The history of the genitive case from Old English onwards, specifically the diversity of genitival functions in Old English, and the subsequent development of the case as determiner marker as claimed in Rosenbach (2002), can be explained coherently by assuming NP status for genitive nominals and nondistinct categorial status of the determiner class in OE. This article argues that OE, as an inflected language, encodes events in terms of roles carried by participants, and that the genitive case signalled partial or nonparticipation of a designated thing. This enabled a genitive nominal to be subordinated to another NP, which gives a coherent gestalt of the two things, and also to function as a full NP elsewhere. At the same time, the loose categorial status of determiners made it possible for a genitive nominal in prenominal position to function as a determiner. After the OE period, the determiner category became more distinct from others, and a genitive nominal, being a member of this category, ceased to be able to function as NP.