INTRODUCTION

Archaeology is based upon labels: from the artefact in the museum case to the culture we are describing, we provide it with a label. This seemingly small and unproblematic description encompasses much: a geographical area, a time period, or a group of people, often all three. Such labelling activity is rooted in the origins of archaeology as a discipline, and in spite of recent disquiet about their usefulness, it is still central to the way we communicate and practise archaeology. The problem is that a label is more than a neutral shorthand, but forms the fundamental core of the way in which we conceptualise the world, past or present. If we describe a pot as Bronze Age, or Minoan, or Anglo-Saxon, we are assigning it a category which fits into our divisions of cultures in the past, and our understanding of their relationship with others. Therefore, if we describe a pot as Roman, we are grouping it together with a series of other pots and artefacts we recognise as Roman. The problem with this approach is that when we are using such a label to describe Roman material culture, we are using it to describe a set of material which is far spread in time and in space. Yet it has increasingly been recognised that when we examine the material culture in detail, there is similarity, but not homogeneity, both in form and patterns of use.