When, in the 1970s, historians of the recent past began seriously to explore the uses of oral history they were, as Alistair Thomson points out in this volume, much criticised for uncritical reliance upon the frailties of human memory. Not all such criticism was misplaced, but, as Thomson describes, the past quarter-century of scepticism and experience has immensely refined the ways in which the method is used and its outcomes interpreted. Yet many historians continue to value documentary over oral sources to a surprising degree, given the extent to which documents throughout history have been derived from oral sources, or were written versions of unspoken memories. If there are serious methodological problems confronting interpretations of the recent past which depend upon memory, such problems arise at least equally for other time periods. The value of the essays which follow, and of the conference at which they were read, is in the focus on the common methodological problems posed to historians and anthropologists of very different time periods and cultures by memory and its oral and written expression: issues of what people do and do not remember, of why and how memory is used to interpret past and present.