An open logic system admitting modification of inference rules (Q1814724)

Under the AGM account of belief change, a belief state is taken to be a set of propositional (or eventually first-order, etc.) formulae (sometimes constrained to be closed under consequence). The author considers taking as a state a set of inference rules over such formulae. Some definitions are given, e.g. of consistency in such a context and of maxichoice revision, and one theorem is proven, saying roughly that we can go from any one consistent state (so defined) to any complete one by a sufficiently long (and possibly transfinite) sequence of maxichoice revisions. Comments: The language used is first-order, but its power appears to be cut back to the propositional level by restricting the main definitions and the theorem to the case that all formulae in rules are closed. The reader should be wary of a number of typos.