In this book I have attempted to create a new agenda for the study of Britain in the last millennium BC. The book consciously sets out, in its structure and content, to direct attention away from the nature of the archaeological record towards the nature of past human societies. This does not mean I am not interested in the archaeological record, and readers will have noted there is a considerable amount of detail in the text, perhaps too much for some people; but the data has to be examined in relation to the people who lived in a particular place at a particular time: ‘the archaeologist is digging up, not things, but people’ (Wheeler 1954b: v). The objective has been to outline the overall constraints of place and time (Chapter 2) and to see how these created a distinctive archaeological record that differed not only from other areas of Britain, but which varied significantly within the region. I examine how people created communities (Chapter 3) and explore how the mechanisms used to organize human relationships, within that society, changed through time. These changes were partly brought about through events outside their control, but always in a way that was affected by their own particular circumstances. I consider how the most ubiquitous architectural form in later prehistory, the house, was used to structure social relationships on a daily basis in relation to the family, and how this provided a template for thinking about the world (Chapter 4). The analysis concludes with an examination of how these societies considered individual freedom and connectedness, and how the complex variability of individual agency provides an internal dynamic to social change that was influenced by external events, but not led by them (Chapter 5). When I originally conceived of this book the structure was reversed: I started with the individual and worked up to the organization of the larger landscapes. At first sight this may sound a more sensible way of presenting the evidence, moving from small-scale structures to large-scale processes, but during the writing of the book I found this did not seem to work.