questions and trace their relationship to

broader international relations and political science theory. I also discuss my ethical

approach and explain how the methods I use are appropriate to both the theoretical

questions and the ethical approach.

I. Explanations for Emergence and Diffusion of the Justice Cascade

The justice cascade is but one of an increasing number of norms that are

cascading or diffusing around the globe. Although scholars have now done a good job

documenting that such cascades are occurring, they have not done such a good job

explaining why or how they get started. If diffusion is one of the major processes through

which the international system changes, it is important to understand not only why and

how ideas and policies spread, but also where the ideas and policies that later diffuse

come from, and why they are first adopted. I argue that norm emergence and norm

diffusion are two different processes that require different kinds of explanation. More

scholarly attention has been devoted recently to diffusion than to emergence. For this

reason, I spend two full chapters of the book (Chapters 3 and 4) talking about the early

emergence of the justice ideas and practices, and only one chapter (Chapter 5) discussing

the process of diffusion.

For the purposes of this study, I consider the norm emergence period to be the

period from 1975, when the first trials occurred in Greece, to 1990. The norm diffusion or

cascade period, on the other hand, is the period from 1990 to the present. Prior to 1990,

trials were used infrequently in transitional countries, and rarely more than one

prosecution was in process in the world in a single year. In 1990, for the first time,

human rights prosecutions were underway in six different countries within three separate

31

regions (Latin America, Africa, and Eastern Europe). After 1990, there was never a year

where prosecutions were in process in fewer than six transitional countries around the

world. Thus, the 1990 cut-off captures the difference between a norm emergence phase,

where practices are beginning but are still somewhat isolated, and a norm diffusion

period, where the practice becomes more commonplace.

In the case of the justice cascade, many of the standard explanations of why

diffusion occurs in international politics – like power, interest, or global culture – don’t

tell us very much about why states are deciding to prosecute human rights violations.

Since the norms and practices of the justice cascade withdraw important legal protections

from powerful state officials, it challenges the notion that power and interest drive

diffusion of these ideas. First, it is not clear whose interests are being served by these

prosecutions, but certainly not those of the perpetrators who continue to hold power in

many parts of the world. Except in a handful of cases, powerful states are not forcing

other states to adopt prosecutions. The countries of Eastern Europe and the former Soviet

Union that wish to enter the European Union need to demonstrate that they are doing

something to address past human rights, but that has not necessarily included

prosecutions. As one scholar of Eastern Europe explains it, "if you want to be a

respectable new democracy, especially in Europe, you do something about your bad past.

You have a truth commission and you do something about the files. It's part of your task

of wanting to belong to Europe."46

Second, in the case of the most important of the international institutions

designed to carry out prosecutions, the International Criminal Court, the most powerful

state in the world, the United States, has done everything possible to undermine it. Third

and finally, if, as the sociological institutionalists would have it, this is simply another

instance of states enacting liberal and individualist norms of global culture, why did it

take them so long to get around to it, and why did they struggle so hard against it?

46 “Bulgaria's painful past still hidden, but for how long?” Matthew Brunwasser, International Herald Tribune, June 2, 2006, http://www.aip-bg.org/library/press/iht_020606_eng.htm

32

Several factors were preconditions that facilitated the justice cascade, including

the third wave of democracy, the end of the Cold War, and the gradual accretion of

human rights treaties, along with the corresponding organizations that provided the legal

and institutional underpinnings for the practice of individual criminal accountability. The

third wave of democracy is a necessary factor to explain both the emergence and

diffusion of the norm of individual criminal accountability.47 The first two cases of

prosecutions, Greece and Portugal, are the first cases of transition to democracy in the

third wave, and the justice cascade follows the third wave of democracy as it moves from

southern Europe to Latin America, and from there to Eastern Europe. But, democracy is

not in any sense a sufficient condition, since the second wave of democratization was not

accompanied by human rights trials, and even in the third wave, fewer than half of the

countries that experienced democratic transitions used human rights prosecutions. The

main question is, then, why do some transitional democracies in the third wave use

human rights trials, while others do not? To answer this question, I compare countries

that first used trials—Greece, Portugal, and Argentina—to other countries that

experienced transition to democracy around the same time but did not prosecute state

officials for past human rights violations.

From 1948 until the present, states have drafted and ratified numerous human

rights treaties. By 1990, when the diffusion period begins, the human rights issue area

was highly legalized and highly institutionalized. While few of these treaties explicitly

said that states had a duty to punish perpetrators of human rights, many recognized that

victims of human rights violations had a right to some kind of legal remedy. These

institutions provided the legal and institutional underpinnings for the justice cascade.

Some treaties, like the European Convention on Human Rights and the American

Convention on Human Rights, created institutions like the European Commission of

Human Rights and the Inter-American Commission and Court of Human Rights, which

would later play an active role pressuring for greater accountability. These treaties and

institutions provided possibilities for action, but in the mid 1970s, when the first trials

47 Huntington, S. (1991). The Third Wave: Democratization in the Late Twentieth Century. Norman, Oklahoma, University of Oklahoma Press.

33

were held, these possibilities had not yet been activated. The activists who helped launch

the human rights trials were often simultaneously responsible for helping to engage and

activate the existing but still dormant human rights institutions.

Some scholars consider these international human rights institutions to be the

handmaidens of powerful and wealthy states. But just as small states were deeply

involved in the inclusion of human rights in the UN Charter, and in the drafting of the

Universal Declaration of Human Rights48, so too were they fundamental in the history of

the justice cascade. It was the institutions that are less dominated by powerful states that

played a more important role, for example the Inter-American Commission for Human

Rights (IACHR) and its sister institution, the Inter-American Court of Human Rights.

Although the IACHR is located in Washington D.C., the United States has never ratified

the American Convention for Human Rights, and is not a member of the Inter-American

Court of Human Rights, based in Costa Rica. Even in the case of an institution full of

powerful countries, like the Council of Europe, it was the smaller “like-minded” states of

the Netherlands and Scandinavia that pressed the European Commission to make its first

major investigation of human rights violations in Greece.

The norm of individual criminal accountability begins to emerge over a decade

before the end of the Cold War, so the resulting change from a bi-polar to a uni-polar

world can’t explain norm emergence. But the diffusion or cascade period begins just

after the fall of the Soviet Union, since the breakup created a new group of transitional

states in Eastern Europe that would adopt transitional justice measures including

prosecutions. The end of the Cold War also created a more permissive atmosphere for

holding former repressive leaders, of whatever ideological stripe, accountable for past

human rights violations. The related trends of democratization, the institutionalization of

human rights, and the end of the Cold War are thus important for understanding the

origins of the justice cascade, but they only provide the background conditions, not a

fully satisfactory explanation. My task is to explain why, in this context of

48 Waltz, S. (2001). "Universalizing Human Rights: The Role of Small States in the Construction of the Universal Declaration of Human Rights." Human Rights Quarterly 23: 44-72.

34

democratization and institutionalization of human rights, some countries adopt

prosecutions and others do not.

The first part of this book demonstrates that the justice cascade follows the

general pattern of a “norm life cycle”.49 The main norms at stake here come directly

from domestic legal systems. The individual criminal accountability model is the

dominant model in domestic legal systems, and it has now penetrated the international

legal arena. So the model of justice is not just diffusing from one country to another, but

rather ideas from the domestic criminal system dealing mainly with common crime are

being used to address the more extraordinary crimes committed by state officials. Norms

scholars have long recognized the importance of powerful domestic norms. In some

cases, these domestic norms may have a prominence that makes them likely candidates

for international norms.50 The new norms studied here originated in domestic settings in

countries in the semi-periphery, but they were mobilized by norm entrepreneurs,

including regional human rights organizations, INGOs, and like-minded governments.

The case studies and quantitative evidence presented in this book suggest that

active domestic norm entrepreneurship is essential to norm emergence, while a

combination of transnational emulation/learning and transnational norm entrepreneurship

contributes to diffusion.51 The first three cases of prosecutions in Greece, Portugal, and

Argentina are all driven primarily by domestic pressures. They were largely unaware of

one another, and they were not drawing on international models. In Portugal, there was

not even a role for a regional human rights organization, but in both Greece and

Argentina, the regional human rights commissions wrote crucial reports on the human

rights violations during the authoritarian regimes that provided evidence and

encouragement for such prosecutions, while not at all determining the outcomes.

49 Finnemore and Sikkink, 1998.50 Peter Katzenstein, (1996) Cultural Norms and National Security: Police and Military in Postwar Japan (Ithaca: Cornell University Press). Ann Florini (1996), D. Lumsdaine, Moral Vision: The Foreign Aid Regime (Princeton University Press, 1993); Finnemore and Sikkink (1998), p. 906.51 This argument is related to that of some studies of diffusion among the American states, which find that both some internal determinants and external diffusion matter for policy adoption.Daley, D. a. G., James (2005). "Horizontal Diffusion, Vertical Diffusion, and Internal Pressure in State Environmental P:olicymaking." American Politics Research 33(5): 615-644.

35

Eventually, in the later cases considered in the group, transnational actors became

deeply involved in the justice cascade. These transnational actors were sometimes

human rights groups that resembled an advocacy network, in that they were inter-

connected groups of individuals bound together by shared values and discourse who

engage in dense exchanges of information and services.52 Individuals in these groups are

often “brokers” linking two or more previously unconnected networks or sites.53 The

transnational justice network was not just your garden-variety social movement, however,

because its central members were lawyers with appreciable technical expertise in

international and domestic law who systematically pursued the tactic of human rights

prosecutions. In this sense, the transnational justice network resembles what political

scientists call an epistemic community, a network of professionals engaged in a common

policy enterprise with "recognized expertise and competence in the particular domain and

an authoritative claim to policy relevant knowledge in that issue or domain."54 But unlike

what the epistemic community literature leads us to expect, states did not turn to the

transnational justice network in situations of complexity and uncertainty for information

to help them understand the situation and their interests. Very often, human rights

victims set up and used the justice advocacy networks, and these networks pursued

justice in the face of governmental indifference or recalcitrance. At other times, they

formed alliances with networks of government officials committed to rule of law and to

human rights. These networks of government officials, what Anne-Marie Slaughter has

called “transgovernmental networks”, often included judges and lawyers in the judicial

branches of governments.55 National judicial authorities cooperated both with human

rights NGOs and with international and regional courts. These linkages were especially

strong in Europe, where the European Court of Justice and the European Court of Human

52 Margaret Keck and Kathryn Sikkink, Activists Beyond Borders: Advocacy Networks in International Politics (Ithaca: Cornell University Press, 1998), p. 2.

53 (McAdam, Tarrow, Tilly, 2001, Tarrow, 2005). Graham, E., Shipan, Charles, and Volden, Craig (December 2008). The Diffusion of Policy Diffusion Research.

54 Peter Haas, "Introduction: Epistemic Communities and International Policy Coordination."

55 Slaughter, A.-M. (2005). "A New World Order."

36

Rights helped create and consolidate networks of judicial officials committed to rule of

law, but they also emerged in Latin America. The Inter-America Commission for Human

Rights, for example, played on unexpectedly important role in the emergence of the

accountability norm. The transnational justice networks thus blended characteristics of

advocacy networks, epistemic communities, and transgovernmental networks.

There was not a single transnational justice network, but various regional

networks that occasionally overlapped and connected into global networks. Some

regions, especially Asia and the Middle East, were underrepresented. The justice

networks were voluntary networks including domestic advocates and officials, so

countries with few domestic human rights NGOs or few international linkages were often

absent. The Latin American justice network shared for the most part a common language

(Spanish) and met occasionally in informal workshops around the hemisphere as well as

during the annual meetings of the General Assembly of the OAS, especially when it was

going to consider important reports of the of the Inter-American Commission of Human

Rights. The Latin American network had strong contacts with U.S. and European-based

NGOs that worked on human rights and transitional justice, but it had very weak

connections, for example, with NGOs based in Eastern Europe, Asia, or the Middle

East.56 Likewise, there is something we could call a European justice network, focused

on transitional justice questions primarily in Europe.

The transnational justice network did not operate in a vacuum. It was possible

because it was part of a broader human rights advocacy network working in the context

of a broad shift in international norms towards greater protection for human rights. At

times, key members of the transnational justice network were simultaneously leaders in

the broader human rights network. The human rights issue area is one long characterized

56 One way to create a map of the Latin American justice network would be by combining the links from web pages of a key justice NGOs working on Latin America, like CEJIL, the Center for Justice and International Law. http://www.cejil.org/enlaces.cfm

37

by increasingly dense relationships among states, intergovernmental organizations, and

nonstate actors.57

The role of such networks in instigating the justice cascade and contributing to its

diffusion comes through quite clearly in all of the qualitative research on the emergence

of human rights norms and institutions, including that presented in Chapters 3 and 4 of

this book. It is very difficult to test quantitatively because we have no measures, or poor

measures, of transnational and transgovernmental networks. Scholars have used data on

domestic human rights organizations or INGOs as proxies in their quantitative analysis

for the presence of such networks, but they are weak proxies not available for all years.

Nevertheless, Hunjoon Kim has shown that the presence of domestic human rights

groups has a positive and highly significant effect on the repeated use of human rights

prosecutions.58

Still, the presence and work of a transnational network is not enough to explain

the rapid diffusion of the justice norm. For more general diffusion of the norm to occur,

the existing model of handling human rights abuses had to be perceived as inadequate or

broken in some way. Before WWII, there had been no accountability for past human

rights violations. The Holocaust revealed the dramatic failures of the sovereign

immunity model and led states to construct the human rights regime, with its state

accountability model. But while naming and shaming and the state accountability model

had some successes, it was often inadequate in dealing with the most repressive states.

By the early 1990s, although the human rights field was thick with norms and laws, there

was still a great weakness in the area of enforcement. The human rights regime had been

launched with high expectations in 1948, but a half-century later, human rights violations

had not subsided, and if anything, the perception was that human rights violations were

on the increase. This was especially heightened by the demonstration effect of the

conflict in the Balkans, since the discovery of concentration camps and genocide in the

57 See Risse, et al., 1999; Ellen L. Lutz and Kathryn Sikkink, “International Human Rights Law and Practice in Latin America, “International Organization 54, 3, Summer 2000, pp. 633-659; Tarrow, 2005.

58 Kim, H. (2007). Why and When Do States Use Human Rights Trials and Truth Commissions after Transition? An Event History Analysis of 100 Countries Covering 1974-2004."

38

heart of Europe fifty years after WWII suggested that the existing regulatory model had

failed. The ineffectiveness of the international response to the genocide in Rwanda in

1994 proved yet another demonstration effect of the failure of regulation to prevent major

human rights violations. These crises provoked a move towards an individual criminal

accountability model at the international level, when the Security Council set up the

ICTY and the ICTR. The cascade or rapid diffusion period of the accountability norm

began in the 1990s in response to these perceived failures of past regulation. Policy

makers are particularly open to new ideas after great crises such as wars or depressions,

and other policy failures. This argument, that new policies are adopted in the context of

perceived crisis or failure of existing models, has long been a staple of the literature on

ideas.59 Diffusion models have focused more on diffusion as a result of success or

effectiveness.60 Thus, Simmons et al. (2006) for example, argue that a key argument for

diffusion of liberalism is that it provides competitive edges to the countries that adopt it.

In the case of the justice cascade, it would appear that the perception of past failure drove

policy innovation rather than the perception of success or the hope of gaining any

competitive edge. The very slogan of the transitional justice movement, “Nunca Más”

(Never again), speaks to this perception of failure that must not be repeated.

But of course, neither failure nor success is fully obvious, and the perception of

the failure of the old model was in part a reaction to publicity efforts by human rights

activists, who drew attention to the failure in Rwanda, the former Yugoslavia, and

elsewhere. The work of norms entrepreneurs was also essential for first imagining that it

would be possible also to hold state officials accountable for past human rights violations,

then struggling to create specific instances of accountability in domestic and foreign

courts, and eventually in the newly created international tribunals. The early adopters of

prosecutions are in the most precarious situations, requiring not only strong norm

entrepreneurs but also a fortuitous combination of domestic and international conditions

to permit experimentation with human rights prosecutions. In particular, countries that

had experienced a “ruptured” transition to democracy that weakened the power of the 59 See for example, Odell 1982, Kowert and Legro 1996, McNamara, 1998.60 Graham, E., Shipan, Charles, and Volden, Craig (December 2008). The Diffusion of Policy Diffusion Research.

39

military and other groups associated with repressive regime were initially more able to

use human rights prosecutions. Because proposing prosecutions of state officials were

seen as dangerous, in early cases of adoption of human rights prosecutions, those

powerful state officials also had to be weakened in the process of transition to

democracy. The four first cases of use of human rights prosecutions all followed what

political scientists have called “ruptured” transitions to democracy, where power holders

are delegitimized by economic or military failures, and are unable to control the

conditions of their exit from power. So, for example, in Argentina, the military was

blamed for defeat in the Malvinas/Falklands war, and this led to a rapid transition where

they couldn’t set the terms. Likewise, the Greek military was seen as responsible for a

military fiasco in Cyprus, leading to Turkish occupation of half of the island. To the

contrary, state officials in countries such as Spain, Brazil, Chile, South Africa, and

Uruguay performed “negotiated transitions”, where they often put in place guarantees to

protect them from prosecution. As the momentum for prosecutions increased, however,

it became possible over time to hold them even in some of the countries that had

negotiated transitions, as Pinochet and Bordaberry discovered in Chile and Uruguay.

Even so, the balance of power between the former elites associated with the authoritarian

regime and democratic elites who lead the transition continues to be an important part of

the explanation for where prosecutions are held.

One key question, then, is: how, where, and why do new ideas and practices

emerge that will later diffuse and take root? The problem with diffusion is that it makes

something so commonplace that it can cloud memories about the ideational terrain prior

to a norms cascade. We forget that not too long ago, proposing prosecutions of powerful

state officials was perceived not just as novel and dangerous, but actually as impossible

or unimaginable. Just as states in the United States were “laboratories of democracy,”

early adopters of transitional justice polices were “laboratories of justice,” which

innovated and experimented with new and different transitional justice practices.

Eventually, many of these practices became models that could be adopted in other

countries where somewhat less fortuitous circumstances existed.

In the second stage of norm adoption, diffusion processes started to set in, and the

most important predictor for adoption of prosecutions was less the internal human rights

40

pressures and more what neighboring countries were doing. Hunjoon Kim found, for

example, that the single most important determinant for whether a country will use a truth

commission or human rights trials is the number of other states in that region that have

previously used a truth commission or prosecutions.61 Early norm adoption is the result

of political struggle and norm entrepreneurs, but later adoption is the result of a

combination of internal demands and external diffusion of models.62 This is what the

norm life cycle argument originally suggested. The research on human rights

prosecutions and truth commissions, however, adds a strong regional dimension to the

norm life cycle theory. Norms emerge and diffuse first within regions and only later are

able to jump from one region to another and go global.

The ideas and practices of transitional justice moved in multiple ways and

directions. Practices of individual criminal accountability used in the domestic legal

system to address common crime were applied within these same domestic legal systems

to state officials accused of human rights violations. These practices then spreak from

one country to another in a process scholars call “horizontal diffusion.” Vertical

diffusion also occurred and took two forms: bottom-up from the national polities to

international organizations, for example, and top-down, from the international back to

different countries.63 Bottom-up vertical diffusion occurs when practices of individual

criminal accountability long used in domestic legal systems are adopted by international

tribunals like Nuremberg, and much later by the International Criminal Tribunal for the

former Yugoslavia (ICTY), the International Criminal Tribunal for Rwanda (ICTR), and

61 Kim, H. (2007). Why and When Do States Use Human Rights Trials and Truth Commissions after Transition? An Event History Analysis of 100 Countries Covering 1974-2004."

62This is similar to processes found in diffusion in the America states, where both determinants internal to the states and external diffusion are part of the explanation for policy change. See, for example, Daley, D. a. G., James (2005). "Horizontal Diffusion, Vertical Diffusion, and Internal Pressure in State Environmental P:olicymaking." American Politics Research 33(5): 615-644.

63 Ibid.

, Graham, E., Shipan, Charles, and Volden, Craig (December 2008). The Diffusion of Policy Diffusion Research.

41

the International Criminal Court (ICC). Top-down vertical diffusion occurs, for example,

when the Inter-American Commission on Human Rights recommends individual criminal

accountability in specific countries, when international tribunals encourage states that

have not yet used individual criminal accountability to do so in their domestic legal

systems, or when they provide specific new legal doctrines or formula that are then

applied in domestic legal fora.64 So, for example, the doctrine from international and

regional human rights law that crimes against humanity are so grave that they cannot be

subject to statutes of limitations was later used by domestic courts in Latin America to

clear away the main legal barriers to domestic prosecution of crimes that often occurred

decades earlier. Domestic courts needed this doctrine from international law to overcome

the statute of limitations. Once they do so, however, domestic criminal law may be

perfectly capable of providing the legal tools to prosecute human rights violations by

state officials.

Vertical and horizontal diffusion do not proceed arbitrarily. Horizontal diffusion

occurs more forcefully within regions than between regions, and bottom up vertical

diffusion happens from innovative regions to international institutions. What region is

innovative varies somewhat by issue-area; in the area of transitional justice, Latin

America has been one of the most innovative regions in the world. Top-down vertical

diffusion happens in turn from international institutions to countries in less innovative

regions or with more intractable national political situations.

Scholars of international politics, sociology, and international law have paid most

attention to top-down vertical diffusion. Indeed, the entire school of sociological

institutionalism is premised on the notion of top-down vertical diffusion from global

culture to domestic societies. The disciplinary and geographical “blinders” of those who

focus on the international level sometimes leads them to assume that all diffusion is top-

down, from global culture or international institutions to domestic politics, or that all

diffusion moves from the North to the South. In the case of justice, scholars have mainly

explored the move from international tribunals to domestic legal systems. For the most

part they have missed the other forms of diffusion discussed here. They tell a story that 64 I’m indebted to Sidney Tarrow and Andrew Karch for drawing my attention to literatures on horizontal and vertical diffusion. Tarrow refers to vertical diffusion as “scale shift” when domestic norms and practices are transferred to the international level.

42

moves from Nuremberg to the ICTY to the ICC, and then into domestic legal systems

through the legal changes made to implement the Rome Statute. But this story leaves out

the domestic and regional developments in-between. Such a story explains the fifty-year

gap between Nuremberg and the ICTY as a casualty of the Cold War. But what this story

misses is that the post-Nuremberg trend toward individual criminal accountability for

human rights violations started twenty years before the ICTY, and it started in the

domestic courts of the individual countries in the semi-periphery where human rights

violations had occurred. The explanation for the justice cascade thus needs to be

attentive both to developments at the international level and to explanations at the

domestic level. Domestically, first in Southern Europe, and then in Latin America,

developments in regional law and in domestic politics permitted human rights activists to

press for greater accountability, and in particularly fortuitous situations, they made

individual criminal accountability possible. Multiple domestic experiences with

individual criminal accountability thus created the backdrop against which the

international community developed new international legal doctrines and fashioned new

international tribunals, especially the ICC.

International relations theorists have increasingly called attention to the role of

region in world politics65, and diffusion theory has identified strong regional diffusion

effects.66 But awareness of the importance of region has not been fully integrated into

norm theory or diffusion theory, nor do we understand well the mechanisms through

which regional diffusion operates. A regional dummy variable often explains important

variation in quantitative models, but we don’t know exactly what region is a proxy for, or

what it means. The notion of an early adopter or late adopter is region-specific, not

global. Models diffuse first within regions. Later there are some country cases that

permit region-spanning, or region-jumping horizontal diffusion. The Greek prosecutions,

for example, did not become a model that could jump regions. The South African Truth

and Reconciliation Commission (TRC), however, was not only hugely important within

the African region, but was also a clear region-jumping model, having broad influence

globally. The anti-apartheid struggle has mobilized the whole globe, and all eyes were

65 Katzenstein, 200566 Simmons, 2009

43

turned to South Africa to see how it would handle the transition. The South African

TRC, with the blessing of Nelson Mandela, and under the leadership of Bishop Tutu, set

the new standard for transitional justice. The South African model implied prosecutions

for those who did not come forward and tell the truth to the TRC but, in practice, has not

resulted in prosecutions.

Some regions have a stronger human rights consciousness that has become

embedded in regional human rights law and institutions; these institutions are an

important part of the explanation for why prosecutions were stronger in some regions

than in others. Specifically, the oldest and strongest regional human rights regime is the

European Convention of Human Rights and the corresponding European Court of Human

Rights, which was a centerpiece in the origins of prosecutions in Greece.

The next strongest regional human rights regime is in the Americas, with its

American Convention of Human Rights, the Inter-American Commission and Inter-

American Court of Human Rights. Compared to Western Europe, where most countries

were democratic, military coups swept through Latin America in the 1970s, and the Inter-

American human rights system existed in uneasy contradiction with the practices of many

member states. The Inter-American Commission managed to play an important role in

this challenging regional environment. In every Latin American country that eventually

adopted prosecutions, the Inter-American Commission had issued an earlier report on

human rights practices that recommended judicial accountability. Finally, Africa has the

third such regional human rights regime, but the institutions are more recent and as yet

weaker than those in Europe and the Americas. There are no regional human rights

institutions in either Asia or the Middle East, and this helps explain the relative paucity of

human rights prosecutions in those regions.

Africa has been the source of horizontal diffusion of the truth commission model

from South Africa to other countries in Africa and elsewhere in the world. In addition,

bottom up vertical diffusion from South Africa to INGOs also occurred. South Africa’s

truth commission, for example, was instrumental in the emergence and practices of the

world’s main transitional justice INGO, the International Center for Transitional Justice.

Some scholars of transitional justice point to the ICTJ as an example of top-down

diffusion, where a “one size fits all” model of transitional justice is being exported from

44

the New York offices of the NGO to countries around the world. They miss, however,

the actual historical diffusion path of the truth commission model associated with the

ICTJ. Truth commissions started in Argentina in 1983 as an autonomous innovation of

policy makers in coordination with the human rights movement there. The Argentine

experience in turn influenced the Chilean truth commission, and both Argentines and

Chileans were invited to South Africa to share their experiences. But each new truth

commission was not simply a copy of what came before but was consciously crafted to

try to avoid earlier problems or to fit the particular situation of the transitional country.

South Africans crafted their own unique version of a truth commission, both building off

the Argentine and Chilean experiences, and adding important new tools, such as the

provision that perpetrators would only receive amnesty if they provided true testimony to

the commission. This model, in turn, helped create the very idea that there was a

category of institutions that could be called “truth commissions.” There is now a

powerful “industry” around the idea of truth commissions and transitional justice more

generally, and, in the words of Leigh Payne, a “memory market” in parts of the world.

But the origins of this industry and its diffusion path call into question many “top down”

understandings of the sources of norm emergence.

To date, the life cycle of the trial norm has cascaded in Latin America and in

Europe, but not elsewhere. The truth telling norm has cascaded in Africa and Latin

America, but not in other regions. Whole regions, like Asia and particularly the Middle

East, are still virtually untouched by the accountability norm.

One important debate in the literature is whether diffusion happens as a result of

a conscious or coordinated action, or whether it results from uncoordinated action. In the

justice cascade, early norm adoption was almost always the result of conscious and

coordinated action. But, at some point in the justice cascade, explicit brokerage

apparently becomes less important and more uncoordinated forms of diffusion take

precedence.

This book proposes a theoretical story about norm emergence and norm effects

that I believe is of relevance to other issues of change in world politics. Norm change

begins with political struggle, often led by norm entrepreneurs. Such initial norm change

then requires fortuitous circumstances to succeed. These norm entrepreneurs are pushing

45

in multiple places, but for their advocacy to translate into policy, a series of more

structural and institutional factors must align, such as the ways in which a ruptured

political transition weakened spoilers and made prosecutions more likely. Once new

norms start succeeding in multiple setting, they provoke imitation and experimentation.

Eventually a model of the new norm emerges – what the appropriate behavior looks like.

This model is increasingly diffused, both through international organizations67 through

INGOs, and through observations of practices of neighboring states.

Widespread norm diffusion, in turn, alters the strategic games for key actors

because it imposes costs and grants benefits to different actors. Some of these actors

believe in the norm, but others don’t, and they only change behavior according to the new

set of incentives and costs. In other words, the model proposed here integrates various

approaches to international relations, incorporating existing theories about norm

emergence, diffusion, and rational action, and eventually internalization of new norms. It

argues that rather than constantly struggle in the paradigm wars between theoretical

models, we need to ask ourselves which theoretical mechanism is appropriate at which

political moment or stage we are studying. But I go further than saying that we need to

mix theoretical models. I propose a specific sequencing of theoretical mechanisms,

connected to the stage of the life cycle of the norm. Most politics is “normal politics”

with existing norms and clear sets of incentives in place for rational action. Such normal

politics takes place both within deep (but accepted, and internalized) norm structures, and

common rules of the game for rational action. Standard theories of international relations

can tell us much about how to analyze these situations. Rational choice provides tools to

analyze short term interactions, and structural constructivism helps us understand how

ideas structure the very understandings that make the game possible. Constructivism is a

theoretical approach that takes seriously the role of human consciousness – including

beliefs, ideas, and norms – in social life. Constructivism is most associated with helping

us understand that in many social situations actors don’t make conscious choices about

how to behave but instead are guided by almost automatic understandings of what is

appropriate behavior in particular circumstances. For example, when an insurgent group

wins a civil war, they don’t think about what kind of entity they should set up, but rather

67 Finnemore, 1993

46

declare themselves an independent state and appeal to world society for admission into

the international system. Constructivists call this form of action a “logic of

appropriateness,” and it can be contrasted with an alternative model of behavior – “a

logic of consequences”--where rational actors weigh costs and benefits of outcomes and

choose actions accordingly. We can call this approach “structural constructivism”

because it shows how the structures that condition international outcomes are not only

material power or wealth, but also ideational structures that have a taken-for-granted

status that nevertheless shaped international politics in equally powerful ways. So, for

example, both rational choice and structural constructivism help us understand why it

was so difficult to initiate individual criminal accountability in the world. Powerful

leaders worked to insure that they had wide latitude for action, including the ability to kill

their opponents with impunity. We can see that it might indeed be in the “interests” of

these leaders to know they could not be held accountable. But, equally powerful, were

the beliefs held not only by leaders but by the general population who could not imagine

that it would be possible to hold leaders accountable. So, these approaches are useful in

helping explain why it took so long for individual criminal accountability for human

rights violations to emerge. But neither of these approaches is good at understanding

why and how change occurs in these structures and rules.

This is because change cannot be understood without seeing how new ideas

emerge and spread. These new ideas often challenge the older ideas that constitute the

ideational structure of the international system. For example, in the issue of prosecutions,

old ideas about sovereignty and sovereign immunity maintained that heads of state should

not and could not be prosecuted. The new ideas about individual criminal accountability

for human rights violations stress that state officials should and could be held

accountable. These new ideas “catch on” through new logics that are not captured by

either a “logic of consequences” or a “logic of appropriateness.” When Argentine

activists first pushed for criminal trials, they were not following any logic of

appropriateness. They were being consciously and explicitly “inappropriate” – and they

knew it, which was why it was a frightening thing to consider at first. These new logics

can be explained by “agentic” constructivism. Like constructivism more broadly, it is

concerned with the role of human consciousness in international politics, but unlike

47

structural constructivism, it is not mainly interested in how deep ideational structures

constrain imagination and action and perpetuate continuity and logics of appropriateness.

Rather, agentic constructivism is concerned with how new ideas emerge and catch on,

and eventually challenge old logics of appropriateness and old logics of consequences. If

successful, such new ideas may eventually create new logics of appropriateness and new

logics of consequences. In other words, they will create new understanding of the ways

states ought to behave, and new understandings of the very national interests of states.

One way that new ideas catch on is through what Thomas Risse has called a “logic of

arguing” where the best argument can sometimes win out. While some new ideas catch

on through this more cerebral process of evaluating the best argument, other ideas catch

on through less cerebral processes, involving struggle, pressure, and trend-following.

Snyder and Vinjamuri discuss a “logic of emotions” in relation to transitional justice,

which stresses “the need for emotional catharsis in the aftermath of social trauma,” but

they make clear that such a logic is not to be trusted and instead advocate a logic of

consequences as the “pragmatic approach to justice.”

But not all new ideas “catch on.” This is the single most difficult issue to explain

in the norms life cycle – why do certain ideas at certain moments in certain places

resonate, grab attention, and become possible? In addition to the cerebral processes of

the logic of arguing, and the pressure politics that make trials a rational response to public

pressure, I will argue that which ideas catch on are also related to deeper ideational

instincts in the human brain. The idea of justice and the need for punishment for those

who violate societal norms is deeply embedded in many societies around the world. So

there is an initial receptivity to demands for justice. But in early adopters, domestic

norms and domestic political conditions must also be propitious for such norm change to

occur. In Greece, for example, human rights prosecutions were more possible because

Greece had a long tradition of using political trials.

As more and more countries begin to adopt new norms, however, domestic

circumstances become less important, and international and regional contexts begin to

play a bigger role. Countries that originally didn’t have the domestic conditions for using

prosecutions are later able to use prosecutions, both because domestic conditions for

prosecutions change with the passage of time, and because the broader adoption of

48

prosecutions in other countries make them more acceptable and less costly to use. While

prosecutions were considered impossible in many transitional countries immediately after

transitions, with the passage of time, conditions changed and prosecutions became not

just possible but likely. Once new normative structures are in place, they alter not just

the way people think, but also the incentive structures for behavior. The effects of the

justice cascade are thus the result of both changing norms and of changing incentive

structures put in place by the possibility of prosecutions.

II. Effects of the Justice Cascade

The results of the analysis summarized in Chapter 5 provide some of the first

quantitative support for the existence of a deterrence effect in the realm of human rights.

In our research, both human rights prosecutions and truth commissions are associated

with improvements in human rights protection, and prosecutions in neighboring countries

have a possible deterrence impact beyond the confines of the single country. These

findings suggest that the impact of prosecution is the result of both normative pressures

and material punishment.

We conceptualize human rights prosecutions as a type of enforcement of existing

human rights norms, and our findings are generally supportive of various theories that

posit that an increase in enforcement of norms, in the likelihood of punishment, and in an

increase in costs of repression should lead to a decrease in actual human rights violations.

But our findings are also consistent with the norms literature,68 which has stressed that

both normative and coercive factors are important for human rights change. Human

rights prosecutions are not only instances of punishment or enforcement but are also

high-profile symbolic events that communicate and dramatize norms. It is difficult to

isolate the material costs of punishment from its social effects as an expression of social

disapproval. Informal social sanctions may follow from the formal sanctions of

prosecutions and can have important effects in political arenas where reputation is

essential.

68 Risse, T., S. C. Ropp, et al., Eds. (1999). The Power of Human Rights. International Norms and Domestic Change. Cambridge, Cambridge University Press.

49

By comparing the impact of truth commissions and prosecutions, however, we are

able to separate out a purely normative mechanism like a truth commission from one like

a trial that imposes both material and social costs. Because truth commissions do not

result in any material punishment of individuals, if only material costs matter, truth

commissions would be unlikely to have any independent effect on human rights

practices. If, however, both social and material costs are important, we would expect to

see that truth commissions have an impact on human rights practices.

The case studies help us understand how normative and strategic factors work

together to bring about changes in practices. The entire first part of the book outlines this

norm emergence story. But, once the new prosecutions and tribunals were up and

running, they started to alter the strategic landscape within which state officials operated,

and a more standard strategic story began to unfold along side of the norms story.

In the past, state officials faced few costs for committing human rights violations;

now they face more, especially in those countries that have been persistent in carrying out

human rights prosecutions. This suggests that there has been an increase in enforcement,

which, in turn, heightens the expected costs of repression for government officials who

make choices about how much repression to use. During the time period examined here,

the benefits that state officials receive from repression have remained constant and the

formal and informal costs of repression have increased. It thus appears that prosecutions

may deter future human rights violations by increasing the perception of the possibility of

costs of repression for individual state officials.

The threat of prosecution is a particularly unambiguous type of communication.

Exactly because the model of individual criminal accountability is drawn from domestic

judicial systems, the meaning of the threat of prosecution is clearer to all involved. The

association is made between human rights violations and common crime. As such, the

threat of prosecution may be a more effective form of communication than foreign policy

statements of states.

Much of the earlier work on enforcement of human rights law, including my own,

has examined the role of NGOs and states in bringing pressure to bear on countries that

violate rights. But the foreign policies of major states have multiple goals, and they

frequently give mixed signals on their commitments to human rights. In the United

50

States, the Secretary of State and the Secretary of Defense, for example, may give human

rights concerns a different priority. Thus, foreign policy communications of the United

States, and other states as well, may often offer mixed signals as regards human rights.

Repressive regimes, in turn, are often ideologically invested in repression and often

embrace ethnic or political ideologies that portray them as the saviors of the nation.

When they receive mixed signals, they are inclined to “hear” or receive those messages

that are most consistent with their world view and self-perception.69 A threat of human

rights prosecution, however, is harder to misinterpret than the standard foreign policy

statement.

Even in the face of such unambiguous communication, for individual leaders and

security forces that have already carried out human rights abuses, the strategic landscape

is straightforward: it is in their interests to prevent both truth-telling and especially

prosecution for past human rights violations. These are so-called “spoilers”, who are

often willing to go to great lengths to prevent prosecution. Neither the threat of

prosecutions nor a truth commission is likely to deter them from future human rights

violations. Indeed, as realists have suggested, the threat of prosecution could entrench

them in power. We don’t even use the term “deterrence” to refer to individuals who have

already committed crimes, since deterrence involves the inhibiting effect of sanctions on

people other than the sanctioned offender. States can only affect their actions by

preemption: by physically removing them from power through detention or

imprisonment.70 Given a choice, spoilers will always prefer no transitional justice at all,

preferably guaranteed by an amnesty. They very often succeed in blocking domestic

prosecutions, through threats, coup attempts, obstructed peace processes, and the like.

This group has been the main concern of scholars like Snyder and Vinjamuri, who point

out the obvious difficulties of such a strategic situation. What they miss, however, is the

bigger and longer strategic and norms game that prosecutions can set in motion.

69 Kathryn Sikkink, Mixed Signals: U.S. Human Rights Policy and Latin America (Ithaca: Cornell University Press, 2004.70 Mendeloff, D. (2004). "Truth-Seeking, Truth-Telling, and Postconflict Peacebuilding: Curb the Enthusiasm?" International Studies Review 6(3): 355-80.

51

In the longer term, future human rights violations can be prevented through the

impact of prosecutions on the new generations of military and police officers, and on

civilian political leaders. This is mainly a domestic mechanism. Young officers who

were not involved in the last round of repression may look at past leaders and draw

conclusions about their future choices. They observe former leaders, perhaps imprisoned

as a result of domestic prosecutions, or with tattered international and domestic

reputations. Future military officers may decide that prosecutions have made repression

and coups too costly for use in the future. How these leaders calculate costs may vary.

Thus, some may interpret the costs of prosecutions mainly in terms of the costs to the

reputation and honor of the military, the police, or the party as an institution; others may

calculate the costs in a more individualist fashion. Prosecutions and truth commission

created sanctions that did not exist before, including not only the possibility of

imprisonment, but also damage to honor and national and international reputation.

The costs of prosecutions are both economic and political, and they are both

personal and institutional. They include formal sanctions (lost wages, litigation fees,

inability to participate in elections while on trial or in prison, etc.) and informal social

and political costs that arise from the publicity surrounding trials (loss of reputation or

legitimacy, and the resulting loss of political and social support). Prosecutions for

individual criminal accountability publicize the new norms. But not all prosecutions

publicize equally. Prosecutions of well known and high-level officials have a much

greater publicity effect than prosecutions of lower level and unknown officials. Thus, the

prosecutions of Pinochet, Milosevic, and the military juntas in Argentina had a much

greater publicity effect that the prosecutions of lower-level police officers in Portugal

after the revolution there in 1974. The impact of the prosecutions of lower level officials

in the ICTY, for example, had a lower publicity effect than the trial of Milosevic or

Karadzic. The type of prosecutions may be less important than the fame and level of the

official being tried.

Second, in order for the publicity effect to be “received,” the potential recipient of

the message must see himself as somehow similar to the individual being prosecuted.

Thus, the trial of Pinochet had a greater impact on other military officers in Latin

America than it did on U.S. officials in Washington D.C. Geographic, cultural and

52

linguistic proximity can all contribute to the perception that actors are similar to those

being tried. Although the prosecutions of the Greek juntas in 1975 were very high level

prosecutions of the top leaders of the military regime, and they resulted in some severe

sentences, they had very little impact in other parts of the world. Neither the Latin

American militaries nor human rights activists had much awareness of the Greek

example. The Argentine military coup happened one year after the Greek prosecutions,

and yet there is no evidence that the Argentine military had any awareness of the Greek

prosecutions, or the long term implications of such a legal move for their own future.

They were aware of the Nuremberg precedent, but they interpreted it as evidence that

only a defeated military would be prosecuted.

III. The Ethics of Human Rights Prosecutions: Principles and Consequences

A book on this topic calls out for a discussion of the ethics of human rights

prosecutions. And yet, it is often difficult for political scientists, indeed for any scientist,

to talk about ethics. Political scientists sometimes believe that if you take an explicit

ethical stance on an issue, it calls into question your scientific objectivity and thus

undermines the credibility of your research. But political scientists may also shy away

from normative theorizing because we have not found satisfactory ways to combine our

ethical commitments with our empirical research. As long as the appeals to address

ethics essentially said “stop being a researcher and become a moral philosopher,” it

wasn’t very attractive to me. I had neither the training nor the inclination to engage in

abstract moral philosophy.71 In this section, I attempt to outline an explicit approach to

the ethics of human rights prosecutions that combines attention to both normative issues

and empirical research findings.

Ethical theorizing is typically divided into two main approaches: deontological or

rule-based approaches and consequentialist approaches. In other words, when we make

ethical judgments, should we be concerned with how well our actions and the actions of

71 I’m indebted to the work of colleagues who have also charted such an approach: Matthew Gibney, The Politics and Ethics of Asylum: Liberal Democracy and the Response to Refugees(Cambridge University Press, 2004); Michael Barnett, Eyewitness to Genocide: The United Nations and Rwanda (Cornell University Press, 2002); Joseph Nye, Nuclear Ethics (New York: Free Press, 1986), J.L Holzgrefe and Robert O. Keohane, (eds), Humanitarian Intervention: Ethical, Legal, and Political Dilemmas (Cambridge: Cambridge University Press, 2003)

53

others conform to preexisting moral rules, or should we take into account primarily the

consequences of those actions? For example, should we be more concerned about

following the rules of international law for punishing human rights violators, or should

we pay more attention to the consequences of holding trials? My answer is “both of the

above.” I believe that when we begin to combine ethical and empirical inquiry,

deontological and the consequentialist concerns are intimately linked. 72 In this book, I

argue that ethical judgment requires both choices of principles and evaluation of

consequences in terms of those principles. In other words, to answer the question “what

to do?” we need to ask not just “what is right” but also “what may work” to bring about

outcomes consistent with my principles. 73 For example, in the case of human rights

prosecutions, it is not enough to ask, “is it right or good to hold human rights

prosecutions?” We also want to know what impact human rights prosecutions have on

actually protecting human rights. The answer to the second question—“what impact do

human rights prosecutions have?”—could affect the answer to the first question—“is it

right or good to hold human rights prosecutions?” Discerning and evaluating

consequences is an inherently comparative and empirical enterprise, and thus empirically

grounded scholars can make an important contribution.74

But in order to weigh consequences, we must first have specified what criteria we

intend to use to evaluate which consequences are most valued and beneficial. In this

book, I will inquire about the consequences for human rights, as defined in current human

rights law. Prominent philosophers like Amartya Sen and Martha Nussbaum use the

fulfillment of rights as a starting point for normative theorizing.75 But philosophers often

begin as if the world were a tabula rasa on which they could construct the set of

principles of their choice. They believe that they must start from scratch in inventing

their central list of rights or capabilities.

72 Similar points have been made by Nye, 1986, and J.L. Holzgrefe, “The Humanitarian Intervention Debate,” in Holzgrefe and Keohane, Humanitarian Intervention, 50-51.73 I am indebted to Richard Price for this particular formulation.74 Richard Price, ed. 2008; Richard Price, 2009.75 Amartya Sen, “Rights and Agency,” in Consequentialism and Its Critics, Samuel Scheffler (ed) (Oxford University Press, 1988), 187-223; Martha C. Nussbaum, Woman and Human Development: The Capabilities Approach (Cambridge: Cambridge University Press, 2000).

54

But it seems illusory to act as though the modern world weren’t dense with

existing norms and laws, and to pretend that we could design the ideal code we could

imagine. I prefer to start with existing human rights principles embodied in international

human rights law, especially the International Covenant on Civil and Political Rights

(ICCPR) and the International Covenant on Economic, Social and Cultural Rights

(ICESCR). When one set of human rights principles come into conflict with another, I

prioritize what human rights law calls the “non-derogable rights” in the ICCPR,

especially the right to life and freedom from slavery and torture. What “non-derogable

rights” mean is that under human rights law there are certain rights that cannot be

violated in any circumstance. Many people think that all rights are non-derogable. But

human rights law has long recognized that in times of emergency governments may need

to restrict certain rights. That does not mean that anything goes in a time of emergency.

Human rights treaties spell out what constitutes an emergency and how governments

should behave in these emergency situations. The ICCPR defines an emergency as one

“which threatens the life of the nation.” So, for example, in some situations of war, a

government may legitimately put some restrictions on the freedom of speech or press. It

cannot, however, enslave or torture prisoners in any circumstance. This was a point

missed by many of the governments considered in this book, including some members of

the Bush Administration in the “war on terror,” who thought that an emergency situation

was a carte blanche to violate the non-derogable right against torture.

This non-derogable rights clause comes as close as I can find to expressing a

genuine international consensus about what rights are most important. The international

community, when drafting one of its central human rights documents, decided that a

small handful of rights could not be put aside under any circumstance. This seems like a

good starting place for normative argument. Starting with actually existing international

norms drafted through exhaustive debate and consultation among many states has the

virtue of being less ethnocentric than having the analyst substitute her own normative

criteria.

Some people say that international law is simply the crystallization of global

power relations where the wealthy North imposes its agenda. But close studies of the

drafting processes of the Universal Declaration of Human Rights, the ICCPR, and the

55

ICESCR reveal that delegates from countries representing diverse cultural, political, and

theological positions debated virtually every phrase in these documents in hundreds of

consultations and meetings.76 In some cases, the deliberation was so extensive that

delegates despaired they would ever finish. In 1948, one delegate complained that they

had been debating the draft of the Universal Declaration for Human Rights for three

weeks and had only adopted 2 out of the 28 articles. Six days later, article three was

adopted. Because delegates hoped to finish before the end of the year, the Committee

began to have night meetings as well. Two months later they completed their

deliberations, and the full Declaration was approved by the General Assembly.77

It is true that many countries, especially in Africa, were not involved in the

negotiations for the UDHR because they were still colonies at the time. But since that

time, most have participated in negotiating many of the other human rights treaties.

Finally, countries also get to signal their consent to human rights law through the process

of ratification, through which each country voluntarily agrees to abide by the treaty.

Secondly, by their ratification of the ICCPR including its non-derogable rights clause,

and the ICESCR, an even wider range of countries has expressed their support for these

principles and norms. So, today, over 150 countries have signaled commitment to these

norms by the voluntary ratification of the ICCPR and the ICESCR, including over forty-

five African countries, which were the single largest group not present at the drafting

stage, since most were still colonies at that time. Finally, thousands of non-governmental

human rights organizations representing citizens from around the world have also

debated and embraced these norms.

Because the deliberative processes that went into the construction of these

international human rights norms was voluntary and widely participatory, these human

rights norms are a more legitimate source of general principles than any I or any other

individual or group of researchers could invent. I do not argue that international law in 76 See Johannes Morsink, The Universal Declaration of Human Rights: Origins, Drafting and Intent (University of Pennsylvania Press, 1999); Mary Ann Glendon, A World Made New: Eleanor Roosevelt and the Universal Declaration of Human Rights (NY: Random House, 2001); Paul Gordon Lauren, The Evolution of Human Rights: Visions Seen (Philadelphia: University of Pittsburgh Press, 1998); Mary Ann Glendon, “The Forgotten Crucible: The Latin American Influence on the Universal Human Rights Idea,” Harvard Human Rights Journal Vol. 16 (Spring 2003).77 Glendon, A World Made New, p. 151.

56

and of itself tells us what is right. Rather, I argue that the manner in which the

international human rights law was drafted provided a process, a way of finding out what

is right.78 First, by consulting as diverse a group of individuals as possible from a wide

range of backgrounds and cultures, it insured the inclusion of a diverse set of views. It

asked states ahead of time to formulate rules to govern their behavior, and then it

provided non-coercive processes through which actors new to the process could give their

consent to those rules through ratification and make reservations to fit their own cultural

or legal situations. The process of making human rights law provided ample

opportunities for making reasoned and principled arguments. It happened in a context of

international power politics, of course, but relatively few exercises of direct power were

used to twist arms and coerce either speech or action. If you look at the debates over

human rights law, you find very few instances where one state threatened others to

change their position on the law. Oddly, it is possible that little force was used because

the human rights regime was often seen as symbolic and simply not important enough to

justify the use of power or resources.

Power is exercised at different levels, and direct coercion or compulsion is only

the most obvious form, as Michael Barnett and Raymond Duvall have reminded us.79 To

the degree that structural power is at work in the drafting of laws about human rights and

accountability, the most striking point is the dominance of an individualist perspective,

and the empowerment of the individual vis-a-vis the collectivity and the state. An

individualist perspective is indeed inherent in a Western liberal legal and philosophical

tradition, but empowering such an individual vis-a-vis her state runs deeply counter to the

structural power of states in the state system. Thus, human rights law presents a

quandary for scholars of structural power. We would expect that when state power

comes up against liberal ideas of individual rights asserted against the state, state power

would win hands-down.

To escape this quandary, international relations scholars sometimes claim (usually

ex-post facto) that something unexpected (like empowering citizens to prosecute state

78 I want to thank the participants at a seminar at the Kennedy School on ethics and human rights, and in particular, John Tasioulas, for helping me think about and formulate my argument in this way.79 Barnett and Duvall 2005.

57

officials) is actually a way of reproducing the state and reaffirming it at the same time. It

is not at all clear to me how a movement to prosecute state officials serves in any general

sense to strengthen, reproduce, or reaffirm the state. Certainly, such legalization

strengthens the judicial branch at the expense of the executive branch. Certainly, as

Stephen Krasner has reminded us, the ratification of human rights treaties is an exercise

of one form of sovereignty, since only sovereign states can ratify treaties, even as it

erodes another kind of sovereignty. In this sense, we can say that states “invited” the

process of individual criminal accountability, even if they were often later unpleasantly

surprised, as in the case of Pinochet, with the results of their actions.

In the entire history of human rights law-making with regard to justice, by far the

most coercive moment occurred quite late in the game, when the United States first

opposed the Statute of the International Criminal Court, and then later tried to undermine

it by seeking bilateral agreements with as many countries as possible saying that they

would not submit US troops to the Court. Here, the United States expended significant

power resources to secure such agreements, providing additional aid, threatening to cut

military economic aid and training, and actually cutting such aid. There is no other

example in the history of human rights law where a powerful country expended such

resources to secure a particular human rights legal outcome. But note that this example

runs counter to the argument that the powerful impose international law. In this case, the

most powerful state of the system tried to block the implementation of human rights law

exactly because it had failed in its efforts to control the drafting process. And yet, the

U.S. campaign against the ICC was oddly ineffective, and even the United States was

forced to admit it eventually had been counterproductive to other U.S. interests.

International human rights law now represents an “overlapping international

political consensus” that we can use to evaluate the consequences of human rights

prosecutions.80 There are still lots of disagreements about human rights in the world.

But there tend to be fewer disagreements about the non-derogable rights. This may be

because psychological research suggests that nothing does greater and more long-lasting

harm to people than when other people intentionally and directly inflict bodily harm on

80 Jack Donnelly, Universal Human Rights in Theory and Practice (2nd ed) (Ithaca: Cornell University Press, 2003), 40.

58

them. The bonds of trust on which human communities are founded are sundered most

completely by the direct and intentional inflicting of pain and suffering of one human on

another. Research on Post-Traumatic Stress Disorder (PTSD), for example, shows that

human induced traumas from torture or child abuse are more severe and more difficult to

recover from than trauma produced by accidents or natural disasters.

Some recent cross-cultural research in psychology suggests humans have a special

aversion to violating bodily integrity. This small core of basic rights that form the “non-

derogable” rights of the human rights regime may be intrinsically appealing to many

individuals. Some research in psychology suggests that most cultures share a common

set of moral “instincts,” including five moral themes that appear again and again in

surveys of individuals around the world. These themes may reflect biologically

programmed behavior in some way, though their interpretation varies greatly among

cultures and across time. One of these principles involves basic notions of fairness that

are at the basis of the justice cascade. In addition, Steven Pinker, psychology professor at

Harvard argues that one common moral attribute is that “people feel that those who

commit immoral acts deserve to be punished. Not only is it allowable to inflict pain on a

person who has broken a moral rule; it is wrong not to, to ‘let them get away with it.’”81

While psychologists can speak of such a “moral instinct”, it is still heresy for a

political scientist. It is perfectly acceptable for political scientists to suggest that all

humans have an innate drive for power or wealth. But shunned is the suggestion that

many humans intrinsically find certain human rights ideas appealing. Yet it does not

strike me as particularly odd to suggest that almost everyone would prefer to be alive

than dead, free than imprisoned, secure than tortured, fed than hungry. Core human

rights norms have resonated so profoundly in the world in part because of this intrinsic

appeal. It is surprising to watch how quickly humans, even those embedded in cultural

scenarios that tell them otherwise, come to believe that they are entitled to something

better.

My argument for the normative primacy of core human rights does not rest nor

depend on such foundational claims. It is sufficient for our purposes that many states,

81 “The Moral Instinct,” New York Times Magazine, January 13, 2008

59

groups, and individuals, through an extended process of international negotiation,

mutually arrived upon a set of international human rights norms and that since that time,

virtually all of these states and many more have further endorsed such norms through the

voluntary ratification of the relevant treaties. But my causal explanation for why these

norms resonate so deeply around the work, and thus why the justice cascade took off

rapidly, is supported by claims about the moral instinct.

International human rights law is comprehensive, but also incomplete. For the

issues that concern us here, the law is not always clear. Major human rights treaties

(ICCPR and the ICESCR) do not establish a clear duty for states to punish perpetrators of

human rights violations. The ICCPR states that victims have a right to a legal remedy,

but it does not clarify that this remedy must involve criminal prosecution. The duty to

punish is still an evolving norm. New norms are ethically and empirically distinct from

existing norms because they do not yet represent any international consensus. Ethically,

they represent a proposal for normative change by one group that has not yet been

broadly endorsed. Empirically, the dynamics of new norms are different from those of

well established norms.82 In our case, the norm against torture is hard law, but the norm

that individuals who engage in human rights violations should be held criminally

accountable and punished is softer. The people and movements we study in this book are

working to “harden” that law through drafting new legal standards, and through their

ongoing practices.

IV. Evaluating Consequences

The significant philosophical debate about consequentialism and the closely

related school of utilitarianism has not hinged on what would seem to be a glaring issue

for any empirical researcher. How can we know with any confidence about

consequences? A large body of (consequentialist) normative thought is based on the

notion that morality depends on what we can reasonably expect to happen. But most

philosophers aren’t in the business of trying to find out what we can reasonably expect to

82 Martha Finnemore and Kathryn Sikkink, “International Norm Dynamics and Political Change,” International Organization, Vol. 52:4 (Autumn 1998), 887-917

60

happen.83 The task is left to empirical researchers to fill the gap. But having filled the

gap, we then need to reconnect our conclusions back to their normative implications.

In the debate over human rights prosecutions, the empirical question of what we

can reasonably expect to happen has important ethical implications. If Jack Snyder and

Leslie Vinjamuri are correct that under certain circumstances, such prosecutions can lead

to more atrocities, sustain conflict, and undermine democracy, a consequentialist reading

would suggest that countries should not carry out such prosecutions.84 If, to the contrary,

as I argue in this book, there is little empirical support for the argument that prosecutions

lead to more atrocities, sustain conflict or undermine democracy, the normative equation

is different.

This may help clarify why, for me, careful empirical research using the best tools

at our disposal is not just a professional obligation, but also an ethical one. Exactly

because the ethical and normative questions are so important, and because many ethical

judgments require some knowledge about reasonable expectations about consequences,

good research is necessary for ethical judgment.

V. Prosecutions and Consequences

The final question that concerns us in this book is: given that human rights

prosecutions are occurring with considerable frequency, what impact or consequences do

they have? A purely deontological approach might say that providing justice for victims

of human rights violations and accountability for perpetrators is such an important

principle that countries should proceed with prosecutions regardless of the consequences.

While I understand and respect that ethical argument, I’m not willing to make it myself.

I agree that providing justice and accountability are worthy goals that should be pursued.

However, the duty to punish past human rights violations does not form part of the core

“non derogable rights” of the human rights regime. There is an emerging norm and some

hard law that states have an obligation or duty to punish individuals who carry out mass

83 For example, Shaw argues that “most expected outcome utilitarians” think about consequences as “calculated by a reasonable and well informed agent based on the available evidence.” William Shaw, Contemporary Ethics, 30. This description sounds like a reference to an empirical researcher.84 Snyder and Vinjamuri, “Prosecutions and Errors.”

61

atrocities.85 But even this law does not specify that states must prosecute all those guilty

of core human rights violations. Just as with domestic criminal law, states have

discretion about which crimes they will prosecute, and they are permitted to use limited

resources to pursue those cases that are most severe, to pardon convicted offenders, and

to offer plea-bargains for accused that cooperate in prosecuting other criminals. Thus,

even the hardest international law on a duty to prosecute would still leave considerable

leeway for states to make decisions about who and what to prosecute. In this case, I

believe that knowledge of expected consequences is important for helping make hard

ethical choices.

Particularly in cases where important and valued principles are at stake, the

presumption should be in favor of the principle, and only substantial and well-established

evidence against the principle would be sufficient to lead us to call it into question. So,

for example, the principle of accountability for past human rights violations is

sufficiently well established that in order to suspend it, we would want to see very

persuasive research that it is counterproductive.

VI. Weighing Consequences

So, how do we go about weighing consequences? I will argue that three

particularly prominent forms of reasoning are often used to evaluate consequences: (1)

comparison to the ideal; (2) counterfactuals; and (3) empirical comparison, both

qualitative in the form of case studies, and quantitative studies on a large number of

countries. I will discuss each of these forms of reasoning below.

Comparison to the Ideal

Some scholars make a distinction between “ideal theory” and “non-ideal” theory,

that is, the difference between theorizing about the ethical ideal vs. theorizing about what

is possible for governments or individuals to do in a non-ideal world.86 I prefer to think

about this distinction as the difference between “comparison to the ideal” – a comparison

of what actually happened to what should have happened in an ideal world – and

85 Juan Mendez, “In Defense of Transitional Justice,” Transitional Justice and Rule of Law in New Democracies, A. James McAdams (ed) (Notre Dame: University of Notre Dame Press, 1997), 5.86 See, for example, Gibney, The Ethics and Politics of Asylum

62

empirical comparison. Comparison to the ideal can be explicit or implicit. Implicit

comparison to the ideal is common in discussions of human rights prosecutions. So, for

example, many people have discussed the flaws in the international prosecutions of the

ICTY in terms of its failure to actually arrest the most senior suspected war criminals;

for its wrongheaded efforts at even-handedness; or because, by giving Milosevic the

chance to represent himself, the Tribunal gave him yet another opportunity to traumatize

his victims, this time when they testified against him. Because these flaws offend the

sensibilities of those committed to justice, they make an argument that it would have

been better to have no prosecutions at all than to have the flawed prosecutions of the

ICTY.

Human rights activists most often engage in comparison to the ideal. Activists in

Argentina, for example, were not satisfied with their historic human rights prosecutions

of their juntas in 1985. Only five of the nine junta members were convicted, and only

two of them received a life sentence. Human rights activists thought that all the junta

members should have been convicted, and that more of them deserved life sentences.

This is often the job of human rights activists, to be a voice of conscience and to

constantly remind us of the distance between our practices and our ideals. But it is not

necessarily the task of the social scientist.

Many scholars also engage in comparison to the ideal. There has been a strong

tendency among academics to engage in comparison to the ideal when it comes to

international and national justice. This comes from a variety of sources. Some of the

critique comes from scholars strongly committed to rule of law principles who find that

human rights tribunals, in general, and international or foreign prosecutions, in particular,

fail to live up to these rule of law principles. Others are committed to the rights of

victims and note how courts of law may retraumatize or undervalue the victims. Some

scholars are critical of any judicial form involving excessively harsh punishment because

they believe in restorative rather than retributive forms of justice.

Gary Bass, for example, in his book on international war crimes tribunals,

explains the ad-hoc tribunals as something the countries of the West designed to assuage

their guilt for failing to intervene and stop human rights violations in countries like the

former Yugoslavia or Rwanda. This is an example of comparison to an ideal. The ideal is

63

that wealthy states should intervene to stop human rights violations in other states. When

they fail to do so, we hold them up against this ideal and find them wanting. When they

do something else (set up war crimes tribunals, in this case) it is again measured against

the ideal (preventing the human rights violations in the first place) and found wanting,

and thus dismissed as flawed.87 Comparison to the ideal is an important form of ethical

reasoning. We need to keep the ability to hold our actual practices up to our ideals and

constantly measure where they fall short. Such reasoning is a powerful pressure for

change in the international system. It is one of the main tools that advocacy groups use in

the world. But, it is also important to be careful how we use this form of the ideal

comparison and clearly distinguish it from empirical comparison. Most importantly,

comparison to the ideal should be explicit rather than implicit. The author should clarify

that the practice or institution in question is being compared not to an empirical example

in the world, but to a set of ideals of what such a practice or institutions should look like,

and those ideals should be explicitly stated and defended.

Counterfactuals

Much important work has focused exactly on the unanticipated consequences of

trials and transitional justice. But it is necessary to do more than simply highlight

unintended or negative consequences. Rather, it is necessary to weigh the positive

consequences of trials against their negative consequences and arrive at some kind of

conclusion about the overall balance. Such a weighing of consequences is neither easy

nor straightforward. It often involves counterfactual arguments about the past, and also

may imply some kind of prediction about the future, or what some people have called a

“future counterfactual.”88

Counterfactuals are subjunctive conditions statements (i.e. - they take the form, if

x then y would have …) in which the first part of the statement is not true. So, for

example, some people make the counterfactual argument that if President Alfonsín of

Argentina had not held prosecutions for past human rights abuses, the military coup

87 Gary Bass, Stay the Hand of Vengeance: The Politics of War Crimes Tribunals (Princeton University Press, 2000).88 Steve Weber, “Counterfactuals: Past and Future,” in Philip Tetlock and Aaron Belkin, editors, Counterfactual Though Experiments in World Politics: Logical, Methodological, and Psychological Perspectives, (Princeton: Princeton University Press, 1996), 268-288.

64

attempts against his government would not have occurred, and democracy in Argentina

would have been more stable. But note that one can almost always make plausible

alternative counterfactual arguments. For example, if President Alfonsín has not held

prosecutions, the military would have been emboldened and the coup attempts would

have succeeded. Counterfactuals are tricky because we cannot determine which of these

counterfactuals is more valid.

For this reason, I prefer comparative empirical research. On the basis of such

research, we can say, for example, that since 1983 when the first prosecutions were

initiated in the region, there have been only three examples of coups in Latin America,

and none was provoked by human rights prosecutions.89 The remaining countries that

used prosecutions have not had a successful coup attempt since the use of prosecutions

and, in many cases, are increasingly considered to house consolidated democratic

regimes. The argument that prosecutions undermine democracy came largely from

observations of a single case: the early coup attempts in Argentina against the Alfonsín

government after it completed prosecutions of the three juntas for past human rights

violations and embarked upon far reaching prosecutions of lower level officials. But

almost twenty years have passed since those failed coup attempts, and Argentina has had

more transitional human rights prosecutions than any other country in the world, enjoying

the longest uninterrupted period of democratic rule in its history.

Empirical Comparisons

The comparison of practices or institutions not to the ideal but to other current or

historical practices can lead to rather different evaluations. For instance, an evaluation of

the ICTY that uses empirical comparisons to other international tribunals might arrive at

different conclusions than would comparison to the ideal. For example, I would argue

that the ICTY is an example of a “successful” international war crimes tribunal, in the

sense that it actually indicted, tried, and convicted criminals using due process,

something that had not happened since Nuremberg. Instead of holding the ICTY up

against an ideal of how international justice might transpire, I compare it to other cases in

the past and the present. How often in the past did powerful states act to punish human

89 These include the ‘self-coup’ in 1992 in Peru, and coups in Haiti in 2004 and in Ecuador in 2000, none provoked by the threat of prosecutions, and all since reversed.

65

rights violations in countries where they were not the victors in war? Never? If they now

do it in the case of ICTY and ICTR, that makes something new in the area of

accountability for past human rights violations. This looks to me like significant

historical change of a progressive kind. I would argue that with the ICTY, there exists

more accountability for past crimes in the former Yugoslavia that there would have been

without the ICTY (a counterfactual argument) and that with all its shortcomings, the

ICTY is still an improvement in some ways on previous international prosecutions (e.g. –

less an application of victors’ justice than at Nuremberg, fewer immense delays with their

attendant violations of the rights of defendants than at the ICTR). Even weighing some

of the negative consequences, I find the overall balance positive, compared to what

would have happened without the ICTY. This is not to say that we should forget the

problems, or be naïve about our expectations for the future. But, neither should we let

our ideals interfere with the actual documentation of change in the system.

Judgment is a result of a combination of the premises and commitments we begin

with and the empirical research results about the consequences of action. This will

ensure that we render judgments that weigh the limits and possibilities of the real world

of politics. The research will often involve difficult counterfactuals and different kinds of

comparisons. Well-intentioned researchers will disagree about results. We can improve

our discussions by being more explicit about our processes of ethical reasoning and by

relating our research findings more explicitly to their normative implications.

This book is a work primarily of empirical comparison. Much of the work here is

based on years of qualitative field research, especially in Argentina and Uruguay, but also

including some research in Guatemala, Chile, Greece, Portugal, Italy, Belgium, Spain, the

Netherlands, and Brazil. Cases were selected to study the origins of prosecutions, by

looking both at the early adopters of prosecutions, and by comparing those to similarly

situated late adopters and non-adopters. So, for example, I compare the early use of

prosecutions in Greece and Portugal to Spain, a country in the same region with a

transition at a similar time that chose not to use human rights prosecutions. Likewise,

Argentina, an early adopter of prosecutions, is compared to Brazil, Uruguay, and Chile,

all countries in the same region with transitions around the same time that adopted

different approaches to transitional justice.

66

The results of the field research were complemented by quantitative analysis –

through the creation of the data set on human rights prosecutions used to test

systematically some of the hypotheses that have emerged from the general literature and

from the qualitative research. Finally, there is some legal analysis and some argument

employing future counterfactuals. In other words, this is a methodologically eclectic

piece of work, driven by the desire to address certain puzzles and answer certain

questions.

The evidence and the theory presented here leads me to conclude that it is time to

put false dichotomies behind us and begin a more nuanced debate about transitional

justice and about change in the international system. The policy choices are not between

truth or justice, between trials or democracy, or between idealists or pragmatists. Instead

it is much more interesting to examine under what conditions trials can contribute to

improving human rights and enhancing rule of law systems, or what sequencing or

judicious combination a transitional justice mechanisms can help build democracy and

resolve conflicts. The theoretical choices are not between rationalism and

constructivism, or between ideas and power. Rather, it is more true to the empirical

stories in this book to see how ideas and rationality interact at different stages and

sequences to produce political change.

67

Daley, D. a. G., James (2005). "Horizontal Diffusion, Vertical Diffusion, and Internal Pressure in State Environmental P:olicymaking." American Politics Research 33(5): 615-644.

Graham, E., Shipan, Charles, and Volden, Craig (December 2008). The Diffusion of Policy Diffusion Research.

Huntington, S. (1991). The Third Wave: Democratization in the Late Twentieth Century. Norman, Oklahoma, University of Oklahoma Press.

Kim, H. (2007). Why and When Do States Use Human Rights Trials and Truth Commissions after Transition? An Event History Analysis of 100 Countries Covering 1974-2004."

Mendeloff, D. (2004). "Truth-Seeking, Truth-Telling, and Postconflict Peacebuilding: Curb the Enthusiasm?" International Studies Review 6(3): 355-80.

Risse, T., S. C. Ropp, et al., Eds. (1999). The Power of Human Rights. International Norms and Domestic Change. Cambridge, Cambridge University Press.

Slaughter, A.-M. (2005). "A New World Order."

Waltz, S. (2001). "Universalizing Human Rights: The Role of Small States in the Construction of the Universal Declaration of Human Rights." Human Rights Quarterly 23: 44-72.