“Vulnerability”: The Trouble with Categorical Definitions in Institutional Ethical Reviews, Forced Migration Research & Humanitarian Practice

Labeling refugees as “vulnerable” creates moral categories that have profound implications for individuals’ and communities’ lives, both in the present and future. Within the international humanitarian regime, vulnerability does not just describe a human condition, but has become a label used to classify some migrants as more deserving of assistance and protection than others. Yet this labeling has also become critical for institutional ethical approval in forced migration research and in humanitarian practice. In this essay, I discuss the use of vulnerability in these domains in order to reflect on some key dimensions in the meeting point between research ethics, humanitarian interventions, and our conduct as researchers more generally. While these domains are intrinsically interconnected, I show that, in practice, they are often disconnected.

To reconsider our application of vulnerability in discussions on research ethics, I present two entry points from which to discuss the use of vulnerability. The first is an international research collaboration in which the Centre for Lebanese Studies (CLS) was a partner. The research analyzed the use of “vulnerability” in the international protection regime for migrants. The second is the work at the CLS to establish its own ethical review process and identify where vulnerability also played a role in formulating the guidelines. From these two points, I seek specifically to reflect on the notion of vulnerability in relation to research with refugees, in ethical review processes, and in ethical scholarship more generally. Critical to my analysis is how ethics is often separated from the political and ethical conditions of the research process as a more procedural and bureaucratic exercise. Through the discussion of how vulnerability was used in the research project and in the ethics framework we are developing at the CLS, I reflect on both how we can create a closer relationship in our research between the procedural and a more holistic ethical consideration, and what implications this will have for the use of the concept of vulnerability.

The focus here is on how vulnerability is used and understood in the specific context of ethics reviews and refugee responses and the implications of using the concept in research and interventions in Lebanon.1 As a concept, vulnerability may be understood both as boundary-drawing and boundary-less.2 As a boundary-­drawing concept, vulnerability is a category-based notion. It suggests an idea that is clear on who should be labeled as vulnerable—even in law. As a boundary-less concept, vulnerability refers to a more generalized condition of being human; it is an open and more blurred notion. As a boundary-less concept, vulnerability is commonly applied to understand the ways in which humans are fundamentally constituted through relations.3 In that relational human condition, there is a primary vulnerability despite the insight that precarity and vulnerability are unevenly distributed among individuals.4

These two broad meanings of vulnerability have created profound tensions and ambiguities in how the concept is used in processes to determine who qualifies for assistance and protection, and for how we approach research ethics and thus methodology and knowledge production more generally. The concept has traveled between refugee situations across the world, but this conceptual journey is not straightforward.

Two points of entry led to the reflections in this essay. The first was a research collaboration on vulnerabilities in the global protection regime; the second was the work establishing an ethical review process at the CLS.5

Between 2019 and 2023, the CLS partnered with six other teams from Belgium, Canada, Germany, Italy, Norway, and Uganda in an EU Horizon 2020 project entitled Vulnerabilities under the International Protection Regime (VULNER). The project sought to understand how the notion of vulnerability was used and understood in the legal protection regime for migrants seeking protection in those partner countries and the ways in which migrants who encounter those regimes experienced it. I came into the project about halfway through its lifetime when I joined the CLS in the summer of 2021.

The research collaboration had developed a relatively standardized and comprehensive approach to case studies in the seven countries, with some measures in place for contextualization of questions and procedures. In its first stage, the project analyzed policy documents and case law and interviewed policymakers, humanitarian and social workers, and legal practitioners. The second stage concentrated on the experience of the protection seekers. In Lebanon, we interviewed fifty-seven Syrian and Palestinian refugees about their experiences with the humanitarian system and governance structures in the country.6

A critical component in the project was the adoption of a common ethical protocol, the VULNER Common Ethics Strategy.7 The guidelines were aligned with common institutional procedures by focusing on the principles of informed consent, confidentiality, “do no harm,” and data protection. In line with standard institutional ethical procedures for research in many contexts across the world, the guidelines took for granted the notion of vulnerability and the assumption that refugees are vulnerable. It was also stated in VULNER’s common ethical framework that the professionals we interview “such as experts, practitioners and other relevant stakeholders but not migrants” are nonvulnerable research participants.8 In addition to the project-specific guidelines, the project also applied the European Commission’s “Guidance Note on Research on Refugees, Asylum Seekers & Migrants” and Christina Clark-Kazak’s “Ethical Considerations for Research with People in Situations of Forced Migration.”9 A final set of ethical guidelines were the respective national guidelines of the partner countries, which differed in level of detail and procedure. The Centre for Lebanese Studies had institutional review board (IRB) approval from the Lebanese American University (LAU), which we were affiliated with at the time of starting the project. It is fair to say that the basis for all these guidelines and principles, including LAU’s IRB, can be traced back to the 1979 Belmont Report (which I will return to).

We saw research ethics as much more than a box-ticking exercise in the project. To ensure ethically sound research beyond the standardized guidelines, there were regular meeting points that enabled continued ethical discussions during the implementation of the project. The discussions generally went far beyond the guidelines formulated. The project also had a specifically dedicated ethics advisor who researchers could consult when they felt the need to discuss ethical dilemmas. The ethics advisor was a senior academic based in the United Kingdom with long-term experience with international research. VULNER’s common ethical framework was vital as it enabled ethical discussions as part of the ongoing project discourse.

In the end stages of the project, I was invited to contribute to the final publication—an edited book—with a reflection on ethical questions associated with the research project.10 Because I joined the project late, I had not been involved in formulating the framework; and to acknowledge my own shortcoming, this was the first time I closely studied VULNER’s common ethical framework and analyzed its content.

I think this lack of more thorough engagement with the framework on my part was most likely because I considered it mainly as just another procedural text that did not contribute much to the deeper and more holistic discussions we had on research ethics in the project. Hence, I had not paid much attention to the tension between the common ethical framework’s definition of refugees as vulnerable and the research we were conducting on critically assessing the use of the notion of vulnerability. In hindsight, I can clearly see the irony of this: taking vulnerability for granted in our ethical guidelines, on the one hand, and critically exploring the meaning of vulnerability in our research, on the other. However, perhaps this can also be seen as a clear indication of how the procedural guidelines are often delinked from the wider knowledge production process and the oftentimes limited value of those guidelines, both in terms of the wider political questions of knowledge production and in the dilemmas that emerge in the practical conduct of the research. An essential lesson from the project was also the value of the ongoing discussions and collective reflections on ethical dilemmas, which are richer and often more helpful than the procedural guidelines put in place at the beginning of a project. A key question I take with me, then, is how could the ethical principles and the research we conducted be further aligned and integrated?

The second entry point to this essay has been the work over the past year with the Centre for Lebanese Studies in creating an ethical review process.

As mentioned above, the CLS previously received its ethical approval from the institutional review board of the Lebanese American University, with which we were affiliated until February 2024. When our agreement with LAU terminated, we saw this as an opportunity to develop our own ethical review process. We hoped that a review process at the CLS would allow for a more contextually sensitive approach, more active learning and discussion on research ethics in our organization, and scholarship that is more directly situated in a relational ethics than the traditional institutional review board process we had followed so far.

As we progressed, we carried out an exhaustive study of the way other institutions handle ethical reviews. While developing the ethical review process, and in the discussions with colleagues both individually and in groups, there were some concerns about our ability to conduct such ethical reviews: What status would publications have if based on an approval from the CLS and not from a university IRB? Would there be implications for colleagues’ potential to publish, and how would it affect their careers? Would funders take this kind of ethical review seriously? Would we have the capacity—and the competence—to conduct such reviews?

Many institutions across the world with an IRB are registered with the U.S. Department of Health and Human Services and their Office for Human Research Protection. However, as far as I have been able to determine, there is no general global condition that institutions must meet for conducting their own ethical reviews. What is important, and is very well described by Mudar Kassis and colleagues, is that there are certain legal and procedural conditions and standards that need to be in place for the institution to be accountable to all implicated parties and that require standardized ethical guidelines.11 As an institution, we need to make sure that we are able to meet these standards. Most CLS research falls under the social sciences and humanities and spans a wide range of methodologies and activities. Another challenge is to make sure that the standards we develop can cover these different approaches and interests. In formulating an ethical review process, we have started a practice of creating more discussion on research ethics among colleagues; hopefully, we can continue and situate these discussions in a broader context than the procedural ethics and toward understanding our work in a broader historical, social, and political framework. We are, I believe, in the middle of that process.

As a small organization, we decided that we would evaluate applications internally as a team and as peer reviewers. There is an application procedure, and two colleagues who are not directly involved in the research will review the application. All colleagues are invited to a seminar to present and discuss the proposals. Amendments are made based on peer reviewers’ and seminar participants’ feedback. We also plan to have discussions at the midway point and upon completion of a project, both to explore dilemmas and for all of us to be involved in the learning. In more sensitive cases, where we as peers may find it challenging to make the decision on whether a project is ready to go ahead, we will invite an external reviewer from outside the CLS who has deep contextual and thematic knowledge and who will review and discuss the project with us. We termed this “tier 2.” We created criteria for when a proposal should go through the tier 2 process, such as questions around safeguarding, data protection, moral issues, and conflicts of interest. What I want to discuss here, however, is how the tier 2 review is initiated when “vulnerable groups” are included in the proposal.12 Our approach was inspired by existing review processes in other institutions, most of which mention “vulnerable groups” in their guidelines. The problem is, as discussed above, that vulnerability as a group-based label should not be taken for granted. Thus, in order to make sense of what to do with vulnerability in research ethics and, more important, how to align better our ethical discussions and ethical review procedures, I will discuss further the use of the concept of vulnerability in ethics reviews and in humanitarian practice, as well as how refugees we interviewed in Lebanon understood the concept.

We may think of research ethics in two different—albeit overlapping—ways: The first is the more standardized institutional system for ensuring that research is conducted in accordance with the institutions’ definition of ethical research, as well as legal conditions and commitments. This procedural understanding of research ethics is often connected with a utilitarian or deontological ethics.13 The second is understanding what an ethical research encounter is, and hence emphasizing a relational research ethics with more attention to the wider historical, social, and political implications of research and knowledge production. As I reflected on in the previous section, the broader relational ethical considerations are seldom integrated into the institutional procedures.

Research ethics as standards and principles emerged in the medical sciences and date back to at least the nineteenth century, resulting after World War II in the Nuremberg Code for permissible medical experimentation on humans.14 Later ethical reviews moved from a disciplinary focus to a more generalized interdisciplinary approach. The Belmont Report, published in 1979 by the U.S. Department of Health, Education, and Welfare following the 1974 National Research Act, is the basis on which many universities, including the Lebanese American University and the American University of Beirut, base their ethical review process. These processes focus on the core principles of respect for persons (and their autonomy), beneficence, nonmaleficence, and distributive justice (ensuring benefits and burdens are shared equitably).15

There has been abundant criticism of institutional review boards and institutional ethical reviews more generally.16 Sheeva Sabati summarizes the critique against IRBs in three points: first, “the inadequacy or irrelevance of evaluating social research through principles designed for positivistic, biomedical modes of inquiry”; second, “the blurriness of institutional ethics and the more bureaucratic control of IRBs” such as the risk aversion, audit culture, and instrumentalist cost-benefit analyses in which universities’ legal liabilities have become the primary concern; and third, that IRBs have become gatekeepers to the advancement of new knowledge.17 This critique is firmly related to neoliberal forms of governing knowledge production, in which ethical reviews become a kind of monitoring mechanism and the audit culture is there as much to protect the institution and the researcher as the research participants.18 This rigid form of ethical review is often incompatible with more dynamic and open research such as, for example, ethnography, action, and activist research.

Vulnerability has to some extent come to be at the center of this approach. The Belmont Report mentions vulnerability three times: first, in the context of voluntariness and the risk of coercion that may occur due to a subject being especially vulnerable; second, emphasizing the need to justify why vulnerable people need to be included in the research by weighing risks and benefits; and, finally, in the closing paragraph of the report, in the context of the selection of subjects, justice/injustice, and the identification of vulnerable groups:

One special instance of injustice results from the involvement of vulnerable subjects. Certain groups, such as racial minorities, the economically disadvantaged, the very sick, and the institutionalized may continually be sought as research subjects, owing to their ready availability in settings where research is conducted. Given their dependent status and their frequently compromised capacity for free consent, they should be protected against the danger of being involved in research solely for administrative convenience, or because they are easy to manipulate as a result of their illness or socio­economic condition.19

Again, as with IRB processes more generally, there has been abundant discussion on the notion of vulnerability in The Belmont Report.20 At the core of this discussion is the problem that vulnerability is poorly defined and that the ethical justification of applying the principle of vulnerability is often vague.21 So far, most critical accounts of vulnerability have focused on providing more comprehensive and nuanced definitions of the sources and types of vulnerability. This has led to different and increasingly more expansive lists of who should be categorized as vulnerable. The Belmont Report includes racial minorities, the economically disadvantaged, the very sick, and the institutionalized. Twenty-three years later, in the 2002 version of the Council of the International Organisation of Medical Sciences’ guidelines, this list expanded substantially to include:

• Those with limited capacity or freedom to consent or to decline to consent . . . [including] children, and persons who because of mental or behavioural disorders are incapable of giving informed consent;

• Junior or subordinate members of a hierarchical group . . . [such as] medical and nursing students, subordinate hospital and laboratory personnel, employees of pharmaceutical companies, and members of the armed forces or police;

• Elderly persons;

• Residents of nursing homes;

• People receiving welfare benefits or social assistance and other poor people;

• The unemployed;

• Patients in emergency rooms;

• Some ethnic and racial minority groups;

• Homeless persons;

• Nomads;

• Refugees or displaced persons;

• Prisoners;

• Patients with incurable disease;

• Individuals who are politically powerless; and

• Members of communities unfamiliar with modern medical concepts.22

However, this category-based approach does not reflect a nuanced understanding of the ethical risks to be taken into consideration when labeling someone as vulnerable, and does not reflect the majority of studies that promote a more nuanced understanding of the categories of vulnerability, rather than moving away from vulnerability altogether.

Samia Hurst has shown that The Belmont Report defines the vulnerable explicitly on grounds of “their dependent status and their frequently compromised capacity for free consent.”23 Thus, the implication is that vulnerable individuals are devoid of agency in the research process and beyond, which adds to the general narrative of vulnerable populations and risks attaching stigma to groups and individuals. This approach has been much discussed in refugee studies in relation to the understanding that refugees are presented as vulnerable victims that “are incapable of determining ethical practices that align with their needs/values.”24 Rather than protecting groups defined as vulnerable, such beliefs deprive people of agency in the research process.

Another critique against the categorical understanding of vulnerability promoted by The Belmont Report is the problem of individualizing risk rather than looking into the deeper analytical and historical frameworks that reproduce inequality and oppression.25 Rather than abandoning vulnerability, many of the discussions on the meaning of vulnerability in ethical reviews call for approaching vulnerability as relational and analytical to understand better the sources of vulnerability.

Despite the unending nature of displacement in Lebanon, a humanitarian approach of what a Kenyan participant in Cindy Horst’s research described as “don’t die survival” continues to dominate the response to refugees under the conditions of temporary presence.26 The host state has no interest in de facto or de jure integration between its people and forced migrants. At the same time, the principles of neutrality, impartiality, and independence continue to be strong ethical imperatives among humanitarian organizations mandated to assist “vulnerable populations.” Based in this deontological ethical approach, humanitarian organizations show little willingness to reflect deeper on the more political role of their own conduct. It is in this context that vulnerability is being used as an active tool for labeling receivers of aid.

In humanitarianism and specifically in the context of refugee studies and policies, vulnerability is a relatively new concept.27 Hande Sözer has shown that vulnerability started to be used in some academic discussions connected to asylum and irregularity in the early 2000s. From 2011, it was used much more widely and specifically as a response to the Syria crisis. In humanitarian settings, vulnerability has often been defined as the diminished capacity of an individual to anticipate, cope with, resist, and recover from the impact of a natural or humanmade hazard.28 Over time, however, we can identify a shift in the use of the notion of vulnerability from this boundary-less concept to a more boundary-drawing concept: Sözer describes this as a semantic shift from understanding forced migrants’ vulnerability to vulnerable forced migrants.29 With this shift, vulnerability stops representing a human condition and becomes a label classifying some forced migrants, resulting in differential access to assistance and protection. Vulnerability becomes a specific form of ordering, with distributive and redistributive aspects closely aligned with the neoliberalization of humanitarian assistance and policy, which emphasize individual responsibility and measuring efficiency and return on investment in interventions.30

Prime examples of this shift include the Vulnerability Assessment Framework (VAF) in Jordan and the Vulnerability Assessment of Syrian Refugees (VASyr) in Lebanon, which identify particular subcategories of refugees as vulnerable. The frameworks were introduced in 2013 to curb the financial and operational burdens of humanitarian engagement; they enabled a prioritization among different groups of refugees who were most in need of assistance at one specific time.31

The category-based understanding of vulnerability is confusing because humanitarian organizations often use arbitrarily designed vulnerability criteria.32 In the same country context and often within the same organization, the criteria for labeling a forced migrant as vulnerable differ from project to project, often depending on the funder’s conditions. Thus, the labeling of vulnerable groups is governed by the financialization of assistance. Generally, vulnerability is used to make decisions to support some vulnerable groups over others, rooted in the argument that funding is limited and difficult choices must be made. Here, economic and neoliberal logics trump any attempt to create a moral and ethical argument. Sigrid Lupieri, for example, shows that older refugees, although often scoring high on the vulnerability criteria, score low on the value for financial vulnerability “due to reduced life span, disproportionate disease burden and limited contribution to the labour market.”33 Thus, vulnerability is tied to what a “real and deserving refugee” looks like.34

The use of vulnerability among humanitarian organizations thus resembles the critical concerns for categorical understandings in research ethics raised in the discussion above. Both contexts justify concerns about applying the label of vulnerability and the less than nuanced understanding of agency that accompanies it. In Lebanon, the notion of “negative coping strategies” is enabled by the language of vulnerability. “Negative coping strategies” or “negative coping mechanisms” are defined by the UNHCR as “begging, borrowing money, not sending their children to school, reducing health expenses or not paying rent.”35 The moral blaming embedded in these descriptions as well as the individuation devoid of reflections on the structural conditions that cause such strategies serve specific interests and are profoundly political. They draw attention away from the structural forces and interests that produce vulnerability by responsibilizing individuals.

Lewis Turner has suggested that we should not conflate the categories of those who have the power to label migrants with the categories that those migrants use themselves.36 Perceiving someone as vulnerable might in itself render them vulnerable, which means that the concept of vulnerability should be assumed to be an external imposition into refugee communities: “it is an imposition with its own politics, its own hierarchies, and its own violence.”37 Thus, Turner suggests that “scholars should not assume that ‘vulnerabilities’ are a relevant category of analysis for refugee communities, but rather should foreground ‘the experiences and knowledges’ of refugee communities themselves.”38

Veena Das warns against “super concepts” like vulnerability; it is all encompassing, having traveled across the globe and been applied in many different contexts and in many different ways.39 The discussions on the use of vulnerability in research ethics and in humanitarianism also show the dangers of imposing the concept on any group, as Turner suggests.40 However, when a concept like vulnerability travels and becomes an active component in decisions on assistance and protection as well as in how knowledge is produced, we should also attempt to understand how people on which the concept is imposed relate to it and how it has taken on meaning among them. In the VULNER project, we discussed the meaning and experience of vulnerability during fifty-seven interviews with Syrian and Palestinian research participants. I will summarize a few of the points that people we interviewed raised, highlighting some of those related to encounters with the humanitarian system. My reflections here are based on my previous work with Maria Maalouf.41

While “vulnerability” has become a key term in the refugee policy and knowledge complex in Lebanon, the term’s exact meaning for people is vague. “Vulnerability” is not easily translated into Arabic. When we used the word هشاشة (literally translated as fragility) in interviews, the word was not generally understood by research participants. We thus spent some time with the research participants to come to a joint understanding of the meaning of vulnerability; the words الضعف حالة (weak condition) and ضعيف (weak) were the closest in people’s reflections. In those extended discussions with research participants about the meaning of vulnerability, they also described it as loneliness—the feeling of being left alone, that there is no help, no one to look after us—and, closely related to loneliness, the feeling of not having a state and living in precarious conditions. Thus, vulnerability was also associated with restrictions that made people feel they were not a full member of their new society. Yasmine from Syria explained: “like when I’m forced to keep quiet when the shaweesh humiliates me because there’s nothing I can do. I could call the UN but they’re not answering us.”42 Discussing the different meanings of vulnerability and the different “vulnerability categories,” such as living in a female-headed household or having a disability, most of our research participants would say that, as refugees, they were not the only ones who were vulnerable in Lebanon: due to the compounded crises of the country, almost “every­one” in the country was vulnerable in some way or the other. Hence, to some extent, vulnerability was not really a relevant category for them.

In our discussions about the use of vulnerability as a criteria for assistance and protection, research participants conveyed the gravity of their experiences encountering the institutions that have the power to label them as vulnerable: they were often described as encounters that created feelings of alienation, deprivation, and nonrepresentation. It was not clear to participants what vulnerability criteria were used to determine their eligibility for assistance. The concept of vulnerability was also confusing because the boundaries of the concept kept changing, over time and between institutions.

Many research participants described their experiences engaging with aid organizations, ultimately, as disappointing. Some complained about being cut from the monthly cash assistance without any explanation, or facing delays in accessing cash assistance. The immense struggles to communicate and get in touch with the organization and the feeling of being neglected meant that many no longer trusted the UNHCR, UNRWA, or other humanitarian institutions.

It is important not to overemphasize the role of aid organizations in refugees’ lives. However, in a system with no institutions representing refugees, as many research participants stressed, they experienced a stronger sense of uncertainty when the governance structures they encountered through the aid system were diffuse and blurred. Alongside changing definitions and prioritizations of the categories of vulnerable refugees, the “opaque benefit distribution methods” are part of an active strategy by many aid organizations to deal with overwhelming needs.43 There are abundant stories of how people call and call organizations to ask for assistance they are eligible for, such as cash assistance, legal assistance, help to register newborn babies, or questions related to applying for resettlement, but they cannot get through. Over the phone, by email, or trying to visit their offices, some refugees spend hours every week trying to secure some assistance, often with little results. Speaking on the phone or traveling to an organization’s office comes with substantial costs, which might make aid even more out of reach for some. Institutions in Lebanon with a mandate to assist “vulnerable” people were also the institutions that contributed to experiences of vulnerability among people interviewed.

Using vulnerability as a category-based concept reproduces standardized understandings of refugees and deprives subjects of agency and personality. I have so far shown some of the ontological, epistemological, method­ological, and practical implications of applying notions of vulnerability in our work: imposing the label of vulnerability on a generalized group or individuals within that group affects the narrative that research can produce and risks rendering people vulnerable. In research ethics and in humanitarianism, despite the challenges with the concept of vulnerability, it continues to be applied, frequently as a response to funder requirements and ethical review boards. Rather than abandoning the concept of vulnerability, scholars have argued for a more nuanced conceptualization of vulnerability in ethical reviews. Similarly, in humanitarian practice, there have been attempts to nuance vulnerability categories. But is it time to leave the concept behind? Das suggests that when words are “projected in new contexts, they might have meaning but they do not have life.”44 In Lebanon, vulnerability is a concept that has meaning, but it does not have life: the concept is not understood to be relevant by people with whom institutions associate the concept.

In the process of reconsidering, nuancing, or moving away from the concept of vulnerability, it is worth looking to Birzeit University’s (West Bank) Research Ethics Learning Portal.45 In their attempt to nuance and contextualize ethical reviews, the notion of “people lacking legal capacity” could necessarily be included in procedural ethics. People lacking legal capacity are defined as:

people who are unable to make informed decisions (due to their age, health, physical or psychological condition). Therefore, their participation in the research requires the approval of and/or mediation of a third legally-approved party such as a parent or legal guardian. This group includes children, people with cognitive or psychological impairments, coma patients and amnesiacs.

Lack of legal capacity is an important dimension and could, to some extent, replace the notion of “vulnerable groups” in ethical review procedures, but it does not avoid the dilemma of categorical violence. When I presented a draft of this essay and the proposal of abandoning vulnerability in our ethical reviews to my colleagues at the Centre for Lebanese Studies, they were not convinced. They argued for the importance in ethical reviews of analyzing the context and the positionality of the people we conduct research with, and that within this analysis, the notion of vulnerability is helpful for understanding the positionality of research participants and researchers.

The discussions among colleagues also emphasized that research ethics is a continuing conversation that needs to take place not just at the outset of a project when we impose predefined categories or descriptions of research participants. It is a conversation to have at all stages of our research, including accessing funding and beyond the project itself, to consider the broader historical, political, and social contexts in which our research is situated. Ethical discussion needs to consider what our aims and interests are, what we consider as knowledge and knowledge production, and the power relations involved in research. In our conversation on ethics at the CLS, we also agreed that we should maintain some form of procedural ethics because it represents a potential for holding each other accountable.46

However, procedural ethics must be more closely integrated with the broader research complex and the ongoing conversation about research, knowing, and knowledge production more generally.47 Formulating an ethical review process in our institution is an attempt to foster an ongoing conversation and integrate procedural considerations with broader ethical discussions that can enable a different language beyond what emerged from The Belmont Report.

The discussion of vulnerability in research ethics is closely connected to the principle of “do no harm.” In The Belmont Report, “harm” is specified as follows:

Many kinds of possible harms and benefits need to be taken into account. There are, for example, risks of psychological harm, physical harm, legal harm, social harm and economic harm and the corresponding benefits. While the most likely types of harms to research subjects are those of psychological or physical pain or injury, other possible kinds should not be overlooked.48

Many scholars have emphasized that such generic understandings of “do no harm is not enough,” particularly in the context of forced migration and violence where harms afflicted may be life threating.49 Forced migration research should not stop at do no harm, but should contribute to ending what causes marginalization, exploitation, and violence.50 Consequently, for scholars within a critical epistemological perspective, research should not be limited to simply maximizing benefits or reducing harms, but should challenge the oppressive structures and power asymmetries that underpin migration contexts, including the research process itself.51 Samer Abdelnour and Mai Abu Moghli suggest a more nuanced understanding of harm, focusing on three of its forms:

objectification, which involves reducing someone to the status of an object or generalized category, or representing people without appreciation for their agency or voice . . . ; violence normalization, which occurs when violence is depicted as immutable, normal, unchangeable, or without considering the consequences for people most impacted by it . . . ; and silencing, which involves the exclusion of marginalized or critical voices, especially those most impacted by violence.52

When asked about their understanding of vulnerability, the refugees we interviewed in Lebanon highlighted nonrepresentation, exclusion, and lack of recognition at the core of their experience. Hence, beyond legal capacity and categories more generally, the questions on the risk of objectification, violence normalization, and silencing could replace considerations of vulnerable groups as categories in our ethical review. Eve Tuck and Michelle Fine, in the context of indigenous rights in research and beyond, suggest the principles of sovereignty, self-determination, and complex personhood as personal rights in research.53

Reflections on a more holistic and nuanced way of discussing vulnerability and harm require a broader and more-relational ethics. A key word is “accountability,” a concept that can entail both the procedural and the more holistic and relational approach to ethics and responsibility. Accountability refers to “two core components: to be responsible for one’s actions and living up to/answering for this responsibility.”54 Accountability is thus a two-way relational process. We can think of this responsibility as embedded in procedures and in legal and bureaucratic control, but accountability is also a broader process of moral and social responsibilities. Both humanitarian principles and research ethics could engage more with this broader accountability through a relational ethics that centers “the self on a web of social relationships” in a concrete context.55 By advocating for a relational approach that encompasses both procedural and broader considerations of ethics in institutional reviews, we can emphasize vulnerability as a boundary-less concept applicable to understandings of being human. It also stresses the “working” of a relationship between researcher and research participants, on the one hand, and between research participants and the society at large, on the other hand.56 In this way, we continue our work at the Centre for Lebanese Studies to move away from the violence of predefined categories.

author’s note

This essay is based on the generous collaboration and discussions with all my colleagues at the Centre for Lebanese Studies. Thank you as well to all the colleagues across the world in the EU Horizon 2020 project Vulnerabilities under the Global Protection Regime for the thoughtful discussions the project allowed for. I would also like to thank my good colleagues in the collaborative project Holding Aid Accountable: Relational Humanitarianism in Protracted Crisis (funded by the Research Council of Norway, project number 301024), which provided inspiration and insights for this essay. Finally, thank you to Lisa Anderson and all organizers and participants in the Research Ethics in the Middle East and North Africa (REMENA) workshop in Doha on January 27–29, 2024.