This paper provides a critique of current conceptions of "organizational memory" as presented in a number of recent studies. It briefly reviews some of the rich and varied contributions from both administrative studies and information systems concerning this topic, while at the same time noting the vagueness of the term as it is commonly used. What is of interest is the pervasiveness and perseverance of this nebulous concept across a wide range of disciplinary endeavors. The paper provides an important re-formulation of one aspect of "memory" that is implicit if not explicit in most current views, i.e. the notion of memory as a passive store, arguing instead for an active, constructive view of "remembering" that has a long, if forgotten history within psychology and other fields. In the final section, some implications of such an approach are discussed, paying particular attention to the need for empirical studies of "memories in use" and the need to focus on the active construction of common information spaces from information repositories, and expanding the domain of discourse to include sociological as well as psychological perspectives on concepts such as memory, learning, remembering, talking, etc. in the context of organizations. This re-formulation of the issues surrounding organizational memory has significant implications for the kinds of computer "support" for this phenomenon which might be possible or feasible, which can only be touched on in this paper, but will be extended in future work.