A Tale of Two Tribes: the Sino-Tibetan origins recorded in historical texts

To determine the common origins of languages and peoples, the most reliable methods include: the comparative reconstruction method and other numerical methods in linguistics, archaeological evidence, and population genetics, according to Wang (1998). When I was writing my book A History of the Chinese Language, I initially wanted to introduce arguments made by Yu Min (1980) for the common origins of the Sino-Tibetan languages and peoples, based on early historical texts. This argument was in chapter 2 of my original manuscript, and the title of chapter 2 was originally titled “A Tale of Two Tribes: Prehistory”, where the “two tribes” refers to the legendary Huangdi and Yandi tribes, as described in Yu Min’s (1980) article. But then based on the criticism from an anonymous reviewer, I agreed to remove that section from chapter 2 and the title was changed to “Where it all began: Prehistory”.

two tribes

The reviewer’s comments are very reasonable, since historical texts are not hard evidence, and any historical text could be distorted by people during the process of transmission. However, if taken with a large grain of salt, a study of such historical texts can actually add to the validity of both the independent findings in other disciplines such as historical linguistics, archaeology and population genetics, and the historical truthfulness of the text itself. Let me explain in a little more detail as to how this is possible.

Suppose the stories of an early nomadic tribe was passed down orally through many generations. There must be some truth in such narratives, although it is hard to separate the facts from fantasy just by looking at the oral legend itself. Also suppose that at some point the oral legend was written down and then continuously transmitted down to our current time, with many possible distortions introduced during the whole process. However, if say, there is a date in such a text about the early origins of the tribe, and it coincides with the established reconstructed time depth in linguistics or the time depth determined by archaeological or genetic research, the truthfulness of this date in the text is more likely to be the case, because this date was recorded or possibly revised long before any modern scientific methods could determine a comparable date. But of course it is still possible for this to be just a coincidence. But if many details in the text coincide with the evidence from other modern disciplines, the truthfulness of the text cannot be denied anymore. In such cases, the historical text and the findings in other scientific disciplines can indeed cross-validate each other. Therefore let us examine the original claims made by Yu Min (1980) here just to entertain the thesis that I proposed above. I think such a blog post should be a good venue for discussions of controversial methodology that has not gained wide acceptance. So here it goes.

Although concrete historical written records of ancient China goes back to the late Shang Dynasty about 3200 years ago as attested in the oracle bone script, there are extensive written records in later times about the pre-Shang era in Chinese history. In many aspects these records resemble legends passed down through generations before they were finally written down, and therefore their accuracy might need disputed. Nonetheless these are valuable information that we can use to gain some knowledge into the prehistory of the Chinese language. Yu (1980) examined such legendary records extensively and critically. He found the following story in the ancient history book called Guóyǔ (“The Discourses of the States”) dated as early as the 5th century BCE:

“In the past, the tribe led by Shàodiǎn intermarried with the tribe Yǒujiǎo. Then Huángdì and Yándì were born. The Huángdì tribe later prospered as a tribal leader along the Jī River, while the Yándì tribe prospered as a tribal leader along the Jiāng River.”

Huangdi, or the Yellow Emperor, is a legendary sovereign in ancient China, whose rule is said to have begun in 2697 BCE. He is commonly regarded as the ancestor of the Chinese people. Yandi, or the Yan Emperor, is also a legendary sovereign. According the story cited above, Yandi was Huangdi’s brother. Then where were the Ji River and Jiang River? Yu (1980) cited from an ancient geography book called Shuǐ Jīng Zhù (“Commentary on the Waterways Classic”) dated to the Northern Wei Dynasty (386-535 CE), saying that the Huangdi tribe prospered first in the region which is currently the Ningxia Hui Autonomous Region in northwestern China. The Yandi tribe prospered in what is the Shaanxi Province, not far from Huangdi’s tribe. Both emperors then led their peoples along the northern bank of the Yellow River eastwards to what is present-day Henan, Hebei and Shandong Provinces. The Huangdi tribe defeated a tribe led by Chī Yóu near Zhuolu, Hebei Province, after which Huangdi was regarded as the leader of all tribes. Although Huangdi ruled the region, the Yandi tribe was also prominent. Therefore these two tribes can be considered the basis of the Huáxià people in the Yellow River region from these early times until the Hàn Dynasty (206 BCE – 220 CE) when the Huáxià people began to be called the Hàn people, which is still the name of the largest ethnic group in present-day China. We mentioned earlier that Yandi hailed from the Jiang River, and therefore he adopted “Jiang” as his name. The Yandi and Huangdi tribes were originally nomadic tribes. After they settled down along the middle and lower Yellow River region, agriculture gradually developed. At the same time, some of the Yandi tribe either went back to their nomadic roots or remained in the western part of China while the majority of the tribal people migrated with Yandi. These nomadic people of the Yandi tribe who stayed in their original homelands were called “Qiang”. Yu (1980) cited evidence from the oracle bone script that a nomadic Jiang tribe was called Qiang. The pronunciations of “Jiang” and “Qiang” must have been quite similar. According to Yu (1980), these two different pronunciations were just differences in dialectal accents between the agrarian descendants of the Yandi tribe and the nomadic descendants of the Yandi tribe. Yu (1980) continues to cite a passage in the Book of Documents (Shàngshū, as early as 11th century BCE), which was the speech delivered by the King Wu of Zhou in front of his army that included some Qiang people. The same story was also recorded in bronze script which was inscribed on bronze ware dated to the early Zhou Dynasty. Thus this speech should be a real historical event. If the Qiang people could understand what the King Wu of Zhou, then maybe as late as the 11th century, the Qiang language and what was to develop into the Chinese language were still mutually intelligible. Thus historical records show, the nomadic people from the Yandi tribe who remained in the western part or returned there after migrating eastward became the Qiang people. There are a series of later historical records about the Qiang people and their kingdoms in western China. Finally, in the Tang Dynasty in the 7th century, the Qiang people formed a kingdom called Tǔbō, i.e. Tibet. Now let’s go back to the agrarian Huangdi-Yandi tribes in the middle and lower Yellow River region. They continued to rule the northern part of present-day China while migrating further south, getting into contact with the inhabitants in what is present-day southern China.

Therefore if all these historical records are accurate, then according to Yu (1980), the Sino-Tibetan people originally lived a nomadic life in an area in present-day Ningxia, Gansu and Shaanxi during the 3rd millennium BCE, i.e. 5000 years ago. Then the majority of these people migrated east first, and gradually became agrarian and built the earliest dynasties in Chinese history. These early agrarian people are the Huáxià people who were the ancestors of the Han people. The Huáxià people further migrated southward to present-day southern China. Some of the original Sino-Tibetan people either stayed or returned afterwards to their homeland in a nomadic lifestyle in the western part of China. They gradually migrated further westward and then southward to the Himalayan region including Tibet. During the migration, the original Sino-Tibetan language became Chinese, Tibetan, and Qiang, and also other Sino-Tibetan languages. Yu (1980) gave a few examples of cognate words between Chinese and Tibetan. For example, the word for “this” in Tibetan is “adi”, while the same word in Old Chinese is written with the character 是 and its modern pronunciation is “shì”. But according to early textual records, the word “shì” might have been pronounced with an initial consonant “d-”.

Yu’s (1980) argument based on historical records is quite convincing, although the time-depth of 5000 years is a little younger than the 6000 years that Wang (1998) obtained by measuring the lengths of phylogenetic trees. But these records were made during much later times and their authenticity is sometimes debated too. Therefore in order to form a stronger argument, we need concrete archaeological evidence. Wang (1998) cites three maps from Chang (1986) showing prehistoric cultures at 9000 BP, 7000 BP and 6000/5000 BP that archaeologists constructed for the early history of China. The map at 9000 BP shows several cultures in the upper and middle Yellow River region, and cultures in the lower Yangtze River region and southern China. These cultures showed no clear evidence of interactions. The map for 7000 BP contains three major cultures that span the whole region of northern China, two cultures in the lower Yangtze River and a vast culture that spans the southernmost coastal area of China. At this stage there was still no major interactions between these cultures. The third map for 6000/5000 BP presents a picture of a network of cultures from north to south that had a significant degree of interaction. Wang (1998) concludes that population movements brought about similarities between these cultures and alongside such migrations, people brought their own languages. Therefore a former linguistic community can split up into different linguistic communities; formerly different linguistic communities can come into close contact with each other so that they may borrow words from each other. Such a scenario accords very well with Yu’s (1980) story. 

So here what is worth noting is exactly the coincidence of the location, migration directions and time depth of both stories: the one reconstructed by Yu (1980) based on historical texts, and the one reconstructed by modern scientific methods reported in Wang (1998). Such coincidences in many details cannot still be just coincidences. I am actually a little amazed by how similar Yu (1980) story is to Wang (1998) in so many aspects.

Interestingly, Yu’s (1980) article has not been given its deserved attention in scholarly field. I think this is probably due to the credibility of his method which relies so much on historical texts which in turn are not hard evidence at all. But with the new development originally reported by Wang (1998), we see that Yu’s (1998) story is actually more convincing.

Therefore a careful study of historical texts combined with cross-reference from modern scientific disciplines can be a rewarding process. If done correctly, such textual evidence can be another good type of evidence for exploring prehistory.


Yu, Min (俞敏). 1980. “汉藏两族人和话同源探索 (An Investigation Into the Common Origins of Peoples and Their Languages of the Han and the Tibetans)”.《北京师范大学学报(社会科学版)》1: 45-53.

Wang, William S-Y. 1994. Glottochronology, lexicostatistics and other numerical methods. In Encyclopedia of Language and Linguistics, Longman.

Wang, William S-Y. 1998. Three windows on the past. In The Bronze Age and Early Iron Age Peoples of Eastern Central Asia, pp508-534.  V.H.Mair, ed.  University of Pennsylvania Museum Publications.


Leave a Reply

Your email address will not be published. Required fields are marked *