PhD Defense - Evan Josselin, "Citizen science in practice and theory : Building on the case of Plankton Planet"

Date/heure : Mardi 9 décembre à 14h00

Lieu : Salle de réunion, rez-de-chaussée, Institut Jean Nicod

Titre : "Citizen science in practice and theory : Building on the case of Plankton Planet" (La science citoyenne en pratique et en théorie : le cas de Plankton Planet)

Abstract : Citizen science is a rapidly expanding field that remains only partially integrated into the normative frameworks of traditional science. While sociology and philosophy of science have begun to examine citizen science—addressing its promises, typologies, and issues of data quality— comprehensive theoretical frameworks capable of capturing its epistemic particularities are still scarce. This dissertation seeks to address that gap by interrogating the status of participants, the scientific activities they perform, and the ways in which citizen science interacts with traditional science. It provides, therefore, both an epistemological analysis of citizen science and practical insights for its practitioners. After using Dewey’s pragmatic framework to establish conceptual anchors to navigate the diversity of citizen science practices, disciplines, and forms of engagement, the analysis turns to a detailed case study of Plankton Planet during its developmental phase (2022–2025). Plankton Planet is a project in which sailors, master students engaged with the French Navy, and scientists collaborate to collect oceanic plankton data using scientific instruments. Drawing on Hasok Chang’s framework, the dissertation demonstrates that participants are genuine epistemic agents : they engage in complex, skill-intensive activities requiring both mental and physical capabilities, as well as adaptability. Such engagement challenges the conventional assumption that data collection is a straightforward task that can be easily delegated to the public. Moreover, participants’ status as epistemic agents prompts a reconsideration of data quality : without the full capabilities of professional scientists, participants cannot be assessed by expert accuracy alone. I propose that data quality should instead be evaluated through a criterion of operational coherence—the extent to which data enable subsequent epistemic activities—and, more specifically, through cohesion in data collection processes that systematically account for potential vulnerabilities. Beyond the fact that participants are fully-fledged epistemic agents, the dissertation emphasizes the fundamentally collaborative nature of citizen science and moves away from Chang’s individualistic view of knowledge production. Using Plankton Planet as a model, it highlights multiple layers of collaboration, heterogeneous goals, and hierarchical divisions of epistemic labor, which together render collaboration in citizen science epistemically fragile and complex. To sustain successful knowledge production, three conditions are essential : epistemic trust, community forums, and practices of care. This framework reconceives citizen science as a form of co-production, in which participants and scientists engage iteratively and collectively, rather than functioning within a linear and unidirectional division of epistemic labor.

Superviseurs et jury :

• Stéphanie RUPHY (ENS-PSL), Présidente

• Anouk BARBEROUSSE (Sorbonne Université), Rapportrice

• Kevin C. ELLIOTT (Michigan State University), Rapporteur

• Olivier IRISSON (Sorbonne Université), Examinateur

• Sabina LEONELLI (Technical University of Munich), Examinatrice

• Valeria GIARDINO (CNRS), Co-Directrice de thèse

• Colomban de VARGAS (Sorbonne Université), Co-Directeur de thèse