The vignette is a valuable and flexible research tool in behavioural sciences and health care. Joel Richman and Dave Mercer outline the variable application of this method and propose a classification in terms of the form and use of the vignette. The discussion is illustrated with an example of the way that vignettes provided access to the discursive structures of forensic mental health nurses in one maximum secure psychiatric hospital. It is suggested that this methodology, within the broader aims of the study, revealed an ordered and taxonomic discourse of evil, with significant implications for professional practice