Many people have wanted to give an account of logical validity which pays some attention to conditions of relevance. If \(X,A \vdash B\) holds, then \(X\) must somehow be relevant to \(A\). Premise combination is restricted in the following way. We may have \(X \vdash A\) without also having \(X,Y \vdash A\) . The new material \(Y\) might not be relevant to the deduction. In the 1950s, Moh (1950), Church (1951) and Ackermann (1956) all gave accounts of what a ‘relevant’ logic could be. The idea