Four rungs
A came before B. A correlates with B. A predicts B. Changing A changes B. Only the last is causation, and usually only intervention shows it.
Confounders and colliders
A third variable can create a correlation from nothing — and controlling for the wrong variable can create one too.
The counterfactual question
Would B have happened anyway? Every causal claim is secretly a claim about a world that did not occur.