Check for updates
Journal of ,
Peace Research
Ethics, empathy, and fear in research on violent conflict
Anastasia Shesterinina ©
Department of Politics, University of Sheffield
Journal of Peace Research 2019, Vol. 56(2) 190-202 © The Author(s) 2018
Article reuse guidelines: sagepub.com/journals-permissions DOI: 10.1177/0022343318783246 journals.sagepub.com/home/jpr
Abstract
The discussion of ethics in the social sciences focuses on 'doing no harm' and 'giving back' to research participants, but does not explore the challenges of empathy and fear in research with participants in political violence and war. Drawing on 180 in-depth interviews on the Georgian-Abkhaz war of 1992-93 collected over eight months between 2010 and 2013 primarily in Abkhazia, but also Georgia and Russia, I argue that researchers can come to empathize with some but fear other participants in past and present violence. These emotional responses can influence researchers' ability to probe and interpret interviews and respondents' ability to surpass strong positions to explore dilemmas of participation in violence. By empathizing with not only 'victims' and 'non-fighters' as I had expected based on my pre-existing moral-conceptual categories, but also participants in the war, I found that individuals adopted multiple overlapping roles and shifted between these roles in the changing conditions of violence. In contrast, failing to empathize with and fearing those who continued to participate in violence after the war of 1992-93 limited my ability to fully appreciate the complexity of their participation, but shed light on the context of violence in contemporary Abkhazia. This analysis shows that reflection on the role of empathy and fear in shaping our interactions with research participants can help advance our understanding of participation in violence and this difficult research context.
Keywords
civil war, empathy, ethics, fear, political violence, qualitative field research
Introduction
The discussion of ethics in the social sciences focuses almost exclusively on the researchers' duty to protect human subjects from any potential harm that may come from participation in research (Wood, 2006). The imperative applies to the ethics board process and as an ongoing responsibility of researchers — from research design to publication (Fujii, 2012). Feminist scholars take a step further, 'insisting] that a researcher cannot be content merely to record another's life story for scholarly publication but must "return the research" to the subject as a means of empowering the informant' (Blee, 1993: 605). Few studies, however, address the ethics of empathy in intensive fieldwork  with participants in
1 Parkinson & Wood (2015: 22, fn. 1) define intensive fieldwork as 'fieldwork that is qualitative and carried out during long-term (six months or more), at least partially immersive stays in the field, incorporating methods such as participant observation, in-depth interviewing, focus groups, and community mapping'. Here I focus on in-depth interviewing.
political violence and war, which poses difficult dilemmas beyond the 'do not harm' principle and feminist ethics of empowerment. Is it ethical to empathize with some - but not other - participants in past and present violence? How does empathy influence the results of field research?
As a maxim, researchers who conduct intensive field-work are expected to develop empathy for research participants. As Thomson (2010: 27) argues, 'we must act beyond the ethical imperative of doing no harm; we must display empathy, look out for the emotional safety of our interviewees'. Yet empathy in research on violence and war is not always straightforward. While it is possible with such groups as the Mothers of Heroes and Martyrs in Nicaragua (Bayard de Volo, 2009) and peasant Rwandans after the genocide (Thomson, 2010), it may not be with 'unloved' and 'repellent' groups (Blee, 1993; Fielding, 1990; Gallaher, 2009), especially those we fear.
Corresponding author:
a.shesterinina@sheffield.ac.uk
Shesterinina
191
In the civil war setting, these groups include political and military elites, insurgent leaders, warlords, and members and supporters of armed units. Researchers can find it difficult to empathize with these groups, but can also develop varying degrees of empathy for these actors in the course of research. Resultant selective empathy, or differential treatment of research participants based on emotions that researchers experience during fieldwork, is an ethical dilemma that deserves careful attention in research on violence and war.
Selective empathy is not only a problem of research ethics, but also shapes research results. As researchers empathize to varying degrees with some research participants and fear others, these emotional responses affect the dynamics of interaction, accounts of research participants, and ability of researchers to probe and interpret the accounts. Empathy that I developed for research participants who fought on the Abkhaz side in the Georgian-Abkhaz war of 1992-93 encouraged respondents to explore issues challenging the official narrative of the war and provided me with insight into the complexities of participation trajectories, where individuals adopted multiple overlapping roles and shifted between these roles in the changing conditions of the war. In contrast, my fear of research participants who were active in recent violence prevented me from empathizing with these respondents and limited my ability to probe their accounts or appreciate the complexity of their roles, which had also most likely overlapped and shifted during and after the war. Yet this fear helped me understand the context of isolation and protracted conflict in postwar Abkhazia.
This article builds on eight months of field research conducted over 2010-13, primarily in Abkhazia, but also Georgia and Russia, and studies of political violence and war in political science, anthropology, and sociology to explore how empathy and fear can impact researchers in conducting interviews and interpreting the results, how displaying these emotions can influence respondents in positive and negative ways, and how researchers can treat these emotions during and after fieldwork. The article begins with a brief assessment of the state of the art on ethics and emotions in research on violent conflict. I then characterize the conditions of fieldwork in postwar Abkhazia and discuss the anticipated ethical dilemmas and assumptions about empathy that I had at the outset of my research, how these assumptions changed in the course of my fieldwork, and how reflection on these changes impacted the results of my study. I present the complex roles that I discovered with participants in the 1992-93 war and analytical challenges with
those active in ongoing violence. The next section draws implications of this reflexive process for research with participants in violence and war. This analysis highlights the relational character of immersive research on violent conflict and contributes to the literature on research ethics and emotions in general and fieldwork on and in settings of violent conflict in particular by drawing attention to empathy and fear in these settings.
Ethics and emotions in research on violent conflict
Do principles of ethics apply in similar ways in contexts of violent conflict, such as civil war, as in other field conditions? How do emotions affect the process of research in these contexts? Social scientists have recognized a range of ethical and emotional challenges and dilemmas that arise in the context of field research on violent conflict.2 The discussion of ethics of fieldwork in societies torn by large-scale violence and war draws on and extends the ethical principles developed in research on human subjects in general (Campbell, 2017). These guiding principles include respect for persons, beneficence, and justice and are applied through informed consent, assessment of risks and benefits, privacy and confidentiality, and fair selection of research subjects (Fujii, 2012: 718). The Belmont Report (National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research, 1979) advanced these principles in response to ethical issues in biomedical and behavioral research in 1979. Their implementation has since been extended to social science field research and has rested with university-based Institutional Review Boards (IRBs) (Yanow & Schwartz-Shea, 2013).
Research ethics beyond 1KB
Scholars of political violence and war have questioned the emphasis on IRB approval as the cornerstone of research ethics in field research and have highlighted the ways in which conflict and post-conflict environments intensify and pose unique challenges surrounding the ethical principles developed in the context of medical research. Protection of human subjects is the paramount ethical concern for scholars of violence and war where
2 See Sriram et al. (2009) and Mazurana, Jacobsen & Gale (2013) on conducting research in areas of violent conflict. See Campbell (2017) for a review of the literature on ethics of research on conflict and violence. On the emotional challenges in the field, see Nordstrom & Robben (1995) and Thomson, Ansoms & Murison (2013).
192
journal of Peace Research 56(2)
repercussions stemming from research can be severe due to 'political polarization, the presence of armed actors, the precarious security of most residents, the general unpredictability of events, and the traumatization through violence' (Wood, 2006: 373). Participation in research in these highly politicized conditions can result in retraumatization,3 local retaliation, and state investigations that can subject interlocutors to imprisonment, torture, and even death for sharing politically sensitive or potentially compromising information. The risks apply not only to research subjects, but also interlocutors who may not qualify as 'subjects' according to IRB, such as field assistants and local colleagues, but who can become vulnerable to surveillance through their perceived association with the researcher (Cronin-Furman & Lake, 2018; Thomson, 2009).
In this context, informed consent is particularly challenging and requires multiple options for interlocutors to refuse to participate, decline to answer questions, answer with or without attribution, and withdraw at any time (Wood, 2006: 379). The risks and benefits of participation can change with shifting political circumstances: when and what is sensitive and who might be vulnerable to the risks resulting from participation in research is not always clear and safeguarding human subjects requires consistent reassessment of risks and benefits during and after fieldwork (Fujii, 2012: 721). Similarly, ensuring privacy and confidentiality means not only securing data from state officials, armed actors, and other groups seeking access to it in the field, but also protecting the interlocutor identities during writing and publication - a critical issue in light of recent transparency debates in political science (Parkinson & Wood, 2015). Thus, researchers have taken significant efforts to protect the collected data at security checkpoints (Wood, 2006: 381), deidentify it and use carefully selected pseudonyms in writing (Fujii, 2012: 721), and present the process of data collection in methodological sections or appendices without compromising participants' backgrounds (Parkinson, 2013; Shesterinina, 2016).
Feminist ethics of care
Recognizing that the traditional values of neutrality, objectivity, and detachment are difficult (if not counterproductive) in intensive fieldwork (Bayard de Volo, 2009; Yanow & Schwartz-Shea, 2006), feminist scholars have advocated a step further, toward the ethics of care
3 Yet some subjects who experience distress in the interviews express relief in recounting their story (Wood, 2006: 377).
based on 'interactive and universal moral respect for every individual'; in this model '[r]esearch should be collaborative, useful to research participants and provide a means of empowerment' (de Laine, 2000: 28, 210).4 Developing empathy5 for research participants has become a general expectation in qualitative work, particularly in-depth interviewing (Rubin & Rubin, 1995: 12) and ethnography (Schatz, 2009: 5). 'Interpretation as a method,' Yanow and Schwartz-Shea (2006: 22) say, 'is conducted as "sustained empathic inquiry" [...], in which empathy constitutes an intentional embracing of the other's meaning.'
Engaged and reflective listening form the foundation of empathy understood as embracing of the respondent's meaning in interview-based work. As an engaged listener, the researcher is open to understanding the history of conflict and violence from the perspective of her research participants. This involves being attentive to the interlocutors' 'spoken and unspoken thoughts and feelings which they do not always articulate in their [... ] interview responses, but which emerge in other ways' (Fujii, 2010: 231). Silences, gestures, and expression of comfort and distress can help care for respondents. Avoiding potentially traumatizing topics and carefully selecting questions based on these signals are among the steps that relate empathy to the researcher's ethical duty to protect subjects from harm.
Empathy, fear, and research results Empathy, however, is not only an ethical concern; it also affects research results. Displaying an interest in comprehending respondents' perspectives and sensitivity to their well-being can shape what respondents say. On the one hand, it can encourage respondents 'to reflect on and even explore [their] ideas, to reveal not only strong views but also worries, uncertainties' (Yanow & Schwartz-Shea, 2006: 118). Empathy can thus invite respondents to confide in the researcher what they would not
4 This can involve drawing attention to marginalized voices, telling multiple, often conflicting accounts of violence, returning research materials to interlocutors, and contributing to social change (Thomson, 2010; Wood, 2006).
5 The term 'empathy' has been used to denote 'a disposition that affects interaction; an intentional attitude; a moral or ethical choice; a specific act of imagining or entering the world of the other; a mode of attending to the other; an automatic response to something seen or heard' (Cameron & Seu, 2012: 284). It has been related to other terms, such as sympathy, rapport, and trust, that can be associated with, but are different from the concept of empathy as embracing the other's meaning adopted here. On sympathy, see Schatz (2009); on rapport, Fujii (2018); on trust, Thomson (2010).
Shesterinina
193
otherwise have revealed. On the other hand, it can provoke respondents to tell the stories that the researcher might want to hear while remaining silent on the details that would present themselves or their group in a negative light (Robben, 1995). Departures from the official narratives justifying past and present violence in the interviews point to a positive effect of empathy on respondents' accounts.
In my research on Abkhaz participation in the Geor-gian-Abkhaz war of 1992-93, empathy was central to inviting respondents to express personal views beyond the official narrative of the war as an offence on Abkhazia by the Georgian forces. Respondents often began their accounts with this narrative, but revealed their nuanced and conflicting positions in the course of the interviews, raising dilemmas of responsibility, betrayal, regret, and violence by the Abkhaz during and after the war. In response to these dilemmas, many adopted a range of overlapping, shifting roles in the war. While I did not expect to empathize with Abkhaz war participants whom I first saw merely as perpetrators of violence against the displaced Georgian population of Abkhazia,6 this insight would not be possible without my expression of interest in and sensitivity to respondents that I developed during fieldwork.
Empathy is a relational, intersubjective process that can affect both what respondents tell and how researchers conduct and analyze the interviews (Hollan & Throop, 2011: 8). During fieldwork, '[h]ow the researcher asks questions changes depending on how he or she feels about the topic or the interviewee' (Rubin & Rubin, 1995: 12). For example, fear of the research setting and subjects can limit researchers' ability to ask sensitive questions or develop follow-up queries to probe the official narratives of violence. As Driscoll (2016: 22) describes fear's paralyzing effect in his research with Georgian warlords, '[s]ometimes I got scared. The interview subjects always noticed. The interview ended shortly afterwards.'7 After fieldwork, the researcher is likely to remember these dynamics of interaction and interpret the findings based on her understanding of the research context gained from these emotions (Blee,
6 Up to 240,000 Georgians were displaced by the war and only 40,000 returned to Abkhazia (Trier, Lohm & Szakonyi, 2010: 21).
7 In some cases, it is possible to hide and overcome fear. Blee (2002: 20) recalls 'the need to display certain feelings' in her work with Ku Klux Klan women in Indiana; she 'mimicked what [she] did not feel [... or] withheld the emotions [she] did feel' to complete the interviews.
2002; Green, 1994; see section on Emotional reflexivity, below).
The weight of emotions is particularly heavy in research on violence and war (Nordstrom & Robben, 1995: 3). Researchers working in conflict and post-conflict contexts can be isolated and exposed to violence, surveillance, intimidation, and their own choices that pose risks to their safety (Sluka, 2012). The emotions of'fear, anger, outrage, grief, and pity [can arise] through observing, suffering, or fearing the effects of violence' (Wood, 2006: 384). Fear is a consistent undercurrent of fieldwork on violent conflict (Diphoorn, 2013; Green, 1994; Thomson, 2009). Fear stems not only from the research environment 'where violence is a key currency', but also research participants, as researchers become vulnerable to their subjects, especially those who have access to coercive means and perceive the researcher as a political resource (Malejacq & Mukhopadhyay, 2016: 1013).
These emotions affect researchers' level of empathy. As Crawford (2014: 542) finds drawing on the neu-roscience literature, 'fear can diminish empathy [... and l]ow levels of empathy can lead to difficulty sympathizing with others, and therefore decreased opportunities for positive interactions'. Blee (1998: 390) speaks of intimidation and fear of her informants, Ku Klux Klan women in Indiana:
Informants constantly highlighted my vulnerability to them, asking whether I was afraid to come see them, whether I was afraid to be in their home. Others suggested that I would face harm if I did - or sometimes if I did not - interview a particular person in the movement.
For Blee (2002: 13; 1998: 390), empathy was a difficult starting point with 'racist activist[s] whose life is given meaning and purpose by the desire to annihilate you or others like you'; 'it was fear, not empathetic connection' that characterized the interviews. Drawing on her fieldwork with right-wing militias in the United States and Northern Ireland, Gallaher (2009: 143) agrees that 'empathy rarely exists for those doing research on repellent groups', particularly in the context where the IRB offers no protection for the researcher. These scholars thus conclude that 'basing interviews on rapport and empathy is helpful for groups that are "conducive, whimsical, or at least unthreatening", but it hardly seems appropriate when the groups are hostile or frightening' (Blee, 2002:12; Gallaher, 2009: 135).
194
journal of Peace Research 56(2)
This was evident in my interviews with participants who were active in ongoing violence in Abkhazia. While I empathized with many respondents who fought in the 1992-93 war, I struggled to develop empathy and feared those whose violent activity persisted into the present day. My ability to ask these respondents about their trajectories freely was limited as a result and departing from the official narrative of the war in my interpretation, to develop a nuanced account of their participation and grasp the complexity of their roles in past and present violence, was difficult. I recalled my fear in the interviews and was not willing to see the conflict from the perspective of these respondents.
Selective empathy
Empathy, therefore, is not as straightforward in intensive fieldwork on violence and war as the general guidelines for researchers suggest. While feminist scholars argue that it is possible even with 'unloved groups' (Sehgal, 2009: 300), it may not be possible, or even desirable, with people active in violence and war. 'Since an ability to sympathize lies at the core of ethnography', Schatz (2009: 8) captures the paradox, 'conducting a study that relied on ethnographic contact with such individuals would be practically and sometimes ethically difficult'. Empathy in this context can oppose personal ethics. 'If you find yourself sympathizing with interviewees who are killers [...], you might begin to have questions about yourself (Rubin & Rubin, 1995: 13). It can also be eluded by 'ethnographic seduction' based on manipulation of appearances 'prominent in research on violent political conflict because the interlocutors have great personal and political stakes in making the ethnographer adopt their interpretations' (Robben, 1995: 84). However, Brouneus (2011: 137) insists, '[e]mpathy does not mean to identify with the other or to become absorbed in the same feeling. It is to understand the other's perspective - even if we do not agree with [... or] are repelled by what is being said.'
The debate as a result centers on the question of whether empathy can and should be afforded to respondents whose world-view is profoundly different from that of the researcher's. Empathy in this context is easily dismissed as a failure of personal ethics or a result of ethnographic seduction. I argue, on the other hand, that the dilemma8 of the ethics of empathy lies not only in
8 I follow de Laine (2000: 3-4) in characterizing an ethical dilemma 'as a problem for which no course of action seems satisfactory [. .. ] Ethical dilemmas [... ] admit of no comfortable outcome but must be lived.'
the researcher's ability or inability to empathize with participants in violence and war, but also in the selective nature of empathy in contexts of violent conflict. Researchers in these contexts can develop varying degrees of empathy, especially due to their fear of the research environment and/or participants. Researchers are therefore able to afford empathy to some - but not other -individuals involved in the interviews, including participants in past and present violence. This depends on the conditions under which their participation took place and how these conditions affect the researcher's emotions in the interviews. Respondents involved in present-day violence, for example, might invoke a different set of emotions than those who participated in a distant war. This shapes how we treat respondents, by asking certain questions and covering certain topics, and what insights we gain through our research as a result.
Emotional reflexivity
Gaining an understanding of the complex environments marked by violent conflict requires a careful reflection on what researchers' emotions in this context mean. Why do we fear some but not other respondents? Why can we hide or overcome these emotions in some field sites but not others? What can we learn from interpreting the empathy and fear that we experience during fieldwork? In a range of conflict environments, answering these questions about the researcher's emotional state in the interviews provided researchers with grounded knowledge of the context itself. For example, in her study of fear 'as a way of life' in postwar Guatemala, Green (1994: 230) did not 'stand apart as an outsider [..., as] any understanding of the women's lives [in Xe-caj] would include a journey into the state of fear'. It was not her research participants whom Green feared, but the relationships and interactions, including with military commanders, in the broader context of militarized Guatemala. Empathizing and experiencing fear with the Xe-caj women was not ethically contentious and shaped Green's knowledge of 'the system of violence and terror' (245). In contrast, Blee (1993: 597, 604) exposed inherent challenges of empathy for individuals 'active in the politics of intolerance, bigotry, or hatred' and 'was prepared to hate and fear [her] informants'. The mutual fear between her and the racist women she interviewed and their unforeseen complexity, which challenged stereotypes about them, helped Blee understand 'what it must feel like to be inside a racist group' (Blee, 2002: 19). These insights would not be possible without reflection on the emotional dynamics in the interviews.
Shesterinina
195
My work in Abkhazia combined elements of these studies. I experienced fear of the research site, an isolated environment torn by war where violence continued for decades, especially along the Georgian-Abkhaz border area, and of some respondents, those active in postwar violence. Similarly to Green (1994), I feared my interactions with Abkhaz military commanders in the context of current violence. Similarly to Blee (1993), respondents active in this violence invoked fear in the interviews. But not all yielded this emotional response. Instead, similarly to Pierce (1995) in her study of gender in law firms, I was able to relate more to some respondents than others based on my understanding of their roles in the war and postwar violence that emerged in the interviews.9 This selective empathy shaped what people told me, whether I could probe their accounts, and how I interpreted the findings as I reflected on what my emotional responses meant in the context of postwar Abkhazia.
Emotional reflexivity was critical to my interpretation of interviews. Had I not paid attention to the changes in my relationship with participants in the distant war of 1992-93, I would not have gained insight on the multiple overlapping roles that they adopted in the war. My reflection on developing empathy for these respondents during my fieldwork suggested the importance of departing from my initial assumptions about respondents based on the moral and conceptual categories of 'perpetrator' and 'fighter' that I assigned to them. Similarly, my interpretation of the fear that I experienced with respondents active in ongoing violence, while preventing me from probing their accounts and fully appreciating the range of roles that they adopted during and after the war, pointed to their role in the persistence of violence and fear in present-day Abkhaz society. It is to these different emotional responses and their ethical and research implications in the Abkhaz case that I turn now.
Research in postwar Abkhazia10
Post-war Abkhazia is a challenging setting for intensive fieldwork due to the isolation of the de facto Abkhaz state, devastation and displacement brought by the war,
9 Pierce's (1995: 197) understanding of the sexualized nature of some interactions with male lawyers, where her 'status as an attractive female became more salient [than that of] graduate student', changed her interviews and interpretations.
10 The fieldwork reported in this article was covered by Ethics Certificate no. HI 1-02222 of 21 September 2011, of the University of British Columbia's Behavioural Research Ethics Board.
and protracted violence that marked the Georgian-Abkhaz border area decades after.11 Few scholars have carried out immersive research in Abkhazia. Those who have worked in the area have in general focused on elite interviews and surveys and recorded the cementing of the official narrative of the war, particularly on the return of the displaced Georgian population12 and Abkhazia's present status as a partially recognized state, with limited access to the outside world except through Russia (Cop-pieters, Darchiashvili & Akaba, 2000; Hewitt, 1996; O'Loughlin, Kolossov & Toal, 2011). The lives of ordinary participants in violence have rarely been a matter of in-depth, face-to-face research, yet could reveal insight beyond the narrative of Georgian aggression in the war to eliminate Abkhazia as a separate political entity and the Abkhaz as its core cultural unit.
My aim in the study was thus to understand the Georgian-Abkhaz war of 1992-93 from the perspective of the regular Abkhaz. This relatively short war was preceded by decades of nonviolent conflict over Abkhaz rights and Abkhazia's status, which changed from Soviet Socialist Republic associated with Georgia through a Union Treaty early in the Soviet period, to Autonomous Republic of Georgia in 1931. The status change, along with the Georgian demographic expansion in Abkhazia and suppression of the Abkhaz language and culture, were underlying the conflict, which culminated in the major intergroup clashes in the context of the falling Soviet Union in the 1980s. This violence polarized the society, where different groups lived side by side, into the Georgian and Abkhaz blocs, as other groups joined one or the other side in daily interactions, contentious politics, and, later, war.13
The war began on 14 August 1992, when the forces of the Georgian State Council, including the Georgian National Guard and paramilitary Mkhedrioni, entered Abkhazia from its administrative border in the east and, the next day, encircled the territory from the Black Sea in the west. These forces passed Gal/i, a district bordering
11 Violence waned with Russia's fortification of the Georgian-Abkhaz border after the 2008 Georgian—Russian war.
12 The term 'refugees', or those who flee by crossing borders, is used in Abkhazia, but 'internally displaced persons', who move within the country, is used in Georgia; 'state border' is used in Abkhazia, but 'administrative border' in Georgia.
13 According to the 1989 census of Abkhazia, of the population of 525,061, Georgians constituted 239,872 (45.7%), the Abkhaz 93,267 (17.8%), Armenians 76,541 (14.6%), Russians 74,914 (14.3%), Greeks 14,664 (2.8%), and other groups 15,959 (4.8%). See Trier, Lohm & Szakonyi (2010). On the history of the conflict, see also Hewitt (1996) and Coppieters, Darchiasvili & Akaba (2000).
1%
journal of Peace Research 56(2)
Georgia and dominated by Georgians, blockaded the eastern center Tqvarchel/i, captured the capital Sukhum/i and western center Gagra, cut access to Russia's border, and left Gudauta in central Abkhazia under Abkhaz control.14 Abkhaz men and women mobilized in response to the Georgian advance in the east and west of Abkhazia, often without armed structure or weapons (Shesterinina, 2016: 423).
My fieldwork followed this mobilization in 2011.15 Moving along the single major road, I conducted primary research in Gagra, adjacent front-line town Pit-sunda, Gudauta, and Sukhum/i and collected secondary materials on mobilization in Tqvarchel/i and Gal/i due to recurrent violence there. I carried out 150 in-depth interviews with individuals from a wide range of pre- and postwar backgrounds and wartime roles across these locales, achieving a balance between respondents who participated in the war as fighters and in the support apparatus and those who fled, hid, or maintained neutrality. To avoid personal and institutional bias (Fujii, 2009), I used multiple network referral and targeted selection strategies. I selected only those respondents who fit the purposes of the study and approached necessary respondents not referred by my networks at their workplace. To address issues of distant war's memory (Wood, 2003), my semi-structured interview plan covered pre- to postwar life histories, combining event and narrative questions (Viterna, 2006), using follow-up probes, and triangulating between the interviews in Abkhazia and additional focus group and interview materials collected in Georgia and Russia in 2013. Participant observation in daily informal conversations, meetings of veterans' and mothers' groups, and war-related events strengthened my interviews, as a way to contextualize responses and develop informed follow-up questions, and helped position respondent accounts in the current social setting.
Anticipated ethical dilemmas
This setting presented a range of potential ethical dilemmas that I prepared for, following the 'do no harm' principle and drawing on my exploratory trip to Abkhazia in 2010. I did not establish local affiliation or seek local assistants during my research. Formal or informal
1 The spelling of proper nouns differs in Georgia and Abkhazia, e.g. Gal vs. Gali. Unless quoting from the original interview, I use combined spelling, e.g. Gal/i.
15 For a detailed discussion of my fieldwork, see Supplementary Materials (Shesterinina, 2016).
affiliation with the de facto Abkhaz government, nongovernmental organizations, or universities in Abkhazia, Georgia, or Russia, the central actors in the Georgian-Abkhaz conflict, could draw unwanted attention to local colleagues at partner organizations and raise suspicion among research participants by signaling my support for an organization's position in the conflict. At the same time, I had to ensure that potential participants in my research would not suffer reprisals from their communities or de facto state. Thus I relied on community leaders for logistical support and sought approval for my research from city or town authorities. In a small society where trust originates in local networks and where a researcher's visibility is high, this suggested that my purposes were locally known and were academic in nature.
The detailed informed consent protocol that I devised for all potential research participants in Abkhazia stressed my role as a researcher, the local approval for my research, and academic benefits of the study. I followed Wood (2006) in offering a range of options in an oral consent and kept a de-identified log in field notes to protect the identities of research participants. However, I foresaw that government officials and nongovernmental leaders who are often involved in elite interviews might expect a written informed consent form and request me to note their name, affiliation, and post in the interview record and subsequent writing. While I prepared a written informed consent form for these individuals, with the option of an oral protocol used with most respondents, regular men and women, following these individuals' requests to disclose their identities could compromise their positions in the highly politicized and changing postwar setting where pressures from Georgia and Russia could pose risks to the research participants who fought in the war and now occupy leadership posts. Hence I have chosen not to disclose their names and to excerpt their responses without identifying details - a position that I have taken with both elite and non-elite interviews to equally protect all respondents.
I drew on fieldwork experience of other researchers in the region (Driscoll, 2016) and did not take note of responses that could compromise respondents' or my security in the context of ongoing tension by sharing details of wartime or postwar violence that could be deemed especially sensitive or compromising the current political leadership, security officials, or other actors. As a result, I did not record responses that noted names or networks relevant to present-day Abkhaz politics, retain the record of these responses, or use them in my analysis or writing. To maintain as representative a sample of the
Shesterinina
197
interviews as possible, I instead sought other respondents in similar wartime roles.
While I anticipated and prepared for the ethical challenges associated with the 'do no harm' principle in the context of postwar Abkhazia, I did not expect other dilemmas that emerged during my field research. Developing empathy for respondents who fought and killed in the 1992-93 war was one ethical difficulty that I faced. With most of the Georgian population displaced and unable to return to Abkhazia since the war, this empathy could be seen as a result of failing personal ethics or ethnographic seduction. Yet in line with the feminist ethics of care and empowerment, I came to see it as part of giving voice to regular men and women whose mobilization stories were rarely recorded but were essential for understanding the conflict. Empathizing with participants in past violence was one part of the ethical challenge. Another part was failing to relate to and foster trust with individuals whose accounts of participation in ongoing violence frightened me and constrained in-depth insight. The following sections discuss the importance of empathy and fear for my research results.
Empathy and the distant war
When I arrived in Abkhazia, I viewed the Abkhaz side in the Georgian—Abkhaz war through the lens of mass Georgian displacement - a common view among observers (Amnesty International, 2010). I followed other scholars of civil war mobilization in assigning potential research participants the 'fighter' and 'non-fighter' categories.16 I expected to empathize with the latter but not the former and with 'victims' but not 'perpetrators' of Georgian displacement. As Baines (2009: 177) observes, the 'categories of "victim" and "perpetrator" are assigned a moral value in the field, where empathy is offered to "victims," while "perpetrators" are treated with fear, suspicion, and mistrust'.
My interviews soon revealed multiple roles within and outside of these pre-existing moral-conceptual categories as individuals combined and shifted between roles and developed overlapping identities beyond the victim-perpetrator divide. Some who fought also protected Georgian families, others who maintained neutrality during the war later endorsed violence, and women who lost their children transitioned between the roles as mothers and indirect supporters of the Abkhaz effort. This
Viterna (2006: 11), for example, identified and interviewed seven guerrilla and seven non-guerrilla women in each of the six communities selected for her study of mobilization in El Salvador.
challenged my expectations about empathy and my categories of 'non-fighter' and 'fighter', 'victim' and 'perpetrator'. I began empathizing to varying degrees with participants in the war in response to this complexity17 and broadened the spectrum of roles in my research.18
This helped me pose sensitive questions about respondents' participation decisions in follow-up to their common initial accounts of evil and cowardice on the Georgian side and good and bravery of the Abkhaz. I was able to ask about the role of Georgian neighbors, rejection of killing, and social repercussions of evasion if/when appropriate. Asking these questions would be difficult had I feared my respondents. In turn, my interest in and sensitivity to the responses prompted respondents to talk about the dilemmas they faced as their families were killed, cities surrounded, and status in Abkhazia threatened. This revealed issues of guilt, responsibility, and moral choice that challenged the official narrative of the Georgian offense against the Abkhaz and brought to bear the Abkhaz agency in the war. In my interpretation of these interviews, I was able to appreciate the conflict from this perspective.
In particular, some Abkhaz sought neutrality. A respondent who adopted this role to preserve friendship on both sides in the war and remained in Abkhazia thereafter, where most of his Georgian friends could not return, captures the dilemma of participation that polarized his networks: 'I took a neutral position and did not go to fight. I had friends on both sides and so if I had taken a gun I would become an enemy to one of the sides' (Interview 77, Fall 2011). The respondent expressed profound regret about this decision in the postwar setting: 'from my current viewpoint, it was not right. If you live here, [you should] defend the land, [even if b] oth sides fight for the interests of their layer of the population.' This acceptance of violence highlighted the changing nature of non-fighter neutrality.
Neutrality was also volatile in the context of wartime polarization and violence. 'There was general mobilization and pressure from both sides' (Interview 77, Fall 2011). A form that neutrality took in this context was defending all locals from violence. 'We made a decision to guard the village including the Georgian population', a fighter who adopted this role demonstrates; 'we all live in one village: all our women, children, elders live in this village. Allowing in these conditions to perpetrate
17 For example, I joined collective mourning with women who lost their loved ones yet supported the war.
18 I extend Petersen's (2001) spectrum of neutrality, unarmed opposition, direct support, and membership in rebellion.
198
journal of Peace Research 56(2)
violence against our neighbors was unacceptable' (Interview 117, Fall 2011). Individuals in this role opposed both the Abkhaz and Georgian armed actors. 'You can kill me but you must help me get the people I have here out of Abkhazia'; a fighter recalls threats by other Abkhaz for rescuing Georgians (Interview 134, Winter 2011). The risk intensified the dilemma of neutrality toward local Georgians.
As the war progressed, neutrality of any form became untenable. 'It was difficult to imagine [local] Georgians who could take up arms against Abkhaz neighbors, but when the Georgian [forces] were left with less and less mobilization resources they involved [them]' (Interview 117, Fall 2011). As a result, respondents found themselves fighting some Georgians while defending others: 'We hid Georgian families [...], but how could we not fight in this situation?' (Interview 117, Fall 2011). Many reported insubordination to Abkhaz orders due to perceived shared humanity with Georgian fighters, whether local or not: 'I was definitely supposed to kill him then but showed him with my eyes to go' (Interview 102, Fall 2011). Individuals thus oscillated between the roles of fighting and neutrality.
However, wartime roles extended beyond these options. Many Abkhaz evaded participation by fleeing or hiding in safe areas to avoid recruitment or social pressure to mobilize. One respondent, for instance, facilitated her son's evasion even if it meant that others would fight and die in his place: 'my 20-year-old son was serving in the Gudauta army [... Many] were hiding their sons there [... ] If volunteers came, they were taken to fight. These ones were not volunteers, however, so they remained [in the Abkhaz army reserve] and lived' (Interview 11, Fall 2011).
Others provided indirect support though social interactions, such as encouragement to fight. Parental blessing is a recurrent example in the Abkhaz case. 'My mother said, "Go to the end!" and I went', fighters say (Interview 9, Fall 2011). Yet others supported the fighting directly in medical, logistics, media, and engineering roles or 'gathered [funds] to provide for the boys' (Interview 85, Fall 2011).
Movement between these roles was a common feature, as in other civil wars (Petersen, 2001; Parkinson, 2013). One reason was formation of the army during the war. 'The Abkhaz population of Gagra had no structure when the war began', a local explains, '[Then leaders] started forming units' (Interview 75, Fall 2011). This shifted individuals between fighting and support roles and mobilized those who formerly were not. Another reason was everyday changes in the context of violence (Fujii, 2009). Individuals who initially maintained
neutrality joined the Abkhaz force after attacks on families and homes: 'The house my father built was burned [... ] Then I joined military actions' (Interview 87, Fall 2011). Those in support roles engaged in the fighting as the war intensified. Field engineers who laid roads to avoid attacks, made dugouts, and cleared minefields were also called to military operations, for example (Interview 14, Fall 2011). In the periods of ceasefire individuals transitioned from army obligations to duties toward families and friends - between combatant and civilian roles. 'I changed out of my military uniform into civilian clothing', a respondent illustrates, 'and went by foot across the bridge [to check on m]y mother, brother-in-law's wife, and nephews' (Interview 100, Fall 2011).
One outcome of these trajectories was the overlapping identities that individuals developed. Respondents rarely referred to the victim-perpetrator divide that I assumed at the outset of the study. They used the term victim concerning others who suffered from the war rather than themselves. The notion of suffering emerged consistendy, but as related to the broader Georgian-Abkhaz conflict that spans the Soviet era rather than the war of 1992-93 in conclusion of this long-term social conflict. At first, respondents emphasized the defensive nature of their actions during the war, reflected in the systematic use of the terms motherland, defense, and freeing to refer to the Abkhaz effort, juxtaposed with aggression, attack, and occupation to describe the Georgian force. 'We were defending against aggression' (Interview 128, Winter 2011), respondents say; 'They occupied the whole of Abkhazia' (Interview 126, Winter 2011). Many, then, acknowledged their part in violence and its repercussions for Georgians, both displaced residents of Abkhazia and fighters from Georgia proper. 'How many hundreds of [Georgian] boys died here' (Interview 127, Winter 2011), respondents admit; 'there are many refugees among them. They are angered, they cannot return home' (Interview 107, Fall 2011).
Adjusting my initial categories of roles and identities and expectations about empathy during fieldwork, displaying empathy for participants in the 1992-93 war and asking sensitive follow-up questions as appropriate, and interpreting individual participation trajectories in the war in light of their complexities thereafter, allowed me to develop insight beyond the official conflict narrative of Georgian aggression and Abkhaz defense - a critical advance from the dominant view of the case.
Fear and ongoing violence
However, developing empathy was not possible with some research participants. Individuals who voluntarily
Shesterinina
199
participated in the protracted hostilities along the Geor-gian-Abkhaz administrative border after the war invoked fear rather than empathy in the interviews. My fear stemmed from the research context and the respondents. One of the few researchers, I felt isolated in Abkhazia where entry and exit are difficult due to its de facto state status. Working unaffiliated with local partners added to the sense of isolation. This choice was ethically necessary, but it meant that I could not call on local partners had I faced threat or intimidation. While hostilities were concentrated in the border area, the threat of violence was present in the locales where I conducted my research given physical proximity of the border in a small territory and general volatility of security in postwar Abkhazia.19 Proximity of violence implied that my safety was not guaranteed and I forged contacts in the security sector to learn the location, time, and nature of violence, select safer interview sites, and devise exit strategies. I frequently experienced intimidation by men in power positions, especially in the spatial confines of the de facto Ministry of Defense. The checkpoint procedure at its entry and interrogation by the more junior officials that preceded my interviews with commanders created a sense of surveillance. High-level officials assumed my affiliation with intelligence agencies and asked me whether I was a spy, a common suspicion in the region (Driscoll, 2016). Some told me that my phone conversations were tapped and I avoided phone communication about my research with research participants and others.
I feared my interactions with respondents active in ongoing violence in this research context. As actors with access to coercive means, I expected these respondents to pose potential threat to my safety and make intimidating remarks.20 In general, I knew about their role prior to the interview and the dynamic of fear was intensified during the interview as they spoke about this role. Many bragged and expressed pleasure and excitement about violent activities, especially when they were prohibited from participation by established norms and procedures. For example, some engaged in the recurrent armed clashes and further civilian displacement after the war outside of the Abkhaz police {milicija) structure
For example, I observed armed men in transit and heard rumors of kidnappings. My room was searched in my absence. 20 Indeed, one of these respondents agreed to meet in a public location for the interview, but changed his mind and asked me to ride with him in his car to an alternative location that he preferred. I accepted the change but faced intimidation in the course of the trip. While this was an extreme case, other interviews involved lesser forms of threat and intimidation.
permitted in the border region by the United Nations (1994). 'I was not allowed [in the border area after the war]', one respondent illustrates, 'Only the milicija was allowed [... ] but I changed into the milicija uniform and went with them' (Interview 87, Fall 2011). Most justified this violence as the right thing to do against the displaced local Georgians returning to their homes in the border area. 'The [returning Georgian] population was not reliable', a respondent explains; 'It was a population torn by war. They were called to fight by Georgia and so once we arrived, we did horrors to them [... ] We went into houses, cleared them, creating a mass psychosis' (Interview 20, Fall 2011).
These responses appalled me. Gallaher (2009: 128) reports a similar reaction to such repellent responses: 'I felt angry and wanted to tell my informants just how wrong I thought they were'. But my fear of these individuals in the context of ongoing violence meant that I could not scrutinize their narratives. I was afraid of asking sensitive follow-up questions that could anger my respondents; the interview was restricted. This was in contrast to my interactions with participants in the 1992-93 war who did not voluntarily continue violent activities. The war was distant and that made it possible for us to look back on and evaluate it through my probes and respondents' exploratory departures, which shifted the conversation beyond the official conflict narrative. With those active in ongoing violence, however, it became clear that their positions were not likely to shift or expose 'spoken and unspoken thoughts and feelings' (Fujii, 2010: 232) and I treated these individuals with suspicion and mistrust.
These emotions carried into the interpretation, as I remembered my fear of these respondents and was not willing to appreciate any complexity of their position, reproducing the official narrative on the status of postwar Abkhazia and limitation on the return of the displaced Georgian population. Hence, participants in ongoing violence expressed a cemented position on Abkhazia's status that the war brought about and stressed the necessity of violence as a way to defend the Abkhaz state. 'Now that we freed Abkhazia, we had to defend it' (Interview 47, Fall 2011). Respondents showed pride in performing border guard duties and military operations to clear the border area of Georgian armed formations active after the war. 'We were not paid for that' (Interview 85, Fall 2011), border guards underline. 'Our cleaning operations were necessary to get rid of partisans' (Interview 25, Fall 2011). Respondents described hostilities in the border area as continued Georgian aggression and 'attempts at military revenge' (Interview 123, Fall
200
journal of Peace Research 56(2)
2011). 'In 1998, Georgians had a task to occupy and cut off the [border] region [... to] establish their jurisdiction there' (Interview 126, Winter 2011). 'Our army repulsed the occupation' (Interview 60, Fall 2011). They celebrate the Abkhaz success: 'We freed all the borders of Abkhazia in the Kodor offensive [during the Russo-Georgian war] in 2008. As a result, we restored the Abkhaz statehood' (Interview 75, Fall 2011).
Respondents demonstrated an equally strong position on the return of the displaced Georgian population. 'Those who ran understood they did not have an option to return peacefully' (Interview 131, Winter 2011). Many blamed displaced local Georgians for supporting Georgian armed groups. 'Not all locals took our side. At night, some changed into uniforms and went against us' (Interview 30, Fall 2011). The desire to punish these locals was a common theme in the interviews. 'There was a desire to go and punish them because we knew that it would not stop there' (Interview 148, Winter 2011). Thus 'the Abkhaz burned houses because Georgians hid there' (Interview 131, Winter 2011) and 'there were many casualties on the Georgian side' (Interview 148, Winter 2011). The sanitized language that respondents used highlighted their uncritical view of the events: 'We did not even say "destroy them" but "push them out'" (Interview 70, Fall 2011).
My inability to engage these accounts meant that they did not bypass the official narrative of the war, to delve into participation dilemmas. I interpreted their roles as critical to the persistence of violence in contemporary Abkhazia without appreciating the likely complexity of their participation. Their cemented position on Abkhaz statehood and Georgian displacement pointed to a justification of ongoing violence that these respondents were unwilling to critically reflect on and challenge. As in Green's (1994) and Blee's (2002) research, my own fear of these respondents was indicative of the normalization of violence in the context of isolation and protracted conflict in postwar Abkhazia.
Conclusion
What implications do these emotional responses have for scholars of violence and war? My interviews in Abkhazia challenge our propensity to afford empathy to all respondents except unloved or repellent groups and draw attention to selective empathy in research with armed actors. I empathized with participants in the 1992-93 war, whom I initially categorized as repellent, but not participants in ongoing violence, whom I feared in the volatile field context of Abkhazia. The ethical dilemma of selective empathy was evident in my differential treatment of
these respondents, namely, the interest in and sensitivity that I displayed for those I empathized with and suspicion and mistrust for those I feared. Common approaches to research ethics, the 'do no harm' imperative and 'give back' feminist ethics of care, do not capture these emotionally challenging and exclusionary practices of fieldwork where some research participants are afforded empathy while others are not.
These emotions shaped what questions I was able to ask, what respondents told me, and how I interpreted the results. Had I not empathized with wartime participants, I would likely have missed their complexity of roles, as illustrated by my interviews with participants in current violence whom I feared. Broadening the spectrum of roles to reflect the experiences of the former was a critical step to a greater understanding of participation in the changing conditions of the war, while reflection on the latter made apparent that fear limited my appreciation of their trajectories in the difficult research context. This shows that answering why we empathize with some but fear other research participants involved in past and present violence can illuminate why we ask sensitive questions selectively, how this impacts respondents' accounts, and what this tells us about our research contexts. Reflecting on how these emotions change in the course of fieldwork can help adjust our assumptions and analytical categories and gain insight that better reflects the experiences of research participants.
My interviews thus highlight the relational nature of emotions in research on violent conflict. Whether researchers are able to relate to or are appalled by respondents affects how we engage with them. On the one hand, our emotions allow us to probe respondents' accounts to varying extent. On the other hand, emotions influence our capacity for engaged listening, which allows respondents to surpass strong positions that they might hold, to explore dilemmas and uncertainties. As researchers get immersed in contexts of deep social conflict to answer important questions about participation in violence and war (Fujii, 2009; Parkinson, 2013; Viterna, 2006; Wood, 2003), reflection on the roles of empathy and fear in shaping our interactions with research participants can lead to findings beyond the pre-existing moral-conceptual categories of 'victim' and 'perpetrator', 'non-fighter' and 'fighter', and advance our understanding of political violence beyond dominant conflict narratives.
Replication data
An Online appendix can be found at http://www.prio. org/jpr/datasets.
Shesterinina
201
Acknowledgements
The author benefited greatly from comments by Peregrine Schwartz-Shea, participants in the APSA 2015 panel on Ethical and Methodological Innovation in Fieldwork, Paolo Sandro, and three anonymous reviewers. Any errors or omissions are the responsibility of the author.
Funding
Support for research and writing is acknowledged from the Canadian Social Sciences and Humanities Research Council, the MacMillan Center at Yale University, the University of British Columbia, and the Liu Institute of Global Issues.
ORCID iD
Anastasia Shesterinina © https://orcid.org/0000-0002-3779-4903
References
Amnesty International (2010) In the waiting room: Internally displaced people in Georgia (https://www.amnesty.org/en/ documents/EUR56/002/2010/en/).
Baines, Erin K (2009) Complex political perpetrators: Reflections on Dominic Ongwen. Journal of Modern African Studies 47'(2): 163-191.
Bayard de Volo, Lorraine (2009) Participant-observation, politics, and power relations: Nicaraguan mothers and US casino waitresses. In: Edward Schatz (ed.) Political Ethnography: What Immersion Contributes to the Study of Power. Chicago, IL: University of Chicago Press, 217-236.
Blee, Kathleen M (1993) Evidence, empathy, and ethics: Lessons from oral histories of the Klan. Journal of American History 80(2): 596-606.
Blee, Kathleen M (1998) White-knuckle research: Emotional dynamics in fieldwork with racist activists. Qualitative Sociology 21(4): 381-399.
Blee, Kathleen M (2002) Inside Organized Racism: Women in the Hate Movement. Berkeley, CA: University of California Press.
Brouneus, Karen (2011) In-depth interviewing: The process, skill and ethics of interviews in peace research. In: Magnus Oberg & Kristine Hoglund (eds) Understanding Peace Research: Methods and Challenges. Abingdon: Routledge, 130-145.
Campbell, Susanna P (2017) Ethics of research in conflict environments. Journal of Global Security Studies 2(1): 89-101.
Cameron, Lynne & Bruna I Seu (2012) Landscapes of empathy: Spatial scenarios, metaphors and metonymies in responses to distant suffering. Text & Talk 32(3): 281-305.
Coppieters, Bruno; David Darchiashvili & Natela Akaba (2000) Federal Practice: Exploring Alternatives for Georgia and Abkhazia. Brussels: VUB University Press.
Crawford, Neta C (2014) Institutionalizing passion in world politics: Fear and empathy. International Theory 6(3): 535-557.
Cronin-Furman, Kate & Milli Lake (2018) Ethics abroad: Fieldwork in fragile and violent contexts. Political Science & Politics 51(3): 607-614.
de Laine, Marlene (2000) Fieldwork Participation and Practice: Ethics and Dilemmas in Qualitative Research. London: Sage.
Diphoorn, Tessa (2013) The emotionality of participation: Various modes of participation in ethnographic fieldwork on private policing in Durban, South Africa. Journal of Contemporary Ethnography 42(2): 201—225.
Driscoll, Jesse (2016) Warlords and Coalition Politics in Post-Soviet States. Cambridge: Cambridge University Press.
Fielding, Nigel G (1990) Mediating the message: Affinity and hostility in research on sensitive topics. American Behavioral Scientist 33(5): 608-620.
Fujii, Lee Ann (2009) Killing Neighbors: Webs of Violence in Rwanda. Ithaca, NY: Cornell University Press.
Fujii, Lee Ann (2010) Shades of truth and lies: Interpreting testimonies of war and violence. Journal of Peace Research 47(2): 231-241.
Fujii, Lee Ann (2012) Research ethics 101: Dilemmas and responsibilities. PS: Political Science and Politics 45(4): 717-723.
Fujii, Lee Ann (2018) Interviewing in Social Science Research: A Relational Approach. Abingdon: Routledge.
Gallaher, Carolyn (2009) Researching repellent groups: Some methodological considerations on how to represent militants, radicals, and other belligerents. In: Chandra Lekha Sriram, John C King, Julie A Mertus, Olga Martin-Ortega & Johanna Herman (eds) Surviving Field Research: Working in Violent and Difficult Situations. Abingdon: Routledge, 127-146.
Green, Linda (1994) Fear as a way of life. Cultural Anthropology 9(2): 227-256.
Hewitt, George B (1996) Abkhazia: A problem of identity and ownership. In: John FR Wright, Suzanne Goldenberg & Richard N Schofield (eds) Transcaucasian Boundaries. London: UCL Press, 190-226.
Hollan, Douglas W & C Jason Throop (2011) The anthropology of empathy: Introduction. In: Douglas W Hollan & C Jason Throop (eds) The Anthropology of Empathy: Experiencing the lives of Others in Pacific Societies. New York: Berghahn, 1-24.
Malejacq, Romain & Dipali Mukhopadhyay (2016) The 'tribal politics' of field research: A reflection on power and partiality in 21st-century warzones. Perspectives on Politics 14(4): 1011-1028.
Mazurana, Dyan; Karen Jacobsen & Lacey Andrews Gale (2013) Research Methods in Conflict Settings: A View from Below. Cambridge: Cambridge University Press.
202
Journal of Peace Research 56(2)
National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research (1979) The Belmont Report: Ethical Principles and Guidelines for the Protection of Human Subjects of Research. US Department of Health, Education, and Welfare. Washington, DC: US Government Printing Office.
Nordstrom, Carolyn & Antonius CGM Robben (eds) (1995) The anthropology and ethnography of violence and sociopolitical conflict. In: Carolyn Nordstrom & Antonius CGM Robben (eds) Fieldwork Under Fire: Contemporary Studies of Violence and Survival. Berkeley, CA: University of California Press, 1—24.
O'Loughlin, John; Vladimir Kolossov & Gerard Toal (2011) Inside Abkhazia: Survey of attitudes in a de facto state. Post-Soviet Affairs 27'(1): 1-36.
Parkinson, Sarah E (2013) Organizing rebellion: Rethinking high-risk mobilization and social networks in war. American Political Science Review 107(3): 418-432.
Parkinson, Sarah E & Elisabeth J Wood (2015) Transparency in intensive research on violence: Ethical dilemmas and unforeseen consequences. Qualitative & Multi-Method Research 13(1): 22-27.
Petersen, Roger D (2001) Resistance and Rebellion: Lessons from Eastern Europe. New York: Cambridge University Press.
Pierce, Jennifer L (1995) Gender Trials: Emotional Lives in Contemporary Law Firms. Berkeley, CA: University of California Press.
Robben, Antonius CGM (1995) The politics of truth and emotion among victims and perpetrators of violence. In: Carolyn Nordstrom & Antonius CGM Robben (eds) Fieldwork Under Fire: Contemporary Studies of Violence and Survival. Berkeley, CA: University of California Press, 81-104.
Rubin, Herbert J & Irene S Rubin (1995) Qualitative Interviewing: The Art of Hearing Data. Thousand Oaks, CA: Sage.
Schatz, Edward (2009) Ethnographic immersion and the study of politics. In: Edward Schatz (ed.) Political Ethnography: What Immersion Contributes to the Study of Power. Chicago, IL: University of Chicago Press, 1-22.
Sehgal, Meera (2009) The veiled feminist ethnographer: Field-work amongst women of India's Hindu Right. In: Martha Huggins & Marie-Louise Glebbeek (eds) Women Fielding Danger: Negotiating Ethnographic Ldentities in Field Research. Boulder, CO: Rowman & Littlefield, 325-352.
Shesterinina, Anastasia (2016) Collective threat framing and mobilization in civil war. American Political Science Review 110(3): 411-427.
Sluka, Jeffrey A (2012) Staying alive while conducting primary research: Fieldwork on political violence. In: Marie Breen-Smyth (ed.) Ashgate Research Companion to Political Violence. Abingdon: Ashgate, 301-325.
Sriram, Chandra Lekha; John C King, Julie A Mertus, Olga Martin-Ortega & Johanna Herman (eds) (2009) Surviving Field Research: Working in Violent and Difficult Situations. Abingdon: Routledge.
Thomson, Susan (2009) 'That is not what we authorized you to do . . .': Access and government interference in highly politicised research environments. In: Chandra Lekha Sriram, John C King, Julie A Mertus, Olga Martin-Ortega & Johanna Herman (eds) Surviving Field Research: Working in Violent and Difficult Situations. Abingdon: Routledge, 108-123.
Thomson, Susan (2010) Getting close to Rwandans since the genocide: Studying everyday life in highly politicized research settings. African Studies Review 53(3): 19-34.
Thomson, Susan; An Ansoms & Jude Murison (2013) Emotional and Ethical Challenges for Field Research in Africa: The Story behind the Findings. London: Palgrave Macmillan.
Trier, Tom; Hedvig Lohm & David Szakonyi (2010) Under Siege: Lnter-Ethnic Relations in Abkhazia. New York: Columbia University Press.
United Nations (1994) Security Council Resolution 937 {SI RES/937) (https://documents-dds-ny.un.org/doc/ UNDOC/GEN/N94/298/25/PDF/N9429825.pdf? OpenElement).
Viterna, Jocelyn (2006) Pulled, pushed, and persuaded: Explaining women's mobilization into the Salvadoran guerrilla army. American Journal of Sociology 112(1): 1-45.
Wood, Elisabeth J (2003) Insurgent Collective Action and Civil War in El Salvador. Cambridge: Cambridge University Press.
Wood, Elisabeth J (2006) The ethical challenges of field research in conflict zones. Qualitative Sociology 29(1): 272-286.
Yanow, Dvora & Peregrine Schwartz-Shea (2006) Interpretation and Method: Empirical Research Methods and the Interpretive Turn. New York: ME Sharpe.
Yanow, Dvora & Peregrine Schwartz-Shea (2013) Interpretive Research Design: Concepts and Processes. London: Routledge.
ANASTASIA SHESTERININA, b. 1986, PhD in Political Science (University of British Columbia, 2014); Lecturer (Assistant Professor) in Politics/International Politics at the University of Sheffield (2017- ); field-based research on the Georgian-Abkhaz war of 1992-93 explores the internal dynamics of armed conflict with a focus on participation dilemmas; parallel research looks at the norms and practices of civilian protection in armed conflict.