This paper discusses the need for a hierarchical arrangement of explanations used by archaeologists to deal with processes that occur over different scales of space and time. The aim is to define what type of hierarchy is needed. In a hierarchical structure of theories, large-scale processes cannot be reduced to small-scale ones and small-scale ones are not determined by large-scale ones. By contrast, a reductionist view holds that the nature of the smallest component sufficiently explains the largest system or is essential for a basic understanding of it. Hierarchical explanations are usual in the biological sciences, the earth sciences and the hard sciences, contrary to much misapprehension about them by humanistic disciplines. The trend is toward these integrative explanatory structures.

A version of hierarchical explanation introduced to history by the Annales school (Braudel in particular) suggests that it is not incompatible with studies of human beings. Biologists such as S. J. Gould have no difficulty in arguing that history – in the sense of successions of unique events – matters in a study of the vast patterns and processes of biological evolution, and is consistent with “science.” It follows that a history/science dichotomy in archaeology or any study of human beings is founded on suspect premises about the significance of uniqueness and determinism. Since biological, geographical, and historical theories are regularly used in archaeology, tacit hierarchical arrangements of theory are accepted but without a rigorously defined hierarchy of explanation.…