What would count as an explanation of the size of China?

There are two striking facts about China.  First, the country is quite large.  Second, the country was remarkably large early in its history, compared to most other political units.  For instance, here is China in 200 AD:

How did this happen?

Or consider a modern version of the puzzle: currently there are over one billion Chinese in one political unit, and a bit of scattering.  And there are over one billion Europeans, spread in fairly significant numbers across about fifty political units.  How did such a fundamental difference come to pass?

I can think of many instructive explanations for China’s early size and unity that are nonetheless derivative.  For instance perhaps a common language for writing played a key role, or perhaps the civil service and the exam system bound the country together.  I don’t mean to gainsay those claims, but they are not fundamental.  In part they are simply alternative descriptions of China’s relatively early unity.  And there still ought to be reasons why those factors were the case, and some of them seem to postdate unity.  On top of that, ideally we would like the explanation to account for China’s periodic descents into fragmentation and sometimes warring chaos.

I can think of a few factors that might count as fundamental, and often they involve economies of scale:

1. There may be greater economies of scale in Chinese agriculture.  One specific hypothesis is that China’s “hydraulic” system of rice irrigation favored a centralized despotic authority (Karl Wittfogel, though I’ve never found this particular view convincing, see also earlier takes on “Oriental Despotism”).

2. There may be economies of scale for fighting land battles with horses.  Alternatively, when it comes to naval warfare — more common for Europe — small countries have a chance to punch above their weight, witness  England and Portugal.

3. China had lower climate volatility than did Europe, and that made it easier for a more stable equilibrium to emerge.  (Or the kinds of climate volatility China had mattered less for its agriculture.)  Big changes in climate, in contrast, periodically overturn political equilibria, most of all when agriculture was a huge chunk of gdp.

4. China has two main, navigable rivers running east to west, the Yellow and Yangtze rivers.  It also has a large space of relatively flat plains.

5. China was formed when the prevailing technologies favored size and scale, and thus size and scale were imprinted onto early Chinese political DNA.  This is a bit like the “inflation” theory of the universe.  (NB: This part of the explanation is arguably “accidental” rather than “fundamental.”)

6. China and Rome are with regard to size and early unity not so different, but China did a better job absorbing the “barbarians” and thus persisted as a larger political unit.

What else?  With some mix of those (and other) factors in place, the more traditional detailed explanations then kick in to promote China’s size as China.

Ideally, an explanation for China’s early size and unity, and why that size and unity bounced back from so many periodic bouts of warring states, should address the following:

a. Why the mountainous Tibet also ended up as a more or less coherent nation-state, and why that too happened fairly early.  That seems to militate against purely rice-based explanations.

b. Why Yunnan was absorbed into China at a relatively late date — the 17th century — but once attached did become a stable part of the country in a manner that other parts of southeast Asia did not mimic.

c. Why Korea remained separate.

d. Why the Khmer empire proved unstable and perished, despite a high level of sophistication and state capacity.

e. Why the Aztec Triple Alliance grew to a much larger size than any political unit in North America at the same time.

What else?

I am grateful to a presentation by Debin Ma, and to comments from the Washington Area Economic History Seminar (recommended!), from a seminar last night.  None of them are implicated in what I have written.  I look forward to Debin’s paper on this topic (here is his earlier 2012 work), and Kenneth Pomeranz is writing an entire book on the question.

Addendum: Here is the Ko, Koyama, and Sng piece (pdf).


Many factors may have contributed, but obviously the most important is the very high average IQ of the Chinese and their horses.

How smart were those horses?

Let's just say their donkeys weren't called "smartasses" for nothing

But were the donkeys book smart or street smart?
I stil don't understand how the horses could take an IQ test.

Multiple choice is easy for horses but they have trouble with the essay questions.

I guess it makes sense, I have seen horses using their hoofs to count. Thanks.

Clever Hans.

Now, I feel blindsided again. So if I understand the consensus here, Chinese Hans are clever because they are born this way, but German Hans was clever because he was well-taught and everyone knows the German training system is the world's best. Are American schools failing American horses? How can America train American horses for the jobs of the future?

I just want to say how great it is that "Thiago" isn't pretending to be a Brazilian any more.

It was a good run, big guy. We all enjoyed your 700-word posts in respond to my six-word trolling. But all good things must end.

I never pretended to be Brazilian, I am, but Brazilian schools are not failing Brazilian horses, this is an American issue.

The early urbanization, specialization, and trade that this large state enabled eventually resulted in the higher IQs.

Not sure about the horses.

Is there some genetic effect of urbanization? Or maybe they are just better at taking tests written by elite urbanites, when living in urban centres as compared to when living on farms?

Indeed, I believe Charles Neigh Murray wrote about this in his magnum opus, the hoof curve. Some liberal idiots think it's due to the nutrition content of Chinese barley, but horse twin studies prove that this is fundamentally false.

Of course, there are those neigh-Sayers who say horse IQ tests are culturally biased towards the Chinese, but this is simply anti-science. I mean, horses can't read anyways.

Does it mean Silicon Valley should be allowed to recruit more horses overseas?

Europe is constituted of the parts of failed empires. China isn't.

China is almost entirely constituted of failed empires: The Shang, Zhou, Qin, Han, Sui, Tang, Yuan, Ming, Qing, and those are just the big ones. The difference is that these failed empires were much bigger than most failed European empires. But an explanation for that is almost exactly the question Tyler is asking.

In what respect were those Chinese dynasties "failures"? Please be specific about what you mean before using that word. I wouldn't agree that their coming to an end, constitutes "failure".

The Song was the other big one.


You have a large fertile area with relatively easy internal transport surrounded by difficult terrain.

Easy internal transport leads to unification. Difficult surrounding terrain keeps potential rivals at bay.

Korea was too mountainous and did not produce enough economic surplus for conquest to be cost effective.

How did a fairly locked out country like Rome manage to conquer the west then? And why are France and Germany separate rather than a single country? I grant that this does explain the independence of islands.

It seems less a question of conquering as of maintaining -- which leads nicely into your second question (for which I have only the most superficial answers: language and the dissolution of Western Rome).

The Med is the easy internal transport that the Romans took advantage of. Look how they spread, it's along the coasts of the sea. Even close by areas like Alpine Gaul weren't conquered until very late. The only exceptions are really northern Gaul and Britain, both of which sloughed off the empire early in its dissolution.

Rome/Italians never have been / will be good seafarers. See costa concordia.

Almost all of romes internal troop movements used land.

The Genoan, Ragusan, and specifically the Venetian Republics say hi. They would also like you to repudiate your bad sentence.

France and Germany are different countries because Charlemagne divided his lands between his sons.

The Rhineland also formed a sort of natural boundary, and it reinforced a cultural boundary created under the Romans: Latinate speakers to the western Germanic speakers to the east. (Minor quibble: It was Charlemagne's grandsons who divide the empire after their father Emperor Louis/Ludwig the Pious died.)

Precisely because the Roman Empire was not landlocked-- look at a map of the Roman Empire. Apart from its northern extension into Europe, and it extension down the Nile, the Roman Empire is based on the Mediterranean. Much of it was in fact acquired so the Romans could protect their commerce, and above all the vital grain trade, on "Mare Nostrum". The push north was mainly done to keep Gaulish and Germanic invaders away from Italy.

I agree that it has similarities to Ancient Egypt with two rivers instead of one.

Exactly. Surrounded by mountains, desert and ocean. Whereas Europe was forced to defend more borders.

Not something I'm an expert on, but a possibility: Confucian philosophy. This fits nicely into the rise of a unified China, and makes a certain intuitive sense as a cause, given the value it places on hierarchy an authority.

U.S. and Canada both have a larger landmass than China. Must be all that Confucianism of the British settlers.

Except that the Spanish settlers beat the English by more than a century, and St. Augustine was founded 52 years before the first British settlement - https://en.wikipedia.org/wiki/St._Augustine,_Florida

The French too were not exactly layabouts, with Quebec being founded one year after Jamestown - https://en.wikipedia.org/wiki/French_colonization_of_the_Americas

"The French too were not exactly layabouts, with Quebec being founded one year after Jamestown"

Quebec was marginal land - barely worth having for the French and not worth seriously defending.

At the Treaty of Paris 1763, the French abandoned Quebec to save their sugar islands in the Caribbean. To this day, Quebecers have a hard time accepting that they were sold out and abandoned by France for sugar.

At the time the fur trade was a pretty big thing.

Sugar is sweet, Quebecois not so much

St Augustine was a naval base only, intended to keep the French and English from pushing too close to the Caribbean and also to fend off pirates (ineffectively) who preyed on the Spanish treasure fleet. There was no serious Spanish attempt to colonize Florida.

Confucianism would indeed have been an important factor in promoting the stability and prosperity that would encourage a larger China. Although there are elements of hierarchy and authority in Confucianism, it could be argued that Confucianism also strongly limited and controlled the exercise of political power, and that these limits along with Confucian virtue ethics offer greater influence on China 's size.

The competency of Chinese government tended to impress European visitors, so maybe Confucianism, maybe the exam system?

Probably more because of geographical attributes. China isn't surrounded by mountains, a desert, and an ocean vs a small sea like the Mediterranean.

Is surrounded rather

I'm comfortable with these being semi-random outcomes. Throw in a little Jeff Goldblum monologue about chaos and strange attractors. Most observers during most of human history could point to a large political unit that was "winning." One hundred years ago that would still be the British Empire.

Controlling roads to and from mines and agricultural commodity producing areas in numerous colonies does not imply any amount of "winning" in other places.

Also, let it be known. There are alternatives to rule or be ruled.

That is patently clear throughout much French and Anglo history and political practice, as compared to, say, the czars and those who followed them. What alternative to "be ruled", in the most extreme sense, was on offer in Soviet Eastern Europe, for example? In East Germany, we became appraised of the Zersetzung, reflecting the operations manual of the oppressive Stasi. But is anything known of whatever arose from further east?

The reason I point this out, is that the language of "winning" in this context (specifically, historical power of Britain) can for some audiences serve as a certain sort of propaganda wherein the overall legitimacy of current governments - which, despite not controlling every twitch or word, have decision making leaders and representatives, regularly chosen by "the people".

They were probably less complacent than europeans. Or more.

'perhaps a common language for writing played a key role'

Considering that speakers of different languages can communicate using a single set of ideographs, it is a bit flippant to say it is not 'fundamental.'

There are many disadvantages to ideographs compared to an alphabet. However, the one major advantage of ideographs is the ability of people who do not share a single spoken word in common to communicate using them. An alphabet based system is inherently reliant on users sharing the same language.

'In part they are simply alternative descriptions of China’s relatively early unity'

Somewhat tautological in attempting to dismiss the point - being part of that early unity is not the same as saying that an ideograph based system was not a fundamental building block in creating that unity.

I had a professor in college who attributed the cultural collectivism or communitarianism of China to some unique hydrological features of the Yellow River. The claim was that the Yellow river carries an unusually high concentration of silt after it passes through the Loess Plateau. This silt than starts to fall out of suspension when the river slows down as it reaches flatter, lower land with less elevation gradient. The silt accumulates on the river bed, creating underwater dams and sometimes building up the river bed to a somewhat higher elevation than the surrounding plains. Hence, when the river floods, it is frequently catastrophic, as the river easily changes course, flooding the most productive farmland in China and causing large famines. Hence attempts to prevent the river from flooding were very important, so important that one of China's mythical founding Emperors Yu the Great, the founder of the Xia dynasty (the first dynasty according to traditional Chinese historians), is credited with having figured out how to effectively control or mitigate the effects of the flooding of the Yellow River. The point being that if there was any weak point in the defenses against flooding, there was a very large risk that the Yellow River would flood, cause famine, and consequently also bring about a fair amount of violence due to the weakened status of the Empire. So if anyone shirked their duties, especially in regards to flood control, it could be putting everyone at real risk. If it is true that this kind of intensely collectivist culture is in some sense endogenous to the landscape, that might explain why China was able to grow so large so early, as people were already culturally attuned to obey authority, as opposed to societies full of more individualistic, unruly, and warlike barbarians. Anyway, that was the theory the professor expounded.

Sorry, I meant to post this in a separate thread here. Ideographs or logographs certainly do make administration of a polyglot empire easier, although there are some phonetic elements within Chinese characters when used to represent Chinese words. If you try and use Chinese characters to write down Japanese words, I am not sure if there remain any phonetic elements. Also, most of the languages spoken in China are from the same language family and have similar grammar, and some unique features that enable Chinese characters to be used to write the multiple languages, such as a lack of declensions, a lack of verb conjugation, and a lack of the use of phonemes used to distinguish singular and plural, and a shared SVO sentence structure. It is actually fairly easy for English speakers to gain some competence in reading Chinese if you can learn what the characters mean.

Kanji (the Japanese logographs you're talking about) tend to have multiple readings. Most have an onyomi (based on the Chinese pronunciation) and a kunyomi (based on the native Japanese pronunciation) with different pronunciations (and frequently subtly different meanings). Some have more than just one of each, a few have only one reading. They also have two other writing systems adapted from Chinese characters, hiragana and katakana, but those are syllabaries rather than logographs.

Anything with an onyomi should be easy to write in Chinese (not sure how much the switch from traditional to simplified would affect this, though), but words that only have kunyomi or are loan-words without kanji (lots of computer terms, for instance) would probably just have to be translated. As far as I know, Chinese doesn't have an syllabaric writing system in common use, unless maybe you count the romanizations like pinyin.

I don't know of any syllabaric writing system in common use in China. I was thinking of the fact that Chinese logographs are often composed of multiple other logographs which contain information about the meaning of the logograph and its pronunciation. I think that there are some patterns to the way that you see this, but I cannot remember the way in which it is most usually seen. My recollection is that a character that is a compound of other characters would with some frequency have a component character on the upper left hand side somewhat related to the meaning of the word, another component character in the upper right, and another beneath the other two which gives some information about the words pronunciation.

It's not unique to that professor. China has been described as a hydraulic empire to the point where the term has its own wikipedia page https://en.wikipedia.org/wiki/Hydraulic_empire

For a good long time educated people in western Europe also shared a common written language: Latin.

India had nearly just as big an area as a single political unit in 250BCE. And was pretty big as a unit (Gupta Empire) in 400AD.

They also had geographical boundaries based on mountains, etc.

This suggests that geography is the main factor.

The Deccan Plateau was the last area integrated into India's ancient kingdoms, just like Tibet.

India was only very occasionally a unified political state: Under the Maurya in the 3rd century BC, and under the Mughals at the end of the 17th century. For most of its history India was divided into distinct polities, some of them rather large (The Sultanate of Delhi dominated the Ganges basin for much of the Middle Ages) and some rather small city states, like the Dravidian trade cities in the south.

Seen elsewhere. The remains of two ethnic Chinese were unearthed in a Roman cemetery in Britain.. Apparently, they were well-traveled.

Link? If true, that is fascinating, which is why I don't believe it.

Phoenician coins were found in Britain, which started much ado about nothing. It showed indirect trade iirc, but not the early direct links of silver to timber/tin that had been hoped for.


Is it accurate to describe China as a single country before Mao? For much of its history, China was dominated by warlords. So was Japan, until the Meiji Restoration unified Japan. Due to the division in China, it was dominated by much smaller countries including Japan. That experience of being dominated by foreigners, including the British as well as the Japanese, contributed to Mao, nationalism, and the economic miracle we know as China today. American conservatives prefer to emphasize states over the national government, to strengthen the former while weakening the latter. Our Austrian friends prefer even greater division, or devolution, the city-state. Can they and the Chinese both be right?

Of course, this argument has been taking place in America since its founding. I just started reading The Framers' Coup by Michael Klarman and it's going to be very good read. The "coup" refers to the Federalists having prevailed in the argument over ratification of the Constitution, which was opposed by many (including George Mason) because of the power granted to the national government, the strongest case made for the national government by Washington et al. that it was essential if America was to survive not only another war with the British, France, or other European powers but an insurrection by anti-government mobs.

I haven't quite got this straight in my mind. Are those who advocate a weak national government subversives who are undermining America and making America vulnerable to its enemies?

"Can they and the Chinese both be right?"
Yes: Singapore is a Chinese city-state ruthlessly rule by technocratic Mandarins.

Lee Kuan Yew wasn't ruthless, he was smart. Smart in the way he ran the government in Singapore and smart in the way he ran Singapore's relationship with China. In a previous comment I alluded to the Five Families and what was required to maintain the peace. Lee Kuan Yew knew how to maintain the peace.

"Lee Kuan Yew knew how to maintain the peace."
So did Salazar and Franco and a number of other Fascist leaders. It does not xhange the fact they lead by terrorizing their peoples.

What terror is there in Singapore?

Complaints about their curtailment of free speech is a little overdone. They don't allow malcontents to engage in demagoguery, but its not like you can't criticize the government if your respectful and informed. The government does log public criticism and occasionally uses it to change policy course if such criticisms are legitimate.

While the various parties certainly try to gain advantages for themselves, they have free and fair elections. The PAP wins primarily because they do a very good job.

Their criminal justice system really only cracks down on people that engage in serious crimes. Most complaints I here about Singapore are from people who don't even know the law (they do not in fact arrest people for littering, though you might get a ticket).

They score pretty high on business freedom indexes and low on taxes.

Where is the terror? Nobody seemed terrorized when I was there.

Yes, it is. There's a reason the opening line of the Romance of the Three Kingdoms is, "The empire, long divided, must unite; long united, must divide." Mao's not an exceptional uniter; he's just a another founder of another dynasty in a long cycle, after the dissolution period of the Opium War through the end of World War 2.

The Tokugawa shogunate ruled all of Japan for hundreds of years before the Meiji restoration. I believe other shogunates ruled all of Japan prior to that.

the ignorance: it burns!

China has two agricultural systems. A rice, irrigation system in the south and a semi-arid wheat dominated system in the north.

This would appear to argue against claims that the rice culture of cooperation on irrigation played much of a role in the nature of the country.

The South of China is different.

but still China.

Europe is a peninsula, made up of smaller peninsulas, themselves often made up of smaller peninsulas. It means it is virtually impossible to control with a large army in a central location. Add the Alps, the Pyrenees, the Channel and the Carpathians into the mix and uniting Europe by force is a fool's errand.

Incidentally this explains why Koran is a separate country (Peninsula), Tibet (analogous to Switzerland).

Another reason is that the river system allows bulk transportation of food across vast areas with uncorrelated food shocks. If there is an earthquake or draught in Province X, the central government can send wheat from unaffected provinces Y and Z by river barge, which is by far the safest and cheapest option. This imposes obvious costso to breakaway provinces.

In Europe, rivers tend to be much shorter, and famine relief will generally involve expensive/dangerous sea travel. Again, South Korea is on a different river system, and I am guessing Tibet rivers are not particularly navigable.

Why isn't Shangdong its own country? Its also a peninsula.

I believe that there are few natural barriers between Shandong and the rest of the North China plain, i.e. nothing comparable to the Amur.

Its also worth mentioning that Shandong is right next to the cradle of Chinese civilization, while the areas adjacent to Korea were more peripheral for most of early Chinese history.

A book I read about Gengis Khan's empire said that the dense European forests at that time might have made further conquest difficult. (Although, it doesn't seem to have stopped the Romans?)

Maybe look into how terrain favored attack versus defense?

Cavalry vs infantry based armies. This is a large part of the reason that Roman eastern expansion stopped where it did. Everything that could be dominated by an infantry centric army was taken. The eastern steppes were ruled by cavalry centric armies, in particular horse archer armies of light cavalry.

In the west, a cavalry based army would suffer disadvantages due to topography and forestry. In the east, Roman legions were annihilated by bow and arrow without ever even closing with the enemy. See Marcus Antonius' dismal attempt at invading the east.

Europe was very poor at the time. Genghis didn't really try.

[Citation Needed]

Dense forests did stop the Romans: they were crushingly defeated by the Germans in Teutoburg Forest in 9 BC, and never tried to conquer Germania again.

This Quora thread on how Korea stayed independent is of interest: https://www.quora.com/How-was-Korea-able-to-resist-China-and-remain-a-separate-nation-state-throughout-history-given-that-it-is-so-close-geographically-to-the-mighty-nation

I have a theory that the reason why Korean dramas are so valued is because Korean language and culture has evolved to be more fierce due to having to defend from China, Japan and Mongolia throughout their history.

Their language took on their historical drama basically. It just sounds more dramatic.

I go for topography
e.g., the Han empire covers pretty much the eastern plains missing the mountainous and highland regions.
Other examples might be the Persian or the Ethiopian empires covering more or less the respective plateaus. The Incan empire streched over the Andens after developing the matching infrastructure (but not over to the lowland plains).
The Espanas lie on the higher sections of Iberia while Portugal covers the western lowlands
and so on

Less disease in China may have made it easier to unify the population. Sickness promotes independence.

Wait, there! What about the role of government investment in infrastructure? The Grand Canal is by far the most profitable public works project of antiquity. It supercharged the Chinese economy for centuries. Whatever its cost, it's paid it back thousands of times over. Those ancient Chinese were smart cookies, indeed.

China has been so large for so long, as opposed to being divided into multiple nation states, because:

1. Shared writing system
2. Shared spoken language - Mandarin - dictated by Mao. It is now the official language of China. India should GB for English. I'm glad I don't speak Irish or Italian.
3. Confucianism - a group oriented ethos that subordinates individual needs to the family and community, respect for tradition and authority, and veneration of past generations.
4. Genghis Khan - his DNA is shared by a significant percentage of the population - he had a thousand(s) concubines and even more daughters.
5. A wide, relatively flat, terrain spanned by an enormous navigable river.
5. Horses because Ghengis.

It is interesting to ponder what that means for the USA and for Europe.

Wow, I didn't know Mao had a time machine and dictated the course of the country thousands of years earlier. Truly, he must be history's greatest leader.

I meant to say: "India should thank GB for English".

Only a small percentage of Indians speak English (I've read 15), and many not very well.

15% is a lot of Indians...

Sure, it helps many Indians get IT jobs, but does nothing for national cohesion. India these days has a de facto two-tier caste system: those who speak English and those who do not (or very poorly.) And there is resentment in India over that, much the same way in which there is resentment in the US of the globalist elite.

For Christ's sake, Cowen, read some Turchin!

In case you don't know, it's offensive to Christians to take the Lord's name in vain.

I assume you don't use the N-word and don't call homosexuals F----ts because you don't want to offend. Please show the same courtesy to others.

It's all about the geography. China has a huge central plain that was conquered quite quickly once someone - the Emperor of Qin State - was strong enough to do so. Until then, China was made up of various independent kingdoms.

Various Chinese and Korean dynasties fought each other, and at times strong Korean dynasties ran much of what is now northeast China and southeastern Siberia. The Mongols tried to conquer both of them, crushing the existing Chinese dynasty but ultimately failing in Korea. As they set up their seat of power in China as the "Yuan Dynasty" in what is now Beijing, the Mongols are regarded by Chinese as a Chinese dynasty. Korea is very rugged and mountainous, and has very harsh winters, both of which make it tough to conquer, especially from the land.

China actually did conquer Vietnam and occupied it for over 1000 years. After that, stronger Chinese dynasties influenced (and occupied) parts of VN on and off until the French came in the late 19th century.

Both Korea and Vietnam purposely abandoned or de-emphasized Chinese characters and developed their own script at least partially to assert cultural independence from China.

What the hell is this Korean propaganda? The Mongols failed to subjugate Korea? What the hell. Korea was made a province. The king kept because he kowtowed so hard the Mongols thought him more loyal than a fellow mongol.

It's fashionable because Trump mentioned that Korea was a vassal state of China. So, the poseurs have to act like they know Korea has always been independent.

How relevant is some instance of having been ruled many many hundreds of years ago?

If you're going to talk about non-independence in Korea, the more relevant example would be Japan. However, in either case, Korea has been far more independent than otherwise. Among other things, they speak neither Japanese nor Chinese, which is sort of proof that external rule was rare or highly incomplete.

The cliodynamics/social evolution crowd has a lot to say about this. China faced a much greater threat from horse-riding pastoralists on the steppe. In premodern times, horse-riding pastoralists had exceptional war-making capacity. Since the main form of wealth in pastoralist societies (livestock) is easily taken in raids, pastoralists are subject to endemic warfare, and the social units that survive are geared towards warfare. The invention of horse archery further increased their war-making capacity. Morris at least calls this the "nomad anomaly", where steppe pastoralists' war-making capacity is far out of proportion with their social development, but I don't know if he coined the term.

So the ecological frontier between settled agriculture and nomadic pastoralism selected for agrarian states that were highly cohesive and functional, since non-cohesive, weak agrarian states would be destroyed by pastoralists. Eventually, one such strong state would be strong enough to turn around and conquer other agrarian states, especially ones far from the agriculturalist-pastoralist frontier, which did not face as strong selection pressure for state and war-making capacity. The Koyama et al. paper also argues that the agriculturalist-pastoralist frontier drove Chinese unification, but the mechanism seems to be different (though not mutually exclusive with the process I talked about) (I've only skimmed the paper so far).

There must be other processes by which states and empires form, since many of them have been built away from such frontiers, as in pre-columbian North America and Mesoamerica and in Western Europe. Rome didn't form on an agriculturalist-pastoralist frontier. But then again, when Rome fell, no one ever managed to unify the Mediterranean basin again.

Agreed! This issue has been thoroughly addressed by Turchin, Morris and others as a response to a plethora of steppe Nomadic raiders. The pastoralists live martial lives with swift horses and learn to organize together to steal from the agricultural communities. The Mongol recipe is simple -- unite as needed to quickly overwhelm any local community and then steal all their surplus. The logical response of agricultural communities (in this case China) was to unite together into a very large empire. Failure to do so threatens survival.

Europe had better natural barriers of forests and mountains.

I am not saying this is the only answer, but it should at least be added to the list, as it what several prominent historians believe.

I would actually almost reverse the question though, as empires were common historically. The better question is why was Europe (and earlier Classical Greece) able to resist unification for so long? As most historians of the Great Divergence suggest, intermediate levels of fragmentation seem to be necessary for rapid cutting edge (as opposed to catch up) progress.

Oh, and BTW, China was fragmented various times throughout history and these were often eras of cultural dynamism (as were the initial stages of unification).

Europe suffered from raiders too, and right up into the Middle Ages: The Vikings by sea and Magyars by land terrorized most of the Continent up until the 10th and 11th centuries. The southern coast was plagued by raids by Moors and Berbers in Africa. And of course eastern Europe fell to the Mongols, the Balkans to the Turks.

Any explanation that relies heavily on rice cultivation has to contend with a couple things:

1) Much of northern China has traditionally relied on wheat and millet rather than rice, and it has historically been northern, rather than rice growing southern, Chinese who more often have been the drivers of China's political unity.

2) Rice didn't have this effect on Southeast Asia.

Talk about Gell-Mann amnesia.

I would've thought that given Mr. Cowen's reading speed he could read all the history books written about China in a week, but apparently there are things he doesn't understand.
First of all that map is wrong; Koguryo was further north, and somewhat bigger.

China is big for the reason that Rome or Persia were big. Empires are efficient. China more so. China also had the advantage that it's vector of conquest was south; and it's always easier to conquer to your south. Gets warmer the further you conquer. And the south is mountainous; hard to build a big state there, while the north is a huge plain; so in every instance of collapse of central authority China usually had a big state on the north and several small states in the south; and the guys on the north always ended up conquering the southern states.

Isn't this discussed at length in Guns, Germs, and Steel? Wasn't the idea there that, in China, you had relatively fewer mountains, thick forests, and coasts forming natural borders (vis a vis Europe) and relatively fewer climate problems, deserts, etc. (vis a vis India)?

Also, regarding Korea, isn't the obvious answer "tons of mountains" ... and the Chinese weren't never really great at that whole boat thing given that there was no pressing need to be good at boats.

Precisely. Topography and physical geography has everything to do with today's political geography.

Why the Aztec Triple Alliance grew to a much larger size than any political unit in North America at the same time.

Did it? Our available sources on the various Mound Builder cultures are so thin (no written history, no visitor chronicles except De Soto in the southern reaches, mostly wiped out by disease and decivilized before anyone else came by) that we have no good idea if the political organization was all city-states, if it had several polities in the north of the same order of magnitude as the Aztecs, or even a single empire that dwarfed the Aztecs. At best we can make guesses on the basis of physical artifacts, but that's pretty weak when none of them have any writing on them.

Mound Builder Civilisation: http://dispatchesfromturtleisland.blogspot.com.au/2012/01/monroe-louisiana-as-source-of-new-world.html

China is not more geographically divided than other regions of the world. One could as well argue that the deep reach of the Baltic into the North European Plain, which is very flat and climactically homogeneous, helped to knit it together, as much as it helped to divide it, much as the Mediterranean did for Rome. The Chinese advantage in rivers and waterways was not exploitable until the Southern parts of China (rich in rivers) and the Grand Canal was constructed, and this occurred relatively late in its history, from the 7th Century on at the earliest.

The cliodynamics school is certainly onto something, and this is very apparent because there is a second geographical anomaly here: why is it that every successful steppic group that expanded into Eurasia nucleated on the borders of China and moved West? The Huns, Xiongnu, Kushans, Xionites, the Gokturks and all their descendant polities such as Khazaria, the Sekjuks and the Golden Horde, the White Huns, Kimeks, Kharakhitai, and of course the Mongols themselves and all their descendant Empires. Every steppic federation that was prominent in Eurasian history traces at least part of its social origins to the borders of China.

Isn't this reverse causality? The Eurasian steppe streches from Hungary to Mongolia so it would make sense that a successful nomadic group -- almost by definition -- would have a history of dominating some or all of this steppe land at some point in its history.

The climate also grows a bit warmer and wetter as one moves west from Siberia and the Gobi Desert. Any cooling period (as in the 6th century AD) tended to send waves of migrants west from the region, as well as south into China and India-- but they were often turned back in battle there. Until the evolution of a strong Russian state in the early modern era there was nothing to shield Europe from these people.

I'm surprised that no one mentioned the essay from David Friedman on the size of nations, which is the topic being discussed here. I'm not sure how much it answers the question of China though, since it's been a very long time I haven't looked at it

Here is the essay: http://www.daviddfriedman.com/Academic/Size_of_Nations/Size_of_Nations.html#fn1

#1: topography, #2: the existence of the Roman Empire and its subsequent dissolution. It's way simple.

You're really trying to force a much more complicated Guns Germs and Steel lately. Everything has to do with agriculture, geography, and the weather. In many countries of Europe, the indigenous people speak extremely different dialects depending on which mountain valley one finds himself in. They have different wedding traditions, as one example, and numerous other cultural traditions.

Another way of putting it is that the Great Man theory of history - that individual humans have any impact on political geography - is in the micro scale, whereas real changes in the macro scale have very little to do with individual humans and much more to do with how humans handle and manipulate their surroundings generally.

I took a class on Chinese history maybe 39 years ago. The actual process of the Chinese empire's southward expansion was left hand-wavingly vague. Was it violent, was it by consent of the locals? It would seem like almost as big a story in Chinese history as westward expansion is in American history, but it seemed much less well-documented than other aspects of Chinese history.

Or at least that was the state of the art in the 1970s.

China had basically reached the modern southern border by 218BC. Completely done by 100BC By conquest, sure, but the locals weren't many, nor very warlike. The area wasn't heavily colonized by Han farmers for millennia, and it wasn't heavily populated at all, so it's not a big part of the Chinese histories themselves. No great battles. Just "natives did something stupid, we sent troops, annexed the place, revolts happened twice a century or so".

The big stories are the westward expansions to Central Asia. Those are very well narrated. Great stories there.

Chinese eventually replaced local languages in southern Chinese (these were apparently related to the languages of Indochina) but very distinct Chinerse dialects-- intelligible with Mandarin only in writing-- did develope there, much are Latin developed into the Romance languages

The actual conquests weren't particularly interesting, but the shift from North to South China is indeed one of the biggest stories in history. For the last thousand years, South China vastly outperformed the North in pretty much everything (academic achievement, business, science, culture, wealth, whatever) except fighting capacity. In other words, being next to steppe nomads may not have had such good long term effects.

Might this have to do with the fact that the South had better access to trade routes reaching Indochina, India and the middle East? The culture emanating from the Pearl River delta has just traditionally been so much more involved in global commerce, as well as emigration.

I have one thought to add. Korea - why did it remain independent?


It was annexed once, barbarians from Manchuria invaded. Tried once again; didn't work out. Korean kingdoms were very nice vassals, the real estate isn't very productive, so there was never a good rationale for invasion.

A possibility:

Gengis Khan conquered all the mongolic and turkic tribes of central Asia, and then, as an unit, they conquered China (meaning that, in the end, China becomes even bigger: now it was China + turco-mongol territories).

In contrast, most of the germanic tribes of central and eastern Europe, instead of being conquered by Attila and his Huns, ran to the Roman Empire, who was splitted between several independent tribes.

Of course, a mystery remains: why central Asia barbarians were more easily unified than central European barbarians?

>why central Asia barbarians were more easily unified than central European barbarians?

Brexit was a death blow to the status of people like Tyler, but I would not completely give up on the unity of European barbarians just yet.

A divided Europe is a united Europe?

Black is white?

I don't understand. And anyways, there is still loads of room for national sovereignty and cultural expression, etc., within the political union.

Central Europe barbarians weren't highly mobile like horse tribes.

Europe has lots of geographic barriers compared to Central Asia.

China is one nation because the U.S. didn't intervene during it's massing the way the U.S. intervened during WWII.

Maybe its the presence of the large offshore island of the UK? It seems like the policy of the English at least since the Norman conquest was to support the second most powerful nation on the Continent to maintain a balance of power. The UK could quickly ship troops and other resources from their stronghold island well protected from invasion, to any particular theatre of war. If they hadn't been there, maybe the Roman empire would have reformed in the 13C and 14C, like it almost did prior to 10C a couple of times, with the various incarnations of the Holy Roman Empire.
Japan is a bit too far from China to play this role for China and separated by the Korean mountains.

Looking at the question: (There are) currently there are over one billion Chinese in one political unit, and a bit of scattering. And there are over one billion Europeans, spread in fairly significant numbers across about fifty political units. How did such a fundamental difference come to pass?

To me there are are actually a couple of questions that it breaks down into.

The first of which - Why Europeans are living in so many political units compared to Chinese? - is answered mainly by the nature of intensive vs extensive agriculture. Europeans spread out across North America, South America, Australia and Europe itself. Their economic development was focused towards opening new agricultural frontiers, and to a lesser degree towards maximizing output within a particular land area.

The question of why so many political units for Europeans is mainly answered by land area and that is answered by the agricultural strategy.

(Though really TC, you should be counting the significant numbers of overseas Chinese in all the SE Asian states here, as well. There's more of them than there are people at all in some small European states).

The Europeans also went through the demographic transition about a 70-100 years earlier than China; that's the shift to fewer, richer people, and if China had gone through DT earlier or the West later, we wouldn't be talking about as similar numbers.

That gets us to the second question - Why are there are about 4-6 states in the European (roll together say France, Germany, Poland, Britain) for the same land area where the Chinese had one ("China Proper")?

I think that's answered by the differing nature of the Chinese state - "The mountains are high and the emperor is far away". The Chinese state was armies (though typically with less share of tax than in Europe) and massive public works (e.g. the Grand Canal), with little intrusion or regulation of common life.

Europe was quite different, with much more developed, intrusive governments that then required a much greater intensity of accountability and measures that limited state size relative to population.

Take Mai Menghua, writing in the late 19th century - "Comparing the Power of China and Western Nations" - "Over, all five continents and throughout all past ages, no ruler has had less power than in present-day China, and no rulers have had more power than in present-day European nations.

In Western countries, the age, birth and death of every person in every household is reported to the officials, who record and investigate it. An omission in a report is punished as a criminal offense. In China, birth, death and taking care of oneself are all personal matters, beyond state intervention.

In Western countries, when property is inherited by descendants, the amount of the property and its location must be reported and registered with the authorities. An inheritance tax must be paid before the property is transmitted to the inheritors. In China, people give and take as they please, and the state is unable to investigate.

In Western countries, when children reach the age of eight, they all go to elementary school. Doting parents who neglect their children's studies are punished. In China, 70-80% of the population is indolent, worthless, uncouth and illiterate, and the state can do nothing to encourage them to improve themselves.

In Western countries, one must go through school to become an official, and unless one does adequately, one cannot make his own way. In China, one can be a slave in the market place in the morning, and bedecked in the robes of high office by evening, and this is beyond the capacity of the state to control.

In Western countries, the currency system is fixed by the court; one country has the pound, another the ruble, and another the franc, but each currency is uniform through the entire country, and no one dares to differ. In China, each of the 18 provinces has a different currency, and the shape of the money is different. The people are satisfied with what they are accustomed to, and the state is unable to enforce uniformity".

The list goes on. A weak state like the Chinese state has a lot more easy extendability over a large geographic area; little tax revenues needed to support a fairly skeleton civil service, and the people do not press for an accountability from a state that does little and asks for little (more an absence of the concern of "No taxation without representation" because there's little taxation, and the consequence is the state does little).

If anyone is interested in reading more on the size of nations, here is a list to start with:

Alesina, Alberto and Enrico Spolaore, "On the Number and Size of Nations," Quarterly Journal of Economics, 112, 1027-1056, 1997.
Dudley, Leonard, The Word and the Sword: How Techniques of Information and Violence have Shaped the World, Blackwell, Cambridge, MA, 1991.
Findlay, Ronald, "Towards a Model of Territorial Expansion and the Limits of Empire," in Garfinkle and Skaperdas (eds.)The Political Economy of Conflict and Appropriation, Cambridge University Press, 1996.
Friedman, David, "A Theory of the Size and Shape of Nations," Journal of Political Economy, Vol. 85, No. 1, pp. 59-77, February 1977.
Tilly, Charles, The Formation of National States in Western Europe, Princeton, Princeton University Press, 1975
Wittfogel, Karl August, Oriental Despotism; a Comparative Study of Total Power, New Haven, Yale University Press, 1957.
Wittman, Donald (2000) The Wealth and Size of Nations, Journal of Conflict Resolution. 44:868

I think you are looking at the question backwards. The real issue is why is Europe fragmented? If you start from 1AD, there were three big empires: Rome, Persia and China. Rome effectively splintered into three pieces which have never been able to get back together: the Germanic kingdoms of Western Europe, Byzantine Eastern Europe and the Caliphate. Of these three, only Western Europe did not coalesce into an empire. There were repeated attempts by Franks, Germans and French to unite Europe but each attempt failed. Persia, China, Eastern Europe and the Caliphate all repeatedly formed large empires and there is the feeling that these places will continue to form empires - just look at the power of the word "caliphate" today. Europe on the other hand is fragmented and continues to fragment. British don't want to be European, Scottish don't want to be with England etc. It is Western Europe that is the odd one out, the only one who does not feel the power of being a united empire. Regional interests are much more powerful than a united identity. As Brexit, WWII, the Napoleonic Wars etc show, there has never been a force strong enough to unite Western Europe either willingly or unwillingly into a single political entity. The same goes with the pieces of the old British Empire that have a similar background : USA, Canada, Ireland, Britain, South Africa, Australia and New Zealand - there is no force that naturally pulls them together, instead there are only forces that pull them apart.

As it happens, I am beginning to read


This is a review of 'The Everlasting Empire: The Political Culture of Ancient China and Its Imperial Legacy' Looks good for part of an answer to the questions.

Yuri Pines
Jingbin Wang

Yuri Pines. The Everlasting Empire: The Political Culture of Ancient China and Its Imperial Legacy. Princeton: Princeton University Press, 2012. 248 pp. $39.50 (cloth), ISBN 978-0-691-13495-6.

Thank you, that looks like a fascinating book. Will order.

I've read a number of books that touch on this subject, but none have been satisfactory. All oversimplified (unsurprising for a 2000 year old civilization). People seem to be searching for single causal factors (Rice & benefits of taming hydrology, external Horse Nomad threats, Confucian bureaucracy) when it seems to be a complex interplay of winner take all internal topography, a shared written language, self replicating elites that recognize the benefits of harmony (times of warfare are bad times, compare to Europe's near incessant wars), Confucian ideals of deference to authority and an educated competent bureaucracy.

Particularly interesting in that review is Pine's point about the Chinese Emperor's total power in theory vs. practice. Theoretical omnipotence vs practical non-intervention offers a partial explanation for a number of oddities in Chinese history. Most of the Western analysis I've read has taken the theory of omnipotent emperor at face value and little time has been spent analyzing the practical or effective limitations on power. Perhaps the lack of reward for innovations that accumulated into the industrial revolution in Europe was partially due to weak protections and rewards for innovators and not simply a resistance to change from an omnipotent state? That seems to be the crucial question, why didn't the changes accumulate? Why did knowledge frequently get lost? That seems to be a different question than a simple tendency to reject foreign influence (as implied by many Western histories). See Landes Celestial Empire: Stasis and Retreat, chpt 21 from Wealth and Poverty of Nations. This excellent book written at Japan’s apparent transcendence completely missed any signs of the incipient Chinese transformation. Interesting in hindsight.

Might be useful to know more about the customary system of taxation. Did it tend to reward unification?

The Human Web (McNeill & McNeill) seems to imply that the early unification of China combined with Confucian culture/ideology/shared written language effectively imprinted central "China" with the benefits of civilization e.g. unity with limited external influence via the silk road and external trading. Attempts at conquest outside the natural boundaries were made (e.g. the horse nomad areas to the North), but it was more cost effective to hire them & trade with them instead. Balance of power which periodically failed.

.. the answer is surely geography and the fact that the many peninsulas and Island(s) of Europe gave birth to numerous civilisations (Greece/Rome/Iberia (twice one Islamic one Christian), UK, Scandinavians and it is the interplay between these and the other European polities like France and Germany that gives rise to the difference.
Perhaps there is an optimum size of state and currently that is the European nation state of 50/70 million. China seems to have a perpetual pattern of Dynastic rise and fall with the Communist party being the modern equivalent of a Dynasty.

Bear in mind that the map shows ancient China at its maximum extent under the Han, and the nation only achieved that size (more or less) one other time in pre-modern era, under the Tang-- unless one counts Mongol China under the Yuan Khans-- but that's really about the Mongols, not the Chinese. For much of its history before the Mongols China consisted only of the core Han areas in the north central part of the country. And often enough it was not a unified state but was divided into anywhere from two to a dozen political states.

Another observation-

My old international politics professor was in China during the outbreak of WWII. Back in the Eighties he used to observe that the Chinese were unlikely Communists as everywhere outside of China they were the merchants and traders. Professor Cowen's opening post doesn't appear to quantify the numbers or scale of the Chinese trading empire. That should probably not be overlooked.

p.s. Being Dutch, that same professor was always nattering on about Trade and Finance. Perhaps one rewarding angle on this issue could be the nature of credit & finance in China. (He was also the one who would always suggest looking at the structure of taxation when trying to understand trade and political social cohesion.)

I was going to recommend his book, but it's gotten kind of pricey: https://www.amazon.com/History-Credit-Western-Gastmann-Albert/dp/B008AUEEXG/ref=sr_1_fkmr0_1?s=books&ie=UTF8&qid=1494892944&sr=1-1-fkmr0&keywords=albert+gastman+credit+trade

That's weird. How can the same book be selling for almost three grand and $45 at the same time on Amazon? Here it is at a reasonable price:

"The first unambiguously modern state, Mr Fukuyama believes, was the Qin dynasty in China, founded in 221BC. Many of the control mechanisms perfected by the Qin had developed during the preceding 500 years or so of the Eastern Zhou dynasty when a host of small warring states across China began to coalesce. Such elements included a merit-based (non-aristocratic) military leadership combined with mass conscription, sophisticated taxation systems and a bureaucracy recruited from a permanent administrative cadre selected on the basis of ability rather than family connection. The Qin simply went much further, assaulting every section of society in its remorseless attempt to establish a form of protototalitarian dictatorship."


Yunnan became a stable part of China because of the movement of almost 1,000,000 people into the area by the government for the express purpose of unity.

I couldn't get enough of your post. I'm not one to usually
comment, but I felt really motivated to let you
know. I even ended up sharing this on my blog!

Comments for this post are closed