[Note:  This is a guest post by Geoffrey Dancy, Assistant Professor of Political Science at Tulane University]

Nearly every civil war negotiation or democratic revolution is now accompanied by a consideration of how to publicly address previous human rights abuses—what practitioners refer to as transitional justice.

Over the last week, Juan Manuel Santos was narrowly reelected president in Colombia on a peace platform. His government must now move forward with a fourth round of negotiations with FARC rebels. Having already tackled land reform, political rights, and the drug trade, this round will involve discussion over the “transitional justice framework,” and must resolve a series of thorny issues like victims’ rights to reparation. Most importantly, the government and rebels will have to address the controversy over whether individual combatants will receive prison time for the many human rights violations they committed during the 50-year-long war. The issue of justice is especially salient following the recent ‘false positives’ scandal–where it was discovered that government security forces over the last decade rounded up and murdered thousands of young men from slums, dressed them as guerillas, and presented their kills to authorities for reward.

Colombia is not an isolated case.  The UN recently applauded the launch of the Truth and Dignity Commission in Tunisia, an amnesty and reconciliation law is at the top of the agenda for the newly elected government in post-coup Papua Guinea-Bissau, and the Central African Republic’s President has formally requested an International Criminal Court investigation into the inter-communal violence that has plagued the country since 2012.

But what do we know about the effectiveness of various transitional justice policies—including trials, truth commissions, and amnesties—in such situations? Based on data collected over the last three decades, a group called the Transitional Justice Research Collaborative (TJRC) has begun to produce some generalizations about which mechanisms  work in these difficult contexts. Here are three of the TJRC’s most important initial findings:

1.     Human rights prosecutions, typically associated with transitional demands for justice, help decrease abuses to physical integrity rights in the long term. One of the challenges with studying the impact of trials for state agents is that of selection: it would seem that any country willing to try rights abusers is already less likely to use repressive violence. However, the TJRC makes three points on this issue. First, prosecutions are not an easy sell: they typically happen following years of untiring efforts by activists to pressure the courts, not because state leaders decide on a whim to have them. Second, when they do take place, they create legal openings that allow for more litigation in the future. And third, this leads to information updating on the part of would-be human rights abusers. The more criminal trials they observe, like the recent proceedings against Guatemalan General Efrain Rios Montt, the less likely they will be to practice torture, indiscriminate killing, imprisonment, or disappearances. These findings hold even when accounting for sources of selection bias like regional culture, state institutional strength, or countries’ levels of democracy.

2.     Amnesties for human rights abuses, typically associated with forgiveness or reconciliation, do not bolster physical integrity protections. Instead, they help fortify protections of civil liberties like the right to vote, form parties, and protest. Why? Countries trying to strengthen basic democratic institutions after a period of autocracy or conflict typically face massive credible commitment problems. Parties are often outlawed, and certain groups are unreasonably targeted for attack. Amnesties are a tool to signal to opposition groups that the government intends on pursuing a kind of politics of mercy, rather than seeking reprisals against their enemies. This seems to create an uneasy relationship for those pursuing  prosecutions. However, the TJRC finds that a choice need not be made between the two. Prosecutions and amnesties can co-exist if designed properly.

3.     Finally, truth commissions do not appear to have much of a lasting impact, even though the calls for truth have been getting louder over the last two decades (the group puts the count of TCs at over 70). One of the reasons that truth commissions do not appear generally to change human rights practices for the better is that there is a combination of anemic and robust truth-telling institutions that have been established (they are bimodal). Some truth commissions, like Liberia’s TRC, do quite a bit to address grievances, whereas others, like Sri Lanka’s recent Lessons Learnt and Reconciliation Commission, are cheap attempts to curry favor with an international audience.

The Santos administration in Colombia has just received a slim mandate to continue peace talks and produce results for victims of a long and awful war. The initial findings from the TJRC suggest that pursuing transitional justice could be beneficial for the country in the long run.  But Colombia and other countries emerging from periods of human rights violations need to realize that just doing something is not enough.  The mechanisms they choose to pursue need to be implemented competently and paired well with each other to achieve the desired goals of stronger democracies and human rights protections.