Narrating Witchcraft:Agency, Discourse, and Power Esther Eidinow and Richard Gordon This collection of thirteen essays—three issues of a special volume of this journal—explores "witchcraft narratives" over time, place, and culture. Most of the contributions are based on papers delivered at a conference entitled "Narrating Witchcraft: Agency, Discourse and Power," held at the University of Erfurt at the end of June 2016, which took up the theoretical challenge of the Lived Ancient Religion project (LAR). This five-year research project, which ran from 2012–17 and which provided funding for the conference, was directed by Professor Jörg Rüpke of the Max Weber Center for Advanced Cultural and Social Studies at the University of Erfurt, Germany, and focused on the individual, situational, and larger global trajectories of the pluralistic religions of antiquity. This special volume, Narrating Witchcraft, takes a similarly multi-dimensional approach: scholars of different disciplines, including ancient, early modern and modern history, anthropology, and gender studies, introduced a broad variety of case studies. In acknowledgement of the theme of lived ancient religion, most of the contributions to this special volume—specifically, the articles that appear in the first two issues and that set up the thematic content—focus on malign magic as perceived and practiced in the "Ancient Mediterranean," which includes the Ancient Near East and connected areas over a time span of some thirty-five hundred years. Articles that continue to explore the theme in the early modern and modern worlds are contained in Issue 3. The original conference was explicitly framed as a project in comparative history with an emphasis on the ancient world. While the conference organizers and participants acknowledged the difficulties of cross-cultural translation, we worked with a shared understanding of the term "witchcraft" along [End Page 1] the lines of what Ronald Hutton recently described as standard scholarly definitions of witchcraft: causing "harm to others by mystical means," or "evil magic and sorcery."1 The papers presented at the conference did (and the articles across the volume do) discuss points of terminology (see especially the insights of Peter Geschiere in Issue 3), and work from a wide variety of primary material, introducing diverse methodologies. But they start from the assumption that the conceptual power of witchcraft beliefs and practices is both diachronic and cross-cultural, encompassing both local and global meanings, and with a historical trajectory that extends back to ancient Mediterranean cultures. This starting assumption has allowed them to focus on exploring witchcraft's dynamic, cross-cultural manifestations: in particular, the question of how, across and within widely differing discursive representations, these manifestations each maintain their own distinctive character, while still achieving a "family resemblance," as Esther Eidinow (this issue) describes it. The question of representations introduces the other key term of the conference and volume title—"narrating"—and our concern with the ways in which witchcraft is both engendered by narratives and, in turn, creates them. Narration here is considered as storytelling, and a narrative as a story; as Aristotle suggests of plot, it is a mimesis of a sequence of events with a beginning, middle and end, of a length that can be coherently remembered, of a size which permits a transformation to occur in a probable or necessary sequence of events, from adversity to prosperity, or vice versa. In terms of contemporary theory, we here orient ourselves toward the general theory of narrative elaborated by Albrecht Koschorke, which is useful because its framework is geared to adapting narrative theory to social science. It also considers cognitive processes, where, in an elementary sense, narrative is considered as a kind of "reduction" of knowledge, in which "narrative patterns thus function similarly to cognitive schemas allowing the surplus of unsorted empirical data to be reducted to typical, easily recognizable forms."2 The reduction is not limited to any particular core of real acts or real things; indeed, its interest for us, as for him, has to do with it the way it can both appear to be, and to manipulate, experiential truth; narrative has the capacity to create both orientation and nonsense, both order and disorder. As Koschorke writes, [End Page 2] Like thinking and speaking in general...