What does it actually mean to decolonize international research? Not as a rhetorical commitment in a grant application, but as a lived practice inside a large, externally funded, multi-country research project?
This is a question that the RelEx project — the Centre of Excellence on Religion and Social Exclusion — forces us to take seriously every day. And the more honestly we engage with it, the more it reveals itself to have the structure of a paradox.
What RelEx is trying to do
RelEx investigates how socially excluded people view and relate to religion across five national contexts: Finland, Ghana, India, Peru, and the Philippines. The project deliberately moves beyond the well-documented WEIRD bias in social research — the tendency to draw conclusions about human experience from populations that are Western, Educated, Industrialised, Rich, and Democratic.
But we have pushed this even further. Across our five contexts, RelEx focuses specifically on three populations that have historically been studied about rather than with: indigenous communities, incarcerated individuals, and LGBTQI+ individuals. These are groups that face intersecting forms of exclusion, that are often considered hard to reach, and whose vulnerability varies significantly depending on national and cultural context.
The ambition is clear. But the structures that make this research possible constantly resist it.
The Catch-22
Joseph Heller’s novel gave me a phrase that captures something real about paradoxical situations — situations where two contradictory conditions mutually cancel out any attempt to resolve them. The classic example: a pilot can be grounded for being insane, but requesting to be grounded proves you are sane, so you cannot be grounded.
Decolonizing international research has exactly this structure.
On the one hand, large-scale cross-cultural research — precisely the kind with the best chance of challenging Western-centric knowledge production — requires substantial external funding. On the other hand, the frameworks that provide that funding require project designs, accountability structures, partnership arrangements, and dissemination plans that are built on and reproduce the very hierarchies you are trying to dismantle. You need institutional legitimacy to give marginalized voices a platform. But claiming that institutional legitimacy is itself a hegemonic position.
Each move toward decolonization risks reinforcing what it critiques. There is no clean position outside the system from which to operate. And there is no idealized blueprint that resolves this. Any attempt to prescribe a universal decolonial protocol would itself reproduce the top-down, universalizing logic it claims to oppose.
What this situation calls for, I think, is both honesty and ongoing, institutionally embedded critical reflexivity. Let me walk through three levels at which this tension manifests in practice.
The methodological level
Research on religion has often — despite growing awareness of the problem — wrongly presupposed a conceptual correspondence between religions across time and cultural contexts. Instruments developed in Western settings have been applied to populations for whom those categories carry very different meanings, or no meaning at all.
In RelEx we have had to tackle this at the level of instrument design. The Faith Q-Sort, and more recently the Worldview Q-Sort, were explicitly developed to take a more bottom-up approach — to allow participants to express their own forms of religiosity and worldview rather than forcing them into predetermined categories. Even so, ensuring genuine cross-cultural validity required sustained, iterative dialogue with colleagues, translators, and community members in each context: double and back translation, cognitive validation, and ongoing critical discussion about what concepts actually mean in specific cultural and linguistic settings.
This process is slow, expensive, and uncertain. And it is never fully complete. What it teaches you — if you stay honest with yourself — is that there is no one-size-fits-all model for cross-cultural research on religion. The tension between universalism and particularism cannot be finally resolved. What you can aspire to is what Michael Lambek calls a moving balance among several distinct epistemological positions: a reflective, plural methodology that resists the temptation to impose premature closure as Ruth Illman, Nurit Novis-Deutsch, Rafael Fernández Hart, and I explored in the edited volume Researching Global Religious Landscapes (Equinox, 2024).
The structural level
Methodological choices do not happen in a vacuum. They are shaped by a larger research design and by funding requirements that carry their own hierarchies.
The internationality of a project is rarely defined purely by scientific necessity. EU funding frameworks typically require specific country partnerships, specific accountability forms, and specific authority structures — creating structural asymmetries from the outset. Non-European partners are routinely allocated less funding and less decision-making authority.
One small but telling example: the accounting procedures at my university were designed for teams with credit cards, stable Finnish average incomes, and the institutional capacity to cover costs and wait for reimbursement. Asking a junior researcher from a low-income country to travel at their own expense and wait months for reimbursement is not an administrative inconvenience. It is a symptom of hidden racial and economic structures that project designs rarely fully anticipate — and a direct contradiction of the EU funding ideal of promoting the careers of young scholars from the Global South.
A different kind of structural vulnerability may emerge if political conditions shift in one of our partner contexts, and local authorities begin viewing the research with suspicion. What are our means, as an internationally coordinated project, to protect our ‘valued’ partners in such cases? These moments and questions reveal what the idealization of international collaboration tends to obscure: including partners in our projects does not automatically benefit them. It can also endanger them — while leaving us with no adequate space to assume responsibility.
The relational and psychological level
The third dimension is relational: the dynamics within research teams and between researchers and participants.
Breaking patterns of dominance and making genuine space for diverse expertise requires more than formally inviting people to speak. It requires actively destabilizing power relations — attending to who defines the research questions, whose theoretical frameworks are treated as the default, whose expertise is recognized in publications and grants, and who bears the costs of building collaboration and trust across very different institutional environments.
This kind of work is difficult to plan for in advance. It requires time — regular workshops, ongoing dialogue, the slow accumulation of trust. Under the pressures of project delivery, within a research design that leaves little space for reshaping the work as it unfolds, there is always an imminent risk of treating partners as data-collection instruments executing a design produced elsewhere.
Yet, this connects with one more layer that tends to be underacknowledged: the psychological costs of research participation for vulnerable communities.
In our fieldwork with LGBTQI+ individuals in Ghana, for instance, same-sex relations carry severe legal penalties and the social consequences of visibility can be devastating. But even in Finland, I know from experience how research can reopen wounds that have only partially healed. Research participants bring histories of violence and exclusion in forms that even a carefully designed research approach cannot fully anticipate or contain.
The risk of re-traumatization is not a marginal concern. It is an inherent feature of conducting fieldwork with populations whose lives have been shaped by sustained exposure to exclusion and harm.
Standard ethics frameworks address this through procedural safeguards — informed consent, anonymity, confidentiality, debriefing protocols. These are necessary. But they routinely overlook what it would actually mean to take seriously the psychological needs of participants whose very visibility as research subjects may constitute a risk, and whose experience of being researched is shaped by histories of marginalization. In such cases, a research ethics framework is not neutral. It can perpetuate the very hierarchies it claims to protect against.
No blueprint — but a commitment
These challenges are not independent problems with independent solutions. They interact and reinforce one another. The funding frameworks that create structural asymmetry also shape methodological choices. The power relations within research teams affect what participants feel safe to disclose. The re-traumatization risk in fieldwork is inseparable from the colonial history that produced the conditions of vulnerability in the first place.
This is the Catch-22. The institutional structures that make large-scale international research with marginalized communities possible reproduce the hierarchies that such research hopes to challenge. And there is no idealized blueprint that resolves this contradiction.
But this is not a reason for resignation. The absence of a blueprint is not the absence of direction. What it calls for is an ongoing, institutionally embedded practice of critical reflexivity. A commitment to keeping the contradictions visible. A willingness to ask, at every stage of a project: who bears the costs here? Whose expertise is being recognized, and whose is being extracted? What would genuine epistemic partnership look like in this specific context, with these specific partners, under these specific conditions?
And perhaps most importantly: a commitment to honesty in how we represent this work — in our publications, in our grant applications, and in public conversations like this one. The research that most needs to be done, with and about the most marginalized communities in challenging contexts, is also the research that carries the greatest structural contradictions, methodological uncertainties, and psychological costs — for participants and for researchers alike. Refusing to acknowledge this is not neutrality. It is a choice to protect the comfort and reputation of the research enterprise at the expense of those we claim to study.
Decolonization is not a methodology that can be fully specified in advance. It is a commitment — one that has to be renewed, negotiated, and honestly addressed throughout the research process.