Jay Greenberg proposes that the underlying assumptions informing the analyst’s work can usefully be viewed as controlling fictions. He describes some of the experiences that led him to think in this way and explains why he prefers to talk in terms of ‘fictions’ rather than models or metaphors and argues that thinking in this way can enable and enliven the analyst’s work, and that those who view it as introducing a paralyzing uncertainty are themselves under the influence of a particular controlling fiction.