Variations on the Ramsey test: More triviality results (Q1100192)

The so-called Ramsey test is a condition for when a conditional proposition is acceptable in a given state of belief. It is often formulated as (R): \(x>y\in A\) iff \(y\in A\dot +x\), where \(A\) is any theory (closed under consequence) and \(A\dot +x\) is the theory obtained by revising \(A\) so as to incomporate \(x\). \textit{P. Gärdenfors} [``Belief revisions and the Ramsey test for conditionals'', The Philosophical Review 95, 81-93 (1986)] observed that so formulated, the Ramsey test immediately implies monotony of \(\dot+\), i.e. (M): \(A\subseteq A'\) implies \(A\dot +x\subseteq A'\dot +x\), which in turn conflicts, in all but ``trivial'' cases, with certain other very natural conditions on \(\dot+\). In the paper under review, he experiments with some weaker forms of (R), e.g. (WR) \(=\) ``if \(x\vee y\not\in A\) then (R)'', and shows that most of these yields a weaker version of monotony of \(\dot+\) that still sufficies to engender similar conflict.