Behind the Study
One of the motivations behind this project was a simple frustration with the way we often study the origins of complex societies. Much of the literature on early states, markets, and urbanization focuses on institutions: governments, taxation systems, bureaucracies, merchant organizations, and legal structures. These institutions are important, but they are also the most visible part of the story. I became increasingly interested in a more fundamental question: what kinds of interaction networks existed before mature states and markets emerged, and could those networks themselves help explain the rise of large-scale social and economic integration?
The Uruk Expansion offered an ideal case study. Between roughly 3900 and 3000 BCE, southern Mesopotamian societies became connected to regions stretching across northern Mesopotamia, western Iran, and southeastern Anatolia. This same period witnessed the emergence of the world's earliest cities, the appearance of writing, growing administrative complexity, and a dramatic increase in long-distance exchange. Yet despite decades of research, scholars continue to disagree about how this integration worked. Was it primarily driven by powerful urban elites seeking access to resources? By colonies and political control? By trade diasporas and migration? Or by more decentralized forms of interaction?
Rather than trying to answer these questions directly, we decided to approach the problem from a different angle. If these different interpretations imply different ways of organizing connections among settlements, perhaps we could learn something by reconstructing the networks themselves.
The project required building what became, to our knowledge, the largest geospatial database yet assembled for the Uruk world. This involved bringing together evidence from decades of archaeological surveys, excavations, and regional syntheses scattered across hundreds of publications. In total, we compiled information on more than 1,700 settlement occupations and nearly 1,000 securely dated and geolocated sites.
A surprisingly large share of the work went into tasks that are often invisible in published research. Archaeologists working in different regions frequently use different chronological systems, terminology, and mapping conventions. Reconciling these differences required a substantial amount of data cleaning and standardization. We also had to make decisions about which sites could be securely dated, which locations were reliable enough to include, and how to deal with gaps in the archaeological record.
The reconstruction of transportation routes presented another challenge. Ancient maps of Mesopotamia do not exist, and rivers, coastlines, and landscapes have changed dramatically over the last five thousand years. To address this problem, we combined archaeological, historical, and geomorphological evidence to reconstruct both river routes and major overland corridors. Importantly, these transportation networks were reconstructed independently of the archaeological evidence used to identify Uruk sites. This allowed us to study how settlements interacted within a broader transportation landscape rather than building connectivity directly from the presence of shared archaeological materials.
Methodologically, the study also provided an opportunity to bring together archaeology and network science. Instead of treating settlements as isolated points on a map, we represented them as nodes connected through transportation routes. We then reconstructed networks for three major phases of the Late Chalcolithic and analyzed how their structure changed over time. To better understand the significance of the observed patterns, we compared the archaeological networks to simulated alternatives ranging from highly localized systems to much more globally connected structures.
What we found was both surprising and exciting. Rather than becoming increasingly centralized over time, the Uruk interaction system evolved toward a more integrated but also more decentralized structure. Local and regional clusters remained important, but they became increasingly connected through a limited number of longer-distance links. In other words, integration emerged not simply through centralized coordination but through the progressive reorganization of connectivity itself.
For me, one of the most interesting aspects of the project is how it connects archaeology to broader debates in historical political economy. Recent research has shown how trade networks influence urbanization, inequality, state formation, and long-run economic development. Our study focuses on an earlier stage of the process: the emergence of the interaction structures that make these developments possible. The findings suggest that before states and markets became formal institutions, the networks capable of sustaining them may already have been taking shape.
More broadly, the project highlights the value of looking beyond institutions alone. States, markets, and bureaucracies do not emerge in a vacuum. They develop within networks connecting people, resources, information, and ideas. Understanding how those networks form and evolve may therefore be essential for understanding not only the origins of the first cities, but also the origins of the institutions that continue to shape human societies today.