Does Transitional Justice Work? Latin America in Comparative Perspective


Despite the recent proliferation of transitional justice practices and scholarship around the world, we know very little about whether and how it achieves its goals of strengthening democracy and reducing human rights violations.  Findings from the Transitional Justice Data Base (TJDB) fill that gap in knowledge.  We confirm assumptions that transitional justice contributes to improvements in democracy and human rights scores.  We challenge existing assumptions, however, that single mechanisms – i.e., human rights trials, truth commissions, or amnesties – achieve those improvements.  Instead, we find that these mechanisms only work to improve democracy and human rights in specific combinations:  trials and amnesties; or trials, amnesties, and truth commissions.  Combining trials and amnesties to achieve political objectives seems counter-intuitive since trials should undermine amnesties and vice versa.  Our analysis leads us to an explanation for these puzzling findings that we term the justice balance.  Specifically, we contend that in certain contexts particular types of amnesties and trials coexist and have a positive impact.  In other contexts, trials follow amnesties and the sequencing of the mechanisms contributes to their positive impact on human rights and democracy.1

In this paper, we present those findings.  We first examine the existing assumptions about transitional justice and its impact on democracy and human rights.  Next, we briefly describe the TJDB, and summarize our statistical findings.  In the fourth section, we use the Latin American cases to illustrate our findings.  The paper concludes with a discussion of the policy implications of our results.

From an abundant literature on transitional justice, four principal claims about its effect on democracy and human rights emerge.2 A maximalist approach advocates a maximal form of accountability – human rights trials – to meet moral, legal, and political obligations to address past wrongs, lay the foundation for the rule of law, deter future human rights violations, and strengthen democracy.  The approach warns that failing to try perpetrators will lead to cycles of retributive violence or vigilante justice.3 International law and treaties, moreover, imposes a duty on countries to prosecute perpetrators of past atrocities.4 When domestic courts fail to do so, regional and international courts sometimes intervene.  Under the principle of universal jurisdiction foreign courts have attempted to hold perpetrators accountable where their own domestic courts have failed to do so.  In sharp contrast, a minimalist approach endorses a minimal form of accountability – amnesty – to promote a stable transition.  Trials, this approach contends, jeopardize democracy by catalyzing spoilers, while amnesty neutralizes these actors, paving the way to stronger democracies and human rights protections.5 A moderate approach promotes truth commissions as a way to moderate the demand for accountability in the vulnerable transitional moment, while avoiding morally and legally questionable impunity measures.  Truth commissions, the approach contends, acknowledge the past and hold perpetrators accountable, but focus on restorative over retributive justice, presenting less of a threat to spoilers.6 A holistic approach holds that no single mechanism can address the many problems societies face after such atrocities.  It argues, instead, that the use of multiple mechanisms will advance human rights and democracy objectives more effectively than a single one.7

Our analysis challenges the maximalist, minimalist, and moderate approaches’ focus on single mechanisms.  While most closely resembling a holistic approach, we contend that only specific combinations of mechanisms advance the democracy and human rights objective.

The TJDB contains data on a variety of transitional justice mechanisms, coded from Keesing’s World News Archive, implemented worldwide during a forty-year period.  To test the hypotheses in the existing literature we examine the impact of three types of mechanisms alone and in combination – trials, truth commissions, and amnesty – in 91 transitions from authoritarian rule from 1970 to 2004.  We measure the outcomes of democracy and human rights using established data.  The Freedom House Project measures change in democracy differentiating between political rights (democratic participation) and civil liberties (institutional democracy).  One limitation of this data source for the proposed project is that it begins in 1980.  In addition, scholars frequently criticize the Freedom House project.8 Despite these limitations, it remains widely used and provides a particularly good measure of the specific aspects of democracy (rights and freedoms) that transitional justice is most likely to affect.  The Political Terror Scale develops five-levels of political terror based on Amnesty International (AI) and U.S. State Department (USSD) reports on human rights practices.  PTS begins in 1976, which limits the observations we include in our analysis, although not to the same degree as other human rights measures.

Using maximum likelihood estimation techniques, we analyzed the data set to determine whether general patterns between mechanisms and outcomes held up to multivariate analysis.  Because the countries included in the sample are not selected randomly, we also took specific measures to control for selection bias.9

Using the TJDB, we find that none of the transitional justice mechanisms reduce human rights violations or improve democracy on their own.  Trials and amnesties do not prove statistically significant for those political objectives.  Truth commissions are statistically significant, but negative. Put differently, using them alone would harm, rather than improve, human rights and democracy scores.  The single-mechanism approaches advocated by maximalists, minimalists, and moderates simply do not hold up to empirical testing.  The holistic approach moves in the right direction, but fails to specify the only two combinations that improve democracy and human rights:  (1) trials and amnesties and (2) trials, amnesties and truth commissions (see Table 1).10

Table 1: Effect of Transitional Justice on Human Rights and Democracy


What explains the effectiveness of these two combinations of mechanisms?  The existing literature coupled with our own analysis of the TJDB offers an explanation to this puzzle.  The transitional context allows for particular combinations of trials and amnesties.  When they co-exist, they create a balance that is conducive to improvements in democracy and human rights.  Trials, as maximalists assert, provide the accountability necessary for strengthening democracy and deterring human rights violations.  Amnesties, as minimalists aver, provide the security from spoilers to protect against authoritarian reversals.  The two mechanisms work in tandem to advance democracy and human rights.

Given that these two mechanisms would appear to undermine each other, we consider two possible transition scenarios in which they can co-exist:  transitions resulting from collapse of the authoritarian regime and negotiated transitions (see Figure 1).

Figure 1.  The Justice Balance


When authoritarian regimes collapse, the new democracy that follows faces fewer constraints on prosecuting perpetrators.  The former regime’s loss of legitimacy leaves few supporters to defend it and its actions.  Those who do, lack sufficient political power to pose a credible threat to democratic rule.  While trials alone could effectively address past violations, our findings show that they are not enough to improve democracy and human rights.  Why are amnesties also necessary for those improvements?

We suggest that amnesty is not politically necessary in the regime-collapse scenario, but it is pragmatic.  Trying everyone is simply not practical.  Attempting to do so will exhaust legal and financial resources, neglecting other projects important to democratization.  Instead of putting all perpetrators on trial, therefore, new democracies strategically limit the number or scope of trials.  By selecting perpetrators or crimes for prosecution, and amnestying the rest, democracies still comply with domestic demand for, and international norms regarding, accountability.  By limiting those trials, they avoid exhausting their economies in retrospective justice projects.11 In sum, in cases of regime collapse, the justice balance advocates complementary processes.

Negotiated transitions offer a different explanation.  The authoritarian regime retains sufficient power in this scenario to influence and even threaten the new democracy.  Trials imply risks for the new government.  Improvements in democracy and human rights depend in this case on initial amnesties to secure the transition.  Once the new democracy develops its institutions and establishes security against former perpetrators, it can begin trials.  Delayed justice offers new democracies the chance to balance accountability with a practical need – security – for amnesty.  Truth commissions might be used for restorative justice purposes at various points in this scenario without jeopardizing the transitional process, though our analysis does not address this point directly.

Latin America illustrates the positive impact of combining trials with amnesties for improved human rights and democracy scores.  Of the 19 transitions in the region, eight included human rights trials.  In seven of those cases, countries balanced human rights trials with amnesties or amnesties and truth commissions.  In those cases of justice balance, the results for democracy and human rights improvements are much higher than in those countries that did not combine amnesties and trials (with or without truth commissions).  Table 2 below illustrates that those countries at the top, which used both trials and amnesties, have stronger improvements in human rights and democracy than those at the bottom that did not follow the justice balance.

Table 2.  Difference in score from year prior to transition and ten years post-transition


These findings suggest that the zero-sum approach to transitional justice in the existing literature misses an important interaction between trials and amnesties that allows for improved human rights and democracy outcomes.  This interaction, moreover, has important policy implications.  First, new democracies and non-governmental organizations advocating transitional justice should avoid promoting any single mechanism as a pathway to democracy and human rights.  The assumptions that trials or amnesties are too dangerous for human rights and democracy, or that truth commissions provide a successful pathway, simply do not hold up to empirical testing.  Holistic approaches offer a more compelling argument, but only when trials and amnesties are balanced, with or without truth commissions.

Second, democracies do not have to choose between trials and amnesties.  Together they offer a pathway to improved democracy and human rights.  Even those governments saddled with self- or blanket amnesties passed by former authoritarian regimes or their vulnerable successors, may initiate trials.  Indeed, delayed justice allows for the justice balance that improves human rights and democracy.

Third, amnesties do not weaken democracy and human rights.  Indeed, as long as partial amnesties allow for some human rights trials, or mobilized human rights or legal communities find ways to circumvent or overthrow amnesty laws, these amnesties can help to strengthen human rights and democracy.

Tricia D. Olsen is doctoral student at the University of Wisconsin-Madison. Leigh A. Payne is Professor of Sociology and Latin American Studies at the University of Oxford, Professor of Political Science at the University of Wisconsin-Madison, and Visiting Professor of Political Science and Global Studies at the University of Minnesota. Andreas G. Reiter is doctoral student at the University of Wisconsin-Madison.


  1. Olsen, Tricia D., Leigh A. Payne, and Andrew G. Reiter, Engaging the Past to Safeguard the Future: Transitional Justice in Comparative Perspective, unpublished book manuscript. []
  2. The Transitional Justice Bibliography finds over 2,000 scholarly works on transitional justice.  See []
  3. Diane F. Orentlicher, “Settling Accounts: The Duty to Prosecute Human Rights Violations of a Prior Regime,” Yale Law Journal 100, no. 8 (1991); Naomi Roht-Arriaza, “State Responsibility to Investigate and Prosecute Grave Human Rights Violations in International Law,” California Law Review 78, no. 2 (1990). []
  4. Ibid.  See also M. Cherif Bassiouni, “International Crimes: Jus Cogens and Obligatio Erga Omnes,” Law and Contemporary Problems 59, no. 4 (1996); Michael P. Scharf, “The Letter of the Law: The Scope of the International Legal Obligation to Prosecute Human Rights Crimes,” Law and Contemporary Problems 59, no. 4 (1996). []
  5. Jack Snyder and Leslie Vinjamuri, “Trials and Errors: Principle and Pragmatism in Strategies of International Justice,” International Security 28, no. 3 (2003); Helena Cobban, Amnesty After Atrocity? Healing Nations After Genocide and War Crimes (Boulder, CO: Paradigm Publishers, 2007). []
  6. José Zalaquett, “Confronting Human Rights Violations Committed by Former Governments: Principles Applicable and Political Constraints,” Hamline Law Review 13, no. 3 (1990). []
  7. ICTJ, “What is Transitional Justice?” []
  8. Munck and Verkuilen 2002. []
  9. In the analysis of how transitional justice mechanisms influence democracy and human rights, a transition was the unit of analysis because we were interested in the final outcome of transitional justice choices.  As such, we included regional variables and also clustered standard errors by country to ensure our findings were robust.  In other analyses – especially those that seek to understand the conditions under which country leaders are most likely to adopt specific transitional justice mechanisms – we were interested in understanding the nuances over time influenced mechanism adoption; thus a country-year was the unit of analysis. For these analyses, we included a propensity score in the models and again, included regional control variables and clustered by country.  See our book manuscript (Engaging the Past to Safeguard the Future:  Transitional Justice in Comparative Perspective) for an in-depth discussion of the quantitative methods used. []
  10. This table represents a summary of the findings found in the “Does Transitional Justice Work?” chapter in our book manuscript Engaging the Past to Safeguard the Future:  Transitional Justice in Comparative Perspective. []
  11. For a more detailed discussion of the political economy of transitional justice see the “At What Cost?” chapter in our book manuscript Engaging the Past to Safeguard the Future:  Transitional Justice in Comparative Perspective. []


Print This Post Print This Post

This entry was posted on Tuesday, December 15th, 2009 at 2:36 pm and is filed under Americas, Human Rights, International Law, Justice, Society, Transitional Justice, Violence. You can follow any responses to this entry through the RSS 2.0 feed. You can skip to the end and leave a response. Pinging is currently not allowed.


One Response to “Does Transitional Justice Work? Latin America in Comparative Perspective”

  1. Mr. King Says:

    Strengthening democracy… Perhaps we should start asking questions about the big picture…

    Do we need a Referendum For A New Democracy?

    Are you concerned about the future of democracy? Do you feel democracy is under attack by extreme greed in countries around the world? Are you sick and tired of: living in fear, corporate greed, growing police state, government for the rich, working more but having less?

    Can we use both elections and random selection (in the way we select government officials) to rid democracy of undue influence by extreme wealth and wealth-dominated mass media campaigns?

    The world’s first democracy (Athenian democracy, 600 B.C.) used both elections and random selection. Even Aristotle (the cofounder of Western thought) promoted the use of random selection as the best way to protect democracy. The idea of randomly selecting (after screening) juries remains from Athenian democracy, but not randomly selecting (after screening) government officials. Why is it used only for individual justice and not also for social justice? Who wins from that? …the extremely wealthy?

    What is the best way to combine elections and random selection to protect democracy in today’s world? Can we use elections as the way to screen candidates, and random selection as the way to do the final selection? Who wins from that? …the people?

Leave a Reply