Testing Theories as a Concern of the Sociology of Science

Abstract

In the context of recent transformations in science, marked by Big Data, computational simulations, and artificial intelligence, the concept of theory testing requires reevaluation. From the universalist ideal of Popperian falsifiability, we are shifting toward forms of testability that are local, situational, and institutionally mediated. The central issue lies in clarifying what it means to test a theory in contemporary sciences, a problem that generates tension between universal and local forms of testability. The aim of this article is to provide arguments in favour of a philosophy of “situated testability,” grounded in an interdisciplinary methodological approach. Following a critical analysis of how testing is currently understood and practiced in science, I argue for a mode of testability that is adequate to the complexity and diversity of contemporary epistemologies, and I seek to answer the question “To what extent does
scientific testing become an institutional and social practice?” Contrary to the universalist and deductivist ideal advanced by traditional philosophy of science, I contend that testability today emerges as a situated, infrastructural, and contextual practice. I thus advocate a pluralist and critical orientation, grounded in contextual experimentalism and in the localization of scientific laws.