A study of 64 archaeological sites across four continents shows that the growth of agricultural and political systems provoked economic disparities, more so in Eurasia than in North America. See Letter p.619
Understanding the processes that broaden the gap between the wealthy and the poor is a concern for social scientists and policymakers worldwide1. Although economists and historians can identify key factors that drive inequality today, it is less clear where its ultimate origins lie and what dynamics first led to economic differences2. Archaeology is in a unique position to address this issue, because it can be used to study multiple ancient civilizations over long periods of time and before the existence of written records. On page 619, Kohler et al.3 present an archaeological study that provides evidence that increasing dependence on agriculture and the growing complexity of political systems intensified wealth inequalities over time.
Archaeologists have long been aware that members of ancient societies did not always share equal access to resources or opportunities for economic advancement4,5. However, identifying variables that reliably reflect the economic status of households and that are comparable across distinct cultures and time periods can pose a tremendous challenge. For example, offerings placed in graves have been suggested6 as a way of understanding differences in the social status of inhabitants of a site, but interred individuals are not always representative of the entire population — formal cemeteries might be reserved for people of high status. Even comparing the types of possession found in ancient houses can be problematic, because it is rare for a family to leave all of its possessions behind on relocating; instead, certain items of value would probably be selected to be taken. Moreover, the way in which value is attributed might vary greatly between societies.
Using archaeological and historical data, Kohler and colleagues demonstrated that a relatively simple and universal parameter, variability of house size within a community, is highly correlated with the degree of social inequality in that group. In societies in which most people have similar economic standing, houses tend to be the same size. But for groups in which some have greater wealth than others, a mix of smaller and larger houses is more common.
The authors analysed house sizes at 62 archaeological sites across North America, Europe and Asia, and two from Africa (Fig. 1). The sites represent a range of economic systems spanning the past 11,000 years, from those of hunter-gatherers to ancient cities. The researchers measured house-size variability at each site using a statistic called the Gini coefficient, which was originally created to study wealth inequality within and between modern societies7. They then ranked the sites on the basis of these relative wealth-inequality scores.
Kohler et al. found greater inequity at agricultural sites than at sites occupied by nomadic hunter-gatherers or horticulturalists (smaller groups that relied on crop cultivation supplemented by some hunting or fishing). This is logical, because the sedentary lifestyles of agriculturalists came with greater possibilities for accumulating material wealth. The archaeologists also noted that larger populations, complex political systems (such as states) and more-authoritarian regimes were all associated with a higher degree of wealth inequality.
It seems intuitive that hierarchical political systems and large populations present greater opportunities for competition and inequity. But Kohler and colleagues made a second, more unexpected finding using an innovative approach to compare the Gini scores of the sites through time. Instead of simply ordering the sites by calendar years, they created a relative order that was based on the total number of years for which agriculture had existed at earlier sites in the region before the initial occupation of the sites selected for the study.
This allowed better comparison of sites that shared similar stages of economic development, but existed at different periods in time (the Eurasian sites of the Old World tended to considerably pre-date the North American sites of the New World). The results showed that Eurasian sites reached significantly higher degrees of inequality than North American ones, even when their respective agricultural economies had existed for equivalent amounts of time.
Although Kohler and colleagues' observations are intriguing, it is essential to bear in mind that this study is only a first attempt to tackle a highly ambitious goal. In some ways, it raises more questions than it answers. For instance, in the sample of sites studied, it is clear that Eurasia's trajectory towards inequality is different from that of North America — what factors could explain this contrast?
Kohler et al. suggest that it is due to key differences in their agricultural economies, notably, the presence of large domesticated animals such as cattle, horses and pigs in Eurasia, and their absence in North America. The authors argue that the use of animals to plough fields would have allowed more-rapid and more-extensive economic growth in Old World societies. However, because not all members of a community would be able to afford draft animals, and the amount of land available for ploughing at each site was not infinite, competition for these resources would tend to exacerbate economic inequalities over time. The use of the animals to transport goods and to serve as mounts in war campaigns might also have been key factors.
Similar ideas regarding differing resource distributions between the Old and New worlds before the two became culturally connected have been proposed to explain differing socio-economic developments between these regions8. But although this hypothesis is interesting, it needs a great deal of further testing and refining. One key step will be to include the Andean societies of South America in future analyses. These New World groups domesticated llamas and alpacas for food, to provide fibres for textiles and as pack animals for the long-distance transport of goods between ecologically distinct regions9. Will Andean wealth inequalities rank closer to those of Old World societies, or to their North American neighbours?
Furthermore, it is important to acknowledge that the 'Old World' was a vast space, characterized by tremendously diverse biogeography, cultures and economies. In Kohler and colleagues' study it is represented by only 25 sites spanning some 9,000 years, which leaves large gaps in time and space. A more-robust sample of sites that includes greater swathes of Europe, Asia and Africa is needed to determine more precisely how the presence of large domesticated animals was related to growing inequality. Other differences between the Old and New worlds, such as more-sophisticated metallurgy in the Old World, and the possibility that some large New World political systems had more-collective forms of governance10,11, may also be fruitful lines of enquiry.Footnote 1
United Nations. The Sustainable Development Goals Report 2017 (UN, 2017).
Piketty, T. Le Capital au XXIe Siècle (Le Seuil, 2013).
Kohler, T. A. et al. Nature 551, 619–622 (2017).
Flannery, K. & Marcus, J. The Creation of Inequality (Harvard Univ. Press, 2012).
Earle, T. & Kristiansen, K. Organizing Bronze Age Societies (Cambridge Univ. Press, 2010).
Pearson, R., Lee, J., Koh, W. & Underhill, A. J. Anthropol. Archaeol. 8, 1–50 (1989).
Gini, C. Variabilità e Mutabilità (1912); reprinted in Memorie di Metodologica Statistica (eds Pizetti, E. & Salvemini, T.) (Libreria Eredi Virgilio Veschi, 1955).
Diamond, J. Guns, Germs, and Steel: The Fates of Human Societies (Norton, 1999).
Bonavia, D. The South American Camelids: An Expanded and Corrected Edition (UCLA-Cotsen Inst. Archaeol. Press, 2009).
Carballo, D. M. Cooperation and Collective Action: Archaeological Perspectives (Univ. Press Colorado, 2013).
Blanton, R. E., Feinman, G. M., Kowalewski, S. A. & Peregrine, P. N. Curr. Anthropol. 37, 1–14 (1996).