Metaphysics of natural species (continued) - Introduction

The first lesson first recalled the requirements of a metaphysical knowledge of nature: bypassing our illusions about modalities and realism; laying down the rules of the method of conceptual analysis, the role ofa priori and intuition; moving on to the a posteriori stage, bringing conceptual analysis into contact with the sciences; saying, in order to achieve a non-scientistic scientific metaphysics, what the metaphysician's commitments are: how he must be able to defend a realism without subscribing to "metaphysical realism", and why his realist commitment implies scientific realism and a form of "reasoned humility". We have identified five characteristics of the dispositional realist metaphysics thus established: a form of realism that is not "metaphysical" but scholastic (inspired by the medievals) and dispositional; a semantic realism - assuming that there are real universals, but not that all universals are real: the real is that which "signifies" something real - forcing us to clarify the concept of causality, to determine the meaning of our dispositional attributions, to understand why the reduction of dispositional attributions to conditionals is inoperative, and why reduction statements cannot express "all" that dispositional predicates signify; a realism of real properties and not just predicates, based on a causal criterion of existence: a scientific (not instrumentalist) realism that admits, as an abductive hypothesis called for by the explanatory necessity of science, certain real universals; an essentialist (not substantialist) but "thin", dynamic and relational or structural realism requiring a redefinition of both essence (conceived, no longer as a static quiddity, pure natural species, simple bundle of habits, but as a disposition-habitude) and causality itself (not just efficient but final or intentional) and laws: dispositions find their intelligibility in the conditional necessity of laws, but laws are a true description of the world only insofar as they are grounded in what things can do (in the sense of real possibilia metaphysically necessary though discovered a posteriori). We then reviewed the difficulties and challenges facing a metaphysics of natural kinds, looking back at the history of the concept of " natural kind " as proposed by Ian Hacking (Collège de France lecture 2001-2006), its problematic translation (species, kinds, natural groups?), the variety of assessments of the reasons for the importance of classifications - stability (P. Duhem), creation (Nietzsche) - and the different levels of difficulty in the natural sciences and humanities. Can the metaphysics of natural species be reduced to a problem of natural "classifications", and can we consider that "natural classifications do not exist" (cf. Hacking, 2006, last lecture at the Collège de France)? Are natural species simply fictions, creations (certainly "relevant"), but not "natural", of our mind? After reviewing the results of last year's conceptual analysis of the wrong criteria for identifying natural species (resemblance, inductive or predictive success, a priori and semantic criteria), and the main findings of an examination of medieval, modern and contemporary history of the question (Duns Scotus, Locke and Leibniz), we recalled the scope and limits of nominalism and conceptualism in the face of a realist perspective, and set the agenda for the main questions to be addressed: what is the relationship between species and essences? What "knowledge" can be claimed at this stage of the inquiry?