Shipibo archaeo‐ethnography: Site formation processes and archaeological interpretation

Abstract The structures of two Shipibo house compounds are considered. One described by DeBoer and Lathrap (1979) represents the on‐going ethnographic context. The other was recently abandoned and therefore represents the archaeological record. A k‐means cluster analysis program is used to investigate the spatial structures of both the ethnographic and archaeological contexts. In doing so, the systemic context becomes a model against which the archaeological setting may be compared. We find that there is a variety of spatial relationships between use and discard locations manifested by a single cultural group, the precise nature of which is constrained by the range of activities conducted by the occupants of the group. Based upon this analysis we suggest that the correspondence between use and discard areas should not only be examined in terms of occupational intensity or enclosed vs. open‐air activities, but should also include a consideration of the variability in the settlement activity organization. Theoretical expectations and sampling considerations are proposed for excavating prehistoric settlements in the South American lowlands.