The root word for civilization is the same as the root word for city. The rise and development of cities is inextricably linked in the archeological record with the rise and development of civilizations. What makes a city a city? And where do cities naturally take root and grow?
From ancient Çatalhöyük to modern Manhattan, most cities share certain features in common. Cities can be seen as a form of advanced social technology, but they also depend on a combination of physical technology and geographic resources in order to develop and grow. Before the rise of rail transportation, this was intimately linked with the geography of water.
What is a city?
A city is primarily defined by scale. In 1992, a British anthropologist named Robin Dunbar proposed that humans have a “natural” group size limit of about 150 individuals based on consideration of primate group sizes in general. Chimpanzees and bonobos, our closest living relatives, are known to form social groups of up to this size as well, although smaller groups are typical.
As a group size, Dunbar’s number corresponds roughly to the typical size of Neolithic farming villages, which began to appear around 12,000 years ago. It’s larger than the typical hunter-gatherer groups and the estimated typical size of bands of Neanderthals. Groups with sizes smaller than Dunbar’s number can function informally because everyone knows everyone else; groups larger than this need to rely on norms, rules, and laws.
Agriculture was required for permanent settlement. Jericho, one occasional contender for the title of oldest city, first experienced year-round human settlement around 11,000 years ago, developing fortifications around 10,000 years ago, with a peak population estimated at anywhere from 200 to 2,000 before being abandoned for the first time.1 Whether or not Jericho qualified as a city during this period as opposed to merely being a fortified village or not is a matter of debate.
In preindustrial periods, most of the population of most of the world lived in settlements no larger than Dunbar’s number multiplied by the average household size; in the absence of sophisticated social technologies, villages with populations larger than Dunbar’s number are the exception rather than the rule.
Living technology
The transition from paleolithic hunter-gatherers to neolithic villages was accompanied by the use of cultivation and domestication. In most cases, wild plants are highly inefficient crops, and wild animals are inefficient sources of food. Neolithic farming villages are believed to have their origins in campgrounds regularly visited by nomadic groups, with the cultivation and domestication of key plants being the single most important technological advancement allowing humans to stay in the same place year-round.
Farming villages may be self-sufficient, but a city will need to draw food in from a significant area surrounding it, which in turn means that the surrounding population must be able to generate a surplus of food. Throughout Eurasia, neolithic farming villages usually precede the development of cities in the same area by thousands of years for the very simple reason that it takes a significant amount of time for wild plants to evolve into high-yield crops capable of producing a surplus.
Çatalhöyük and the neolithic cities
The strongest known contender for the title of earliest city is Çatalhöyük, which was founded in the Anatolian peninsula (modern-day Turkey) roughly 9,000 years ago and abandoned after roughly 1,500 years of habitation. The population of Çatalhöyük peaked at roughly 10,000 and likely averaged on the order of 5,000 throughout its entire existence. Given the scale and density of the site, it is hard to describe it as anything other than a city.
Other than arguments based on physical technologies, the primary argument against Çatalhöyük being the first city is that it lacked the highly hierarchical social structures seen in other early cities.2 While the construction and organization of Çatalhöyük is physically sophisticated, the buildings are relatively uniform in size and style. Skeletal evidence suggests a relatively egalitarian distribution of food resources.
Importantly in the development of our understanding on how cities form, Çatalhöyük as a site predates metallurgy. One of the principal economic arguments for the rise of urbanization is that it supports increased specialization of labor. Çatalhöyük and other urban-scale Neolithic sites, such as Mehrgarh in the Indus Valley (also founded around 9,000 years ago).
With more precise dating and archaeological work, it has become very clear that the ancient Sumerian cities were also founded with Neolithic technology. Eridu, for example, was founded roughly 7,500 years ago and was inhabited for approximately 5,000 years. The Cucuteni-Trypilia culture in modern-day Ukraine shows large numbers of urban-scale settlements dating back to roughly 6,000 years ago.
There are also New World examples of cities built in the absence of metalworking technology, such as Cahokia (on the junction of the Mississippi and Missouri rivers) or Huaricanga (founded in Peru roughly 5,500 years ago, nearly 2,000 years before the earliest evidence of South American copper smelting).
In Egypt and China, evidence of urbanization comes much closer to the Bronze Age, with the evidence pointing to 5,500-6,000 years for Egyptian urbanization and around 4,000-4,500 years for Chinese urbanization.3 This might be due to a less complete archeological record (or a more thoroughly erased archaeological record).
Water is life
Both Çatalhöyük and Eridu were founded near rivers (respectively, the Çarşamba and the Euphrates). In both cases, the river changed course and the city was abandoned. Çatalhöyük and Eridu are not the only abandoned cities to be found next to dry riverbeds. This brings us to the next major point about cities.
All cities require a significant supply of fresh water. Stored rainwater and wells tapping into groundwater are rarely sufficient to meet that need. In the pre-industrial age, supporting an urban-scale population requires a highly reliable and highly continuous source of fresh clean water for drinking, as well as large volumes of fresh water for agriculture and sanitation.4
Unlike farming villages, cities also require trade and transportation networks to support their populations. Food must be brought into the city from outlying farms; cities cannot feed themselves, meaning that they usually include skilled specialists who produce goods for trade.5 Prior to the invention of railroads, water transport was the fastest and most efficient method of transporting goods, animals, and people.
Cities founded near river mouths or river junctions have especially broad access to trade; cities also tend to develop near rapids or falls that block an otherwise navigable river. Once a river gains its first city, it is likely to gain many more; this happened along the Tigris and Euphrates rivers first, and then the Nile. Rapid propagation of Mycenean urbanization likely happened via sea transport.
Third, as the ancient fortifications of Jericho suggest, most cities eventually face security concerns. Major bodies of water present significant strategic obstacles. An opposed army cannot easily cross a river or lake; even a narrow body of water is a defensive obstacle as effective as a city wall.
Pre-industrial cities almost all grew up on rivers and freshwater lakes. The cities that grew to be major centers of power have very often been cities located at junctions of rivers (such as Cahokia), where the rivers flow into the sea (such as Ur and later Athens) or in positions rendered highly defensible by lakes and rivers (such as Tenochtitlan and Paris).
Conversely, there are very few places in Eurasia where the geography of a site is naturally suitable for cities where cities did not develop organically on their own. The main exceptions to this pattern are places in the world where the technology of urbanization was unusually slow to develop, such as in North America.
Some estimates of both the relative continuity and early population of Jericho may be impacted by contemporary political and religious concerns. The general consensus appears to be that Jericho grew, shrank, and was likely entirely abandoned several times. With its relatively small supply of reliable fresh water, Jericho usually qualified as a village or town rather than a city.
See in particular Prateek Dasgupta’s four-part series on the rise of cities - the first installment offers Childe’s definition of a city, which includes truly long-range trade and writing. The second part discusses protocities.
Egypt is an unusual case in that hieroglyphs directly document the existence of major urban centers to circa 3200 B.C.E., but archaeological evidence has yet to push the date of urbanization back significantly from that point. Chengziya is one main candidate for a first Chinese neolithic city with an estimated founding date of 2600 BCE; large numbers of urban-scale settlements appear roughly 4,000 years ago along with local bronzeworking.
Agriculture may, in some cases, be adequately supported by less reliable sources of water, such as rainwater and seasonal flooding.
Or at least skilled specialists in the fine art of collecting taxes from outlying villages.