An intra-site spatial analysis of the Late Paleoindian Mackenzie I Site (DdJf-9), near Thunder Bay, Ontario

Abstract

This thesis presents the results of the intra-site spatial analysis of the Mackenzie I (DdJf-9) Late Paleoindian site, located ~30 kilometers east of Thunder Bay, Ontario. The Mackenzie I assemblage consisted of a wide range of artifacts and formal tools suggesting a repeatedly occupied, large-scale, habitation site. The large sample size allowed for the application of in-depth spatial analysis techniques that had not yet been applied to Lakehead Complex habitation sites. The primary goal of this thesis is to determine whether statistically significant and diagnostic patterns exist which represent discrete activity areas, and spatial organization. To accomplish this, six study areas were determined through the visual inspection of artifact distribution. Locations were selected along the site periphery to avoid the more intensely occupied, and potentially less interpretable, areas of the site. After visually comparing multiple aspects of the assemblage, various statistical tests were used to determine the presence or absence of significant patterning, the optimal number of clusters, and cluster composition. Observed artifact patterning was then interpreted using diagnostic patterns discussed in the ethnoarchaeological and archaeological literature. The mixed-methods approach applied within this thesis allowed for the direct identification of lithic manufacturing and tool finishing/re-working activities within the study areas. Additionally, indirect evidence based upon various tool types suggests the presence of more diverse activities (e.g. hide, wood, antler, and bone working, small-scale butchering, and food preparation). Tools associated with these activities were likely not being used within the study areas, but rather represent the manufacture or refurbishing of such tools which were then used elsewhere. Results suggest that many of these lithic activities were co-occurring within the same activity area regardless of tool type or the intended purpose of the tool. The close vertical and horizontal association of various tool types suggests that these lithic activities were not spatially segregated. Conversely, discrete patterns were detected between in-situ activity zones and refuse middens. Evidence within some clusters suggests the consistent reuse of middens across separate knapping events, and possibly separate occupations. The segregation and reuse of middens indicates the purposeful organization of space. This study enabled the exploration of a methodology that could detect spatial organization given the taphonomical processes and poor preservation affecting Boreal forest sites, as well as data derived from Cultural Resource Management (CRM) mitigations. The intra-site spatial analysis of the Mackenzie I site expands upon our knowledge of Late Paleoindian occupation in northwestern Ontario, and provides a better understanding of the spatial organization and behavioural patterns of Mackenzie I’s past inhabitants.