As the editor of the Journal of World Prehistory I am supposed to know what the spatio-temporal boundaries of my publication are. Spatially it is easy: the world. Temporally it is not easy: the archaeologically-studied cultural phenomena of the world include not only the Oldowan, the Lengyel and the Hohokam (random examples), but also the Tasmanians, the Moriori and the urban residents of Tucson, Arizona. The last are, obviously, a modern, urban and 'historical' community; nevertheless the data from Rathje's Tucson Garbage Project surprised sociologists and historians by measuring what people actually did with objects rather than what they believed they did: archaeology here uncovered a sort of 'prehistory of the present'. But what is the history-prehistory distinction? Does it appear as a complex chronocline that we must apply like a map of barymetric pressure over the globe, with tight circles covering late fourth millennium BC Egypt and third millennium Mesopotamia, and a series of ever widening concentric rings, finally reaching the furthest corners of Australiasia, Polynesia and Amazonia at the present time? As the saying goes: 'that's all very well in practice, but it won't work in theory'. This paper will examine the various meanings of prehistory (and Urgeschichte) and makes some tentative proposals for a servicable working definition for 21st-century transdisciplinary scholarship.