How Nomad States Form and Develop

By Ruilin Shi


How nomadic states form has always been a debatable topic. I would like to present some scholars’ work about the formation of the Xiongnu (third century BCE to third century CE), and explain where I agree and disagree, particularly with Owen Lattimore’s theory of state formation in the northern Chinese frontier zone.


  1. How have Chinese scholars viewed the problem, and how have their views changed?

The first work concerning how the Xiongnu formed is the Shiji (Records of the Grand Historian of China), by Sima Qian during the Han dynasty. He thought that “the ancestor of Hsiung-nu was a descendant of the rulers of Hsia dynasty by the name of Ch’un-wei.”[1] This is a very naive guess, and does not rely on much evidence. However, because of the special position of Sima Qian in Chinese culture, almost all scholars of imperial China down to the later Qing dynasty followed and developed this idea.

The first influential scholar who amended this idea was Wang Guowei, in the early 20th century. He believed the Xiongnu were descended from other barbarians mentioned in Shiji, such as the Shanrong, Xianyun and Hunzhou.[2] This idea was common in Chinese studies of the Xiongnu in the 20th century, and most scholars agreed with this theory. Another classic work of Chinese Xiongnu study, Xiongnu Tongshi (General History of Xiongnu) developed this idea, contending that the Xiongnu formed from a number of different of nomad groups from the Yinshan area and absorbed some nomads from the Yellow River area. The evidence Lin Han (author of Xiongnu Tongshi) used in this study is quite interesting. He mentions that a surname used by the Xiongnu is the Chinese character for their original nomad group, similar to the way the surname Han was taken from the social group they used to belong to before they were mixed with another group.[3] I think Chinese and Western scholars should think more about this clue.

Having made contact with Xiongnu studies by Russian and Mongolian scholars and acquired a lot of new archaeological evidence since the 1980s, more and more scholars nowadays are interested in the theory that the Xiongnu formed on the far northern steppe. Artifacts show the huge cultural differences between different Xiongnu regions. Most of those scholars believe that there was a powerful nomadic culture in the far northern Mongolian steppe during the Shang dynasty, which absorbed nomadic groups like the Xianyun and the Guifang in the period before Qin dynasty, and finally became an ethnic group in the later Warring States period.


  1. Western ideas about the formation of the Xiongnu

There are two very popular ideas about the formation of the Xiongnu in the west: Owen Lattimore’s theory and Thomas Barfield’s theory. These two ideas are almost opposite to one another with respect to the relation between China and the development of the Xiongnu. Barfield stated that the Xiongnu were the “shadow empire” of a strong China.[4] This is a very interesting idea, but it is too simplistic, and lacks sufficient evidence over the long term. In contrast, Turchin points out that steppe states did come into existence when China was weak and divided. Moreover, Turchin states that nomadic culture in the steppe developed parallel with agrarian culture in China; unfortunately, there is not enough evidence to show that this was happening on the steppe in the Xiongnu period.[5]

Catrin Kost’s theory combines independent and dependent aspects of development. Kost has something in common with Chinese scholars regarding the formation of the Xiongnu: they agree that part of the Xiongnu were gathered by the nomadic group that used to live in the Yellow River area and moved to the north when they were attacked by the powerful Chinese states to the south. Kost states that the nomads formed alliances to enable them to attack their neighbors more effectively.[6] This idea seems popular among Western scholars.

What interests me most is the cycle of nomad rule posited by Lattimore. This cycle can be briefly described in four steps: powerful nomads appear; these nomads divide; some of them denomadize; conflicts arise among nomad groups.[7] This statement is fine in that it describes the appearance and disappearance of nomad states in a very generalizable way, but it is too abstract and does not show the details of how the Xiongnu formed as a specific nomadic group. I would like to follow Lattimore’s idea and restate the nomad cycle in my own way.


  1. My theory of nomad formations in the China–Mongolia region

3.1 The significance of the Great Wall as a geographical boundary

In the works of Barfield and some other scholars, they have noticed that Qin state was so strong that it did not need to build the Great Wall to defend itself. Moreover, they point out the cultural significance of the Great Wall and how it was used as the boundary of Chinese or Central Plain culture. Profoundly, the wall has significance as a border not only with respect to different cultures, but also in an economic, ecological and military sense. Economy always has a strong relation to the vegetation in a specific area. If we compare the position of the Great Wall to the Vegetation Map of China, we can see that the Great Wall was built near the margin of the warm temperate forest, which was the northernmost vegetation type suitable for agriculture. The different life styles in the different zones may underlie the differences in culture.

(Map: )

Moreover, the vegetation of different areas also conditions the advantages of different military units on either side of the border, and they will have to spend more if they want to control the land on the other side. If cost exceeds benefit, people will not go to the expense. Different vegetation types correspond to natural military boundaries of different cultures. Groups near the boundary will spontaneously change from seminomadic into nomads or agriculturalists. Unified countries near the boundary may divide into several countries, some choosing agricultural life and some choosing nomadism. In the Warring States period, Zhongshan changed from nomadic to agricultural.[8] Zhao, an important Han country, change their clothing and life style to become nomads, as recorded in the Hufu Qishe in Chinese history books. These two countries show the two types of change on the Great Wall margin.


3.1.1 The significance of the Great Wall as a chronological boundary

Before the Qin dynasty, Zhao, Yan and Jin expanded their territory to the Great Wall area. Chinese culture had never expanded into areas that were not suitable for agriculture, which means that before the later Warring States period, after Chinese expelled nomads or absorbed semi-nomads and took control of land, the nomads could still move north to where the land was still suitable for agriculture, and mix with another nomads or semi-nomads. The fact is that when nomads and semi-nomads lived inside of the Great Wall area, they never formed a powerful alliance that could fend off an attack from strong Chinese countries. This could be because of the great differences among those nomads. In response to the weakness of their nomadic neighbors, strong Chinese countries would attack the nomads again and again to get more land for agriculture.

But the situation is quite different when Chinese countries reach the Great Wall area. These countries were not very interested in the steppe, which is not suitable for agriculture, and where it is more and more difficult to expel nomads who can go deeper and deeper into the steppe. The situation of the nomads who were expelled is also different. When they moved north, they did not have a choice of life style; they had to become “poor” and “pure” nomads.[9] Because of the position of the Gobi Desert, sometimes they had to move far to the north, to the steppe beyond the desert. This is why I think the building of the Great Wall or the time when Chinese countries reached the Great Wall area is very important for periodization.


3.2 The cycle of nomad rule

Based on the significance of the Great Wall as both a geographical and a chronological boundary, I would divide the cycle of nomad rule into a Pre Great Wall Period and a Great Wall Period.


3.2.1 The Pre Great Wall Period

In this period, there were many nomads living inside of the future Great Wall zone, in North China and in the far northern steppe. But they were largely separated and did not have a very powerful state. We start the cycle when Chinese defeat a nomadic alliance or a single clan:

  1. Nomads were defeated by Chinese.
  2. The clan was divided into two parts: one part chose the agricultural life and became Chinese, and the other part chose to run away and move to the northern area, and mixed with other clans.
  3. The new clan chose their life style, and some of them favored agricultural life and become semi-nomadic people.
  4. This caused weakness in the military power of the clan. Start again at 1.


Repeating the cycle several times might form a powerful alliance which can defeat some weak Chinese states, but it would still not be powerful enough to defeat strong Chinese states. They would be defeated by the Chinese and separate into small clans again. The Zhunwang Rangyi (‘Respect the king of Zhou and expel the barbarian’) War led by Jin and Qi from 680 to 580 BCE is quite a good example to show the cycle[10]: the Jin expelled several different clans of the Di and Rong, and some became followers of Jin and married with Jin rulers (both Jin Xigong’s and Jin Wengong’s mothers were the daughters of the chiefs of Di and Rong). After 30 years of war, several clans of Di gathered into a powerful alliance, and this alliance was able to conquer some small Chinese states. Thus began a long period of war with Jin and Qi, which ended in defeat; the alliance separated, and some nomads moved to the north and some became followers of Jin and Qi.

This cycle is quite simple from the perspective of the Chinese. It is a long-term, and never-ending war between Chinese and different kinds of barbarians. Sometimes they met nomads (Xiao Yu ding) and sometimes they met semi-nomads (Shiji: Zhou Muwang). The great poetry collection of the Zhou dynasty, Shi Jin, contains a huge number of poems about the war between Zhou and the Hu, Di, and Rong, which is evidence of the never-ending war.


3.2.2 The Great Wall Period

This period is much more complex than the Pre Great Wall Period. Again we start with Chinese defeating clans or an alliance near the Great Wall area:

  1. Chinese defeat nomadic or semi-nomadic people near the Great Wall area.
  2. Some of them became followers of the Chinese, and another part moved to the northern Yin Shan area or the far northern steppe.
  3. Since Chinese military force could not reach the far northern steppe, steps 1 and 2 repeat several times; the nomads in the steppe can develop without attacking and gather more and more people from different clans and life styles.
  4. Because of their common life style and some interest in trade, the northern steppe nomads would form a powerful alliance.
  5. Some of these alliances expanded their territory to the south, and met Chinese countries. Because of significance of this boundary discussed in 3.1, these alliances would be in stalemate with Chinese countries. Even if they were sometimes expelled by the Chinese, the far northern steppe nomads could still develop.
  6. The above step repeats several times. After a weak clan or alliance was expelled to the far north, and the far north steppe nomads developed sufficiently, then this alliance would try to gain more allies among clans of far north to restore their power. At this time, if there was an ambitious chief, he would try to build a powerful state and rule all the people in the north. This is what Modun of Xiongnu, Taba Gui of Northern Wei (Xian Bei), and Elterish of the Second Turk Empire did.
  7. Then the nomad state would expand to the south, to the the Great Wall area and even further south.
  8. The next step is as Lattimore states: division of the society, denomadization, conflict, and defeat and expulsion by Chinese, and separation into small clans.


The Xiongnu, the Wusun, and the Donghu were the first alliances generated in this cycle, and the Xiongnu became the first nomadic state in the Great Wall Period. After the collapse of the Xiongnu in the later Han dynasty, there was no powerful alliance to the north of China until the Jin dynasty (265-420 CE). Later, small nomadic powers established within the Jin dynasty, and the Qian Qin, the Northern Wei, and the Ruan Ruan developed one by one in the north.

In this cycle, the far northern steppe is very important. It is the petri dish and hometown of the nomadic culture. The “poor but pure” nomads always lived in the area, developing their culture without attacking the Chinese and absorbing many different clans from the south, some of whom were nomads and others semi-nomads, workers, and merchants. These new migrations were the source of the development of the far northern steppe. Di Cosmo discusses the problem of comparing nomadic and Chinese ideas about the colors of the four cardinal directions and the administrative meaning of “left” and “right.” He challenged the idea that these ideas among the Xiongnu show that they got their culture from Chinese.[11] I agree with Di Cosmo; I do not believe the Xiongnu got these ideas from China directly. They may have developed the same system during the 1000 years they were in the far north, or they may have got it from the people who moved to the far north.

The far northern steppe was like a tree, and the nomads were like dew on the tree: when enough dew gathers, it drops to the ground. Some of the drops are absorbed by the ground, while others evaporate and become dew again the next day. Every drop was a developing nomadic group.


[1] Burton Watson (trans.), Records of the Grand Historian of China: The Shih Chi of Ssu-ma Ch’ien, vol.2 (New York: Columbia University Press, 1961), 155.

[2]Ma Changshou, Beidi yu Xiongnu (Shanlian Bookstore, 1962), 3.

[3] Li Han, Xiongnu Tongshi, (Renmin Press, 1983), 2-4.

[4] Thomas Barfield, The Perilous Frontier: Nomadic Empires and China (Oxford: Blackwell, 1989).

[5] Nicola Di Cosmo, “Ethnogenesis, Coevolution and Political Morphology of the Earliest Steppe Empire: The Xiongnu Question Revisited,” in Xiongnu Archaeology Multidisciplinary Perspectives of the First Steppe Empire in Inner Asia, edited by Ursula Brosseder and Bryan K. Miller (Bonn, 2011), 43-44.

[6] Catrin Kost, The Practice of Imagery in the Northern Chinese Steppe (5th-lst centuries BCE) (Bonn, 2014), 69-82.

[7] Owen Lattimore, Inner Asian Frontiers of China (New York: American Geographical Society, 1940), 511-529.

[8] Kost, “The Practice of Imagery,” 71.

[9] Lattimore, Inner Asian Frontiers of China, 522.

[10] Ma Changshou, Beidi yu Xiongnu (Shanlian Bookstore, 1962), 14-21.

[11] Di Cosmo, “Ethnogenesis, Coevolution and Political Morphology,” 47.


About Stephen Norris