Not impossible" vs. "guaranteed possible" in fusion and revision

Abstract

In daily life we have two kinds of knowledge at our disposal, pieces of information ruling out what is known to be impossible on the one hand, and case reports pointing out things which are indeed possible. The fusion of the first type of information is basically conjunctive, while it is disjunctive in the other case. The second type of information has been largely neglected by the logical tradition. Both types can be pervaded with uncertainty. The paper first describes how the two types of information can be accommodated in the possibility theory and in the evidence theory frameworks. Then it is shown how the existence of the two types of information can shed new light on the revision of a knowledge base when receiving new information.