RésuméJe me demande ici si les conceptions pragmatiques de la vérité peuvent être réconciliées avec les intuitions ordinaires quant à la portée de la bivalence. Je soutiens que les pragmatistes sont conduits à accepter une distinction du genre «type / occurrence» entre les formes d'une investigation et ses instanciations particulières, sous peine de banaliser leur vérificationnisme. Néanmoins, même la conception révisée que j'examine échoue à sauver les approches épistémiques de la vérité de certaines conséquences peu plausibles.

In a 1993 paper, I argued that empirical treatments of the epistemologyused by scientists in experimental work are too abstract in practice tocounter relativist efforts to explain the outcome of scientificcontroversies by reference to sociological forces. This was because, atthe rarefied level at which the methodology of scientists is treated byphilosophers, multiple mutually inconsistent instantiations of theprinciples described by philosophers are employed by contestingscientists. These multiple construals change within a scientificcommunity over short time frames, and these different versions ofscientific methodology (...) can determine the outcome of a controversy. Iillustrated with a comparatively detailed analysis of the methodologyused by biologists debating the existence of an entity called thebacterial mesosome between the mid-1950s and the mid-1970s. This 1993piece has drawn several critiques in the philosophical literature. Inthis present piece I respond to these critiques and argue that they failto address the core argument of the original paper, and I reflectfurther on the methodologies of philosophers of science pursuingempirical or `naturalistic' epistemology. (shrink)