Hostname: page-component-848d4c4894-sjtt6 Total loading time: 0 Render date: 2024-06-17T13:39:22.704Z Has data issue: false hasContentIssue false

Chinggis Khan Defeated: Plano Carpini, Jūzjānī and the Symbolic Origins of the Mongol Empire

Published online by Cambridge University Press:  16 October 2020

EHESS, Paris.


This article aims to clarify an obscure passage in Plano Carpini's text, and subsequently in C. de Bridia's one, referring to a crushing defeat of Chinggis Khan, which has so far not been identified with certainty. The record of such a defeat is found in identical terms under the pen of Jūzjānī, and it actually appears that this strange narrative follows the pattern of the Mongol myth of origin, which is also common to the myths of the Türks, of the Kimeks and others. Here the argument is made that these accounts written outside the Mongol territory are therefore not only the result of confusion and distortion, contrary to what has long been thought. They testify to the existence of a legend of Chinggis Khan, built in an imperial propaganda effort directed at all the nomadic subjects of the Mongol Empire, and which placed the birth of the empire and the story of the origins contained in the myth on the same symbolic level.

Research Article
Copyright © The Royal Asiatic Society 2020

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)


1 Carpini, Plano, Ystoria Mongalorum quos nos Tartaros apellamus; Storia dei Mongoli, (ed.) Menestò, Enrico et al. (Spoleto, 1989), pp. 259, 274Google Scholar; translation Dawson, Ch., “History of the Mongols”, in The Mongol Mission: Narratives and Letters of the Franciscan Missionaries in Mongolia and China in the Thirteenth and Fourteenth Centuries, (ed.) Dawson, Ch. (London, New York, 1955), pp. 23, 31Google Scholar; translation Tanase, Th., “Histoire des Mongols”, in Dans l'Empire mongol, (ed.) Tanase, Th. (Toulouse, 2014), pp. 93, 105Google Scholar.

2 Thus, for example, the defeat of the Mongols against the Christians of India ruled by the Priest John and his copper fire-breather mannequins, the meeting of Chinggis Khan's army with dogmen, the passage through the magnetized mountains of the Caspian where the peoples of Gog and Magog are imprisoned, and then the arrival in a country where men live underground, because of the infernal noise that the sun makes when it rises; Plano Carpini, Storia dei Mongoli, pp. 258–263; translation Dawson, “History of the Mongols”, pp. 22–25; translation Tanase, “Histoire des Mongols”, pp. 92–96. A little further on, Plano Carpini recounts how the Mongols, returning from their expedition in Hungary, met the Parossits, who only feed with smoke, men with ox feet and dog faces, and others, finally, whom he calls Cyclopeds, having only one arm and one leg, and who move around doing the wheel; Plano Carpini, Storia dei Mongoli, pp. 272–274; translation Dawson, “History of the Mongols”, pp. 30–31; translation Tanase, “Histoire des Mongols”, pp. 104–105.

3 Boyle, J. A., “The Alexander Legend in Central Asia”, Folklore LXXXV, n°4 (1974), p. 221Google Scholar; Guéret-Laferté, M., Sur les routes de l'Empire mongol. Ordre et rhétorique des relations de voyage aux XIIIe et XIVe siècles (Paris, 1994), pp. 299300Google Scholar; Aigle, D., The Mongol Empire between Myth and Reality. Studies in Anthropological History (Leiden, 2014) pp. 5152Google Scholar. See also the introduction to Tanase, Th. (ed.), Dans l'empire mongol (Toulouse, 2014), pp. 2934Google Scholar.

4 Yourtchenko, A., “Ein asiatisches Bilderrätsel für die westliche Geschichtsschreibung. Ein unbekanntes Werk aus dem 13. Jahrhundert (Der ‘Tschingis Khan-Roman’)”, Zentralasiatische Studien XXVIII (1998), pp. 4849Google Scholar in particular.

5 On the authenticity of C. de Bridia's text, see Sinor, D., “Mongol and Turkic Words in the Latin Versions of John of Plano Carpini's Journey to the Mongols (1245–1247)”, in Mongolian Studies, (ed.) Ligeti, L. (Budapest, 1970)Google Scholar, and more recently Guzman, G., “The Vinland Map Controversy and the Discovery of a Second Version of the Tartar Relation: The Authenticity of the 1339 Text”, Terrae Icognitae XXVIII (2008)Google Scholar.

6 The men living underground, in the country where the sun rises in a great crash, are called the Narayrgens, which he translates as “men of the sun”, from nara(n), “sun”, and irgen, “people”. The land of the dogmen is called Nochoy Kadzar, from Noqai, “dog”, and qajar, “land, country”. Men with ox feet and dog heads are called Ucorcolon, “Beef feet”, from üker, “ox feet”, and köl, “foot”, or Nochoyterim, “Dog heads”, terigün meaning “head”; Bridia, Hystoria Tartarorum, (ed.) Alf Önnerfors (Berlin, 1967), pp. 10, 13, 16; translation Painter, G. D., “The Tartar Relation”, in The Vinland Map and the Tartar Relation, (ed.) Skelton, R. A., Marston, T. E. and Painter, G. D. (New Haven, London, 1965), pp. 64, 70, 74Google Scholar; translation Tanase, Th., “Histoires des Tartares”, in Dans l'Empire mongol, (ed.) Tanase, Th. (Toulouse, 2014), pp. 171, 175176Google Scholar. In general, all the etymologies proposed by C. de Bridia for the names of peoples or places, including those of real peoples, are correct, or at least plausible and explainable by Turkish or Mongolian. See Sinor, “Mongol and Turkic Words”.

7 Guéret-Laferté, Sur les routes, p. 299 n. 39.

8 As Painter put it in his introduction to the editing and translation of Bridia's text (Painter, “The Tartar Relation”, p. 49), taken up by de Rachewiltz, I., Papal Envoys to the Great Khans (London, 1971), p. 107Google Scholar, and Yourtchenko, “Ein asiatisches Bilderrätsel”.

9 The stories in this chapter would indeed have been reported in the Mongol court, as Plano Carpini repeats twice: Plano Carpini, Storia dei Mongoli, pp. 259, 274; translation Dawson, “History of the Mongols”, pp. 23, 31; translation Tanase, “Histoire des Mongols”, pp. 93, 105. See also Yourtchenko “Ein asiatisches Bilderrätsel”, p. 49.

10 Painter, “The Tartar Relation”, p. 49; Rachewiltz, Papal Envoys, p. 107. According to Boyle, “The Alexander Legend”, the Alexander Romance was known to the Mongols, and more generally to Eurasian nomads, thanks to the dissemination in Central Asia by Nestorian missionaries of its version in Syriac, as well as that of a text entitled the Christian Legend concerning Alexander, also known as the Syriac Legend concerning Alexander. However, I have serious reservations about this theory; see below. On the Christian Legend concerning Alexander in particular, and the spread of the Alexander Romance in the East, see Czeglédy, K., “The Syriac Legend Concerning Alexander the Great”, Acta Orientalia Academiae Scientiarum Hungaricae VII, no2/3 (1957)Google Scholar, and more recently, E. Van Donzel and Schmidt, A., Gog and Magog in Early Syriac and Islamic Sources. Sallam's Quest for Alexander's Wall (Leiden, 2009)Google Scholar.

11 Plano Carpini, Storia dei Mongoli, p. 255; translation Dawson, “History of the Mongols”, p. 20; translation Tanase, “Histoire des Mongols” pp. 89–90.

12 Bridia, , Hystoria Tartarorum, (ed.) Önnerfors, Alf (Berlin, 1967), p. 6Google Scholar; translation Painter, “The Tartar Relation”, p. 58; translation Tanase, “Histoires des Tartares”, p. 168.

13 On the identification between Kitai/Kitat and the inhabitants of northern China, see for example de Rachewiltz, I., The Secret History of the Mongols: A Mongolian Epic Chronicle of the Thirteenth Century (Leiden, 2006), p. 300Google Scholar.

14 Tanase, “Histoire des Mongols”, p. 89 n. 35 (“cette campagne ne se laisse pas identifier”).

15 Secret History [hereafter SH], § 129; translation Rachewiltz, The Secret History, p. 54.

16 Rashīd ad-Dīn, Jāmi‘ at-tavārīkh: Tārīkh-e Ghāzānī, (ed.) M. Rowshan and M. Mūsavī (Tehran, 2016), p. 299; translation W. M. Thackston, Classical Writings of the Medieval Islamic World. Persian Histories of the Mongol Dynasties, Vol. III: Rashiduddin Fazlullah (London, 2012), p. 115; Shengwu Qinzheng Lu: “Shengwu Qinzheng Lu jiaozhu 聖武親征錄校注”, in Wang Guowei yishu 王國維遺書, Vol. XIII, (ed.) Wang Guowei (Shanghai, 1983) [hereafter SWQZL], p. 10a; translation Pelliot, P. and Hambis, L., Histoire des campagnes de Chinggis Khan (Leiden, 1951), pp. 3537Google Scholar.

17 Ratchnevsky, P., Genghis Khan. His Life and Legacy (Malden, 2006), pp. 4647Google Scholar.

18 Ibid., p. 49. Interestingly, it is just after the discord between Temüjin and Jamuqa broke out, and before Dalan Baljut, that Rashīd ad-Dīn places an episode he had already mentioned earlier – and that he certainly places in Chinggis Khan's youth, but not necessarily in his adolescence as the author of the SH does: the capture by the Tayichi'ud, who were at the centre of a coalition with Jamuqa and other groups to defeat Chinggis Khan. He would have been defeated, abandoned by his companions, and finally captured. The account of Rashīd ad-Dīn, even if it makes Dalan Baljut a victory, is therefore perhaps not contradictory with the idea of a “gap” in the life of the Mongol ruler; Rashīd ad-Dīn, Jāmi‘ at-tavārīkh, pp. 163–164, 296; translation Thackston, Rashiduddin, pp. 66, 114.

19 Ratchnevsky, Genghis Khan, p. 49.

20 Ibid., p. 50; Zhao Gong 趙珙, Mengda beilu: “Mengda beilu 蒙韃備錄”, in Wang Guowei yishu 王國維遺書, Vol. XIII, (ed.) Wang Guowei p. 3a; translation Olbricht, Pinks et al. “Meng-Ta Pei-Lu. Ausführlische Aufeichnungen über die Mongolischen Tatan von Chao Hung 1221”, in Meng-ta Pei-lu und Hei-ta shih-lüeh, (ed.) P. Olbricht, E. Pinks et al. (Wiesbaden, 1980), p. 12.

21 Rashīd ad-Dīn is in fact almost the only one to ascribe Qalaqaljid Elet (Qalāljīt Elet قلالجین الت in his text) as a defeat for Chinggis Khan: Rashīd ad-Dīn, Jāmi‘ at-tavārīkh, p. 347; translation Thackston, Rashiduddin, p. 132. The Yuanshi 元史 presents it as a victory in its history of Chinggis Khan but as a defeat, although without naming it, in Ja‘far's biography Khwāja: Yuanshi 1: 10, 120: 2960 [all the Chinese dynastic histories are quoted according to the standard Zhonghua shuju edition]. The SWQZL and the SH (which does not, however, mention the Baljuna Covenant) present it as a victory: SWQZL, p. 37a; SH, § 171; translation Rachewiltz, The Secret History, p. 92. Juvaynī telescopes the two events into a battle that would have taken place near a stream called Baljuna, from which Chinggis Khan emerged victorious, although his forces were smaller in number: Juvaynī, Tarīkh-i jahān-gushā: The Ta'rīkh-i-Jahān-gushā of ‘Alā’u'd-Dīn ‘Atā Malik-i-Juwaynī, (ed.) Mīrzā Muḥammad Qazvīnī (Leiden, 1912–1937), I, p. 27; translation Boyle, J. A., History of the World Conqueror (Cambridge [Mass.], 1958), p. 37Google Scholar. Now Pelliot describes the battle as “une victoire à la Pyrrhus, qui fut peut-être une défaite” (Pelliot and Hambis, Histoire des campagnes, p. 46), and de Rachewiltz concludes from Chinggis Khan's withdrawal and the diminished position in which he subsequently found himself that the confrontation “was in fact a reverse for the Mongols” (Rachewiltz, The Secret History, pp. 623–624). See also Cleaves, F. W., “The Historicity of the Baljuna Convenant”, Harvard Journal of Asiatic Studies XVIII, no3–4 (1955)Google Scholar, and below.

22 Painter, “The Tartar Relation”, p. 59 n. 1.

23 Rashīd ad-Dīn, Jāmi‘ at-tavārīkh, pp. 84 ff.; translation Thackston, Rashiduddin, p. 36.

24 Rachewiltz, Papal Envoys, p. 107.

25 Yourtchenko “Ein asiatisches Bilderrätsel”, pp. 49–50 n. 12 (“Unklar bleibt, wie es dazu kam, daß Veteranen der Feldzüge Tschingis Khans, die abends am Lagerfeuer saßen, Legenden über die Feldzüge ihrer mongolischen Herren verfaßten, wenn das mongolische Heer in jenen Legenden vollständige Niederlagen”), pp. 80–85.

26 Guéret-Laferté, Sur les routes, p. 298 (“ajouts purs et simples d’évènements imaginaires”).

27 Plano Carpini, Storia dei Mongoli, p. 431.

28 An episode reminiscent of Chinggis Khan's youth, reported in the SH, during which he hid for nine days in a thicket to escape his Tayichi'ud enemies; SH, § 79–80; translation Rachewiltz, The Secret History, pp. 22–23.

29 Hayton, “La Flor des Estoires de la Terre d'Orient”, in Recueils des Historiens des Croisades. Documents latins et français relatifs à l'Arménie, Vol. II, (ed.) É. Dulaurier, pp. 147–154.

30 I will address this point in particular in a future article.

31 Jūzjānī, , Ṭabaqāt-i Nāṣirī, (ed.) Habībī, ‘Abd al-Hayy (Kabul, 1964), II, p. 99Google Scholar; translation Raverty, H. G., Ṭabaqāt-i Nāṣirī: A general history of the Muhammadan dynasties of Asia, including Hindustan; from A. H. 194 (810 A.D.) to A.H. 658 (1260 A.D.) and the irruption of the infidel Mughals into Islam (London 1881), II, p. 937Google Scholar, slightly modified.

32 Jūzjānī, Ṭabaqāt-i Nāṣirī, II, pp. 99–100; tr. Raverty, Ṭabaqāt-i Nāṣirī, II, pp. 937–959.

33 SH, § 53; translation Rachewiltz, The Secret History, p. 10, to cite just one example. Besides, the dynastic name Jin 金 means “gold” in Chinese.

34 Golden, P. B. Introduction to the History of the Turkic Peoples (Wiesbaden, 1992), p. 73Google Scholar.

35 Jūzjānī, Ṭabaqāt-i Nāṣirī, II, p. 98; translation Raverty, Ṭabaqāt-i Nāṣirī, II, p. 936.

36 Raverty, Ṭabaqāt-i Nāṣirī, II p. 937n. 9.

37 Rashīd ad-Dīn, Jāmi‘ at-tavārīkh, pp. 137–139; translation Thackston, Rashiduddin, pp. 56–57, slightly modified.

38 Rashīd ad-Dīn, Jāmi‘ at-tavārīkh, p. 202; translation Thackston, Rashiduddin, p. 81, slightly modified.

39 Abū'l-Ghāzī was able to consult abundantly the Persian sources, and Rashīd ad-Dīn's Jāmi' at-tavārīkh in the first place, during his exile to the Safavids. But he also collected many oral traditions from Kazakhs and Kalmüks, his neighbours. His work is therefore not only a pale copy of the text of Rashīd ad-Dīn, but presents many variants and original contributions; see Spuler, B., “Abū’l-Ghāzī Bahādur Khān”, in The Encyclopaedia of Islam, 2nd edition. (Leiden, 1954–1960), I., pp. 120121Google Scholar.

40 Abū’l-Ghāzī, , Shajara-ye Turk: Histoire des Mongols et des Tatares, (ed.) Desmaisons, P. I. (St Petersburg, 1874), pp. 3334, 59Google Scholar;. translation Desmaisons, pp. 33, 63, slightly modified.

41 SH, § 1; translation. Rachewiltz, The Secret History, p. 1.

42 See also the version of Mustowfī Qazvīnī in his Tārīkh-i Guzīda (completed in 1330), which closely follows Rashīd ad-Dīn, but adds a variant according to which Qiyan and Nüküz were two women, who mated in the Ergene Qūn with a wolf, which is then not without reminding one of the origin myths of the Kyrgyz: Qazvīnī, Mustowfī, Tārīkh-i guzīda, (ed.) Navā’ī, ‘Abd al-Ḥusayn (Tehran, 1983), pp. 562563Google Scholar; Roux, J.-P., La religion des Turcs et des Mongols (Paris, 1984), pp. 193194)Google Scholar. There is also a version very similar to that of Mustowfī Qazvīnī in the Muqqadima written by Sharaf ad-Dīn Alī Yazdī around 1419–1420: Yazdī, “Muqaddima”, in Ẓafarnāma, (ed.) Seyyed Sa‘īd Mīr Moḥammad Ṣādeq and ‘Abd al-Ḥosayn Navā’ī (Tehran 2008), pp. 55–57. On the revival of the myth of Ergene Qūn among Persian authors after Rashīd ad-Dīn, see Dobrovits, M., “The Turco-Mongolian Tradition of Common Origin and the Historiography in Fifteenth Century Central Asia”, Acta Orientalia Academiae Scientiarum Hungaricae XLVII, n°3 (1994), pp. 272Google Scholar ff. There is also a version in Shajarat al-Atrāk, a text in Persian probably compiled at the court of the Timurid sovereign Ulugh Beg (1411–1449): Shajarat al-Atrāk; tr. Miles, W., The Shajrat ul-Atrak, Or the Genealogical Tree of the Turks and Tatars (London, 1838), pp. 3843Google Scholar.

43 Boyle, J. A., “Some Thoughts on the Sources for the Il-Khanid Period of Persian History”, Iran XII (1974), p. 186Google Scholar; Jackson, P., The Mongols and the West, 1221–1410 (Harlow, London, New York, 2005), p. 149Google Scholar; Aigle, The Mongol Empire, p. 35.

44 See on this subject the excellent and very complete analysis of Devin DeWeese, many of whose elements are included here: DeWeese, D., Islamization and Native Religion in the Golden Horde. Baba Türkles and Conversion to Islam in Historical and Epic Tradition (University Park, 1994), pp. 3950Google Scholar, 273–287, 494–502 in particular.

45 Gardīzī, , Zayn al-Akhbār: Tārīkh-i Gardīzī, (ed.) Ḥabībī, ‘Abd al-Ḥayy (Tehran, 1984), pp. 549551Google Scholar; translation Martinez, A. P., “Gardīzī’s Two Chapters on the Turks”, Archivum Eurasiae Medii Ævi II (1982), pp. 120121Google Scholar, slightly modified. It can be noted that the survivors’ flight motif is doubled here, between on the one hand Shad's voluntary flight to escape his death, and on the other hand the wandering of the seven Tatars, which allows them to escape the destruction of their people.

46 ad-Dawādārī, Ibn, Kanz ad-durar wa jāmi‘ al-ghurar, Vol. VII: ad-Durr al-maṭlūb fī akhbār mulūk banī Ayyūb: Die Chronik des Ibn ad-Dawādārī. Siebter Teil, Der Bericht über die Ayyubiden, (ed.) ‘Āshūr, Sa‘īd ‘Abd al-Fattāḥ (Cairo, Wiesbaden 1972), p. 230Google Scholar. I would like to thank Thomas Bédrède and Paul Neuenkirchen for translating for me these pages from Ibn ad-Dawādārī.

47 Ibn ad-Dawādārī, Kanz ad-durar, pp. 228–231. See also Haarman, U., “Alṭun Ḫān und Čingiz Ḫān bei den ägyptischen Mamluken”, Der Islam LI (1974), pp. 2226Google Scholar, and DeWeese, Islamization, pp. 278–282.

48 Beishi 99: 3285; translation Sinor, D., “The Legendary Origins of the Türks”, in Folklorica: Festschrift for Felix J. Oinas, (ed.) Zygas, E. V. and Voorheis, P. (Bloomington, 1982), p. 224225Google Scholar, amended. See also Zhoushu 50: 907; translation Mau-Tsai, Liu, Die Chinesischen Nachrichten zur Geschichte der Ost-Türken (T'u-Kue) (Wiesbaden, 1958), p. 5Google Scholar and Suishu 84: 1863.

49 The stage of the appointment of the chief is doubled here, between the election of Ashina on the one hand, and the exit of the cave under the leadership of Axian shad on the other hand.

50 Pelliot, P., “Neuf notes sur des questions d'Asie centrale”, T'oung Pao XXVI (1929), p. 214n. 2Google Scholar.

51 Sinor, “The Legendary Origins”.

52 DeWeese, Islamization, pp. 275–278. Boyle's theory that the motif of the cave or enclosed place of the Turk and Mongol myths ultimately goes back to the legend of the Wall of Alexander, built to enclose the peoples of Gog and Magog, which was transmitted to Central Asia through the Syriac version of the Alexander Romance, must be rejected; Boyle, “The Alexander Legend”, repeated in Jackson, The Mongols, p. 150, and Tanase (ed.), Dans l'empire mongol, p. 32. That Nestorian missionaries brought with them to the Central Asian nomads, in addition to the Bible, the Christian Legend concerning Alexander is an ad hoc hypothesis without any tangible basis, and is contradicted by chronology. Indeed, the Christian Legend concerning Alexander, which is the oldest attestation of the theme of the imprisonment of Gog and Magog by Alexander, was elaborated around 629–630 as a propaganda document in favour of Heraclius, the recent winner of the Persians: Czeglédy, “The Syriac Legend”; Van Donzel and Schmidt, Gog and Magog, pp. 18–22. However, the Beishi and the Zhoushu were compiled at the beginning of the Tang dynasty, in the years 630–650, and are based, for their parts concerning the Türks, on informations collected under the Northern Qi (550–577) and the Northern Zhou (557–581) dynasties. Therefore, it is impossible that the origin myth of the Türks, and furthermore that of the Mongols, was developed under the influence of the legend of Alexander (as already noted by DeWeese, Islamization, pp. 273–274n. 79), especially since at least a certain number of elements of this myth go back well beyond the Türks. Thus the motif of the ancestral cave alone dates back at least to the Tuoba-Wei; see Ford, R. A., “The Gaxian Cave 嘎仙洞 Inscriptions: The Perpetuation of Steppe Traditions Under the Northern Wei Dynasty”, Archivum Eurasiae Medii Ævi XX (2013), pp. 2466Google Scholar. And the theme of the nourishing she-wolf is already present in the myth of the Wusuns reported by the Shiji 史記 and the Hanshu 漢書; see de La Vaissière, É., “Iranian in Wusun? A tentative reinterpretation of the Kultobe inscriptions”, in Commentationes Iranicae. Vladimiro f. Aaron Livschits nonagenario donum natalicium, (ed.) Tokhtasev, S. and Lur'e, P. (St. Petersburg, 2013), pp. 320325Google Scholar. In addition, the two traditions differ in their meanings. The cave or enclosed place, in the myths of Eurasian nomads, symbolizes, sometimes explicitly, the uterus where the future community is gestating, and from which it must emerge to continue to grow, in what appears to be a true birth. It is therefore an ancestral place in two ways: not only as a place of ancestors, but also as a maternal ancestor itself, or motherland; see DeWeese, Islamization, especially pp. 43–44, 273 ff. The Wall of Alexander, on the other hand, is a wall protecting sedentary people from nomads, whose legend is undoubtedly based on a reality common in the Iranian world, in particular (it should be noted that, in all likelihood, the Syriac version of the Alexander Romance was translated from a Pehlevi version; Czeglédy, “The Syriac Legend”, p. 241), as for example the wall that surrounded the Bukhara oasis to this end. It is undeniable that the legends composing the Alexander Romance have circulated in Central Asia, as Plano Carpini's text attests. It is not said, however, that all of them have circulated in one direction. One may even wonder if it was not the myth of nomadic origin that influenced the Wall of Alexander legend. Nevertheless, there is no indication that the Nestorians were solely responsible for these circulations. On the contrary, the Mongolian version of the Alexander Romance, dating from the early 14th century and featuring a certain Sulqarnai (i. e. the Coranic name of Alexander, Dhū’l-Qarnayn), seems to testify to the late transmission of the complete account to the Mongols, and to the fact that it took place throughout the Muslim world; see Cleaves, F. W., “An Early Mongolian Version of the Alexander Romance”, Harvard Journal of Asiatic Studies XXII (1959), pp. 199CrossRefGoogle Scholar.

53 Tekin, T., A Grammar of Orkhon Turkic (Bloomington, 1968), p. 233Google Scholar; translation p. 265.

54 The title of shad referred to the rank immediately below that of qaghan in the Türk hierarchy, and was granted to members of the sovereign's immediate family to perform essentially military functions. See Clauson, G., An Etymological Dictionary of Pre-Thirteenth-Century Turkish (Oxford, 1972), p. 866Google Scholar, and Golden, P. B., “The Türk Imperial Tradition in the Pre-Chinggisid Era”, in Imperial Statecraft: Political Forms and Techniques of Governance in Inner Asia, Sixth-Twentieth Centuries, (ed.) Sneath, D. (Bellingham, 2006), pp. 5152Google Scholar. This is the title borne by Axian shad, and it is also obviously found in the name of the hero of the kimek myth.

55 Tekin, A Grammar, p. 235; translation pp. 267–268.

56 Zhoushu 50: 908; translation Liu, Die Chinesischen Nachrichten, pp. 5–6, slighly modified.

57 Sinor, “The Legendary Origins”, pp. 223–224, 230–232, 235–236.

58 Devin DeWeese too seems to consider them as such: DeWeese, Islamization, pp. 496–498. It should be noted that DeWeese, like Sinor, also analyses a third legend which is not however a myth of origin of a political community, and which therefore does not interest our purpose. See also Kljaštornyj, S. G., “Problemy rannej istorii plemeni türk (Ašina)”, in Novoe v sovetskoj arxeologii, (ed.) Krupnov, E. I., Arcihovskij, A. V., Voronin, N. N. et al. (Moscow, 1965), p. 279Google Scholar.

59 Although the text of Zhoushu is not as explicit on this point as Sinor writes, who attributes to the Chinese author the expression “another tradition” to designate the second narrative, probably following Liu Mau-Tsai's translation: Liu, Die Chinesischen Nachrichten, p. 5. It is a little overtranslated. The paragraph in question begins with the characters huo yun 或云: huo 或 may alternatively mean “or, either” or “someone, some”. I have translated here by “some also tell”, in order to reflect both meanings simultaneously. In general, Sinor's arguments for making these two stories two separate legends seem unconvincing to me. The comparative tables, which he draws up to illustrate the fact that there is no relationship between the two legends (or rather, in this case, the three; Sinor, “The Legendary Origins”, pp. 231–232), do not demonstrate much, and even if he states that “Others may prefer a different choice or may whish to add to or delete from the list; the essential differences cannot really be bridged”, there are significant points of comparison. In addition to the motive of the ancestor wolf, we find the initial destruction of the people or the State at the beginning of the two stories; if the cave of the first story does not appear in the second, the mountain (where the cave is located?) is indeed present in it (and the mountain is indeed an element belonging to the structure of the myth, as illustrated by the role of the Qara Tagh in the legend of the First Man reported by Ibn ad-Dawādārī: Ibn ad-Dawādārī, Kanz ad-durar, p. 219–27; DeWeese, Islamization, pp. 280–281). Sinor notes that the two passages also have in common the election of the chief on his merit, and that the names of Ashina and Axian-she are found in both, but that the fact that the two distinct characters of the first are no longer one and the same person in the second must have been the trace of a “compromise between two traditions which may originally have been hostile” (Sinor, “The Legendary Origins”, p. 230). I think it is again the exact opposite: it has already been noted that the reason for the election of the leader, common to the founding myths of the steppe, was split in the first story, between Ashina and Axian-she (cf. n. 45). We can hypothesize that he was originally one and the same character, or even that this duplication is a distortion that occurred somewhere in the chain of transmission from the legend to the Zhoushu (for the name of Axian-she and its meaning, see below). We would more readily refer to Devin DeWeese's extensive comparative table, which takes more elements into account, and brings the Türk myth into resonance with other legends of steppic origin, including those already mentioned here; DeWeese, Islamization, pp. 508–509. Most of my disagreements with Sinor's interpretations stem, on the one hand, from his attention, which I think is too great, almost exclusively to the motive of the wolf, to the detriment of others that are just as essential, and on the other hand from his desire to historicize at all costs the Turk legends transmitted by Chinese sources.

60 Tekin, A Grammar, p. 233; translation p. 264.

61 If the expression “They [Tengri and the earth and water spirits] held my father Elterish Khagan, and my mother El Bilge Khatum, at the top of heaven and raised them upwards” must be taken in a figurative sense, Elterish and El Bilge being politically elevated above the people as their rulers, the image should not be without reminding the contemporary reader familiar with the türk myth the she-wolf held from the ground and carried into the sky by a spirit. Similarly, the image “the soldiers of my father, the kaghan, were like wolves” must have resonated strongly in the mind of the reader, aware that the Türks considered a wolf to be their ancestor, and this all the more so since the guard of the khagan, at least during the First Türk Empire, was called, according to Chinese sources, the Böri (fuli 附離), “the Wolves”; Zhoushu 50: 909; tr. Liu, Die Chinesischen Nachrichten, p. 9.

62 And not to Abangbu, whom the legend says is stupid. When the text of Zhoushu says that “Bangbu and his brothers were all stupid” 謗步等性竝愚癡, it is of course necessary to hear with the exception of Yizhini. This one is the son of a wolf (wether it is a wolf or a she-wolf is not clear from the chinese text), an obvious sign of his divine origin. He is distinguished (bie 別: this term and the fortune that favours Yizhini are also reminiscent of the distinction enjoyed by Elterish and El Bilge compared to the rest of the people; see previous note) by its power to control wind and rain, whic is commonly associated with sovereignty, particularly by the use that can be made of it in war. I would also be inclined to see behind the “fortune”, qi 氣 (“breath, manifestation of the soul or of the spirit”, but also “fortune, destiny”; Liu, Die Chinesischen Nachrichten, p. 6, translates it by “Geisterhauch”) of the Chinese text, the qut, or “sacred fortune”, of the Türks and the Uighurs; on this, see Roux, La religion, pp. 158–161. On the divine character of the wolf, see Roux, La religion, pp. 188–95. On the association between control of the elements and sovereignty, see Molnár, , Weather-Magic in Inner Asia (Bloomington, 1994)Google Scholar. It should be noted that the ability to control rain and wind most often comes from the possession of a particular stone, called yat or yada, whose Islamic sources often locate its origin in hard-to-reach places surrounded by mountains; see the discussions on this subject in Boyle, J. A., “Turkish and Mongol Shamanism in the Middle Ages”, Floklore LXXXIII, n°3 (1972), pp. 187193Google Scholar, and especially Molnár, Weather-Magic, pp. 1–59 in particular.

63 Seventeen, like seventy and seven hundred, are compounds of seven, a number that can be found with the seven Tatars of the origin myth of the Kimeks or the legend reported by Ibn ad-Dawādārī, with the seven survivors among the Mongolian nobles in Plano Carpini's account, or with the seven Mongol nations in Hayton's text. It should also be noted that in the Beishi version of the second story, Abangbu has seventy brothers (possibly explaining the translation error in Sinor, “The Legendary Origins”, p. 226); Beishi 99: 3286. Undoubtedly, the figure seven has strong symbolic power among the nomadic peoples of Central Eurasia, and Roux, J.-P., “Les chiffres symboliques 7 et 9 chez les Turcs non musulmans”, Revue de l'histoire des religions CLXVIII, n°1 (1965)Google Scholar multiplies the examples (pp. 35 ff.); the theories he puts forward to explain this symbolic charge (pp. 49–53) nevertheless seem to me to be unfounded, and we can only, in the present situation, note the recurrence and the importance of this figure in our sources.

64 Tekin, A Grammar, p. 234; translation p. 266.

65 Ibid., p. 235; translation p. 268.

66 Golden, Introduction, p. 119n. 26; Shiji 110: 2890.

67 With the exception of she 設 which, like shidu 師都 in Yizhini-shidu 伊質泥師都, transcribes the Turkish title shad. The specificity of this title easily explains why it would not have been translated.

68 This hypothesis seems to be reinforced by the fact that Abangbu 阿謗步 is called Bangbu 謗步 in the following sentence, which would tend to show that 阿 is used here as a surname, xing 姓. It could nevertheless be objected that 阿 is a character regularly used to transcribe proper names, especially foreign ones, and not only Ashina; thus, in the same text, the Afu River 阿輔.

69 This hypothesis is all the more tempting if, as we have said before, Ashina and Axian shad are the same character (see note 59): in the first story, Ashina is elected king because he is “the most intelligent”, zui xian 最賢; and, according to what I suppose, this is precisely what his name, or rather his title, Axian, means, as the second story specifies. To my knowledge, the Turkish word Bilge is only transcribed in Chinese, in the form of pijia 毗伽, from the compilations of the Jiu Tangshu 舊唐書 and the Xin Tangshu 新唐書, in 945 and 1060 respectively, which would mean that at the time of the compilation of Zhoushu and Beishi, in the middle of the 7th century, the practice would have been to translate it and not to transcribe it. However, it cannot be excluded that the authors did indeed wish to transcribe an approaching turkish name, albeit vaguely, by choosing the character xian 賢, precisely because it denoted the wisdom that they could detect as a trait of the character

70 Jiu Tangshu 194: 5173; translation Liu, Die Chinesischen Nachrichten, p. 171.

71 Jiu Tangshu 194: 5170–3; translation Liu, Die Chinesischen Nachrichten, pp. 168–171. The ascension to the throne of Bögö following his father, to the detriment of his cousin Bilge, was in principle in violation of the lateral succession in force among the Türks, from the eldest brother to the youngest, then from the sons of the eldest to the sons of the youngest. However, that Tonyuquq, the old counselor of Elterish, Qapaghan, and finally Bilge, was first among Bögö's supporters (and was the only one spared), seems to indicate that legitimacy was on the latter's side.

72 Tekin, A Grammar, pp. 231–232; translation pp. 261–262.

73 Zhoushu 50: 910; translation Liu, Die Chinesischen Nachrichten, p. 10, modified. The Suishu 隋書 adds with regard to the Western Türk khagans, who could not go themselves, that “every year they sent an [some?] official to the cave where their ancestors lived to offer a sacrifice” 歲遣重臣向其先世所居之窟致祭焉; Suishu 84: 1877; translation É. Chavannes, Documents sur les Tou-kiue (Turcs) Occidentaux (St Petersburg, 1903), p. 15.

74 The text then goes on to talk about an annual ceremony bringing all the Türks together on the banks of the Tamir, a tributary of the Orkhon, and then about a sacred mountain 400 or 500 li from the Ötüken. It can be assumed that if the cave had not been in the Ötüken, it would have been mentioned later in the text. Roux, however, places it in the Altai, which is said to be the place of origin of the Türks, as suggested by the first version of the myth as reported by the Zhoushu, as well as the Suishu 84: 1863; tr. Liu, Die Chinesischen Nachrichten, p. 40; Roux, La religion, p. 151.

75 Pace Sinor, D., “A propos de la biographie ouïghoure de Hiuan-tsang”, Journal Asiatique CCXXXI (1939), pp. 552553Google Scholar et Idem, “The Legendary Origins”, pp. 235–236, who, against the opinion of Pelliot, “Neuf notes”, p. 214n. 2, thinks that the grotto, called ku 窟, where the ceremony was held, has nothing to do with the cave, called xue 穴, where the wolf took refuge, but with the troglodytic habitats of the first Türks, which would be mentioned in passing in the Tongdian 通典 (it should be noted however that in the translation he gives of 突厥窟北: “North of the caves of the T'ou-kiue”, the plural is totally arbitrary). And to note, in support of this: “nothing in Legend A [i.e. the first version of the myth] suggests that the Türks’ ancestor was born in a cavern” (sic!); Sinor, “The Legendary Origins”, p. 236. The cave where the khagan and the nobles go once a year to sacrifice to the ancestors is the same as the one where the legendary she-wolf found refuge, in the sense that it concretely represents the mythical cave within the centre of the imperial Türk power, thus sacredized; this is what DeWeese calls the “nationalization” of the myth; DeWeese, Islamization, p. 277 n. 87.

76 Tekin, A Grammar, p. 234; translation p. 267.

77 Of course it cannot be completely excluded that the Kimeks had more or less directly knowledge of the Orkhon inscriptions and of their contents, and drew inspiration from them to create their original myth. However, this is less likely. The two hypotheses are not, in any case, exclusive to each other.

78 de La Vaissière, É., “Away from the Ötüken: A Geopolitical Approch of the seventh Century Eastern Türks”, in Complexity of Interaction along the Eurasian Steppe Zone in the First Millenium CE, (ed.) Bemman, J. and Schmauder, M. (Bonn, 2015), pp. 458459Google Scholar.

79 Ibid., p. 459.

80 DeWeese, Islamization, pp. 160–161.

81 The form Kelüren itself is regularly found in the SH, where it alternates with Kerülen. The place of residence of the pre-Chinggisid Mongols was between the Kerülen and the Onan. The two rivers originate in the sacred mountain of the Burqan Qaldun, the burial place of Chinggis Khan, a mountain that the SH explicitly identifies as a refuge: it is there that, by the will of Heaven, the still young Temüjin escapes the hunt of his Merkid enemies; SH, § 100–103; translation Rachewiltz, The Secret History, pp. 31–33. The Kerülen River flows into Lake Hulun, which in turn feeds the Argun River (Ergūne), whose name has been approximated by some to that of the Ergene Qūn: thus J., Tamura, “The Legend of the Origin of the Mongols and Problems concerning their Migration”, Acta Asiatica XXIV (1973), pp. 119Google Scholar, who presents an attempt to historicize the legend of the Ergene Qūn.

82 Although Börte Chino is not, strictly speaking, the direct ancestor of Chinggis Khan: the latter is indeed descended from Bodonchar, one of the three sons that Alan Qo'a, the widow of the descendant of Börte Chino, Dobun Mergen, had from a ray of light entering through the upper opening of her tent in the form of a man, and leaving it in the form of a yellow dog (it is the incarnation of Tengri). SH, § 17–22; translation Rachewiltz, The Secret History, pp. 3–5. Rashīd ad-Dīn, Jāmi‘ at-tavārīkh, p. 204; translation Thackston, Rashiduddin, p. 82. See also the discussion in Aigle, The Mongol Empire, pp. 126 ff.

83 Rashīd ad-Dīn, Jāmi‘ at-tavārīkh, p. 202; translation Thackston, Rashiduddin, p. 81; Abū’l-Ghāzī, Histoire des Mongols, pp. 34, 59; translation Desmaisons, pp. 33, 63.

84 Jūzjānī, Ṭabaqāt-i Nāṣirī, II, p. 100; tr. Raverty, Ṭabaqāt-i Nāṣirī, II, pp. 953–954. As it will seen later, it is relevant that this path is discoverned by Chinggis Khan's Muslim follower and merchant Ja‘far Khwāja.

85 Abū’l-Ghāzī, Histoire des Mongols, p. 34; tr. Desmaisons, p. 33.

86 See Rachewiltz, The Secret History, p. 224.

87 See for example Altan Tobči; translation Bawden, The Mongol Chronicle Altan Tobči (Wiesbaden, 1955), p. 113.

88 Aigle, The Mongol Empire, p. 133.

89 Hamayon, R., “Shamanism in Siberia: From Partnership in Supernature to Counterpower in Society”, in Shamanism, History and State, (ed.) Thomas, N. and Humphrey, C. (Ann Harbor, 1996), pp. 8384Google Scholar.

90 SH, § 1; translation Rachewiltz, The Secret History, p. 1.

91 Oghuz Nāma; translation Bang, W. and Rachmati, G. R., “Die Legende von Oγuz Qaγan”, Sitzungsberischte der Preuβischen Akademie der Wissenschaften, Phil.-Hist. Klasse XXV (1932), pp. 1021Google Scholar.

92 Michael the Syrian, Maktbānūt zabnē; translation Chabot, J.-B., Chronique de Michel le Syrien, patriarche jacobite d'Antioche (1166–1199) (Paris, 1905), III, pp. 151155Google Scholar; see the discussion in DeWeese, Islamization, p. 278.

93 Jordanes, De origine actibusque Getarum; translation O. Devillers, Histoire des Goths (Paris, 1995), p. 49.

94 Weishu 1: 2; translation Holmgren, J., Annals of Tai: Early T'o-pa History According to the First Chapter of the Wei-shu (Camberra, 1982), p. 52Google Scholar. The first place chosen to settle obviously refers to the enclosed space of the myth of origin.

95 Juvaynī, Tarīkh-i jahān-gushā, I, p. 45; translation Boyle, History of the World Conqueror, p. 61. This same word, which is also pronounced by the Oghuz animal guide, can be found in Rashīd ad-Dīn, precisely with regard to the exit of the Mongols from the Ergene Qūn: köch karda’ānd کوچ کرده اند, “they emigrated”; Rashīd ad-Dīn, Jāmi‘ at-tavārīkh, p. 139; translation Thakston, Rashiduddin, p. 56.

96 Monte Croce, Liber Peregrationis; translation Kappler, R., Pérégrination en Terre Sainte et au Proche-Orient. Lettres sur la chute de Saint-Jean d'Acre (Paris, 1997), pp. 9293, 98–101Google Scholar; the parallel with the Hunnic legend reported by Jordanes is blatant. On the guiding animals in Centrasiatic traditions, see Spinei, V., “Preliminary Notes on the Legend of the Ritual Hunt of the Guiding-Animal in the Mythology of the Eurasian Tribes and the Surroundings Peoples in the Middle Ages”, in Central Eurasia in the Middle Ages. Studies in Honor of Peter B. Golden, (ed.) Zimonyi, I. and Karatay, O. (Wiesbaden, 2016)Google Scholar.

97 An hypothesis already formulated in DeWeese, Islamization, p. 496 n. 8.

98 Hayton, “La Flor des Estoires”, p. 153. The Mount Belgian, which closes the passage to the Mongols, and at the foot of which lies the sea to be crossed, must be identified with Mount Burqan Qaldun, at the sources of the Onan and of the Kerülen, according to Aigle, The Mongol Empire, p. 61, or with the river or lake Baljuna, according to Stang, H., “The Baljuna Revisited”, Journal of Turkish Studies IX (1985), p. 230Google Scholar. It is in fact probably a confusion between the two places. In any case, at least if we agree with Pelliot's opinion that the lake Baljuna was located in the Kerülen basin (Pelliot and Hambis, Histoire des campagnes, pp. 46–48), this probably confirms that the Onan-Kerülen region was assimilated to the original enclosed place.

99 Aigle, The Mongol Empire, pp. 60–61. I therefore do not fully agree with Stang, “The Baljuna Revisited”, p. 230, that the episode of the partition of waters “must, alas, be discarded as apocryphal”: the withdrawal of waters by divine intervention is probably an addition by Hayton, but the crossing itself, certainly not. For an episode of Möngke's campaign against the Qipchaqs that may have inspired this story, see Juvaynī, Tarīkh-i jahān-gushā, III, p. 11; translation Boyle, History of the World Conqueror, p. 554.

100 SH, § 1; translation Rachewiltz, The Secret History, p. 1.

101 Roux, La religion, pp. 110 ff.; Aigle, The Mongol Empire, pp. 173 ff.

102 In the version of the legend of the Ergene Qūn contained in the epic poem called the Shāhanshāh Nāma, commissioned by the Ilkhan Abu Sa‘īd (1316–1335) to Aḥmad-i Tabrīzī, it is not stated who in particular discovers the iron vein, but the solution of melting the ore to clear a passage is given by an individualized character, although it is not explicitly mentioned that he was a blacksmith; see Boyle, “Some Thoughts”, pp. 186–187.

103 Hayton, “La Flor des Estoires”, p. 148.

104 Rubruck, Itinerarium; translation Jackson, P., The Mission of Friar William of Rubruck. His Journey to the Court of the Great Khan Möngke 1253–1255 (London, 1990), pp. 124, 247Google Scholar.

105 Pachymeres, Syngraphikai historiai; tr. Laurent, V., Relations historiques (Paris, 1984), II, p. 444445Google Scholar.

106 Nuwayrī, Nihāyat al-ārāb fī funūn al-ādab, (ed.) Sā‘id ‘Abd al-Fattāḥ ‘Āshūr (Cairo, 1985), XXVII, p. 302; Ṣafadī, Kitāb al-wāfī bi-al-wafayāt, (ed.) Shukrī Fayṣal (Wiesbaden, 1981), IX, p. 199; Ibn ad-Dawādārī, Kanz ad-durar, p. 231.

107 Ibn Baṭṭūṭa, Riḥla; translation Charles-Dominique, P., “Voyages et périples”, in Voyageurs arabes: Ibn Faḍlān, Ibn Jubayr, Ibn Baṭṭūṭa et un auteur anonyme, (ed.) Charles-Dominique, P. (Paris, 1995), p. 715Google Scholar.

108 Rachewiltz, The Secret History, p. 332; Pelliot 1959, pp. 289–290; pace Sinor, “The Legendary Origins”, pp. 248–249. Chance does things well here, which gave the future conqueror the name of the enemy his father had just defeated, according to an ancient Mongol tradition; see SH, § 59; translation Rachewiltz, The Secret History, p. 13. It is also to be noted that the Chinese rendition of his name, Tiemuzhen 鐵木真, includes the character tie 鐵, meaning “iron”.

109 Pelliot, P., Notes on Marco Polo (Paris, 1959), I, p. 290Google Scholar.

110 Pace Sinor, D., “Random Remarks on Metallurgical Themes in Pre-Modern Inner Asia”, in Scripta Ottomanica et Res Altaicae: Festschrift für Barbara Kellner-Heinkele zu ihrem 60. Geburtstag, (ed.) Hauenschild, I., Schönig, C. and Zieme, P. (Wiesbaden, 2002), p. 307Google Scholar, who is therefore doubly mistaken in asserting that “Although it would harmonize with the Ergene qun legend, I have my misgivings concerning the age-old etymology which would link Temüjin, Chinggis Khan's personnal name, with the Mongol word temürči ‘blacksmith’. At best it is a popular etymology based on assonance.”

111 Popov, A. A., “Consecration Ritual for a Blacksmith Novice among the Yakuts”, The Journal of American Folklore XLVI, n°181 (1933)Google Scholar; Roux, J.-P., “Fonctions chamaniques et valeurs du feu chez les peuples altaïques”, Revue de l'histoire des religions CLXXXIX, n°1 (1976), pp. 8386Google Scholar; Idem, La religion, pp. 79–80.

112 The elimination of the great shaman Teb Tenggeri, as reported by SH, might be interpreted by this position of the supreme sovereign, whose power is also religious, and who does not suffer from an intermediary in his relationship to Heaven and the supernatural world: SH, § 245; translation Rachewiltz, The Secret History, p. 172. See the discussion in Roux 1987, pp. 153 ff. It would perhaps be tempting to see in the supreme title of the Xiongnu, chanyu 单于 > EMC: *dān-γwaγ, behind which probably lies the title tarkhan (Pulleyblank, E. G., “The Consonantal System of Old Chinese”, Asia Major IX, 1962, pp. 91, 256257Google Scholar), the ancient expression of this duality of monarchical power among Eurasian nomads, between political and sacred royalties. However, it would seem that the meaning of darqan/tarkhan as “blacksmith” is a derivation of the darqan status that blacksmiths, and more generally artisans, received in the Mongol Empire, hence the meaning of “blacksmith, craftsman”: Atwood, Ch. P., Encyclopedia of Mongolia and the Mongol Empire (New York, 2004), p. 133Google Scholar.

113 Rashīd ad-Dīn, Jāmi‘ at-tavārīkh, p. 139; translation Thackston, Rashiduddin, p. 57, amended. Thackston's translation “to gather ironsmiths” is based on the text of Karīmī's edition which has qowm-i āhangarān قوم آهنگران “a group of ironsmiths”: Jāmi‘ at-tavārīkh, (ed.) B. Karīmī, Tehran: 1959, I, p. 110. We should however prefer the version of Rowshan and Mūsavī (whose edition is based on better manuscripts), dam-i āhangarān دم آهنگران , “ironsmiths’ bellows”. That is to say, it is the members of the Chinggisid family themselves who act as blacksmiths.

114 Abū’l-Ghāzī, Histoire des Mongols, p. 34; translation Desmaisons, p. 33, amended. I would like to express my deepest thanks to Marc Toutant for having kindly translated for me this passage, as well as all the other excerpts from the text of Abū'l-Ghazī cited in this article. It should be noted that the Shajarat al-Atrāk stipulates that the Mongols used to return each year to the Ergene Qūn to extract iron: Shajarat al-Atrāk; translation Miles, The Shajrat ul-Atrak, pp. 45–46.

115 Popov, “Consecration Ritual”, pp. 262 ff. It can also be noted that, in the myth of the Ergene Qūn, the melting of the iron mine by means of seventy bellows made from the skins of seventy horses and sheep, as Rashīd ad-Dīn tells us, and arranged in seventy different places, as Abū’l-Ghāzī tells us (seventy being a numinous number), and that it was blown at the same time, has all a magic operation; see below. It is also interesting to note that, according to the SH Chinggis Khan made his nightguards the guardians of, among other things, the gü'ürge which Rachewiltz translates as “drums”: SH, § 232, Index to the Secret History of the Mongols, (ed.) I. de Rachewiltz (Bloomington, 1972), p. 134; translation Rachewiltz, The Secret History, p. 160. It would seem that etymologically, the term refers to skins swollen with air in order to serve as floats to cross rivers: it is this idea of crossing, with that of stretched skin, which would be at the origin of a semantic shift towards the drum, and first of all that of the shaman, for whom it is a means of passage towards the other world. Now another meaning of gü'ürge is that of “bellows”: the ambiguity of the term indicates here again the shamanic nature of the very act of forging; see Skrynnikova, T. D., “Sülde – The Basic Idea of the Chinggis-Khan Cult”, Acta Orientalia Academiae Scientarum Hungaricae XLVI, n°1 (1992–1993), p. 54Google Scholar.

116 Zhoushu 50: 907; translation Liu, Die Chinesischen Nachrichten, p. 5. See also Suishu 84: 1863; translation Liu, Die Chinesischen Nachrichten, p. 40.

117 Roux, La religion, pp. 72, 80.

118 Jūzjānī, Ṭabaqāt-i Nāṣirī, II p. 199; tr. Raverty, Ṭabaqāt-i Nāṣirī, II, pp. 1077–1078. See also the passage describing the moment before the “exit” of the Mongols from Kelurān, during which Chinggis Khan gathered his soldiers at the foot of a mountain and made them fast and invoke Tengri for three days, while he remained isolated in his tent, a rope around his neck: he came out on the fourth day, announcing that Tengri had granted him victory; Jūzjānī, Ṭabaqāt-i Nāṣirī, II, pp. 100–101; translation Raverty, Ṭabaqāt-i Nāṣirī, II, p. 954. This passage is parallel to that of Hayton's Flor des Estoires, describing the Mongol army at the foot of the Mount Belgian, kneeling nine times in front of the sea, then spending the whole night in prayer for a passage to open; Hayton, “La Flor des Estoires”, p. 153.

119 See the discussion on this subject in Amitai, R., “Did Chinggis Khan Have a Jewish Teacher ? An Examination of an Early Fourteenth-Century Arabic Text”, Journal of the American Oriental Society CXXIV, n4 (2004), p. 698699Google Scholar.

120 See note 52.

121 See DeWeese, Islamization, pp. 273–287 in particular, on the equivalence between the oven in which Baba Tükles entered and emerged unscathed, symbolizing the birth of the new Muslim community of the Golden Horde, and the enclosed place from which the new political group emerged in several Central Asian traditions.

122 Abū'l-Ghāzī writes that “a tax in wood and coal was imposed on all the inhabitants of the country”: Abū’l-Ghāzī, Histoire des Mongols, p. 34; translation Desmaisons, p. 33.

123 Cleaves, “The Historicity”.

124 See the passages in question in Cleaves, Cleaves, “The Historicity”, pp. 370–372.

125 On the different versions of Baljuna's account, see Cleaves, “The Historicity”, pp. 396–419, and Stang, “The Baljuna Revisited”.

126 It seems to me, although the Yuanshi compiler apparently did not realize what it was about, that it is the Mongol cooking technique called boodog which is described here: “Cette technique consiste à désosser une bête […] en ne pratiquant qu'une ouverture aussi réduite que possible de la peau, le plus souvent au niveau du cou. La viande est préparée, découpée et assaisonnée pendant qu'un grand nombre de pierres, de préférence galets de rivière, sont mis à chauffer dans un feu. Une fois les pierres chaudes, les morceaux de viande, des oignons, de l'ail et des herbes aromatiques, parfois un peu d'eau, sont replacés dans la peau avec les pierres brûlantes. L'ouverture est alors refermée et la bête est placée sur le feu, grillée de l'extérieur pendant qu'elle cuit à l’étouffée de l'intérieur. La cuisson varie évidemment selon la taille de la bête, mais est assez rapide (quelques dizaines de minutes). L'ouverture du boodog est un moment de grande jubilation, la dégustation du bouillon précédant la distribution des morceaux de viande”; Legrand, J., “La cuisine mongole”, in Cuisines d'Orient et d'ailleurs : traditions culinaires des peuples du monde, (ed.) Aufray, M. and Perret, M. (Grenoble, Paris 1995), p. 124Google Scholar. I would like to thank Pierre Marsone and Adrien Dupuis for their helpful comments on the translation of this passage.

127 Yuanshi 120: 2960; translation Cleaves, “The Historicity”, p. 387.

128 Rashīd ad-Dīn, Jāmi‘ at-tavārīkh, p. 347; translation Thackston, Rashiduddin, p. 132, slightly modified. Rashīd ad-Dīn puts the moment when Chinggis Khan and his companions drank the water of Baljuna a little further in his story, during a second visit to this place that he is alone to mention: Rashīd ad-Dīn, Jāmi‘ at-tavārīkh, pp. 354–355; translation Thackston, Rashiduddin, p. 135.

129 DeWeese, however, has not failed to point this out, and even considers it to be an essential motif in the structure of the myth: DeWeese, Islamization, pp. 43–44, 273–286, 494–506. Should we go as far as to recognize Baljuna behind Jūzjānī's Balīq Jāq, from Turkish balïk, “mud” and the onomatopoeia chak indicating the idea of shock, from which may be “pressed mud” (to squeeze out the water? see also the OT. çaǧ çaǧ expressing the sound of flowing water)? See Clauson, An Etymological Dictionary, pp. 336, 403–5. Raverty proposes the meaning of “fast, violent spring”: Raverty, Ṭabaqāt-i Nāṣirī, II, p. 937n. 8. Baljuna's name itself is to be likened to the Turkish balchïq, “mud, swamp”: Pelliot and Hambis, Histoire des campagnes, p. 38.

130 Plano Carpini, Storia dei Mongoli, pp. 263–234; translation Dawson, “History of the Mongols”, p. 25; translation Tanase, “Histoire des Mongols”, p. 96; Bridia, 1 Hystoria Tartarorum, (ed.) Alf Önnerfors (Berlin, 1967), pp. 11–12; translation Painter, “The Tartar Relation”, p. 66; translation Tanase, “Histoires des Tartares”, p. 173.

131 In the Türk myth the young boy is fed by the she-wolf with meat, and Naduliu she saves the group by making fire. In the Kimek myth Shad shares his game with the seven Tatars. In Ibn ad-Dawādārī the young wild boy hunts for the seven refugees: Ibn ad-Dawādārī, Kanz ad-durar, p. 229.

132 Juvaynī, Tarīkh-i jahān-gushā, I, p. 27; translation Boyle, History of the World Conqueror, p. 37; Yuanshi 1: 10.

133 Yuanshi 122: 3008; translation Cleaves, “The Historicity”, p. 398.

134 Quoted in Cleaves, “The Historicity”, p. 409.

135 See Cleaves, “The Historicity”, pp. 400, 409–410.

136 Quoted in Cleaves, “The Historicity”, pp. 417–418.

137 Read 漫 for 曼.

138 Yuji, “Jingzhou lu zongguan Niegutai muzhi xian 靖州路總管捏古台墓誌餡”, in Quan Yuan wen 全元文 (Nanjing, 2004), XXVI, pp. 525–526. Pelliot quotes this passage in his commentary on the SWQZL, but he only had access to it through the simplified version of Tu Ji's Mengwuer Shiji 蒙兀兒史記 (152: 14a-b): Pelliot and Hambis, Histoire des campagnes, pp. 138–139. I would like to express my deepest thanks to Pierre Marsone for having provided me with the original text of this stele, and for having clarified its obscurities.

139 We can also notice that we find some elements present in this passage in the SH, when Temüjin, pursued by the Merkid, finds refuge on the Mount Burqan Qaldun, and gives thanks to it by kneeling nine times, his belt around his neck: SH, § 102–103; translation Rachewiltz, The Secret History, pp. 32–33. In Yu Ji's text as in Hayton's work Burqan Qaldun and Baljuna are therefore merged into each other (see note 98).

140 In an article under preparation I will discuss the conclusions that can be drawn from this present study regarding the representations that Eurasian nomads themselves had on their own socio-political organisation.

141 Hartog, F., Le Miroir d'Hérodote : essai sur la représentation de l'autre (Paris, 1991), p. 27Google Scholar.

142 See Bensa, A., La fin de l'exotisme. Essais d'anthropologie critique (Toulouse, 2006), p. 35Google Scholar.

143 Dobrovits, “The Turco-Mongolian Tradition”.