Sumer and the First Cities: What First City Really Means

There is no settlement that everyone agrees was the first city, and the reason is not missing evidence. It is that "city" is a definition, and the candidates win or lose depending on which definition is applied. Asked with a criterion attached — the oldest settlement with monumental public architecture and administrative record-keeping, say, or the oldest permanent settlement above a given population — the question has good answers. Asked without one, it has none, and any source that supplies a single name without saying what test it used has quietly made the choice for you.

What a city has to have

Archaeologists generally look for several things together rather than any one of them. Population well beyond a farming hamlet. Occupational specialisation, so that some people are not producing their own food. Public buildings that serve more than one household group. Administration of some kind, usually visible as seals, tokens, standardised containers or written records. And a hinterland — a surrounding countryside organised to feed the centre.

Each of these leaves different traces and each fails in different ways. Population is estimated from house counts and density, which requires assumptions. Specialisation is inferred from tools and workshop debris. Administration is the criterion that most favours Mesopotamia, because clay tablets survive where perishable recording systems do not. That last bias is worth holding on to, because it shapes the standard answer more than any argument does.

The Mesopotamian candidates

The southern Mesopotamian case is strong and well documented. UNESCO's Ahwar of Southern Iraq listing covers seven components — three archaeological sites and four wetland marsh areas — and describes the archaeological cities of Uruk and Ur and the Tell Eridu site as part of the remains of Sumerian cities and settlements that developed in southern Mesopotamia between the fourth and third millennia BCE, in the marshy delta of the Tigris and the Euphrates.

That formulation repays attention: cities and settlements, developing over a span of roughly a thousand years, in a wetland landscape. It describes a process rather than a founding. Nippur, another of the great Mesopotamian centres, shows the same depth. The Institute for the Study of Ancient Cultures at the University of Chicago, which began excavating there in 1948 and continues the project today, describes Nippur as a sacred centre of the god Enlil whose temples, neighbourhoods, public buildings, tablets and artefacts preserve more than 6,000 years of urban history. Six thousand years of continuous urban occupation is not a moment of invention; it is a landscape that was urban for longer than everything since has lasted.

You will often see Uruk given flatly as the world's first true city, including in reference material built for learners. The claim is defensible under a specific criterion — scale combined with monumental architecture and early administrative technology — and indefensible as a bare superlative. State the criterion, and it becomes an argument other people can engage with.

Çatalhöyük: large, dense, and arguably not a city

The clearest demonstration that the definition is doing the work sits in southern Anatolia. UNESCO describes Çatalhöyük as a 37-hectare site formed by two mounds, with the taller eastern mound containing eighteen levels of Neolithic occupation between 7400 BC and 6200 BC, including wall paintings, reliefs and sculptures, and the western mound carrying Chalcolithic occupation from 6200 BC to 5200 BC.

That is thousands of years older than Uruk, and the settlement was large and densely built. But the same description is careful: the site provides important evidence of the transition from settled villages to urban agglomeration, maintained in the same location for over 2,000 years, and it features a unique streetless settlement of houses clustered together and entered from above. A place with no streets, no obvious public monumental buildings and no administrative apparatus fails several standard criteria while passing the size test comfortably. Whether that makes it a very large village or a different kind of city is exactly the argument, and it is a real one.