Geography and Chinese History – The Fractured Land Hypothesis

Occasionally I come across attempts to explain the broad course of Chinese history in reference to China’s geography. These arguments tend to focus on the unified empires of Chinese history. Always contrasting Chinese history with the European experience, they suggest that China’s political unity and Europe’s perpetual disunity are reflections of the unbroken terrain of the first and the disparate geography of the second. Two prominent examples can be found in Paul Kennedy’s Rise and Fall of the Great Powers: Economic Powers and Military Change, 1500-2000 and Jared Diamond’s Guns, Germs, and Steel: The Fate of Human Society. Quoting Professor Kennedy first: 

For this political diversity Europe had largely to thank its geography. There were no enormous plains over which an empire of horsemen cold impose its swift dominion; nor were there any broad and fertile river zones like those around the Ganges, Nile,Tigris and Euphrates, Yellow, and Yangtze, providing food for masses of toiling and easily conquerable peasants. Europe’s landscape was much more fractured, with mountain ranges and large forest separating the scattered population centers in the valleys; and its climate altered considerably from north to south and west to east. This had a number of important consequences. For a start, it both made difficult the establishment of unified control, even by a powerful and determined warlord, and minimized the possibility that the continent could be overrun by an external force like the Mongol hordes. 

Conversely, this variegated landscape encouraged the growth, and the continued existence, of decentralized power, with local kingdoms and marcher lordships and highland clans and lowland town confederations making a political map of Europe drawn at any time after the fall of Rome lool like a patchwork quilt. The patterns on that quilt might vary from century to century, but no single color could ever be used to denote a unified empire. 

Europe’s differentiated climate led to differentiated products, suitable for exchange; and in time, as market relations developed, they were transported along the rivers or the pathways which cut through the forests between one area of settlement and the next. Probably the most important characteristic of this commerce was that it consisted primarily of bulk products timber, grain, wine, wool, herrings, and so on catering to the rising population of the fifteenth century Europe, rather than the luxuries carried on the oriental caravans. Here again geography played a crucial role, for water transport of these goods was so much more economical and Europe possessed many navigable rivers. Being surrounded by seas was a further incentive to the vital ship building industry, and the later Middle Ages a flourishing maritime commerce was being carried out between the Baltic, the North Sea, the Mediterranean and the Black Sea.” [1]

And here is Mr. Diamond’s version of the same thesis:

China’s frequent unity and Europe’s perpetual disunity both have a long history. The most productive areas of modern China were politically joined for the first time in 221 BC and have remained so for most of the time since then. China has had a single writing system from the beginnings of literacy, a single dominant language for a long time, and substantial cultural unity for 2,000 years. In contrast, Europe has never come close to political unification. Still splintered into 1,000 independent statelets in the 14th century, into 500 statelets in 1500, reduced to a minimum of 25 states in the 1980s, these have increased again to nearly 40 at the time of writing. Europe still has 45 languages, and even greater cultural diversity. The disagreements that continue to frustrate even modest attempts at European unification through the EU are symptomatic of Europe’s ingrained commitment to disunity. 

Map of the coastlines of China and Europe, 
drawn to the same scale.

Source: Jared Diamond, 
Guns, Germs and Steel, p. 415.

To understand China’s loss of political and technological pre-eminence to Europe we must understand why China has remained united, and Europe fragmented. Maps provide the clue. Europe has a highly indented coastline, with five large peninsulas that are island-like in their isolation, and all of which evolved independent languages, ethnic groups and governments: Greece, Italy, Iberia, Denmark and Norway/Sweden. China’s coastline is much smoother, and only the nearby Korean peninsula attained separate importance. Europe has two islands (Britain and Ireland) large enough to assert their political independence and maintain their own languages and ethnicities. Indeed one is big and close enough to become a major independent European power. China’s two largest islands, Taiwan and Hainan, are each less than half the area of Ireland; neither was an independent power until Taiwan’s emergence in recent decades. Japan’s geographic distance from the Asian mainland kept it much more isolated politically than Britain has been from main-land Europe. Europe is carved up into independent linguistic, ethnic and political units by high mountains (the Alps, Pyrenees, Carpathians and Norwegian border mountains), while China’s mountains east of the Tibetan plateau are less formidable barriers. China’s heartland is bound together by two long navigable river systems in rich alluvial valleys (the Yangtze and Yellow rivers), and it is joined from north to south by relatively easy con-nections between these two river systems. As a result, China very early became dominated by two huge geographic core areas of high productivity, which gradually fused into one. Europe’s two biggest rivers, the Rhine and Danube, are shorter and connect less of Europe. Unlike China, Europe has many scattered “core” areas, none big enough to dominate the others for long, and each the centre of an independent state.” [2]

Both historians contend that Europe’s fractured physical geography led to its fractured political divisions, and that the lack of such clear geographic divisions explains China’s political unity. For simplicity’s sake let us call this the “Fractured Lands Hypothesis.

I once found this hypothesis very convincing. Then I began to study Chinese history. The more I learn about China’s past the less convincing this argument is. A few of its serious problems are outlined below.

NOTE: While preparing materials for this post I came across Peter Turchin’s note Why Europe is Not China.” I was surprised to find that he agrees with my assessment and makes many of the same points I planned to make in here using the same historical examples I intended to use! (Up to an including the reference to the film Red Cliff.) There is no need to reproduce his material here so please read his post before continuing on with this one. 

The Fractured Land Hypothesis’ explanation of Chinese unity fails on two counts:

#1 – China’s geography is not homogenous – it is just as fractured as the rest of the world.

The idea that China’s geography is unhindered by major geographic, topographic, or ecological barriers is faulty. It is often based on maps like the one used by Jared Diamond above. The map cleverly conceals China’s geographic diversity by only distinguishing the land from the sea. All of China is depicted in homogenous white. Alas, there is more to a continent than its coast line. A different picture emerges when we look at China from a topographic perspective:

A Topographical Map of Eastern China.

Source: Adapted from World of Maps
 Also see this topographical relief map viewing China from the sea, 

On the basis of topography alone we can divide Eastern China into several discrete areas: the first is the Central Plain, following along the lower reachers of the Huang, comprising modern Southern Shanxi and Hebei, and North Eastern Henan, and Shandong in its entirety. Ancient sources often called this area Guandong (关东), or “East of the Passes.” Directly to its West, over the Taihang Mountains, was Guanzhong (关中), or “Within the Passes,” comprising parts of Western Henan and most of Shaanxi today. The elevation is higher here than in the Central Plain; its population is centered on the Wei River Valley, easily seen on the topographical map above.

During the classical age of China’s history (c. 771 BC-220 AD) these two regions were the major “centers” of the Chinese world. Passage from one center to another was difficult, limited to a few mountain passes. The Hangu Pass was the most important of these; firm control of the pass was a strategic priority for every warlord claiming either region. It has been the site of dozens of battles since antiquity.

The Hangu Pass (函谷關), dividing Guanzhong and Guandong.

Source: ‘Hovering Month.’ “Hangu Pass Pictures.” Posted on Mafengwo Travel Guides, Henan Travel section. (Chinese language source).

South of Guanzhong is the Sichuan Basin. 100 million people call the modern states of Sichuan and Chongqing home today; before the state of Qin (c. 800-221 BC) conquered the region its inhabitants were the Ba and Shu peoples. Ringed by mountains, the Sichuan Basin has only a few points of entry: in the North through Hanzhong, and in the East by way of the Yangtze. The Qin conquered Ba and Shu before beginning their rampage across the Central Plain (some have suggested that their conquest of the Sichuan Basin gave them the resource boost they needed to do so [3]), but over the course of subsequent Chinese history imperial control over the Basin often faltered, with dynasties abandoning the region to indigenous tribes and kingdoms in times of great stress or collapse.

Annual precipitation in China. Notice the difference between the Yangtze and Huang River Plain.

Image Source:, “Climate Map – Annual Precipitation.”

The lower reaches of the Yangtze river form another major region in Eastern China. Like the Central Plain, the Yangtze region is an alluvial plain dominated by the river which runs through it. Partly on the basis of their similar topographic profile, some geographers will throw the Yangtze and Central Plain together in one large “Northern Plain” region. I think this is unwise. The topography of the two regions is similar, but the ecology of each are very different. These differences ultimately derive from climate. The Yangtze basin is both hotter and wetter than the Central Plain; every year twice the amount of rain falls into the lower reaches of Yangtze than falls into the lower reaches of the Huang. The ecological effects of humidity, precipitation, and temperature can be seen in the types of crops farmers are able to grow in each region:

China’s most productive rice and wheat growing regions.

Source: USDA Office of Chief Economist Country Files

The ancient Chinese certainly considered the two regions distinct. During the Spring and Autumn (771-476 BC) and Warring States (475-221 BC) periods,  Chinese literati and statesmen from Guandong believed that men from Chu, Yue, and Wu –  all states founded along along the Yangtze plain – were semi-barbarians that were not fully a part the Chinese cultural sphere. Waves of Han immigrants over the centuries would change this perception; by the time of the Song Dynasty (960-1279) China’s capital rested on the banks of the Yangtze.  

Karst hill country in Guilin, Guangxi.
Source: Wikipedia.

The final region of Eastern China stretches from the Yangtze River Plain in the North to Vietnam in the South. This is the region often called the “Southern Hills” by geographers. Of all the regions of “China proper” the Southern Hills took the longest to incorporate into both the Chinese cultural and political sphere. The early dynasties did not bother trying to conquer most of the region (see this map of the Han Dynasty) and many parts of Guangxi and Guangdong were not incorporated until the Qing Dynasty (1644-1912). Today the more remote parts of Fujian, Zhejiang, Guangdong, and Guanxi still are dotted with ethnic minority villages. As with the Basques of the Iberian Peninsula, the region’s rough geography allowed ethnic minorities to persist in large numbers despite deliberate Sinicization efforts  on the part of the Chinese state. 

Mountain valley in the countryside of Fujian. Source: 
China Travel Photo of the Day: Earthen Fortress of Fujian.” 27 October 2011.

This brings us to the second point.

#2 – China has often been fractured politically along geographic lines.

The empire, long divided, must unite; long united, must divide. Thus it has ever been.[4] So begins Luo Guanzhong’s famous historical epic Three Kingdoms. One of China’s “Four Great Classical Novels,” the stories contained inside are some of the most beloved and well known of the entire East Asian intellectual tradition. It both influenced and captured Chinese attitudes towards their history as the inevitable cycle between dynasty and disarray.

Luo Guanzhong had the benefit of hindsight. He wrote during the early Ming Dynasty (1388-1664), when China was strong and unified. He knew that China would unite once again. The historical characters whose adventures he embellished did not have this foresight. They were the first to live through an imperial collapse. By the time the disunity of the”Six Dynasties” ended Chinese civilization had passed more years divided than united!   

China’s fractured geography played a major role in China’s periods of political disunity. The Three Kingdoms Luo Guanzhong wrote about are a case in point. With the collapse of the Han Dynasty the three warlords who arose from the wreckage built their kingdoms around the basic geographic barriers of East China. Liu Bei, leader of the kingdom of Shu, made the Sichuan Basin the center of his operations; the Sun clan claimed the kingdom of Wu, protected by the Yangtze, and Guanzhong and Guandong were captured by the brilliant maneuvers of Cao Cao, founder of the Wei kingdom. Despite his brilliance, Cao Cao had to give up his hopes of universal domination when the fleet he pulled together to invade Wu was burned by fireships at the Battle of Red Cliffs. Without his fleet he had no way to get at his enemies; the mountains of Sichuan and the waters of the Yangtze were barriers too large for the conquerer to overcome.  

The Three Kingdoms of storied fame.
Source: Wikimedia. 

Rivers would prove the end of many conquerers. Cao Cao was defeated on the banks of the Yangtze; the Southern Song Dynasty made their stand against the Liao, Jin, and Mongols on the banks of the much smaller Huai:

The Song Dynasty (Green) retreated behind the Huai River when nomads from Manchuria overwhelmed their forces on the Central Plain. 

Image Source: GeoNova. “Song and Jin Dynasties.” Accessed 29 April 2013. 

Mountain ranges also played their part. The much contested Hangu Pass has already been mentioned; the hills and mountains of South China were no less important, ensuring that the region took millennia to conquer and incorporate. The ecological gap between the North and South only made the task harder. A great deal of historical research has shown the difficulty migrant and conquering people have establishing themselves or stamping out local cultures in lands ecologically different than their own [5]; the Chinese of the Northern plain found the challenge no easier than the rest of humanity. The numerous Chinese dialects of the Southern Hills region suggests how limited this “universal” cultural assimilation really was. 

A map of Sinitic languages in China – notice that most are South of the Yangtze.

Source: Wikimedia.

All of these things present serious problems with the Fractured Land Hypothesis. A close examination of the geography of East Asia suggests that there is no geographic feature capable of explaining the divergent paths of European powers like Germany, France, and the Netherlands that cannot be found in China. Chinese unity did not come because of its geography. It came in spite of it.  


[1] Paul Kennedy. Rise and Fall of the Great Powers: Economic Powers and Military Change, 1500-2000 (New York: Oxford University Press). 1989. p. 17-19

[2] “Disunity is Strength.” Prospect Magazine. 20 July 1997. A slightly adapted version of these two paragraphs is contained in the epilogue to Guns, Germs, and Steel. 

[3] The most notable argument of this type is found in Steven Sage. Ancient Sichuan and the Unification of China. (Albany: SUNY Press). 1992

[4] Luo Guanzhong. Three Kingdoms, A Historical Novel. Vol. I  Trans. Robert Moss. (Berkely: University of California Press). 1991. p.1

[5] By far the best (and entertaining) introduction to this research is Alfred Crosby. Ecological Imperialism: The Biological Expansion of Europe, 900-1900. (Cambridge: Cambridge University press). 1992. 

Leave a Reply to Unknown Cancel reply


Good stuff. Both of those authors are largely focused on the West and its development. I would still say, however, that Europe was definitely more fractured both geographically and politically than China. You have effectively broken it down into 3 pieces, but China does still have a clear history of centralization that western Europe didn't see once Rome fell. Sure, the Holy Romans might have tried, but Europe was largely much more fractured due a large number of factors, geography being one of them. China might have had divisions, but these were nowhere near the size of Europe's.

The point of this post was not to claim that Europe is less politically fragmented than China is. The last 1000 years of history prove that this is not true.

But an idea of interest: lets wind the clock back to 500 AD. Which place is the more fragmented of the two? The answer would be East Asia – and not only that, but someone living in 500 could truthfully claim that Europe had a longer history of unified empire than China!

This is part of the reason I am so skeptical of the Fractured Land Hypothesis. It explains things between 700 BC-600 AD poorly. Ecology and climate did change during those times, but topography and coast lines did not. If Europe's "fragmented" geography was not conducive to unified Empire then Rome should never have existed. (And as Peter Turchin points out, the Med. Sea unifies its world in a way few geographic features of the Earth can). If China's geography inevitably leads to empire then the hundreds of polities of the Spring and Autumn, dozens kingdoms of the Warring States, and the divided reigns of the Six Dynasties and Southern Song should never have happened.

Chinese history suffers from an extreme case of hindsight bias. We treat what did happen as inevitable – but I suspect that if the tape were wound back and played again a different show would play. It is significant that China's periods of 'disunity' grew smaller and smaller as time wore on – it was not physical geography that changed, but the institutional and cultural geography of the land, each dynasty making it easier for the next to arrive. Things like the Grand Canal are physical expressions of this. China's geography did not create a unified polity – a unified polity literally unified China's geography!

The whole process is so much more contingent than most Chinese would admit. The first unifying dynasty – the Qin, in the 220s – lasted only one generation. It is worth remembering that the peoples they conquered considered themselves just as distinct as most Europeans would have in the High Middle Ages. The Qin conquered the lot, but lacked the institutions they needed to hold the whole thing together. They were pulled apart by the empire they conquered.

I often wonder what would have happened to China if the Han Dynasty had followed in their foot steps. The early Han certainly did not take their position for granted – you cannot read Sima Qian's Grand Records without appreciating how insecure Han Gaozu and the other early dynasts felt. Even their liberal, decentralized political order could not stop regional rebellion from occurring. Had the Rebellion of the Seven Kingdoms succeeded, what would have happened? Would a unifier emerged? Would China exist as we now know it?

I cannot say. But today's Chinese are in more debt to the structures set up by the Han than is readily recognized.

I think reducing the scale gives a clue. How the whole was unified might be best considered by seeing how the component three kingdoms came to be unified and beyond that how the seven kingdoms came to be unified into the three kingdoms etc.

I think the standard pattern is human societies build up until they're stable as a unit at a kind of maximum scale which by default is around the level of tribe / city-state level contained within some natural geographical limit.

This patchwork of polities is relatively easy to conquer piecemeal because a conqueror can bite off one polity at a time. The polity that begins the process needs a significant initial advantage but after that the process snowballs as tributary soldiers from each conquered polity are added to the conqueror's forces.

If this was happening with multiple starting polities at the same time then they would expand until they eventually bumped into each other and there were no more independent polities to be easy pickings.

Some of the initial advantages one of the conquering polities might have are
– good leader
– gold or silver mine
– military advantage e.g. access to horse archers
– fertile river valley
and some of these advantages are more consistent over time than others.

I think the Persian empire and Alexander's war against it exemplifies this idea – an empire made up of a collection of polities who are both small enough to be relatively easy to conquer and which can be just as easily be coerced into a Macedonian empire as a Persian one and which will separate back into independence as soon as the conquering polity's military advantage disappears.

Of those advantages the most sure and certain long-term is being centred on a particularly fertile river valley. Whatever else happens in the interim with the rise and fall of great leaders, silver mines and horse archers (among others) a polity centred on a particularly fertile river valley is likely to keep recurring as a conquering polity simply because of manpower and money.

So it seems to me you could probably pick the *eventual* geographical winners of the first stage of Chinese unification e.g. the seven kingdoms stage, from the get go even if might take a thousand years or two – or at least be able to choose the list of possible contenders from within each geographical unit.

If you apply that idea to Europe then you don't really have any polities with the consistent advantage of a core region built around a very fertile river valley until after the heavy plow is introduced. Once that happens parts of Europe get a re-run of the Chinese experience e.g. France is constructed by bolting Bretons, Gascons and Burgundians onto the core French polity centred in the Paris basin.

Before the heavy plow Rome was the closest to that but its advantage was based on captured fertility like Egypt rather than its own core.

So it seems to me the big difference between Europe and China is China had core regions centred on high fertility river valleys much earlier than Europe. Various unifying attempts were made in both China and Europe based on shorter-term advantages like good leaders but its only the ones built on core high fertility regions that eventually stick.

When high density agriculture finally made it to Northern Europe a process similar to China's began.

The second difference is why did it stop. Why didn't Rhine, Seine, Thames, Rhone, Loire centred polities carry on swallowing each other and forming ever bigger polities like China did?

Perhaps it simply had less time to do so or perhaps it's something to do with hbdchick's idea and outbreeding created a larger size of default polity that was harder to swallow e.g. three city-state sized polities might be easy to individually swallow but three city-states who saw themselves as part of a nation is more difficult.


"So it seems to me the big difference between Europe and China is China had core regions centred on high fertility river valleys much earlier than Europe. Various unifying attempts were made in both China and Europe based on shorter-term advantages like good leaders but its only the ones built on core high fertility regions that eventually stick."

Being built around a high-fertility river valley does not seem to provide any benefits not provided by building up around a high-fertility coastline. In fact, as most folks since Mr. Thayer Mahan have noted, the Roman empire existed because of that human highway we call the Mediterranean.

The second difference is why did it stop. Why didn't Rhine, Seine, Thames, Rhone, Loire centred polities carry on swallowing each other and forming ever bigger polities like China did?

Now that is the truly interesting question. It did not stop in the classical world, of course; Rome conquered the lot. But after, why did it universal empire fail and fail again?

Victoria Hui Tinbor's War and State Formation in Ancient China and Early Modern Europe is a good book on this topic (explicitly comparing the ancient Chinese situation with the Early Modern European one, spending the better part of 60 pages devoted to your question) and an excellent place to start. But be warned: the book is a wonderful piece of scholarship. It is also utterly unreadable.

Had the empire of Carlemagne not been divided and had the Franks not effectively given Rome away to the Papal States, and had the Franks proceded to annexe the nearby 'OTL' political centres of Britain and Iberia, maybe European history would have been much more like that of China.

For a further rather belated response to this blog post, in the form of a paper by Mark Koyama, see:"Fractured-Land and the Puzzle of Political Unification and Fragmentation" – January 2019"Fractured-Land and the Puzzle of Political Unification and Fragmentation" – July 2020 version of the same

Effectively the paper is based on countering the claim that Europe was not more "fractured" than China by establishing that:

A) emphasizing the role of heavy forest cover in disrupting the Central European plain.

B) using a spatially explicit "Civilization" / "4x" style grid model to model state formation, to argue that while China may have more ruggedness overall, the position of its ruggedness was less situated to inhibit state formation.

I'm still not totally sure I find it convincing. they find that their model recreates dynamics, but I wonder if it isn't overfitting.

It also has some errors in the Near East and South Asia, which are fairly significant, as a cost of optimizing for Europe vs China. A model that doesn't explain much about the formation of states in the first region in which states formed and preceded those in China by literally *thousands* of years is not much cop as a model.

And also wonder if forest cover is really as endogenous as they claim – isn't it perhaps an effect as much as a cause (unlike mountains, rivers, etc, which cannot be moved)? Perhaps expanding Celts would have worked harder to chop down those forests with iron tools, if the Romans hadn't got to them first, and then created a nice opening for Germanics.

I have read both books, Kennedy's and Diamond's, only that +15 years have passed between the first read (Kennedy) and the last (Diamond). Today, they rest together, side by side, in my bookshelf. By college studies, I'm a human geographer myself, so I'm unconsciously biased towards accepting the Geographic factor.

I might look at China geography in a different way than explained above, while certainly it can be considered internally as fragmented and interrupted by natural accidents as Europe, it also can be considered as a massive chunk of land relatively compact and more/less circular, when you look at it against the context that surrounds it: the massive Pacific Ocean to the East, with relatively few accidents (islands and such) in the proximity of Chinese shores, the massive mountain ranges and deserts to the West and Northwest, hot humid tropical jungles to the South and the cold void of Siberia to the North.

Within such strong external barriers, the Chinese diverse immensity can be also considered as a vast but yet homogeneous territory, in the same way that we in the US look at our country: while recognizing the physical differences between hot/humid/sea level South Florida and alpine mountain ranges in Colorado.

Perhaps the case of Japan can be seen as something in between Europe and China. Thousand years ago it looked as fragmented, or more, than Europe. According to Diamond book, Chinese sources described Japan as the land of 100 small kingdoms always warring each other. Being Japan more common landscape the mountain ranges and thick forests, the political fragmentation can be considered as heavily influenced, if not determined, by the rugged terrain.

However, Geography also confined the Japanese peoples to their own islands, they didn't have anywhere to go, so their fate was to unite in one nation, one way or the other, process that ended in the second half of XIX century, curiously the same historic period, approximately, when US and Germany were finally solidly unified into the political entities they are these days.

Perhaps China process to reach that stage, was similar to Japan's, only that happened in a continental scenario and 2,000 years before.