From Data to Conditions of Knowledge: A Trauma-Informed, Non-Extractive Methodology for Multi-Context Research on War Rape Survivors ()
1. Introduction
Research on conflict-related sexual violence has generated a substantial body of qualitative literature focused on survivor experiences, trauma, memory, and postwar consequences. Interviews remain one of the central methodological approaches within this field and are frequently described as the product of clearly defined research designs characterized by stable procedures, coherent sampling strategies, and relatively controlled conditions of data collection (Ellsberg & Heise, 2005; Wood, 2006). These descriptions often fail to fully capture the realities of conducting qualitative research in environments shaped by war, forced displacement, institutional instability, social fragmentation, and ongoing political uncertainty. Access to participants is frequently unstable and shaped by changing conditions of safety, relational trust, institutional mediation, cultural expectations, and the ethical complexities inherent in engaging with survivors of extreme violence (Fujii, 2012; Guillemin & Gillam, 2004).
Qualitative data based on survivor testimonies are particularly valuable because they provide access to dimensions of violence, memory, silence, and social experience that cannot be fully captured through quantitative methods or institutional records. Survivor narratives illuminate how trauma is lived, interpreted, negotiated, and remembered within specific relational, political, historical, and cultural conditions.
Research conducted in these contexts unfolds under conditions that are neither neutral nor methodologically uniform. Interviews emerge through ongoing negotiations of trust, timing, vulnerability, and perceived risk, and the resulting narratives carry the imprint of those negotiations. Survivors do not speak within a social or relational vacuum. Their testimonies take shape within specific institutional, political, interpersonal, linguistic, and cultural environments that influence not only what can be articulated but also how experiences become narratable in the first place (Fujii, 2012; Miller, 2004).
Narrative structure, emotional intensity, silence, fragmentation, and disclosure practices are therefore inseparable from the conditions under which testimony is produced. Yet despite increasing recognition of ethical complexity within trauma research, methodological discussions continue to focus predominantly on procedural dimensions of qualitative inquiry—including interview guides, recruitment techniques, coding strategies, and trauma-informed interviewing practices—while giving comparatively limited attention to the broader conditions of narration and knowledge production themselves (Temple & Young, 2004; Squires, 2009).
At the same time, critiques of extractive research practices have become increasingly influential within scholarship addressing violence, inequality, humanitarian intervention, and postcolonial relations of power (Kovach, 2009; Smith, 2012). These critiques challenge the assumption that knowledge can be separated from the relational and political conditions through which it is generated. Within research involving survivors of conflict-related sexual violence, this concern acquires particular urgency because participation may expose individuals to risks extending far beyond the immediate interview encounter itself, including retraumatization, social exposure, institutional vulnerability, stigmatization, or political consequence (World Health Organization [WHO], 2001). Decisions concerning whether, when, and how survivors choose to engage in conversation are therefore shaped not only by methodological design but also by ongoing ethical assessments related to safety, recognition, emotional regulation, and anticipated consequences of disclosure. Under such conditions, testimony cannot be understood as transparent access to experience independent from the environments in which narration occurs.
The present study engages directly with these methodological and ethical tensions by examining how knowledge about war rape survivors becomes possible across heterogeneous conflict and post-conflict settings. Drawing on a cumulative dataset of 402 qualitative interview and testimony items collected between 2007 and 2026 across Central and East Africa, the Balkans, Eastern Europe, Latin America, Asia, and diaspora contexts in Western Europe and North America, the article develops a trauma-informed, non-extractive methodological framework for multi-context qualitative research on conflict-related sexual violence. The material encompasses formal research interviews, clinical narratives, humanitarian and programmatic encounters, survivor-initiated testimonies, and field-based disclosures emerging through memorial visits, community interaction, and longitudinal relational engagement. This diversity is not approached as a methodological weakness requiring correction or standardization. Rather, it is treated as an intrinsic characteristic of qualitative trauma research conducted under unstable political, institutional, relational, and historical conditions.
Differences in narrative form are central to this inquiry. Some testimonies are extended, layered, emotionally reflective, and temporally expansive. Others are highly condensed, fragmented, structured, or carefully controlled, at times reduced to a small number of factual statements. These variations do not necessarily correspond to communicative ability, educational background, memory failure, or willingness to disclose. Instead, they frequently reflect differences in contextual conditions: perceived safety, anticipated institutional use of testimony, relational trust, linguistic mediation, emotional regulation, political recognition, social stigma, and the broader environment surrounding the interaction itself (Pavlenko, 2005; Dewaele & Costa, 2013). Narrative depth therefore cannot be treated as a stable indicator of credibility, psychological processing, or testimonial completeness. It highlights the social, political, and emotional conditions that shape the possibility of narration. The article argues that methodological heterogeneity within research on conflict-related sexual violence should not be interpreted as a deviation from methodological rigor but as a structural consequence of the conditions under which testimony emerges. Conventional qualitative frameworks frequently assume relatively stable interview environments, coherent procedural structures, and reproducible conditions of participant engagement. Such assumptions become difficult to sustain in trauma-centered fieldwork involving war rape survivors, where disclosure practices are shaped by displacement, institutional mediation, multilingual communication, ongoing insecurity, and the uneven possibility of trust itself. Variability in testimony—including silence, fragmentation, temporal inconsistency, narrative compression, and shifting forms of disclosure—is therefore approached not as methodological failure but as analytically meaningful evidence concerning the conditions of knowledge production.
To engage systematically with this complexity, the study develops a trauma-informed and non-extractive methodological framework grounded in the recognition that qualitative data are generated through situated interactions shaped by ethical, relational, neurobiological, political, and historical forces. The article introduces a tri-axial classification structure organized across temporal-historical, contextual-political, and interview-typological dimensions. This framework enables the interpretation of heterogeneous qualitative material without erasing the contextual specificity of individual encounters. Rather than seeking methodological uniformity, the framework prioritizes contextual interpretation, relational ethics, and attentiveness to the conditions under which survivor testimony becomes possible.
The shift proposed here is both methodological and epistemological. Attention moves away from understanding qualitative data as neutral information collected from participants and toward an understanding of testimony as situated knowledge emerging under specific relational, institutional, linguistic, and political conditions. From this perspective, methodological rigor does not depend upon eliminating variability but upon the capacity to interpret variability itself as structurally meaningful. Such an approach opens space for forms of qualitative inquiry that remain responsive to complexity while acknowledging that heterogeneity, silence, fragmentation, and testimonial adaptation are not peripheral methodological problems to be corrected but constitutive features of trauma-centered research in contexts shaped by violence and displacement.
2. Neurobiological Foundations of Trauma and Their Methodological Implications
Research on conflict-related sexual violence frequently relies on narrative data while assuming that experiences can be elicited, structured, and interpreted within stable communicative frameworks. This assumption becomes difficult to sustain when considered in relation to the neurobiological effects of trauma. Exposure to extreme violence affects memory formation, emotional regulation, and the capacity for verbal articulation, shaping how experiences are encoded, stored, and later communicated. These processes have direct methodological implications for qualitative research, particularly in how data are generated and interpreted.
Recent neurobiological research demonstrates that traumatic experiences are associated with alterations in neural systems involved in memory and affect regulation. Stress-related processes influence the interaction between the amygdala, hippocampus, and prefrontal cortex, affecting both encoding and retrieval (Davis & Hamner, 2024). These changes do not produce stable, linear memory traces. Instead, trauma recall is often characterized by individualized and disorganized neural patterns that differ from those associated with non-traumatic memory. Evidence further suggests that trauma-related memories may function as a distinct form of cognitive representation, marked by fragmentation, intrusion, and discontinuity rather than coherent narrative structure.
These findings challenge methodological assumptions that treat narrative coherence as an indicator of reliability or completeness. In qualitative research, accounts that are non-linear, fragmented, or inconsistent are often interpreted as methodological limitations or as evidence of memory distortion. Neurobiological evidence suggests that such features are consistent with the way traumatic experiences are encoded and retrieved. Fragmentation, temporal disjunction, and variability in recall are not peripheral characteristics of the data; they are intrinsic to the processes through which traumatic memory operates.
The effects of stress on memory extend beyond initial encoding. Research indicates that retrieval processes are also shaped by contextual and physiological conditions, leading to variability in how experiences are recalled across time and settings (Schwabe, 2025). Memory is not a fixed record but a dynamic process that is reconstructed in relation to current conditions. This variability has direct implications for qualitative interviews, where narratives are produced within specific relational and environmental contexts. Differences in narrative form across interviews, including shifts in detail, structure, and emphasis, reflect not only individual experience but also the conditions under which recall occurs.
Mechanisms of memory suppression and control further influence narrative expression. Studies indicate that trauma can affect the ability to access or regulate memory, shaping the extent to which individuals engage with specific aspects of their experience (Mary et al., 2020). In interview settings, this may be reflected in selective disclosure, avoidance of certain topics, or the presentation of highly condensed accounts. Such patterns are often interpreted as reluctance or inconsistency. Within a neurobiological framework, they can be understood as regulatory processes that manage the interaction between memory, emotion, and perceived safety.
The relationship between trauma and language introduces an additional layer of complexity. Under conditions of stress, the capacity for verbal articulation may be altered, with greater reliance on sensory or affective forms of memory. This affects how experiences are translated into narrative form, particularly in contexts where participants are asked to provide detailed verbal accounts. The expectation of linear, descriptive narration may not align with the ways in which traumatic experiences are stored and accessed.
These dynamics intersect with the conditions of the interview itself. Factors such as perceived safety, relational trust, and environmental stability influence the activation of neurobiological systems involved in memory retrieval. When participants experience a sense of safety, there may be greater capacity for reflective processing and narrative integration. In contexts perceived as uncertain or threatening, responses may be shaped by heightened arousal or protective regulation, resulting in constrained or fragmented narratives. The interview setting therefore becomes part of the neurobiological context in which memory is accessed and expressed.
Silence and minimal disclosure can also be understood within this framework. Decisions not to elaborate may reflect regulatory processes aimed at maintaining emotional stability or avoiding reactivation of distress. These responses are frequently interpreted in qualitative research as absence of data or lack of engagement. Neurobiological perspectives suggest that they may represent adaptive strategies shaped by the interaction between memory systems and current conditions. Methodologically, this requires a shift in interpretation, recognizing silence as meaningful rather than deficient.
Linguistic factors further shape these processes. Interviews conducted in a shared non-native language, or through interpreters, introduce both distance and constraint. Research indicates that emotional processing may differ across languages, with second-language use sometimes associated with reduced affective intensity (Dewaele & Costa, 2013; Pavlenko, 2005). This linguistic distance can facilitate disclosure by creating a degree of separation from the emotional immediacy of the experience. At the same time, it may limit the expression of nuance and culturally specific meaning. The resulting narratives reflect the interaction between neurobiological processes and linguistic mediation.
These insights do not replace qualitative methodology. They refine it. They indicate that features such as fragmentation, variability, and silence should not be treated as methodological problems to be corrected. They are expected outcomes of the interaction between trauma, memory, and context. Incorporating neurobiological perspectives into methodological design supports an approach that aligns analytical expectations with the conditions under which narratives are produced.
The implications for this study are direct. The variability observed across interviews is consistent with the neurobiological organization of traumatic memory and with the conditions of field engagement described in subsequent sections. Narrative form cannot be evaluated independently of these factors. A methodology that recognizes the relationship between trauma, context, and narration provides a more accurate basis for interpretation. This perspective underpins the analytical framework developed in the following sections, where qualitative data are examined in relation to the conditions of their production rather than against fixed standards of coherence or completeness.
3. Dataset as Methodological Object
The present study approaches the dataset itself as the primary object of methodological analysis. The material consists of approximately 402 qualitative interviews and testimony-based encounters conducted across multiple geopolitical and historical contexts, including Central and East Africa, the Balkans, Eastern Europe, Latin America, parts of Asia, and Western Europe. Rather than emerging from a single standardized research design, the dataset developed cumulatively over time through different forms of engagement with survivors of conflict-related sexual violence, including formal research interviews, clinical encounters, humanitarian and programmatic work, survivor-initiated disclosures, and field-based interactions. The material spans distinct historical periods ranging from post-Second World War legacies to contemporary conflicts, each shaped by differing conditions of political recognition, institutional response, collective memory, and social silence.
The dataset does not conform to conventional expectations of methodological uniformity commonly associated with qualitative research. Interviews varied substantially in duration, structure, relational dynamics, setting, linguistic mediation, and institutional context. Some narratives developed gradually across repeated encounters, while others consisted of highly compressed disclosures emerging under conditions of uncertainty, displacement, or perceived risk. Certain testimonies unfolded within structured or semi-structured interview frameworks, whereas others emerged spontaneously through relational interaction or situational dynamics. Rather than treating such variation as methodological inconsistency requiring correction, the present study interprets narrative heterogeneity as analytically meaningful and inseparable from the conditions under which disclosure became possible. Qualitative scholarship has long emphasized that interviews are co-produced through interaction, positionality, and context (Fujii, 2012; Guillemin & Gillam, 2004). Within research involving conflict-related sexual violence, these dynamics become intensified through issues of stigma, safety, institutional power, displacement, and the potential consequences of disclosure (Ellsberg & Heise, 2005; Wood, 2006).
The dataset incorporates material generated across several domains of practice, each associated with distinct relational and institutional conditions. Research interviews were conducted within formally defined projects organized around specific thematic or geopolitical concerns. Clinical encounters emerged within therapeutic relationships in which narratives unfolded gradually through processes of trust, stabilization, and longitudinal engagement. Humanitarian and programmatic interactions developed within service-oriented settings where experiences of violence were often embedded within broader discussions concerning survival, displacement, access to support, and social reintegration. Survivor-initiated testimonies emerged outside formal recruitment structures, frequently through participants’ own decisions to disclose experiences in response to relational trust or perceived ethical safety. Each of these modes of engagement produced distinct testimonial forms shaped by differing expectations, risks, institutional environments, and anticipated consequences of narration.
In addition to formally structured interviews, approximately 30% of the dataset emerged through field-based and relational encounters occurring during memorial visits, humanitarian engagement, site-based interaction, and community dialogue in locations directly connected to histories of mass violence. These interactions took place within memorial spaces, local institutions, community centers, and post-conflict environments where disclosure emerged through shared presence, spatial proximity, and relational interaction rather than predefined interview procedures. Conversations with memorial workers, community members, local staff, and survivors frequently evolved into survivor-centered narratives or contextual testimonies concerning violence, loss, memory, and post-conflict social conditions. Similar encounters emerged across multiple geopolitical settings in which participants chose to speak in response to the symbolic, historical, or affective significance of the environment itself.
Testimonies generated through such encounters differed significantly from those produced within formally structured interview settings. They were typically unscripted, interactional, and relationally negotiated rather than procedurally organized through predefined questioning. Both narrative form and disclosure depth were shaped by the spatial, historical, emotional, and institutional conditions within which interaction occurred. Memorial and post-conflict environments functioned not merely as passive research locations but as active spaces in which memory could be preserved, activated, negotiated, and constrained simultaneously. Spatial context therefore influenced not only what could be articulated but also the conditions under which narration became psychologically and socially possible. Research on political violence has demonstrated that testimony is co-produced through relational negotiation and situational dynamics (Fujii, 2012). The present study extends this understanding by conceptualizing field-based and place-based encounters as distinct modes of qualitative data generation in which relational, spatial, affective, and symbolic conditions play constitutive methodological roles.
The inclusion of such material required careful methodological and ethical consideration. Informal or spontaneous interactions are often excluded from qualitative datasets because they do not conform to expectations of procedural standardization, replicability, or controlled interview conditions. However, excluding such encounters risks narrowing what is recognized as valid qualitative evidence by privileging only testimonies produced within formalized research structures. The present study adopts a different methodological position. Field-based encounters are treated as integral to understanding how knowledge about conflict-related sexual violence emerges across unstable and heterogeneous environments. Such interactions frequently provide access to forms of testimony that might not emerge within formal interview contexts, particularly where participation carries political, social, institutional, or psychological risk.
The decision to include these testimonies is grounded in a trauma-informed and non-extractive methodological orientation. Within this framework, qualitative data are not treated as neutral objects collected independently from the conditions of their production. Rather, testimony is understood as situated knowledge emerging through relationships shaped by ethics, context, institutional mediation, power asymmetries, linguistic conditions, and perceptions of safety. Decolonial and Indigenous methodological scholarship has emphasized that knowledge production cannot be separated from questions of relational accountability, representation, and responsibility (Kovach, 2009; Smith, 2012). These concerns become especially significant within research involving survivors of violence because participation itself may expose individuals to retraumatization, institutional vulnerability, social stigma, or political consequence. Engagement must therefore remain attentive not only to procedural consent but also to autonomy, contextual sensitivity, emotional readiness, and the broader consequences of disclosure.
By positioning the dataset, itself as a methodological object, the study shifts analytical attention away from data collection alone and toward the conditions under which testimony and knowledge become possible. From this perspective, narrative variability, fragmentation, silence, compression, testimonial adaptation, and contextual instability are interpreted not as methodological failures but as constitutive features of qualitative trauma research conducted within environments shaped by violence, displacement, institutional uncertainty, and political instability. The framework developed in the following sections builds upon this position by proposing a systematic structure for engaging heterogeneous qualitative material without erasing the specificity of the encounters through which it emerged.
Analytic Workflow
Each dataset item was reviewed according to three analytic axes: temporal-historical context, contextual-political environment, and interview typology. Metadata recorded for each item included conflict or post-conflict setting, approximate period of violence, date or period of encounter, mode of engagement, location, institutional context, language or translation mode, presence of interpreter, degree of structure, longitudinal or single-encounter status, and ethical pathway. Items were assigned to categories according to the dominant conditions shaping the encounter. Where cases occupied more than one category, the primary classification reflected the condition most strongly shaping disclosure, while secondary features were retained analytically. Ambiguous cases were not forced into rigid categories but treated as methodologically significant examples of overlap between research, clinical, humanitarian, and field-based modes of knowledge production.
Distribution of the 402 qualitative dataset items included in the study across research interviews, clinical narratives, humanitarian and programmatic encounters, survivor-initiated testimonies, field-based encounters, interpreter-mediated interviews, shared non-native language interviews, and longitudinal or repeated engagements. The figure illustrates the methodological heterogeneity of the corpus and the multiple relational and institutional conditions under which testimony emerged.
To support the analysis of 402 interviews conducted across diverse conflict and post-conflict settings, this study introduces a Tri-Axial Classification Framework that accounts for the conditions under which testimony emerged. The framework organizes interviews according to three interrelated dimensions: temporal-historical context, contextual-political setting, and interview typology. Consistent with a trauma-informed and non-extractive methodology, differences in narrative form, depth, and dis-closure are interpreted as reflections of the conditions shaping testimony rather than as indicators of data quality or methodological inconsistency (Figure 1).
4. Case Illustrations: Conditions of Testimonial
Trauma-Approach
The following case illustrations are included not as anecdotal supplements to the methodology but as integral demonstrations of the conditions under which testimony emerged across different conflict and post-conflict environments. Rather than functioning as illustrative examples detached from the methodological framework, these encounters reveal how disclosure developed relationally through trust, timing, emotional regulation, symbolic space, institutional context, and perceptions of safety. In each setting, testimony emerged unevenly and unpredictably, shaped not by standardized interview procedures alone but by evolving interpersonal and environmental conditions that influenced whether narration became psychologically, socially, and ethically possible.
Case Illustration A: Relational Trust, Field Presence, and the Conditions of Disclosure
During fieldwork conducted in Rwanda and the Democratic Republic of Congo in 2018, I visited the Kigali Genocide Memorial Centre to review archival materials and deepen contextual understanding of sexual violence during the genocide.
Figure 1. Overview of corpus composition.
Access to the memorial environment was not established through formal institutional recruitment alone. Prior to the visit, I had been introduced to members associated with the memorial through earlier contact with a genocide survivor diaspora in the United States. One individual who had previously worked for many years within the memorial environment facilitated my introduction and remained connected to the institution. This relational pathway significantly shaped the conditions of access. I did not arrive as an unknown external researcher entering the space without prior connection. Rather, I entered through an already existing network of trust and recommendation. My presence had, in part, been socially validated before my arrival through individuals already recognized within the memorial community.
This relational positioning became methodologically significant during the fieldwork itself. After several days at the memorial site, I requested technical assistance related to computer access and archival materials. During this interaction, the staff member assisting me gradually began speaking about personal experiences connected to the genocide. The interaction was not initiated through a formal interview protocol, nor did it emerge from direct questioning concerning personal trauma. Instead, disclosure developed relationally within the shared environment of the memorial itself. We moved into a quiet room where the conversation evolved into an extended testimony lasting more than one hour, involving highly painful details, emotional reflection, and declarations connected to experiences of violence and loss during the genocide.
Several days later, upon returning to the memorial center, another individual approached me directly and asked why I was collecting these stories and what I intended to do with them. The conversation gradually shifted from questions about my role toward personal disclosure, eventually leading to the sharing of additional narratives and introductions to surviving family members. Again, the interaction did not emerge through structured recruitment procedures but through relational negotiation, contextual trust, and the symbolic meaning attached to the memorial environment itself.
These cases illustrate several methodological conditions central to field-based qualitative research on conflict-related sexual violence and genocide-related trauma. First, readiness to disclose highly traumatic experiences was not determined solely by the presence of a researcher or by formal interview structure. Disclosure appeared to depend on multiple interacting variables, including prior recommendation, relational trust, perceived ethical intention, social validation through known intermediaries, and the presence of an external witness perceived as willing to listen and carry testimony beyond the immediate local environment. The fact that introductions occurred through trusted relational networks significantly altered the social conditions under which narration became possible.
Second, the memorial space itself functioned as an active methodological environment rather than a neutral backdrop. The site carried historical, symbolic, and emotional significance that shaped both interaction and memory activation. Testimonies emerged not simply through questioning but through shared presence within a space dedicated to remembrance, mourning, and historical witnessing. The environment contributed to the movement from ordinary interaction into trauma narration.
Third, this example demonstrates how qualitative data in field environments frequently emerge outside formalized interview procedures. The testimonies were unscripted, relationally negotiated, and contingent upon evolving interpersonal dynamics rather than predefined methodological control. Such encounters challenge narrow distinctions between “formal interview” and “field interaction,” demonstrating that knowledge production in post-conflict environments often depends on trust-building processes that cannot be fully standardized or predicted in advance.
Finally, the case highlights the importance of understanding testimony not as information extracted from participants but as disclosure emerging under particular relational and ethical conditions. The narratives became possible through accumulated trust, contextual positioning, perceived safety, and the recognition of the researcher as an outside witness capable of listening without immediate institutional judgment or local political entanglement. From a methodological perspective, the interaction illustrates how field-based testimony is co-produced through environment, relational mediation, symbolic space, and the social conditions surrounding disclosure itself.
Case Illustration B: Trauma-Informed Methodology, Survivor Readiness, and the Ethics of Listening in Postwar Kosovo
During fieldwork conducted in Kosovo in 2021, I met with survivors of wartime sexual violence related to the 1998-1999 conflict, as well as local NGO professionals engaged in long-term advocacy and psychosocial support for survivors. In addition to formal institutional meetings, I participated in several community and social gatherings organized within local networks connected to postwar survivor support initiatives. These informal settings became methodologically important because they created relational conditions through which observation, trust, and familiarity could gradually develop outside formal interview structures.
Following one community event held in a public social environment, one participant approached me privately and stated that they felt ready to share their experiences related to wartime sexual violence. Importantly, this disclosure did not emerge through scheduled recruitment procedures, direct questioning, or prearranged testimony collection. Rather, the participant-initiated contact after observing my interactions within the community and after becoming familiar with the broader purpose of my work. The approach reflected not only willingness to speak but also the participant’s own assessment of relational safety, emotional readiness, and trust.
From a trauma-informed methodological perspective, the decision to proceed with the testimony required prioritizing the participant’s control over the interaction instead of prioritizing data collection itself. The conversation initially focused on determining whether the participant wished to continue speaking, as well as when and under what conditions they felt comfortable doing so, before proceeding with a formal interview. Together we identified a quieter and more private setting that would allow the participant to maintain agency over pacing, emotional boundaries, and the direction of disclosure. The participant remained informed throughout the process that they could pause, redirect, limit, or stop the conversation at any point without consequence.
The resulting testimony extended over approximately two hours and included highly emotional reflections concerning rape during the war, pregnancy resulting from sexual violence, child adoption, social stigma, and the long-term consequences of survival in the postwar environment. The narrative did not unfold in a linear or uninterrupted manner. Periods of detailed narration alternated with silence, emotional pauses, hesitation, and shifts away from traumatic material toward broader reflections about life after the war. From a trauma-informed perspective, these interruptions were not treated as obstacles to testimony or as indicators of inconsistency. Rather, they were understood as part of the survivor’s ongoing regulation of emotional safety during disclosure.
The methodological significance of this encounter lies precisely in the fact that the testimony emerged outside rigid procedural scheduling. During this field visit, I was not conducting systematic prearranged testimony collection according to fixed recruitment timelines. Instead, the disclosure emerged because the participant independently determined that sufficient trust and safety had developed to make narration possible. In this context, the role of the researcher shifted away from extracting information and toward creating conditions under which testimony could occur ethically and without coercive pressure.
This case illustrates several core principles of trauma-informed methodology in research involving survivors of sexual violence. First, survivor readiness cannot be assumed or procedurally produced through interview design alone. The willingness to disclose traumatic experiences often depends upon relational trust, observation of the researcher’s behavior within the community, perceived ethical intention, emotional safety, and the participant’s own timing. Second, trauma-informed interviewing requires recognition that disclosure is not simply verbal production of information but an emotionally regulated process shaped by vulnerability, risk assessment, memory, shame, and social consequence. The pacing of narration, moments of silence, fragmentation, and emotional shifts therefore become methodologically meaningful aspects of the testimony itself rather than problems requiring correction.
Third, the case demonstrates that ethical qualitative work with survivors of sexual violence requires flexibility rather than rigid procedural control. The interview structure emerged responsively around the participant’s needs, boundaries, and emotional state. The methodological priority was not maximizing disclosure but maintaining conditions in which the participant retained agency throughout the interaction. This included attention to privacy, pacing, emotional regulation, and the participant’s right to determine the limits of narration.
Finally, the encounter highlights a broader methodological principle within trauma-informed and non-extractive research: testimonies are not simply “collected.” They emerge relationally under specific social, emotional, and ethical conditions. In postwar environments marked by stigma, silence, and unresolved trauma, the ability to speak frequently depends upon whether survivors perceive the listener not merely as a researcher but as a trustworthy witness capable of receiving testimony without exploitation, judgment, or institutional pressure.
Case Illustration C: Trauma Vulnerability, Ethical Withdrawal, and Methodological Limits in Work with Yazidi Survivors
Between 2019 and 2023, I engaged in ongoing attempts to establish ethically appropriate communication with Yazidi survivors of sexual violence perpetrated during ISIS attacks against the Yazidi community. Initial contact occurred primarily through individual referrals, organizational recommendations, survivor networks, and humanitarian connections. These interactions unfolded through email correspondence, phone communication, and several Zoom-based conversations. The participants were geographically dispersed across multiple locations, reflecting the broader realities of forced displacement and diaspora fragmentation within the Yazidi community. Some individuals remained in Northern Iraq awaiting resettlement, while others had already relocated to countries including the United States, Germany, and Scandinavian states.
From the beginning, these interactions revealed methodological and ethical complexities that differed significantly from many other field contexts. Although a number of individuals initially expressed willingness to share their experiences, sustained engagement frequently revealed profound psychological vulnerability, instability, and ambivalence surrounding disclosure. Across the broader period of contact, I communicated with approximately sixteen individuals who at various stages indicated possible readiness to participate in testimony-based research or narrative sharing. However, in most cases, the process did not proceed to formal testimony collection.
The reasons for non-participation were not procedural but psychological and relational. In many interactions, survivors demonstrated signs of severe emotional distress, fluctuating readiness, exhaustion related to repeated retelling of traumatic experiences, fear of retraumatization, concerns regarding privacy and exposure, or uncertainty regarding the long-term consequences of disclosure. In several cases, survivors initially agreed to proceed with testimony and completed aspects of the consent process. However, during the course of interaction it became increasingly clear that continuing the interview process risked placing additional psychological burden on individuals already living under conditions of significant emotional strain and displacement-related instability.
As a result, I made the methodological and ethical decision not to incorporate several partially collected testimonies into the research database, despite the existence of preliminary consent agreements. This decision was grounded in a trauma-informed understanding that formal consent alone does not automatically establish ethical readiness for participation. Within trauma-centered qualitative methodology, the capacity to tolerate disclosure may shift throughout the interaction itself, particularly among survivors exposed to prolonged captivity, sexual violence, forced displacement, and ongoing uncertainty. Ethical responsibility therefore required continuous reassessment of participant well-being rather than reliance solely on procedural completion of consent documentation.
Ultimately, only one participant both completed the full testimony process and maintained sustained readiness to proceed under the ethical conditions required for inclusion within the study. This outcome became methodologically significant in itself. The limited number of completed testimonies highlighted the importance of recognizing non-disclosure, withdrawal, hesitation, and emotional instability as meaningful methodological realities within trauma research, rather than indicators of recruitment failure or insufficient participant engagement. The work with Yazidi survivors represented one of the most psychologically complex areas of engagement across the broader project. The interactions demonstrated that willingness to disclose experiences of sexual violence cannot be treated as stable, linear, or fully predictable. Readiness to narrate trauma often fluctuated in response to emotional state, displacement conditions, family pressures, prior experiences with humanitarian or media systems, and cumulative exhaustion from repeated requests to testify. In several cases, survivors appeared caught between competing desires: the wish to have their suffering recognized by the outside world and the simultaneous need to protect themselves from renewed psychological exposure.
This case illustrates a central principle of trauma-informed and non-extractive methodology: ethical qualitative research may require discontinuing data collection when continuation risks compromising participant well-being.
The absence of testimony, interruption of disclosure, or refusal to proceed should not automatically be understood as lack of cooperation or methodological limitation. In research involving survivors of severe sexual violence and genocide-related trauma, the decision not to speak—or the inability to continue speaking—may itself represent an important expression of survivor agency, emotional self-protection, and psychological survival.
5. The Problem of Methodological Coherence
Research on conflict-related sexual violence is frequently represented as methodologically coherent, with clearly defined samples, structured interview protocols, and stable procedures for data collection and analysis (Ellsberg & Heise, 2005; Wood, 2006). Published accounts often describe qualitative interviews as the outcome of planned and systematic methodological processes, emphasizing clarity, consistency, and procedural control. Such representations align with broader academic expectations in which methodological transparency is associated with rigor, reliability, and validity. However, these accounts do not fully capture the realities of conducting research in environments shaped by conflict, displacement, institutional instability, and ongoing political uncertainty.
Fieldwork in such settings unfolds through continuous negotiation and adaptation, with limited procedural stability. Access to participants frequently depends upon relationships with local institutions, community members, humanitarian actors, interpreters, and informal social networks, all of which shape who becomes available for participation and under what conditions engagement becomes possible. Researchers often rely on intermediaries to establish contact, facilitate communication, and mediate social trust. These relational dynamics influence not only the possibility of participation but also the form and depth of disclosure itself. Opportunities for interviews may emerge unexpectedly, shift over time, or disappear entirely due to changes in security conditions, institutional constraints, participant vulnerability, or perceptions of risk. Under such circumstances, sampling cannot be understood as a fixed methodological procedure. It becomes an evolving and context-dependent process shaped by instability, negotiation, and relational access.
Interviews generated within these environments are similarly contingent and interactional. Narratives concerning violence do not emerge solely through standardized questioning but take form through relationships shaped by trust, power, cultural expectations, emotional regulation, and situational dynamics. Research on political violence has demonstrated that interview material is co-produced through interaction, with both participant and researcher shaping the structure, pacing, and content of testimony (Fujii, 2012). Survivors may alter narrative form depending on the perceived role of the researcher, the purpose of the interaction, anticipated institutional consequences, and the emotional or political safety of disclosure. The same individual may produce markedly different forms of testimony across settings, offering extended and reflective narratives in one context while limiting disclosure to highly condensed factual statements in another. Such variation should not be interpreted as inconsistency or unreliability. The emergence of testimony is influenced by fluctuating psychological, social, and political conditions that determine when and how speech becomes possible. These dynamics become even more complex in multilingual environments where communication depends upon interpreters whose role extends beyond linguistic translation alone. Interpreters frequently mediate meaning, emotional tone, cultural reference, and relational atmosphere simultaneously. Translation within qualitative research is therefore not a neutral transfer of information but an interpretive and interactional process through which meaning is continuously negotiated and reformulated (Temple & Young, 2004; Squires, 2009). The presence of interpreters influences how questions are framed, how responses are conveyed, and how emotional emphasis is distributed throughout the narrative. Consequently, both the structure and affective texture of testimony may be shaped by linguistic mediation itself.
The social positioning of interpreters further influences these interactions. Gender, age, social status, perceived political affiliation, and community embeddedness may significantly affect participant comfort, trust, and willingness to disclose experiences of sexual violence. In many settings, survivors appeared more willing to discuss highly sensitive experiences with interpreters of a specific gender, while limiting or restructuring disclosure in the presence of others. Interpreters are rarely neutral linguistic instruments detached from the environments in which research occurs. They are often socially situated actors embedded within local networks and institutional relationships that participants recognize and respond to. These dynamics may facilitate communication by creating familiarity and trust, but they may also constrain disclosure if interaction is perceived as socially exposed or emotionally unsafe (Kapborg & Berterö, 2002; Squires, 2009).
The interview therefore becomes a triadic relational process in which meaning is co-produced across participant, researcher, and interpreter. Questions of validity cannot be reduced to linguistic equivalence alone. Although techniques such as back-translation may address textual consistency, they cannot fully capture the relational, affective, and situational dimensions shaping communication within trauma-centered interviews. Meaning emerges through interaction, negotiation, emotional regulation, and contextual interpretation. In research involving conflict-related sexual violence, where disclosure is deeply shaped by stigma, institutional vulnerability, fear, and political risk, these dynamics become methodologically central rather than peripheral.
Ethical considerations are inseparable from these processes. Within research on conflict-related sexual violence, ethics cannot be reduced to procedural approval or the initial acquisition of consent documentation. Decisions concerning whether to initiate, continue, redirect, pause, or terminate interviews frequently depend upon ongoing assessments of participant vulnerability, retraumatization risk, emotional regulation, social exposure, and potential institutional consequences (World Health Organization [WHO], 2001). Researchers may modify interview structure, avoid certain topics, or abandon procedural plans entirely in response to participant distress or changing situational conditions. Such adjustments are not deviations from methodology but integral aspects of ethical qualitative practice within trauma-centered fieldwork. At the same time, these ethical adaptations introduce variability that is rarely visible within formal methodological descriptions.
Despite these realities, published research often presents methodological procedures as stable, coherent, and consistently implemented. Sampling strategies, interview structures, and analytical procedures are frequently described in ways that emphasize order, replicability, and procedural consistency. While such representations align with disciplinary expectations regarding methodological rigor, they may obscure the contingent, negotiated, and relational character of fieldwork in conflict settings. The gap between reported methodology and lived research practice therefore raises important questions concerning how qualitative research involving survivors of violence is represented, interpreted, and evaluated within academic knowledge production.
The retrospective construction of methodological coherence carries significant analytical risks. When contextual variability remains unexamined, differences in narrative structure, emotional tone, length, or detail may be misinterpreted as reflections of individual credibility, memory quality, or psychological functioning rather than as responses to differing relational and institutional conditions. Highly structured or condensed narratives may be treated as incomplete, while extended accounts are implicitly privileged as more authentic or reliable. Such assumptions introduce unacknowledged hierarchies into interpretation, reinforcing institutional expectations—particularly within humanitarian, legal, and asylum systems—that equate testimonial detail and consistency with truthfulness.
Silence is especially vulnerable to such misinterpretation. The absence of elaboration, hesitation, or refusal to disclose is frequently understood primarily through psychological frameworks associated with trauma, repression, or avoidance. Although these interpretations may at times be relevant, they can obscure the broader social and political conditions shaping narration. Survivors may limit disclosure because of stigma, fear, institutional mistrust, community exposure, or uncertainty concerning how testimony may be used. In environments where speaking carries emotional, social, or political risk, restraint may represent a deliberate and protective form of regulation rather than an individual deficit. Without attention to these contextual dynamics, silence risks being psychologized in ways that detach testimony from the environments in which it emerges.
The emphasis on methodological coherence also shapes what becomes recognized as valid qualitative evidence. Material generated outside formal interview settings—including disclosures emerging during field visits, humanitarian interaction, memorial engagement, or long-term relational contact—is frequently excluded because it does not conform to expectations of standardization or procedural control. Such exclusions narrow the scope of qualitative inquiry by privileging forms of testimony that align with formalized research structures while marginalizing relationally emergent forms of narration. As a result, important dimensions of how knowledge about violence becomes possible may remain analytically invisible.
Addressing these limitations requires a reconsideration of methodological assumptions within qualitative research on conflict-related sexual violence. In this field, rigor cannot be equated with procedural uniformity or methodological control alone. Methodological rigor requires careful consideration of the relational, linguistic, institutional, political, and ethical conditions that shape the production of testimony. Methodological transparency must therefore extend beyond procedural description to include negotiation, instability, translation, emotional regulation, and contextual mediation as constitutive dimensions of qualitative knowledge production.
The central argument advanced here is that methodological coherence within research on conflict-related sexual violence is frequently constructed retrospectively through forms of representation that align with disciplinary expectations while obscuring the realities of trauma-centered fieldwork. Such constructions risk distorting the processes through which survivor testimony emerges and contribute to interpretive frameworks that privilege certain narrative forms over others. The approach developed in the following sections responds to this problem by treating heterogeneity not as methodological weakness but as analytically significant evidence concerning the conditions under which knowledge about violence becomes possible.
An additional layer of complexity emerges in interviews conducted within a shared non-native language, most commonly English in this study. In many interactions, neither the participant nor the researcher communicated in their first language. This linguistic positioning influenced both the structure and emotional dynamics of disclosure. Research in cross-language qualitative inquiry suggests that communication in a second language may create forms of emotional and cognitive distancing that enable articulation of experiences that might otherwise feel overwhelming or culturally constrained in one’s native language (Dewaele & Costa, 2013; Pavlenko, 2005). Within research involving sexual violence, this distancing may function as a protective mechanism allowing participants to narrate traumatic experiences without full affective immersion.
Field observations further suggested that some participants experienced the use of English as a space of relative anonymity and emotional distance from local systems of judgment and stigma. Communicating in a shared second language with an external researcher sometimes reduced the social immediacy associated with local linguistic expressions surrounding sexual violence, particularly in settings where such terminology carried highly stigmatizing or culturally embedded meanings. At the same time, communication in a non-native language introduced significant constraints. Vocabulary limitations, differences in semantic range, and the absence of culturally specific expressions often led to narrative simplification, restructuring, or partial translation of emotional meaning (Pavlenko, 2005; Squires, 2009). Testimony produced under such conditions was therefore shaped simultaneously by the enabling and limiting dimensions of linguistic distance, both of which require careful methodological consideration in the interpretation of qualitative narratives across contexts.
6. Toward a Non-Extractive, Trauma-Informed Approach
Research on conflict-related sexual violence has often proceeded from methodological assumptions that treat qualitative data as independent of the relational, institutional, and political contexts in which they emerge. Within many qualitative traditions, interviews are approached primarily as instruments for eliciting, organizing, and analyzing information, with methodological rigor associated with procedural consistency and systematic data collection. In environments shaped by violence, displacement, institutional instability, and social vulnerability, however, such assumptions become difficult to sustain. Engagement with survivors does not occur within neutral research conditions, and narratives do not emerge independently from the relational, ethical, political, and emotional environments in which disclosure takes place. A different methodological orientation is therefore required—one capable of accounting for how knowledge about violence is produced under conditions of constraint, vulnerability, and instability.
The approach developed in this study is grounded in two interrelated methodological principles: a non-extractive orientation to qualitative inquiry and a trauma-informed understanding of testimonial production. These principles are not treated as separate frameworks but as mutually constitutive dimensions of a broader methodological position centered on the conditions through which disclosure becomes possible.
6.1. Non-Extractive Methodology
A non-extractive methodology challenges the assumption that knowledge can be obtained through the systematic collection of information from participants while remaining detached from the relationships and environments through which narration emerges. Conventional qualitative approaches often position interviews as sources of data that can be accessed, recorded, coded, and interpreted with limited attention to the broader social and political conditions shaping disclosure. Critiques emerging from decolonial and Indigenous methodological traditions have questioned these assumptions, arguing that such approaches may reproduce asymmetries of power by positioning participants primarily as providers of data while obscuring the researcher’s own role in shaping interaction and interpretation (Kovach, 2009; Smith, 2012).
Within a non-extractive framework, interviews are understood as relational processes through which knowledge is co-produced rather than passively retrieved. What becomes narratable within an interview depends not only on the participant’s experiences but also on how the interaction is structured, how the researcher is perceived, and how the anticipated purpose of the encounter is understood. Testimony may be shaped by expectations concerning whether narratives will be used for research, legal documentation, humanitarian advocacy, psychosocial support, or public representation. Participants may therefore emphasize, withhold, reorganize, or limit aspects of their experiences in response to the relational and institutional conditions surrounding disclosure. Such shifts should not be interpreted as distortion or unreliability but as expressions of the situated nature of testimonial production.
A non-extractive orientation also requires recognition of the limits of disclosure itself. Not all experiences can be articulated within a particular interaction, and not all participants will choose—or be able—to narrate experiences of violence fully. Decisions not to speak, to remain silent, or to disclose only selectively must therefore be understood as analytically meaningful aspects of the data rather than as absences requiring completion. These limits are shaped by safety concerns, emotional regulation, social stigma, institutional vulnerability, mistrust, and the anticipated consequences of testimony. Treating such boundaries merely as missing information risks imposing methodological expectations that fail to align with the lived realities of survivors of conflict-related sexual violence.
Context-dependent narration is central to this methodological position. The same individual may produce substantially different forms of testimony across settings, relationships, and institutional environments. Narratives articulated within therapeutic encounters may differ significantly from those emerging in legal proceedings, humanitarian programs, or field-based interactions. These variations do not necessarily reflect instability or inconsistency in experience. They reflect the relational, emotional, and political conditions that shape how testimony becomes possible within specific contexts. Analysis must therefore remain attentive to multiplicity and variation, treating differences in narrative form as meaningful features of qualitative material rather than as methodological problems requiring standardization.
6.2. Trauma-Informed Methodology
A trauma-informed methodology extends these considerations by examining how experiences of violence shape engagement with the research process itself. Within this framework, qualitative data are understood as deeply influenced by conditions of safety, trust, timing, emotional regulation, and perceived risk. Participation in an interview is not treated as a neutral or purely procedural act. Narration is shaped by continuous assessments regarding safety, the boundaries of disclosure, the extent to which experiences can be articulated, and the anticipated consequences of speaking. These assessments are shaped simultaneously by prior experiences of violence and by present social, institutional, and relational realities.
Safety constitutes one of the central conditions shaping disclosure. Participants may evaluate risk in relation to interview location, institutional affiliation, confidentiality, the presence of interpreters or community members, and the potential social or political consequences of speaking. Under such circumstances, survivors may produce highly structured or emotionally controlled accounts, limit detail, avoid specific topics, or disengage from narration altogether. These responses should not be reduced to narrowly psychological interpretations of avoidance alone. Rather, they frequently represent contextually situated strategies of protection and emotional regulation shaped by the environment within which disclosure occurs.
Trust operates alongside safety as a dynamic and relational process rather than as a fixed precondition of interviewing. The development of trust may depend upon perceptions of the researcher’s ethical positioning, relational conduct, institutional affiliations, use of intermediaries, and broader engagement within the community. Trust may fluctuate throughout the interaction itself or evolve gradually across repeated encounters. Where trust remains limited, testimony may stay concise, factual, or highly formalized. Where trust develops over time, participants may engage in more extended, emotionally reflective, or layered forms of narration. Such differences should not be attributed to individual communicative capacity alone but understood in relation to the conditions under which disclosure becomes emotionally and socially possible.
Temporal conditions also shape testimonial form. Narratives articulated shortly after experiences of violence may differ substantially from testimonies emerging years or decades later. Current life circumstances, prior experiences of disclosure, social recognition, displacement, and ongoing insecurity all influence how experiences are recalled and communicated. Repeated engagement may produce shifts in narrative structure, emphasis, emotional tone, and interpretive meaning. These transformations should not automatically be interpreted as evidence of memory instability. Rather, they reflect the evolving ways in which traumatic experiences are processed, situated, and communicated across time.
Within a trauma-informed framework, silence is not interpreted as the absence of data. Hesitation, pauses, refusal to elaborate, or selective disclosure are understood as meaningful responses shaped by context, vulnerability, and perceived consequence. Silence may function as emotional protection, social regulation, resistance, or an attempt to maintain control over the conditions of narration. Similarly, fragmentation within testimony is not approached as methodological deficiency or inconsistency. Fragmented narration reflects the conditions under which experiences are remembered, emotionally processed, and articulated, including the impact of trauma on memory, communication, and affective regulation. Attempts to force coherence or narrative completeness risk obscuring the significance of fragmentation itself as part of the testimonial process.
6.3. Implications for Methodological Practice
Taken together, these considerations require a reconceptualization of qualitative data within research on conflict-related sexual violence. Testimonies cannot be understood as discrete informational units collected through standardized procedures independently from the circumstances of their production. Testimonies emerge through interactions shaped by relational, linguistic, institutional, ethical, political, and emotional conditions. Disclosure occurs under constraint, where possibilities for narration are influenced by safety, trust, translation, social recognition, institutional mediation, and perceived consequence.
Recognizing qualitative material as co-produced under such conditions carries several methodological implications. First, it requires approaches capable of engaging systematically with variability without imposing artificial standards of uniformity upon survivor narratives. Second, it demands analytic strategies attentive not only to what is said but also to how, when, and under what conditions testimony emerges. Third, it requires transparency regarding the factors shaping data generation, including linguistic mediation, non-native language use, field conditions, institutional settings, and relational dynamics influencing disclosure.
This approach does not abandon methodological rigor. Instead, it redefines rigor in relation to the realities of trauma-centered fieldwork. Rigor lies not in the elimination of variability but in the capacity to engage critically and systematically with complexity, instability, and the limits of knowledge production. A trauma-informed and non-extractive methodological framework therefore provides a basis for aligning qualitative research practice with the relational, ethical, and political realities of conducting research in environments shaped by violence and displacement. The framework developed in the following section builds upon these principles by offering a structured approach for organizing heterogeneous qualitative material across temporal, contextual, and interactional dimensions.
7. Tri-Axial Classification Framework
The preceding sections have demonstrated that qualitative data in research on conflict-related sexual violence emerge under conditions that vary across historical periods, political environments, linguistic settings, and modes of engagement. This variability is not incidental to the research process. It reflects the conditions under which testimony becomes possible, including differences in safety, institutional mediation, relational trust, displacement, memorialization, and social recognition. The central methodological challenge is therefore not to eliminate heterogeneity through standardization but to develop a structure capable of organizing variation systematically without erasing the specificity of the encounters from which narratives emerge.
The present study addresses this challenge through a tri-axial classification framework designed to organize heterogeneous qualitative material across three interrelated dimensions: temporal-historical location, contextual-political environment, and interview typology. The framework does not impose a single interpretive model upon the data or reduce testimonies to fixed categories. It is providing an analytical structure through which differences in narrative form, disclosure practices, and testimonial configuration can be situated and interpreted in relation to the conditions of their production.
The three axes are analytically distinct but operate in combination. Each testimony may be located at the intersection of these dimensions, allowing for a more precise understanding of how narrative structure is shaped by temporal distance, political environment, institutional conditions, and relational context. The framework therefore functions not merely as a classificatory system but as a methodological device linking qualitative material to the circumstances through which knowledge about violence is generated.
7.1. Temporal-Historical Axis
The temporal-historical axis situates testimonies within broader historical trajectories of violence, memory, recognition, and post-conflict transformation. This dimension recognizes that narratives are shaped not only by individual experience but also by temporal distance from the event and by the changing social, political, and institutional environments within which experiences are recalled and communicated. Accounts produced shortly after violence differ from those articulated decades later, not only in content but also in emotional structure, narrative organization, interpretive framing, and conditions of disclosure.
Four broad temporal categories emerged across the dataset:
This category includes testimonies linked to earlier conflicts and their long-term aftermath, often narrated decades after the original events. Such accounts are frequently shaped by intergenerational transmission, prolonged social silence, partial institutional recognition, and unresolved historical trauma. Temporal distance may create space for reflection and narrative elaboration while simultaneously preserving longstanding constraints related to shame, stigma, or political denial.
This category includes contexts such as the Balkan wars and the Rwandan genocide. Narratives emerging from these settings are situated within periods marked by increasing international recognition of wartime sexual violence alongside continuing local tensions, contested memory, and uneven postwar reconstruction. Testimony may therefore reflect both the immediacy of post-conflict conditions and the influence of expanding legal, humanitarian, and commemorative frameworks.
This period includes conflicts in parts of Central and East Africa and other regions affected by prolonged instability. Narratives are frequently shaped by continuing insecurity, fragile institutional protection, displacement, and sustained humanitarian presence. Disclosure may occur within programmatic or service-oriented environments where testimony is closely connected to access to support, medical care, or psychosocial intervention.
This category includes testimonies emerging under conditions of active violence, recent displacement, or unresolved political instability. Narratives produced within these environments are often highly constrained and shaped by immediate concerns regarding safety, surveillance, retaliation, and future uncertainty. Under such conditions, disclosure may remain fragmented, emotionally regulated, or strategically limited.
The temporal-historical axis highlights the dynamic relationship between memory and context. Changes in narrative form across time should not be interpreted as evidence of inconsistency or unreliability. They reflect evolving conditions of recognition, safety, emotional processing, institutional response, and social meaning. They reflect evolving conditions, safety, emotional processing, institutional responses and collective interpretation. Narrative transformation across historical periods therefore becomes analytically significant for understanding how experiences of violence are remembered, communicated, and situated within changing environments.
7.2. Contextual-Political Axis
The contextual-political axis organizes testimonies according to the broader socio-political environments within which violence occurred and disclosure became possible. This dimension recognizes that political context shapes not only the experience of violence itself but also the conditions under which survivors may speak, remain silent, or regulate disclosure.
Several broad contextual categories emerged within the dataset:
These environments are characterized by large-scale systematic violence and, in some instances, formalized processes of memorialization and institutional recognition. Narratives may be influenced by commemorative practices, public acknowledgment, transitional justice initiatives, and collective memory structures. At the same time, local social dynamics and unresolved trauma may continue to shape the limits of disclosure.
This category includes wars marked by identity-based violence, including conflicts in the Balkans. Narratives within these settings are frequently shaped by contested histories, political polarization, community affiliation, and uneven legal recognition. Disclosure may remain constrained by fears concerning social exposure, political interpretation, or intergroup tensions.
In environments where violence continues or instability remains unresolved, testimony often emerges under conditions of immediate vulnerability. Participants may restrict disclosure because of concerns regarding retaliation, surveillance, institutional collapse, or physical insecurity. Accounts produced under such circumstances are often concise, highly controlled, and strategically regulated.
These settings involve violence associated with state actors or political systems restricting freedom of expression. Narratives may remain partial or carefully managed because disclosure itself carries social, legal, or political risk. Silence, hesitation, or minimal testimony may therefore function as protective strategies rather than indicators of absent experience.
This category include testimonies generated within asylum systems, humanitarian structures, and resettlement environments in Western Europe and North America. Narratives are frequently shaped by institutional expectations requiring detailed, coherent, and legally legible forms of testimony. Participants may therefore adapt narrative structure in response to perceived bureaucratic or legal demands, emphasizing some aspects of experience while limiting others.
The contextual-political axis demonstrates that narrative form cannot be interpreted independently from the environment in which disclosure occurs. Differences in testimonial length, emotional tone, fragmentation, detail, and coherence frequently reflect varying conditions of safety, institutional recognition, political surveillance, and social legitimacy. A brief or highly controlled narrative in one setting may reflect acute contextual constraint, while a more elaborated testimony in another may become possible through relative safety and recognition. Interpretation therefore requires attention not only to what is narrated but also to the political and institutional conditions shaping the possibility of speech itself.
7.3. Interview Typology Axis
The interview typology axis focuses on the modes through which qualitative material was generated and the implications these modes carry for narrative structure and disclosure practices. This axis recognizes that testimonies emerging through clinical encounters, formal interviews, survivor-led disclosures, or field-based interaction cannot be treated as methodologically equivalent because each form of engagement produces distinct relational and communicative conditions.
The typology axis is organized across four dimensions: purpose, structure, format, and duration.
By purpose
Conducted within formally defined research designs organized around thematic, geographic, or analytical questions. These interviews were often semi-structured and shaped by the objectives of the broader study.
Emerging within therapeutic relationships where testimony unfolded gradually through processes of trust, stabilization, emotional regulation, and ongoing care. Narratives within these settings frequently developed across extended periods of relational continuity.
Initiated by participants themselves outside predefined interview protocols. Such disclosures often emerged through perceived opportunities for witnessing, recognition, or emotional necessity rather than formal recruitment procedures.
Arising during memorial visits, humanitarian engagement, community interaction, or site-based presence in locations connected to histories of violence. Narratives within these contexts frequently emerged relationally through shared environment and symbolic space rather than through direct questioning alone.
By structure
Organized around predefined questions intended to ensure procedural consistency across participants.
Combining thematic guidance with conversational flexibility, allowing participants greater influence over narrative direction and emphasis.
Prioritizing participant-led storytelling with minimal intervention from the researcher.
By format
Conducted one-on-one and allowing for focused relational engagement.
Involving multiple participants, where testimony may be shaped by collective dynamics, mutual witnessing, or social regulation.
Emerging within broader social environments where distinctions between interview, conversation, and relational engagement remained fluid.
By duration
The duration dimension reflects not simply the length of interaction but the degree of relational continuity shaping testimonial development.
Single interactions often occurring under constrained field conditions or limited access. Narratives in this category were frequently concise and situationally bounded.
Involving one or several meetings allowing for partial development of trust, clarification, and narrative elaboration without extending into long-term engagement.
Consisting of repeated interactions across extended periods, often within clinical or sustained research relationships. These encounters allowed narratives to evolve over time through changing conditions of trust, interpretation, memory, and emotional processing.
Duration is not treated as a hierarchy of testimonial value. Longer engagement may facilitate narrative expansion but may also introduce new constraints, reinterpretations, or shifts in meaning. Similarly, brief encounters may produce highly significant testimony, particularly within environments where prolonged engagement is impossible or unsafe. Duration therefore functions as an indicator of interactional context rather than as a measure of narrative quality or credibility.
Integration of Axes
Figure 2 provides an overview of the 402 qualitative dataset items included in the corpus, illustrating the diversity of data sources and conditions under which testimonies were generated. The distribution highlights the methodological complexity of the dataset, including the use of interpreters, multilingual contexts, survivor-initiated disclosures, field-based encounters, and longitudinal engagements across multiple settings.
Each testimony within the dataset may be situated at the intersection of these three axes. For example, a brief disclosure emerging during a memorial-site encounter in a post-genocide environment reflects a distinct configuration of temporal, contextual, and typological conditions. A longitudinal clinical narrative generated within a diaspora setting reflects a different configuration shaped by sustained relational engagement and institutional mediation.
The framework enables systematic comparison across these configurations without reducing heterogeneous narratives to a single methodological standard. It allows patterns of testimonial variation to be examined in relation to changing conditions of disclosure while preserving the specificity of each encounter. By
Figure 2. Tri-axial classification framework for structuring heterogeneous qualitative data on war rape survivors.
organizing qualitative material across these intersecting dimensions, the framework aligns methodological analysis with the principles developed in the preceding sections. Testimony is approached not as neutral data extracted independently from context but as situated knowledge emerging through conditions of violence, displacement, memory, relational trust, and institutional constraint.
To facilitate systematic analysis across heterogeneous qualitative material, the study developed a tri-axial classification framework linking testimonies to three interrelated dimensions: temporal-historical conditions, contextual-political environments, and interview typology. The framework conceptualizes each interview not as a standardized unit of data collection but as a situated encounter produced under specific historical, relational, institutional, linguistic, and political conditions. Rather than eliminating variability, the framework structures heterogeneity analytically, allowing narrative differences to be interpreted in relation to the conditions shaping disclosure and testimony production (see Table 1).
Table 1. Summary of the tri-axial classification framework: axes, subcategories, and methodological significance.
Axis |
Subcategories |
Methodological Significance |
1. Temporal-Historical Axis |
Post-WWII/Legacy contexts |
Historical distance, intergenerational silence, memory transmission, unresolved trauma |
1990s conflicts |
Emerging international recognition, legal developments, transitional justice |
2000s-2010s conflicts |
Ongoing instability, humanitarian presence, fragile institutional protection |
Contemporary contexts |
Active conflict conditions, acute risk, immediacy of political and social consequences |
2. Contextual-Political Axis |
Genocide contexts |
Mass violence, collective targeting, official or partial recognition |
Ethnic conflict contexts |
Identity-based violence, contested narratives, politicized memory |
Ongoing violence contexts |
Continuing insecurity, restricted disclosure, unstable environments |
State repression contexts |
Fear of surveillance, retaliation, institutional constraint |
Diaspora/migration contexts |
Displacement, asylum systems, legal expectations, resettlement dynamics |
3. Interview Typology Axis |
Purpose: Research, clinical, survivor-led, field-based |
Distinct relational and institutional conditions shaping disclosure |
Structure: Structured, semi-structured, narrative |
Different levels of procedural flexibility and participant-led narration |
Format: Individual, group, community-based |
Influence of relational and collective dynamics on testimony |
Duration: Short, medium, longitudinal |
Degree of relational continuity and evolving trust over time |
The framework conceptualizes qualitative testimony as produced under conditions of negotiation, relational trust, institutional mediation, translation, and contextual constraint. Variability in narrative form—including fragmentation, silence, brevity, or extended narration—is therefore interpreted not as methodological inconsistency but as evidence of the differing conditions under which disclosure becomes psychologically, socially, and politically possible. This approach supports a trauma-informed and non-extractive methodology by linking narrative structure to the environments in which testimony emerges.
To further operationalize the tri-axial methodological framework developed in this study, Figure 3 presents an integrated overview of the dataset across temporal-historical, contextual-political, and interview typology dimensions. The figure demonstrates how qualitative interviews and testimonies were situated within specific historical periods, geopolitical conditions, and modes of engagement, including research, clinical, survivor-led, humanitarian, and field-based
Figure 3. Tri-axial classification framework and interview dataset by axis description.
encounters. It additionally illustrates the diversity of interview formats, durations, and conditions of knowledge production across multiple conflict and post-conflict environments.
The figure highlights that qualitative data within trauma-centered research are not standardized or context-free. Instead, testimonies emerge under distinct relational, institutional, linguistic, political, and ethical conditions that shape disclosure practices, narrative structure, emotional regulation, and testimonial depth. By organizing interviews through intersecting methodological axes, the framework enables systematic analysis of heterogeneous qualitative material while preserving the contextual specificity of each encounter (see Figure 3).
Integrated visualization of the study’s tri-axial methodological framework linking temporal-historical conditions, contextual-political environments, and interview typology dimensions across the 402-item qualitative dataset. The figure includes dataset distribution by country and conflict context, interview modes and formats, duration patterns, and notes on conditions shaping testimony production. It illustrates the methodological principle that qualitative narratives on conflict-related sexual violence are produced under conditions of negotiation, instability, institutional mediation, displacement, relational trust, and contextual constraint, rather than through uniform or standardized procedures.
8. Field Conditions and Data Formation
Qualitative data in research on conflict-related sexual violence do not emerge within controlled or methodologically stable environments. It is produced within field conditions shaped by conflict, displacement, institutional instability, humanitarian intervention, migration, and ongoing political uncertainty. These conditions influence not only access to participants but also the structure, depth, pacing, and emotional organization of testimony itself. Understanding how qualitative data are formed therefore requires attention to the environments in which disclosure becomes possible, including conflict zones, humanitarian and programmatic settings, clinical contexts, memorial environments, and migration or asylum systems. Across these domains, methodology does not function as a fixed procedural design imposed uniformly in advance. Instead, it develops responsively through negotiation with the conditions of the field.
In active conflict environments and post-conflict regions marked by continuing insecurity, qualitative data generation is shaped by instability, fear, restricted mobility, and the ongoing possibility of violence. Access to participants frequently depends upon local intermediaries, community members, humanitarian workers, non-governmental organizations, and informal gatekeepers who mediate trust and determine whether engagement becomes possible. These relationships influence not only recruitment but also the broader relational conditions surrounding disclosure. Safety considerations affect where interviews can occur, how long interactions may continue, who can be present, and which topics participants perceive as safe to discuss. Under such circumstances, testimonies may remain brief, fragmented, interrupted, or highly regulated. Participants may intentionally limit disclosure because of fears concerning retaliation, surveillance, community exposure, or institutional consequence. Narratives generated under these conditions often emphasize essential factual elements while minimizing emotional elaboration or contextual detail. Such forms of narration should not be interpreted as evidence of diminished experience, reduced memory, or limited communicative capacity. They reflect the immediate conditions under which speech becomes socially, politically, and psychologically possible. Humanitarian and programmatic environments introduce a different but equally significant set of methodological dynamics. Interviews conducted within NGO programs, psychosocial services, women’s centers, medical initiatives, or community-based support structures are shaped by institutional frameworks and the practical realities of service provision. Participants may engage with researchers while simultaneously navigating systems related to food access, shelter, healthcare, legal documentation, psychosocial support, or economic survival. Under these circumstances, narratives concerning sexual violence are often embedded within broader discussions about displacement, caregiving, poverty, illness, family separation, or access to assistance. The institutional environment may create opportunities for disclosure by providing relative safety, legitimacy, and organizational support. At the same time, these settings may generate implicit expectations concerning how experiences should be narrated, particularly where testimony is linked—directly or indirectly—to eligibility for services, humanitarian recognition, or program participation.
Participants may therefore adapt narrative structure in response to perceived institutional priorities, emphasizing certain forms of suffering while minimizing others. Such adaptations do not necessarily reflect manipulation or strategic distortion. They reveal the extent to which testimony is shaped by the social and institutional conditions within which it is produced. The resulting qualitative material frequently combines personal memory with the structural logic of humanitarian systems, where narratives are organized not only around lived experience but also around bureaucratic legibility, vulnerability categories, and institutional recognition.
Clinical environments represent another distinct mode of qualitative data formation characterized by longitudinal engagement, emotional continuity, and relational depth. Within therapeutic settings, narratives frequently unfold gradually over extended periods through processes of trust-building, stabilization, emotional regulation, and ongoing care. Participants may revisit experiences repeatedly across sessions, introducing additional detail, reframing earlier interpretations, or altering the emotional organization of their testimony over time. The qualitative material generated within such settings therefore differs substantially from that produced through single-session interviews. They reflecting static accounts of experience, clinical narratives emerge through iterative processes in which meaning is continuously negotiated, reconstructed, and emotionally processed.
This longitudinal dimension introduces important methodological implications. Changes in narrative structure across time should not automatically be interpreted as inconsistency or unreliability. Instead, they often reflect shifts in emotional safety, therapeutic stabilization, social context, or the participant’s evolving relationship to memory itself. Ethical considerations become especially significant within clinical environments because disclosure is closely connected to processes of care and psychological support. Decisions regarding what is shared, delayed, or withheld are frequently shaped by emotional readiness, therapeutic pacing, vulnerability, and the perceived safety of the relationship. Testimony produced in these contexts therefore cannot be separated from the relational and ethical conditions through which narration develops.
Migration, asylum, and diaspora settings introduce additional layers of complexity into the formation of qualitative data. Interviews conducted within countries of resettlement frequently unfold within legal and bureaucratic systems that impose specific expectations regarding narrative structure, consistency, and evidentiary detail. Survivors may already have participated in multiple institutional interviews related to asylum applications, immigration procedures, medical evaluations, or humanitarian assessments before entering research encounters. Under such conditions, participants may become acutely aware that detailed, linear, and coherent narratives are often institutionally rewarded and treated as indicators of credibility.
These pressures significantly influence testimonial form. Participants may organize narratives strategically around chronology, factual precision, or legally recognizable categories of violence while limiting ambiguity, emotional complexity, uncertainty, or contradiction. Concerns regarding the consequences of inconsistency may encourage highly controlled forms of narration in which testimony is shaped simultaneously by memory and by awareness of institutional evaluation. As a result, narratives generated within asylum and migration systems often reflect not only experiences of violence themselves but also prolonged exposure to bureaucratic structures demanding repeated articulation of trauma under conditions of judgment and surveillance.
Field-based encounters occurring during memorial visits, humanitarian travel, community engagement, or site-based interaction represent an additional mode of qualitative data formation. These encounters are often unscripted and emerge through shared presence within environments marked by historical, symbolic, or emotional significance. Narratives may develop spontaneously in response to memorial spaces, collective rituals, informal conversation, or the emotional atmosphere surrounding sites associated with violence and loss. Such interactions frequently produce concise yet highly significant forms of testimony shaped by immediacy, relational trust, and contextual meaning rather than by formal interview procedures.
The absence of predefined structure does not diminish the methodological significance of these encounters. On the contrary, they demonstrate how qualitative testimony may emerge through environmental and relational conditions that formalized interview settings cannot fully reproduce. Memorial spaces, post-conflict landscapes, and community environments often function not merely as passive research locations but as active contexts influencing memory activation, emotional engagement, and the possibility of narration itself. Testimony generated through such encounters therefore reflects the interaction between place, relational dynamics, symbolic meaning, and historical memory.
Across all of these environments, language and translation play constitutive roles in shaping qualitative data formation. Communication frequently occurs through interpreters or through the use of a shared non-native language, introducing additional layers of relational and linguistic mediation. Translation is not a neutral transfer of information from one language into another. Rather, meaning is continuously negotiated through linguistic choices, emotional interpretation, cultural framing, and interactional dynamics (Temple & Young, 2004; Squires, 2009). The presence of interpreters may influence both the structure and emotional texture of testimony, shaping what participants choose to disclose and how experiences are communicated.
Similarly, interviews conducted in a shared second language may alter emotional distance, narrative pacing, and expressive possibility. Some participants appeared to experience communication in English as creating a degree of emotional separation from culturally embedded expressions of shame or stigma associated with sexual violence. At the same time, non-native language communication frequently constrained nuance, cultural specificity, and emotional precision (Dewaele & Costa, 2013; Pavlenko, 2005). Qualitative data produced under such conditions therefore cannot be interpreted independently from the linguistic environments through which narration occurs.
Taken together, these field conditions demonstrate that qualitative data in research on conflict-related sexual violence are generated through contingent, relational, and context-dependent processes rather than through uniform methodological procedures. There is no single pathway through which testimony emerges. Each setting introduces distinct possibilities, constraints, institutional pressures, and ethical considerations that shape how participants engage with researchers and what forms of disclosure become possible. Methodological decisions concerning sampling, interview structure, pacing, duration, and interpretation are therefore inseparable from the realities of the field itself.
The key implication is that methodology within trauma-centered qualitative research must be understood as emergent rather than fully predetermined. Methodological practice develops responsively through ongoing negotiation with instability, access limitations, ethical tensions, emotional vulnerability, and shifting field conditions. Attempts to impose rigid procedural uniformity risk obscuring the relational and contextual factors shaping qualitative data production and, in doing so, misrepresent the nature of testimony itself.
This perspective reinforces the central argument of the present study. Qualitative data concerning conflict-related sexual violence are not collected uniformly across contexts but produced through interactions shaped by violence, displacement, institutional mediation, language, trust, and political uncertainty. Recognizing these conditions is essential for developing methodological approaches capable of engaging with complexity while maintaining ethical and analytical rigor.
9. Ethics as Structural, Not Procedural
Ethical considerations in research on conflict-related sexual violence are frequently presented as procedural obligations addressed through institutional review processes, informed consent documentation, confidentiality protocols, and measures designed to minimize participant harm. Although such safeguards remain essential, they do not fully capture the ethical realities of conducting research in environments shaped by violence, displacement, political instability, institutional vulnerability, and ongoing social risk. In these contexts, ethics cannot be understood solely as compliance with predefined procedural standards operating externally to the research process. Rather, ethical conditions shape whether participation becomes possible, what forms of disclosure emerge, how narratives are structured, and which testimonies remain constrained, interrupted, or unspeakable. Ethics therefore function not as a separate dimension of methodology but as a constitutive condition of qualitative data production itself.
Ethical oversight within the present study varied across the different modes of engagement represented in the dataset. Material originating from formally structured research projects was conducted under institutionally approved ethical procedures appropriate to the relevant setting, period, and form of data collection. Clinical narratives emerged within therapeutic contexts governed by principles of confidentiality, emotional stabilization, professional responsibility, and ongoing relational consent. Humanitarian and programmatic interactions unfolded within support-oriented environments in which disclosure frequently occurred alongside processes related to psychosocial care, service access, community participation, or survivor advocacy initiatives.
Field-based testimonies required additional retrospective ethical evaluation because many emerged outside formally organized interview environments. In such cases, inclusion depended upon careful assessment of voluntariness, contextual sensitivity, absence of coercion, anonymization feasibility, and evaluation that incorporation into the study would not compromise participant dignity, psychological well-being, social position, or safety. The inclusion of material generated through memorial visits, humanitarian engagement, or relational field encounters therefore required ethical consideration extending beyond procedural models of consent alone.
Across all forms of engagement, consent was approached not as a singular administrative event but as an ongoing relational process unfolding throughout interaction. Participants retained the ability to pause, redirect, limit, refuse, or terminate disclosure at any stage of engagement. Decisions concerning continuation, emotional pacing, and depth of narration were continuously negotiated rather than assumed to be permanently established through initial agreement. This distinction becomes especially significant within trauma-centered qualitative research, where willingness to speak may fluctuate in response to emotional activation, relational trust, perceptions of safety, fatigue, or changing assessments of vulnerability. Ethical engagement therefore required continuous responsiveness to participants’ emotional and contextual conditions rather than reliance upon procedural completion of consent documentation alone.
A central implication of this position is that ethical practice directly shapes the form and limits of qualitative data. Participants frequently reassessed disclosure boundaries throughout interviews, choosing to elaborate, remain silent, avoid specific details, or shift the direction of conversation in response to evolving emotional and relational conditions. Such movements were not treated as disruptions to the research process but as meaningful aspects of testimonial regulation. Ethical methodology in this context required preserving participant control over the scope, pacing, and direction of narration rather than prioritizing maximal disclosure or narrative completeness.
Risk management formed an inseparable component of this process. Within research on conflict-related sexual violence, risks extend beyond the possibility of emotional distress alone. Disclosure may carry social, familial, institutional, legal, or political consequences, particularly in environments where stigma, discrimination, retaliation, or surveillance remain ongoing concerns (World Health Organization [WHO], 2001). Participants often navigated these risks actively while determining what could be disclosed, how experiences should be narrated, and which aspects of violence required concealment or emotional regulation. Ethical practice therefore involved ongoing assessment not only of psychological vulnerability but also of the broader social and political implications of participation.
Under such conditions, methodological flexibility became ethically necessary. Interviews occasionally required modification, interruption, redirection, or termination in response to participant distress, emotional exhaustion, environmental instability, or changing perceptions of safety. These adjustments should not be understood as deviations from methodological rigor. Rather, they constituted integral components of trauma-informed qualitative practice. Ethical responsiveness shaped not only the interaction itself but also the character of the qualitative material ultimately generated.
Power asymmetry further complicated these dynamics. Researchers frequently occupy positions associated with institutional legitimacy, academic authority, international mobility, and control over representation within scholarly, legal, or policy-oriented frameworks. Participants, particularly survivors of conflict-related sexual violence, may simultaneously occupy positions marked by displacement, economic precarity, institutional dependency, legal uncertainty, or social marginalization. These asymmetries influence how interviews are perceived and how testimony is structured. Participants may attempt to anticipate researcher expectations, shape narratives toward perceived institutional priorities, or regulate disclosure according to assumptions concerning credibility, usefulness, or consequence.
Recognition of these asymmetries does not eliminate them. However, it allows for more reflexive engagement with how knowledge is produced through unequal relational conditions. Efforts to reduce exploitation therefore involved emphasizing transparency regarding research aims, maintaining participant control over engagement, minimizing coercive pressure toward disclosure, and recognizing limits to what could ethically be asked, recorded, or interpreted. Ethical responsibility extended beyond data collection alone to include decisions concerning analysis, representation, anonymization, and dissemination.
Within this framework, survivor agency is understood as both active and constrained. Participants are not approached as passive providers of information but as individuals continuously shaping testimonial form through decisions concerning narration, silence, emotional pacing, and disclosure boundaries. At the same time, these decisions occur within environments structured by violence, stigma, institutional vulnerability, and unequal power relations. Agency therefore cannot be reduced to unrestricted autonomy. Rather, it is expressed through the navigation of constraints and the management of exposure within conditions that may severely limit available choices.
The inclusion of material originating outside formal research interviews introduced additional ethical complexity. As discussed in previous sections, portions of the dataset emerged through field-based encounters, memorial interaction, humanitarian engagement, or relational exchanges not initially designed as formal research procedures. The incorporation of such material required careful evaluation concerning voluntariness, contextual appropriateness, participant understanding, and the potential consequences of inclusion. In situations where these conditions could not be ethically secured, material was excluded regardless of possible analytical significance.
Safeguards applied to such material included anonymization, removal of identifying contextual details, attention to symbolic and community vulnerability, and continuous evaluation of whether inclusion risked exposing participants to harm, stigmatization, or institutional consequence. The boundaries separating clinical, humanitarian, memorial, and research environments frequently remained fluid in practice, particularly within long-term field engagement. Participants themselves did not always distinguish sharply between these domains. This ambiguity required heightened ethical sensitivity regarding how testimonies were interpreted, categorized, and represented within academic analysis.
Viewing ethics as structural rather than procedural also transforms how qualitative narratives are interpreted analytically. If ethical conditions shape the possibility of disclosure, then silence, fragmentation, brevity, emotional regulation, and selective narration cannot be interpreted solely as methodological limitations or individual psychological phenomena. Such features may instead reflect the ethical, social, institutional, and political environments within which testimony emerged. Interpreting narratives without reference to these conditions’ risks reproducing decontextualized understandings of survivor testimony and obscuring the structural constraints shaping what can be spoken.
This perspective aligns with broader discussions within qualitative methodology emphasizing the relational and situated character of knowledge production (Guillemin & Gillam, 2004). The present study extends these discussions into the specific context of conflict-related sexual violence, where ethical considerations become inseparable from questions of vulnerability, representation, institutional power, and testimonial possibility. Ethics are therefore conceptualized not as external procedural safeguards applied to research after methodological design but as constitutive conditions shaping the production, limits, and interpretation of qualitative knowledge itself.
The central argument advanced here is that ethical conditions influence not only how research is conducted but also what forms of testimony become narratable at all. Ethics shape whether disclosure emerges, how narratives are organized, what remains fragmented or silent, and which experiences can safely enter qualitative analysis. Recognizing this relationship requires methodological approaches capable of engaging with uncertainty, vulnerability, and the limits of representation while maintaining both analytical rigor and ethical accountability.
Ethical Pathways and Consent
The dataset was generated across several ethically distinct data streams. Formal research interviews followed explicit consent procedures, including explanation of the purpose of the study, voluntary participation, the right to withdraw, and anonymization. Clinical narratives were included only when later research use could be ethically justified without compromising confidentiality, dignity, or safety. Humanitarian and programmatic encounters were incorporated only when disclosure was voluntary and when the material could be anonymized and interpreted without exposing participants to social, institutional, or political risk. Field-based testimonies were assessed retrospectively through a trauma-informed ethical lens. Material was excluded where voluntariness, contextual grounding, or ethical justification could not be established. Consent was therefore treated not as a single procedural event but as an ongoing ethical process shaped by participant readiness, safety, and the relational conditions of disclosure.
10. Analytical Strategy
The analytical approach adopted in this study was designed to engage with the heterogeneity of the dataset while maintaining systematic rigor. Given the diversity of contexts, interview types, languages, and conditions of data generation, analysis could not rely on a single standardized coding scheme. Instead, it proceeded through cross-context comparison and pattern recognition, grounded in the tri-axial framework introduced above. This approach is consistent with qualitative traditions that treat analysis as iterative, interpretive, and context-sensitive rather than mechanically procedural (Braun & Clarke, 2006, 2021; Saldaña, 2021). The analysis focused on both narrative content and the conditions under which narratives were produced.
For each dataset item, a structured metadata profile was recorded to support tri-axial classification. Metadata included:
temporal location of the conflict;
geopolitical context;
interview environment;
type of engagement;
language and translation mode;
duration and continuity of interaction;
participant positionality;
conditions surrounding disclosure.
Classification proceeded iteratively rather than through rigid categorical assignment. Interviews were situated simultaneously across the temporal-historical, contextual-political, and interview typology axes. In cases where testimonies overlapped categories, classification prioritized the dominant conditions shaping narrative production rather than forcing singular categorization.
Ambiguous or hybrid cases were retained within the analytical framework rather than excluded. For example, narratives emerging within humanitarian programs could simultaneously contain survivor-led and semi-structured interview characteristics. Similarly, longitudinal therapeutic encounters could shift across narrative forms over time. Such overlap was treated as analytically meaningful because it reflected the instability and relational complexity of field conditions.
The analytical process therefore emphasized interpretive positioning rather than mechanical coding. The objective was not to reduce interviews into fixed categories but to identify patterns linking narrative form to the conditions under which narration became possible.
The first stage involved situating each interview within the three axes of the framework: temporal-historical, contextual-political, and interview typology. This classification did not function as the final analysis. It provided an organizing structure through which each testimony could be understood in relation to the conditions of its production. By locating each interview within a specific configuration of time, context, and mode of engagement, it became possible to examine how variations in narrative form corresponded to differences in field conditions, institutional context, and relational dynamics. This reflects broader qualitative guidance that comparison across cases should preserve context rather than dissolve difference into abstract categories (Fujii, 2012; Maxwell, 2013; Yin, 2018).
Cross-context comparison was then conducted by examining how similar experiences were articulated across different configurations. Accounts produced in settings characterized by institutional recognition were compared with narratives generated in environments marked by ongoing risk, repression, displacement, or legal uncertainty. Attention was directed not only to themes but to narrative form: length, sequencing, emotional register, degree of detail, pauses, compression, and shifts in emphasis. This approach draws on narrative and thematic traditions that recognize meaning as located not only in what is said, but also in how accounts are structured and delivered (Riessman, 2008; Braun & Clarke, 2021).
Pattern recognition focused especially on silence, fragmentation, and narrative variability. Silence was treated as analytically significant, not as an absence of data. Moments where participants paused, declined elaboration, changed direction, or limited disclosure were examined in relation to safety, trust, stigma, and institutional context. This approach is supported by scholarship emphasizing that silence in qualitative research can carry meaning and requires reflexive interpretation rather than automatic classification as missing data (Müller, 2024; Poland & Pederson, 1998). In research on sexual violence, this is especially important because disclosure may expose survivors to social, emotional, legal, or political consequences (Campbell et al., 2009; World Health Organization [WHO], 2001).
Fragmentation was approached in a similar way. Discontinuous accounts, shifts in chronology, partial disclosures, and abrupt changes in emotional tone were not interpreted as methodological weaknesses or signs of unreliability. They were analyzed as forms of narration produced within specific conditions of memory, trauma, language, and safety. Trauma-informed research cautions against treating non-linear or affectively altered narratives as deficient; such forms may reflect both the impact of trauma and the constraints of the interview setting (Campbell et al., 2009; Isobel, 2021). In this study, fragmentation was therefore examined in relation to the interaction between personal memory, relational trust, and the socio-political environment.
Narrative variability constituted a central focus of the analysis. Differences in length, coherence, and detail were examined through the tri-axial framework. Short, highly structured accounts were often associated with settings where participants faced legal expectations, limited time, restricted safety, or uncertainty about how their testimony would be used. Extended narratives more often emerged in contexts of sustained engagement, relational trust, or survivor-led disclosure. These patterns were not treated as indicators of narrative value. A brief statement could carry substantial methodological significance, just as a long account could be shaped by institutional pressures. Narrative depth was therefore analyzed as context-dependent, not as a measure of credibility, education, or communicative capacity.
Linguistic mediation was also incorporated into the analytical process. Interviews conducted through interpreters, or in English as a shared non-native language, were examined with attention to how meaning was negotiated across linguistic and cultural boundaries. Cross-language qualitative research has shown that translators and interpreters are not neutral conduits but active participants in the production of meaning (Squires, 2009; Temple & Young, 2004). The analysis therefore considered how gender, age, perceived affiliation, social trust, and the use of a second language shaped disclosure and narrative form. This was especially relevant where English created a degree of emotional distance that could either enable safer disclosure or reduce affective and cultural nuance.
Throughout the analytical process, interpretations were developed reflexively. This involved attention to the researcher’s role, the conditions of access, the presence of intermediaries, and the ethical constraints surrounding each encounter. Reflexivity was treated as part of analytic rigor, particularly because research on political violence and sexual violence is shaped by power, risk, and positionality (Fujii, 2012; Guillemin & Gillam, 2004). The aim was not to produce a single generalized account of survivor testimony but to identify patterns in how narratives are formed under different conditions.
The analytical strategy therefore moved beyond content analysis alone. It examined the relationship between testimony and the conditions of its emergence. By analyzing silence, fragmentation, and narrative variability across temporal, political, and interactional contexts, the study demonstrates that these features are not incidental. They are central to understanding how knowledge about conflict-related sexual violence is produced. This approach supports the broader methodological claim of the paper: qualitative data in this field are not merely collected; they are generated through interaction, constraint, and ethical negotiation.
11. Positionality
Within qualitative research on conflict-related sexual violence, the role of the researcher cannot be separated from the conditions under which testimony becomes possible. Access to participants, the development of trust, the pacing of disclosure, and the structure of narratives are shaped not only by broader political and institutional environments but also by how the researcher is perceived within the interaction itself. Positionality in the present study is therefore approached not as a reflexive addendum to methodology but as a constitutive methodological condition influencing both data generation and interpretation. The researcher’s role as clinician, qualitative researcher, long-term field participant, and witness shaped the relational environments through which testimonies emerged across different contexts.
The dual positioning of the researcher as both clinician and researcher introduced overlapping but distinct modes of engagement. Within clinical settings, interactions were shaped by therapeutic principles emphasizing stabilization, emotional regulation, continuity, and attention to participant safety. Narratives frequently unfolded gradually across extended periods of relational engagement, with disclosure shaped by processes of trust, care, and emotional pacing. In research-oriented settings, however, participants often understood the interaction differently, recognizing the encounter as connected to documentation, testimony, advocacy, or broader analytical interpretation. These differing expectations influenced how narratives were structured and what forms of disclosure became possible. In some contexts, the clinical dimension facilitated trust and enabled more layered and reflective forms of testimony. In others, the perceived research function encouraged more controlled, factual, or condensed forms of narration shaped by assumptions regarding institutional use or representation.
The position of witness also played a central methodological role throughout the study. Engagement with survivors of conflict-related sexual violence involves receiving testimonies shaped by experiences of war, displacement, social rupture, and loss. Witnessing within this context cannot be understood as neutral observation detached from relational consequence. It is involves occupying a position shaped simultaneously by listening, interpretation, representation, and ethical responsibility. Participants frequently evaluated not only whether they wished to disclose but also whether the researcher could be trusted to receive and carry testimony beyond the immediate interaction without exploitation, dismissal, or institutional misuse. The role of witness therefore influenced the conditions under which disclosure emerged, particularly in settings where participants associated testimony with political invisibility, historical denial, or prior experiences of institutional betrayal.
Long-term engagement across multiple geopolitical and institutional settings further shaped the research process. The dataset developed cumulatively over an extended period through repeated interaction within conflict zones, humanitarian environments, memorial spaces, diaspora communities, and clinical settings. This continuity significantly influenced access, participation, and testimonial form. In certain contexts, ongoing presence facilitated relational familiarity and trust, allowing participants to revisit narratives, elaborate experiences, or engage in more reflective forms of disclosure across time. In other contexts, limited duration of engagement restricted interaction to brief or situational encounters, resulting in more constrained and highly regulated forms of testimony. Differences in narrative depth therefore reflected not only individual participant characteristics but also the temporal conditions of the researcher’s involvement and the relational continuity made possible within specific settings.
Positionality also shaped access to participants and the composition of the dataset itself. Entry into field environments frequently depended upon relationships with local institutions, community members, humanitarian organizations, memorial staff, clinicians, interpreters, and informal intermediaries who mediated social legitimacy and trust. Access was therefore never neutral or procedurally guaranteed. It was shaped by the researcher’s perceived intentions, prior engagements, institutional affiliations, and broader relational positioning within existing social networks. These dynamics influenced which forms of testimony became accessible, which participants felt safe engaging, and under what circumstances interviews or disclosures could occur.
Trust emerged as a particularly significant relational dimension of positionality. Participants continuously evaluated the researcher’s ethical orientation, institutional role, emotional responsiveness, and perceived capacity to receive testimony safely. Trust did not function as a stable or universal condition. Rather, it shifted across settings and interactions, developing unevenly in response to context, language, relational continuity, and social mediation. In multilingual environments, trust was additionally shaped by the presence of interpreters or by communication through shared non-native languages, both of which influenced emotional expression, disclosure pacing, and narrative structure. The variability observed throughout the dataset was therefore inseparable from these relational conditions.
The influence of positionality extended directly to the form of testimony itself. The same participant might narrate experiences differently depending on whether the interaction was perceived as therapeutic, humanitarian, legal, memorial, or research-oriented. Narratives were not produced independently from the research relationship but emerged through interaction shaped by trust, perceived consequence, emotional safety, and expectations regarding representation. Recognizing this relational dimension does not diminish the validity of the qualitative material. Rather, it situates testimony within the social and methodological processes through which knowledge about violence becomes possible.
Within this study, reflexivity therefore functions not merely as a requirement for methodological transparency but as an analytical tool for understanding how qualitative data are produced under conditions shaped by ethics, relational dynamics, institutional structures, and power asymmetries. This perspective aligns with qualitative approaches emphasizing the situated and co-produced character of knowledge generation (Fujii, 2012; Guillemin & Gillam, 2004). At the same time, it extends these discussions into the specific context of conflict-related sexual violence, where testimony emerges through highly unstable relational, political, and emotional environments. Positionality is therefore integrated into the methodological framework not as a secondary reflection on the research process but as part of the structural conditions shaping access, disclosure, silence, and narrative form.
12. Limitations
The methodological framework developed in this study emerged from a heterogeneous, multi-context dataset generated across different geopolitical environments, institutional settings, and modes of engagement over an extended period of time. Such an approach necessarily introduces limitations that require explicit consideration. These limitations include the non-standardized character of the dataset, the retrospective integration of material generated across different purposes and contexts, the challenges associated with translation and cross-language communication, and the uneven distribution of qualitative material across regions, conflicts, and historical periods. At the same time, these limitations reflect the conditions under which research on conflict-related sexual violence becomes possible and should therefore be understood not simply as methodological weaknesses but as structural features of trauma-centered fieldwork itself.
The dataset does not conform to a single standardized research design. Qualitative material emerged through multiple forms of engagement, including formal research interviews, clinical encounters, humanitarian interaction, survivor-initiated disclosure, and field-based testimony generated through memorial visits and community engagement. Consequently, interviews varied substantially in duration, structure, relational continuity, institutional context, and disclosure depth. Conventional qualitative methodology often prioritizes procedural consistency in order to maximize comparability and analytic control (Yin, 2018). In the present study, however, comparability is established through the tri-axial analytical framework rather than through methodological uniformity at the level of data generation. Although such variability limits conventional forms of generalizability, it allows for analysis that remains responsive to the unstable conditions under which testimony emerges in contexts shaped by violence, displacement, and institutional fragility.
A further limitation concerns the retrospective integration of material collected across different periods and purposes. The dataset was not initially produced as a single unified research project but developed cumulatively through long-term engagement across diverse environments and forms of interaction. Retrospective synthesis carries the risk of imposing coherence upon qualitative material generated under differing methodological and relational conditions (Maxwell, 2013). This risk is addressed in the present study through explicit attention to the conditions of data production and through systematic positioning of testimonies within temporal-historical, contextual-political, and interactional dimensions. While retrospective integration may reduce the precision of direct comparison across certain cases, it also enables identification of broader methodological patterns that would remain less visible within isolated single-context studies.
Translation and cross-language communication introduce additional methodological constraints. Interviews were conducted across multiple linguistic environments, frequently involving interpreters or communication through shared non-native languages. Translation within qualitative research is not a neutral transfer of information but an interpretive process shaped by cultural meaning, emotional nuance, linguistic limitation, and relational mediation (Temple & Young, 2004; Squires, 2009). Differences in vocabulary, symbolic expression, and culturally embedded understandings of violence may influence how experiences are narrated and interpreted. The presence of interpreters introduces further complexity because communication becomes mediated through an additional participant whose positionality, social role, and relational presence may shape both disclosure and interpretation. These dynamics limit the extent to which testimony can be approached as direct or transparent representation of experience while simultaneously reinforcing the central methodological argument that qualitative data are produced relationally under specific linguistic and contextual conditions.
The distribution of qualitative material across regions, conflicts, and time periods is also uneven. Certain geopolitical contexts are represented by a larger number of interviews because of extended engagement, greater relational access, or sustained field presence, whereas other settings are represented by smaller numbers of testimonies shaped by limited access, instability, or ethical constraints. From the perspective of conventional sampling logic, such imbalance may appear methodologically problematic. Within the context of conflict-related sexual violence research, however, uneven distribution reflects the realities of field access, political instability, displacement, survivor vulnerability, and fluctuating conditions of safety. The dataset therefore mirrors the asymmetrical conditions through which qualitative testimony becomes available rather than an artificially balanced sampling design detached from field realities.
These limitations underscore the broader methodological difficulties associated with conducting qualitative research in environments shaped by conflict, forced migration, institutional fragility, and trauma-related vulnerability. They also reinforce the need for methodological approaches capable of engaging systematically with instability, heterogeneity, and contextual variation rather than attempting to eliminate them through procedural standardization alone. The framework developed in this study does not seek to resolve variability or impose artificial methodological coherence upon qualitative material generated under divergent conditions. Instead, it provides a structured approach for analyzing heterogeneity while preserving the contextual specificity of the environments through which testimony emerged.
Reframing these limitations as structural conditions rather than methodological failures shifts analytical attention away from ideals of procedural uniformity and toward the realities of qualitative fieldwork involving survivors of violence. Non-standardization, retrospective integration, linguistic mediation, and uneven distribution are not anomalies external to the research process. They are integral features of conducting trauma-centered qualitative inquiry across unstable and unequal environments. Recognizing this allows for a more accurate representation of how knowledge concerning conflict-related sexual violence is generated and supports methodological approaches aligned with the relational, ethical, and political realities of the field.
13. Conclusion
This study has examined a central methodological challenge within research on conflict-related sexual violence: how to engage systematically with qualitative material generated across divergent historical periods, political environments, institutional settings, linguistic conditions, and modes of engagement. Existing qualitative approaches frequently associate methodological rigor with procedural consistency, coherent sampling structures, and standardized forms of data generation. The analysis developed throughout this study demonstrates that such coherence often reflects retrospective methodological representation more than the realities of trauma-centered fieldwork itself. Research involving survivors of conflict-related sexual violence unfolds through instability, negotiation, relational mediation, ethical constraint, and uneven access, and these dynamics become embedded within the structure of the qualitative material produced.
The tri-axial framework introduced in this study was developed in response to this methodological problem. By organizing qualitative material across temporal-historical, contextual-political, and interview typology dimensions, the framework provides a systematic structure for analyzing heterogeneous testimony without erasing the conditions under which narratives emerge. Testimonies are approached as situated forms of knowledge shaped by specific configurations of violence, displacement, institutional mediation, relational trust, linguistic negotiation, and emotional regulation. This structure allows for comparison across diverse forms of qualitative material while preserving the specificity of individual encounters and the environments within which disclosure became possible.
A central contribution of the study lies in reconceptualizing methodological heterogeneity as an intrinsic feature of qualitative research conducted in contexts shaped by violence and instability. Differences in narrative structure, emotional expression, fragmentation, silence, duration, and disclosure depth are not automatically interpreted as indicators of inconsistency, unreliability, or methodological weakness. Instead, these variations are examined in relation to the conditions under which narration becomes socially, politically, ethically, and psychologically possible. Narrative form is therefore understood as inseparable from the relational and contextual environments shaping testimony itself.
The study further argues that qualitative knowledge concerning conflict-related sexual violence is condition-dependent. Testimony does not emerge independently from the circumstances of its production. Narratives are generated through interactions shaped by institutional frameworks, power asymmetries, ethical negotiation, translation, social risk, and perceptions of safety. The presence of interpreters, communication through shared non-native languages, humanitarian structures, asylum systems, clinical relationships, and memorial environments all influence how experiences are articulated, regulated, and interpreted. Attention to these conditions is therefore essential for any methodological approach seeking to engage responsibly with qualitative material generated within trauma-centered fieldwork.
This perspective carries important implications for methodological practice. It calls for qualitative approaches capable of responding flexibly to changing conditions of access, vulnerability, trust, and disclosure rather than relying exclusively on ideals of procedural standardization and methodological control. It also requires greater transparency concerning the realities of fieldwork, including the relational, linguistic, institutional, and ethical factors shaping data generation itself. Within this framework, methodological rigor is defined not through uniformity alone but through the capacity to engage systematically and reflexively with complexity, instability, and the limits of knowledge production.
The framework developed here is not intended to replace existing qualitative methodologies. Its purpose is to extend methodological approaches used in trauma-centered research by providing an analytical structure capable of accommodating the variability inherent in multi-context qualitative inquiry. Although developed within research on conflict-related sexual violence, the framework may also hold relevance for other areas of study involving displacement, political violence, forced migration, institutional vulnerability, and social marginalization, where testimony emerges under conditions of instability and constraint.
At the same time, the study has intentionally maintained a methodological focus. The analysis does not seek to construct a generalized theory of trauma or to present survivor testimonies primarily as empirical findings concerning violence itself. Instead, attention has remained directed toward the processes through which qualitative knowledge is generated, negotiated, constrained, and interpreted across differing environments of disclosure. Establishing such a methodological foundation is necessary for future research seeking to engage responsibly with complex qualitative material produced under conditions of violence and displacement.
Ultimately, the study argues for a shift in how qualitative research on conflict-related sexual violence is conceptualized. Qualitative data cannot be understood simply as information extracted from participants through stable methodological procedures. Testimony emerges through relational, political, linguistic, institutional, and ethical conditions that shape what can be narrated, how experiences are communicated, and which forms of knowledge become possible. Recognizing these conditions allows for methodological approaches that engage more accurately with the realities of trauma-centered fieldwork while remaining analytically rigorous and ethically accountable. The study therefore proposes a movement away from extractive models of data collection and toward an understanding of qualitative knowledge as produced through the conditions under which testimony itself becomes possible.