Peter Railton

It has been the dominant view that probabilistic explanations of particular facts must be inductive in character. I argue here that this view is mistaken, and that the aim of probabilistic explanation is not to demonstrate that the explanandum fact was nomically expectable, but to give an account of the chance mechanism(s) responsible for it. To this end, a deductive-nomological model of probabilistic explanation is developed and defended. Such a model has application only when the probabilities occurring in covering laws can be interpreted as measures of objective chance, expressing the strength of physical propensities. Unlike inductive models of probabilistic explanation, this deductive model stands in no need of troublesome requirements of maximal specificity or epistemic relativization

Statements about a person's good slip into and out of our ordinary discourse about the world with nary a ripple. Such statements are objects of belief and assertion, they obey the rules of logic, and they are often defended by evidence and argument. They even participate in common-sense explanations, as when we say of some person that he has been less subject to wild swings of enthusiasm and disappointment now that, with experience, he has gained a clearer idea of what is good for him. Statements about a person's good present themselves as being about something with respect to which our beliefs can be true or false, warranted or unwarranted. Let us speak of these features as the descriptive side of discourse about a person's good

Although the standard theory and actual practice of cost-benefit analysis are seriously defective, the general idea of making social policy in accord with an aggregative, maximizing, consequentialist criterion is a sensible one. Therefore it is argued, against Bantz, that interpersonal utility comparisons can be meaningful, and, against both Bantz and MacLean, that quantitative overall assessments of expected value provide a presumptively rational basis for social choice. However, it does not follow that introducing cost-benefit tests into the political or legal process would always be optimal: recognizing some quite stringent legal rights against involuntary exposure to pollution or risk may actually promote cost-beneficial results more reliably than cost-benefit tests employed in very imperfect circumstances

Nous pouvons tous, je crois, reconnaître la justesse de la thèse d'Aristote à l'effet que le véritable raisonnement pratique a pour résultat non pas une simple croyance à propos du caractère désirable, ou même du caractère obligatoire, d'un acte, mais plutôt l'initiation effective d'une action. Cette thèse donne lieu à une énigme : comment la délibération, archétypiquement une inférence propositionnelle rationnelle , peut-elle logiquement aboutir à un acte ? L'action présuppose la motivation, mais la motivation est une force appétitive active, et non pas une proposition concernant le désir ou le caractère désirable d'un acte . La raison pratique n'est pas similaire aux substances qui peuvent manifestement être mélangées les unes aux autres, mais elle est censée constituer une fusion normative régie par des principes — le choix est un appétit délibéré , comme le dit Aristote. Comment ceci est-il possible ? Les premiers pas vers une réponse peuvent être trouvés à un endroit surprenant : chez Kant. Dans la Métaphysique des Moeurs , Kant présente une théorie du désir comme appétit pour un objet par le biais d'une idée non-sensorielle de cet objet, et ce, par opposition à une inclination , un appétit direct basé sur la sensation. Kant emploie cette notion du désir pour expliquer à la fois la volonté et la possibilité d'une action accomplie par devoir : la volonté appartient à la faculté de désirer, et le devoir est l'action motivée par un désir pour un acte par le biais d'une idée appropriée de cet acte. Kant nomme notre susceptibilité à ce type de désir laquelle constitue une condition nécessaire à la moralité le sentiment moral . Ce sentiment rend compte du caractère pratique de la délibération morale et de son rapport au plaisir et à la douleur pratiques d'une façon qui aurait, je crois, plu à Aristote. Le devoir est donc un exemple d'un appétit délibératif . Ce schème kantien n'est pas simplement intéressant d'un point de vue historique : il s'accorde remarquablement bien avec certains travaux récents en psychologie motivationnelle et hédonique.We all can, I think, appreciate the justness of Aristotle's view that genuine practical reasoning ends not in a mere belief about the desirability, or even obligatoriness, of an act, but in the actual initiation of action. This yields a puzzle: how can deliberation, archetypically a rational, propositional inference , logically terminate in an act? Action presupposes motivation, yet motivation is an active appetitive force, not a proposition about desire or desirability . Unlike substances can obviously be mixed together, but practical reason is supposed to a principled, normative fusing — choice is deliberated appetite , as Aristotle puts it. How is this possible? The beginnings of an answer can be found in a surprising place: Kant. In The Metaphysics of Morals Kant gives a theory of desire as appetite for an object through a non-sensory idea of it, in contrast to inclination , a sensation-based direct appetite. Kant uses this notion of desire to explain both will and the possibility of action from duty : will belongs to the faculty of desire, and duty is action motivated by a desire for an act through an appropriate idea. Kant terms our susceptibility to this sort of desire moral feeling , a necessary condition for morality. It accounts for the practicality of moral deliberation, and its linkage to practical pleasure and pain, in a way Aristotle would, I think, have found congenial. Duty, then, is an example of deliberated appetite . The Kantian scheme is not merely of historical interest: it fits surprisingly well with recent work in motivational and hedonic psychology.