Peace Science Made Accessible, Understandable, and Useful.

When Do Nonviolent Uprisings Prompt Mediation?

When Do Nonviolent Uprisings Prompt Mediation?

Photo credit: USAID in Africa

This analysis summarizes and reflects on the following research: Svensson, I. & Lundgren, M. (2018). From revolution to resolution: Exploring third-party mediation in nonviolent uprisings. Peace & Change, 43(3), 271-291.

Talking Points

  • Mediation has been attempted in 19% of conflicts involving nonviolent uprisings from 1970 to 2014.
  • Mediation appears to be more likely in cases where nonviolent uprisings display greater costs and risks, especially of violent escalation.
  • Nonviolent uprisings with radical flanks have a higher mediation rate—35%, as opposed to 14% for those uprisings without radical flanks—suggesting that movements with radical flanks may create greater incentives for mediation due to the “greater risks of negative externalities.”
  • The more extreme the level of state repression against the nonviolent uprising, the more likely attempted mediation is, with 27% of nonviolent uprisings facing extreme state repression being mediated (compared to 8% of those that do not face state repression).


With all the recent scholarly attention devoted to nonviolent resistance, surprisingly little has been dedicated to studying the relationship between nonviolent uprisings and mediation attempts to resolve the conflicts they bring to the surface. The authors here note that this gap in scholarship may be due, in part, to the compartmentalized research fields of nonviolent resistance and conflict resolution/management. While the former is concerned with (nonviolent) conflict escalation and identifying the conditions necessary for nonviolent movements to succeed, the latter is concerned with conflict de-escalation and identifying the conditions necessary for both sides (typically armed) to stop their violence and reach a negotiated settlement. But, as other thinkers (such as King, Curle, Lederach, and Dudouet) have noted, nonviolent uprisings are themselves part of a broader conflict transformation process, where negotiation to establish a new socio-political order is actually a principal objective—once an unequal distribution of power has been rectified through nonviolent struggle.

The authors of the present research, therefore, wish to bring together these research fields to frame their investigation into the question, “Under what conditions does third-party mediation occur in nonviolent uprisings?” Focusing on the time period 1970-2014, the authors compare two datasets on nonviolent campaigns to a dataset on mediation attempts in nonviolent uprisings to explore the factors influencing when mediation occurs in response to nonviolent uprisings.

As there are numerous reasons why governments or opposition groups may not wish to engage in mediation (including, for the government, the appearance of their loss of control or a gain in legitimacy for the opposition and, for the opposition, a loss of “revolutionary momentum” and a risk of settling the conflict before an adequate shift in power has occurred), existing mediation scholarship suggests that parties will only support a mediation effort if the conflict has become sufficiently costly for them. Likewise, mediators themselves are more likely to intervene when a conflict seems likely to “create negative externalities, including the risks of escalation to civil war and of wider instability and chaos.” The authors hypothesize, therefore, that the incidence of attempted mediation in response to nonviolent uprisings will be higher in situations that have become costly for the parties and wherethe risk of (violent) escalation is higher.

When the authors look at the data, they note, first, that the number of nonviolent campaigns per year worldwide varies considerably across the 1970-2014 time period, starting low, peaking around the end of the Cold War, and then decreasing again until another peak in the 2010s (during the Arab Spring). The incidence of mediated nonviolent uprisings varies similarly, revealing a somewhat steady mediation rate—19% of nonviolent uprisings—across this time period. This fairly low mediation rate suggests either that conflict parties often see mediation as too disadvantageous or that “nonviolent campaigns do not generate sufficiently high risks to incentivize mediation.” To explore those cases where mediation is attempted, however, the authors examine a few variables of interest: “geography, characteristics of the uprising, and the origins of the mediator.”

Of these, the most theoretically interesting is the second. The authors consider characteristics of nonviolent uprisings—including the presence of “radical flanks” in a movement and the severity of state repression—that may lead uprisings to be at greater risk for violent escalation (hence incentivizing mediation). Although they define “radical flanks” as “actors characterized by more extreme demands and methods relative to other protest groups,” they do not spell out more clearly how a demand or a method counts as “extreme” enough to constitute an actor as part of a radical flank. (In addition, the authors depart from Chenoweth and Schock who instead look at “violent flanks,” focusing solely on the means employed, as most of the nonviolent campaigns examined already have radical demands.) Whomever they are including in this category, the authors find that nonviolent uprisings with radical flanks have a higher mediation rate—35%, as opposed to 14% for those uprisings without radical flanks—suggesting that these may create greater incentives for mediation due to the “greater risks of negative externalities.” Likewise, they find that the more extreme the level of state repression against the nonviolent uprising, the more likely attempted mediation is, again suggesting that those cases with a higher “risk of escalation and widespread harm” are more likely to be mediated.

Overall, therefore, the authors find tentative support for their hypothesis that mediation is more likely in cases where nonviolent uprisings display greater costs and risks. In the end, they point out that mediation can be an important “conflict prevention” tool for ensuring that nonviolent uprisings do not escalate into violence and for facilitating a constructive conflict trajectory. The difficulty is, of course, preventing possible violence while not intervening with mediation so soon that the status quo is simply reinstated, in cases where the nonviolent movement has not yet had ample time to rectify the inequalities it was created to address.

Contemporary Relevance

This research raises some important and provocative questions about the relationship between nonviolent resistance, violent escalation, and mediation efforts. First, it draws attention to the fact that, despite the proven track record of nonviolent resistance to create change (where violent resistance often fails) and to fundamentally transform the status quo, violence is still better able to spark certain types of response—in this case, attempts at mediation—than nonviolence is, simply because it has the reputation of being more “disruptive,” “risky,” or “threatening.” We’ve all heard the saying, “If it bleeds, it leads,” referring to news coverage of conflicts worldwide. What is going on here is a similar dynamic: potential mediators tend to pay attention and see it as worth their while to intervene when a conflict looks like it is at risk for violent escalation. Of course, there may be good reasons for this: people are physically harmed when violence is employed by either side (or both sides) in a conflict, and this fact by itself may merit greater attempts at intervention (mediation, in this case) to stop the violence one way or the other. But something else is also going on when members of the regime being challenged do not initially take the threat of a nonviolent movement as seriously as they would the threat of a violent rebellion—and therefore don’t think their situation is costly enough to warrant accepting mediation. In short, deeply engrained assumptions about violence and nonviolence persist, despite recent research indicating that regimes have more to fear (in terms of holding onto power) from nonviolent movements than from violent ones.

Second, the research makes us ask ourselves whether mediation is, in fact, a desirable outcome in nonviolent uprisings. On the one hand, as the authors note, it may help prevent an escalation to violence if initiated at the right time (and if parties respond favorably). On the other, premature mediation efforts risk preempting gains for justice that might have been made if the nonviolent movement had been given more time to build power and withdraw power from the opponent regime.

These questions are brought into relief by recent events in Nicaragua, where an anti-government protest movement has emerged and grown since mid-April of this year. Although initially formed in opposition to cuts in social security, the movement shifted focus to attempting to oust President Daniel Ortega and institute broader democratic reforms once the regime started violently repressing the largely nonviolent movement. The Nicaraguan church began facilitating a dialogue process between the regime and the protest movement, but as the death toll has mounted—about 300 people had been killed by the end of July—church leaders have found it increasingly difficult to remain neutral, vacillating between their role as mediators and their need to stand with those being victimized by the government. So, while the Nicaraguan case illustrates the tendency of mediators (in this case domestic mediators) to step in when a conflict is escalating to violence, it also underscores an additional finding not noted in the research: that while mediators in a civil war context may be able to remain impartial, as both sides are responsible for comparable levels of violence, mediators in a conflict where one side is engaged in nonviolent resistance and the other in violent repression may find it more difficult to remain similarly impartial. In the case of the Nicaraguan priests, they can affirm their roles as both mediators and “protectors” of movement activists (see Malkin and Robles article under Continued Reading), but the government clearly no longer sees them as impartial once they have taken on the latter role, which has implications for the success of mediation efforts.

Furthermore, we can ask whether it would be beneficial for a mediation process to proceed at this point, when the nonviolent resistance movement may still have gains to make in terms of bringing other segments of Nicaraguan society to its side and putting greater pressure on the regime for significant changes. That said, such shifts in support and power become less likely the more violent flanks emerge in the movement in reaction to government repression, as illustrated by the use of homemade weapons in the city of Masaya to defend neighborhoods from government incursions. Although the violence in such instances is clearly asymmetrical, its use by anti-regime activists—though understandable in the face of government repression—makes it harder for those who support the Ortega regime to change sides and ultimately shift the balance of power—and may even further entrench support for the regime, as it re-affirms Ortega’s representations of these activists as “terrorists” against whom the country must be defended.

Practical Implications

There are implications here for both mediators and journalists. For local and international mediators, the challenge is to make one’s services available without intervening too soon (before a severe power imbalance has been rectified), lest one unwittingly serve the status quo. Perhaps nonviolent movements can simply be trusted to refuse mediation if its offer comes too early, however, so what matters is that mediators be sensitive to these dynamics when they communicate with the various parties, not that they hesitate to offer their services at all. For journalists, the key is to report just as attentively on—and take just as seriously—nonviolent uprisings as they would violent rebellions, avoiding outdated tropes in their writing about how nonviolence merely operates via the persuasion of the adversary and instead emphasizing the shifts in power that are occurring in a society facing nonviolent resistance. Furthermore, journalists should be careful not to speak in general terms about “protests turning violent” but rather should specify to whom the violence is attributed—violent flanks of the movement or government forces—as this detail matters very much in terms of how such conflicts are perceived on the world stage and how they precipitate shifts in power. These moves will help parties and observers gain a more accurate understanding of what is going on and make it more likely that mediation—when it does emerge in response to nonviolent uprisings—better serves the dual purposes of justice and peace.

Continued Reading

Nonviolent Resistance and Conflict Transformation in Power Asymmetries By Véronique Dudouet. Berghof Research Center for Constructive Conflict Management, September 2008.

Nicaragua Protests Grow Increasingly Violent, 100 Killed Since April By Frances Robles. The New York Times, May 31, 2018.

Inside Nicaragua’s Protest Movement By Brent McDonald, Neil Collier, and Ben Laffin. The New York Times, July 22, 2018.

Nicaragua Clergy, Siding with Protesters, Becomes ‘Terrible Enemy’ of Ortega By Elisabeth Malkin and Frances Robles. The New York Times, July 22, 2018.

Shoot to Kill: Nicaragua’s Strategy to Repress Protest By Amnesty International. May 29, 2018.

Keywords: nonviolent/civil resistance, mediation, conflict resolution, violent repression, violent flanks


Next article Social Capital, the Reintegration of Ex-Combatants, and Peacebuilding in Post-War Liberia
Previous article When Countries Import More Weapons, Are They More Likely to Go to War?