Hegemonic articulation, paradigm incommensurability, and the discursive exclusion of Humanities and Social Science Research in the South African Open Science Policy 2026

The South African Open Science Policy 2026, published by the Department of Science, Technology, and Innovation, is formally committed to the democratisation of knowledge and epistemic equity. Yet a critical reading reveals that the Policy reproduces, through its own discursive architecture, the epistemological hierarchy it claims to dismantle. This article analyses the Policy as a discursive text, drawing on Laclau and Mouffe’s discourse theory and Foucault’s concept of the regime of truth to explain how a policy committed to democratisation can simultaneously enact epistemic exclusion. The primary argument is threefold. First, the Policy constructs “Open Science” as a nodal point around which a chain of equivalences is built from STEM-derived norms, including reproducibility, FAIR data principles, journal metrics, and natural science models of citizen science. Second, we argue that this hegemonic articulation reconstitutes a hidden antagonism between positivist and interpretivist epistemological traditions, reawakening a binary opposition that the paradigm wars of the twentieth century had ostensibly resolved. Third, the incommensurability between these traditions means that applying STEM-derived criteria universally to all research constitutes not evaluation but category error institutionalised as governance. Six dimensions of epistemological closure are examined: the reproducibility criterion, the FAIR data mandate, the marginalisation of CARE principles, the protective rather than epistemic treatment of indigenous knowledge, the structural absence of humanistic output forms, and the natural-science-centred design of the citizen science framework and its consequences for participatory research traditions. Five policy interventions are proposed through which the Policy’s stated commitment to epistemic equity could be made operational across the full disciplinary spectrum.

Competing interests

The authors declare no competing interests.

Ethics

Ethical approval was not required as no human participants were involved in this research.

Informed consent

This article does not contain any studies with human participants performed by any of the authors.