1. Manipulating the divine – an introduction Jesper Frøkjær Sørensen & Anders Klostergaard Petersen Concepts in jeopardy In his classic of anthropology, Evans-Pritchard described how the Central African people Azande explain all types of misfortune as caused by the actions of a witch. Witchcraft, mangu, provides “a natural philosophy by which the relations between men and unfortunate events are explained and a ready and stereotyped means of reacting to such events” (Evans-Pritchard 1937: 63). It is a local explanatory principle that deals with experiences of contingency – the famous “second spear” (umbaga) that, while recognizing the material causes of event (its “first spear”), allows the Azande to moor all unfortunate aspects of events securely to the social realm as ultimately being caused by human agency. However, as a cultural model and an explanatory principle, witchcraft cannot be seen. Nobody is ever observed performing witchcraft, even if anyone can be accused of being a witch. Therefore, in addition to the performative statements involved in accusations, in Evans- Pritchard’s view the stability of the belief depends on two types of concrete practices. The first one is constituted by oracles, soroka, used both to diagnose an event resulting from witchcraft and to identify the witch. The second one is represented by magic, ngua, used as a means to punish the alleged witch in cases where direct physical retribution is not an option. The three institutions, witchcraft, oracles and magic, form a triangle, each one supporting the other two and, thereby, creating a stable pattern of cultural representations and practice. But was Evans-Pritchard really justified in translating these local concepts into ‘witchcraft’, ‘oracle’ and ‘magic’? While Evans-Pritchard’s meticulous field-studies and his acute sense of local conceptual details have rightly placed his work among the all-time classics of anthropology, modern readers will look in vain for explicit reflections on the use of these core concepts. 1 Since then, much water has run under the bridge 2 and postcolonial critiques have rightly 1 In chapter two (affixed as an appendix in the abridged reprint), Evans-Pritchard provided a list of Zande concepts translated into English, but his reflections upon this process are limited to the dry observation that “I am not anxious to define witchcraft, oracles and magic as ideal types of thought, but desire to describe what Azande understand by mangu, soroka, and ngua” (1937: 8). 2 Of particular interest is the role of Evans-Pritchard’s work in the so-called ‘rationality debate’ in the late 1960’s and early 70’s. For a collection of the key contributions, see Wilson (1970).