Settlement or Settlements?

Just as the United States Civil War has been said to have been fought over a point of grammar (whether to say ‘the United States are’ or ‘the United States is’), there is an important conceptual difference between a system of settlement and a system of settlements. Most, though not all, settlement system analyses have been based on the notion of settlements as basic units of analysis, and a one-for-one correspondence between archaeological sites and human settlements has often been assumed (usually implicitly). Sites are frequently classified into a settlement typology, with different settlement types reflecting functional differences and particular kinds of relationships between settlements. Since settlement system analysis has frequently been an integral part of research on the emergence and development of complex hierarchical societies, the settlement typologies utilized have often reflected the hierarchical relationships between communities that constitute regional-scale political and economic organization in such societies - schemes like hamlet, village, town or district center, and city or regional center.

Such procedures have been criticized for ignoring evidence of human activities that can be found outside what are conventionally identified as archaeological sites. In their more extreme forms, these criticisms have gone so far as to deny that human settlements or communities even exist. This is an obvious exaggeration, since the essential notions of such settlement typologies are derived from concepts of different kinds of communities. These have amply demonstrated their usefulness in characterizing settlement variability which is readily observable ethnographically. Typologies of settlement can become so central to settlement system analysis, however, that careful attention may be required to recognize that, in at least the best of such work, sites are not the basic observational units in the field but are defined post hoc on the basis of artifact densities and distributions, and the typological classification is explicitly justified in these terms, as well as in terms of archaeological features visible on the surface. It is also true that in some cases, archaeologically observed distributions of settlement evidence defy characterization in terms of discretely identifiable settlements or local communities. In such instances, some commonly pursued modes of settlement system analysis cannot be applied.