Philosophy
The Logic of Relevance Logic and the Relevance Condition
Quick fact
Classical logic accepts the argument 'If pigs can fly, then I am the king of France' as valid, but relevance logic rejects it because the premise and conclusion share no content—a violation of the relevance condition.
Why this is interesting
You know that if you drop a glass, it breaks—but did you know that in classical logic, from 'the moon is made of cheese' you can derive '2+2=5'? How can that be?
Read the full explanation
Understanding The Logic of Relevance Logic and the Relevance Condition
Start with everyday implication. We usually think 'if P then Q' means there is a connection between P and Q. But classical logic defines implication purely by truth tables: 'P → Q' is false only when P is true and Q is false. This leads to paradoxes like 'a false statement implies anything' (from P being false, P → Q is always true) and 'a true statement is implied by anything' (if Q is true, P → Q is true for any P). These are not just curiosities; they allow arguments that feel utterly irrelevant. For example, 'It is raining, therefore it is raining or the sky is green' is valid, but in relevance logic we demand that premises actually help establish the conclusion. The core idea is the relevance condition: for an argument to be valid, the premises must be genuinely relevant to the conclusion—typically, they must share at least a variable or predicate. This simple constraint filters out the irrelevant validities.
A deeper explanation
Relevance logic works by redefining the notion of entailment. In classical logic, entailment is just necessary truth-preservation. In relevance logic, we add a syntactic and semantic requirement: premises and conclusion must be relevantly connected. The most common syntactic condition is variable sharing: in any valid implication 'A → B', the antecedent A and consequent B must share at least one propositional variable. This blocks the paradoxes immediately. For instance, 'P and not P ⇒ Q' is invalid because Q shares no variable with the antecedent. Semantically, relevance logics use models where an implication is true at a world only if the consequent holds in a related world—often using a ternary accessibility relation. This allows the truth of 'A → B' to depend on the information content connection, not just on the truth values of A and B. The practical importance is that relevance logic provides a more accurate model of real reasoning, especially in areas like computer science where systems must not infer irrelevant or spurious consequences from inconsistent premises. It also links to paraconsistency, as relevance logics often reject the principle of explosion, but the motivation is different: relevance first, consistency later.