Mobilizing in Uncertainty: Collective Identities and War in Abkhazia

ABSTRACT

In this reflection, Anastasia Shesterinina introduces her new book, Mobilizing in Uncertainty: Collective Identities and War in Abkhazia (Cornell University Press, 2021). We invited Nina Caspersen at the University of York, Jesse Driscoll at the University of California, San Diego, and Edward Schatz at the University of Toronto to comment on the book. Their commentaries are followed by a response from the author.

How do ordinary people navigate uncertainty to make mobilization decisions in civil war? This question motivates my book on Abkhaz mobilization decisions in the Georgian-Abkhaz war of 1992–1993. While existing approaches to mobilization in civil war start from the assumption of potential participants’ knowledge of the risks involved in mobilization, 150 in-depth interviews that I collected with participants and nonparticipants in the war in Abkhazia demonstrate that these risks were not well understood when Georgian forces entered Abkhazia on August 14, 1992. Instead, the events that marked the beginning of the war were characterized by intense uncertainty. Despite decades of intergroup conflict that preceded the war, the advance of Georgian forces into Abkhazia came as a shock for most regular men and women in Abkhazia. ‘Tanks entered all of a sudden on August 14,’ witnesses remember. ‘People were at work, at the beach. It was like thunder in the middle of a sunny day.’ The Georgian advance ruptured everyday life in Abkhazia and posed with unprecedented urgency the questions of whether this was a war, who was threatened, by whom, and to what extent, and how to act in response.

What followed was a complex social process of information filtering, which I call collective threat framing, whereby people drew on their shared understandings of conflict and their roles in it, or collective conflict identities that they developed before the war, to make sense of the violence and decide whether and how to mobilize. National leaders articulated the notion of Georgian forces as threatening Abkhazia and the Abkhaz, which local authorities adapted to the needs of local defense across the territory, but these frames were consolidated into mobilization decisions with immediate social networks of family and friends that individuals were embedded in at the time. In this process, people came to perceive the anticipated risk, or threat, associated with the Georgian advance in different ways, as directed to their own safety, that of their families and friends, their locality, or the broader group, and adopted different roles, from escaping the fighting alone to mobilizing to areas of high-intensity fighting together with their immediate social networks, based on whom they understood to be threatened and mobilized to protect. Variable threat perceptions and mobilization decisions had lasting effects on how the war in Abkhazia progressed and how people continued to mobilize during and after the war.

These findings have implications for our understanding of mobilization, civil war, and conflict more generally and for how we study these processes. They challenge a fundamental assumption in conflict research, that individuals know the risks involved in mobilization and calculate their decisions to participate or not in a civil war and in what capacity based on this knowledge. The book shows that individuals do not simply choose to fight or not based on a given notion of risk, but come to perceive threat in different ways affected by earlier experiences of conflict and by social networks at the time of mobilization and act differently based on whom they understand to be threatened and mobilize to protect. Underlying this argument is the recognition that ordinary people experience intense uncertainty when war breaks out in their communities. Why some potential participants join the fighting and in what capacity, but others do not, cannot be grasped without knowing how they interpret the reality they face. This requires broadening the concept of mobilization from a focus on recruitment of fighters into armed groups to appreciating it as an ongoing process involving organization of and participation in collective action that shapes people’s shared understandings of conflict and their roles in it. This also requires broadening the focus from civil war as an isolated phenomenon to appreciating the importance of pre- to postwar experiences of conflict from the perspective of the socially embedded actors involved.

In this brief reflection, I discuss how I conducted the research underlying the book, how this research generated a novel focus on uncertainty in civil war mobilization and the sociohistorical approach that I developed as a result, and how this approach can help further advance research in this area.

Studying Civil War Mobilization

To understand how ordinary people make difficult mobilization decisions in the uncertainty of the war’s onset, I carried out immersive field research with the very participants involved. Secondary materials, archival and news sources, and elite interviews are important to the overall goals of this research, but these sources rarely document the processes by which people arrive at their decisions. Nor can we get at these processes by inferring people’s willingness to mobilize from civil war outcomes, observed behavior, assumption of interests, or retrospective assignment of grievances that could have affected these decisions given the history of intergroup conflict. At the same time, careful selection of primary materials is crucial to forming the sufficient comparative basis that is necessary to make systematic conclusions about these processes and these materials should be contextualized and substantiated with additional sources. This book is based on my careful selection of the case of Abkhazia, locales within this case, and participants in the interviews that I collected in these locales and a wide range of supporting materials.

Abkhazia is a case of long-term intergroup conflict that culminated in a civil war,¹ with protracted violence and challenges of a political resolution in the postwar period. After the Russian Empire established control and displaced the majority of Abkhaz in the nineteenth century, Georgian-Abkhaz conflict evolved in the context of the repopulation of Soviet Abkhazia as this relatively small territory was formally integrated into the Georgian state structure within the broader Soviet Union. Everyday tensions, including arguments, bending of social customs, and brawls, nonviolent contention, exemplified by letter writing to the Soviet center in Moscow, public gatherings, and strikes, and violent clashes, particularly the Georgian-Abkhaz clashes of 1989 that sparked in the context of separatist mobilization and spread across Abkhazia, injuring and killing hundreds of people, characterized the Soviet period.

However, mobilization for war was not expected in Abkhazia.² The clashes of 1989 demonstrated the dominance of the Georgian group and the repressive capacity of the Georgian state. Only Soviet troops could stop the violence at that time and many Abkhaz participants were dismissed from office and criminally charged. The Abkhaz were at a significant disadvantage in manpower and arms when the war began. The population of 5 million in Georgia and the 240,000 Georgians in Abkhazia greatly exceeded the 93,000 Abkhaz. Georgia inherited a large share of Soviet weapons in the South Caucasus and while it did not have a functioning army in 1992, its forces were more numerous and better armed than any resistance the Abkhaz could have mounted, even with other non-Georgian minorities in Abkhazia. An inflow of foreign fighters and Russian support strengthened the Abkhaz force in the course of the war, but this support cannot explain mobilization at the war’s onset, when Georgian forces immediately captured most of the territory of Abkhazia, with casualties observed on the Abkhaz side, and the potential of external support was unclear. In these conditions, at least one thousand Abkhaz mobilized at the war’s onset and up to 13% of the population mobilized in the course of the war. This puzzling outcome and a wide range of conflict dynamics around this outcome make Abkhazia particularly suitable to studying mobilization in civil war.

This case is also characterized by variation in subnational structural conditions and territorial control established during the war, which could have affected ordinary people’s mobilization decisions (Kalyvas, 2006). Proximity to the Russian border in western Abkhazia, where people could flee at the war’s onset and external help could come from, and the former Soviet military base in Gudauta, where some weapons and a hiding place could be accessed in central Abkhazia, contrasted with the situation in the east, where the administrative border with Georgia facilitated immediate Georgian blockade of the area. After Georgian forces crossed the Ingur/i³ River in the east on August 14, Georgian marines landed in seaside Tsandrypsh/Gantiadi in the west on August 15, to establish control over the east and west of Abkhazia by August 18, with the center left to the Abkhaz side. The Abkhaz regained the area adjacent to Russia’s border in October 1992 with the help of foreign fighters and the map of territorial control changed, facilitating the formation of the Abkhaz army.

I selected four primary and two secondary field sites to leverage this variation across space and time within Abkhazia (see Figure 1 below). I studied Abkhaz mobilization in the east in predominantly Georgian Gal/i, besieged Tqvarchal/Tqvarcheli, and the capital, Sukhum/i, which Georgian forces controlled during the war. I conducted fieldwork in Sukhum/i and collected interviews with former residents and additional data on the less accessible secondary field sites of Tqvarchal/Tqvarcheli and Gal/i in other areas of Abkhazia and Georgia. I worked on Abkhaz mobilization in the west in Gagra, which Georgian forces captured early in the war, and Pitsunda, which was contested due to its proximity to the Gagra front line, but both of which the Abkhaz regained in October 1992, and Gudauta, which was under Abkhaz control in the course of the war.

Figure 1. Research sites.

Source: Reprinted with permission from Shesterinina, 2021, Fig. 1.1. © Copyright Cornell University.

The book is based primarily on 150 in-depth interviews with 142 individuals that I collected in these areas in the fall and winter of 2011, following an exploratory field trip in 2010, which helped me make important ethical decisions, such as the choice to rely on local networks of trust rather than formal affiliation, that guided my core fieldwork. My combined referral and targeted respondent selection strategy achieved a balance between individuals who had been recruited into the Abkhaz armed structure, the Abkhaz Guard, before the war, those who mobilized for war spontaneously, without prior recruitment, and those who did not fight and reflected differences in gender, age, and pre- and postwar backgrounds in the interviews. This variation helped me address alternative explanations of mobilization in this case and understand whether and how people’s prewar experiences and postwar affiliations affected their wartime mobilization decisions and what they told me about the war.

Our semi-structured conversations walked through respondents’ life histories in the context of conflict. Questions on childhood focused on family stories and early memories of intergroup relations. Reflections on prewar adulthood explored daily intergroup interactions in the familial, neighborhood, educational, and employment settings. The interviews then covered step-by-step recollections of the first days of the war—where respondents were, how they learned about the war, whom they talked to, and what actions they took—as well as narratives on whether people anticipated a war, how they viewed Georgian forces, and what motivated them to act. Discussion of wartime and postwar mobilization concluded the interviews. My observations in national and local war-related events, meetings of mothers’ and veterans’ organizations, communal celebrations, and everyday life, original news and document archives, and secondary materials, including comparable interviews collected by other researchers during the war and midway between the war and my field research, often with the same respondents, and interviews with Georgians displaced by the war and experts that I collected in Georgia and Russia in 2013 contextualized Abkhaz accounts.

These extensive materials from an understudied but critical, outlier case of Abkhazia where mobilization for war is difficult to explain using existing theories (see ‘A sociohistorical approach’ below) offer a richly detailed set of mobilization trajectories, with sequences of individual actions situated in the context of structural differences across the territory, social networks at the time of mobilization, and the history of intergroup conflict, and point to a general process of mobilization as understood by the participants themselves that sheds new light on our approaches to mobilization in civil war.

Uncertainty in Civil War Mobilization

A key insight that emerged from this research is the recognition of uncertainty that defines ordinary people’s experience of mobilization at the war’s onset. In designing this research, I followed other scholars who view mobilization as a problem of overcoming the high risks of repression, injury, and death that potential participants face in settings of political violence and civil war. This literature starts from the assumption that individuals have the knowledge, even if limited, of the risks involved in mobilization and goes on to isolate push and pull factors that drive individuals to mobilize despite these risks. Having selected the field sites where differential access to weapons, escape routes, and hiding places and patterns of Georgian and Abkhaz territorial control could have posed distinct risks to potential participants, I expected respondents to calculate their mobilization decisions based on these risks and differently so in the east and west of Abkhazia.

None of my respondents engaged in such calculation, however. In contrast, nearly all described the situation at the time of the Georgian advance as one of intense uncertainty. August 14, 1992, began as a regular day for most men and women in Abkhazia. People set out for their usual activities and made plans for the following days. It was ‘regular life’ and ‘everything was in order,’ respondents sum up. The sudden appearance of Georgian troops, equipped with tanks and artillery and supported by helicopter fire, interrupted current and planned activities. Traffic stopped, people left work and ran to the streets, and traditional gathering places overflowed with crowds. The ordinary Abkhaz experienced profound confusion over the Georgian advance. A woman who later joined the Abkhaz force as a nurse demonstrates the rupture of everyday life that the Georgian advance produced: ‘That day I was making jam. I stood in the garden and cooked on the fire. We usually make a lot of jam. My daughter ran in and said, “The war started!” I asked, “War? With whom?” I froze.’

The recurring references to uncertainty surrounding the Georgian advance were one of the most surprising aspects of my field research. This ethnographic surprise (Bayard de Volo, 2013, p. 220) suggested to me that potential participants did not have the knowledge of risk that observers often assume or attribute retrospectively. ‘No one understood what was going on—how serious it was, how long it would last, whether it was a war,’ a man who fought on the Abkhaz side captures it well. The ordinary Abkhaz could not make out the meaning of the Georgian advance. Many did not believe that a war could start in Abkhazia and interpreted the events as a clash similar to that of 1989, hoping for protection from the disintegrating Soviet troops, as in the past. The events could also be interpreted as a policing action by Georgia. The advance took place as criminal activity was rampant on the railroad that crossed Abkhazia. Yet people doubted that Georgian forces ‘came to guard the railroad … with tanks.’ Georgian forces could equally have entered Abkhazia to pursue supporters of President Gamsakhurdia ousted from Tbilisi who were ostensibly hiding in Abkhazia with kidnapped Georgian officials in the context of the ongoing civil war in Georgia. Yet the Abkhaz feared that they arrived to ‘settle the problem of Abkhazia once and for all.’ As a result, people did not know whether, in what capacity, and, most importantly, for whom to mobilize in response to the advance. Was one’s own or one’s kin’s safety a priority over that of the Abkhaz group or the population at large, including its Georgian part?

Taking seriously the dilemmas that Abkhaz men and women confronted during the first days of the war revealed that uncertainty rather than risk-based calculation characterizes the context of mobilization when war disrupts ‘everyday routines and expectancies’ (Snow et al., 1998, p. 2). As Straus (2006, p. 65) similarly describes the situation in Rwanda in 1994, ‘[t]he president’s [Habyarimana’s] assassination and the resumption of war ruptured the preexisting order, creating a feeling of intense crisis and uncertainty in local communities.’ ‘A picture of daily life [was] suddenly transformed by civil war,’ Fujii (2009, p. 78) corroborates in the Rwandan case. The intense uncertainty that arises as a result puts a premium on the urgency of mobilization decisions because violence can have different meanings. In this context, people come to understand violence in multiple ways rather than know its nature and make mobilization decisions based on their variable interpretations instead of selecting from a range of options associated with potential outcomes through cost-benefit calculation.

This form of uncertainty thus differs from the classic understanding of the term as ‘general unreliability of all information’ in war (Clausewitz, 1976, p. 140). In the classic view, uncertainty is related to risk in that actors do not know the outcome, but make choices among a range of possibilities and their probability of success. Uncertainty dissipates once relevant information becomes available (Fearon, 1995). Still, misperception, among other cognitive constraints, can challenge decision-making under strain (Jervis, 1976; Tversky & Kahneman, 1974). My notion also differs from the uncertainty that is normalized in everyday life through the culture of fear (Green, 1999), violence in the realm of the ordinary (Das, 2007), and existential stress (Finnström, 2008) in the context of intergroup conflict and from the ongoing uncertainty in protracted fighting, where people develop expectations about the occurrence of violence and how to act in response (Arjona, 2016). No such expectations exist when war disrupts everyday life in major ways and people have to navigate the intense uncertainty that results from this rupture in order to make their mobilization decisions.

A Sociohistorical Approach

How did the Abkhaz navigate this uncertainty to arrive at different decisions, from fleeing to Russia, hiding in Abkhazia, or rarely defecting to the Georgian side, to providing direct or indirect support or fighting in one’s locality or areas of utmost intensity, such as the capital? Another ethnographic surprise that emerged in my fieldwork was that so many of my respondents, even if they were highly politicized and participated in Abkhaz activism before the war, fled the territory when the war began, often with relatives and friends. In turn, many others without an activist past remained in Abkhazia and mobilized on the Abkhaz side. This challenged the theoretical expectation that I had of the importance of prior activism for future mobilization (McAdam, 1986). I also expected that grievances (Gurr, 1970; Horowitz, 1985), social and economic incentives (Petersen, 2001; Weinstein, 2007), and pleasure of collective action against injustice (Wood, 2003) could explain mobilization in Abkhazia. But these factors were similar for most of my respondents, whereas their mobilization decisions varied. Likewise, the security that armed groups provide to their members (Kalyvas & Kocher, 2007) could not explain these decisions as the Abkhaz side could not offer the skills and resources that could have increased participants’ chances of survival relative to nonparticipants at the war’s onset. The Abkhaz often mobilized unarmed and suffered immediate losses.

What, then, could explain the Abkhaz mobilization decisions? How respondents spoke about their perceptions of risk that emerged soon after Georgian forces entered Abkhazia gave me an indication of the underlying process. This process involves three main stages: formation of collective conflict identities in the prewar period; their invocation in collective threat framing at the war’s onset, which affects individuals’ threat perceptions and mobilization decisions; and their continued evolution during the war, with implications for the ways in which conflict participants treat wartime and postwar outcomes. This process forms the foundation of my sociohistorical approach to mobilization. In this approach, people do not operate with a given notion of risk, but come to interpret the anticipated risk, or threat, of mobilization differently by drawing on the history of engagement with intergroup conflict and the social setting in which they find themselves at the time of mobilization. They act differently based on whom they understand to be threatened and mobilize to protect their own safety by fleeing, hiding, or defecting to the stronger side in the war; the safety of their families and friends by fleeing, hiding, or joining the fighting together with these segments of their group, often in their home locales; that of their localities by joining the fighting in the back or front lines in their cities, towns, and villages; or the broader group by mobilizing into areas of high-intensity fighting that the overall course of the war depends on.

The process of mobilization for war, therefore, starts long before the war, with the formation of collective conflict identities, which I define as self-perceptions in relation to and as part of broader conflict. Participation in and observation of everyday confrontation, political contention, and violent opposition—mobilization repertoires characteristic of societies marked by conflict that individuals are exposed to by merely living in the context of conflict—shape people’s shared understandings of conflict and their roles in it. These aspects of collective conflict identities are not static but evolve in the course of conflict, particularly in response to intergroup violence. They develop not in isolation but within the social networks with which individuals experience conflict and these experiences relate individuals to one another and their group in powerful ways. Conflict can be seen as part of daily life, public intergroup interaction, or the realm of the political elite and as nonviolent or violent, stagnant or escalating, and requiring mobilization or not. People can view themselves as active participants, supporters of the cause, and nonparticipants observing the events and transition between these roles over time.

In Abkhazia, decades of participating in and observing everyday arguments, national movement activism, and intergroup clashes, informed by collective memories of Abkhazia’s 1931 political status change, repression of the Abkhaz elite and population, Georgian demographic expansion, the closure of Abkhaz schools, prohibition of the Abkhazian language, and the rewriting of Abkhazia’s history to diminish the role of the Abkhaz, shaped the view of Georgian-Abkhaz conflict as one aimed at the dissolution of the Abkhaz identity in the dominant Georgian mass. Absorbed early on, this shared understanding of conflict affected prewar intergroup relations, as the underlying political issues were taboo and created tension between the Abkhaz and Georgians in private and public. But it was the violent clashes of 1989 that dramatically changed the Abkhaz perspective on conflict and their part in it: ‘We were now certain about their [Georgians’] hatred toward the Abkhaz. This was one of the factors that helped us unite.’ The clashes polarized and militarized the society. Teams split and armed groups were formed, with participants in the clashes and those not previously active joining these groups.

When the war began after three years of relative calm that was preserved despite the presence of armed groups in Abkhazia, these shared understandings did not disappear but were invoked across the national, local, and quotidian settings to make sense of the Georgian advance by addressing the questions of whether this was a war, who was threatened, by whom, and to what extent, and how to act in response. This collective threat framing mechanism follows the steps of articulation of threat posed by imminent violence by the national elite, local adaptation of national messages that in general shifts threat framing to fit local needs of defense, and consolidation of information into collective mobilization decisions within the quotidian networks of family and friends. Hence, the Abkhaz national leader Vladislav Ardzinba drew on the history of Georgian-Abkhaz conflict in his address to the population to articulate the threat of Georgian forces to individual safety, that of the Abkhaz group, and the population of Abkhazia as a whole. Respected local authorities widely cited this message to mobilize support at the local level, but did not simply adopt the broad articulation of threat, instead negotiating collective action with their respective communities to direct it to their cities, towns, and villages rather than areas of utmost intensity.

The emergence of collective threat framing affected how the ordinary Abkhaz perceived the threat of the Georgian advance. People realized that a war had started, rather than a clash similar to that of 1989 or a Georgian policing action, and that mobilization was necessary in response. This framing resonated with the view of Georgian-Abkhaz conflict and their roles in it that they developed before the war. Yet people did not know how to act. It was with immediate networks of family and friends individuals were embedded in and trusted that collective threat frames were consolidated into mobilization decisions. These small groups of people who shared prewar conflict experiences mobilized together to protect those segments of society that they perceived to be particularly threatened, from individual safety to the group at different levels of aggregation, based on the patterns of intergroup conflict that they experienced in the past. The resulting mobilization trajectories were often surprising from the perspective of existing explanations as many politicized individuals fled Abkhazia to protect their own safety or that of close family and friends, directed by their immediate networks, whereas others who were not active before the war stayed in Abkhazia and mobilized together with their immediate networks to protect their localities or the broader group in areas of high-intensity fighting.

As people adopted different roles in the war, their collective conflict identities continued to evolve. How threat was framed at the war’s onset influenced people’s shared understandings of conflict and their roles during and after the war. Abkhaz fighters’ self-perception as defenders based on the framing of Georgian forces as an aggressor was reinforced during wartime battles and postwar defense from the ongoing Georgian threat. As the war progressed, the Abkhaz realized that their group could be eradicated if they were to lose the war from the brutality of Georgian forces, mobilization of many local Georgians against them, and serious losses among the Abkhaz notwithstanding external support in the fighting. With the group’s survival at risk, some instances of the fighting came to be viewed as particularly important. The captures of Gagra early in the war to gain access to a lifeline, Russia’s border, and of the capital, Sukhum/i, at the end of the war to reestablish the Abkhaz government were critical wartime victories, ones that the Abkhaz protected as they displaced most of the Georgian population, secured the administrative border with Georgia, and fought in the postwar decades for the territory that they considered to be their own.

Taken together, participants understood these aspects of their prewar, wartime, and postwar mobilization as part of the broader national liberation struggle, which culminated in the recognition of Abkhazia by Russia and a few other states after the ‘liberation’ of the last area of Abkhazia under Georgian control, the Kodor/i Gorge, in 2008. In this context, the loss of Abkhaz life, exclusion of displaced Georgians, and dependence on Russia are perceived as a necessary price to pay for developing as a de facto state independently from Georgia.

Future Research

This detailed analysis of the Abkhaz case demonstrates how the sociohistorical approach that views mobilization for war as part of the broader pre- to postwar conflict can help advance research in the area. Looking closely at mobilization before, during, and after civil war from the perspective of the actors involved provides insight into a range of conflict processes that scholars are grappling with. How people join collective action, how nonviolent contention turns violent, and how conflicts unfold over time to transform actors’ identities with lasting effects on conflict-affected societies are some of these processes. I identify one mechanism—collective threat framing at the war’s onset, which draws on the prewar conflict history and the social networks at the time of mobilization and has effects on the wartime and postwar processes. Future research should further develop our understanding of how civil war is related to broader conflict beyond this mechanism and the logic of escalation common in conflict studies. Placing ordinary people’s experiences of conflict at the center of analysis is essential to achieving this goal as these actors’ decisions are central to these processes.

Disclosure Statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by Social Sciences and Humanities Research Council of Canada [Grant Number 756-2014-0568] and [Grant Number 767-2009-2189].

Notes

1 The term civil war is rarely used in Abkhazia, where the broader Georgian-Abkhaz conflict is distinguished from the war of 1992–1993. However, the war set off with the entry of Georgia’s forces into its autonomous Abkhazia, unfolded with the participation of the local population in Abkhazia, and became internationalized with the engagement of foreign fighters and Russian support, which are common characteristics of civil war.

2 In an earlier study, Beissinger (2002, p. 222) found that there was little chance that ‘a group with the structural characteristics … of the Abkhaz would have engaged in separatist mobilization,’ defining it as an anomalous case in the region.

3 The English spelling of proper nouns differs in Georgia (e.g., ‘Enguri’) and Abkhazia (e.g., ‘Ingur’). I use the spelling common in academic research (e.g., ‘Ingur/i’).