Transitions to and from civil war

Civil wars are highly complex processes involving a myriad of actors whose interactions shape transitions to and from war in non-linear ways

Anastasia Shesterinina

This piece is re-posted from APSA Comparative Politics Newsletter; the original publication can be read here.

Civil wars are highly complex processes involving a myriad of actors, including states, nonstate armed groups, local populations, and external forces; their interactions with each other and the structural conditions they face; and profound changes in these interactions over time.

The literature on civil war has examined the main stages of onset, duration, and termination of war, with a focus on the recurrence of war in the aftermath of armed conflict. More recent turns in the literature to the micro-foundations of individual and group behavior, organizational structure, and enduring legacies of civil war, have enriched our understanding of mobilization, patterns of violence, rebel governance, armed group resilience, and social transformations as a result of civil war, among other dynamics.

Relatively less attention in civil war studies has been paid to transitions to and from civil war, which are commonly associated with the fields of social movements and conflict resolution, respectively, with attempts to bridge the stages of conflict and sub-field divides seen as one of the promising directions in future research on civil war (Cederman and Vogt 2017).

How do conflicts turn violent? How do civil wars unfold over time? How do distinct dynamics of civil war affect the post-war potential for peace? These are some of the driving questions in this research programme where I situate my work. In this essay, I discuss how micro-level analysis of individual trajectories of participants in these processes from the pre- to the post-war stages of conflict can help further advance our understanding of these questions.

I focus on transitions from pre-war conflict to civil war – drawing on my research on the Georgian-Abkhaz war of 1992-1993 – and on transitions from civil war to post-war conflict – drawing on my research on Colombia since the signing of the Final Agreement for Ending the Conflict and Building a Stable and Lasting Peace by the National Government and the Revolutionary Armed Forces of Colombia—People’s Army (FARC-EP) in 2016. While these kinds of transitions have often been characterized as part of escalation and de-escalation of conflict, I find that transitions to and from civil war are more complex and non-linear.

Civil wars emerge from different pre-war interactions between the actors involved, and these interactions do not necessarily escalate in nature, intensity, and scale in advance of the fighting. Conflicts transform – rather than simply de-escalate – in the postwar period, where existing and new patterns of violence that result from changing conditions coexist with attempts at peacebuilding. Careful attention to participants’ lived experiences of these processes reveals how actors’ identities evolve through these processes, to shape their relations in the course of conflict.

This fieldwork-intensive research highlights the potential of examining multiple paths that civil wars follow for our grasp of transitions from pre- to post-war, which my new Civil War Paths project, “Understanding Civil War from Pre- to Post-War Stages: A Comparative Approach,” funded by a £1.2m UK Research and Innovation Future Leaders Fellowship, will turn to over the next years.

To better understand these transitions, we will trace evolving interactions between social groups, their organization and participation in conflict, and the ways in which these experiences shape their identities. Placing actors’ interactions at the center of analysis will help us identify the dynamics, from pre-war contention and non-state armed group formation to state fragmentation and external intervention, that lead to civil wars in different ways, with implications for how the fighting unfolds and how conflicts transform thereafter. In this effort, fieldwork with participants will show how the very actors involved perceive one another, interpret their relations, and adapt to changing circumstances in the course of conflict. Comparing distinct paths across cases will point to a range of dynamics beyond escalation and de-escalation that scholars of civil war should focus on in the future.

From pre-war conflict to civil war

Transitions from non-violent to violent conflict have been most extensively studied by social movement scholars in the context of contentious politics (McAdam, Tarrow and Tilly 2001). Drawing on this tradition and innovations in data, such as the Nonviolent and Violent Campaigns and Outcomes (NAVCO) dataset, recent contributions have highlighted the effectiveness of non-violent resistance, which can prevent violent insurgency (Chenoweth and Stephan 2011). Scholars have used this and other data to explain why some non-violent protests turn to violence (Gustafson 2020), including civil war (Ryckman 2020).

Civil war studies have taken inspiration from this tradition to explore the ways in which state repression and dissent interact in the lead up to civil war (Lichbach et al. 2004; Sambanis and Zinn 2005; Young 2013) and have identified different paths to civil war based on this interaction (Davenport et al. 2006).

This literature has shifted attention beyond structural variables associated with civil war onset and beyond the period of civil war, to focus on escalation from non-violent contention and violence short of war to civil war. Comparative evidence, however, has challenged the linear escalation narrative (Lawrence 2010). Civil war is not always an outgrowth of non-violent and violent conflict, and the dynamics of pre-war conflict can affect civil war in other ways – for example, by shaping actors’ identities.

As I demonstrate in my book, Mobilizing in Uncertainty: Collective Identities and War in Abkhazia, decades of participation in and observation of everyday confrontation, political contention, and violent opposition shaped and reshaped what I call “collective conflict identities,” or shared understandings of conflict and one’s role in it, before the Georgian-Abkhaz war of 1992-1993. These repertoires were widespread in Soviet Abkhazia, yet the war broke out abruptly, following a period of relative calm when few violent events took place after the inter-group clashes of 1989.

The pre-war non-violent and violent conflict, therefore, did not “escalate” to civil war, but was important for other reasons. The history of the conflict provided a shared framework for understanding the Georgian advance into Abkhazia in August, 1992, and situated potential participants in the war in relation to one another and the broader group as they were making their mobilization decisions. These aspects of collective conflict identities influenced how people perceived the threat of the Georgian advance and whom they mobilized to protect as a result, with implications for how the Georgian-Abkhaz war unfolded.

To arrive at these conclusions, I traced nearly 150 individual trajectories from pre- to post-war, collected through life history interviews during eight months of immersive fieldwork in the area and contextualized with additional primary and secondary materials. Here I highlight three micro-level transitions that participants’ lived experiences of the conflict helped me parse out. Whereas the escalation framework would suggest inter-group antagonism increasing over time, a subtler picture of inter-group relations emerged from my interviews.

In the context where Abkhaz, Georgian, and other groups were highly integrated in familial, neighborhood, and organizational settings, everyday life in Soviet Abkhazia was characterized by constant inter-group interaction. People grew up, studied, worked, celebrated, and grieved together, but the underlying conflict structured their relationships. The collective historical memory of socalled Georgianization offered the Abkhaz a set of shared understandings of the conflict as one aimed at the dissolution of the Abkhaz identity in the dominant Georgian mass.

Political issues were considered conversational taboos and, once raised, led to arguments and even brawls among relatives, friends, acquaintances, and strangers. Inter-group tension was normalized and was also felt in jokes, rumors, insults over group belonging, and bending of customs of neighborly solidarity. Systematic experiences of everyday confrontation cemented Abkhaz views of the conflict, shaped how people interpreted and acted on their regular social roles through the lens of the conflict, and transformed relationships.

Refraining from sensitive subjects was one conflict avoidance strategy that people adopted to maintain their familial and friendship ties. However, this strategy drew people apart as they could not address the issues that concerned them with each other.

Conflict avoidance became particularly difficult as parallel repertoires of political contention and violent opposition deepened societal polarization and generated a new process of militarization after the first violent clashes of 1989.

The Abkhaz had long participated in political contention, from letter writing to Soviet authorities, to demonstrations, and strikes, paving the way for the formation of the Abkhaz national movement in the 1980s, during the period of political opening in the Soviet Union. It is in this context that the first Georgian-Abkhaz clashes took place in Abkhazia. Yet the framework of escalation struggles to capture the micro-level dynamics of the clashes.

While the Abkhaz gathering to demand the restoration of Abkhazia’s political status preceded the clashes, the trigger of violence was a different, albeit related, issue of the opening of a Sukhum/i branch of the Tbilisi State University, which was perceived as further threatening the Abkhaz identity by dividing the Abkhaz State University and creating greater educational opportunities for non-Abkhaz students. As Abkhaz activists attempted to prevent entry exams, the clashes started spontaneously, with an attack on an Abkhaz photojournalist recording a Georgian protest elsewhere, and spread across Abkhazia.

The experience of inter-group violence split formerly integrated teams in employment, education, and government institutions and armed groups were formed on both sides in the conflict. Inter-group relations transformed from underlying but largely contained tension to open division as a result. Despite the presence of armed actors, no large-scale violence took place thereafter.

The Abkhaz Guard formed after the clashes had skirmishes with the Georgian paramilitary group, the Mkhedrioni, that now had a branch in Abkhazia and obstructed unwanted crossings from Georgia proper in the Georgian-Abkhaz border area. However, most guards were released from duty shortly before the Georgian advance into Abkhazia in August, 1992. The advance, therefore, came as a surprise after years of relative calm in Abkhazia – contrary to the escalation argument. The preceding conflict shaped how people made sense of the advance and positioned them in relation to one another in powerful ways as they decided whether and how to mobilize in response. In the uncertainty created by the sudden entry of Georgian forces, the events could have been interpreted as a potential clash similar to that of 1989, policing of the frequently looted railroad, or pursuit of supporters of Georgia’s ousted president who were ostensibly hiding in Abkhazia. Yet, these alternative framings did not resonate with Abkhaz experiences of the conflict.

Instead, the Abkhaz leaders’ framing of the advance as a threat corresponded to these experiences and was adapted to the needs of local defense and consolidated into mobilization decisions within small groups of relatives and friends, to direct protection to those segments of society that people perceived to be threatened. This process resulted in a war that lasted over a year and displaced most of the Georgian population from Abkhazia, with protracted violence and no political resolution over this de facto state since the Abkhaz victory in the war in 1993.

From civil war to post-war conflict

So far, I have shown that the escalation argument falls short in explaining some transitions from pre-war conflict to civil war and that paying attention to participants’ experiences of conflict can help uncover more complex and non-linear paths to civil war. Here I turn to transitions from civil war, which are commonly associated with de-escalation.

Scholars of conflict resolution have charted a path of de-escalation from war to ceasefire, peace agreements, normalization, and reconciliation, being mindful of setbacks and relapses that can occur along the way (Ramsbotham et al. 2011, 13), and have emphasized that conditions must be “ripe” for negotiations to succeed, pointing to stalemate as a situation when parties might be willing to talk (Zartman 1995). Recent studies have demonstrated the importance of mediation and peacekeeping for establishing mutually acceptable terms of peace agreements and overcoming commitment problems (Walter 2002), thereby contributing to the reduction of violence (Hultman et al. 2014; Beardsley et al. 2019). Overall, de-escalation has been found to significantly affect the potential for peace in post-war societies by reducing the chances of war recurring in the future (Hegre et al. 2017).

However, by focusing on the reduction of violence, typically measured by arbitrary battle-death thresholds, the de-escalation approach has overlooked different ways in which conflicts continue when the fighting subsides and conflict and peace coexist and coevolve (Campbell et al. 2017).

My research on Colombia, where I have collected life histories of former FARC-EP mid-level commanders (mandos medios or “middle managers” in Spanish) along with expert interviews and secondary materials, suggests that, as in the case of post-war Abkhazia, the armed conflict in Colombia has continued in multiple ways since the signing of the 2016 peace agreement and new conflict dynamics have emerged in the process.

During the negotiations, while the armed conflict with the FARC-EP “de-escalated,” the National Liberation Army (ELN), now the largest remaining armed group in Colombia, propped up violent activities. Since then, the ELN has grown in size and extended operations, including into the territories that the FARC-EP had controlled, which highlights weak state presence in many of these territories. Moreover, illicit economies have provided ample opportunities for existing and new armed groups to operate, especially in areas associated with lucrative drug trade and illegal mining. Armed groups that have emerged from the demobilization of the FARC-EP and ex-FARC-EP members who have remobilized since the peace agreement have been part of this landscape. These actors, particularly former mid-level commanders who have continued violent activities, have been viewed as “spoilers” of peace, obstructing the de-escalation of violence (Stedman 1997). However, not all ex-mandos medios have followed this trajectory and many have instead played critical roles in the negotiations, disarmament, and peace agreement implementation (Shesterinina 2020).

Below I unpack the category of “mid-level commander” to show – following Campbell et al. (2017) – that violent and cooperative activities have coexisted within the same group of participants in this war-to-peace transition and that participants’ self-perceptions have transformed over time, to embed some within the peace process.

The category of “mid-level commander” is a grey area in the internal structure of the FARCEP, where a distinction could be made between rank-and-file combatants, commanders and deputies of military units, and general command of the organization. It is commanders and deputies of intermediary structures, from squad to front, who were in charge of most dayto-day operations of the organization and who could be considered mandos medios, connecting rank-and-file combatants to top commanders. A typical mid-level commander trajectory included at least two years of training, observing FARC-EP regulations, and performing assigned tasks, which if successful could lead to progression to mando medio roles, starting with deputy of a squad.

Those who were sanctioned for not respecting the rules of the organization rarely progressed. Once in the role, mandos medios were responsible for regulating everyday life of their unit, ensuring that their subordinates had access to basic necessities, addressing their problems, maintaining discipline, and assigning tasks based on broader strategy. They were also responsible for working with local populations under their control and securing cash flows for the organization.

As a result, mid-level commanders developed necessary skills and military, economic, and social standing within the organization and among local populations to be able to attract their subordinates and continue violent activities, even as their organization disarmed. Indeed, reports abound not only in Colombia (Zukerman Daly 2014), but also in other contexts (Themnér 2011) where mid-level commanders became “spoilers” undermining the peace process through violent activities or serving as intermediaries between the “spoiler” elite and ex-combatants.

Having occupied a central role in the organization, mid-level commanders might feel unrepresented in peace agreements negotiated by top commanders to secure their own interests and designed for rank-and-file combatants in terms of disarmament, demobilization, and reintegration (DDR) (Theidon 2016). Hence, experts have called on national and international actors to develop special DDR programmes to prevent mid-level commanders’ return to violence (FIP 2010). Yet no rank-based strategy has been developed, and a number of former FARC-EP mid-level commanders formed or joined dissident groups during and after the negotiations. While attention has focused on those former FARC-EP mid-level commanders who have abandoned the peace process, their number is relatively low. In fact, most have remained in the peace process, according to my interviews.

More importantly, this focus has masked the coexistence of violent and cooperative activities within this group of participants in the Colombian post-peace agreement transition. Surely some former FARC-EP mid-level commanders have continued violent activities, but others have played critical roles in the peace process since the negotiations in Havana. During the negotiations, mandos medios became a key communication channel, transmitting information from top commanders at the negotiations table to rank-and-file combatants.

They discussed the contents of the negotiations with their units point by point, to address doubts and fears among combatants – disarmament could make them vulnerable to retaliation from the state, other armed groups, and local populations who have suffered from FARC-EP violence; their skills could not translate to livelihoods outside of the armed struggle; and they could be betrayed in the process, including by their organization. When the peace agreement was signed in 2016, they similarly explained “the laying down of arms” – a concept that did not sit well with life in arms that combatants had led, often for decades – and managed disarmament through practices of subordination that they previously developed as commanders responsible for their units and new roles that some acquired as part of formal mechanisms established to support the process.

After these initial stages of the war-to-peace transition, many former FARC-EP mid-level commanders have used their skills and standing to organize everyday life and mobilize productive projects in collective reincorporation areas. New leaders outside of the former hierarchical structure have emerged in these settings, including through conflicts between ex-combatants, but former mid-level commanders have continued to lead some ex-combatant communities, often as elected representatives. Despite divisions within the FARC-EP political party, they have also adapted their intermediary position to support implementation of the peace agreement by representing the party and their communities in dialogue and monitoring institutions.

Finally, former mid-level commanders have led a range of peace initiatives beyond the formal peace process, repurposing their wartime experience for the advancement of peace, for example, by producing knowledge on territorial dynamics of violence and developing local protection measures through community training. As a result, these individuals’ self-perceptions have transformed, from mandos medios responsible for their units and the operation of the FARC-EP as an armed organization, to local leaders responsible for their communities. These roles – and the relationships among ex-combatants that they entail – have kept them committed to the peace process despite setbacks in peace agreement implementation, rampant violence against social leaders and FARC-EP ex-combatants, and ongoing stigmatization of former mid-level commanders due to current and historical precedents of rearmament among this group in Colombia.


This essay has considered transitions to and from civil war and has challenged linear escalation and de-escalation arguments, showing that research on participant trajectories can uncover complex processes whereby conflicts do not simply escalate or de-escalate but rather follow multiple micro-level dynamics and transform actors’ identities. In the case of Abkhazia, years of relative calm preceded the Georgian-Abkhaz war of 1992-1993, which broke out unexpectedly, prompting potential participants to draw on their experiences of inter-group conflict to make sense of the violence and decide whether and how to mobilize in the uncertainty of the events.

In the case of Colombia, the armed conflict continued past the signing of the 2016 peace agreement, exposing variation within the same groups of participants, such as former FARC-EP mid-level commanders, some of whom have used their skills and standing to continue violent activities, whereas others have repurposed their wartime experiences for leadership in the peace process. However, these are only few among a range of paths that civil wars follow from pre- to postwar.

For example, Krause (2018) convincingly demonstrates conflict escalation in communal wars in Indonesia and Nigeria, where violence escalated from pogroms to battles and joint attacks of mobile gangs and militias with participation of civilians. Similarly, Dudouet (2013) shows an alternative path of de-escalation whereby armed groups in a variety of settings shifted to non-violent tactics as a result of leadership changes, re-evaluation of goals and means, and exposure to new repertoires of action. Future research should move beyond general arguments on escalation and de-escalation to theorizing under what conditions escalation and de-escalation dynamics might hold among other paths to and from civil war.

This is a central aim of my Civil War Paths project, which will draw on fieldwork-intensive comparison of a number of cases that vary on how civil wars emerged, unfolded, and ended or transformed carried out by a team of researchers at the new Centre for the Comparative Study of Civil War to better understand diverse transitions across the stages of conflict.


Beardsley, Kyle, David E. Cunningham, and Peter B. White. 2019. “Mediation, Peacekeeping, and the Severity of Civil War.” Journal of Conflict Resolution 63(7): 1682-709.

Campbell, Susanna P., Michael G. Findley, and Kyosuke Kikuta. 2017. “An Ontology of Peace: Landscapes of Conflict and Cooperation with Application to Colombia.” International Studies Review 19(1): 92-113.

Cederman, Lars-Erik, and Manuel Vogt. 2017. “Dynamics and Logics of Civil War.” Journal of Conflict Resolution 61(9): 1992-2016.

Chenoweth, Erica, and Maria J. Stephan. Why Civil Resistance Works: The Strategic Logic of Nonviolent Conflict. New York: Columbia University Press, 2011.

Davenport, Christian, David A. Armstrong II, and Mark I. Lichbach. 2006. “From Mountains to Movements: Dissent, Repression and Escalation to Civil War.” International Studies Association Annual Convention.

Dudouet, Véronique. 2013. “Dynamics and Factors of Transition from Armed Struggle to Nonviolent Resistance.”
Journal of Peace Research 50(3): 401-13.

Fundación Ideas para la Paz (FIP). 2010. “Mandos medios de las FARC y su proceso de desmovilización en el conflicto colombiano: ¿una apuesta para la paz o para la Guerra?” Informe nº 10. Accessed on February 23, 2021.

Gustafson, Daniel. 2020. “Hunger to Violence: Explaining the Violent Escalation of Nonviolent Demonstrations.”
Journal of Conflict Resolution 64 (6): 1121-45.

Hegre, Håvard, Håvard Mokleiv Nygård, and Ranveig Flaten Ræder. 2017. “Evaluating the Scope and Intensity of the Conflict Trap: A Dynamic Simulation Approach.” Journal of Peace Research 54(2): 243-61.

Hultman, Lisa, Jacob Kathman, and Megan Shannon. 2014. “Beyond Keeping Peace: United Nations Effectiveness in the Midst of Fighting.” American Political Science Review 108(4): 737-53.

Krause, Jana. 2018. Resilient Communities: Non-Violence and Civilian Agency in Communal War. Cambridge, UK:
Cambridge University Press.

Lawrence, Adria. 2010. “Driven to Arms? The Escalation to Violence in Nationalist Conflicts.” In Rethinking Violence: States and Non-state Actors in Conflict, edited by Erica Chenoweth and Adria Lawrence, 143-72. Cambridge, MA: MIT Press.

Lichbach, Mark I., Christian Davenport, and David A. Armstrong II. 2004. “Contingency, Inherency, and the Onset of Civil War.” Annual Meeting of the Midwest Political Science Association.

McAdam, Doug, Sidney Tarrow, and Charles Tilly. 2001. Dynamics of Contention. Cambridge, UK: Cambridge University Press.

Ramsbotham, Oliver, Hugh Miall, and Tom Woodhouse. 2011. Contemporary Conflict Resolution: The Prevention, Management and Transformation of Deadly Conflicts. Cambridge: Polity.

Ryckman, Kirssa Cline. 2020. “A Turn to Violence: The Escalation of Nonviolent Movements.” Journal of Conflict
Resolution 64(2-3): 318-43.

Sambanis, Nicholas, and Annalisa Zinn. 2005. “From Protest to Violence: An Analysis of Conflict Escalation with an Application to Self-determination Movements.” American Political Science Association Annual Conference.

Shesterinina, Anastasia. 2020. “Committed to Peace: The Potential of Former FARC-EP Midlevel Commanders as
Local Leaders in the Peace Process.” SPERI Briefs. Accessed on February 23, 2021. http://speri.dept.shef.

Shesterinina, Anastasia. 2021. Mobilizing in Uncertainty: Collective Identities and War in Abkhazia. Ithaca, NY: Cornell University Press.

Stedman, John. 1997. “Spoiler Problems in Peace Processes.” International Security 22 (2): 5-53.

Theidon, Kimberly. 2016. “Peace in Colombia: A Time to Believe?” Current History 115 (778): 51-6.

Themnér, Anders. 2011. Violence in Post-Conflict Societies: Remarginalization, Remobilizers and Relationships.
London: Routledge.

Walter, Barbara. 2002. Committing to Peace: The Successful Settlement of Civil War. Princeton, NJ: Princeton
University Press.

Young, Joseph K. 2013. “Repression, Dissent, and the Onset of Civil War.” Political Research Quarterly 66(3): 516-32.

Zartman, I. William. 1995. Elusive Peace: Negotiating an End to Civil Wars. Washington, DC: Brookings Institution.

Zukerman Daly, Sarah. 2014. “The Dark Side of Power-Sharing: Middle Managers and Civil War Recurrence.” Comparative Politics 46(3): 333-53.