The archaeology of Fjordland Archipelagos : mobility networks, social practice and the built environment
Investigation of the archaeological record of hunters and gatherers has been frequently concerned with the origins of social complexity. Yet, 'social complexity' is not a straightforward variable, and the category 'complex hunter-gatherer' may create more problems than it solves. Rejection of the category does not, however, eliminate nor account for real variation in the social organisation and archaeological signatures of hunter-gatherers. Archaeological analyses of hunter-gatherer economies have frequently considered time-budgeting constraints associated with the production, storage, and redistribution of surpluses to be central. However, examination of these time constraints show that they are not necessarily a constraint upon the development of social complexity, but are an expression of the relationship between individual humans and their environment. Spatial and temporal constraints are manifested through the individual's body, and are expressed through technology, settlement pattern, and mobility practice. Some spatial approaches in archaeology, such as locational analysis, have focused on the individual monad but few have done so in a manner that adequately expresses the possibilities and constraints of the individual and individual agency. Instead, most analyses have cast the individual as either a simple optimising 'Homo economicus' making rational decisions within a neutral environment, or as subject to a highly normative culture, or both. It is argued in this thesis that reconceptualising the individual as living within a 'habitus' may be conducive to understanding some aspects of the archaeological record. In particular, conceiving of the individual - environment relationship as one of non-Cartesian mutualism leads to an appreciation of the paired importance of the mobile individual in a built environment. From this perspective, a case study from Vancouver Island on the Northwest Coast of North America is introduced. The pmedian model in a Location-Allocation analysis is applied to a network formed by transportation linkages between 238 habitation zones, created by clustering 576 archaeological sites. It is shown that centrality of place within a network matters, as the more central places are also larger sites, but this pattern only occurs at a spatial scale difficult to reconcile with deliberate optimising behaviour. It is therefore concluded that this descriptive spatial geometry is irreconcilable with any plausible underlying generative social geometry based on either normative cultural rules or deliberate optimisation. Recognition that the built environment is an interrupted process rather than a planned, finished product, allows one to avoid imposing the 'fallacy of the rule:' in this case ascribing to the inhabitants of the study area a totalised decisionset for site location and intensity of use based on the location-allocation solution sets. Instead, it is argued that the observed spatial patterning in the case study is better seen as the archaeological signature of long-term, wide-scale, practical activity of individuals within a landscape of habit. The result is the discovery of an important threshold in the spatial scale of the culturallyperceived environment. Discussion follows of the implications of this thesis for the interpretation of social complexity, for the predictive modelling of site location, for the establishment of relevant spatial units of analysis, and for such familiar spatial ecological variables as 'population density,' on the Northwest Coast and elsewhere.