ᠬᠦᠢᠰᠦᠲᠣᠯᠣᠭᠠᠢ ᠶᠢᠨ ᠪᠢᠴᠢᠭᠡᠰᠦ ᠶᠢᠨ ᠲᠡᠦᠬᠡᠨ ᠣᠷᠴᠢᠨ The Historical context to the Huis Tolgoi inscription -Étienne de La Vaissière （ᠠᠩᠭ᠍ᠯᠢ ᠬᠡᠯᠡ ᠪᠠᠷ）

The Historical context to the Hü︔ is Tolgoi inscription É︔ tienne de la Vaissiè︔ re EHESS, Paris ︵Draft version︶ Although the reading of the Hü︔ is Tolgoi inscription by the joint and tremendous work of Dieter Maue, Mehmet Ö︔ lmez and Alexander Vovin is a gigantic step forward, still its content is not fully understood, and the other stone could not be read. As a historian, I will propose a likely hypothesis on the origin of the inscription, but obviously this hypothesis is conditional to further readings, confirmations and interpretations. I think caution is very much needed here. What we can rely upon are, first, external data, like the place of the discovery, the nature of the site, the choice of the script and of the language︔ and internal data, from the content of the text, that is mainly titles, proper names and some isolated words. The first important data is the place of discovery. It is in a lateral valley on the left bank of the Tuul river. It is my great pleasure to thanks Prof. Osawa for the precise GPS coordinates of the discovery spot, North 48, 08, 14,8 East 103, 09, 49,4, 1648m height. 1 Actually the site is slightly closer to the Orkhon than to the Tuul itself. But it is in the Tuul river system and drainage basin. In the context of pastoral economy the control of a valley was a major component of power. We are dealing with a power located in the broader Tuul valley, not in the Orkhon ︵although close to it︶ , the Selenga, the Tamir or farther West the Altay valleys. We have very few archaeological details on the site, a circular stone mound with a depression in the center on the northern part of a small hill. The total diameter of the mound is 18 m, the diameter of the central part is 14 m, stones were found on the bottom of the depression : eight fragments of stone slabs were found, may be from a funerary stone box. 2 The choice of Brahmi is striking. Brahmi was the script in use in the Tarim basin from the fifth to the tenth century, where it superseded the earlier Kharosthi. It has strong links with Buddhism. In the Mongol steppe, the only other inscription in Brahmi is the Brahmi part of the Bugut inscription in memory of the Turkish qaghan Tatpar, dated 582, at about 170 km as the crow flies to the west 1 See Takashi OSAWA, Kosetsu SUZUKI, R. Munkhtulga, 2019/12. Bichees II: Report of Researches on Historical Sites and Turkic Inscriptions in Mongolia from 2006 to 2008. Japan- Mongolian Joint Expedition Project “Bichees II“. Ulaan baatar: SOFEX Co., p. 93. 2 Zholdasbekov, M., Sartkozhauly, K. 2006. Atlas Orxonskix pamjatnikov. Astana: Kultegin, p. 35. from Hü︔ is Tolgoi. After that only Sogdian and Runic script were used. It is very much unlikely that a late revival of Brahmi script took place in the steppe and I think we can safely bet on a 5 th to 7 th century inscription, that is, in the Rouran and Early Turk period. As regard the language of the inscription, the major fact is that it is close to the mainstream Mongolic language. While it was generally assumed that during the 5 th to 7 th c. the Mongol speaking groups should be sought in Eastern Mongolia and Manchuria, here we have one ruler making use of a Mongolic language in Central-Western Mongolia. It can be : -a lingua franca, an unlikely hypothesis as Sogdian clearly was the lingua franca of the period︔ -an imperial language : Rouran or Xianbei are the most likely candidates︔ -the language of the political group itself︔ The question of the identity of languages between Bugut and Hü︔ is Tolgoi is open and cannot be solved currently. Not being a linguist I cannot say anything on the language itself. But, anticipating on the following, I should say that it is doubtful that it is the language of the tribe itself, and that from a historical point of view the most likely choice is Rouran, as the main non- Turkic imperial language of the steppe in this period, a choice on par with that of Brahmi, Buddhism being very important among the Rouran elites. There are some other possibilities and the issue is certainly not settled. All these data, except for the linguistic affiliation of the language, were known to Prof. Ō sawa when he proposed to identify the ruler at the origin of the inscription as the small qaghan Anluo. 3 Anluo is indeed known from the Chinese sources as a competitor to the Turkish throne immediately after the death of Tatpar, and to have settled on the Tuul river. His history matched two main sets of data, the location of the site, and the use of Brahmi as in Bugut. This was a perfectly sound reasoning, which only the knowledge of the content of the inscription could modify. If we now turn to the internal data, what do we read︖ I think the key point from a chronological point of view is the mention on lines 5 and 10 of the qaghan of the Turks Niri. This points without any doubts to Niri qaghan, who reigned from 595 to 604. He was a grandson of Muqan qaghan and from his base on the Northern Tianshan he claimed to be the heir of the whole Turkic empire. He sent an embassy to Byzantium in 595, corresponded with the Chinese princesses, fought against his enemies the Eastern Turks. He died, and his yabghu and qatun with him, in 604, killed by the Tiele. 4 His memorial is in the Tekes valley, with the still unpublished Mongolkü︔ re inscription in Sogdian narrating his life and giving the date of his death. With the mention of Niri qaghan we are one generation remote from the events of 581-2 and it seems difficult that Anluo was at the origin of this inscription. It should be contemporary of later than Niri. An other name in the text is famous : Anagui, the same name as the last Rouran qaghan. However no readable title is provided after his name in the inscription, and even if there was a title we have no special reasons to think that it was a qaghanal one. There are other Anagui mentioned in the 3 Osawa, T.,“Historical significance on the coexistence of languages, cultures and cult-believes under the early old turkic kaghanate from the Ö︔ tü︔ kä︔ n yï︔ š to the Tianshan regions,“ in Nasledie L.N. Gumileva i sovremennaja evrazijskaja integracija. Trudy IX Evrazijskogo nauchnogo foruma, posvjashchennogo 110-letiju so dnja rozhdenija L’va Nikolaevicha Gumileva, Astana: ENU, 2012. 4 la Vaissiè︔ re, E. de. 210. Maurice et le Qaghan : à︔ propos de la digression de Thé︔ ophylacte Simocatta sur les Turcs. Revue des É︔ tudes Byzantines 68: 219-224. sources in the Turkish period, especially Byzantine ones. 5 Anagui does not provide us with a chronological clue, while there was only one Niri qaghan of the Turks. However this cannot be an inscription of Niri qaghan: the steppe inscriptions were all established at the center of the political entities they celebrated, while there are 1800 km as the crow flies from the center of the Empire of Niri in the Tianshan and this inscription. It cannot have been created under the orders of Niri. It mentions Niri but is not from Niri. The obvious choice is that it is a Tiele inscription, narrating, among other facts, the stunning defeat of Niri and his heir. A detail in the text might actually point to this defeat : on line 5 are mentioned the qatun and the princes together with Niri qaghan: we know from the Chinese sources that Niri was defeated with his heir, and from the Mongolkü︔ re inscription that his wife died with him : «︔ the lord Nä︔ ri Qaghan, together with the queen, departed to [the gods]. »︔ says this text. This might be reflected in our inscription on this line, with a direct parallel in the tenth line of the Northern side of the Shine Usu inscription «︔ I seized Ö︔ zmish Qaghan. There I caught his wife ︵Qatun︶ . Thereafter, the Tü︔ rü︔ k people has ceased to exist. »︔ 6 We should here introduce the Tiele chapters of the Chinese sources, especially the Suishu chapter, contemporary to Niri. It has a long depiction of the Tiele tribes, extending up to the Azov sea, and beginning with: «︔ From the north to the Tola River they were: Pugu, Tongluo, Weihe, Bayegu and Fuluo︔ the chieftains are equally erkin, a little bit farther to the west there were the Mengchen, the Turu, the Hesijie, the Hun, the Huxie, and other tribes who altogether had 20 000 elite warriors. »︔ 7 In other words the region were the inscription has been discovered was the very center of the Tiele tribes, precisely when they came into a full fledged war against Niri qaghan, who tried to subdue them. We have archaeological proof of this as indeed a Pugu tomb of the second half of the 7th c., with its Chinese epitaph, has been recently excavated just north of the Tuul, near Zaamar, circa 80 km to the east of Hü︔ is tolgoi. There is a further proof of this in the text. The actual ruler involved in the creation of the inscription is most probably named at the beginning, bodi-satva tö︔ rö︔ -k kagan bṳ d-a kagan-u. There is no Pusa Kehan in the Chinese sources. However these sources mainly mentioned the titles given or recognized by the Chinese courts. There are plenty of examples of qaghan not mentioned as such in the official annals. For instance in the 640’s à︔ propos a Uighur chief, the Chinese sources wrote that “Tumidu, the Xielifa of the Uighurs, was awarded the title «︔ great general who likes improvement. »︔ But Tumidu had already entitled himself qaghan.“ 8 Taken that bias into account, there is one Nomadic chief entitled Bodhisattva right at the right place and the right period in our sources, the early Uighur leader Pusa, that is Bodhisattva. He is not entitled qaghan in the Chinese sources, but Xielifa. 9 The turkish-speaking Uighur were 5 There is a Anagai, chief of the tribe of the Utigurs in Menander’s depiction of a Byzantine embassy to the Turks in 575-6 ︵Fragment 19.1, in Blockley, R. C. 1985. The History of Menander the Guardsman, Liverpool: F. Cairns, p. 173.︶ 6 Text in Moriyasu, T., Ochir A. 1999. Provisional report of researches on historical sites and inscriptions in Mongolia from 1996 to 1998. Tokyo: Society of Central Eurasia Studies, p. 183. I am wondering if the Bargol of line 6 might not be a place name, the Barkul near Hami, and might be the place of Niri’s defeat, admittedly an unproven hypothesis. 7 Dobrovits, M. 2011. The Altaic world through Byzantine eyes: Some remarks on the historical circumstances of Zemarchus › journey to the Turks ︵AD 569–570︶ . Acta Orientalia Academiae Scientarum Hungaricae 64-4: 375. 8 Chavannes, É︔ . 1903. Documents sur les Toukiue ︵Turcs︶ occidentaux. Saint Pé︔ tersbourg: A. Maisonneuve, p. 91. 9 Jiu Tangshu 195.5195-6: emerging at that time as a leading tribe within the Tiele confederation, in agreement and then competing with the Xueyantuo. We are told in the Jiu Tangshu that Pusa was at the origin of the power of the Uighur within the Tiele confederation. He was still active at the end of the 620’s, but we do not know when his reign began. I think that the Hü︔ is tolgoi inscription is a memorial to his life. He is the Bodhisattva khagan of the beginning, who, dying, became Buddha︔ he might have played a great role in the defeat of Niri qaghan 20 years earlier. We are told in the Xin Tangshu that his political activity, that is wars, have been long, even before the death of his father. Crucially, we are told also by the Xin Tangshu that he established his camp in on the Tuul river valley 樹牙獨樂水上, in the broader valley of which the inscription was discovered. 10 The timespan between Niri’s defeat and the end of his reign, in the early 630’s, is compatible with him playing an important role in Niri’s demise. Unfortunately not much is understood in the later part and many details of the inscription : can hä︔ -r gä︔ -n on line 6 be the name of the famous mother of Pusa, Wuluohun 烏羅渾, pronounced ʔ ə raγwə n, who played an important political and military role in his realm︖ I do not think that the Tuwa several times mentioned in the inscription, with a qaghan on line 9, might be the Dubo, a Northern people of hunter gatherers with only a very limited political role. Earlier in the text the tö︔ rö︔ x qaghan would nicely fit the context of a Tiele qaghan, if all the Chinese transcription of Tiele would not converge to teg-reg and if A. Vovin had not proposed reborn, which nicely fit the context. However the «︔ tṳ wa par ︵or –nar︶ kagan to ̤ ro-x kagan-un »︔ combining tuwa and torox is puzzling, and may be ‘reborn’ is less convincing here. We should also think to the possibility of an intermediate language, for instance an inscription in Rouran giving the Rouran name of the various groups, while the Chinese would made use of Tuoba Wei transcription. I have no solution to propose on all of this. The political situation was extremely complex, with the Turkish qaghans reasserting themselves on part of the Tiele tribes after the death of Niri qaghan, to be defeated again after that, and the Xueyantuo playing an important role, entitling themselves qaghans. 11 This might or might be not be reflected in our inscription which logically should have put the emphasis on the victories of Pusa, not on his defeats. However, this inscription and the interpretation I propose fit perfectly with the recent improvements of our understanding of the history of the steppe. In a recent article, «︔ Away from the Ö︔ tü︔ ken »︔ I have demonstrated that the Turk power was absent from Mongolia during most of Translated in Chavannes, É︔ douard. 1903. Documents sur les Toukiue ︵Turcs︶ occidentaux. Saint Pé︔ tersbourg: A. Maisonneuve, p. 89-90. 10 Xin Tangshu 217b.6111-2: 回紇姓藥羅葛氏，居薛延陀北娑陵水上，距京師七千里。眾十萬，勝兵半之。地 磧鹵，畜多大足羊。有時健俟斤者，眾始推為君長。子曰菩薩，材勇有謀，嗜獵射，戰必身先，所 向輒摧 破，故下皆畏附，為時健所逐。時健死，部人賢菩薩，立之。母曰烏羅渾，性嚴明，能決平部事。回紇繇是 寖盛。與薛延陀共攻突厥北邊，頡利遣欲谷設領騎十萬討之，菩薩身將五千騎破之馬鬣山，追北至天山，大 俘其部人，聲震北方。繇是附薛延陀，相脣齒，號活頡利發，樹牙獨樂水上 。 11 Chavannes, É︔ douard. 1903. Documents sur les Toukiue ︵Turcs︶ occidentaux. Saint Pé︔ tersbourg: A. Maisonneuve, p. 91-6. the 7th c. 12 The center of power of the Eastern Turkic qaghans was in the Huanghe loop close to the Yinshan, south of the Gobi. It is only in the 690’s that Tonyukuk lead raids to reconquer the North. During most of the 7th c. Mongolia was mainly the power base of various Tiele tribes, among them the expanding power of the Uighurs. A Uighur palace of the 7th c. might have been recently identified in the Orkhon valley. 13 The Uighur inscriptions of the 8th c. have been overlooked when they told us that that «︔ my ancestors had reigned for eighty years. »︔ as the Tariat inscription has it. 14 The date ot the Turkish reconquest of the North is not precisely known, it should have been within the reign of Ilterish ︵682–691︶ , maybe in 685 or 6