[Originally posted at SF Novelists.]
A while back, a friend of mine was running a roleplaying game set in the Pathfinder universe. The game took place in the city of Korvosa, which — according to the official writeup — is a substantial place. It is the royal seat, and its seven districts support three different military groups (the city guards, the Hellknights, and the Sable Company of hippogriff riders), a thieves’ guild, a major gambling enterprise, an even more major necropolis, a fighting academy, a mage academy, a mundane university, and a thriving sea trade . . .
. . . all with a population less than one-seventh the size of Elizabethan London.
As I said to my friend, “Add another zero to that number, and I might believe it.”
Korvosa’s population is 28,000 people. In 1600, London had approximately 200,000 — and it didn’t support all the things Korvosa is supposed to have (though it did have other things that tend to get ignored by game writers). The discrepancy between the two makes me laugh like a drain.
I think there’s two things at odds here. One is that everybody knows a Fantasy City ™ must have a bunch of things that make for fun plots, like thieves’ guilds and city guards and all the other trappings of your Dungeons & Dragons style world. The other is that everybody knows that cities in the past were way smaller than they are today.
Or were they?
I’m not going to pretend there was anything five hundred years ago that resembled modern-day New York City (population eight million and then some) or Tokyo (thirteen million). But Elizabethan London had two hundred thousand. Tenochtitlan was roughly the same, as was Moscow. Beijing had a population of four hundred thousand in the early thirteenth century; this climbed to nearly a million in the mid-fourteenth. These would be entirely respectable cities in the modern United States, albeit no competition for the top ten. But they run against our default assumptions about the past, and so we end up with this: the weird contradiction of a Fantasy City that has all the plot-generating amenities, but none of the population to support it. (Even Waterdeep, that vast metropolis of the Forgotten Realms, apparently holds a hundred and thirty thousand people.)
This wouldn’t bug me half so much if the writers didn’t try to put numbers on these things. Another roleplaying game, Legend of the Five Rings, models its setting on historical Japan, but on a geographic scale that dwarfs that island chain. I’m willing to spot them the notion that the Lion Clan fielded an army of half a million men during the Clan War; my admittedly patchy knowledge of Chinese military history says that if you’ve got a big enough country and a well-organized bureaucracy to administer it, you can hit those kinds of numbers. But L5R is written by freelancers, and sometimes things fall into the cracks between them: the core rulebook for the fourth edition claims that some of the Minor Clans consist only of “a couple dozen samurai” (and their associated peasants). That isn’t a clan, at least not in L5R terms; it’s barely an extended family. My mother has nearly three dozen cousins all on her own. The descriptions given for the nature of a Minor Clan don’t jibe at all with the notion of a few dozen people — more like a few hundred.
The thing is, I actually like it when settings give me concrete detail. Few novels are going to bring up actual population numbers, of course — that’s more the kind of thing I expect from a role-playing game, where the text discusses practical matters in far more direct, out-of-character terms. But the principle applies to other things, too. I like the concrete detail . . . but the risk of giving it to me is, I might spot when it’s implausible. So that means you have to do your research, which is a lot of work, and maybe it’s easier just to skip the math and give me the fluff.
Otherwise, I may grumble at you to add another zero.