War both reduced and increased inequality over the past ten thousand years.

Scholars are divided over the long-term effects that war has had on inequality. Some have argued that conflict grows the gap between rich and poor. Others counter that violence levels out wealth differences. The GINI Project Database is a large global sample of archaeological data on house sizes created to investigate what factors influenced economic inequality over long periods of time, including warfare. Over 39,000 individual residential units were coded as having fortifications present or absent, with about a third in fortified settlements (n = 13,372) and two-thirds in unfortified settlements (n = 25,897). We compared residential disparity (differences in residential unit sizes within a settlement) at sites around the world (n = 770) dating as far back as 10,000 y ago. We found strong support for the expectation that conflict was linked with increasing residential disparity (i.e., wealth inequality), specifically when governance was less collective and the main factor limiting agricultural production was available land. However, we also found long periods, especially in the earliest eras represented in the database, when fortified settlements had residential disparity less than or equal to unfortified settlements. These early societies tended to be more collective with available labor limiting agricultural production. We speculate that in these communities, the relative value of coalition building was higher, whereas in cases where conflict was associated with rising residential disparity, elites found a way to leverage their wealth to protect property. These contradictory models help explain why war co-occurs with increasing inequality in some cases and decreasing inequality in others.