Semantics for Pollock’s Defeasible Reasoning

Abstract

Many researchers have proposed argumentation-based rear-soning as a viable alternative to reasoning systems with a flat epistemological structure. Perhaps one of the longest standing approaches has been in the Oscar project, led by John Pollock. Unfortunately, without a formal semantics, it is often difficult to evaluate the various incarnations of defeasible reasoning. We provide a semantics for Pollocks defeasible reasoning in terms of Bondarenko et al.’s unified framework for default reasoning. We also indicate some internal inconsistencies between the motivation behind and definition’s governing Pollock’s system.

An extended version of this paper with full proofs of all theorems is also available. The reader should feel free to contact the authors for a copy.