BETWEEN JANUARY AND MAY 1981, I was interviewing people in Manchester (north west England) about the supranormal-ESP, psychic powers, the influence of the dead on and in the mundane world, and so on (Bennett 1985). During the course of these interviews, the respondents spontaneously told 153 narratives. The storytelling patterns revealed in this corpus have led me to doubt the universal applicability of the most commonly used model of oral narrative structure (Labov and Waletsky 1967) and to look for alternative ways of defining and describing narrative. As background to the discussion, I want first to give a brief outline of the context in which the stories were volunteered and to summarize the conventional Labovian definition and model. The main