Virtual Information in the Light of Kant’s Practical Reason

In (D’Agostino M, Floridi L, Synthese 167:271–315, 2009) the authors face the so-called “scandal of deduction” (Hintikka J, Logic, language games and information. Kantian themes in the philosophy of logic. Clarendon Press, Oxford, 1973). This lies in the fact that the Bar-Hillel and Carnap theory of semantic information implies that tautologies carry no information. Given that any mathematical demonstration and more in general every logical inference in a first-order language can be reduced to a tautology; this would imply, that demonstrations bring no fresh information at all. Addressing this question (D’Agostino M, Floridi L, Synthese 167:271–315, 2009) offers both: (i) a logical model for a strictly analytical reasoning, where the conclusions depend just on the information explicitly present in the premises; and (ii) a proposal for the ranking of the informativeness of deductions according to their increasing recourse to so called “virtual information”, namely information that is temporarily assumed but not contained in the premises. In this paper I will focus on the status of virtual information in its connection with the Kantian philosophical spirit. Exploiting the standard Kantian difference between theoretical and practical reason, my aim is to show that the access to virtual information is due to what Kant calls practical reason rather then to the theoretical one, even though the effects of its deployment are purely theoretical, i.e. don’t lead an agent to any moral action but just to acquiring new information.