Caught in the middle: Local researchers’ experience of mundane ethical harms in crisis settings

Case studies

Each case study location (the Balkans, Colombia, Nepal) is characterised by protracted conflict and varying scales of disaster, accompanied by significant academic research and ethical concerns related to being ‘over-researched’ (Kappler, 2013; Kelly, 2021). Layered into these three case studies are histories of direct and semi-colonisation: Colombia and the Balkans were directly ruled under the Spanish, and the Austro-Hungarian and Ottoman Empires, respectively. Although Nepal was not under direct colonial rule, it was nonetheless influenced by the British colonial regime in India and remained subject to a system of internal colonisation under a hereditary regime (1846–1941), followed by three decades of international isolation under an autocratic Monarchy (1960–1990; Whelpton, 2005). These histories and experiences of colonisation and authoritarianism have influenced ‘ways of knowing’ about these regions. In the Balkans, ethical concerns centre on external misrepresentation and epistemic domination; in Nepal, they are intertwined with the challenges of working in the contested environment of post-war and recurring disasters; and in Colombia, they relate to navigating active, complex, and often violent conflict. Collectively, these cases demonstrate that research ethics in conflict or disaster settings cannot be approached through universalised frameworks alone, but must be responsive to specific political, historical, and social dynamics that underpin each context. As such, the Balkans, Colombia and Nepal each offer distinct yet instructive contexts for examining research ethics in conflict-affected and post-war settings.

In particular, in the Balkans it is the violent dissolution of Yugoslavia in the 1990s that has shaped contemporary research dynamics, with a body of conflict and peacebuilding literature predominantly produced by Northern scholars (e.g., Belloni, 2020; Chandler, 2006), whose limited grasp of regional complexities often led to reductive and stereotypical portrayals (Kaplan, 1993). This underscores the ethical risks inherent in externally driven knowledge production. Nepal’s contemporary history presents a similar trajectory, with a peace and conflict literature focussed on the Nepalese Civil War (1996–2006), which sought to dismantle monarchical and centralised governance structures. Its exposure to recurrent disasters, notably the 2015 Nepal earthquakes that attracted international aid and academic attention, adds an additional layer of complexity that positions it as a crucial case for examining the ethics of research in environments shaped by both conflict and disaster.

Colombia’s internal armed conflict has been protracted, multifaceted, and deeply entangled with organised crime, gender-based violence, and structural inequalities (GMH, 2013; González González, 2014). Despite the 2016 peace agreement between the Colombian government and the Revolutionary Armed Forces of Colombia-People’s Army (FARC-EP), violence persists and involves various non-state actors and emergent guerrilla groups. The country has experienced multiple disasters, such as the 1985 Nevado del Ruiz volcanic eruption and resulting mudslide that left 25,000 people dead. Colombia exemplifies the ethical challenges of conducting research in environments where peace is partial and violence persists in fragmented forms.

Procedural ethics in context

To better understand mundane harms, it is worth examining procedural ethics in these three contexts. As discussed above, procedural ethics serve as a means by which ethics norms are formalised and enforced, often reflecting the institutional logics and power structures of dominant research regimes. In creating a comprehensive dataset of global regulatory ethics frameworks, Tapscott and Machón (2025: 796) demonstrate that ‘countries worldwide have adopted versions of a regulatory structure designed for biomedical and clinical research, rather than the types of ethical considerations raised by social scientists and conflict researchers’. Further, they show how these regulations are adopted inconsistently. This is mirrored in our case studies since, as in other places, procedural ethics review does not guarantee ethical practice.

In the Balkans, developing and institutionalising ethical practices in the social sciences, especially in C&D research, is a relatively new and voluntary phenomenon and regulations vary across countries in the region. Bosnia and Herzegovina, for instance, has no mandated formal ethics review process (Tapscott and Machón, 2025). In Serbia, beginning in 2005, all health institutions are obliged to establish ethics committees, as prescribed in the Law on Health Care, while the social sciences continue to operate without mandatory ethics review (Petrović, 2015). Many academic institutions have established voluntary ethics committees, but they often remain performative. Until recently, it was common for researchers to write their own ethics approvals, and ER at universities and research institutions existed only on paper. In the last decade, however, this has changed for two reasons. First, mid-career scholars educated in the gN, mostly in the US and UK, attained management positions at domestic institutions and introduced practices similar to those of their alma maters. Additionally, after years of destruction and isolation, the region has slowly started to integrate into the European Higher Education Area. This has increased opportunities for international cooperation and research funding and means following the rules and procedures of research donors and partners, including ethics procedures. Thus, although the Balkan research context differs from that of gN institutions in terms of available resources, institutional support, and research traditions (Džuverović and Tepšić, 2020), the main approach in this process has been to copy-paste the practices and norms of gN institutions. Many researchers still see ER as a ‘Western invention’ or ‘necessary evil’ required to secure funding (see Kovačević, 2008), and in practice, fieldwork often differs from that proposed in ER applications, assuming no one will check. While this discrepancy can be attributed to a disconnect between standards and context, a full explanation must also consider researchers’ perceptions of ethics review and its necessity.

In Colombia, ethics guidelines are rooted in the health sector but have become more inclusive of various disciplines, including social sciences. Resolution 8430, issued by the Ministry of Health, established standards for health research and subsequently became the national benchmark (Ministerio de salud, 1993). Notably, this resolution defined protections for vulnerable groups within research protocols. The Colombian Ministry of Science, Technology, and Innovation currently serves as the primary regulatory body and leads the implementation of research ethics and scientific integrity policies across both public and private sectors. The ministry has issued regulations concerning specimen collection, biodiversity, access to genetic resources, and the protection of indigenous and local knowledge (Castañeda-Ruiz et al., 2020). A senior researcher with four decades of experience pointed out, however, that for most of her career, informed consent was nonexistent; instead, it was trust with communities that governed the research relationship (Interviewee C3). Despite better developed policy frameworks, a gap exists between these policies, ethics guidelines, and research practice, evidenced by the reduction of ethics standards to checklists and the lack of institutional support, funding, and training for university ERs. Consequently, ERs are often insufficient, highlighting the need for better regulation, and especially ER training about the protocols themselves and its follow up; participant protection, and community-based protocols.

In Nepal, as in the other two contexts, biomedical ethics regulations tend to dictate the ethical landscape, and social science ER is inadequate and hardly enforced. As one interviewee with extensive international disaster research experience said, ‘People take [research] ethics very casually here’ (Interviewee N5). A major policy institute report affirmed this statement, suggesting research misconduct in Nepal is increasing due to the lack of adequate and enforceable codes of conduct and associated oversight bodies (Policy Research Institute, 2024). Nepal’s Health Research Council is the primary formal ER board, recognised by the Forum for Ethical Review Committees in the Asian and Western Pacific Region (FERCAP) since 2019. Universities in Nepal have their own ethics boards but they predominantly focus on health-based research. In general, humanities and social science human subject research tends to undergo less comprehensive ER processes; sometimes it is fast-tracked or even sidelined. The University Grants Commission of Nepal, the body responsible for developing policies and monitoring and evaluation guidelines for universities, published a ‘Good academic research practices’ (University Grants Commission (Nepal), 2021), but it focuses on the ethics of publications, not research conduct. In some institutions with ER, the process is generic with regard to consent and maintaining anonymity. Other newer research institutes have developed more rigorous ER mechanisms. Overall, however, ER in Nepal’s academic sector remains inconsistent and underdeveloped.

Absent strong and consistent national ER in all three cases, foreign researchers continue to defer to their own ER since requirements vary between countries and universities. Clearly, the existence of regulations does not guarantee ethical practice, especially without effective accountability mechanisms. As one interviewee summarised: ‘If there is no cop, you don’t drive 50 miles [per hour]. In the last project they had an ethics committee which never met during the project. Not a single time. So, it is only about the form, not about the substance. No one really cares once you get the approval’ (Interviewee B3).

The dynamic ethics-context interface and knowledge hierarchies

‘Context’ emerged as a critical factor in local understandings of ethics and ethical misconduct. To avoid harm, ostensibly researchers should understand the social, political, and cultural context of the communities they research. Yet, the way scholars choose research sites and produce research outputs, and thus create knowledge, as promised in the procedural ethics, is rarely subject to the dynamic nature of the contextual realities. Instead of being viewed as a fixed and standard entity, our interviewees defined context as a dynamic force and sought to connect it with the ethics of knowledge production. In other words, the unequal context in which research and research partnerships occur is seen as capable of facilitating the exclusion and de-legitimisation of local research expertise. However, in some instances, context is viewed favourably, allowing local researchers to claim their expertise and legitimacy in knowledge generation over their international counterparts.

In our interviews, local researchers often found that their international counterparts displayed a ‘colonial way of thinking. [International researchers] think that [their] way of working and doing things . . . is the same everywhere else’ with regard to ethics and collaborations more generally (Interviewee B1). In Nepal, the broader aid-receiving context has led to much research being donor-driven, focussing on evaluating aid interventions. Such research typically adheres to predefined international aid standards instead of local norms. Moreover, our research suggests that commissioned research does not consistently respect ER principles, such as keeping informants safe and away from potential risks, avoiding re-traumatisation of victims or providing researchers to work free of donor influences, despite researching sensitive topics (e.g. violence and trauma among conflict-affected communities, discrimination in disaster relief distribution). One example given was that insufficient attention is paid to the consent of research participants in the publication and promotion of photographic evidence, widely favoured by aid agencies. ⁶ Local researchers report obtaining ER approval does not guarantee sensitive conduct, nor do international researchers necessarily pick up on the hidden transcripts of interviews: ‘Local researchers understand social and human aspects and they are more sensitive in dealing with subjects (for example what kind of questions they should ask, when they ask, how they ask, and so on). Also, if you have witnessed some violent events, you are much more careful in conducting research’ (Interviewee N4). Others, however, warned against over-reliance on local researchers as if they are inherent experts about all cultural and linguistic realities: ‘Foreigners think that one Nepali will understand all of Nepal, but that is not true. We are not homogeneous’ (Interviewee N4). Researchers from the capital, Kathmandu, may be equally prone to misjudge local nuances as foreigners. As one stated, ‘We need to understand that all these international policies sometimes are not contextual – you cannot see Kathmandu from Geneva and you cannot see local communities from Kathmandu’ (Interviewee N4).

Colombian researchers found themselves caught between helping donors understand the context and the challenges of research in a context of violence. For instance, one interviewee reported that donors requested research findings after an escalation in violence, not understanding that research participants may be afraid to answer questions because of the violence (Interviewee C2). Donors want research reports and findings, but for local researchers, navigating the context can be as important as the results. This navigation requires familiarity with the visible and invisible community rules as well as respect for and safety of the communities (and researchers). Lacking this contextual knowledge can lead to different perceptions of ethical conduct. Following implementation of the 2016 Peace Agreement, some former combatants were elected to public office. Donors, however, refused to fund peacebuilding projects in those municipalities because they were unfamiliar with the legal status of these former combatants. One participant stated: ‘There is a significant difference between my work and my organisation’s vision. These tensions are very typical within such ethical processes’ (Workshop C participant).

Yet, when it comes to contextual factors of research, there was a clear sense amongst local researchers that ‘Most of the internationals do not fully understand the context, they cannot understand shades of grey and the nitty-gritty of problems. They are still colonising facts they acquire by interpreting data from their epistemological position and understanding of the context’ (Interviewee B4). This leads local researchers to assume international researchers possess only superficial regional or place-based contextual knowledge and, in the process, assert their authority over contextual dynamics. One stated, ‘They know a lot about the region from the literature, but they lack peculiarities that come from the lived experiences in the field’ (Interviewee B5). In Colombia, local researchers found themselves both translating and facilitating international counterparts’ understanding of the complexities of the country’s internal armed conflict: ‘more than the language barrier, these [international] researchers found themselves overwhelmed by the amount of information they didn’t understand. I remembered some researchers that wanted to visit a FARC dominated zone and they felt they didn’t have the instruments to understand the context’ (Interviewee C1).

There is also resentment of the dominance of English-language publications (Roth, 2019). As one stated, ‘There is a lot of local knowledge, but if it is not in English or not well written and they cannot publish internationally, the knowledge stays hidden’ (Interviewee B3). Even if well-known regionally, local researchers and their work are often not acknowledged externally. Additionally, local researchers often lack either the experience or language skills to publish in gN publications, and the incentive to do so, since their institutions do not necessarily reward publication in these venues.

The ethics of over-researched communities

A second type of mundane ethical harm in conflict and disaster-affected settings relates to repeated data collection about communities experiencing difficult circumstances. We identified two interconnected ethical issues here: first, the burden of repeated participation in data collection, related to time, repetition, and potential traumatisation for individuals who experience violence or disaster; and second, the expectation that participating in research would lead to tangible changes or increased services to the community. Workshop participants and interviewees across all three contexts consistently identified this as a major concern in their work.

As topics become popular and (different) researchers return to collect more data on existing problems, they collectively contribute to the problem of over-researched societies and communities (Kappler, 2013; Kelly, 2021). Although there are no established criteria for identifying an over-researched society, it is generally understood to refer to a context in which ‘too much fieldwork has been conducted, too much attention has been paid to a particular community or issue, and, by implication, some research could be deemed repetitive, unnecessary or redundant’ (Kelly, 2021, 49). In these instances, researchers ask questions of and listen to the same, often English-speaking, voices. This does not mean that being ‘over-researched’ is a consistent experience. Instead, it is a highly localised phenomenon that communities experience differently. For instance, Mostar and Srebrenica in Bosnia and Herzegovina, Vukovar in Croatia, the regions of Urabá and Montes de María in Colombia as well as other conflict-affected communities in Nepal, and also Northern Ireland, Cyprus and Rwanda, have experienced significant levels of ‘research fatigue’ (Workshop C) or ‘participant fatigue’ (Workshop B). This fatigue manifests in communities feeling exploited, exhausted by, or increasingly distrustful of researchers, largely due to the continuous influx of studies focussed on conflict, violence, aid, and transitions from war to peace or immediate relief and reconstruction efforts after disasters (Fast et al., 2025; Kappler, 2013; Kelly, 2021). The fact that qualitative researchers tend not to share their data from crisis-affected spaces (Vogel et al., 2024: 23–26) further intensifies the process of repeated data collection.

One factor here is attention to a few, high-profile topics that stimulates more research, but that may illustrate a lack of contextual knowledge that could identify other subjects deserving of more scholarly inquiry. One interviewee summarised it this way: ‘There is some kind of fetishisation of research because everybody wants to see the same things because they read books written by global Northern researchers; nobody wants to research something relevant and contextual [to us]’ (Interviewee B5). Importantly, participants stated that even ‘over-researched’ communities were not always opposed to research participation, showing enthusiasm for sharing other positive aspects of their culture, such as music, that researchers too often overlook in favour of the negatives, such as stereotypes or violence (Workshop B and N). This problem is not limited to academic research, since humanitarian and development organisations frequently conduct needs assessments, programme evaluations, and operational research in the same communities. These efforts, driven by different objectives, contribute to over-research, particularly when characterised by limited data-sharing (Workshop C).

Some, especially from CSO participants, lamented over-collecting data when the problems seemed obvious, suggesting instead that organisations should spend money on responses over more research. One participant described it as a drive to ‘over-scientificate’ problems in Bosnia-Herzegovina that are well known to policy makers (Workshop B). Another participant shared a post-disaster context example, where the acute deprivation of affected populations was immediately evident. Nevertheless, local evaluators were tasked with collecting data framed as ‘stories’ from affected populations, thereby positioning them in a complex ethical and professional dilemma (Workshop N). The frontline data collectors were required to justify both the rationale for, and the implications of, extracting narratives that merely confirmed the easily observable social realities of the community (Workshop N).

In other instances, local researchers are tasked with collecting photos of disaster victims as part of the growing international push for alternative ways of knowing, such as photo or video storytelling in disaster research. One interviewee with extensive experience of researching disaster-affected communities highlighted concerns about the lack of ethics guidelines surrounding the collection, ownership and dissemination of photos in Nepal. He shared his anxiety related to challenges from community members while taking photos: ‘They will say, you had taken pictures and what did you do with it’ (Interviewee N2). This illustrates the pressure local researchers face in justifying the use and value of alternative forms of knowledge production. As the literal face of the research process, local researchers confront challenges in navigating the tension between respecting the privacy of community members in contexts of hardship, and fulfilling external expectations tied to securing funding for relief interventions. The requirement to document and report on self-evident conditions not only strained their community relationships but also jeopardised the sustainability of future research and engagement activities in the same area. This raises the issue of how to ethically define questions, design research, and identify research locations in conflict or disaster contexts.

A second and related ethical issue of (over-)researching is the expectation of communities that their involvement should result in change or action by external actors. Speaking about a particular Nepali community, one local researcher (Interviewee N4) observed:

They have also been overly researched. The same person is interviewed over and over again. It is my personal ethical principal that I will not promise them anything that I cannot deliver in future.

Scholars usually consider knowledge production as their primary goal, with the potential to change future interventions as a secondary contribution. Accounting for multi-year projects and publication cycles, however, it is often impossible for communities to see a link between their participation and future change. Local researchers, who are connected to communities or return to them for (new) projects and data collection, directly face the expectations for change, regardless of whether their international partners are aware of them. As one participant stated, ‘Local researchers (. . .) are left to deal with the expectations, frustrations and resistance of the local communities long after foreign researchers are gone’ (Workshop N). One Nepali researcher changed their approach to research and communication to mitigate expectations: ‘My ethics is to tell people exactly what we are doing. Not even to talk about the future of what might happen’ (Interviewee N4).

The ethics of recognition

Our last point relates to the ethics of recognition. This came through in two ways: first, at the level of local researchers, and the all-too-common lack of acknowledgement of their contributions in (international) collaborations; and second, at the level of research participants, who object to the blanket approach of data anonymisation that can silence their voice and agency.

Across all three case studies, local researchers reported countless occasions of receiving only a ‘thank you note at the end of the page’ (Interviewee B5), and instances where researchers are not acknowledged in the final product (Workshop C and N). This was true ‘even though they [global Northern researcher] got the context, data, interviews all because of’ the local researchers (Interviewee B5). Some cynically reported that ‘They [international researchers] get together for a dinner, drink a lot of alcohol and in the morning, they say they will write the paper/book on my idea. They have natural advantage in writing [in English] but crippling disadvantage in knowing the context’ (Interviewee B5). This goes against academic convention to give credit based on contribution. Participants in Colombia also questioned the unequal relationship between foreign and local researchers, which even undermines local research. There seems to be a social order where the expert is the one who comes and goes, but not the local researcher who has all the information. ‘For foreigners, you are just one professional, nothing more’ (Workshop B).

In other cases, workshop participants reported that drafts they wrote were, slightly edited, published as journal articles by gN scholars. Fearing lost future employment, and therefore a vital income, or unsure where or how to report this type of plagiarism, local researchers have not challenged such unethical practice (Zapata Cancelado et al., 2025). While there might be instances where local researchers wish to remain anonymous because of national politics and repression, or a desire not to be associated with certain research projects and themes, ethical practice and academic convention suggests local researchers should retain agency and that these decisions should follow from an open conversation between all parties, ideally before the start of the research.

Second, and somewhat surprisingly, workshop participants pushed against the tendency to anonymise subjects in reporting research findings. Many ER processes encourage or even require anonymising research subjects, particularly in politically volatile situations, to protect them from harm. Many researchers, however, reported that participants demanded explicit acknowledgement as a condition of participation, as they felt this was one of the few areas of agency and advocacy left to them (Workshop C). In Colombia, one researcher reported that communities are growing more distrustful of researchers and research extractivism, as they see no tangible changes from these studies, saying ‘they are absolutely right to distrust us, since they don’t get something out of it’ (Interviewee C3). Thus, research can serve to build trust that extends beyond the scope of a specific project. Exerting a choice to be recognised implies a more equal relationship, an acknowledgement of the needs of the communities and a corresponding contribution from researchers in response to those needs. This contribution need not be limited to material resources; it could include offering expertise and professional support. In Colombia, for instance, one community chose to favour research processes that emphasised co-creation and capacity building. These approaches valued their active participation and produced tangible benefits for the community (e.g., works of art, public infrastructure, local expertise, local history book) that went beyond academic publications (Workshop C).