How will human societies evolve in the face of the massive changes humans themselves are driving in the earth systems? Currently, few data exist with which to address this question. I argue that archaeological datasets from islands provide useful models for understanding long-term socioecological responses to large-scale environmental change, by virtue of their longitudinal dimension and their relative insulation from broader biophysical systems. Reviewing how colonizing humans initiated biological and physical change in the insular Pacific, I show that varied adaptations to this dynamism caused diversification in social and subsistence systems. This diversification shows considerable path dependency related to the degree of heterogeneity/homogeneity in the distribution of food resources. This suggests that the extent to which the Anthropocene modifies agroeconomic land surfaces toward or away from patchiness will have profound sociopolitical implications.