Equity in science: A beautiful lie?
When partnerships aren’t neutral
by Similo Ngwenya (UCL/RoRI)
For as long as I can remember, I’ve been drawn to understanding inequalities in science; how they manifest, how they are reproduced, and why they persist. These questions sit at the heart of many debates about research systems, collaborations, and the politics of knowledge production.
Recently, I came across a reflection by ecologist Dolors Armenteras: “Equity in science is a beautiful lie.”
It’s a provocative - even uncomfortable - claim. But it forces us to ask: is it really a lie? And if so, what does that mean for the partnerships, collaborations, and systems we keep building?
Since engaging with metascience, a systems approach that examines the incentives, structures, and institutional arrangements shaping research practice, I’ve started seeing questions of equity through a systems lens. This perspective helps reveal the underlying mechanisms that make inequity so persistent, especially in partnerships between researchers from the Global North and Global South, and encourages us to pay closer attention to how we evaluate research, define excellence, choose publication outlets, and design funding calls.
In essence, partnerships are embedded within systems that allocate resources, authority, visibility, and legitimacy. These systems determine whose work is recognised, who gets to lead, and which knowledge is taken seriously in policy spaces. If we are serious about equitable partnerships, the real task is to interrogate the systems that shape them because the question is not simply whether individual partnerships are equitable, but whether the systems in which they operate allow equity to emerge in the first place.
Before diving deeper into the systems perspective, I want to take a brief detour through the work of Pierre Bourdieu, whose sociology of science is grounded in the idea that science, like any other sphere, is a social field of forces, struggles, and relationships. In a previous project examining the epistemic effects of inequality in forest research, my cohort drew deeply on Bourdieu’s insights. He argues that actors in the scientific field compete for scientific authority - the socially recognised capacity to speak and act legitimately in scientific matters. Success in this competition depends not only on intellectual ability but also on the amount and forms of capital actors possess. Researchers and institutions accumulate different types of capital: economic, scientific, social, symbolic - and these are distributed unevenly across the global research system.
Those who occupy dominant positions in the field tend to set the standards for what is perceived as “quality science.” The properties of their work – the methods they use, the topics they study, the journals in which they publish – come to define what is seen as legitimate scientific practice. New entrants, or those in more marginal positions, are then judged against these pre-set standards.
Coming back to partnerships, researchers rarely meet on neutral ground. They enter collaborations from very different starting positions, shaped by the forms of capital they’ve accumulated. These positions influence how roles, authority and recognition are distributed. Over time, these unequal dynamics can become internalised through what might be called a collaboration habitus where researchers gradually learn, often unconsciously, what roles are expected of them. This includes who leads proposals, who collects data, who writes, who publishes, and who speaks to policymakers. Inequality becomes normalised and routine. It is rarely questioned.
While habitus can explain how certain roles in partnerships become accepted and normalised informally, the evaluation processes of research has already formalised them since they have ranked researchers and institutions long before collaborators meet. Whether through administrative evaluation, which governs promotions, funding decisions and institutional rankings, or within-science evaluation, such as peer review that determines what constitutes recognised knowledge, researchers enter collaborations with unequal reputational capital. The evaluation system relies heavily on global quantitative metrics and standards for this capital, which disadvantages those working in settings where resources are limited. In this way, inequality is present not only within partnerships, but also in the broader evaluative systems that shape scientific careers.
Turning to scholarly publishing, it is worth noting that, despite being presented as a neutral academic activity, it is heavily influenced by evaluation pressures. Universities and funders incentivise metrics such as paper and counts, H-indices and journal impact factors. Scientific work is judged by its performance in elite outlets, and researchers are encouraged to publish in these outlets. The National Research Foundation (NRF) of South Africa‘s individual rating system is a good example of this, as it explicitly encourages researchers to produce “high-quality” outputs in high-impact journals. Is this inherently bad? Not necessarily. However, it does influence behaviour and priorities in ways that deserve scrutiny, particularly in applied, context-specific settings.
In these application-oriented contexts, it helps to distinguish merit from relevance. ‘Merit’ indicates that African researchers meet global quality standards, even if these standards are not always appropriate for African contexts. ‘Relevance’ is assessed against local expectations and needs. A system that primarily rewards merit, as measured by global metrics, can inadvertently downplay relevance, particularly when local knowledge, multilingual outputs or policy-proximate genres do not easily fit into the most visible journals.
When open science, with its practices of open data and open access to publications, is introduced to the publishing landscape, the situation becomes even more complex. Open access has shifted the business model from readers paying to authors paying article processing charges (APCs), creating additional complications around embargoes and licensing. For many researchers in the Global South, where countries still struggle to meet the widely discussed target of allocating 1% of GDP to R&D, APCs and compliance costs are unaffordable; (a previous post on this Substack examined the harms of this open access business model in India). The scarcity of R&D resources makes it harder to participate on equal terms and also changes how authors negotiate authorship, leadership roles and outlet choices in collaborative projects.
The picture is no better when it comes to funding and policy systems. Funding calls originating in the Global North often include criteria that perpetuate inequalities, either intentionally or not. These include requirements relating to publication outlets, partnership structures for eligibility and expectations of international collaboration. For instance, the US National Institutes of Health (NIH) require projects with a foreign component, including those undertaken outside the US, to submit a specific Foreign Justification and meet additional eligibility and review criteria that apply only to foreign applicants. These conditions often result in leadership being channelled to US organisations, with others positioned as sub-awardees. Such stipulations cascade through the system, shaping evaluation criteria, publishing strategies, and impact expectations.
Publishing outlet requirements and APC-driven models can also influence authorship order and visibility, with researchers from well-resourced institutions being better able to secure prestigious authorship positions in elite journals. Partnership criteria can also incentivise what we might call compliance-driven or nominal collaboration, where partners are added primarily to meet call specifications or unlock eligibility.
These relationships secure grants, but do not necessarily enable equitable workload distribution, decision-making or credit allocation. Once funding has been awarded, its influence on the direction of research continues long after a project has ended. International funding, in particular, does more than enable research; it configures the systems within which research operates, shaping agendas, incentives, and decision-making structures.
I do not claim to have all the answers to the issues I have raised. But what follows are a few recommendations to open up possibilities (and there’s certainly room to add more):
Acknowledge the interdependence of incentives. Publishing incentives do not operate in isolation; they interact with evaluation, funding and policy uptake. Reforms should therefore be aligned across these domains so that the signals sent out are coherent rather than contradictory.
Adopt context-sensitive assessment and value multilingual outputs, whether these be policy briefs, datasets, software and community-engaged reports alongside journal articles, so that locally relevant contributions are visible and count. Combining narrative with numbers allows experts to contextualise bibliometric information. However, combining quantitative and qualitative indicators requires sensitivity to gaps in data, disciplinary variation and contextual differences; for example, informed peer review can help, but it does not always benefit early-career researchers, those working in marginalised fields or those tackling problems that are difficult to evaluate by conventional means.
Reduce the costs of openness. Prioritise Diamond Open Access (OA) publishing models and investment in African-led platforms and repositories to ensure that open access is financially sustainable and equitable. While fee waivers can provide temporary relief, they should not be the long-term mechanism through which African researchers access open publishing, as waivers ultimately reinforce the unequal power dynamics that this piece interrogates. At the same time, recognise the wide range of publishing practices, including preprints, regional journals, multilingual work, policy outputs and other knowledge forms.
Reward relevance as part of excellence by encouraging journals, funders and universities to explicitly recognise relevance to local health, education, environmental or economic challenges as a core dimension of excellence, not an optional add-on.
Equity in science isn’t just about good intentions; it’s about the systems that reward, recognise, and reinforce certain actors over others. We need to seriously interrogate these structures to move equity from a “beautiful lie” to a tangible possibility.