How many buildings in a medieval city?

Random thought

With a thousand year old city, it might be fun to have a "Ghost Quarter" of ruins that are rumored to be haunted, where no one goes . . . except Rogues making deals by night, adventurers, etc.
 

log in or register to remove this ad

howandwhy99

Adventurer
I believe old cities became cramped because they took over the role of castle in defense of the local population. We ran to the "leader" because he had a stone tower and we didn't. (And boy did he ever Lord it over us). ;p

Cities had armies camped in them and/or walls to protect them. Maybe just the inner quarters, but something defensible when the barbarians came to slaughter and steal in force.

A large city with a small population may have trouble defending its walls. It may only defend interior walls and leave others open. Outer quarters may be less populated because they aren't as safe. But they are cheaper to live in too, larger living conditions, etc. So many take the risk. It's not like living out in the surrounding farms who high tail it when siege armies arrive too.
 
Last edited:

Yaarel

He Mage
I find it helpful to refer to medieval ‘cities’ as ‘towns’ to keep their small size clear.

On average, the diameter of the townwalls are about a kilometer across. In other words, about 10 city blocks.

The streets are often like a web of main streets with labyrinthine alleyways between them.

Often there are unprotected parts of the town outside the townwalls. During a seige these residents will flee into the townwalls.
 
Last edited:

gamerprinter

Mapper/Publisher
When I created the original hand-drawn map of the city of Kasai as a commission for the Paizo Publishing Jade Regent Adventure Path, The Empty Throne module, I didn't know what the intended population was supposed to be - Paizo didn't tell me. It turned out to be double what I created, but I created a population close to what you say your city might have. I created Kasai to have about 40,000 population. While there would certainly be multi-storied, larger buildings to contain many family units, as well as crowded brothels and less standard housing for particular subcultures, I guess that most buildings were single family units with an average of 5 individuals per household - of course this covers a single inhabitant or an extended family of 10+. So at 5 people per smaller home, I created roughly 8,500 hand-drawn buildings. I also included a more densely populated ghetto area on the north and northwest of the main city.

If you decide that the average per household is higher, and that there are more multi-family structures, then your number of buildings should be greatly reduced. In a city with a smaller population than previous, that might allow a smaller number of people per structure, but also requiring you to create a greater number of largely empty buildings. If you're creating a map for yourself, I don't envy you, it's a lot of work. Kasai took me 16 hours to draw in four 4 hour sessions, at a high rate of speed, and I'm a pro. It will probably take you longer.

The northeast quarter of the city has thicker lines, as I had to use a different pen to draw it in. It was several weeks wait to finish this last quarter as Neil Spicer had yet to send me his version of the palace of the Jade Regent so I could match its' design with my drawing.

kasai-hand.jpg
 

Mercurius

Legend
What a gorgeous map, Michael - and kudos to you for spending 16 hours on it. I'm glad you put so much thought into the ratio of population to structures, because I think many fantasy city maps aren't drawn with this in mind. Your city of 40,000 has far more buildings than many cities I've seen done for D&D (mainly Forgotten Realms) with higher populations.

This is also useful for me as I'm currently working on two settlements, one of 1,000 and one of 18,000. I can look at your map and figure that the latter would have a bit less than half as many buildings. The former is harder to say, as it is a small town with a different density.

But looking at Magical Medieval Europe, it says that a small down has roughly one building per two people. That seems like a very low density to me, but I figure that it includes buildings that aren't inhabited, small sheds etc. Still, it is hard to imagine a small town of 1,000 with 500 buildings. I might go for more like 3-400.
 

Hussar

Legend
I suppose it depends on what you consider a building really. Is an outhouse a building? Chicken coop? Tool shed? I dunno. But, that might account for the ratio.
 

Celebrim

Legend
But looking at Magical Medieval Europe, it says that a small down has roughly one building per two people. That seems like a very low density to me, but I figure that it includes buildings that aren't inhabited, small sheds etc. Still, it is hard to imagine a small town of 1,000 with 500 buildings. I might go for more like 3-400.

I'm normally quite the fan of Magical Medieval Europe, but that figure seems way off. Medieval buildings were very heavily occupied by modern standards. Eight to twelve people in a very small space would not at all be unusual, and generally domestic residences doubled as shops or 'factories'. The master might sleep upstairs, and the apprentices and servants in the work area, the cook by the fire in the kitchen and the scullion on the floor. Heavy occupation makes all sorts of sense for an economy barely above the subsistence level. You have reduced heating needs, reduced building costs, and so forth. And since no one works after dark because most people can't afford lighting, and during the day everyone not involved in domestic labor is out in the fields, you don't need a lot of space. If you have a barn, the stablehands and grooms sleep in it.

So I generally use 8-12 people per building for cities, particularly because D&D buildings tend to be rather large and modern. In a city, I might well have something like 40 people to the building, on the assumption that they are large apartments similar to roman villas or modern town homes. Really, the density in cities is probably greater than 1 person per 100 square feet of floor space.

In terms of land usage, medieval cities could hit 50,000 persons per square mile, and 30,000 wouldn't be unusual.
 

Mercurius

Legend
Yeah, it seems a bit off, @Celebrim (although in truth it ranges from 1 building per 2-3 people, depending upon the size of the settlement). But let's just think about the average 10 buildings.

1 public building - 0 inhabitants (e.g. a temple, although possibly a groundskeeper would live there, but let's assume they don't)
1 warehouse - 0 inhabitants
1 shop - 0-3+ inhabitants (the owner and family might live upstairs, but might not)
1 tavern - 0-3+ actual inhabitants, although again in some situations servants might live there
1 wealthy residence - 1-12+ (depending upon size of family, servants, guards, etc)
2 merchant/guild class - 1-10+ (as with wealthy residence)
3 laborer homes - 1-20+ (could be anything from a shack with one person to an apartment type building with several families)

So that's ten buildings, a total of 3-45 inhabitants - really a broad range. Now if we include structures of all kinds, sheds and outhouses, then I could see how the 1 structure per 2-3 people would actually make sense. The point being, perhaps only about a half of all structures are actually inhabited - maybe a bit more, but you've probably got a sizeable number of buildings that aren't inhabited.
 

Celebrim

Legend
Yeah, it seems a bit off, @Celebrim (although in truth it ranges from 1 building per 2-3 people, depending upon the size of the settlement). But let's just think about the average 10 buildings.

1 public building - 0 inhabitants (e.g. a temple, although possibly a groundskeeper would live there, but let's assume they don't)
1 warehouse - 0 inhabitants
1 shop - 0-3+ inhabitants (the owner and family might live upstairs, but might not)
1 tavern - 0-3+ actual inhabitants, although again in some situations servants might live there
1 wealthy residence - 1-12+ (depending upon size of family, servants, guards, etc)
2 merchant/guild class - 1-10+ (as with wealthy residence)
3 laborer homes - 1-20+ (could be anything from a shack with one person to an apartment type building with several families)

So that's ten buildings, a total of 3-45 inhabitants - really a broad range. Now if we include structures of all kinds, sheds and outhouses, then I could see how the 1 structure per 2-3 people would actually make sense. The point being, perhaps only about a half of all structures are actually inhabited - maybe a bit more, but you've probably got a sizeable number of buildings that aren't inhabited.

The big problem with that list is that the proportions are all wrong. Public buildings probably constitute 1% or less of the buildings in a town or city. Warehouses again probably constitute less than 1% of buildings - most town won't even have one. You'll probably only see them in ports that receive goods in large quantities. Taverns probably constitute less than 2% of all buildings, but when they exist they are someone's home. Actual period inns and taverns look more like what we think of as bed and breakfasts or boarding houses than hotels and restaurants. In fact, very commonly they'd be referred to as "public houses" or "pubs". So real population here could be like 1-20+. Shops are 90% of the time also households, so uninhabited shops are a small percentage of buildings and not 25% of the total number of shops, and the maximum occupation ought to be the same as standard domiciles. In fact, if it is a shop keeper, then you are in the educated/skilled class and already in the upper 10% of income (particularly because of guild monopolization). That is a wealthy residence!

So real numbers might look something like.

1 Public Building - 0(?) inhabitants
1 Warehouse - 0(?) inhabitants
1 Unoccupied Mill, Craft or Guild Hall - 0(?) inhabitants
1 Inn (~6 inhabitants)
1 Wealthy Residence (~16 inhabitants)
10 Wealthier Merchant Homes/Shops (~8 inhabitants)
90 Laborers Homes (~12 inhabitants, usually an extended family covering 3-4 generations, working as teamsters, day servants, crafters assistants, farmers, wood cutters, green sellers, spinsters, etc. generally with multiple incomes per household, including older children)

105 buildings = 1182 persons > 9 inhabitants per building.

To get that number down you have to note this is primarily an agrarian economy, so there are barns, dove coots, smoke houses, out houses, spring houses, summer kitchens, carriages houses, lumber rooms, and sheds of various sorts with average occupation below 1 and probably near enough to zero not to matter. (Though can I imagine servants with less desirable jobs, goose girls and swine herds and so forth, are often using them as bed chambers particularly when they can sleep with the animals for warmth.) Or you might have the numbers slightly lower per household. Still, I can't imagine those numbers tripling or even doubling the number of buildings in a town, as ownership of those things indicates advancing wealth and status and the vast majority of those buildings are going to be outside the town proper in the more intensive farming areas. Additional buildings were usually noted in the Doomsday book (because they were valuable for tax purposes). Not even every rural farm had them, much less within the crowded area of a walled city.

And to the extent that we could get those numbers down, we could also go the other way and get numbers up. That public building could be a barracks, with 40+ inhabitants, or a temple housing a high status priest (a 'bishop' type) with his various servants (a cook and a groundskeeper, certainly), his dependents (an unmarried sister, his young nephew), a bell ringer, various lesser priests (deacons, lectors, cantors, acolytes, etc.), or a temple might house 24 vestal virgins, or a temple might house 40 temple prostitutes, 20 eunuchs, and various priests/priestess and their servants. A public building might be the mayor's home.
 
Last edited:

Mercurius

Legend
Thanks @Celebrim, that's quite helpful. Adding it all together, I think a range of something like 8-12 people per building, or roughly 10, works. If we want to include all possible structures, we could take half that. If it isn't perfectly accurate, it is close enough and manageable. This fits in with @gamerprinter's map, although he's clearly not including outhouses and chicken-shacks in that map! By your estimation, his map would better fit a city of 80-100,000.
 

Remove ads

Top