Kuznets at -7000: Is there a really long-term relationship between growth and inequality?

We use archaeological data on house sizes to generate estimates for economic inequality and economic growth from near the beginning of the Holocene to about the first millennium AD. At worldwide scales these variables are positively related, but patterns are more divergent at regional levels. Crosssectional regression shows that the formation of central-place hierarchies and development of landesque capital are important in generating both wealth and wealth inequality; development of farming is also important to the generation of wealth. Iron smelting detracts from wealth inequality whereas copper smelting detracts from the generation of wealth. Examination of three well-known interaction zones (Bronze Age West Asia, the Classic Maya world, and first-millennium-AD Britain) shows surprisingly regular transformations of the relationship between economic growth and inequality on millennial time scales.