“The Research on the Identification Between Tiele (鐵勒) and the Oγuric Tribes” — Cheng Fangyi (2012)

 

“The Research on the Identification Between Tiele (鐵勒) and the Oγuric Tribes” — Cheng Fangyi

Archivum Eurasiae Medii Aevi, 2012

 

ABSTRACT

 

The subject of this article is to try to identify the Oγuric tribes, which appeared in Western literature beginning the second half of the fifth century, with the Tiele (鐵勒) tribes from Chinese literature. To prove this, we find other similarities between the Oγuric and Tiele tribes apart from the standards that define a nation (race, language, custom, etc.).

 

First, we demonstrate that Oγur is Oγuz, the former being the form with rhotacism of the latter, although the time of the occurrence of rhotacism has not been fully established. At the same time, Oγuz is identified as Wuhu (烏護)\Wuhe (烏紇), and even Huihe (回紇、Uyγur). This allows scholars to connect the Oγuric tribes with the Tiele tribes.

 

The second method of establishing the connection is to identify On-Oγur in Western literature as On-Uyγur which appears in the Runic Inscriptions in the ancient Turkic language and the Central Asian literature of Arabia and Persia. The main methodologies will consist of comparing different historical data from the West and East and the identification of the On-Oγur and On-Uyγur tribes.

 

The Oγuric tribes is the name for the collective tribes whose ethnic names contain the element of Oγur, such as Saraγur (Σαράγουροι), Uγur (Οὔγωροι),[1] Onoγur (Ὀνόγουροι), Kutriγur and Utigur. Most scholars consider Oγur as the rhotacism of Oγuz, which was explained as a clan/tribe by James Hamilton.[2] “Sara” or Šara in “Saraγur” (Šara Oγur) means “yellow” or “white,” and can even be translated as “western.” “On” in Onoγur is the word “ten” in Turkic.[3]

Peter Golden described the Oγuric tribes in The History of Early Inner Asia according to data found in the Byzantine literature. This paper sums up the main features of the Oγuric tribes as follows:

1. When the Oγurs first appeared in Western sources, they occupied the Western Siberian and Kazakh steppes. Their language was sufficiently distinct from Common Turkic, so that it is possible to trace some of the effects of their influence on the Finno-Ugrian peoples through loanwords taken from them.
2. The sole surviving element of Oγur is the Chuvash language, which has been heavily influenced by its neighboring Finno-Ugric tongues.
3. The economic structure of Oγur society, with its attendant cultural and lifestyle expressions, appears to have been relatively distinct from those of most Turkic nomads. Oγur society placed a greater emphasis on agriculture, urban development and, in general, more peaceful pursuits.
4. In addition, the Oγur tribes exploited the proximity to the forest-steppe zone and the high potential of the fur trade; these were factors which contributed to development of their societies.[4]

Today most scholars, such as Peter Golden and James Hamilton, et al., identify the Oγur tribes as part of the Tiele (鐵勒) tribal confederation of Chinese literature. However, the difference between the identification was rarely mentioned by the scholars. In fact, because of the shortage of Oγur material, we can only approximate when and where the Oγurs tribes existed, but we cannot create an integrated chain of the historical evidence to prove the identification. Therefore, many scholars have utilized comparisons between Turkic, Chinese and Greek phonetics to do so. The most systemic paper on this subject is Toquz-Oγuz et On-Uyγur by James Hamilton.

 

The first method on this subject is to prove that Oγur is Oγuz, and the former is the rhotacism of the latter, although the time of the precise occurrence of the rhotacism remains debated. At the same time, Oγuz is shown to be Wuhu (烏護)\Wuhe (烏紇), even Huihe (回紇, MC ɦuoi-ɦet,[5] Uyγur). In this way, scholars can make a connection between the Oγuric and Tiele tribes.

 

The second objective is to identify the On-Oγur in Western literature with the On-Uyγur, who appear in the Runic Inscriptions and the Central Asian literature of Arabia and Persia. To do so, we compare historical data between the West and East and identify the tribes under the name of On-Oγur and On-Uyγur. The latter identification is accomplished in a way similar to how most Chinese scholars proved that the Oγur are the Tiele in the Monograph on the Tiele  (隋書·鐵勒傳).

 

This paper will focus on the two methods, and offer other new explanations on this subject.

 

1. THE APPEARANCE OF THE TIELE TRIBE AND RESEARCH ON THE NAME

 

The Tiele appeared in Chinese literature from the 6^th century to the 8^th century, and were recorded as Tiele or Chile (敕勒). The New T’ang Shu (新唐書) claims that Tiele is a mistaken form from Chile. Most scholars believe the two names are just different Chinese characters to describe the same Turkic word; the same is true with Di (狄), Dili (狄曆), Dingling (丁零, 丁靈) and Tujue (突厥, CE thuotkʷjɐt,Türk). Different scholars such as Ma Changshou and Shiratori Kurakichi disagree on what the specific Turkic word is. This topic will be addressed later.

 

Duan Lianqin tried to prove that the Dingling, Gaoche or High Carts, and Tiele were the same ethnos using ethnology. His research focused on the historical connection between the three tribes and he also discovered similar customs between the three tribes. He considered the Tiele as a different ethnos from Turk, according to the Marxist definition of ethnos given by Joseph Stalin.[6] But other scholars like Toru Haneda,[7] Ma Changshou[8] et al. considered the Tiele to be the same ethnos as the Turks.

 

Now, we will discuss the Turkic word that is represented by the transliteration Tiele (MC thet-lәk). There are several opinions:

1. Töliš (Tölöš)

V. Thomsen, P. A. Boodberg and Emmanuel-Édouard Chavannes all identified Tiele as Töliš (Tölöš). Töliš (Tölöš) came from The Runic Inscription, on the eastern surface of The Kültegin Inscription, East, line 13:

[They] gathered together seven hundred soldiers. After they had numbered seven hundred men, he brought them to order and trained people who had lost their state and their kagan, people who had turned to slaves and servants, people who had lost the Turkic institutions, brought to order and followed the rules of my ancestors. Then, he gave the Töliš and Tardush peoples state-organizations.[9]

 

The earlier scholars in the West, like V. Thomsen, all considered Töliš and Tarduš as the names of tribes, the former being Tiele, and the latter being Yan-tuo (延陀). Conversely, Chinese scholars like Cen Zhongmian and Wang Jingru denied that they were Tiele according to the Chinese literature, and inferred Töliš and Tarduš to be names of political organizations. In his research, Cen Zhongmian regarded Töliš and Tarduš as two districts dealing with the East and the West, respectively.[10] However, Wang Jingru according to New T’ang Shu considered Töliš and Tarduš as two districts dealing with the South and the North, respectively.[11] It is extracted from “Xueyantuo Zhuan” (The Biography of Xueyantuo, 薛延陀傳) of New T’ang Shu:

 

(Yinan, 夷男) had a powerful army of 200,000, and sent his two son Daduse (大度設, MC dɑi-duo-ɕiɛt, tarduš) and Tulisi (突利失, MC thuot-li-ɕit, Töliš) to take charge of them. They were called the Southern and Northern Division.[12]

 

From the two scholars’ research, it can be concluded that Tiele is not Töliš.

 

2. Tegrek

 

Gerard Clauson was the first to postulate that “Tiele is tegrek,” and was supported by Peter Golden. In Golden’s article Some Thoughts on the Origins of the Turks and the Shaping of the Turkic Peoples, he stated:

The Oγuric groupings were, by the 460s, part of a larger loose and still ill-defined confederation of nomadic tribes extending in an arc across Eurasia from Southern Siberia and Northern Mongolia to the Western Eurasian steppes. This confederation was called in Chinese T’ieh-le, a rendering perhaps of tegrek (Turk. tegrek “rim, ring” and hence “wheel”), which may have been an Altaic term for “cart” (cf. Mong, tergen, telegen, terge).[13]

 

This viewpoint may stem from the other name for the Tiele, “The High Carts” (Gaoche, 高車). Shiratori Kurakichi also made a similar point in his article, The Research on the Origin of Xiongnu,[14] stating that Dingling, Dili, Chile and Tiele were merely different Chinese characters used to simulate the same Mongolian word–tegre or terege which means “cart.”

3. Tägräg

 

Hamilton asserted that Tiele is Tägräg in Toquz-Oγuz et On-Uyγur. Alemany also supported it in his Early Byzantine Sources on the Oγuric Tribes in the Northern Black Sea Area. Hamilton also concluded that Tiele is Tägräg from the other name of the Tiele, “The High Carts” (Gaoche). He tried to make a connection between the Tiele and the Pazyryk culture, stating that Russian archaeologists found a high cart in the Altai tomb of the Pazyryk culture which had big and light wheels made with tree branches. [T]he time of the Pazyryk culture was 6^th to 2^nd bce while the existence of the cart was confirmed to be 252–238 bce. However, this is inconsistent with the time of the appearance of “The High Carts (Gaoche)” in Chinese literature, which was in the 4^th century ce. Also, Hamilton incorrectly reported the size of the cart. He said the diameter of the cart’s wheel was 2.15 meters.[15] But the correct diameter is 1.5 meters,[16] meaning the cart is not as high as Hamilton believed. From these two observations, it is unreasonable to connect the Pazyryk culture with “The high cart” (Gaoche, 高車) and Tiele.

 

The Turkic word, tägräg, means a “semicircular hoop.” Hamilton also stated that tägräg did not seem to be used as a specified noun except in this situation.[17] With all considerations, Tie-le being tägräg is unconvincing.

 

4. Türk

 

This viewpoint is shared by many scholars today, including Toru Haneda, Onogawa Hidemi,[18] and Geng Shimin.[19] They also believed that Di, Dili, Dingling, Chile and Tujue (Türk, in the Sui and Tang dynasty) all came from the Turkic word Türk. “Türk” means “powerful” and “strength,” and its plural form is “Türküt.” The reason for the difference in Chinese characters is that the names the tribes called themselves and the names that the tribes were called by other tribes were different. This fact unites all the Chinese names such as Di, Dili, Dingling, Chile, Tiele and Tujue with the explanation that these names are the different names used at different times by the Chinese and other nationalities to describe the Turkic races. Zhou Weizhou[20] claimed this point was the key to resolving the many complicated names in Chinese literature for the same group of Turkic speaking peoples living in the Mongolia plateau and its surrounding area.

 

Onogawa Hidemi explained the relationship of the names of Tiele and the Eastern and Western Türks during the Sui and Tang Dynasty according to the names used by the Türks in the Orkhon and Yenisey inscriptions. He thought that from the inscriptions, Türk was used to refer to the Eastern Türks, and On-oq and Toquz Oγuz referred to the Western Türks and the nine clans of Tiele in Mobei (the north of the desert, 漠北). Säkiz Oγuz and Altï Oγuz meant “Tiele of eight clans and six clans.” At the time of the Türk Qaghanate, Türk was not a joint name for the Turkic peoples, but only for the Eastern Türks. The Western Türks and every clan of Tiele all had their own names. Hidemi verified that the time of the birth of the distinction of Türk and On-oq was after the year of Zhen Guan (貞觀, 627–649 ce). After that time, according to V. V. Bartold, the name “Türk” contained some political meaning. Meanwhile, the consciousness of being one of the Turkic nationalities was the common potential consciousness among the On-oq, Türk and every clan of the Toquz Oγuz. In the inscription “The Toquz Uyγur ay-tängridä qut bolmïsh alp bilgä qaghan,” the qaghan was called “Türkische Weltenherrscher” in the part written in the Sogdian language. It is unreasonable unless we consider that all the clans under the name of Tiele and Tujue (Türk) had the sense that they were of Turkic nationality.[21] Hidemi considered that Tiele and Tujue (Türk) came from the common name of Türk, while Tiele and Tujue (Türk) were used after the establishment of the Türk Qaghanate.

 

Although there were many opinions about the meaning of Tiele, a number of scholars have concluded that Tiele came from Türk, and the same for Di, Dili, Dingling, Chile and Tujue. Hidemi believed that the Tiele tribes were scattered in a large area, and deduced that the origin of the Ashina (阿史那, MC ʔɑ-ʂɨ-nɑ) family also belonged to Tiele. It is conceivable that it was the Tiele tribes, who were led by the Ashina, who built the nomadic state of Türk. Tujue (Türk) and Tiele did not have a difference in their meaning until the nomadic state was built. However, according to the records in The Monograph on the Tiele in the Sui Shu (隋書·鐵勒傳), Tiele seems to include only the Turkic tribes which were not the tribes composing the Türk State. As the Eastern and Western Türk Qaghanates developed, especially the Western one, some tribes in the Tiele went under the name of the Türk Qaghanate gradually, while some tribes in the Tiele union remained in obscurity and their names stayed as they were. Hidemi mentioned that it was difficult to find a connection between the Tiele tribes, which were active during the Sui and Tang Dynasty, and the Tiele tribes in former dynasties such as Northern Wei (Bei Wei北魏 386–534), because their tribes had completely different names. There were great changes in the political conditions in Mobei (漠北), i.e., “the Northern Desert” (the region beyond the Gobi), such as the extinction of the state of Gaoche (高車) and Rouran (柔然) and the rise of the Türk Qaghanate. These great changes made some tribes in the Tiele become weaker, after which they were annexed and replaced by other tribes. The mixture of these tribes in Tiele is a distinct feature in Mobei, and the phenomenon lasted a very long time.[22]

 

Golden holds a different opinion than Hidemi on the composition of the Tiele tribes. He believed the Tiele Union included Mongolic as well as Turkic groupings. They were brought into the Xiongnu Union involuntarily and remained recalcitrant vassals. The Dingling (丁零) or Tiele tribes were concentrated in Northern Mongolia and the Irtysh region, extending to Lake Baikal and Middle Yenisey. The association of terms for “carts” with Turkic nomadic groupings in the designation Gaoche (High Carts), one of the Chinese terms used for Tiele and later the Uyγurs, were derived from them. Golden thought that they were largely in possession of the Kazakh steppe-lands, supplanting Iranian nomads after 350 ce. And after 460 ce, as we have seen, the Oγuric groupings had established themselves in the Black Sea steppes. Golden mentioned the Tiele and their various outcroppings, because they became an important component of the early Türk state. He also had described the distribution of the Turkic peoples in the era prior to the establishment of the Turkic state, which would unite most, if not all, of the Turkic peoples.[23]

 

Thus, from Hidemi’s assertion that Tiele was the transcription of Türk, or Golden’s claim that Tiele represented Tegrek, we can deduce that the Tiele tribes were widely distributed. Golden considered the Oγuric tribes that spread to the steppe of the Black Sea and South Europe as a part of the Tiele. Although the explanation offered by Hidemi is focused on the tribes of Tiele in Mobei and did not mention the Oγuric tribes in the West, it would be natural to deduce that the Oγuric tribes belonged to Tiele according to his point that Tiele came from Türk. Ma Changshou holds the viewpoint that Chile and Tiele was the common name of all the tribes speaking Turkic, and what was recorded in The Monograph on the Tiele in the Sui Shu was a correct acknowledgement by the Chinese about the scattering of the Turkic-speaking tribes of the Western Region (Xiyu, 西域) and the Eurasian steppe. This viewpoint confirms that the widespread distribution of the Tiele, as noted in The Monograph on the Tiele in the Sui Shu, is based on the data of the Chinese sources.

 

Therefore, the subject of this paper, “The identification between Tiele and the Oγuric tribes,” is a reasonable assumption regarding the components of Tiele (including elements such as the language they used) according to the conclusions noted above. The Tiele were the widely scattered, Turkic-speaking tribes in the Eurasian steppe (or as Golden stated, the Tiele union included Mongolic as well as Turkic groupings). Also, it was widely agreed upon by the Western scholars that the Oγuric tribes were Turkic-speaking. However, this paper proceeds to closely compare the tribes under the name of Oγuric tribes and Tiele, with the goal to find a more detailed connection between them.

 

2. IDENTIFICATION BETWEEN OΓUR AND OΓUZ

 

The proposition that oγur was the rhotacism of oγuz was suggested in earlier times by the Altaic linguists. In his Einführung in die altaische Sprachwissenschaft, the Altaic linguist G. J. Ramstedt notes differences found between Ancient Turkic and Chuvash belonging to the Turkic language branch of the Altai language family:

 

According to language or ethnicity the Chuvash people today must come from these people (Bulgar). The Avars, which were mentioned together with the Huns were mostly Mongolian, but not the Turks as Pelliot said. And the Chuvash also have many distinguishing features of Proto-Turkic… the Chuvash was the only remnant of the language of the ancient Huns who called themselves Bulgar. The language keeps many ancient features, such as -l- and -r-, which correspond to -š- and -z- of Turkic. But it separated itself from other Turkic dialects, and developed independently and gradually.[24]

 

The Oγuric tribes were confirmed as the ancestor or predecessor of the Bulgar. In this way, the connection between Oγur and Oγuz was established. Golden believed that the only living survivor of Oγur material is Chuvash, which has been heavily influenced by neighboring Finno-Ugric tongues.[25] He described the differences between the languages that the Oγuric tribes used and the Common Turkic, and tried to explain them in the article Some Thoughts on the Origins of the Turks and the Shaping of the Turkic Peoples. Golden postulated three main reasons, but all contain some differing viewpoints regarding the differences:

 

(a) An earlier form of the language, one that broke away before a series of phonetic changes occurred in Common Turkic or, conversely, (b) an early grouping of Turkic peoples whose language underwent a series of phonetic innovations (some of which may have been reflected in ancient Turkic borrowing in Mongolic) that distinguished it from common Turkic and then migrated westward; or even (c) a separate, Para-Turkic tongue (i.e., one stemming from the same ultimate source that produced Proto-Turkic).[26]

 

Golden believed that the Proto-Turkic and Ancient Turkic periods extended from approximately 3000 bce to 500 ce. At the end of the period, the distinct Oγuric forms (with r for z, e.g., Oγuric tenger/common Turkic tengiz “sea” and l for š) began to surface. The precise chronological details of these changes are still uncertain,[27] and it appears that there are some overlaps within the two groups, that is, certain characteristics of change in the Oγuric language can also be found in some Common Turkic dialects, but they are not always fully realized.  The assertion that oγur was the rhotacism of oγuz has been demonstrated.

 

Oγur was a reflection in the Oγuric tribes’ language, and oγuz was the corresponding word in the Common Turkic. Golden deemed that oγuz initially meant “kindred, related tribes,” and later had the meaning of ethnonymic status, such as “Toquz Oγuz.” Golden believed that the Oγuric peoples moved westward from Inner Asia as Priscus recorded, who reports a migration that began in Inner Asia ignited by the Avars’ expansionist activities circa 450 ce. By 463, they were already in the Pontic steppes and in contact with Constantinople. According to some of the early Turkic loanwords in Mongolic, Golden also believed that the Turkic-speaking community was located near the Mongolic speakers before their scattering and migrations. Thus, the time of scattering and migration of the Turkic-speaking tribes was prior to 450 ce.

 

However, D. Sinor, after analyzing material on the early Turks in the West Central Eurasia, believes that most of these Turks, who used an -r- Turkic language, could not come from distant Mongolia where the Turks spoke a -z- language. Also, he deemed the time of scattering and migration of the peoples using Turkic to be much earlier and longer.[28] In fact, in Ramstedt’s book Einführung in die altaische Sprachwissenschaft, not only Chuvash and Mongol have the corresponding of “l” and “r” with “š” and “z” in the Common Turkic, but also Tungus and Korean. It seems that the language used by the Oγuric tribes was an earlier form of Turkic and the series of phonetic changes occurred in Common Turkic because of the influence of other languages. Thus, the commencement and duration of scattering and migration of the Turkic-using tribes may be as early and lengthy as Sinor said. The time and details cannot be confirmed at this time.

 

Hamilton did not agree with the point that oγur was the rhotacism of oγuz. He stated in Toquz-Oγuz et On-Uyγur:

 

Il n’en reste pas moins que l’on ne connaît, en fait de textes turcs anciens rhotacisants, que quelques inscriptions de pierres tombales de la fin du XIIIe et du début du XIVe siècle, qui proviennent justement du pays čuvaš sur la haute Volga. Qui plus est, on n’a jamais rencontré nulle part de variante Oγur du nom Oγuz, en dehors de l’exemple contesté des sources byzantines.[29]

[The fact remains that only a few tombstone inscriptions from the end of the 13^th and the beginning of the 14^th centuries are known, in fact from rhotacizing ancient Turkish texts, which come precisely from the Chuvash country on the upper Volga. Moreover, no Oγur variant of the name Oγuz has ever been encountered anywhere apart from the disputed example of Byzantine sources.]

Hamilton said that we cannot pay too much attention to this assumption until there exists a definite sign of the rhotacism around the 5^th century. Based on the material available to him at that time, his point is reasonable; the connection of rhotacism was established in Altaic Linguistics, and the occurrence of the linguistic data used to prove the connection was much later than the time of the Oγuric tribes. Ramstedt also did not use oγur-oγuz as an example for the rhotacism. We cannot confirm the assumption until there are new materials, but we cannot reject it either. Recent studies, however, suggest that rhotacism was already present some centuries before 450 ce.[30]

 

Hamilton deduced the origin of oγuz according to the comparison and identification between the terms oγuš in the Uyγur literature and oγuz in the Inscriptions of Orhon. This idea is from Pelliot. The key point of Hamilton’s deduction was why “-š-” in the oγuš became “-z-.” He used “the Homophonic phenomenon” in Turkic to solve the problem. He stated:

 

A l’origine, il y aurait eu une confédération appelée des “neuf clans,” c’està-dire toquz oγuš. L’expression étant bientôt prise comme nom propre, cependant, la tendance générale du turc à l’harmonisation et sa prédilection pour les paires de mots “à écho” auraient joué pour faire passer Toquz-Oγuš, qui fait l’impression d’une rime manquée, à Toquz-Oγuz, rime parfait. Il se serait produit, en somme, une assimilation harmonique entré Toquz et Oγuš, qui ont la même assonance, sur le modèle d’innombrables paires de mots rimants en turc, à peu près comme qarγadi alqadi est devenu qarγadi arqadi “il a maudit.” Ainsi, alors qu’on a généralement supposé le contraire, le nom Toquz-Oγuz aurait précédé le nom Oγuz, qui n’aurait été au début que la forme abrégée du nom de la confédération des “Neuf Clans.”[31]

[Originally, there must have been a confederation called “Nine Clans,” i.e., Toquz Oγuš. The expression having soon been taken as a proper noun, however, the general tendency of Turkish towards harmonization and its predilection for “echoing” pairs of words must have played a role in changing Toquz-Oγuš, which gives the impression of a failed rhyme, to Toquz-Oγuz, a perfect rhyme. There must have occurred, in short, a harmonic assimilation between Toquz and Oγuš, which have the same assonance, on the model of innumerable pairs of rhyming words in Turkish, such as qarγadi alqadi became qarγadi arqadi “he cursed.” Therefore, while it has generally been assumed to the contrary, the name Toquz-Oγuz must have preceded the name Oγuz, which at first must have been only the abbreviated form of the name of the confederation of the “Nine Clans.”]

So, Hamilton made his chain of deduction: Oγuz ＜ Toquz-Oγuz ＜ Toquz-Oγuš, but this chain is not stable.

 

Tang Huiyao (唐會要) has a record of the construction of the “three clans”:

 

Qarluq (葛邏祿, MC kɑt-lɑ-luk) was a Turkic clan originally. It was in the area of Beiting (the north of Northern Yard Government, 北庭) and the west of the Jin Mountain (金山)… There were three clans in Qarluq. They were Moula (謀剌, 謀落), Pobu (婆蔔) and Tashili (踏實力), and they all went to the capital of China and submitted themselves to the rule of the empire. In the second year of Xianqing (顯慶) (657–658), the government set up three Dudu (Turk. tutuq prefectures of management, 都督) for them, which were the Yin Mountain (陰山), the Great Desert (大漠) and Yuanchi (元池). And the Dudu were the leaders of the three clans. The three clans lived beneath the Eastern and Western Türks, and they always changed their position according to the power of the Eastern and Western Türks. Later on, they migrated a short distance to the south, and called themselves the “three clans” san xing (三姓).[32]

 

The “three clans” was also recorded in the Shine-Usu inscription as “üč qarluq,” and affirmed by Geng Shimin.[33] There were many combinations of  “number + name of clan” such as “otuz tatar”[34] and “five clans of Wuman (五姓烏蠻)”[35] in the ancient Turkic inscriptions and Chinese literature. So the “clan” xing (姓) in “nine clans” (九姓) and “three clans” (三姓) may not come from Oγuz or Qarluq. The latter revealed that the combination in the ancient Turkic inscriptions such as Oγuz in “Toquz Oγuz” must be the name of the clan or the clan union.

 

The “three clans” (三姓) in Tang Huiyao (唐會要) was the union of three clans: Moula (謀剌, 謀落), Pobu (婆蔔) and Tashili (踏實力), none of which can stand for the whole of the Qarluq alone. They were so clearly divided that the Tang government set up a prefecture for each clan. This demonstrates the political structure inside “three clans” (三姓) and “nine clans” Jiu Xing (九姓). In the years of Zhen Guan (貞觀, 627–650), the Tang government set up six prefectures and seven districts (Zhou, 州) for the “nine clans”  (九姓).[36] There were actually at least eleven clans in the “nine clans” during the Tang dynasty.[37] Thus, in actuality the “three clans” and the “nine clans” meant that there were three or nine big clans in the clan union, but there also existed other small clans within.

 

In the quote from Tang Huiyao (唐會要), the author did not add any name before or after the “three clans” (三姓) as in the “nine clans of the Tiele” (九姓鐵勒 or 鐵勒九姓). Akio Katayama mentioned that the Tiele in the “nine clans of the Tiele” (九姓鐵勒or鐵勒九姓) was added by the Chinese.[38]  But in the ancient Turkic inscriptions, there always was a name of a clan added behind the combination; Toquz Oγuz was the abbreviation for Oγuz, and Üč Qarluq for Qarluq. So, Oγuz must be the name of one clan or union like Qarluq, but not one of nine clans in the “nine clans” (九姓)/Toquz Oγuz.

 

Up to this point, Pelliot and Akio Katayama’s deduction that the “Toquz Oγuz” in the ancient Turkic inscriptions corresponds to the “nine clans” (九姓) in the Chinese texts is convincing. However, the deductive chain of Hamilton, Oγuz ＜ Toquz-Oγuz ＜ Toquz-oγuš, is unconvincing.[39]

 

Moreover, because of his deductive chain, Hamilton tried to deny the identification between Oγuz and Wuhu (烏護, MC ʔuo-ɦuo). However, the identification is agreed upon by some scholars.[40] Hamilton listed three reasons for the disagreement:

1.  Il n’est pas du tout démontré qu’il faille couper la liste de transcriptions de norms de tribus t’ie-lo, en grande partie non identifiée.
2. Au lieu de Wou-hou烏護, leçon du Pei che, le Souei chou a Wou houan 烏讙, ce qui ressemble à une transcription du nom des Wou-houan烏桓 ou Wou-wan烏丸, peuple bien connu à l’époque des Han et des Trois Royaumes, dont il subsistait encore des restes au Xe siècle.
3. Wou-hou qui rend en fait uγu ou oγu, serait une transcription incomplétude de «Oγuz»; l’on pourrait y voir, tout ausssi bien, par exemple, oγul, oγuš, oγur, etc.[41]

1. There is no evidence whatsoever to cut the list of largely unidentified t'ie-lo tribal norm transcripts.
2. Instead of Wou-hou 烏護, a Pei che lesson, the Sui Shu has Wou houan 烏讙, which looks like a transcription of the name of the Wou-houan 烏桓 or Wou-wan 烏丸, people well known at the time of the Han and the Three Kingdoms, of which there were still remnants in the tenth century.
3. Wou-hou which actually renders uγu or oγu, would be an incomplete transcription of “Oγuz”; we could see there, just as well, for example, oγul, oγuš, oγur, etc.

The point that the “Toquz Oγuz” in the ancient Turkic inscriptions corresponds to the “nine clans” (九姓) in the Chinese literature has been discussed by many scholars. Here this paper will discuss the negative opinion offered by Hamilton.

 

Although there was a long and undivided list of the tribal names in the Monograph on the Tiele in the Sui Shu, Wuhu (烏護) appeared not only in the Monograph on the Tiele in the Sui Shu, but also in the Tang Chou and Bei Shi (Pei Che, 北史). So Wuhu (烏護) can be verified in other works. The Monograph on the Tiele in the Sui Shu and the Monograph on the Tiele in the Bei Shi are practically identical. We know that the Sui Shu was finished earlier than the Bei Shi. The author of the Monograph of the Tiele in the Bei Shi cited almost every word in the Sui Shu. Hence, “Hu 護” in Sui Shu and “Huan 讙” in Bei Shi, which look similar, should be the same. This kind of mistake was common in ancient Chinese texts. The third reason Hamilton offered looks like sophistry. The transliteration of Wuhu can be oγul, oγuš or oγur, but Wuhu in Chinese literature was always a tribal name. If we want to find its correspondence in ancient Turkic, it must also be a tribal name, not other nouns.

 

3. IDENTIFICATION BETWEEN ON-OΓUR AND ON-UYΓUR

 

The identification between On-Oγur and On-Uyγur was made by Hamilton in Toquz-Oγuz et On-Uyγur. At first, Hamilton compared and identified Oγur and Uyγur using phonetics. The similarity between the two words was obvious; they both include the same part: “γur.” About the preceding part, he wrote:

 

Pour ce qui est de la correspondance phonétique, il y abien Uiγur et Uiγor dans deux manuscrits de Menandros. A supposer, toutefois, que ce soient là des fautes de copiste pour Uniγur et Uniγor, il reste encore, pour expliquer que les byzantines aient noté (On-)Uyγur par (On-)Uγur\Oγur\Iγur, le fait qu’il n’y avait pas de son uy dans le grec de l’epoque byzantine. De ce fait, les byzantine, ayant à rendre dans leur langue «Uyγur», auraient recouru tantôt à u\o, tantôt à i, pour représenter la diphtongue uy du turc, à laquelle leurs oreilles n’étaient pas accoutumées.[42]

[As for the phonetic correspondence, there are many Uiγur and Uiγor in two manuscripts of Menandros. Supposing, however, that these are copyist’s errors for Uniγur and Uniγor, there still remains to explain that the Byzantines noted (On-)Uyγur by (On-) Uγur\Oγur\Iγur, while there was no uy sound in Byzantine Greek. As a result, the Byzantines, having to render “Uyγur” in their language, would have resorted sometimes to u\o, sometimes to i to represent the Turkish diphthong uy, to which their ears were not accustomed.]

Thus, Hamilton identified the Uyγur (回鶻, MC ɦuoi-ɦuot) with the Oγuric tribes which appeared in Western literature from the 60s of the 6^th century (560s) to the 7^th century ce. However, in all six inscriptions of the Türk Qaghanate, the Uyγurs only appeared once in 734 ce (in the Bilge Kagan inscription). In line 37 of the eastern surface, it was written:

 

… fought. I speared their army. Those who gave up, joined me, became my nation. Those who had died – died. Having crossed the Selenga, after wintering in Qaraghan, we destroyed their buildings… retreated… The Uighur Eltäbär with about one hundred soldiers fled to the east…[43]

 

Therefore, there must be another name in the Turkic inscriptions to name the clans which included the Uyγur; this name was Toquz Oγuz.

 

However, in Chinese literature the name of Uyγur appeared earlier. And the precondition was the validity of the identification between Uyγur, Weihe (韋紇, MC ɦʷɨi- ɦet) and Yuanhe (袁紇, MC ɦʷjɐn- ɦet). Yuanhe appeared in the Northern Wei (北魏) Dynasty. Oγur, then, appeared in Byzantine literature almost at the same time as the Uyγurs appeared in Chinese literature.

 

But On-Uyγur (十姓回鶻或十姓回紇) never appeared in Chinese literature. The “ten clans” Shi Xing (十姓) in Chinese literature always referred to the “On Oq,” the Western Türks. Jiu Xing Hui He (九姓回紇) and Jiu Xing Hui Hu (九姓回鶻), which were both used in titles for the Khan of the Uyγur Qaghanate, appeared in the Chinese literature. They should be translated as “the Uyγur clan in the nine clans (九姓),” just as the Uyγur Qaghanate was called “Tughuzghuz” in Arabian and Persian literature at a similar time period. At that time, the Uyγur Qaghanate was called “Toquz Oγuz (九姓)” and “Tughuzghuz” by the Chinese and the Central Asians.[44]

 

“On Uyγur” appeared once in the inscription of the Uyγur Qaghanate on the Moyun Chor monument built circa 759 ce. The Moyun Chor monument (Bayan Chor/Shine Usu inscription) recorded the achievements of the second ruler of the Uyγur Qaghanate, Moyun Chor. Line 3 of the northern surface reads:

 

su... anta : qalmïšï : budun : On Uyγur Toquz Oγuz : üze : yüz // yïl olurup : ... Orqun ögüz : ...

left the people (in the…), upon the people of Ten Uigurs and Nine Oguzes,  they ruled for one hundred years… the Orkhon river…[45]

 

The English translation of the above quotation was done according to the Chinese translation of it by Geng Shimin, who translated the sentence just as Ramstedt did.[46] However, Hamilton understood this sentence in another manner. He considered “On Uyγur” a subject, so the translation became «Les On-Uyγur, people (comprenant) ceux de […] qui étaient restés dans la région (?) de la rivière [Selenga?], régnérent pendant cent ans sur les Toquz-Oγuz, et...». From his translation he deduced that On Uyγur had existed and ruled for one hundred years in 760 ce. After the ruling was established, the Tiele tribes union had formed and fought against the Türks. At the same time, according to the “qui étaient restés dans la region (?) de la rivière [Selenga?]” in his translation, he conjectured that another part of On Uyγur had migrated from the region, implying that the migration occurred around 460 ce.

 

Hamilton’s argument was constructed upon his understanding of the sentence on the Moyun Chor monument without other materials. However, Geng Shimin and Ramstedt’s viewpoints can be validated with Chinese literature. “Over the people of Ten Uigurs and Nine Oguzes, they ruled for hundred years” can be explained as the reign of the former Türk Qaghanate over the Tiele. The former Türk Qaghanate was founded 552 ce when the chieftain Tümän/Tuman (Tumen, 土門, MC thuomuon) defeated the Rouran and named himself Yili Qaghan (伊利可汗), and it lasted until its collapse in 630 ce when the Illig Qaghan (頡利可汗) was defeated by the Tang general, Li Jing (李靖 d. 649). The Türk Qaghanate had ruled the Tiele for almost one hundred years. There are other correlations between the Moyun Chor monument and Chinese literature.[47] Thus, Hamilton’s translation and explanation are questionable, and his deductions based on the translation are unconvincing.

 

Hamilton referred to On Uyγur “est encore attesté dans un manuscript en écriture manichéenne trouvé à Qočo (Kao-tch’ang), un centre uyγur depuis la fin de VIIe siècle.” The time period of the manuscript he commented on could be from the 9^th or 10^th century. Actually, Akio Katayama has said that On Uyγur “appeared more than ten times in several Uyγur manuscripts and the Qočo mural excavated in Eastern Turkistan, and two or three times in the unpublished manuscripts in Berlin according to a private letter from Peter Zieme.”[48] Hamilton made a conclusion just on the basis of the only one appearance, stating:

Ce texte confirme que pour les Uyγur eux-mêmes, qu’ils fussent de l’Orkhon aux VIIe-VIIIe siècle ou de Qočo aux IXe-Xe siècle, la forme plein de leur nom, tout au moins lorsqu’on voulait insister sur son aspect politique, était On-Uyγur, «les Dix Uyγur.»[49]

[This text confirms that for the Uyγur themselves, whether they were from Orkhon in the 7^th-8^th century or from Qočo in the 9^th-10^th century, the full form of their name, at least when one wanted to insist on its political aspect, was On-Uyγur, “the Ten Uyγur.”]

 

Hamilton did not discuss in his article whether On Uyγur was the name with which the On Uyγur referred to themselves or the name given to them by others. We know the Uyγur Qaghanate was called “Toquz Oγuz (九姓)” and “Tughuzghuz” by the Chinese and Central Asians, so On Uyγur is most probably the name which was used by the Uyγur to refer to themselves.

 

The final reference on On Uyγur cited by Hamilton is from The Compendium of Chronicles by Rashid al-Din, which was completed in the 14^th century. The following paragraphs include the connections pertaining to Hamilton’s deduction:

 

In these mountainous areas, there were ten rivers in one place and nine rivers in another place. In ancient times, each tribe of the Uyγurs lived along the rivers, the mountains and the plain. Those tribes who lived along the ten rivers were called On-Uyγur, and those who lived along the nine rivers were called Tūğūz-Uyγur (Toquz Oγuz). Those ten rivers were called ʼūnorǧūn (On Orqon), and their names were as follows: ʼyīšl(i)k, ʼūt(i)nk(a)r, būqīz, ʼūzq(u)nd(u)r, tūlār, tārdār, adar, ʼūj -tā-bin, qamlančū and ʼūtīkān.

 

There were nine tribes living along the previous three rivers, and five tribes living along the following four rivers. The tribe living along the ninth river qamlančū was called ʼūng, and the one living in the riverside of ʼwtīkān was called qamaq-ātikūz. Besides the tribes living along these rivers, there were still 122 tribes living in the same area, but their names were unknown.

 

After several months, years and eras, the Uyγur tribes there still didn’t have a designated supreme ruler (pādishāh). Every tribe chose their leader through violent fighting in any case. At last, all the tribes held a meeting for the common interest, and made a decision which was “we survive until we have our common ruler (pādishāh) who can command every tribe in every aspect.” So all the tribes agreed to choose a man whose name was M(a)ngūbāī from the most clever tribe ʼyišl(i)k and awarded him the title of ʼīl-ʼīlt(i)b(i)r. Also they chose another man who displayed good behavior from the tribe of ʼūzq(u)nd(u)r and awarded him Kūl ʼīrkīn. They let the two persons be the supreme ruler (pādishāh) of the whole community and every tribe. Their clan ruled for hundreds of years.[50]

 

The Uyγur tribes’ process of selecting their leader was recorded in the Arabic literature of the 14^th century. Hamilton considered the ten rivers as the tribes’ name for On Uyγur, which will be discussed later.

 

In the quotation, Hamilton connected the Uyγur tribes’ process of selecting their padishah with records from the Monograph on the Tiele in the Sui Shu. The record in Sui Shu is as follows:

 

In the first year of Da-Ye (大業, 605), Chu Luo (處羅) Qaghan conquered the Tiele tribes, and placed a heavy imposition on these tribes. He also suspected the tribe of Xue Yan Tuo (薛延陀) and was afraid they would revolt. So he gathered hundreds of their leaders and killed them all. These tribes all rose in rebellion against Chu Luo, and chose the Si Li Fa (俟利發, MC ɖʐɨ-li-pʷjɐt, eltäβär) Si Jin (俟斤, MC ɖʐɨ -kɨn, irkin) Qi Bi (契弊, a tribe’s name) Ge Leng (歌楞) as Yi Wu Zhen (易勿真) Qaghan living in the Tan Han (貪汗) mountain. They also elected Ye Zi (也咥), who was the Si Jin (irkin) in Xue Yan Tuo, and they called him the little Qaghan.[51]

Hamilton considered ʼīl-ʼīlt(i)b(i)r and kūl-aīrkīn in The Compendium of Chronicles to be Silifa (eltäβär, 俟利發) and Sijin (irkin, 俟斤) in Sui Shu. He also equated the leader-elections from the two materials and believed the two records were related to the same election. He stated that the records in these materials cohered with the On Uyγur rule over the Toquz Oγuz for one hundred years.[52] When Chinese records are examined carefully, it is found that Chu Luo Qaghan was the Qaghan of the Western Türks and the revolt happened around the Altai Mountain, far away from Mobei (North of the Desert, 漠北). The Chinese sources did not even mention the Uyγurs in the rebellion. Also the Qaghan of Qibi (契弊, a tribe’s name) and Xueyantuo (薛延陀) were liberated from their Qaghan later by the power of Shekui (射匱) Qaghan of the Western Türks.[53] Therefore, the “ruling for hundred years” had never happened.

 

The Chinese scholar Qian Boquan (錢伯泉) used the record in The Compendium of Chronicles in conjunction with The Monograph on the Uyγurs in the New Tang Shu (新唐書·回鶻傳) to discuss the origins of the Uyγurs. The record in The Monograph on the Uyγurs in the New Tang Shu is as follows:

 

There was a man whose name was Shijian (時健) Sijin (irkin), and the people chose him as their leader at first. He had a son called Pusa (菩薩)[54] who was brave and very intelligent. Pusa adored hunting and led his men in every battle in which he defeated every enemy he met. So his men were all scared and submitted to him. But he was banished by Shi Jian. After the death of Shi Jian, all the people in the tribe thought Pusa was excellent and chose him as the new leader. Pusa’s mother was Wu Luohun (烏羅渾) who was strict and fair, and could deal with the things inside the tribe. So the Uyγurs began to prosper.

 

The Uyγurs attacked the northern border of the Türk Qaghanate with the Xueyantuo and Illig Qaghan (頡利可汗) sent Yugushe (欲穀設) to lead a cavalry force of one hundred thousand to fight against them. Pusa led a cavalry force of five thousand to defeat them in the Malie (馬鬣) mountain, pursued and attacked them north to the Tianshan mountain (天山). He captured most people in that cavalry and gained considerable fame. So he attached himself to Xueyantuo and they helped each other. Pusa called himself as “Huo Xielifa” (活頡利發) and set up his palace aside the Dule river (獨樂水, MC duk-lak, Tuγlar). In the third year of Zhen Guan (貞觀, 629), he came to T’ang Court and offered local products. At that time, the Türk Qaghanate collapsed, only the Uyγur and the Xueyantuo were the most powerful. After the death of Pusa, the leader of the Uyγurs, who was Hulu (胡祿) Silifa (eltäβär) Tumidu (吐迷度), and other tribes attacked Xueyantuo, made it broken. So the Uyγurs took the places of the Xueyantuo, and its boundary crossed the Helan mountain (賀蘭山) in the south.[55]

 

Qian Boquan considered Huo Xielifa (活頡利發) and Sijin (irkin) as the Chinese transliterations of ʼīl-ʼīlt(i)b(i)r (Il-Eltäbäβer) and kūl-ʼīrkīn (Kül Irkin). He also believed that ʼīl-ʼīlt(i)b(i)r and kūl-ʼīrkīn corresponded to the early leaders of the Uyγurs, Hulu (胡祿) Silifa (eltäβär) Tumidu (吐迷度) and Huo Xielifa Pusa (菩薩). The prototype of ʼīl-ʼīlt(i)b(i)r could be the creator of the first Uyγur Qaghanate, Hulu Silifa Tumidu. The prototype of kūl-ʼīrkīn could be Huo Xielifa Pusa, because Huo (活) is the transliteration of kūl, which was the common prefix for the titles of Uyγur and Türk leaders.[56] In addition, Sijin (irkin) was the title of Pusa’s father, Shijian, so Qian Boquan deduced that the first Uyγur Qaghanate (which was built by Tumidu) came from the ʼyīšl(i)k, which is “Yaγlaqar” in Chinese literature.[57]

 

Qian Boquan’s deduction is more reasonable than Hamilton’s. Although we cannot verify each detail of Qian’s connection, we know that the record in The Compendium of Chronicles has some relationship with the origin of Uyγur. There are still questions as to where the relationship came from and what “On Uyγur” (“Ten Uyγurs”) actually denotes. When we read the entire record about the Uyγurs in The Compendium of Chronicles, we find that the name Uyγur was created by Oγuz. The name of Uyγur was given by Oγuz as a result of a fight against a heretic.[58] Although the story was reconstructed with several religious elements, it is still possible to discern the basic structure of the story, which was that Uyγur came into being later than Oγuz, and perhaps Uyγur was detached from Oγuz.

 

The recorded fact in The Compendium of Chronicles that the Uyγurs lived along several rivers was also mentioned in the Terhyin inscription erected around 756 ce. Line 4 of the western surface of the monument reads:

 

… And all the people in the four direction were ruled (by me), my enemy lost his good luck… between the eight (rivers), there are my meadow and farmland. Selenge, Orkhon, Tuγla and other five rivers, make me feel happy. In that place, between the two rivers which are Qarγa and Burγu, I am living and grazing.[59]

 

In the above quotation, the Uyγurs lived between eight rivers, so the river was an important element with which the Uyγurs used to describe themselves. The ʼīlʼīlt(i)b(i)r and kūl-aīrkīn in The Compendium of Chronicles, which were considered as the equivalent of Silifa (eltäβär) and Sijin (irkin) by Hamilton and Qian Boquan, were not original names that should have been in a story about the origin of a nation. Silifa (eltäβär) and Sijin (irkin) were used as leaders’ names or even titles from the time of Rouran Qaghanate, and they continued to be used in the Türk Qaghanate and many other tribes. The record in The Compendium of Chronicles may have some relationship with the early history of the Uyγurs recorded in Tang Shu (唐書).[60] It is possible that the record in The Compendium of Chronicles was created to be consistent with the historical memory of the Uyγurs. Using existing names (or titles) is another piece of evidence. Hence, we can tell that the original story was created or rediscovered.[61]

 

From the record about the On Uyγur in the Moyun Chor monument and the detailed and systematic story about the On Uyγur, we can deduce that the name of On Uyγur was created after the foundation of the Uyγur Qaghanate in the middle of the 8^th century. After becoming the ruling clan in the Mobei (漠北), the collective consciousness of the Uyγurs was awakened gradually, which made the people yearn for their own distinct name and story about their origin. The story about the river was mentioned in the Terhyin monument, but was not a systematic story. Until The Compendium of Chronicles of the 14^th century, these elements, such as the rivers, were composed together as a story about the origin of the Uyγurs. Also, we noticed that the Uyγurs were detached from the Oγuz and the story about the origin takes after the detaching.

 

So, we know that the first appearance of On Uyγur was in a simple title, which was created by the Uyγur themselves because of the awakening of their collective consciousness. But the essence of On Uyγur was increased gradually with more and more elements. On Uyγur may have come from the name that the Uyγurs called themselves. “On” (meaning “ten”) could perhaps be the number of the tribes composing Uyγur or merely chosen out of desire to be bigger than Toquz (nine).

 

Although Hamilton made the connection of On Uyγur and Onoγur in regards to phonetics, the difference in the periods of existence of On Uyγur and Onoγur and the meaning of On Uyγur make the association questionable.

 

In order to prove his connection between On Uyγur and Onoγur, Hamilton compared the names of the tribes in On Uyγur and Onoγur and claimed that they were the same. The tribes of On Uyγur’s names are derived from Chinese literature, which are The Biography of the Uyγurs in Old Tang Shu (舊唐書) and New Tang Shu (新唐書). The differences between Old Tang Shu and NewTang Shu are cited below:

 

Old Tang Shu

 

In the time of Kai Yuan (開元, 713–741 ce), the Uyγurs rose gradually… there were eleven Dudu (prefectures of management, 都督) which belong to the clans of  the “Nine Clans.” The first is Yaγlaqar (藥羅葛, MC jjɐk-lɑ-kɑt) which was the family name of the Qaghan; the second was Hou Tuo Ko (胡咄葛, MC ɦuo-tʷɑt- kɑt); the third was Hou Lo Wou (咄羅勿, MC tʷɑt-lɑ-kɑt); the fourth was Mo Ko Si Ki (貊歌息訖, MC mˠɛk-kɑ-sɨk-kɨt); the fifth was A Wou Tchö (阿勿嘀, MC ʔɑ-mjut-ʈˠɛk); the sixth was Qazar/Qasar (葛薩, MC kɑt-sɑ); the seventh was Hou Wou Sou (斛嗢素, MC ɦuk-ʔuot-suo); the eighth was Yo Wo Ko (藥勿葛, MC jjɐk-mjut-kɑt ); the ninth was Hi Ye Wou (奚耶勿, MC ɦei-jja-mjut).[62] There is one Dudu for one tribe. And two clans were incorporated when Zhi Xi Mi (枝悉密, it was the mistake for Basmïl, 拔悉密, MC bʷɑt-sit-mˠit) and Qarluq (葛邏祿) were defeated. And they also got their own Dudu. All the clans were called by the unifying name of Eleven Clans.[63]

 

New Tang Shu

 

The Uyγurs received the place of the “Nine Clans.” The “nine clans” were Yaγlaqar, Hou Tuo Ko, Kiu Lo Wou, Mo Ko Si Ki, A Wou Tchö, Qazar/Qasar, Hou Wou Sou, Yo Wo Ko and Hi Ye Wou. Yaγlaqar was the family name of Uyγur, with Boqu, Hun (渾), Qi Bi (契苾), Bayïrqu, Toŋra, Izgil, which were six equal kinds of barbarians enlisted into their number. Later, when the Basmïl and Qarluq were defeated, there were eleven clans altogether that got their own Dudu and called themselves by the name of Eleven Clans.[64]

 

The Old Tang Shu and New Tang Shu records show two groups of names: “The inside nine clans” and “The outside nine clans.” The outside nine clans were Bayïrqu, Toŋra, Izgil, etc. The inside nine clans was the group whose names began with Yaγlaqar, which only appeared in these two instances in the Chinese literature. The inside nine clans were the nine clans belonging to the Uyγurs; the outside nine clans were the clans different from the Uyγurs but perhaps shared similar statues with the Uyγurs.[65]

 

Akio Katayama has a different opinion about the division of the clans. He believed the two groups of names referred to the same group of clans. The groups of names beginning with “Yaγlaqar” were the family names of the clan leaders; Bayïrqu, Toŋra, Izgil, etc. were the formal names used by other clans.[66]

 

In any case, it can be agreed upon that whether it be the division of inside and outside nine clans, or family names and formal names, the two groups of names were two different systems of names, which cannot be mixed up. When Hamilton had compared the names of the tribes under the name of On Uyγur, he mixed them up and placed them in one group, which is incorrect.

 

Another group of names that Hamilton used for comparative purposes were the names of the ten rivers in The Compendium of Chronicles, which we discussed earlier. We said the essence of “On Uyγur” was increased gradually with more elements—the river was one of them. In the quotation from The Compendium of Chronicles, after the description of the ten rivers, it is stated that:

 

There were nine tribes living in the previous three rivers, and five tribes living in the following four rivers. The tribe living along the ninth river qamlančū was called ʼūng, and the one living in the riverside of ʼūtīkān was called qamaq-ātikūz. Besides the tribes living along these rivers, there were still 122 tribes living in the same area, but their names were unknown.

 

The author of the material did not consider the ten rivers to be the names of tribes, because he discussed the tribes after his description of the ten rivers. Despite the fact that the author did not list all the names of the tribes, we still cannot consider the rivers’ names as the names of the tribes. The most obvious example of misinterpretation was that Hamilton regarded the name of Tūlār as Bulγar without considering Toγla, the famous river in Mongolia.[67]

 

Considering the mixing up of the two groups of names and using the name of rivers as the tribes’ names, we can confidently assert that Hamilton’s comparison and identification of the tribes’ names did not help his identification of the On Uyγur with the Onoγur.

4. CONCLUSION

 

1. According to the meaning and transliteration of Tiele, it is reasonable to conclude that the Oγuric tribes belonged to Tiele, despite the claims that Tiele in Turkic rendered Türk or tegrek. All scholars are in agreement that all of the tribes of the Tiele spoke Turkic. Therefore, according to this definition of Tiele, the Oγuric tribes, which were of the Turkic language branch, belonged to Tiele.

 

2. It is credible that Toquz Oγuz of ancient Turkic inscriptions corresponds to “The nine clans” (九姓) in Chinese literature, which refers to the main nine clans in the clan union. Hamilton’s conclusion that Oγuz came from Oγuš is questionable.

 

3. We can deduce that the language used by the Oγuric tribes was an earlier form of Turkic. Common Turkic (which is different from Oγuric) may be formed from the influence of various other languages during the scattering of Turkic-using tribes and the internal Turkic development. As Sinor said, “The time of the scattering and migration of the tribes using Turkic was much earlier and longer,” but the time and the details cannot yet be confirmed.

 

4. On Uyγur was formed after the creation of the Uyγur Qaghanate, which was in the middle of the 8^th century. On Uyγur was a simple title in the beginning, which was constructed by the Uyγurs themselves at the dawn of their collective consciousness. However, the meaning of “On Uyγur” was compounded gradually over time. We then conclude that On Uyγur came from the name that the Uyγur used to refer to themselves. On (ten) is possibly the number of tribes of Uyγur or was simply chosen as a name because it is larger than Toquz (nine).

 

5. Although Hamilton made the connection between On Uyγur and Onoγur in terms of phonetics, the difference in time periods of On Uyγur and Onoγur and the meaning of “On Uyγur” makes the association questionable. The comparison and identification of the tribes’ names made by Hamilton cannot support his claim that On Uyγur and Onoγur are the same tribes.

 

APPENDIX

 

Tiele Zhuan in the Sui Shu (the Monograph on the Tiele in the Sui Shu):[68]

 

The ancestors of the Tiele were the descendants of the Xiongnu.[69] There were many clans among the Tiele who were compactly distributed along the valley from the east of the Western Sea.[70]

 

In the North of the Tola river,[71] there were Boqut (Pugu, 僕骨, MC buk-kuot)[72] Toŋra (Tongluo，同羅, MC duŋ-lɑ),[73] Uyγur (Weihe, 韋紇, MC ɦʷɨi- ɦet), Bayirqu (Bayegu, 拔也古, MC bʷɑt-jja-kuo) and Fuluo (覆羅, MC phək-lɑ), whose leaders were all called Irkin (Sijin, 俟斤, MC ɖʐɨ-kɨn) by themselves. And there were other clans such as Mengchen (蒙陳, MC muŋ-ɖin), Turuhe (吐如紇, MC thuo-ɲjɷ-ɦet), Siqit (Sijie, 斯結, MC sie-ket), Qun (Hun, 渾, MC ɦuon) and Huxue (斛薛, MC ɦuk-siɛt). These clans had a powerful army of almost 20,000 men.

 

In the west of Hami (Yiwu), North of Karashahr (Yanqi), and close to Aqtagh (Bai Mountain)[74] there were Qibi (契弊, CE khet-biɛi),[75] Boluozhi (薄落職, CE bɑk-lɑk-tɕɨk), Yidie (乙咥, CE ʔˠit-tet),[76] Supo (蘇婆, CE suo-bʷɑ), Nahe (那曷, CE nɑ-ɦɑt), Wuhuan (烏讙, CE ʔuo-hʷjɐn),[77] Hegu (紇骨, CE ɦet-kuot), Yedie (也咥, CE jja-tet)[78] Yunihuan (於尼讙, CE ʔuo-ɳi-hʷjɐn)[79] and so on. These clans had powerful army of almost 20,000 men.

 

In the Southwest of Altai Mountain (Jin Mountain), there were Xueyantuo (薛延陀, CE siɛt-jiɛn-dɑ),[80] Dieleer (咥勒兒, CE tet-lək-ɲie), Shipan (十槃, CE ʥipbʷan), Daqi (達契, CE thɑt-khet) and so on, which have an army of more than 10,000 men.

In the north of Samarkand, close to Ade river,[81] there were Hedie (訶咥, CE hɑtet),[82] Hejie (曷嶻, CE ɦɑt-dzɑt),[83] Bohu (撥忽, CE pʷɑt-huot),[84] Bigan (比干, CE pi-kɑn),[85] Juhai (具海, CE gju-həi), Hebixi (曷比悉, CE ɦɑt-pi-sit),[86] Hecuosu (何嵯蘇, CE ɦɑ-ʣɑ-suo), Bayewei (拔也未, CE bʷɑt-jja-mʷɨi),[87] Keda (渴達, CE khɑt-thɑt)[88] and so on, which have an army of more than 30,000 men.

 

In the east and west of Deyihai (得嶷海),[89] there were Sulujie (蘇路羯, CE suoluo-kjɐt),[90] Sansuoyan (三索咽, CE sɑm-sɑk-ʔet),[91] Miecu (蔑促, CE mettshjuok),[92] Longhu (隆忽, CE lju-huot)[93] and so on, more then 8,000 men.

 

In the east of Fulin (拂菻), there were Enqu (恩屈, CE ʔən-kjut),[94] Alan (阿蘭, CE ʔɑ-lɑn),[95] Beirujiuli (北褥九離, CE pək-nuok-kɨu-lei), Fuwenhun (伏嗢昬, CE bɨu-ʔʷˠɛt-huon)[96] and so on, almost 20,000 men.

In the South of Northern Sea, there were Dubo (都波, CE tuo-pʷɑ)[97] and so forth.

 

Although there were so many different names of the clans, they were all called Tiele as a whole. There was no ruler among them, and they belonged to the Eastern and Western Türks separately. They lived in unsettled places, and moved along with the water and grass.

 

They were good at shooting on horseback, and were fierce and cruel, especially greedy. They live on plundering. The clans close to the west do several kinds of cultivating, and breed more cattle and sheep than horses.

 

Since the establishing of the Türk state, the Tiele help the Türks by participating in battles everywhere, and subdue all the groups in the North.

 

At the end of the Kaihuang (開皇, reign period: 581–600), Jinwang (晉王)[98] led a Northern expedition, which made Qimin Qaghan submit, and smashed Bujia Qaghan. Because of this, the Tiele dispersed.

 

At the first year of Daye (大業, 605), Chuluo (處羅) Qaghan[99] conquered the Tiele tribes, and imposed a heavy imposition on these tribes. He also suspected the tribe of Xueyantuo (薛延陀) and was afraid they would revolt. So he gathered hundreds of their leaders and killed them all. These tribes all rose in rebellion against Chuluo, and chose the eltäβär (俟利發, Silifa) Irkin (俟斤, Sijin) Qibi (契弊, a tribe’s name) Geleng (歌楞) as Yiwuzhen (易勿真) Mohe (莫何) Qaghan living in the Tanhan (貪汗) mountain. They also elected Yezi (也咥), who was the Irkin (Sijin) in Xueyantuo, and they called him the little Qaghan.

 

As Chuluo Qaghan became frustrated, Mohe (莫何) Qaghan began to become powerful. Mohe was very brave and unsurpassed, and had great fortitude. So he was welcomed by his people, and was feared by the neighboring country. The countries of Hami (Yiwu, 伊吾), Qoco (Gaochang, 高昌, MC kɑu-tɕhjɐŋ), Karashahr (Yanqi, 焉耆, MC ʔˠiɛn-gˠi) all attached themselves to him.

 

Their customs were mostly like those of the Türks. The differences were that the husband should stay in his wife’s family, and could not go home until the birth of his children. Also the dead were to be buried.

 

In the third year of Daye (607), Tiele sent an envoy and tribute to the court, and never stopped contact from that year.

REFERENCES

Cen Zhongmian 1958: Tu Jue Ji Shi (The History of the Turks), Beijing: Zhong Hua Shu Ju.

Duan Lianqin 1988: Dingling, Tie le and Gaoche, Shanghai: Shanghai Renmin Press.

Feng Chengjun 1962: Xi Yu Nan Hai Shi Di Kao Zheng Yi Cong, Beijing: Shang Wu Yin Shu Guan.

——1980: Xi Yu Di Ming (The Toponym of Xi Yu), Beijing: Zhong Hua Shu Ju.

Geng Shimin 1980: Wu Gu Si Ke Han De Chuan Shuo (The Tales of Oghuz Khan), (Xinjiang: Xin Jiang Ren Min Press).

——2005: Gu Dai Tu Jue Wen Bei Ming Yan Jiu (Research on the Ancient Turkic Inscriptions) (Beijing: Zhongyang Minzu Daxue Press.

Lin Gan 2007: Tu Jue Yu Hui He Shi (The History of the Turks and Uyγurs) (Hohhot: Nei Meng Gu: Nei Meng Gu Renmin Press).

Liu Ge 1991: “E Er Hun Tu Jue Wen Bei Min Yu E Er Hun Hui Hu Shi (The Orkhon Turkic Inscriptions and the History of the Uyγurs on the Orkhon),” in Xin Jiang Wen Wu, No. 3, pp. 56–63.

Ma Changshou 1957: The Turk and the Turk Khanate (Shanghai: Shanghai Renmin Press).

Rui Chuanming 1998: Gu Tu Jue Bei Ming Yan Jiu (Research on the Ancient Turkic Inscriptions) (Shanghai: Shanghai Gu Ji Press).

Shiratori Kurakichi 1938–1940: Sai Wai Shi Di Yi Chong (The Collective Translations of the History and the Geography out of the Strategic Pass), trans. Fang Zhuangyou, (Beijing: Shang Wu Yin Shu Guan).

——1923: “The research on the origin of Xiongnu,” in Shi Xue Za Zhi, No 18. Zhou Weizhou 1983: Chi le and Rou ran (Shanghai: Shanghai Renmin Press).

Onogawa Hidemi 1978: “The Research on Tie le,” trans. Wang Enqin, in Min Zu Shi Yi Wen Ji (Collection of Translated articles on the History of Nationalities), Vol. 6.

GLOSSARY
Ashina 阿史那		On Uyγur 十姓回鴨或十姓回結
Akio Katayama 片山章雄		Pobu 婆卜
Abe Takeka 安部健大		Pusa 菩薩
A Wou Tchö 阿勿喷		Qočo, Kao-tch'ang 高昌
Boqu 僕固		Qibi 契弊,契法
Bašmül 拔悉密		Qazar/Qasar 葛薩
Bayïrqu 拔野古或拔固		Qian Boquan 錢伯泉
Bai Na Ben 百衲本		Qarlug 葛邏祿
Cen Zhongmian 岑仲勉		Shiratori Kurakichi 白鳥庫吉
Chile 敕勒		Sui Shu 隋書
Chuluo 處羅		Silifa 侯利 發, Eltäßär
Daduse 大度設		Sijin 侯斤, irkin
Da Ye 大業		Shijian 時健
Di 狄		Shekui Qaghan 射匮可汗
Di li 狄曆		Toŋra 同羅
Dingling 丁零,丁靈		Tumidu 吐迷度
Duan Lianqin 段連勤		Tian Mountain 天山
Dule river 獨樂水, Tuylar		Tanhan Mountain 貪汗山
Five clans of Wuman 五姓烏蠻		Tümän, Tumen 土門
Gaoche 高車		Three clans 三姓
Geng Shimin 耿世民		Ta Shili 踏實力
Geleng 歌楞		Tang Hui Yao 唐會要
Great Desert 大漠		The North of Desert 漠北
Golden Mountain 金山		Tiele 鐵勒
High Cart, Gaoche 高車		Toru Haneda 羽田亨
Hi Ye Wou 奚耶勿		Tulishi 突利失
Hou Wou Sou 斛唱素		Tujue 突厥
Hou Tuo Ko 胡昭葛		Wuhu 烏護
Hou Lo Wou 咄羅勿		Wuhe 烏紇
He Lan mountain 賀蘭山		Wang Jingru 王靜如
Hu Lu 胡祿		Western Region 西域
Huo Xie Li Fa 活頡利發		Wou Houan 烏桓
Hun 渾		Wouwan 烏丸
High Cart 高車		Weihe 韋結
Huihe 回結		Wu Luohun 烏羅軍
Illig Qaghan 頡利可汗		Xian Qing 顯慶
Izgil 思結		Xueyantuo 薛延陀
Rouran 柔然		Yantuo 延陀
Kai Yuan 開元		Yuan Chi 元池
Li Jing李靖		Yuanhe 袁結
Ma Changshou 馬長		Yili Qaghan 伊利可汗
Moula 謀刺,謀落		Yi Wu Zhen 易勿真
Malie mountain 馬鬣山		Ye Zi 也喔
Mo Ko Si Ki 貊歌息訖		Yin Mountain 陰山
New T'ang Shu 《新唐書》		Yugushe 欲穀設
Northern Yard Government 北庭		Yaγlaqar 藥羅葛
Nine clans of Tiele 九姓鐵勒或鐵勒九姓		Yo Wo Ko 藥勿葛
North Wei Dynasty 北魏		Zhou Weizhou 周偉洲
Nine clans 九姓		Zhen Guan 貞觀
Qarluq 葛邏祿		Zhen Guan 貞觀
Onogawa Hidemi 小野川秀美		Zhiximi 枝悉密

BIBLIOGRAPHY

Abe Takeka, 1986: Xi Huihu Guo Shi De Yanjiu (The Research of the Western Uyγur Qaghanate), trans. Song Suying, Ling Song Chun and Xu Bofu (Urumqi: Xinjiang Renmin Press).

 

Akio Katayama 1986: “Guanyu Toquz Oγuz He Jiuxing De Ruogan Wenti (Several Questions about Toquz Oγuz and Nine Clans),” trans. Xing Yulin, Liu Shize, in Shi Jie Min Zu (World Ethno-National Studies), No.5, pp. 44–47

 

Alemany, Agustí: “Early Byzantine Sources on the Oghuric Tribes in the Northern Black Sea Area,” personal communication from the author.

 

——2000: Sources on the Alans, A Critical Compilation. (Leiden: Brill).

 

Allsen, Thomas T. 2003. Culture and Conquest in Mongol Eurasia (Cambridge: Cambridge University Press)

 

Cen Zhongmian 1958: Tujue Ji Shi (The Collective History of the Turks) (Beijing: Zhong Hua Shu Ju).

 

Duan Lianqin 1988: Dingling Gaoche Yu Tiele (Dingling, Tiele and Gaoche) (Shanghai: Shanghai Renmin Press).

 

Feng Chengjun 1962: Xi Yu Nan Hai Shi Di Kao Zheng Yi Cong (Collection of The Translations of Research on the History and Geography of the Western Region and Southern Sea) (Beijing: Shang Wu Yin Shu Guan).

 

——1980: Xi Yu Di Ming (the Toponym of Xi Yu) (Beijing: Zhong Hua Shu Ju).

 

Gabain, A. Von. 1974: Alttükische Grammatik (Wiesbaden: Otto Harrassowitz Verlag).

 

Geng Shimin 1980: Wu Gu Si Kehan De Chuanshuo (The Tales of Oghuz Khan), (Urumqi, Xinjiang: Xin Jiang Ren Min Press).

 

——2005: Gu Dai Tujue Wen Bei Ming Yan Jiu (The Research on the Ancient Turkic Inscriptions) (Beijing: Zhongyang Minzu Daxue Press).

 

Golden, P. B. 2006: “Some Thoughts on the Origins of the Turks and the Shaping of the Turkic Peoples,” in Contact and exchange in the ancient world, edited by Victor H. Mair (Honolulu: University of Hawai’i Press), pp. 136–157.

 

——1994: “The Peoples of the Russian Forest Belt,” in Sinor, D. (ed.), The Cambridge History of Early Inner Asia (Cambrige: Cambridge University Press), pp. 229–255.

 

Hua Tao 2000: Xiyu Lishi Yanjiu (Ba dao Shi shiji) (The research on the Western Region’s history, the eighth to tenth Century) (Shanghai: Shanghai Guji Press).

 

Hamilton 1962: “Toquz-Oγuz et On-Uyγur,” in Journal Asiatique, CCL/1, pp. 242–63.

 

Li Yanshou 1983: Bei Shi, BNB (Beijing: Zhong Hua Shu Ju).

 

Liu Xun et al 1986: Jiu Tang Shu (Old Tang Shu) (BNB, Beijing: Zhong Hua Shu Ju).

 

Lin Gan 2007: Tujue Yu Hui He Shi (The History of Turk and Uyγur), (Hohhot: Nei Meng Gu: Nei Meng Gu Renmin Press).

 

Liu Ge, 1991: E Er Hun Tujue Wen Beiming Yu E Er Hun Huihu Shi (The Orkhon Turkic Inscriptions and the History of the Uyγurs on the Orkhon), in Xin Jiang Wen Wu, No. 3, pp. 56–63.

 

Macartney, C.A. 1944: “On the Greek Sources for the History of the Turks in the Sixth Century,” Bulletin of the School of Oriental and African Studies 11/2, p. 266–275.

 

Ma Changshou 1957: Tujue Ren He Tujue Hanguo (The Turks and the Turk Qaghanate) (Shanghai: Shanghai Renmin Press, 1957).

 

Moravcsik, G. 1930: “Zur Geschichte der Onoguren”Ungarische Jahrbücher 10, 53–90. 

 

Ou Yang Xiu and Song Qi 1987: Xin Tang Shu (New Tang Shu), BNB (Beijing: Zhong Hua Shu Ju).

 

Onogawa Hidemi 1978: “Tiele Kao (The Research on the Tiele),” trans. Wang Enqin, in Min Zu Shi Yi Wen Ji (Collection of Translated Articles on the History of Nationalities), Vol. 6.

 

Qian Boquan 2005: Weiwuer Ren De Zuyuan Chuanshuo Yanjiu (Research on the tales of the Origin of the Uyγurs), in Ka Shi Shi Fan Xue Yuan Xue Bao, No. 5, 2005, pp. 24–29.

 

Ramstedt, G. J. 2004: Einführung in die altaische Sprachwissenschaft, trans. Zhou Jianqi (Hohhot, Nei Meng Gu: Nei Meng Gu Jiao Yu Press.

 

Rashid al-Din 1983: The Compendium of Chronicles (Shi Ji), trans. Yu Dajun and Zhou Jianqi (Beijing: Shangwu Yinshu Guan).

 

Rui Chuanming 1998: Gu Tujue Bei Ming Yan Jiu (Research on the Ancient Turkic Inscriptions) (Shanghai: Shanghai Gu Ji Press).

 

Shiratori Kurakichi 1938–1940: Sai Wai Shi Di Yi Chong (The Collective Translation of the History and the Geography out of the Strategic Pass), trans. Fang Zhuangyou, (Beijing: Shang Wu Yin Shu Guan).

 

——1923: “The Research on the Origin of the Xiongnu,” in Shi Xue Za Zhi, No 18.

 

Sima Guang 1956: Zi Zhi Tong Jian (Beijing: Zhong Hua Shu Ju).

 

Sinor, D. 1946–7: “Autour d’une Migration de Peuples au Ve Siècle,” Journal Asiatique 235, pp. 1–77.

 

—— (ed.), 1990: The Cambridge History of Early Inner Asia, (Cambridge: Cambridge University Press).

 

—— 1997: “Early Turks in Western Central Eurasia,” in Studia Ottomanica: Festgabe für György Hazai zum 65. Geburstag, ed. Barbara Kellner-Heinkele and Peter Zieme, (Wiesbaden: Otto Harrassowitz Verlag).

 

Toru Haneda, 1957: “The Research on the Relationship of the Toquz Uyγur and Toquz Oγuz,” in Dong Yang Xue Bao, Vol 9, No 1.

 

Wei Shou 1984: Wei Shu, BNB (Beijing: Zhong Hua Shu Ju).

 

Wei Zheng 1982: Sui Shu, BNB (Beijing: Zhong Hua Shu Ju).

 

Wang Pu 1955: Tang Hui Yao, BNB (Beijing: Zhong Hua Shu Ju).

 

Wang Jingru, 1938: “Tujue Wen Huihe Ying Wu Wei Yuan Pi Jia Ke Han Bei Yi Shi” (Translation and Explanation on the Turkic Monument of the Uyγur Intelligent Brave Majestic Bilge Qaghan), in Fu Ren Xue Zhi, Vol 7, No 1–2.

 

Zhou Weizhou 1983: Chile Yu Rouran (Chile and Rouran) (Shanghai: Shanghai Renmin Press).

ABBREVIATIONS

BNB = Bai Na Ben, 百衲本.

 

RATI = Geng Shimin: Gu Dai Tujue Wen Beiming Yanjiu (Research on the Ancient Turkic Inscriptions) (Beijing: Zhongyang Minzu Daxue Press, 2005).

 

RT = Onogawa Hidemi, “Tiele Kao (The Research on the Tiele),” trans. Wang Enqin, in Min Zu Shi Yi Wen Ji (Collection of Translated Articles on the History of Nationalities), Vol 6, 1978.

 

STOTSTP = Peter B. Golden, “Some Thoughts on the Origins of the Turks and the Shaping of the Turkic Peoples,” in Victor H. Mair (ed.) Contact and Exchange in the Ancient World, (Honolulu: University of Hawai’i Press, 2006).

 

TOOU = Hamilton: “Toquz-Oγuz et On-Uyγur,” in Journal Asiatique CCL/1 (1962) pp. 23–63.

 

SQTONC = Akio Katayama, “Guanyu Toquz Oγuz He Jiuxing De Ruogan Wenti (Several Questions About the Toquz Oγuz and Nine Clans),” trans. Xing Yulin, Liu Shize, in Shi Jie Min Zu (World Ethno-National Studies), No. 5 (1986), pp. 44–47.

---

[1] The original word in the manuscript of Priscus was Οὔγωροι, which the scholars thought was a mistake for Ουγωρ. See James Hamilton, “Toquz-Oγuz et On-Uyγur” (TOOU for short in the following annotations), in Journal Asiatique, 1962, p. 33.

[2] James Hamilton, TOOU, p. 34.

[3] D. Sinor, “Autour d’une Migration de Peuples au Ve siècle” in Journal Asiatique, 1946–1947, p. 5; Agustí Alemany, “Early Byzantine Sources on the Oghuric Tribes in the Northern Black Sea Area,” personal communication given by the author.

[4] Peter Golden,“The Peoples of the Russian forest belt,” in The Cambridge History of Early Inner Asia, Cambridge: Cambridge University Press, 1990, pp. 234–235.

[5] MC = Middle Chinese. On the Middle Chinese reconstruction, see Pan Wuyun: Hanyu Lishi Yinyun Xue (Chinese Historical Phonology), Shanghai: Shanghai Jiaoyu Press, 2000. All reconstructions of the Middle Chinese in this article are from Pan Wuyun.

[6] Duan Lianqin, Dingling Gaoche Yu Tiele (Dingling, Tiele and Gaoche), Shanghai: Shanghai Renmin Press, 1988, pp. 1–21. Cf. J. V. Stalin, Marxism and the National Question in J.V. Stalin, Works (Moscow: Foreign Languages Publishing House, 1953), 2, p. 307: “A nation is a historically constituted, stable community of people, formed on the basis of a common language, territory, economic life, and psychological make-up manifested in a common culture.”

[7] Toru Haneda, “The research on the relationship of the Toquz Uyγur and Toquz Oγuz,” in Dong Yang Xue Bao, vol. 9, No 1, 1957.

[8] Ma Changshou, Tujue Ren He Tujue Hanguo (the Turk and the Turk Qaghanate), Shanghai: Shanghai Renmin Press, 1957, pp. 2–5.

[9] The translation mainly comes from the website: [does not exist anymore], and changed sightly according to the Chinese translation of Geng Shimin. Geng Shimin, Gudai Tujue Wen Beiming Yanjiu (Research on the Ancient Turkic Inscriptions, RATI for short in the follow annotations.), Beijing: Zhongyang Minzu Daxue press, 2005, p. 124. The name is also read as Tölis.

[10] Cen Zhongmian, Tujue Ji Shi (The Collective History of Turk), Beijing: Zhong Hua Shu Ju, 1958, pp. 894-895.

[11] Wang Jingru, “Tujue Wen Huihe Ying Wu Wei Yuan Pi Jia Ke Han Bei Yi Shi” (Translation and Explanation on the Turkic Monument of the Uyγur Intelligent Brave Majestic Bilge Khan),” in Fu Ren Xue Zhi, Vol. 7, No1–2, 1938.

[12] “Xue Yan Tuo Zhuan” (the biography of Xue Yan Tuo), in New Tang Shu, BNB edition, Juan 217.

[13] Peter B. Golden, “Some Thoughts on the Origins of the Turks and the Shaping of the Turkic Peoples” (= STOTSTP in following citations), in Contact and exchange in the ancient world, edited by Victor H. Mair, Honolulu: University of Hawai’i Press, 2006, pp.136–140.

[14] Shiratori Kurakichi, “The research on the origin of Xiongnu,” in Shi Xue Za Zhi, No. 18, 1923.

[15] Hamilton, TOOU, p. 51.

[16] State Hermitage Museum: Southern Siberia/Pazyryk.

[17] Hamilton, TOOU, p. 51.

[18] Onogawa Hidemi, “The Research on Tiele (Tiele Kao)” (= RT in following citations), trans. Wang Enqin, in Min Zu Shi Yi Wen Ji (Collection of Translated articles on the History of Nationalities), vol. 6, 1978, pp. 29–36.

[19] Geng Shimin, RATI, p. 3.

[20] Zhou Weizhou, Chile Yu Rouran (Chile and Rouran), Shanghai: Shanghai Renmin Press, 1983, p. 7.

[21] See Onogawa Hidemi, RT, p. 29.

[22] See Onogawa Hidemi, RT, p. 36.

[23] Golden, STOTSTP pp. 136–140.

[24] The quotation was translated from the Chinese edition of G. J. Ramstedt, Einfuhrung in die altaische Sprachwissenschaft. Mémoires de la Société Finno-Ougrienne, 104/1–3, Helsinki, 1957–1966, trans.Zhou Jianqi, Hohhot: Nei Meng Gu: Nei Meng Gu Jiao Yu Press, 2004, pp. 9–10.

[25] Peter Golden, “The Peoples of the Russian Forest Belt,” in The Cambridge History of Early Inner Asia, Cambridge: Cambridge University Press, 1990, pp. 234–235.

[26] Golden, STOTSTP, pp.138–140.

[27] The division into Oγuric (West Old Turkic) and Common Turkic (East Old Turkic) has now been dated to after the 3^rd century bce and up to the 3^rd–5^th century ce, see A. Róna-Tas and Á. Berta, West Old Turkic. Turkic Loanwords in Hungarian, Turcologica 84 (Wiesbaden: Harrassowitz, 2011), II, pp. 1112–1113.

[28] D. Sinor, “Early Turks in Western Central Eurasia,” in Studia Ottomanica: Festgabe für György Hazai zum 65. Geburstag, Herausgegeben von Barbara Kellner-Heinkele und Peter Zieme, Wiesbaden: Otto Harrassowitz Verlag, 1997, p.167.

[29] Hamilton, TOOU, p. 38.

[30] See note 27.

[31] Hamilton, TOOU, p. 25.

[32] Tang Hui Yao, Juan 100, Beijing: Zhong Hua Shu Ju, 1955, pp. 1787–1788. The “three clans” also appeared twice in the New Tang Shu as “three clans” and “three clans Yabγu.” We know that Yabγu was a high ranking Inner Asian title used by Turks and others. Yabγu was added by the Chinese.

[33] Geng Shimin, RATI, p. 195.

[34] Geng Shimin, RATI, pp. 121, 124.

[35] New Tang Shu, BNB edition, Juan 222.

[36] The Chinese recorded two dates as 646 CE and 647 CE.

[37] One record in Tang Hui Yao used “eleven clans” to substitute for “nine clans” to describe the same thing. Tang Hui Yao, Juan 94.

[38] Akio Katayama,“Several Questions about Toquz Oγuz and Nine Clans” (= SQTONC in following citations), trans. Xing Yulin, Liu Shize, in Shi Jie Min Zu (World Ethno-National Studies), No. 5, 1986, pp. 44–47.

[39] Meanwhile, Professor Peter Golden made a detailed and convincing argumentation about the origin of the words “Oγuz” and “Oγur.” He found that they all derived from the root oğ or oq ~ uq, an early term for a kinship grouping. The Chinese character “xing 姓” accurately rendered these words. For more information, see Peter B. Golden: “Oq and Oğur ~ Oğuz” in Turkic Languages (forthcoming).

[40] The identification between Oγuz and Wuhu was not agreed upon by some scholars, such as Chavannes, who considered Wuhu as Uyγur. However, the identification was denied by Hamilton in TOOU (pp. 30–31). Wang Jingru and Liu Meisong also denied this identification on the basis of the location of Wuhu and Uyγur. Many scholars such as Wang Jingru, Liu Meisong and Duan Lianqin all considered Wuhu as Oγuz. Liu Meisong, Liang Tangshu Huihe zhuan Huihu zhuan shuzheng (The Commentary on the Monograph on the Uyγur in the New and Old Tangshu), Beijing: Zhongyang Minzu Xueyuan Press, 1988, p. 5–7; Duan Lianqin, Dingling Gaoche Yu Tiele (Dingling, Tiele and Gaoche), Shanghai: Shanghai Renmin Press, 1988, p. 322.

[41] Hamilton, TOOU, pp. 31–32.

[42] Hamilton, TOOU, p. 38. Von Gabain had mentioned that the «diphtongue uy» was not a real diphthong. A. Von. Gabain, Alttükische Grammatik, Wiesbaden: Otto Harrassowitz Verlag, 1974, p. 51.

[43] The Bilge Qaghan inscription, the translation was mainly from [URL does not exist anymore], and the translation was changed a little according to the Chinese translation of Geng Shimin. Geng Shimin, RATI, p. 163.

[44] According to the systemic research regarding the Tughuzghuz by Hua Tao (华涛), he considered that the “Tughuzghuz” in the Arabic and Persian literature before the ninth century meant the Uyγur Qaghanate in Mobei (漠北), and after that the “Tughuzghuz” sometimes also means the Uyγurs in Qočo. Hua Tao, Xiyu Lishi Yanjiu (Ba dao Shi shiji) (Research of the western region’s history, the eighth to tenth century), Shanghai: Shanghai Guji Press, 2000, pp. 101–133.

[45] “The Moyun Chor monument,” the Turkic and the translation was mainly from [does not exist anymore], and the translation was changed slightly according to the Chinese translation of Geng Shimin. Geng Shimin, RATI, p. 194.

[46] G.J. Ramstedt, “Zwei Uighurische Runeninschriften in der Nor-Mongolei,” Journal de la Société finno-ougrienne, XXX, 3, 1913. See Hamilton, TOOU, p. 59.

[47] See Liu Ge: E Er Hun Tujue Wen Beiming Yu E Er Hun Huihu Shi (the Orkhon Turkic Inscriptions and the History of the Uyγur in the Orkhon zone), in Xin Jiang Wen Wu, No. 3, 1991, pp. 56–63.

[48] Akio Katayama, SQTONC, p. 52.

[49] Hamilton, TOOU, pp. 39–40.

[50] The quotation was translated from the Chinese edition of The Compendium of Chronicles. Rashid al-Din, The Compendium of Chronicles (Shi Ji), trans. Yu Dajun and Zhou Jianqi, Beijing: Shangwu Yinshu Guan, 1983, p. 239. It was compared with the Persian edition: Jāmiʻat-Tawārīkh, ed. M. Rowshan and M. Mūsawī (Tehran: Nashr Alburz, 1373/1994), I, pp. 139–140.

[51] Sui Shu, Tiele Zhuan, Juan 84.

[52] Hamilton, TOOU, pp. 40–41.

[53] Old Tang Shu, Tiele Zhuan, BNB edition, Juan 199.

[54] Pusa (菩萨) is a name of Buddha. But most scholars such as Lin Gan and Geng Shimin do not think that the Uyγurs in their imperial period in Mongolia were Buddhists. The Uyγurs were Shamanists at the beginning, and converted to Manicheism in the 8th century.

[55] Xin Tang Shu Hui Hu Zhuan (the biography of Uyγur in the New Tang Shu), Juan 217.

[56] Qian Boquan mentioned that kūl means “big lake.” Professor Peter Golden reminded me that there is a different, homophonous word in Turkic, probably denoting the idea of “famous” that is used in titles. And it was only when this ancient meaning was forgotten that Turks made the folk etymology comparing it to köl. So the point of Qian Boquan is questionable.

[57] See Qian Boquan, Weiwuer Ren De Zuyuan Chuanshuo Yan Jiu (The Research on the Tales of the Origin of the Uyγur), in Kashi Shifan Xueyuan Xuebao, No. 5, 2005, p. 24–29.

[58] See Rashid al-Din, The Compendium of Chronicles (Shi Ji), trans. Yu Dajun and Zhou Jianqi, Beijing: Shangwu Yinshu Guan, 1983, pp. 238–9. The story about Oγuz in The Compendium of Chronicles was revised from The Legend of Oγuz Khan. The earliest existent version of The Legend of Oγuz Khan is kept in the National Library in Paris. Its number is Suppl. turc, 1001. The Legend of Oγuz Khan, trans. Geng Shimin, Xinjiang Renmin Press, 1980.

[59] The quotation is based on the Chinese translationof the inscription by Geng Shimin:Geng Shimin, RATI, p. 208.

[60] However, the authors of The Compendium of Chronicles paid much attention to searching and using the living oral materials. In the preface of The Compendium of Chronicles, Rashid al-Din said that “All the oral legends and tales of every nationality should be respected… the statements (in this book) were the records from the famous book of every nationality, which were not changed, revised or amended by ourselves. These records all came from the oral legends and tales in every nationality, which were familiar to every family, and from the statements of the authentic scholars and wise men in every nationality” (Rashid al-Din, The Compendium of Chronicles, p. 93). Therefore, the original story about the Uyγurs may come from the oral tales or records of the Uyγurs. Meanwhile, the editor of the Tang Shu may have also received the original story in the same way.

[61]There was another original story about the Uyγurs recorded in Juvayni’s The History of the World-Conqueror and two tablets unearthed in China (Yi duhu Gaochang wang shi xun bei 亦都护高昌王世勋碑 and Liaoyang deng chu xing zhongshusheng zuo chengxiang yi nian zhen gong shendao bei 辽阳等处行中书省左丞相亦辇真公神道碑). It is a tale about that the first leader of the Uyγur who was said to have been born between two trees. However, this story was not recorded in the text of The Compendium of Chronicles, because the story had some features of Manicheism and was preposterous to the authors. Juvayni, The History of the World-Conqueror, trans. He Gaoji, Zhongguo Renmin Daxue Press, 2012, pp. 43–45.

[62] The spelling of the names of nine clans comes from Hamilton. Hamilton, TOOU, pp. 41–44.

[63] Jiu Tang Shu Hui He Zhuan (The Biography of Uyγur in the Old Tang Shu), BNB edition, Juan 195.

[64] Xin Tang Shu Hui Hu Zhuan (The biography of Uyγur in the New Tang Shu), BNB edition, Juan 217.

[65] See Lin Gan, Tujue Yu Huihe Shi (The History of Turk and Uyγur), Hohhot: Nei Meng Gu: Nei Meng Gu Renmin Press, 2007, pp.157–8.

[66] Akio Katayama, SQTONC, pp. 50–53.

[67]In the Chinese edition, the translator mentioned that Doctor E. Bretschneider, Medieval Researches from Eastern Asiatic Sources (1888, reprint: New York, 1967), pp. 240, 255, 259, 260, also did text research on adar, qamlančū and ʼūtīkān, and had confirmed the three were river and place-names.

[68] The original Chinese version of this translation is based on the version of Zhonghua Shuju in 1973. Because there was no punctuation in the traditional Chinese book, the names of the clans in the biography are not the same when read by different scholars.

The Tiele appeared in Chinese literature from the 6^th century to 8^th century, and was recorded as Tiele or Chile (敕勒). The New Tang Shu (新唐書) claims Tiele is mistaken from Chile. Most scholars believe the two names are just different Chinese characters to describe the same Turkic word; the same is true with Di (狄), Dili (狄曆), Dingling (丁零, 丁靈) and Tujue (Türk, 突厥). Different scholars such as Ma Changshou (馬長壽) and Shiratori Kurakichi disagree on what the specific Turkic word is, such as Töliš, tegrek, tägräg and so forth. The most popular viewpoint about the Turkic word of Tiele is Türk. This viewpoint is shared by many scholars today including Toru Haneda, Onogawa Hidemi, and Geng Shimin. They also believed that Di, Dili, Dingling, Chile and Tujue (Türk, in the Sui and Tang dynasty) all came from the Turkic word Türk. “Türk” means “powerful” and “strength,” and its plural form is “Türküt.” The reason for the difference in the Chinese characters is that the names, which the tribes called themselves, and the names, that the tribes were called by other tribes, were different. They argue that all the Chinese names such as Di, Dili, Dingling, Chile, Tiele and Tujue are the different names used at different times by the Chinese and other nationalities to describe the Turkic peoples. Zhou Weizhou claimed this point as the key to resolving the many complicated names in Chinese literature for the same group of Turkic-speaking peoples living in the Mongolia plateau and its surrounding area.

[69] In the Chinese traditional chronicle sources, Xiongnu was generally cited as the ancestors of many clans. So we should view it as denoting the Xiongnu of the Han dynasty era. Xiongnu here just means “nomad” in the northern steppe.

[70] Saguchi Toru (佐口透) considered the Western Sea as the Lake Balkhash or the Aral Sea. Saguchi Toru, Beifang Minzu Shi Yu Menggu Shi Yiwen Ji (The collected work of the translations on the History of the Northern Nations and Mongolia, 北方民族史與蒙古史譯文集), Kunming: Yunnan Renmin Press, 2003, p. 87.

[71] Tola river, which was also known as Tula or Tughla river, has several different names in Chinese materials, but they all describe the same sound of Tola (Tula), such as Du luo 獨洛 in Sui Shu, Du le 獨樂 in the New Tang Shu, and Tu la禿剌 in Yuan Shi.

[72] Pugu 僕骨 was also recorded as Pugu 僕固 in Tang Dynasty.

[73] Toŋra was the name of Tongluo (同羅)in the Turkic inscriptions.

[74] Bai Mountain is the east ridge of the Tian Mountain.

[75] Qibi (契弊) maybe was the Jiepi (解批) in Gaoche. Jiepi can be found in the biography of Gaoche in Weishu (魏書).

[76] Saguchi Toru considered Yidie (乙咥) as Čigil, one of clans inside the Geluolu (葛邏祿)/Qarluqs. The History of the Northern Nations, Yunnan Renmin Press, p. 88.

[77] Wu Huan烏讙, 讙 was viewed as a mistake for 護 (the traditional Chinese of 護). In the biography of the Tiele in Beishi (北), 烏護was used in the same sentence.

[78] Saguchi Toru considered Yedie (也咥)as Yädiz or ädiz. The History of the Northern Nations, Yunnan Renmin Press, p. 88.

[79] Yunihuan (於尼讙) was recorded as Wunihu (烏尼護) in the biography of the Tiele in the Beishi (北史).

[80] Xueyantuo was a very powerful tribe within the Tiele and it had a close relationship with the Uyγurs. Xueyantuo participated in the attack activity against the Eastern Türks, which was recorded in the Biography of Uyγur in the New Tang Shu and Old Tang Shu.

[81] Ade River was the transliteration of Atil (Etil, Itil), which means big river. And Ade River was considered the Volga or Syr Darya. Zhang Xinglang said that Ade river (Attil, Idil), which was first mentioned in the description of Zemarchus, was the name of Volga. And the Arabic geographer Istakhri recorded that the Khazars lived in the riverside of the Atil River. Zhang Xinglang, Zhong Xi Jiaotong Shiliao Huibian (The Collection of the Sources on the Communication Between the West and China, 中西交通史料彙編) Vol.1, Beijing: Zhonghua Shuju, 1977, p. 67–79.

[82] Rui Chuanming considered Hedie as Akatir (Akatizir, Katzir) mentioned by Priscus, which was defeated by the Saraguri. Rui Chuanming: “Kangguo Bei Ji Ade Shui Diqu Tiele Buluo Kao” (The Research on the Tiele Clans in the North of Samarkand and the Area of Ade River, 康國北及阿得水地區鐵勒部落考), Tiedao Shiyuan Xuebao (鐵道師院學報), 1990.04, pp. 15–22.

[83] Hejie (曷嶻) was recorded as Hejie (曷截) in the biography of the Tiele in Beishi (北史). Zhang Xinglang thought that the sound of Hejie was like Hesa (曷薩), which denoted the Khazars. Cen Zhongmian considered Hejie as Kotzageri, which was a tribe allied to the Avars that escaped from the Türks. Rui Chuanming thought Hejie was another name of Kedie.

[84] Zhang Xinglang considered Bohu as an inaccurate transliteration of Bolghar. This point was approved by Cen Zhongmian.

[85] Bigan (比干) was recorded as Biqian (比千) in the biography of the Tiele in Beishi (北史). Zhang Xinglang thought Biqian maybe was Pecheneg, which was opposed by Cen Zhongmian and Rui Chuanming. Cen Zhongmian thought Bigan was the right name, and considered Bigan as Baγgun in a Syriac source (Pseudo-Zacharias Rhetor). Rui Chuanming pointed out that the Pechenegs, who immigrated to the west in the middle of 8^th century, cannot be Biqian because what the events recorded in the Sui Shu took place at the end of 6^th century or the beginning of 7^th century. Rui Chuanming considered Biqian as Bashkir (Bascart, Biscart, Bashghir, Bashghuit, Bashghur). He also thought Juhaihe (具海曷) was a name of a clan which was the descendent of the “Kuhe-Hege” (窟賀-賀葛) clan, which was composed by Sheruohege (設若賀葛氏) and Kuhe (窟賀氏).

[86] Cen Zhongmian considered Juhaihebixi (具海曷比悉) as Kipchak in Yuan Dynasty, and he thought it may be concerned with Yancai (奄蔡). Rui Chuanming considered Bixi (比悉) as Barsil (Berzilia, Barselt), which was among the Hun clans in the North of the Caucasus.

[87] Rui Chuanming considered Hecuo (何嵯) as Khazar. Cen Zhongmian and Rui Chuanming all considered Suba as Sabir. Rui Chuanming mentioned that Sabir was pronounced as Suvar in Chuvash.

[88] Keda (渴達) was recorded as Yeda (謁達) in the biography of the Tiele in the Beishi (北史).Zhang Xinglang considered Bayeweikeda (拔也未渴達) as a single word for the name of the Burtas. Because he thought that the sound of the two words was alike if we pronounced Bayemokeda quickly. Cen Zhongmian considered Yemo as Yamāk (Yemek) in the book of Kāsγari. He also considered Keda as Kourtargar in the Syriac source (Pseudo-Zacharias Rhetor). Rui Chuanming considered Yewei as Yan Guo (嚴國) in Xiyu Zhuan of Hou Han Shu and Xirong Zhuan of Weilue.

[89] Saguchi Toru and Shiratori Kurakichi considered Deyihai (得嶷海) as lake Balkhash. Friedrich Hirth and Cen Zhongmian considered Deyihai as the Caspian Sea. Zhang Xinglang considered Deyihai as Ural.

[90] Cen Zhongmian considered Sulujie as Sarourgour in the Syriac source (Pseudo-Zacharias Rhetor).

[91] Zhang Xinglang considered Sansuoyan as Saksin in the lower reaches of the Volga.

[92] Zhang Xinglang considered Miecu as the Mokshas. Cen Zhongmian considered Yanmiecu (咽蔑促) as a single word for Hermichions in Rouran.

[93] Cen Zhongmian considered Longhu as Ciras.

[94] Shiratori Kurakichi considered Enqu as Ugur, which Zemarchos met, or Ογοr which was recorded by Theophylactus Simocatta. During the Sui Dynasty, Enqu was located between the Caspian and the Black Sea.

[95] Zhang Xinlang considered the Alan as Yancai in Shi Ji and Hou Han Shu.

[96] Shiratori Kurakichi transliterated Fuwenhun to “biuk-uən-Xuən,” which was Bular, in the Tang Phonology. He also considered Beirujiuli as Baškir (Baškurt) in the same way. Zhang Xinglang agreed with this identification. Cen Zhongmian considered Jiuli as Koula (Kholas) in the Syriac source (Pseudo-Zacharias Rhetor). And he also considered Wenhun as the Ouarchonites recorded by Menander.

[97] Dobo is the ancient name of the Tuvans.

[98] Jinwang was Sui Yang Emperor (隋煬帝, r. 604–618) in the Sui dynasty.

[99] Chuluo (處羅) Qaghan was the third emperor of the Eastern Türk State in the period of 619–620 ce.