The problem of ethnography has cast a long shadow over the practice of media and cultural studies. The disputes concerning the ethics and epistemological coherence of fieldwork that split anthropology in the 1980s seem to have transferred some of their force to recent debates about how we can study media audiences across the world. It would, however, be a mistake to discuss the epistemology of the media audience as if the audience researcher carried a pale version of the colonialist’s historical burden, although the rhetoric of some attacks on audience research suggest this (Hartley, 1987, 1996; Nightingale, 1996). For that would obscure a more interesting question, whose significance stretches well beyond media analysis: what kind of “location-work” (Gupta & Ferguson, 1997b, p. 5) will enable us to address the locational complexity which characterises all social and cultural phenomena today, not least those marked by the multi-directional flows of media images?