How Humans Invented the City and the State

Sometime between 1922 and 1934, the British archaeologist Leonard Woolley dug into a low mound on the lower Euphrates in southern Iraq and found himself standing inside a city that had already been ancient when Plato was a boy. The mound was Ur, one of the oldest of the Sumerian capitals. Beneath its tells lay the Royal Cemetery, and inside it, the gold-laden tomb of a woman the inscriptions called Puabi, buried around 2600 BCE with her headdress of gold leaves and lapis lazuli still in place. Woolley had not simply found a rich grave. He had walked, spade in hand, into the deep history of urban life on Earth, into a world of temples, scribes, kings, and crowds already functioning two and a half thousand years before classical Greece existed.

That single excavation poses the question this article is about. Cities feel so natural to us that more than half of all human beings now live in one, yet for almost all of our species' existence there were no cities, and no states either. Somewhere along the line, people stopped living only in villages of a few hundred and began assembling settlements of tens of thousands, organized around institutions that had never existed. How did that happen, and where?

The City That Came Before Athens

If Ur was the cemetery that announced the discovery, Uruk was the city that started it all. Founded around 3500 BCE on the lower Euphrates in southern Mesopotamia, Uruk is generally regarded as the world's first true city, and the qualifier "true" matters, because it distinguishes Uruk from the large farming settlements that preceded it. Uruk was not just big; it was dense, internally differentiated, and organized around a monumental center, and it kept growing, reaching somewhere between forty thousand and eighty thousand inhabitants by around 2900 BCE.

To feel the weight of that number, set it beside a more familiar city: Athens, at the height of Plato's Academy two and a half millennia later, was smaller than Uruk had been at its peak. Sitting on a flat plain between the Tigris and the Euphrates, Uruk had crossed a threshold of scale that later Mediterranean city-states would not match for millennia. The Sumerians who built it were the same people who, a few centuries on, would bury Queen Puabi at Ur with a retinue of attendants and a hoard of worked gold, a coherent urban culture elaborating itself across the southern floodplain for the better part of a thousand years.

Six Cradles That Never Met

It would be tempting to treat the city as a single invention that spread outward from Mesopotamia, the way a technology diffuses from the place that made it, but the archaeological record refuses that tidy story. Cities and states arose independently in at least six different regions, on separate continents, among peoples who had no contact with one another whatsoever. Mesopotamia, Egypt, the Indus Valley, China's Yellow River basin, Mesoamerica, and the Andes each produced their own urban revolution, on their own timetable.

Scholars call this pristine state formation, meaning the emergence of cities and states from scratch, without any preexisting state nearby to copy or to be conquered by. The distinction matters, because the vast majority of states in history are secondary: they formed in reaction to, in imitation of, or under pressure from states that already existed. Only a handful of cases are pristine, and they are the ones that tell us about the underlying conditions rather than about borrowing. That the same broad package (dense settlement, monumental construction, social hierarchy, centralized authority, and usually some system of record-keeping) appeared independently in the Old World and the New suggests that something about agricultural societies, once they reach a certain density and surplus, tends to push them in this direction. It was not inevitable, but the repetition across continents is one of the most important facts in human prehistory.

The Archaeologist Who Named the Revolution

The person who named this transition was an Australian archaeologist named V. Gordon Childe. In his 1936 book Man Makes Himself, Childe coined the phrase urban revolution to describe the move from village to city, deliberately echoing the Industrial Revolution to signal a change of comparable magnitude in how human beings organized themselves. In a short 1950 essay in the journal Town Planning Review, also titled "The Urban Revolution," he laid out ten diagnostic criteria for deciding whether a settlement counted as a city.

Childe's ten points became the standard checklist of early urbanism: settlements larger and denser than any before, full-time specialists who did not grow their own food (craftsmen, priests, officials), the concentration of an agricultural surplus, monumental public buildings, a ruling class exempt from manual labor, systems of writing and recording, the beginnings of exact sciences such as arithmetic and astronomy, sophisticated art, long-distance trade in raw materials, and a political organization based on residence in a territory rather than on kinship. The essay's central claim is easy to state: the urban revolution did not simply produce a bigger village, but something categorically new, a different kind of settlement dependent on a fundamentally different organization of human labor. That insistence on a qualitative break, not merely a quantitative one, is what makes the essay a founding document of the field.

From Monuments to Landscapes

For a long time the archaeology of early cities was essentially the archaeology of their grandest buildings: dig the temple, clear the palace, photograph the ziggurat, and call it a day. The figure who broadened the lens was Robert McCormick Adams of the University of Chicago, who in The Evolution of Urban Society (1966) and especially in Heartland of Cities (1981) systematized the surface survey of Mesopotamian landscapes, walking the plains and recording the scatter of potsherds that marked where people had once lived rather than excavating a single spectacular site.

The shift in method produced a shift in understanding. Adams showed that a city was never isolated; it sat at the center of a web of towns and villages, all strung along the irrigation channels that made farming possible on the dry plain. Settlements clustered where the water ran, and when the rivers changed course over centuries, as the unstable Mesopotamian channels frequently did, whole constellations of communities shifted with them. Adams thus turned the origin of cities from a story about kings and temples into a story about ecology, agriculture, and the moving water of the floodplain, with the city as the densest knot in a much larger fabric of rural life.

Writing, Temples, and the Machinery of Order

One invention appears in the record almost exactly when the cities do, and it is central to how those cities worked: writing. In Mesopotamia, proto-cuneiform, wedge-shaped marks pressed into clay, emerged around 3200 BCE, and in Egypt, hieroglyphs appeared around 3100 BCE, very nearly simultaneous with the urban transition itself. The timing is no coincidence, because the earliest written records are overwhelmingly administrative: lists of grain, counts of livestock, allocations of labor and rations. Writing was, in its origins, a tool for managing the surplus and the people of a complex society, and only later became a medium for literature and law. The full alphabet came much later: the first true one, simplifying Egyptian signs down to roughly twenty-two consonant letters, appeared in the Sinai and the Levant around 1700 BCE and is the distant ancestor of nearly every alphabet in use today.

The cities also had a recognizable institutional skeleton, organized around four anchors. There was the ziggurat, the tiered temple platform that rose in stepped terraces above the plain and dominated the skyline for miles. There was the temple itself, which functioned as the central economic redistributor, taking in produce and paying out rations, less a place of private prayer than the hub of the economy. There was the palace, the emerging residence of a royal house whose power grew alongside, and sometimes against, the temple's. And there was the scribal school, where the administrators who ran all of this were trained to read and write the hundreds of cuneiform signs. Temple, palace, ziggurat, and scribe together formed the operating system of the early city, the machinery through which a settlement of tens of thousands could be fed, taxed, and governed.

Why a City Is Not the Same as a State

It is easy, and common, to use the words "city" and "state" as if they meant the same thing, but they do not, and keeping them apart is one of the most useful distinctions in the whole subject. A city is a settlement pattern: a dense, mixed-class population living together in a compact space, with specialists and a built environment no village possesses. A state, by contrast, is a political institution: a centralized authority that holds coercive power, the ability to compel and punish, over a defined territory and the people in it. The first describes how people are arranged on the ground, the second how power is organized over them.

The two usually go together, which is why we conflate them, but they are logically separable, and the archaeological record contains a case that proves it. The cities of the Indus Valley civilization, flourishing in what is now Pakistan and northwestern India in the third millennium BCE, were genuinely urban, with large planned settlements, gridded streets, standardized baked bricks, sophisticated drainage, and dense, differentiated populations. Yet they show remarkably little sign of the things we associate with a centralized state: no grand royal palaces, no monumental tombs glorifying individual rulers, no obvious standing army, no iconography of kings dominating subjects. The Indus thus stands as the canonical example of cities that may have existed without a state in the strong sense, organized through some other arrangement of authority that we still do not fully understand.

A Lineage That Never Quite Broke

The Sumerians had their own memory of where it all began, and they wrote it down. A document known as the Sumerian King List, copied across many tablets from around 2100 BCE onward, opens when kingship "descended from heaven" and then traces the institution through a sequence of cities that held it in turn, with Uruk holding it for centuries. The list mixes mythic reign-lengths of thousands of years with later, plausibly historical kings, so it is not a reliable chronicle, but as a cultural artifact it is precious, because it shows a people consciously remembering that kingship and the city had origins, that the political order they lived under had begun at a particular time.

What began at Uruk around 3500 BCE has, with countless interruptions, collapses, and local reinventions, continued essentially without a final break to the present day. The institutional package Childe described in 1950, dense population, full-time specialists, concentrated surplus, monumental construction, record-keeping, and territorial political authority, is recognizably the same one that organizes Lagos, Mumbai, Shanghai, and São Paulo right now. These megacities differ from Uruk in almost every detail of technology and scale, and yet the underlying form, a crowd of strangers fed by a hinterland and governed by specialized institutions, is the one worked out on a Mesopotamian floodplain more than five thousand years ago. When Woolley uncovered Puabi's tomb, he was not looking at a curiosity from a vanished world, but at the early chapters of the one we still live in.

Key Takeaways

The first true city was Uruk, founded around 3500 BCE on the lower Euphrates and holding forty to eighty thousand people by about 2900 BCE, larger than Athens at the height of Plato's Academy; the urban revolution that produced it was named by V. Gordon Childe, whose 1950 essay set out ten diagnostic criteria for what counts as a city and argued it was not a bigger village but a categorically new settlement dependent on a new organization of labor. Cities and states arose independently in at least six pristine cradles (Mesopotamia, Egypt, the Indus Valley, China's Yellow River, Mesoamerica, and the Andes), and writing appeared almost simultaneously, as proto-cuneiform around 3200 BCE and Egyptian hieroglyphs around 3100 BCE. Robert Adams's surveys, above all Heartland of Cities (1981), shifted the field from monuments to landscapes, while the Mesopotamian city ran on four institutions: ziggurat, temple, palace, and scribal school. Crucially, a city (a dense, mixed-class settlement) is not the same as a state (a centralized coercive authority over territory), and the Indus Valley, urban but apparently lacking a state, shows the two were separate inventions; the urban form first worked out at Uruk now houses more than half of humanity.