On the Very Idea of a Thought Experiment

Jean-Yves Goffi (1) , Sophie Roux (2)

1 PLC - Philosophie, Langages et Cognition
2 La République des savoirs : Lettres, Sciences, Philosophie

If imagination is free to wander about in a thought experiment, without anything equivalent to the principle of reality able to intervene, is it ever possible that a thought experiment would not work per se? Of these two options, only one is real: -- Either all thought experiments are successful, and there is no need to be a stubborn falsificationist to conclude that the very idea of a thought experiment should be abandoned. -- Or there is a distinction between successful and unsuccessful thought experiments, and the idea of a thought experiment could have a certain epistemological depth. In that sense, showing that there is a distinction to be made between successful thought experiments and unsuccessful thought experiments and explaining why there is such a distinction aims not only at making a distinction for the sake of 'taxonomophilia'. The point is also to show that the idea of a thought experiment is a substantive idea that has a certain epistemological depth.

Mots clés: thought experiment; Galileo; Mach; Aristotle; Thomson; Stevin; Descartes

Domaines: Histoire, Philosophie et Sociologie des sciences

Jean-Yves Goffi, Sophie Roux. On the Very Idea of a Thought Experiment. Thought Experiments in Historical and Methodological Contexts, Brill, pp.165-191, 2011. ⟨halshs-00807059⟩