Yinshan, Ordos, Orkhon, Tonyuquq; Etienne de la Vaissiere

                                      Yinshan, Ordos, Orkhon, Tonyuquq; Etienne de la Vaissiere

                                                    

                                                              Étienne de la Vaissière

When in 607 the Chinese emperor went to the north to see him in his residence, the meeting took place at Yulin, close to Tuldikh’s court. This is not a Southern court, as opposite to a wouldbe Northern court in Mongolia: it was never described as such3 . Moreover, the hapless Tuldikh asked the emperor for the permission to be assimilated to the Chinese people, to wear their dress, to have houses built. As made clear by this meeting, he was only a puppet qaghan, humiliated in front of his own officers. The text further states that the quite angry emperor demanded Tuldikh to pacify the North, a task which was still to be done, instead of trying to become Chinese (仍璽書答 啟民, 以為磧北未靜, 猶須征戰, 但使好心孝順, 何必改變衣服也; Suishu 84: 1874; Liu 1958, 62–63). 

The Chinese answer reveals the limitations of Tuldikh’s rule: he was only in control of what was for the Chinese a buffer zone, the Yinshan and the upper bent of the Huanghe. He was the Great Qaghan of the Türks only in the eyes of the few tribes he managed to gather, and in the eyes of the Chinese historiographers. He even did not dare negotiate himself with some Korean ambassadors at his court.

The situation changed with his son Shibi. He was much more powerful than his father and we do have a quite precise idea of what went on in the north during his time: the ancestral land of the Altai was dominated by the Western Türks while the tribes in the Ötüken and farther to the east submitted to Shibi (Xin Tangshu 217b: 6134; Chavannes 1903, 95). The Eastern Empire undoubtedly retook control of parts of the North. However, Shibi did not reside there, the political center was south of the Gobi. In the Ötüken and Mongolia were Tiele and Xueyantuo tribes, not the Türks themselves. 

The very beginning of the “Old Tang History” chapter on the Eastern Türks, which describes the scope of their empire at the end of the Sui in a synthetic way, clearly designated the Yinshan as their center: “they looked down with pride from the Yinshan and could think but with contempt of the Chinese empire” (高視陰山, 有輕中夏之志; Jiu Tangshu 194a: 5153, Liu 1958, 132 [transl. É.d.l.V.]), then in turmoil after the failure of the Korean wars.

This perfectly corroborates what we know of the activities of Shibi, always in the south, and of Xieli, his brother. An Eastern court is known under Shibi farther east along the Chinese frontier, but no Northern court4 . Shibi and then Xieli raided, year after year, northern China in summer, while in an empire centered north of the Gobi he would have been supposed to be in the north and not in the southern pastures, which were used in winter. The most he did was to send a shad to the north in order to quell the Tiele, with ultimately very little success, as the shad was beaten and expelled (Jiu Tangshu 109: 3289; Liu 1958, 263; see also Chavannes 1903, 89). How could a shad have been in charge of the Ötüken region if this was the center of the Turkic power? We do not have the slightest indication of any of these qaghans in the north. 

Their center, their economic basis were the Yinshan and raiding, and in this regard the famous text describing the growing revolt of Xieli’s own troops, put in the field every summer, and which would have preferred a more tranquil pastoral life, is interesting: “Xieli entrusted everything to the various Hu (Sogdians) and put his own people at a distance. 

The Hu are grasping and presumptuous and by nature uncertain and changeable. So the laws were multiplied and the army put into motion year after year. The people of the nation [i.e. the Türks] resented it and the tribes deserted” (Jiu Tangshu 194a: 5159 [transl. É.d.l.V.]). The economy of the upper social strata of this late ‘Eastern’ – Southern would be more exact – Empire was based on pillaging.

The power in the North was no longer hold by the Türks, but by the Tiele, and as subtribes the Xueyantuo (up to 646) and the Uyghur. This is precisely what the Uyghur inscriptions of the eighth century are saying. This fact has been overlooked because in everybody’s mind the natural power in the North was the Turkic one, even if it was an accepted truth that the Tiele tribes were unruly. However, Chinese sources are clear on the fact that various Tiele qaghans had a strong hold on power for almost half a century: first the Qibi Geleng (606 – before 620), then the Xueyantuo Yinan (628–645) and his son Bazhuo (645–646), and after the defeat of the Xueyantuo at the hand of their former allies, the Uyghur qaghan Tumidu 吐迷度 (646–648).

The Chinese sources do not mention any Tiele qaghan in the north after that, even if they emphasize that the Uyghur were now the dominant power. Turkic sources, however, do mention Uyghur qaghans. In the Uyghur inscriptions of Shine Usu, Tes, or Tariat, a leitmotiv is the fact that before the Türks were able to reestablish power in the 680s, over the span of 70 to 80 years the North was controlled by the ancestors of the Uyghur qaghans, that is precisely after Turan’s death and up to the second Turkic qaghanate created by Qutlugh in 682.

Take for instance the Uyghur Tariat inscription: “I heard that my ancestors had reigned for eighty years. The land of Ötüken and the land of Ögräsh, between these two lands, around the Orkhon River […] for seventy years they have reigned” (Katayama 1999, 171). This claim is sustained by the inscriptions of their foes of the second Turkic Empire: to the north of the nascent empire, in the early 680s, the enemy is Baz Qaghan, the qaghan of the Nine Oghuz confederation, headed by the Uyghur (Bilge Qaghan inscription eastern side, line 14)5 . Whatever the source, the Eastern qaghans are clearly confined south of the Gobi, not north of it, where the Uyghurs hold power more or less under Chinese control6 . Some Turkic tribes still lived as nomads north of the Gobi, but they had submitted to the Tiele power7 .

When placed in its proper geopolitical setting, we see that the Chinese policy at the end of the sixth century was to create a buffer zone and to declare this zone an empire as if an actual Eastern Turkic Empire had survived after Turan’s death. The historiographical trap is here: the Chinese made the better of a bad bargain because Tuldikh simply failed to meet the expectations. They claimed that their champion was the Great Qaghan while in fact he has been very much ejected from the political scene in the steppe, in which the Tiele and the Western Empire played the main roles. When chaos erupted in internal Chinese politics, this buffer zone indeed turned into a powerful Southern Empire under Shibi and Xieli. 

This, however, was a completely unexpected development which proved if anything that it was extremely dangerous to make use of this northern Ordos region to settle nomadic allies; in Antiquity the Ordos has been the very first region where the Xiongnu people grew powerful. The Eastern Empire under Shibi and Xieli, a very long-stretched empire following the Chinese frontier towards the east, lived by pillaging northern China. There was no apparent attempt to put the court back to the north of the Gobi. It was very different from the first Turkic Empire and cannot be seen as a legacy of this first empire, but as a consequence of the Chinese answer to the complex Ashinas policies.

This takes us to the end of the seventh century, to the revolt of Qutlugh (Qutluγ) in 682, who was soon to become Ilterish Qaghan (Fig. 2). There is no question that the revolt was rooted in the south. The base of the Türks were the Yinshan and the Black River region, that is the region around Hohhot: Czeglédy demonstrated in 1962 that the Čoγay-quzï/yïš of the Tonyuquq inscription, where the rebels rallied and created their empire, were none other than the Yinshan (Čoγay has the same meaning as Yin, shaded, northern slope of a mountain) and that the capital of Qaraqum of the same inscription is Heisha cheng of the Chinese text  both meaning black sands), a settlement situated on the northern slopes of the Yinshan (Czeglédy 1962, 55–56)8 . 

A simple list of all the topographical names mentioned in the Chinese sources shows beyond any reasonable doubt that, speaking of the geography, the second Eastern Empire was initially a revival of the empire of Xieli. From the Yinshan they raided over and over again all the northern prefectures of China during one generation (lists of raids in Liu 1958, 433–439; Skaff 2012, 302–312).

However, this is clearly described as a feature of the past in the Orkhon inscriptions. The recurrent message in these texts is the praise of the Ötüken, the residence of the Turkic qaghans in the final years of the Turkic Empire. Thus, one century after Tuldikh’s flight from north to south an unnoticed major, reversed geographical shift from south to north must have been taken place in the organization of the second Turkic Empire.

The reason clearly lies in a devastating strategic defeat of the Türks, which could not be explicitly recognized in such propaganda texts as the official inscriptions: in fact the Türks were forced to leave the Yinshan by a military move of the Chinese. In 708 the Chinese army cut the Yinshan Türks from the south by establishing three fortified points north of the Huanghe. We would like to know much more about this move but the text is not precise (Jiu Tangshu 194a: 5172; Liu 1958, 169). As the Yinshan and the Hohhot Valley have been the heart of the Turkic power since Tuldikh, this Chinese advance in the context of a revived Tang dynasty modified the balance of power. The consequence was immediate and we see a complete end to the annual raids of the Türks on the Chinese northern frontier. Never again would they come to pillage the prefectures of the northern Ordos region or northern Shanxi. The economic strategy of the Türk Empire had once again to change.

Such efficiency of these three fortified points is in a way quite mysterious. Qaraqum was on the other side of the Yinshan and no text mentions a direct conquest of the Yinshan proper. The Türks were not actually expelled from the Yinshan but left them. Maybe they were now too close to the Chinese border for it to be safe for the court. Or it might be that the Yinshan themselves were insufficient to sustain an actual imperial center if it was cut from the Huanghe and Heihe agricultural plains, or/and the raiding hinterland further south9 . Conversely, the complete end of the raids after the building of these three forts even in regions much farther east along the frontier, which might have been reached from the north, suggests that in fact the Yinshan were the point of departure of the previous eastern raids, not the north.

However, this strategic defeat, whatever its economic and political importance, was mitigated by the evolution of the empire under the influence of Tonyuquq. Contrary to the early decades of the seventh century, the qaghans, and especially their main counselor Tonyuquq, did show interest in the northern part of their empire before having to leave the southern part in 708. The chronology of this conquest of the north is not firmly known, it should have been within the reign of Ilterish (682–691), maybe in 685 or 686 before the great eastern campaign of 687 in Hebei – the conquest of the Ötüken is mentioned in Tonyuquq’s inscription just before some raids in Shandong up to the sea10.

The importance of the northern part of the empire as a center of power is likewise unknown: Tonyuquq, who wrote after the events to defend his ideas, should be read cum grano salis when he noted that he “led the [qaghan?] and the Türk people to the Ötüken land”). What becomes clear, however, is that most of the activities of Ilterish Qaghan were concentrated to the south and east of the empire, not in the north: out of 19 raids attributed to Ilterish by Tonyuquq only five were against the Oghuz, and the first of these might have taken place south of the Gobi. The north was first of all Tonyuquq’s domain and it is from there that he directed raids toward the Qirqiz and the On Ok: as Tonyuquq wrote himself “I settled in the Ötüken land”, not the qaghan.

Qapaghan Qaghan (691–716) had a broader vision and much more means: in 698 we see for the first time the words 黑紗南庭 “Southern court at Qaraqum” in the Chinese sources (Jiu Tangshu 194a: 5169; Liu 1958, 162). Clearly in 698 a situation that had lasted over almost one century ended and the empire of Qapaghan was now established – as was the empire of the first Türks at its climax – on both sides of the Gobi. However, this was to survive only for a very short period as the Chinese move of 708 put an end to it. Having to withdraw to the north, and having no longer the usual resources, Qapaghan frantically tried to submit all unruly tribes in every direction except for the south. He used the Ötüken as a base, only to be killed a few years later in 716. The great 711–712 expedition to the west might have been triggered by the necessity to find an alternate place for pillages as China was now out of reach.

So it was up to Bilge Qaghan and Kül Tegin to make the best of a bad situation. In this regard, the message of the Orkhon inscription is pure political propaganda, and a close look at what they actually say does confirm this interpretation. When Bilge Qaghan inherited the throne, or rather took it from his cousin, the Türks were weak, poor, and desperate. According to the Or khon texts, they had had to migrate to the west and the east. These texts are long political appraisals of the Ötüken and Orkhon regions as opposed to the Čoγay mountain and the Tögültün Valley, because the relocation from the south was not voluntary and peaceful: “if you go to the south you will die” the text says, as opposed to “if you stay at the Ötüken then the caravans will come”. This is an attempt to justify the reversal of one century of Turkic history during which the Čoγay and Tögültün were the actual home of the Türks, an attempt to conceal that the change was ultimately effected by a Chinese move and a Turkic defeat. The political message of the Orkhon inscriptions is much clearer once put in this century-long perspective.

Similarly the call to Bumın and Ishtemi in these texts belongs to the same plea, to retrace a mystified Ötüken history of the Türks. In spite of the fascinating but both deceptive and highly political Orkhon inscriptions, the first nearly fifty years of the eighth century (685–743) might be regarded as a quite limited period of Turkic power in the north within a quarter of a millennium (603–840) of actual Tiele and Uyghur domination.

3 In fact Czeglédy, without dealing with the actual geopolitical situation, had already realized in an article published in 1962 that all the data from the Chinese sources point only to a court in the south in the Yinshan for Tuldikh and his heirs (see Czeglédy 1962, 66–67).

4 Czeglédy 1962, 67 is wrong in this regard. There is no Northern court mentioned in the sources, only an Eastern one.

5 On the Nine Oghuz see Golden 1992, 156. Baz Qaghan means submitted qaghan: at that time it can only mean submitted to the Chinese, in perfect agreement with what the Chinese sources and the archaeology (see next note) show. Later, the Türks would certainly not have recognized a qaghan of the Nine Oghuz but would do their best to submit and integrate them.

6 See the recent excavations of Chinese-style tombs in Mongolia dating from the period of Chinese control of the Tiele tribes, e.g., the tomb of a Pugu chief from the Tiele confederation (see Danilov et al. 2010, non vidi). For another tomb see Sartkozhauly 2011. My sincere thanks to Jan Bemmann for these references.

7 The submission of some former Turkic tribes to Xueyantuo qaghans is well known, see the attempt of the Ashina Hubo to settle in the north: he had to submit to the Xueyantuo before trying briefly (648–650) to revive an Eastern Qaghanate in the north after Tumidu’s death, only to be crushed by a Uyghur-Tang alliance.

8 See also Suzuki 2011 for an attempt to clarify the location of Qaraqum as well as a depiction of the economic basis of the Türks in the Yinshan. My sincere thanks to Michael Drompp for drawing my attention to this article and sending it to me.

9 A side effect of this line of thought might be an attempt to identify the still mysterious Tögültün yazï of the Orkhon inscriptions. It is clear that it is associated with the Yinshan in a binomial expression “the Čoγay Mountain and the Tögültün plain”. Various scholars have tried to localize this toponym (see Li 2011, who proposes a broad meaning of the whole upper region of the Huanghe including the Ordos plateau). I am wondering if the two alluvial valleys of the Huanghe and the Heihe are not to be understood under that name: there is still at their confluence a county with the quite close name of Tuoketuo 托克托 – earlier Tuoketun 托克屯? –, and the easternmost fortified point was built around this area in 708. J. Jeong suggested to identify Tögültün yazï with the Tümed steppe (see Li 2011, 375 fn. 6), a solution rather similar to mine as the Tümed steppe is just north of Tuoketuo. The Chinese conquest would have cut the Čoγay from the Tögültün while both of them would have been necessary to sustain an imperial center south of the Gobi.

10 Differently Skaff 2012, 309 who would rather place the conquest of Mongolia between 688 and 693 due to the lack of raids on the Chinese frontier during these years.

https://www.academia.edu/23883202/AWAY_FROM_THE_%C3%96T%C3%9CKEN_A_GEOPOLITICAL_APPROACH_TO_THE_SEVENTH_CENTURY_EASTERN_T%C3%9CRKS