I’m a pretty big city-enjoyer, as some people might have noticed, what with living in a pretty large one and having been born in an even bigger one. Cities aren’t just great for living in either. They’re also a fantastic source of plots, characters and story hooks for writers. The city – especially the big city – is a stock setting for a wide variety of stories for a reason. They provide the people, places, and social dynamics for basically any kind of story you could want, so it’s no wonder that so many worldbuilders and creators fill their settings with big cities to set their stories in.
Unfortunately, this leads us to something of a problem. When a creator wants to set up a big city in their setting, they generally want to make it as big as they can, firstly because big population numbers make for an impressive figure on the page, and secondly because the bigger a city is, the more space it has for new stories to be told, new dark corners to explore, and new neighbourhoods to go through. The thing is, our current understanding of cities comes from what’s been sort of a historical outlier. Our heavily industrialised and mechanised societies are what allow for the massive multi-million people megacities which dominate our world today, and a setting which is firmly set in (for example) a pseudo-medieval fantasy setting will by definition lack a lot of the factors which make cities the size of our modern ones possible. As a result, a lot of worldbuilders working in that kind of setting massively overestimate how big their cities are ‘allowed’ to be, which leads to what seems less like an authentic pre-industrial setting and something more like the modern world, with technology replaced by magic.
If that’s the kind of thing you’re going for, then that’s fine – but that means I expect to see how the magic in your setting makes agriculture and transport vastly more efficient than it was in our own pre-industrial societies, or else I’m going to assume you haven’t done your homework.
So, let’s get started.
First of all, the question “How big is my fictional city?” is actually two questions: “How big could my fictional city be?” and “How big should my fictional city be”. The first of those questions is a lot easier to answer than the second, because ultimately that question is about one particular factor: carrying capacity.
The “carrying capacity” of a particular piece of land is the amount of people it can sustainably support over a long period of time. This doesn’t just mean the fertility of the soil, the availability of rain, or the presence of harsh terrain and weather (though it does mean all of those things), but also the presence of people able and willing to farm the land, and the tools with which they can do it with. This is because while fertile land can certainly sustain sparsely populated farming villages, it’s the use of machinery, fertiliser, irrigation, and other infrastructure which allow for more efficient agriculture, to the point where farmers aren’t just able to grow enough food to feed themselves, but also enough surplus to trade for the services of specialty workers. At the lowest level of surplus, this means supporting itinerant specialists who have to travel from farming settlement to farming settlement, because the surpluses these farmers support isn’t enough to sustain a specialist permanently. At higher levels, we start to see those specialists set up shop in these settlements, which is where you get specialty village craftsmen like carpenters or blacksmiths. Finally, at high enough levels of surplus, enough of those specialists gather in one place to make it a centre of commerce, where farmers from the entire region are able to bring their surpluses to sustain such a wide array of specialists, that other specialists which rely on the work of those specialists can also set up shop.
And then you have the beginnings of a city.
However, it’s important to bear in mind that without mechanisation, even the biggest surpluses are still going to only support a small fraction of the total population. A good rule of thumb is that before mechanised agriculture, we’re looking at ten farmers supporting one town or city-dweller. If the land is really fertile, or if the first stages of mechanisation (things like mechanical reapers or threshing machines) are in place, then that factor may be more like five to one instead – still a far cry from the one to three or four factors that we generally see in most developed countries today.
There are, of course, ways to extract more food than what is strictly surplus – usually by confiscating food from farmers under threat of force. This is, however, a short-term solution, mostly because if you steal the food that farmers need to live, you’re going to start losing farmers (and thus, agricultural capacity) pretty quickly.
Ultimately what this means is that you can figure out a city’s maximum potential sustainable population by figuring out the carrying capacity of all the regions that feed it. This usually means the city’s immediate hinterland, but in some cases, it could also include external territories with very efficient transport links (Classical Rome relied heavily on Egyptian grain, brought to the city’s port at Ostia by vast flotillas of mediterranean cargo ships). The important thing is that all these regions had the necessary infrastructure to carry that produce to the city economically and quickly enough to avoid spoiling. Once you have that total carrying capacity, you need to have at least a vague idea of how developed the agricultural system is in those regions. If farmers are digging with their bare hands on sandy clay, they’re going to have a lot less in the way of a surplus than centralised agriculture with seed drills and oxen-drawn reapers working chernozem or terra preta.
Once you’ve worked out that surplus, then you have the maximum sustainable population of your city.
Of course, that doesn’t mean your city will be able to comfortably feed all of those people. Some of that surplus will be going to smaller towns and market villages which will support services closer to the farms themselves. Other parts of that surplus will be wasted – either in the case of economic inequality (as the rich eat and hoard more than they need) or commerce, as the city’s infrastructure network also allows it to export goods elsewhere. However, these values can be tweaked and ballparked, they don’t provide anywhere near as much of a hard limit as agricultural carrying capacity does.
That brings us to the second question: how big should your city be?
The answer to this question has two factors leading into it: economics, and politics. The economic factor is simple enough: people go where there’s work and where they can make money. If your city has the easiest way of doing that for itinerant and otherwise moveable populations, those populations will gravitate to that city in particular. If a particular city has an industry which makes a lot of money and employs a lot of people, then a lot of people will show up to work there. If it doesn’t, then they’ll go to another city, or stay home. This is ultimately what determined which cities grew exponentially during the Industrial Revolution and which ones didn’t: the cities with easy access to raw materials, power sources, transit networks, and enough space to make all those things work together became industrial centres. Likewise, when railroads were being built across the US and Canada, cities competed with each other to try and get those transcontinental lines to stop near them, since that meant their businesses and industries would be linked to a transit network which let them transport goods in bulk long distances for cheap.
This brings us to the political factor. While there are some economic constants which public policy can’t change, there are some which it can. At the local level, figures with direct authority over things like taxes and commercial incentives can tip the scales in a way which makes certain industries more profitable than they’d be otherwise – usually as a means of attracting those industries and the jobs and tax money they bring. At the level of a state, there’s the possibility of a central government seeing the potential for some kind of growth or advantage in a particular location before it becomes self-evident enough to attract private investment – or because the benefit serves interests of national defence or policy rather than profit: The Russian imperial capital of St. Petersburg was built to support the empire’s policy of westernisation. Brasilia and Washington DC were built to create a central administrative capital free from existing regional rivalries. Halifax was founded as a fortress by the British; while Siberia is filled with cities which were not built because of industrial advantage, but because the Soviet government intended to build those advantages out of raw potential.
There’s also the effect of international politics, which is to say, the fortunes of the country which that city represents. The capital of a big powerful empire is likely to be a big powerful city, not just because empires tend to need a heavily populated metropole to expand and become empires in the first place, but also because political power – and the wealth of the political elite – draw in people wanting to profit off that wealth and influence. An imperial court needs guards, dressmakers, tailors, shoemakers, architects, gardeners, footservants, and all kinds of other people who can make a living providing the members of that court with all the luxuries and fineries which they expect. Likewise, those people need to be fed, clothed, and housed. That’s not even counting any kind of administrative civil service which might actually have to run the empire from the capital as well.
Of course, this works the other way too. Civil wars and imperial decline hit cities hard. When the countryside’s full of rampaging armies, it becomes a lot harder to get food and goods from the countryside to the city or from cities to each other. The intricate networks of infrastructure and transportation which cities rely on to exist fall apart, meaning that a lot of people in them have to leave, or starve – because the food’s no longer coming in reliably from the countryside, and the industries which used pay everyone no longer have reliable trade routes to export their products.
There’s a reason why Constantinople was the largest city in Europe when the Eastern Roman Empire was at its height, only to decline as the Empire did, and then surge in population again when it was taken over by the even bigger, wealthier, more powerful Ottoman Empire.
As always, these are more a process than hard and fast numbers. The actual maximum size of the cities in your setting depends on the precise level of technology, development, and any other sort of supernatural factors which you might want to include – which means you can use this process to tailor your numbers to your setting. In theory, this means you should be able to come up with some pretty good numbers, assuming your fundamental assumptions about total population and agricultural output are solid.
Paul Wang 