XIONGNU (Hsiung-nu), the great nomadic empire to the north of China in the 2nd and 1st centuries BCE, which extended to Iranian-speaking Central Asia and perhaps gave rise to the Huns of the Central Asian Iranian sources.

Origins. The Xiongnu are known mainly from archaeological data and from chapter 110 of the Shiji (Historical Records) of Sima Qian, written around 100 BCE, which is devoted to them. Comparison of the textual and archaeological data makes it possible to show that the Xiongnu were part of a wider phenomenon—the appearance in the 4th century BCE of elite mounted soldiers, the Hu (Di Cosmo, 2002), on the frontiers of the Chinese states which were expanding to the north. The first mention of the Xiongnu in Chinese sources dates to 318 BCE. Archaeologically, these Hu cavalrymen seem to be the heirs of a long development (the Early Nomadic period, from the end of the 7th to the middle of the 4th century BCE), during which the passage from an agro-pastoral economy to one dominated at times exclusively by equestrian pastoralism had taken place. Among these peoples, in the 4th and 3rd centuries BCE the Xiongnu occupied the steppe region of the northern Ordos as well as the regions to the northwest of the great bend of the Yellow River. Numerous archaeological finds in Inner Mongolia and in Ningxia demonstrate the existence of a nomadic culture that was socially differentiated and very rich, in which both iron and gold were in common use and which was in constant contact, militarily as well as diplomatically and commercially, with the Chinese states (in particular Zhao to the southeast). This culture seems to have directly inherited those which had preceded it in the region, but on a wider scale (the Xiongnu pottery of Xigoupan or Aluchaideng is a developed form of the Early Nomadic pottery of Taohongbala; development continued on the site of Maoqinggou from the Early Nomadic period to the Xiongnu period, with a transition to nomadism sometime around the sixth century BCE; for a different view see Minaev, 1996, p. 11, who, on the basis of funerary materials, places the origins of the Xiongnu further to the east, south of Manchuria (Upper Xiajiadan culture), and the response of Di Cosmo, 1999, pp. 931-34). We do not know what influence the Saka cultures to the west may have had in this development.

The name of the Xiongnu (ʿχiʷongʿnuo according to Karlgren) has also been related to other tribal names of Antiquity which have a rather similar pronunciation, following the vague identifications of the Shiji (Pritsak, 1959, strongly criticized by Maenchen-Helfen, 1961). Furthermore, the language of the Xiongnu has been the subject of the most varied hypotheses based on the few words, mainly titles or names of persons, which have been preserved in the Chinese sources: Altaic, Iranian (Bailey, 1985, which has not been followed) and Proto-Siberian (Ket). At present, the hypothesis of Pulleyblank (1962) in favor of Ket seems to be the most well-founded (Vovin, 2000), although it is by no means certain that all of the tribal groups of the confederation belonged to the same linguistic group nor that the late Xiongnu distich was representative of the language (Di Cosmo, 2002, pp. 163-65).

The Xiongnu Empire. The Xiongnu confederation was destabilized in 215 BCE by the offensive campaign of a China recently unified by the Qin, who sent the general Meng Tian to occupy and fortify the pastoral areas of the Ordos and to drive the Xiongnu and their shanyu Touman to the north. The Xiongnu tribes reunified under the charismatic figure of Touman’s son Maodun in 209, crushed the Emperor Gaozu, forced him to sign a humiliating treaty in 198, and reoccupied the Ordos. The status quo then prevailed until 134 BCE, a period during which the Xiongnu secured their pre-eminence over the steppe societies of East Asia. This period was brought to an end by the initiative of the Chinese, who expelled the Xiongnu to the north of the Gobi in 121 and 119 BCE.

In the domain of archaeology, the military domination of the Xiongnu gave rise to the following phenomena: First, the development of a proto-urbanization within the Xiongnu sphere as, for example, fortresses in the excavation of Ivolga on the Selenga by A. Davydova (1995); villages, whether fortified or not, as in Dureny (Davydova and Minyaev, 2002), where handicraft and agriculture were practiced in addition to animal husbandry, which has led to a rereading of the Shiji and Han shu in which it is actually specified that the shanyu must have had towns built in order to preserve the grain they received as tribute, and that they constructed a capital (Minaev, 1996); secondly, the enrichment and very clear sinicization of the contents of the tombs of the Xiongnu aristocracy, such as the tombs of Noin-Ula, a royal cemetery, investigated in 1924 by Kozlov (Rudenko, 1962), in which were found numerous deluxe objects imported as much from China as from the Iranian-speaking West, or the tombs excavated recently at Egiin Gol (Giscard, 2002).

The Xiongnu military domination also gave them control of Central Asia and put them in direct contact with the Iranian-speaking populations: In 176 BCE Maodun crushed the Yuezhi of Gansu, then subdued the Wusun, Loulan, the Hu Jie and “twenty-six peoples” of the region. In 162 the shanyu Laoshang again crushed the Yuezhi refugees in the valley of the Ili and forced them to migrate to the southwest into sedentary Iranian-speaking Central Asia (Sogdiana, Bactriana). At that time all of Central Asia recognized, at least formally, the suzerainty of the Xiongnu: “whenever a Xiongnu envoy appeared in the region [i.e., western Central Asia] carrying credentials from the Shanyu, he was escorted from state to state and provided with food, and no one dared to detain him or cause him any difficulty” (Shiji, tr. Watson, p. 244). Nevertheless, their control was primarily exercised in the northeast of the Tarim Basin and Turfan, with the Lob Nor as a western frontier: The Office of the Commander in Charge of Slaves, responsible for raising taxes and corvées, was established near Karashahr (Qarašahr). Control of the West seems to have been limited to the collection of tribute from the Wusun (Dzungaria) and Kangju (middle Syr Darya and Sogdiana), while further to the south the Yuezhi (Bactriana) were hostile to them.

Having been pushed back to the north by the Chinese, the Xiongnu entered into a period of internal divisions, during which the shanyu rapidly succeeded each other and the local Xiongnu kings fought over the central government. To the west, between the years 115 and 60 BCE, the weakening of the Xiongnu confederation gave rise to a struggle between the Chinese and the Xiongnu for control of the western regions. The principal events of this struggle included the missions of Zhang Qian in search of alliances in 137 and 115 BCE, the raid on Farḡāna (Ferghana) by a Chinese army in 101 BCE, and the battles for control of the region of Turfan (Jushi) between 67 and 60 BCE. The Office of the Commander in Charge of Slaves lasted until 60 BCE, at which time it was replaced by its Chinese equivalent, the Protectorate General of the Western Regions (see Yü Ying-Shih, p. 133).

In 57 BCE the disintegration of the confederation led to its division between five and then two shanyu, one in the South (Huhanye) who submitted to China in 53 BCE, the other (Zhizhi) controlling the North and West. The latter, finally taking refuge in Kangju, carved out a kingdom in the valley of the Talas and was defeated there by the Chinese general Zhen Tang in 36 BCE, an episode that marks the farthest advance of the Xiongnu and Chinese armies into the Iranian-speaking West (Daffinà, 1969).

The ensuing peaceful period ended when the Xiongnu took advantage of troubles in China (reign of Wang Mang, 9-23 CE) and widely recaptured control of the West before once again splitting into two groups, the Southern Xiongnu and the Northern Xiongnu, in 48 CE. The first group took refuge in the north of China in 50 CE, giving rise to areas of Xiongnu population within the frontiers between Taiyuan and the Yellow River that would endure for several centuries. Their last shanyu disappeared at the beginning of the 3rd century, but the Xiongnu, though highly sinicized, preserved their identity and played a major role in the disturbances and plundering that put an end to the Jin dynasty in North China at the beginning of the 4th century (Honey, 1990).

While the Northern Xiongnu for a time succeeded in playing a role in the West (their armies intervened at Khotan and Yarkand after 61 CE), China regained control of the region of Turfan in 74 CE and chased them from Mongolia: the shanyu took refuge in the Ili valley in 91 CE, while many Northern Xiongnu tribes surrendered to China and were settled within the frontiers. The Northern Xiongnu, with several thousand men, continued to intervene at Hami and in the region of Turfan throughout the first half of the 2nd century. We know nothing of their fate: in the Wei Lue, written in the middle of the 3rd century, the Xiongnu are completely absent from the piedmont north of the Tianshan (Chavannes, 1905).

Xiongnu and the Huns. Could these Xiongnu have given rise to the Huns who appeared on the Volga from the year 370 CE before they invaded Europe? The question is highly controversial and has been the subject of numerous works since de Guignes first proposed the identity of the two groups in 1758. The reference in one of the very early Sogdian documents conventionally called the Ancient Letters of 313 to the Xwn pillagers of Luoyang, where the Chinese sources speak of the Xiongnu, seemed to be decisive evidence in favor of this identification (Henning, 1948) before O. Maenchen-Helfen attempted to prove on several occasions that the two were unrelated, mainly using archaeological data, but also via critical examination of the texts. While we are indebted to the latter for having demonstrated the complexity of the Hun question (Maenchen-Helfen, 1973), and while his prudence has in the main prevailed (see for example Sinor, 1989, Daffinà, 1994, and also a recent synthetic view in Golden, 1992, pp. 57-67 and 77-83), an attempt will be made here to show that Maenchen-Helfen’s reasoning, quite valid from an ethnic point of view (the Huns were basically composed of a conglomerate of peoples), cannot be accepted in terms of political identity (la Vaissière, 2005b).

First, it seems possible to prove that the names are indeed identical. In 313 it was a Sogdian merchant writing in the Gansu corridor who, in a letter to a correspondent at Samarqand, described with precision the plundering of the Southern Xiongnu in China and called them Xwn, a name which must be connected to that of the Huns (Henning, 1948, a point conceded by Maenchen-Helfen, 1955, p. 101). Should it be connected to that of the Xiongnu (ʿχiʷongʿnuo)? This connection poses no problems to specialists in Chinese phonology (Pulleyblank, 1962, p. 139), and above all it is difficult to see what other origin one could give to this name which is always given as an equivalent to that of the Xiongnu in all of its first Central Asian occurrences: aside from the Sogdian Ancient Letter, one must also cite the Buddhist translations of Zhu Fahu (Dharmarakṣa), a Yuezhi of Dunhuang, who in 280 CE translated the Tathāgataguhya-sūtra from Sanskrit to Chinese, and rendered Hūṇa by Xiongnu (Lévy, 1905, pp. 289-90, the Sanskrit version is no longer extant, but there exists a Tibetan version which gives Hu-na), and then did the same in 308 in his translation of the Lalitavistara (of which a Sanskrit version is extant) (Daffinà, 1994, p. 10).

The Sogdians had been acquainted with the Xiongnu since the extension of their empire to western Central Asia in the 2nd century BCE, and one can no longer doubt the quality of the evidence (contra Sinor, p. 179, who presupposes that the name of the Huns is generic without asking why, and above all Maenchen-Helfen, 1955, who unconvincingly attempts to disparage this eyewitness testimony by comparing it to purely theoretical examples); solid reasons are required if we are to consider that the Sogdian merchant and the Yuezhi monk of Dunhuang did not give them their real name, for reasons which remain unknown (despite Parlato, 1996).

But these texts do not imply that the Huns of Europe or Central Asia after 350 were themselves descendants of the Xiongnu: one can imagine that the name Xwn or Huna — an accurate word for describing the Southern Xiongnu who plundered North China in the 4th century as well as the ancient Xiongnu, known as far as India — may then have been used again for very different nomadic peoples. Further still, we have proof of such usage: in Sogdiana in the 8th century, the Turks are sometimes named Xwn (Grenet, 1989), and certain Huns of some Khotanese texts could not be Xiongnu (an example in Bailey, 1949, p. 48). But this generic name did not develop from nothing, and only the Xiongnu hypothesis can account for it.

Moreover, the Wei shu, taking up information precisely dated to 457, states: “Formerly, the Xiongnu killed the king (of Sogdiana) and took the country. King Huni is the third ruler of the line” (trans. Enoki, 1955, p. 44). This leads us to place the “Xiongnu” invasion of Sogdiana in the first half of the 5th century. Here, too, there is hardly any reason to doubt this direct testimony stemming from the report of an official Sogdian envoy in China (see Enoki, 1955, pp. 54-57). Also, the personal names found in the Sogdian caravaneer graffiti of the Upper Indus (3rd to 5th century CE, see Sims-Williams, 1989 and 1992) frequently include the first or last name Xwn, whereas it no longer exists in the later corpus of texts (the Chinese documents of Turfan), which reflects the presence of Hun invaders in Sogdiana and the fusion of the populations (la Vaissière, 2004, pp. 77-78; 2005, pp. 81-82) during a precise period of time.

Nevertheless, this still does not imply that the invaders were Huns/Xiongnu, but at least that they claimed to be. In order to proceed further, it is first necessary to stress the extent to which the testimony of the Wei shu concerning the Xiongnu in the West is isolated. Beginning with the hypothesis of de Guignes, scholars have sought on several occasions to identify a westward migratory movement of the Xiongnu. For a long time, the episode cited above (the establishment of the shanyu Zhizhi in the valley of the Talas) was used for that purpose: the Xiongnu who accompanied Zhizhi to the West were considered to be the ancestors of the Huns of Europe. This is impossible, since the Chinese sources emphasize the small number of these Xiongnu (Daffinà, 1969, pp. 229-230). Even if one takes into account the latest reports of the Northern Xiongnu north of the Tian Shan (153 CE), two centuries still separate them from the invasion of Sogdiana, while we have no reason to suppose the existence of a westward movement of the Southern Xiongnu. But other passages in the Wei shu speak of “remnants of the descendants of the Xiongnu” as western neighbors of a branch of the Rouran (Juanjuan), to the northwest of the Gobi around 400 CE (Wei shu, 103.2290). This information is not without interest, for it implies the survival of a Xiongnu identity far to the north, well beyond the field of vision of the Chinese sources, in the very place one would expect to find Huns/Xiongnu shortly after some of the main body of their troops had passed the Volga and others the Syr Darya, leaving these small groups behind them.

From an archaeological point of view, there are now few doubts that the Hunnic cauldrons from Hungary are indeed derived from the Xiongnu ones. Moreover, they were used and buried on the same places, the banks of rivers, a fact which proves the existence of a cultural continuity between the Xiongnu and the Huns (Erdy, 1994; de la Vaissière, 2005b).

The Huns of Central Asia thus consciously succeeded the Xiongnu and established themselves as their heirs, and an authentic Xiongnu element probably existed within them, although it was probably very much in the minority within a conglomerate of various peoples. This is the only hypothesis that accounts for all of the known facts given the current state of our information.

The Xiongnu/Huns in Central Asia in the 4th century. We shall not deal here with the Huns who appeared around 370 CE on the Volga prior to their invasion of Europe (see Maenchen-Helfen, 1973), nor with the Central Asian dynasties of the 5th century that are sometimes called Hunnic in the sources (on the Kidarites, who appeared only from the year 420, see Grenet, 2002; see KIDARITES). In Central Asia, the first references that must be taken into account date from 350 CE. Ammianus Marcellinus then mentions for the first time in his narrative the eastern enemies of the Persians, the Chionites (XVI, 9, 3-4). In 356, Šāhpur II fought against the Chionites in the East, then concluded an alliance: the king of the Chionites, Grumbates, participated in the siege of Amida (Diyarbakir) in 359 (XIX, 1, 7). This name is now attested around the year 430 in the Bactrian documents for a prince of the kingdom of Rōb (north of the Hindu Kush) in the form Gurambad (Sims-Williams, 1997, 13 doc. 4). The influence of Avestan Hyaona might suffice to explain the form Chion, which is divergent in comparison with Xwn, Huna, and Hūṇa. These Chionites could have reached the Sassanid frontiers in 350 CE. Boris Marshak, for other reasons (1971, p. 65) considers the Chionites to have been not Huns but mountain-dwellers of the Hindu Kush, an hypothesis supported by the fact that Ammianus Marcellinus, who knew of both the European Huns and the Chionites, never makes the connection between them (Sinor, p. 179), and possibly by the name of Gurambad. It would then be necessary to situate the invasion of the Xiongnu/Huns one or two generations later: at that time the Armenian sources show that the Sassanids, between 368 and the death of Šāpur II (379), were defeated by a “king of the Kushans” reigning at Balḵ- (Faustus of Byzantium, V, vii and V, xxxvii, tr. Garsoian, 1989, pp. 187-98 and 217-18), and the Kushano-Sassanid kingdom collapsed at that time. Moreover, it was at the end of the 4th century that the alchonno of Kapisa and Gandhara began to strike their own coins re-using those of Šāpur II, which would be perfectly consistent chronologically (on these coins see Alram, 1996, who associates them with the Kidarites, and more generally Göbl). But this reading is contested as some coins can be read alchanno, a name which should be linked with the indian legend rājālakhāna.

This invasion and those that followed it shattered the sedentary economy of Iranian-speaking Central Asia: Bactriana, ravaged for more than a century (until the expansion of the Hephtalites in the middle of the 5th century), declined as the principal center of population and wealth, due as much to the nomadic offensives as to the vigorous Sassanid resistance, while Sogdiana, which had been rapidly conquered, recovered. In Bactriana, all of the available data agree in giving the idea of a sharp decline of the region in the course of the period from the second half of the 4th century to the 6th century: neglected irrigation networks (valley of the Wakhsh), multiple layers of burning (Chaqalaqtepe), abandonment of sites (Dil’beržin Tepe, Emshi Tepe), barren layers in the stratigraphy of sites (Tepe Zargarān at Balḵ-), necropolises over ancient urban areas (Termeḏ, Dal’verzintepe), sacking (Karatepe). In contrast, to the north the populations of the region of the Syr Darya, whether of the delta (Džetyasar culture) or of the middle course (Kaunči culture), seem to have taken refuge in Sogdiana under Hun pressure and rapidly returned the abandoned lands to cultivation. Conversely, the sites of the Džetyasar culture were widely abandoned, and at the middle course of the Syr Darya the city of Kanka diminished to a third of its initial surface area (Grenet, 1996, de la Vaissière, 2002, pp. 105-16). It may also be noted that the sites of Džetyasar are close to the areas in which the Western sources place the European Huns prior to their crossing of the Volga. These peoples who arrived from the north added to the local Sogdian populations, which did not disappear. Sogdiana rapidly rebuilt itself in the 5th century under a stable Xiongnu dynasty, and then under the Kidarites.


M. Alram, “Alchon und Nēzak. Zur Geschichte der iranischen Hunnen in Mittelasien,” in La Persia e l’Asia Centrale da Alessandro Magno al X secolo,Atti dei Conveigni Lincei, 127, Rome, 1996, pp. 519-54.

H. W. Bailey, “A Khotanese Text concerning the Turks in Kantsou,” Asia Major, NS 1, 1949, pp. 28-52.

Idem, “Iranian in Hsiung-nu,” in Monumentum Georg Morgenstierne, I, Leiden, 1981, pp. 22-6.

Idem, Indo-Scythian Studies: being Khotanese Texts, VII, Cambridge, 1985, pp. 25-41.

T. J. Barfield, “The Hsiung-nu Imperial Confederacy: Organization and Foreign Policy,” Journal of Asian Studies 41/1, 1981, pp. 45-61.

É. Chavannes, “Les Pays d’Occident d’après le Wei Lio,” T’oung Pao, series II, vol. VI, 1905, pp. 519-71.

P. Daffinà, “Chih-chih Shan-Yü,” Rivista degli Studi Orientali, 1969, 44/3, pp. 199-232.

Idem, “Stato presente e prospettive della questione unnica,” in S. Blason Scarel, ed., Attila. Flagellum dei?, Rome, 1994, pp. 5-17.

A. Davydova, Ivolga archaeological complex. Part 1. Ivolga fortress (Archaeological Sites of the Hsiungnu, 1), Saint Petersburg, 1995.

A. Davydova, S. Minyaev, Complex of Archaeological Sites near Dureny (Archaeological Sites of the Hsiungnu, 5), Saint Petersburg, 2002.

N. Di Cosmo, “The Northern Frontier in Pre-Imperial China,” in M. Loewe and E. L. Shaughnessy, eds., The Cambridge History of Ancient China, chap. 13, Cambridge, 1999, pp. 885-966.

Idem, Ancient China and its Enemies. The Rise of Nomadic Power in East Asian History, Cambridge, 2002.

K. Enoki, “Sogdiana and the Hsiung-nu,” Central Asiatic Journal 1, 1955, pp. 43-62.

M. Erdy, “An Overview of the Xiongnu Type Cauldron Finds of Eurasia in Three Media, with Historical Observations,” inB Genito, ed., Archaeology of the Steppes, Naples, 1994, pp. 379-438.

N. Garsoian, tr., The Epic Histories Attributed to P’awstos Buzand (Buzandaran Patmut’iwnk), Cambridge, Mass., 1989.

P.-H. Giscard, “Le premier empire des steppes,” in J.-P. Desroches, ed., L’Asie des Steppes d’Alexandre à Gengis-Khan, Paris, 2002, pp. 135-41.

R. Göbl, Dokumente zur Geschichte der iranischen Hunnen in Baktrien und Indien, 4 vols., Wiesbaden, 1967.

P. Golden, An Introduction to the History of the Turkic Peoples,Turcologica 9, Wiesbaden, 1992.

F. Grenet, “Les ‘Huns’ dans les documents sogdiens du mont Mugh (avec un appendice par N. Sims-Williams),” Études Irano-aryennes offertes à Gilbert Lazard, Cahier de Studia Iranica, 7, 1989, pp. 165-84.

Idem, “Crise et sortie de crise en Bactriane-Sogdiane aux IVe-Ve s. de n.è.: de l’héritage antique à l’adoption de modèles sassanides,” in La Persia e l’Asia Centrale da Alessandro Magno al X secolo,Atti dei Convegni Lincei, 127, Rome, 1996, pp. 367-90.

Idem, “Regional Interaction in Central Asia and North-West India in the Kidarite and Hephtalite Period,” in N. Sims-Williams, ed., Indo-Iranian Languages and Peoples,Proceedings of the British Academy 116, London, 2002, pp. 203-24.

W.B Henning, “The Date of the Sogdian Ancient Letters,” BSOAS, 12/3-4, 1948, pp. 601-15.

D.B Honey, The Rise of Medieval Hsiung-Nu: The Biography of Liu-Yüan,Papers on Inner Asia, 15, Bloomington, 1990.

É. de la Vaissière, Histoire des marchands sogdiens,Mémoires de l’IHEC, 32, Paris, 2004.

Idem, Sogdian traders. A History, Leiden and Boston, 2005

Idem, “Huns et Xiongnu,” Central Asiatic Journal, 49, 2005b, pp. 3-26.

S. Lévy, “Notes chinoises sur l’Inde V. Quelques documents sur le bouddhisme indien dans l’Asie centrale,” Bulletin de l’École Française d’Extrême-Orient 5, 1905, pp. 253-305.

O. Maenchen-Helfen, “Pseudo-Huns,” Central Asiatic Journal l, 1955, pp. 101-16.

Idem, “Archaistic Names of the Hiung-nu,” Central Asiatic Journal 6, 1961, pp. 249-61.

Idem, The World of the Huns. Studies in Their History and Culture, Berkeley, 1973.

B. I. Marshak, “K voprosu o vostochnykh protivnikakh Irana v V v.” [On the problem of the Oriental enemies of Iran in the fifth century], Strany i narody Vostoka, 10, 1971, pp. 58-66.

S. Minaev, “Archéologie des Xiongnu en Russie. Nouvelles découvertes et quelques problèmes” Arts Asiatiques 51, 1996, pp. 5-12.

S. Minyaev, Derestuj Cemetery (Archaeological Sites of the Hsiungnu, 3), Saint-Petersburg, 1998.

S. Parlato, “Successo euroasiatico dell’ etnico žUnni’,” in La Persia e l’Asia Centrale da Alessandro Magno al X secolo (Atti dei Convegni Lincei, 127), Rome, 1996, pp. 555-66.

O. Pritsak, “Xun, der Volksname der Hsiung-nu,” Central Asiatic Journal 5, 1959-60, pp. 27-34.

S. I. Rudenko, Kul’tura Khunnov i Noinulinskie kurgany [The Culture of the Huns and the Kurgans of Noin-Ula], Moscow, 1962.

N. Sims-Williams, Sogdian and other Iranian Inscriptions of the Upper Indus, I-II, Corpus Inscriptionum Iranicarum, II/III, London, 1989-1992.

Idem, New Light on Ancient Afghanistan. The Decipherment of Bactrian, London, 1997.

D. Sinor, “The Hun Period,” in D. Sinor ed., The Cambridge History of Early Inner Asia, Cambridge, 1990, pp. 177-205.

A. Vovin, “Did the Xiong-nu Speak an Yeniseian Language?” Central Asiatic Journal 44-1, 2000, pp. 87-104.

B. Watson, tr., Sima Qian, Records of the Grand Historian: Han Dynasty II, New York, 1993.

Ying-Shih Yü, “The Hsiung-nu,” in D. Sinor, ed., The Cambridge History of Early Inner Asia, Cambridge, 1990, pp. 118-49.

November 15, 2006

(Étienne de la Vaissière)

Originally Published: November 15, 2006

Last Updated: November 15, 2006