CEFISES Seminar: Doan Vu Duc, “An interactionist account of the problem of demarcation. How to conceive an empirically informed framework of scientific rationality?”
Series: OLOFOS
Speaker: Doan Vu Duc (UNamur)
Title: “An interactionist account of the problem of demarcation. How to conceive an empirically informed framework of scientific rationality?”
Abstract
One consequence of people’s mistrust in science is the proliferation of certain pseudosciences that aim to “fight down some scientific theory or branch of science” (Hansson, 2021). Although one may have a rough idea of what a pseudoscience looks like (astrology, for instance), the philosophical task of demarcating it from science has been a difficult one. Throughout the 20th century, philosophers attempted to define the nature of science by means of a single criterion, until Laudan dismissed the discussion as “a pseudo-problem” (Laudan, 1983). Yet the problem has resurfaced.
Philosophers of science today increasingly embrace the societal responsibility of distinguishing science from pseudoscience. In the face of pseudoscience and disinformation, demarcation has become necessary for both political and educational purposes. Despite certain difficulties, the gradualist approach is a promising strategy.
Within this context, my aim is to challenge the meaning and nature of scientific rationality in a demarcationist fashion (unlike, for instance, Feyerabend). I defend an account of demarcation grounded in the Argumentative or Interactionist Theory of Reason (ITR). This theory of cognitive science from Hugo Mercier and Dan Sperber holds that reason is a tool evolved for a specific function: producing arguments or justifications and evaluating them in interactive contexts. On this view, science is one such context, where argumentation is particularly fast and dynamic, allowing reason to operate in the very environment for which it evolved. Scientists strive to persuade the community with the strongest possible arguments while being equally motivated to rigorously evaluate those of their peers.
If this framework helps resolve key problems (e.g., distinguishing bad science from pseudoscience), it also raises seemingly relativistic concerns. If scientific justification is simply what convinces the community—i.e., what survives peer evaluation—how can we secure the truth-conduciveness of science? Since truth-conduciveness grounds the educational role of science, it remains a non-negotiable element of any viable account. I therefore defend an account of the problem of demarcation that I call epistemic interactionism (EI) for which I propose two interpretations (one strong and one pragmatist) that I wish to defend in this talk.