When beginning to write, the first thing to decide is who are to be the readers of what we are about to write. It is most unlikely that a satisfactory dialogue can be developed between author and readers unless the latter have been correctly identified from the very beginning. As Matthews, Bowen and Matthews (2000: 99) perceptively remark, ‘The essential difficulty [in writing] is in trying to ensure that the thoughts created in the mind of the reader are the same thoughts that were in your mind’. Consequently, the writer must shape what is written to suit the intended audience. Most obviously, this means there will be considerable variation in technical and theoretical content and attendant explanation: specialists in the writer's field will want the details of the basic data but will not need their context explained; in contrast nonspecialists will need to know what the data mean but will not want their minutiae. Similarly, the structure of a piece of writing – its paragraph and sentence forms, its prose style, its use of metaphor, its choice of words, its type of illustrations and tables and so on – should be varied to suit the potential readers. In particular, as archaeological writers, we must try to avoid what some have perceived as a problem in social and cultural anthropological writing, which has been accused by one of its own profession of ‘hibernating in a difficult language’ (Eriksen 2005: 1).