I give arguments supporting the claim that one of the most prominent methodological results of French conventionalism – rejection of the possibility of a crucial experiment in mature empirical sciences – was formulated simultaneously by Pierre Duhem and Gaston Milhaud in 1894. Thus, I attempt to question the standard approach in philosophy and methodology of science, which attributes the said result exclusively to Duhem. I am building my case of Milhaud’s true contribution to the debate on the rejection of the existence of the experimentum crucis, made in his PhD thesis Essai sur les conditions et les limites de la certitude logique.

The end of the nineteenth century was the period when revolutionary scientific discoveries challenged well-established theories, forcing both philosophers and scientists to ask questions about the nature and certainty of scientific knowledge. A group of French scientists not only performed a thorough critique of contemporary science and its history but proposed a new model that adequately described the development of scientific knowledge. Gaston Milhaud made a significant contribution to this new description of knowledge creation. He is however rarely mentioned in the context of the theory of knowledge and remains overshadowed by his famous colleagues. Despite the fact that more than a hundred years have passed since the conventionalist philosophy of science was formulated, H. Poincaré’s, P. Duhem’s and G. Milhaud’s positions have not gained much popularity beyond the circle of philosophers of science. This article briefly outlines personal relationships within French conventionalist circle, presents important results of Milhaud’s analysis, and the reasons why philosophers do not recognize the role he played in creating a new model for the development of scientific knowledge.