ABSTRACT. Pragmatists and those with affinities for pragmatism, such as Wittgenstein, have argued that the sorts of rules or norms fit to constitute anything as a community, including linguistic norms, must be more than formal, must be embedded in practice. It is argued that linguistic norms or rules cannot be construed as explicit rules without risking an infinite regress, and that the following of implicit rules is shown only in communal practices apart from which distinctions between correct and incorrect cannot take hold. The rules or norms in question are well taken as something more than rules of thumb. Arguments designed to provide an adequate account of rule following without reference to normatively constituted community are criticized.