Amphithéâtre Marguerite de Navarre, Site Marcelin Berthelot
Open to all
-

Abstract

The second lesson clarified the criteria invoked to determine whether there are "articulations" in nature that are more natural than others (cat, money, carbon, electron, planet) to which our classification systems would correspond perfectly: resemblance, predictive and explanatory success, a priori criteria, semantic criteria. We have shown their fragility, and the possibility of finding the concepts of species legitimate only if we lower our metaphysical pretensions a notch: by preferring, for example, in the case of chemical species, to the criterion of "hierarchy", that of explanatory and predictive role; or by admitting the idea of species with "vague" contours. As far as semantic criteria are concerned, we pointed out the fragility of the simplicity criterion (for example, "H2O" and "the element whose atomic number is 79" intuitively designate natural species - if this is also true of "water" and "gold" - but they are not semantically simple. We began to discuss the suggestions, made in the 1970s by Saul Kripke or Hilary Putnam, of a possible systematic overlap between semantic and metaphysical criteria, and concluded this first stage of reflection: 1) by recalling the empiricists' (Locke-Hume) contestation of the possibility (linked to the Cartesian rationalist model) of a priori knowledge of natural species; 2) by already indicating two possible parries: make better use of the concept of a posteriori necessity (Kripke); rethink essentialism, avoiding, of course, a return to outdated forms of neo-aristotelianism (hardly in keeping with scientific advances), but also avoiding reducing essence to a pure modality. For we must distinguish between :

Df1: F is a necessary property of a ssi a a F in all possible worlds that include a. E.g.: to say that water is essentiallyH2O, means that in any world where there is water, it is composed ofH2O(and that in any world where the right H20configuration is found, it is water); and :

Df2: F is an essential property of a if being F is constitutive of theidentity of a; which is the definition taken up, in an Aristotelian wake, by the medievals, then Locke. The essence of a thing is what determines what kind of thing it is, and what then determines whether it belongs to this or that species. This is why anyone who denies that things have an essence would have to deny the existence of a specific difference between him or her and, say, a donkey or a cabbage (Thomas Aquinas).

References