What's up with ancient cities?
This is a spinoff from my last post. It didn’t really end up fitting there, so I put it here instead.
I used to have a model in my head that was something like: more efficient agriculture → denser population → urbanization. I thought of cities as being the logical outcome of more efficient agricultural practice. But this clashes with history in weird ways. The ancient world seems to have been more urbanized than the medieval world, even though the medieval world was more sophisticated in agricultural technique.
What sort of density do you need to have a city? Let’s consider the geometry of it. Suppose a city of about 5,000 family units, which would be about 40,000 people (a big city in pre-modern terms). The rule of thumb is that a typical family unit needs about five acres of land to survive. So our city needs about 25,000 acres of land to support itself. Let’s further constrain ourselves and say we can only realistically cultivate about a third of the land, so we really need 75,000 acres. If we convert that to square miles, that comes out to about 117 sqmi. That’s equivalent to a circle with a radius of just a bit over six miles. That’s a distance you can walk in about two hours.
You only really need something like half your labor force to actually work in the fields. So a lot of people can spend all their time in the city. Another chunk of people can live in the city and walk out to the fields. Then others might sleep further out from the city to be closer to their fields, splitting their time in some way between the city and the hinterland.
I think this actually explains a lot about the ancient world that I used to find confusing. If your culture values living in a city, it’s not, strictly speaking, that hard to do it. It just means some people have to walk a bit further sometimes. Realistically, I think you still find lots of non-urban settlements in most societies, so it’s unusual for everyone to clump up in cities, but there’s still a lot of variation in how “clumpy” your population can be as it spreads across the cultivable countryside.
The above sort of “natural variation” doesn’t get you to ancient Rome of course, which is estimated to have had something like a million people at its peak. That sort of population concentration requires a robust seaborne logistics network that can funnel vast amounts of grain into a port. But it does explain how a place like Greece could be covered in little city-states. The Greeks valued living in cities, so they did it.
So why do people live in cities? Today, we often think in terms of amenities you get in a city. But my impression is that the very earliest cities were more about collective defense. You’re worried about being attacked, so you build some walls to protect everyone. It’s logical at that point to put your house inside the walls so you’re not worried about it. Lots of people end up living inside the walls, at which point you’ve invented the city. Once you have a city, you start to build other sorts of cultural institutions around the thing you’ve made, and eventually the walls are less important than all the cultural stuff (at least until someone attacks, and then the walls feel really important again).
The decline of urbanism in medieval Europe makes more sense to me when I think this way. It’s not that they forgot how to make cities work or something. They just didn’t value it the same way the ancient world did. And the security situation was different enough that the peasantry weren’t as desperate to get their house behind a city wall.