Liminal

1The papers presented in this first volume of Préhistoires de la Méditerranée have in common a concern with the wider communicative practices out of which writing systems first emerged. Their point of departure, in many cases, is the limited relationship between writing and spoken language that characterizes many, and perhaps all, of the world’s earliest known scripts. In Egypt and the Middle East, where the earliest scripts later developed into more extensive systems of language notation, this point of departure is now uncontroversial. By contrast, in other parts of the world (notably Pakistan and India, heartlands of the Harappan civilization) the limited relationship between early script and language has become a subject of rival claims concerning the heritage of modern nation-states and their roots in a “civilized” Bronze Age past (as recently reported in Science vol. 306, 2004: The Indus Script—Write or Wrong?: 2026-2029).

2Claims of the latter sort derive their rhetorical force from an outdated notion of literacy as the dividing line between civilized (or progressive) and non-civilized (or barbaric) peoples. By contrast, the papers in this edition take seriously the diversity and complexity of non-literate communicative practices, as a necessary precursor for understanding the social and cultural milieux out of which the earliest scripts emerged. As they demonstrate, this requires close attention to the material substrates and archaeological contexts of sign-bearing media, as well as the interface between writing and numeracy. Studies of this kind are often most productive when undertaken in a comparative framework, of the sort attempted here. Some of the conclusions reached in the individual papers that follow will undoubtedly prove controversial, and there is much that one could disagree with. Collectively, however, they point the way forwards to a new understanding of the emergence of writing, in which archaeologists (working constructively alongside epigraphers) are now beginning to play a lead role.