Dynamic optimization of horticulture among the Muscogee Creek Indians of the southeastern United States

This paper reviews archaeological applications of optimization theory applied to resource use. A model of agricultural corn production is derived for a single Muscogee Creek Indian town (CA 1725–1825) which was situated in the southeastern United States. The corn productivity model is based on soil depletion, soil variation, known locations of horticultural fields, population size, and nutritional requirements. The corn productivity model shows that corn production varied significantly between 1725 and 1825. The residents of the town shifted their horticultural gardens in order to maintain a nutritional minimum. The model was compared to the marginal value theorem and expectations derived from risk minimization models. Residents of Cussetuh abandoned their horticultural fields well before the instantaneous long-term average rate of production for the habitat. The model shows that the residents of Cussetuh were risk minimizers and were not maximizing the long term average rate of corn yield in their gardens.