Yet another Demography Thread

[DominusNovus]

I know there has been plenty of discussion about whether or not the demographics of Westeros (south of the Wall) make sense. Generally, opinions seem to be less than positive on that account. However, I was looking at some of the numbers, and I think they're not as bad as is sometimes argued.

First, I started with the size of Westeros, and found this analysis fairly convincing:

https://www.quora.com/How-large-are-the-seven-kingdoms-of-Westeros

https://www.quora.com/How-big-is-Westeros-compared-to-our-world/answer/Sunil-Kumar-Gopal

Which puts Westeros in at just about 9.3 million km^2, just a tad smaller than the United States. As for population, the most common consensus seems to be around 40 million, with a fairly wide margin of error. Interestingly enough, the US population in 1870, the first census after the Alaksa purchase (meaning the first census after the US reached something similar to its present total area) has the US population at 38.9 million, of whom, the overwhelming majority (38.2 million) lived in states, rather than territories.

So, overall, the US of 1870 was about as densely populated as the Seven Kingdoms when the books start. Now, obviously, the US was much more urbanized and technologically advanced. That said, it also included Alaska, which was (and is) far less densely populated, and really skews the US's numbers. And the majority population had settled there only in recent years. I think its reasonable for a Westeros that had been inhabited, more or less, by the same basic population (First Men, Andals, and Rhoynar, all coexisting mostly in similar political arrangements) for millennia, to have a stable population that is in the same range. 

[Seaserpent]

5 hours ago, DominusNovus said:

I know there has been plenty of discussion about whether or not the demographics of Westeros (south of the Wall) make sense. Generally, opinions seem to be less than positive on that account. However, I was looking at some of the numbers, and I think they're not as bad as is sometimes argued.

First, I started with the size of Westeros, and found this analysis fairly convincing:

https://www.quora.com/How-large-are-the-seven-kingdoms-of-Westeros

https://www.quora.com/How-big-is-Westeros-compared-to-our-world/answer/Sunil-Kumar-Gopal

Which puts Westeros in at just about 9.3 million km^2, just a tad smaller than the United States. As for population, the most common consensus seems to be around 40 million, with a fairly wide margin of error. Interestingly enough, the US population in 1870, the first census after the Alaksa purchase (meaning the first census after the US reached something similar to its present total area) has the US population at 38.9 million, of whom, the overwhelming majority (38.2 million) lived in states, rather than territories.

So, overall, the US of 1870 was about as densely populated as the Seven Kingdoms when the books start. Now, obviously, the US was much more urbanized and technologically advanced. That said, it also included Alaska, which was (and is) far less densely populated, and really skews the US's numbers. And the majority population had settled there only in recent years. I think its reasonable for a Westeros that had been inhabited, more or less, by the same basic population (First Men, Andals, and Rhoynar, all coexisting mostly in similar political arrangements) for millennia, to have a stable population that is in the same range. 

I think its not right to compare it to US. George had midevil europe in mind and even the expansion of the roman empire. So you have to compare KL to London and london has 500.000 between 1350 and 1500, so in midevil europe. The size of armies and everything adds up to a european origin. if you know how many people live in the cities, you can guess how many will live in the rural areas and the whole of westeros. I think 60 million people is a better guess. The reach has an army of 100.000 and that is around 1%-1,5% of total population. So the reach has between 7 and 10 million people. This also depend on a lot of other factors like pre or post-war times. 

[Vashon]

Neat, now account for the effects of the effd up weather patterns. That should about halv it 

[DominusNovus]

4 hours ago, Seaserpent said:

I think its not right to compare it to US. George had midevil europe in mind and even the expansion of the roman empire. So you have to compare KL to London and london has 500.000 between 1350 and 1500, so in midevil europe. The size of armies and everything adds up to a european origin. if you know how many people live in the cities, you can guess how many will live in the rural areas and the whole of westeros. I think 60 million people is a better guess. The reach has an army of 100.000 and that is around 1%-1,5% of total population. So the reach has between 7 and 10 million people. This also depend on a lot of other factors like pre or post-war times. 

The urbanization rates are totally screwy, I agree. Westeros is very unurbanized (possibly implausibly so). That said, I do think that the US is a fair comparison, even though it is an industrial society by 1870, because so much of that land was utterly untouched. There is no Alaska in Westeros, not even the North. Heck, in 1870, most of the West was barely settled (not to dismiss the American Indians, but their numbers were quite low). Do we have anywhere in Westeros that is regarded as barren? Pretty much Dorne, and thats it. I think that the US could support that same population at a lower technology level. After all, Europe, combined is only slightly larger than the US (10.1 million km^2), and in the middle ages, it tended to have a comparable population to Westeros. If you'll permit me using wikipedia (I know, I know, but all the other sources are in the same ballpark):

https://en.wikipedia.org/wiki/Medieval_demography

We see the population ranged from 56 million to 90 million between 1000 and 1500, per that source. Others vary, but not by a large degree.

 

59 minutes ago, Vashon said:

Neat, now account for the effects of the effd up weather patterns. That should about halv it 

I would say that that has more of an impact on the urbanization rate. The population is far more diffuse than would otherwise be likely. Which means that the population is more protected from the harsh winters - it is easier to feed a small village than a great city.

Its also quite possible that the population is far more volatile than our world. Actually, not just possible, almost certain. Winter probably kills off more, but summer allows for more bounty and more growth, and the simple fact that while the population is more likely to be trimmed during hard winters, the infrastructure isn't, not to mention that in the areas with the hardest winters, its the oldest that tend to die first ('gone hunting' and all that). I expect every spring and and early summer to be something of a mini-baby boom. Also likely is that the population will see something of a material improvement come each spring, as property gets freed up - think something like the Black Death in Europe (which Europe recovered from in a century).

My guess is that Westeros' population graph would zigzag quite a bit, but if you average it out over each seasonal cycle, you'd probably get a value not too far off from that common estimate of 40 million.

[Seaserpent]

21 hours ago, DominusNovus said:

The urbanization rates are totally screwy, I agree. Westeros is very unurbanized (possibly implausibly so). That said, I do think that the US is a fair comparison, even though it is an industrial society by 1870, because so much of that land was utterly untouched. There is no Alaska in Westeros, not even the North. Heck, in 1870, most of the West was barely settled (not to dismiss the American Indians, but their numbers were quite low). Do we have anywhere in Westeros that is regarded as barren? Pretty much Dorne, and thats it. I think that the US could support that same population at a lower technology level. After all, Europe, combined is only slightly larger than the US (10.1 million km^2), and in the middle ages, it tended to have a comparable population to Westeros. If you'll permit me using wikipedia (I know, I know, but all the other sources are in the same ballpark):

https://en.wikipedia.org/wiki/Medieval_demography

We see the population ranged from 56 million to 90 million between 1000 and 1500, per that source. Others vary, but not by a large degree.

 

I would say that that has more of an impact on the urbanization rate. The population is far more diffuse than would otherwise be likely. Which means that the population is more protected from the harsh winters - it is easier to feed a small village than a great city.

Its also quite possible that the population is far more volatile than our world. Actually, not just possible, almost certain. Winter probably kills off more, but summer allows for more bounty and more growth, and the simple fact that while the population is more likely to be trimmed during hard winters, the infrastructure isn't, not to mention that in the areas with the hardest winters, its the oldest that tend to die first ('gone hunting' and all that). I expect every spring and and early summer to be something of a mini-baby boom. Also likely is that the population will see something of a material improvement come each spring, as property gets freed up - think something like the Black Death in Europe (which Europe recovered from in a century).

My guess is that Westeros' population graph would zigzag quite a bit, but if you average it out over each seasonal cycle, you'd probably get a value not too far off from that common estimate of 40 million.

I agree with most what you are saying and the harsh climates and wars play a big part in these societies. I think between 294 AC and 300 AC could be a difference of 6 million over Westeros. 

[DominusNovus]

5 hours ago, Seaserpent said:

I agree with most what you are saying and the harsh climates and wars play a big part in these societies. I think between 294 AC and 300 AC could be a difference of 6 million over Westeros. 

That sounds like a reasonable number to throw out there, but I'd adjust your latter year a bit further out. Winter's just started, and despite how screwed everyone seems to be, it will take awhile for that to set in. Of course, if we look too far out, then you have the Others causing further... demographic fluctuations.

Well, depending on what you count wights as.

[Free Northman Reborn]

 Not really sure how a population recovers a Winter’s losses in 3-5 years of summer. I mean, even if you have a bunch of births for 4 years, all those babies are still toddlers by the time Winter hits again. So I can only assume the die off isn’t really significant unless you have a 5+ year Winter, which doesn’t happen very often.

[Seaserpent]

9 hours ago, Free Northman Reborn said:

 Not really sure how a population recovers a Winter’s losses in 3-5 years of summer. I mean, even if you have a bunch of births for 4 years, all those babies are still toddlers by the time Winter hits again. So I can only assume the die off isn’t really significant unless you have a 5+ year Winter, which doesn’t happen very often.

Interesting, but the weather plays a very big role in population, but there are also very long summers. it works both ways!

[Vashon]

On 6/5/2019 at 7:55 PM, DominusNovus said:

I would say that that has more of an impact on the urbanization rate. The population is far more diffuse than would otherwise be likely. Which means that the population is more protected from the harsh winters - it is easier to feed a small village than a great city.

 

This is the only part I'd agree with you on, and only because it would be cheaper for storage and transportation wise to have large, and cheap, storage in the immediate vicinity of consumption, with a quirky feature of Westeros being massive storage units even in villas of no more than a few hundred people. 

Everything else? No, a long winter of 20 months or longer is very good at killing indiscriminately, but a multi year long summer, say 5 years, does nothing to repair the damage. People take 15 years to make and be big enough for real labor, regardless of the propensity for absurdly young marriages more likely to kill a young wife than produce an heir in ASOIAF and its fanfics, but a winter could and would snuff out all that excess growth in an instant. And a Baby Boom/Baby Bust cycle isn't as relevant in a pre-modern context, even with what appears to be cheap and plentiful abortifacents. Which did sort of exist anyways

The birth rate would remain high and stay high, with most every hetero sexually active woman attempting to carry at least pregnancies to term in her life, even in the most economically constricted lands, outside of sexual slavery forcing abortions or severe damage to the relevant organs, and those that can popping out as many kids as won't outright kill them

 

Only in the North, Beyond the Wall, and the Iron Islands would there ever be shockingly unstable population numbers over a century, outside of plague. I'd even say that Planetos is already doing crop rotation and several of the other agricultural revolutions that are relevant without steam power. Because while everyone forgets this, mild winters, even harsh ones, serve a purpose, letting the ground lie fallow, kiling whole swathes of insects, replenishing ground water with snow melt, burying and breaking down fallen leaves into soil, it all acts as a giant, free fertilizer and free insecticide. 

The North and Far North would have ups and downs related to devastating wars, and overhunting/overfishing, less pronounced on the more southern shores but still present, and then upswings when the plentiful game on land and in the sea allows for increased trade and consumption. 

The Iron Isles directly correlated with internal political stability and external military success, basically how much resource extraction they can obtain from the Greenlands. A lot of people might bring up the old Guns and Butter speech, but its not so straightfoward when you plan to use the guns to get the butter. And I don't think the Ironborn could feed their families ever so well purely on trade as they did on conquest. 

Only in very, very long summers, summers so long it starts being counterproductive, would there maybe be population trend of more surviving children and teens, but all or most would die off when the storage for feeding them doesn't exist

Which means the Reach might be as devastated as the Riverlands if the winter lasts long enough

[LadyOfCastamere]

Frankly I am bad with numbers, but still it surprises me that around 40m seems to be the consensus. If that is true, then why do the armies of each region seem so small? If I remember correctly the individual armies seem to usually count between 10.000 and 60.000 (Highgarden being an exception with 100k). Summing it up, it would give us a total number of round about 220.000 soldiers - which seems little to me if there are seriously 40 million people around. Obviously, there's women and children and elderly people, but there should still be enough to raise much larger armies, especially in medieval times where the next war was always safe to come.

In short:
40 million total population
-12 million (female population aged 16 - 50)
-16 million (children below 16)
-2 million (elderly or sick men incapable of fighting)
__________________________________________
10 Million men aged 16 to 60, capable of fighting.

I'm aware not everyone is trained a soldier and there will be less men than women due to previous war tolls, but neverthless, 220k seems so super little.

I would have personally argued that the population number is a lot lower, perhaps around 16-18 million, with a high density in the big cities, ports and regions like Highgarden/Westerlands and rather little density in the North and Dorne.

[Ran]

The old "rule of thumb" for pre-modern medieval Europe was that approximately 1% of the population could be mobilized. Years ago GRRM agreed that a figure approximately  around 400,000 could be representative of the total mobilizable strength of the Seven Kingdoms, under notional ideal conditions, with the caveat that these would be the sort of figures a maester might understand but these weren't necessarily 100% accurate (a caveat wisely put in, as he later revealed that Dorne had inflated its strength as a defensive measure).

Other methods of determing the populace, including estimating population density from approximately similar medieval regions or using some information on cohort recruitment in the Middle Ages also yielded ~40 million figures, give or take 10%, and separately  an academic demographer doing his voodoo found that a 34.5 million figure, give or take, is one of the better fits for the evidence and certain constraints he put on his analysis. Which is pretty close to that 40m figure, all considered. (Which amused me, since he opened up his article scoffing at the 40m figure as something amateurs came up with, and then in the end his own figure isn't that far off...)

[DMC]

25 minutes ago, Ran said:

Which amused me, since he opened up his article scoffing at the 40m figure as something amateurs came up with, and then in the end his own figure isn't that far off...

I recall (one of the times) when you cited that article.  All and all it seemed like a convincing analysis, but yeah it was funny that the lede was very misleading.  Gotta get attention somehow, right?

[Corvo the Crow]

Since I asked this in small questions yesterday...

keep in mind the numbers we get are rounded and are almost always rounded up

Quote

 A hundred thousand men ride in his khalasar,

...

. "I'd let his whole khalasarfuck you if need be, sweet sister, all forty thousand men, and their horses too if that was what it took to get my army.

...

Drogo had called his khalasar to attend him and they had come, forty thousand Dothraki warriors and uncounted numbers of women, children, and slaves. 

...

Your lord husband alone counts forty thousandmounted warriors in his khalasar."

...

"It was four hundred years ago or more, when the Dothraki first rode out of the east, sacking and burning every town and city in their path. The khal who led them was named Temmo. His khalasar was not so big as Drogo's, but it was big enough. Fifty thousand, at the least. Half of them braided warriors with bells ringing in their hair.

Almost all the males of fighting age will fight, So roughly 40 percent of this population is males of fighting age. There are slaves, of course but not every one will own a slave and they are mostly sold and not kept. Couldn’t find the quote but there are old men as well, riding behind carts/wagons.

Not exactly what you were looking for, but could help some. Using this as a template for Westeros and the %1 figure generally accepted and as confirmed by Ran to have been thought by GRRM as well, you may get something.

[cpg2016]

On 6/5/2019 at 8:55 PM, DominusNovus said:

The urbanization rates are totally screwy, I agree. Westeros is very unurbanized (possibly implausibly so). That said, I do think that the US is a fair comparison, even though it is an industrial society by 1870, because so much of that land was utterly untouched. There is no Alaska in Westeros, not even the North. Heck, in 1870, most of the West was barely settled (not to dismiss the American Indians, but their numbers were quite low). Do we have anywhere in Westeros that is regarded as barren? Pretty much Dorne, and thats it. I think that the US could support that same population at a lower technology level. After all, Europe, combined is only slightly larger than the US (10.1 million km^2), and in the middle ages, it tended to have a comparable population to Westeros. If you'll permit me using wikipedia (I know, I know, but all the other sources are in the same ballpark):

https://en.wikipedia.org/wiki/Medieval_demography

We see the population ranged from 56 million to 90 million between 1000 and 1500, per that source. Others vary, but not by a large degree.

 

I would say that that has more of an impact on the urbanization rate. The population is far more diffuse than would otherwise be likely. Which means that the population is more protected from the harsh winters - it is easier to feed a small village than a great city.

Its also quite possible that the population is far more volatile than our world. Actually, not just possible, almost certain. Winter probably kills off more, but summer allows for more bounty and more growth, and the simple fact that while the population is more likely to be trimmed during hard winters, the infrastructure isn't, not to mention that in the areas with the hardest winters, its the oldest that tend to die first ('gone hunting' and all that). I expect every spring and and early summer to be something of a mini-baby boom. Also likely is that the population will see something of a material improvement come each spring, as property gets freed up - think something like the Black Death in Europe (which Europe recovered from in a century).

My guess is that Westeros' population graph would zigzag quite a bit, but if you average it out over each seasonal cycle, you'd probably get a value not too far off from that common estimate of 40 million.

Why do you find the lack of urbanization surprising?  I think it makes a lot of sense.  At the height of the medieval period there were a couple of large cities in Europe (Constantinople and Cordoba come to mind, both being in the 400-500,000 range at some point) and a lot of smaller places that were closer to the 25-50,000 range.  And then a handful of medium size places, like Paris or Milan, which clocked in in the low six figures.

Which fits in with what we know.  We're told of several large cities; Kings Landing and Oldtown are in that half a million range, you've got the smaller "big" city in Lannisport at maybe half that, and then a few named small cities like Gulltown or White Harbor which are probably 50-100,000 people.  We don't get a ton of time or information on the smaller market/port towns, but places like Maidenpool or the "many" but unnamed market towns in the Reach are probably approximately the size of London in the Middle Ages.  

GRRM isn't super interested in giving us a travelogue of the size of every place in Westeros, we hear a little about the important ones, when it matters.  KL has to be big enough that the mob has significant political influence.  Oldtown has to be as big for worldbuilding reasons.  Other than that... why should ew know?  And you're 100% right that wonky seasonal fluctuations play a big role.

Martin is notoriously bad at math, and the constraints of the size of the series means that he's not giving us superfluous demographic detail.  But when it comes to the larger cities everything makes a reasonable amount of sense.  And think about individual castles.  If the more important lords have keeps that are garrisoned with a couple hundred men, which seems right, then you have to expand that by a factor of at least 2 or 3 to account for non-military personnel like servants, blacksmiths, cooks, etc and their families.  Seems possible that many castle towns have populations that might be at least a few thousand people.

[Br16]

On 6/27/2019 at 3:24 AM, cpg2016 said:

Martin is notoriously bad at math

It's even worse when the power of coinage is concerned. I've read a detailed calculation/comparison on reddit that concluded the Hand's Tourney prizes were the equivalent buying power of hundreds of millions of dollars, which is just ridiculous. 

[Corvo the Crow]

On 7/7/2019 at 12:25 PM, Br16 said:

It's even worse when the power of coinage is concerned. I've read a detailed calculation/comparison on reddit that concluded the Hand's Tourney prizes were the equivalent buying power of hundreds of millions of dollars, which is just ridiculous. 

I’d like to see that, can you post the link here please?

[Lord Varys]

14 hours ago, Corvo the Crow said:

I’d like to see that, can you post the link here please?

No idea about the post, but it is kind of obvious if a horse basically costs just about single golden dragon. That likely implies you can get a decent house for 50 dragons, and a proper farm for a hundred dragons.

Tens of thousands of dragons are an incredible fortune.

[Sigella]

On 6/7/2019 at 12:23 PM, Free Northman Reborn said:

 Not really sure how a population recovers a Winter’s losses in 3-5 years of summer. I mean, even if you have a bunch of births for 4 years, all those babies are still toddlers by the time Winter hits again. So I can only assume the die off isn’t really significant unless you have a 5+ year Winter, which doesn’t happen very often.

Nah, presume you give birth one year into spring -then you have 2-4 more years of spring to be a toddler + 3-5 years of summer +at least two years of plentiful autumn. Giving you an age of 7-11 when the first winter starts. 

 

edit: You also have to consider that starvation wouldn't start in the very first year of winter too.

[Free Northman Reborn]

3 hours ago, Sigella said:

Nah, presume you give birth one year into spring -then you have 2-4 more years of spring to be a toddler + 3-5 years of summer +at least two years of plentiful autumn. Giving you an age of 7-11 when the first winter starts. 

 

edit: You also have to consider that starvation wouldn't start in the very first year of winter too.

My point is that the new births in the Spring/Summer don’t have a chance to reach adulthood and procreate before the next Winter hits.

So every child that dies in Winter reduces the number of future adults who can procreate and expand the species, but each child that is born in Summer does not increase that number again, as all of them will still be children in the next Winter, subject to renewed die offs.

So long term the population will shrink from repeated Winter die offs.

This is not a function of Winter being worse than Summer is good, but is instead a function of the slow human reproductive cycle. If Long Winters happen once in 30 years, that is probably survivable - one would have to do the math to be sure. But not if it happens every 10 years.

That’s why I say most Winters are probably fairly mild - lasting 2-3 years with sufficient food in storage and minimal die off - with 5 year Winters happening maybe once a generation.

[Sigella]

9 hours ago, Free Northman Reborn said:

My point is that the new births in the Spring/Summer don’t have a chance to reach adulthood and procreate before the next Winter hits.

So every child that dies in Winter reduces the number of future adults who can procreate and expand the species, but each child that is born in Summer does not increase that number again, as all of them will still be children in the next Winter, subject to renewed die offs.

So long term the population will shrink from repeated Winter die offs.

This is not a function of Winter being worse than Summer is good, but is instead a function of the slow human reproductive cycle. If Long Winters happen once in 30 years, that is probably survivable - one would have to do the math to be sure. But not if it happens every 10 years.

That’s why I say most Winters are probably fairly mild - lasting 2-3 years with sufficient food in storage and minimal die off - with 5 year Winters happening maybe once a generation.

Ah ok I see