The paper discusses problems in defining explicitation (and implicitation) in translation for the purposes of descriptive translation studies, drawing on empirical material and theoretical concepts such as frame theory or figure/ground alignment. Inconsistencies in different approaches taken to the twin concepts in prescriptive theory as well as descriptive research are pointed out. The first main focus of the paper is the connection between explicitation/implicitation and specification/generalization, which is shown to be more complicated than usually described in literature. The second main focus of the paper is the borderline between explicitation/implicitation on the on hand and addition/omission on the other. It is argued that the borderline is intrinsically fuzzy, due to the cognitive mechanisms involved in retrieving implicit information that can be modelled by means of Fillmore's frames, understood as non-accidental networks of salient and less salient contiguities. It is argued that explicitation and implicitation should be treated as prototype categories.