The seventh lesson began by looking back at one of the strongest arguments in favor of epistemic optimism, and clarifying the epistemic value of abduction. Objections to abduction were recalled: supposed confusion between the logical (justification) and psychological (discovery) planes; between the power to guess and inference (the surprising fact C, is observed; but if A were true, C would be self-evident; hence, there is reason to suspect that A is true). We explained how abduction could be a good method and be used as an inference, if not to the best, at least to a "good" explanation.
We then analyzed the form we believe the scientific realist's metaphysical commitment should take, resting on four essential theses 1. A causal theory of properties. 2. A conditional dispositionalist conception of laws. 3. A dispositional realism that fears neither a certain quidditism, nor final causality, nor the necessity of certain laws. 4. A realism convinced of the need to take into account all the categorical furnishings of the world. Starting from the principle that a project aimed at the metaphysical knowledge of nature must know how to count things, we then sketched out the steps that such an inquiry into the real properties of nature should follow (the more detailed analysis being carried out in parallel in the seminar). In particular, this should make it possible to rehabilitate certain properties long held to be obscure, eliminated and then simply reduced, before finally being held to be "real", even "essential", namely dispositional properties. This made it possible to put forward objections against both "categoricalism" (D. Armstrong) and "pandispositionalism" (D.H. Mellor), as well as initial arguments in favor of dispositional realism . A number of avenues emerge:
- Against metaphysical realism, we need to look for a form of scholastic realism.
- We need to start by carrying out a logical and semantic analysis that reveals the meaning of our attributions of dispositional properties, making the necessary distinctions between properties and predicates, and between certain kinds of properties and certain kinds of predicates, between the concept of "power" and the properties by virtue of which things have the powers they have. at the same time, we need to question the correspondence between this distinction and that between dispositional and non-dispositional predicates, or between dispositional and categorical properties or predicates: does "categorical", for example, mean "unconditional"?
- Then we need to provide precise criteria for identifying properties, and assess the nature of the link (direct or indirect?) between dispositional attributions and conditional subjunctives, and the consequences of the difficulties posed by non-verifiable conditionals, "antidotes" and "treacherous dispositions".
