This paper discusses démarcheurs, intermediaries in the second-hand car markets in Cotonou, Bénin. An ethnographic case study shows how démarcheurs make a profit by creating barriers between car buyers and sellers. In this way the paper presents an alternative interpretation of intermediaries: not as brokers of market information in fragmented business networks but as skilful information manipulators who pretend to be important. By analysing how démarcheurs interpret the car business as an information game, the paper tries to make understandable the cultural logic of their economic behaviour. This shows that market information is socio-culturally constructed knowledge and that intermediaries play a crucial role in its construction.

This article addresses the contrasting pull of two tendencies in anthropology: (a) calls to redress the purification of human from non-human actants and (b) calls to denaturalise notions of borders as things, foregrounding borderwork. The resulting dilemma – do we treat people and things as equivalent actants on a ‘flat’ plane or not?– is explored through an ethnographic exercise on the border that divides Sarajevo. This case study crystallises methodological possibilities, implications for critique and matters of accountability presented by either path. Ultimately, I argue, a focus on things is productive insofar as it functions within a focus on human practice.

An international idea is that the world must be ‘prepared’ for any disaster situation. Among the many tools and practices that contribute to this frame, the paper focuses on exercises intended to prepare for natural disasters: real-scale simulation exercises. The object of this paper, based on several studies conducted at sites overseen by the UN natural disaster reduction agency (ISDR) and on a field study of a simulation in Peru in November 2010, is these exercises and their purposes and outcomes. It also explores the conditions of possibility for their ethnography.