Contemporary discussion about the ontology of society identifies two groups of perspectives. One of them, associated with Searle, includes rules in the inventory of elements that constitute social reality. The other one, associated with Smit, Buekens, and du Plessis, claims that rules can be reduced to more fundamental units. Despite the fact that both perspectives seem equally efficient in describing institutional phenomena, we identify both flaws in the viewpoint that dismisses rules and reasons to prefer the alternative position.