Households and Hierarchy: Domestic Modes of Production in Leeward Kohala, Hawai'i Island

ABSTRACT

This article describes archaeological evidence for the transformation of Hawaiian society from the fifteenth to nineteenth centuries AD. Using archaeological testing of households coupled with high-resolution survey data, we trace changes in the domestic mode of production in late pre-contact Hawai'i. These analyses yield insights into the transformation of Polynesia's most highly stratified society. The traditional land units (ahupua'a) of Makiloa and Kālala, located on the arid, leeward coast of Kohala, Hawai'i, are investigated with both survey and excavation, and detailed information pertaining to subsistence, household extent, and material culture are reviewed. Changes in the economy and configuration of households from the fifteenth to nineteenth centuries AD are discussed, as are the appearance of elite residences in the later periods. This microscale perspective on the evolution of Hawaiian economy and society provides a necessary complement to a macroscale perspective of human ecodynamics in the Hawaiian archipelago.

Keywords:

• pre-contact
• agriculture
• human ecodynamics
• Polynesia

ACKNOWLEDGEMENTS

The 2007–2008 survey and excavations in Makiloa-Kālala were funded by National Science Foundation Human Social Dynamics program grant BCS-0624238. Additional financial support was provided by the Class of 1954 Fund of the University of California, Berkeley and the University of Auckland. We thank the State of Hawai'i Department of Land and Natural Resources, and the State Historic Preservation Office for permission to carry out archaeological research on State lands. Students from the University of Auckland and the University of Hawai'i HARP field school project assisted with the survey and excavations, and students from The Ohio State University assisted with the laboratory analyses. Robin Connors assisted throughout the excavations. Warren Sharp of the Berkeley Geochronology Center carried out the U-series dating of coral samples. We also wish to thank Michael Graves and Mark McCoy of the HARP project.

Notes

Throughout this paper we use the term ‘feature’ rather than the term ‘site’. This is in contrast to many conventional approaches in archaeology where ‘sites’ contain ‘features’. Our use of the term ‘feature’ is an intentional rejection of this conventional site-based approach (sensu Dunnell and Dancey 1983; Kirch 1985; Weisler and Kirch 1985). Recording the archaeological record at the feature scale allows for greater precision, and it removes the problems associated with over-applying a functional classification to spatially contiguous or adjacent features. This is particularly useful for the Hawaiian archaeological record, in which many features (such as the kauhale residential system) had specific functions, but were often in close spatial proximity to one another.