So I'm working on building a city in my world and I've run into an issue with scale/size. I have the general idea of how large the city will be (~8 km diameter) and I am having trouble coming up with a number of residents.

As for technology and resources, they have indoor plumbing, cobblestone roads, with most buildings being primarily stone. This city is a thriving center for trade and has abundant resources. Magic used in construction is possible, though limited. They do have access to glass and steel, but using large quantities is very expensive. And their knowledge of architecture is much like our own.

So, with these resources, how many people could comfortably live in a defined 8km space?

  • 1
    $\begingroup$ There's a few important questions which need answered. I presume initially you're thinking of something like a medieval city, maybe like the old cities of Europe? But if magic can be used in construction, does that mean skyscrapers (or bunker cities) are possible? Either would throw out the limits one would expect from the aforementioned scenario. $\endgroup$ – inappropriateCode Aug 2 '16 at 15:29
  • 1
    $\begingroup$ Magic in this world is limited. So only very important buildings will incorporate magic into the design. And even then it will be as little as possible. Magic in this world is seen like a resource. Only the incredibly rich or powerful use magic as a convenience. Some of the infrastructure might use magic (guild buildings, palace, roads etc) but it is not generally used for residences. $\endgroup$ – Killzone Aug 2 '16 at 15:39
  • $\begingroup$ Not a real answer, but maybe you might find value in this tool: donjon.bin.sh/fantasy/demographics The numbers seem more or less reasonable. $\endgroup$ – Nex Terren Aug 2 '16 at 15:41
  • $\begingroup$ You said magic roads. Magic roads can make a huge, huge difference. As an example, a "road" network enchanted with 10x normal speed and magic sanitation would approximate a street-car era city size. (Ideally with a similar "highway" or "railway" network to get food into the city) Weather/crop growth magic (en mass) could increase farm yields. Something else to help with the labor-heavy process of harvesting food gets you most of the way to industrial era populations (one interesting question: what to do with all that surplus farm labor? Have them etch blood runes to fuel the roads?) $\endgroup$ – Yakk Aug 3 '16 at 2:39

Your technology level seems to be roughly 1700s-ish...I'm not sure if that is what you were going for but thats the impression I got from your question.

Your city is 8km across, so for a circle with radius 4km, you get 50km^2 area. For a 8km edge square you get 64 km^2. Here are your best comparisons that I could find good data on:

Asian Cities

Vijayanagar in South India had city walls enclosing 25km^2, and outer walls enclosing 650km^2 which included farmland, gardens, and suburban residences for the rich. It had a population of 500,000 in 1500AD, most of that probably within the inner walls.

Beijing's old city walls came in two sets, the inner city and outer city. These were not one inside the other, but two city walls right next to each other, about 2km apart. Together they enclosed an area of about 50 km^2, almost a perfect 8x8 km square except for the gap between the two sets of walls. Both walls were completed by 1550. Beijing's population from then until 1800 ranged from 700,000 to 1.1 million, though by that time many people probably lived outside the walls.

The density of these two cities would have been about in the ballpark of 20,000/km^2, so that is probably a good benchmark for your city.

Middle Eastern Cities

These cities tended to by hyper-dense, and remain so today. Cairo is one of the densest cities in the world, and pre-civil war Aleppo and Damascus were as well.

Istanbul didn't grow beyond the Theodosian Walls until the 1800s, except for Galata, across the Golden Horn. I measured on a map and estimate the area to be about 10 km^2. Istanbul's population from the 1550s to 1800 was probably in the 500,000-700,000 range for a density of 50,000-70,000/km^2.

I couldn't get good are data for Cairo, but I think its density was in the 50,000km^2 range also at its peak in the 1300s-1400s. Its footprint was pretty small. Cairo had 'skyscrapers' like the Roman insulae, and may have been built up vertically more than any other city before the 1800s.

European Cities

European cities only had desely populated city cores (of 2-5 km^2) and had more extensive suburbs than China and South Asia. Its tough to find a European city as big as yours with a reliable census figure.

Naples, Italy was 484,000 people at the census of 1861 in an area of 117 km^2.

The central bits of London (defined here as the City of London, plus Finsbury, Shoreditch, Bethnal Green, Stepney, Holborn, Southwark, and Bermondsey) has an area of 29 km^2 and population in 1801 census of 401,000. Greater London had 959,000 people in an area of 5223 km^2.

Paris's administrative limits cover 106 km^2. Its population ranged from 420,000 in the early 1600s, to about 630,000 right before the french revolution.

So for Europe, Naples and Paris's density was in the range of 5,000/km^2 for a larger area, while London's inner core had a density about 15,000/km^2, or a little less than the Asian cities. If you want your capital to have more parkland and suburbs, or have a more European/Renaissance character, go with those numbers.

  • $\begingroup$ @kingledion Good point, I have removed my previous comment, and now that you have added the actual size, +1 $\endgroup$ – Kevin Wells Aug 2 '16 at 23:36

If the city is 8km wide, and we can assume it's roughly a circle, then it's approximately 50km² in area. Taking from this site we can approximate 14,942 people per square kilometer. This totals to 750,000 people.

I'm inclined to think this number is way too high for a couple of factors that scale up faster than population as you increase population:

  • Sanitation requirements
  • Transportation requirements
  • Resource logistics (transporting food and water to and from the city)
  • Civil management (law enforcement, taxes, etc)

Also a few factors scale more or less directly:

  • Food requirements
  • Water requirements

Further Paris during the (very late) medieval ages is estimated to be 250,000. I can't hunt down an acre/square km size for it during this period, but it was a larger city.

Demographics in detail are hard to come by, so these are just approximate numbers. Keep in mind however many people you place in the city, expect a 5-to-1 up to 10-1 ratio of nearby farmers to city citizens.

  • $\begingroup$ Look at descriptions of Paris City Walls for estimate of areas at different times. $\endgroup$ – user2448131 Aug 2 '16 at 17:29

A million people is about the maximum for a city before the invention of mechanised transport, since if it gets any bigger then people can't walk in from the suburbs on any realistic timescale. That's with population densities much higher than in most modern cities, and your 8km radius sounds about right. See Rome in 200AD, Baghdad in 900AD, Kaifeng in 1200AD or London in 1800AD. All around 1 million inhabitants. But they won't be comfortable, because of the need for a very high population density to squeeze everyone in close enough to the centre. If you want to increase that, you need better transportation technologies, not better building technologies.

See this article for some more details on the largest historical cities.

  • 1
    $\begingroup$ Do you have any citations for those figures? Any evidenced historical data would be great to help contextualise the answer. $\endgroup$ – inappropriateCode Aug 2 '16 at 15:32
  • 1
    $\begingroup$ 1800s London with a population of 1.1 million was actually Greater London, with an area of around 600sq miles (1550sq km). $\endgroup$ – TheBloodyPoet Aug 2 '16 at 16:22
  • $\begingroup$ @TheBloodyPoet Modern-day Greater London has an area of 600 square miles. Pre-railway London was very much smaller, and most of what is now the Greater London built-up area was still farmland and small villages. $\endgroup$ – Mike Scott Aug 2 '16 at 16:29
  • 1
    $\begingroup$ @inappropriateCode I've added a link citing my source for those population estimates. $\endgroup$ – Mike Scott Aug 2 '16 at 16:32
  • 1
    $\begingroup$ @inappropriateCode Wikipedia has your data! This list is VERY well cited (with over 300 footnotes!) and reference the best researchers like Tertius Chandler and George Mokeski. $\endgroup$ – kingledion Aug 2 '16 at 17:06

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.