Rethinking Open Science: Towards Care for Equity and Inclusion

Growing pains: more Open Science is exacerbating its contradictions

Open Science (OS) represents a new mode of doing science based on cooperative work and new ways of sharing knowledge, often through digital technologies or other collaborative tools. Also, as expressed in a UNESCO Recommendation in 2021, there is hope that OS will “serve to widen access to scientific knowledge for the benefit of science and society and […] promote opportunities for innovation and participation in the creation of scientific knowledge and the sharing of its benefits” (UNESCO, 2023).

Given these potential benefits, activities such as open-access publications, data sharing, and citizen science have been promoted and have gained ground, particularly over the last decade. However, recent analyses of OS have found some worrying trends: yes, OS is spreading, but it is doing so in a way that questions expectations that it may lead to more equity and enhance the societal impact of science.

Overall, one gets the impression that something has gone wrong with the current modes of developing OS: on the one hand, current Open Science is leading to more inequity and on the other, the societal impact of the current ways of doing Open Science is unclear or limited.

First, on inequity, researchers in rich universities and countries (as is my case) now have the privilege of being relatively more visible than colleagues in poor-resource contexts, because our institutions can pay (often expensive) fees to publish via Open Access. While this makes some knowledge accessible, it goes against the basic principle that scientific contributions should be judged and made visible according to their scholarly merit, not because of the wealth of the authors. Therefore, many stakeholders believe that the pay-to-publish model (what used to be called “gold” or hybrid OS) is corrupting the research system. This model also undermines publishers with “diamond” Open Access (free to publish and read), particularly in regions like Latin America and Eastern Europe. As a result, even in Western Europe the tide is shifting from pay-to-publish to institutional support to diamond Open Access journals.

Second, according to a recent review, so far there is little understanding of the societal benefits of OS. However, current evidence does suggest that citizen science and other participatory approaches, such as interactions with policymakers and stakeholders, are the main ways research contributes via social arenas. In other words, social impact seldom happens via papers or data, but mainly through social interactions that mediate the “transfer” of knowledge back and forth between societal actors and researchers. These findings question the current focus on investment in technological platforms of many OS policies.

Following these insights, I will argue that the conceptualization and promotion of OS need to be rethought to serve the goal of providing epistemic justice.

Open Science as a transformation, but in which direction?

As we have seen, two main drivers exist for the development of OS. First, the digitalization of information has brought about new methods for producing, communicating, and storing scientific knowledge. Second, the expectation that these new methods would facilitate science–society interactions has been associated with critiques of the social impact of science and the hope of making research more responsive to societal needs, demands, and aspirations.

Very different agendas for OS implementation have been developed depending on why we pursue OS and what we believe OS can accomplish. Some of the visions focus more on efficiency gains within the research system, others on the development of platform technologies, on widening access to information, or are concerned primarily with participation. While, in principle, these visions were expected to run in parallel and complement each other, implementation has brought out tensions and contradictory dynamics.

If we understand OS as transformation of the research system, each vision of OS pushes research in a direction incompatible with other visions. For example, the development of OS in terms of information platforms very often enters into tension with OS as inclusion and participation, since some sections of the global population do not enjoy a context or have the capacities that allow them to participate via these platforms. Or, for example, more Open Access via a pay-to-publish model is antagonistic to OS in terms of equity (because researchers in low-resource contexts cannot pay) and integrity (because the rigor of the review systems of some pay-to-publish journals, such as Frontiers or MDPI, is questionable).

In summary, there is not one OS future to be reached but disparate potential futures that would lead to certain types of OS but not others. The question to be asked is thus not about the extent of progress towards more OS but about what types of OS are developed and adopted, by whom, and with what consequences.

Philip Mirowski, a political economist of science, has warned that mainstream OS, associated with information infrastructure, is closely associated with “platform capitalism” (cf. the “surveillance capitalism” of Soshana Zuboff) and involves dangers similar to those of social media such as Google and Facebook: control of public research information at different stages of the research process (from lab notes to publications to evaluation analytics) by oligopolistic companies such as Elsevier, Clarivate or Springer-Nature with the power to shape collective behavior and visions of science. Those companies, often with the support of US and European policies (e.g., the early Plan S), are not only extracting wealth from the Global South, but are in a position to produce representations of science that may reinforce Global North hegemonies in terms of making its main scientific issues, disciplines, languages, values, and cultural perspectives more visible.

Nevertheless, in parallel and friction with these platforms, collective initiatives are being developed both in the Global South and the Global North that offer alternatives for diverse and inclusive OS trajectories; for example, La Referencia, Participatory Research in Asia, the Public Knowledge Project, or the Barcelona Declaration. The question remains as to which are more aligned with epistemic justice within the alternative OS futures.

Open Science by whom? Open Science for whom?

The 2021 UNESCO Recommendation on OS has been crucial in redefining OS by establishing equity and collective benefits as key values of the directions to be pursued. Building on the principle in the Universal Declaration of Human Rights that: “Everyone has the right freely to participate […] and to share in scientific advancement and its benefits”, the Recommendation sees science as a global public good, and the “openness” in OS, as the means by which to make knowledge genuinely public and global.

However, as argued by Michel Callon, science is not a conventional public good because substantive investment in capacities is required to participate not only in its production but also in its reproduction, maintenance, and use. Any citizen can potentially breathe clean air (a public good) with no special effort and without even necessarily being aware of it. However, to participate in producing and using scientific knowledge, previous knowledge, and complementary resources and capabilities are indispensable.

For example, concerning expertise: we may have access to scientific publications on cancer. Even so, in the event of a diagnosis, only experts can use those scientific papers to decide on the appropriate therapies. The rest of us need to rely on reports aimed at a general audience; therefore, these materials (rather than scientific papers) are key for knowledge sharing.

Concerning resources, researchers in middle–low-income countries might be assumed to have access to websites with scientific data. In practice, however, they often cannot use them because data analysis requires some infrastructure or specific personnel that they cannot afford; in the worst scenarios, they may have expensive, poor, or blocked (due to sanctions) internet connections.

In short, scientific information accessible online often cannot be mobilized for good purposes, particularly in the Global South. Making scientific products (articles, data, software, etc.) openly accessible can benefit organizations and companies that have strong capabilities and resources. Even so, targeted efforts of “transfer” and adaptation are needed for this knowledge to reach and benefit most populations in the world. Just making knowledge electronically accessible favors mainly those who already have access and does not foster participation or sharing in the benefits of science for much of the world population. This is why the direction that OS is taking by focusing on free access to scientific products is not leading to more equity and epistemic justice.

Contextualizing openness: from ‘access to outputs’ to ‘connections’

Despite all this, alternative forms of doing OS can lead to equity and impact. The Open and Collaborative Science in Development Network (OCSDNet), led by Leslie Chan, made the case that openness needs to be contextualized. It is only in a specific context that researchers and stakeholders can develop the specific forms of participation and communication that render scientific knowledge valuable to specific social groups, e.g., marginalized social communities.

This contextualization cannot be achieved simply by focusing on making research products digitally accessible. Instead, as Sabina Leonelli has argued in a recent book, the focus should be on knowledge exchange processes between researchers and social communities. These processes will often benefit from open access to digital products. Nonetheless, the particular forms and platforms used will vary depending on the participants in a given knowledge exchange process.

The movement for Open Science has offered the promise of epistemic justice. Many activists think that private actors, mainly oligopolistic publishers, have hijacked current developments, but perhaps so too have large research infrastructures in powerful disciplines (e.g., genomics and high energy physics). To regain its emancipatory power and care for equity and inclusion, OS needs to redefine itself, not in terms of products and technological platforms (so many of which are owned by industry or “big science”), but in the very processes of knowledge exchange in more humble settings across a much more comprehensive range of human communities.

Ismael Ràfols, UNESCO Chair in Diversity and Inclusion in Global Science, Leiden University, The Netherlands, INGENIO, University of Basque Country (CSIC-UPV), and Universitat Politècnica de València, Spain <i.rafols@cwts.leidenuniv.nl>