This work writes Khan throughout, in lemmata and in running text, without exception. Where the sources or the literature write khagan, qaghan, qaγan or kağan, the form they give is held in the frontmatter field rank_attested of the person concerned and in the list above. This article is the one place where the differing spellings, the history of the title and the ranks around it are worked out.

Etymology

The title of the supreme ruler of the steppe from the fifth century onward is written 可汗 in the Chinese histories and 𐰴𐰍𐰣 in the Göktürk inscriptions, and is read khagan, qaghan or kagan in modern scholarship.[1] The shorter form khan, 汗, appears in the Chinese sources beside it and is the form that survived into every later Turkic and Mongolian language. Whether the two were distinct ranks in the sixth century, the khagan supreme and the khan subordinate, or one title under two spellings, is not settled; in the Göktürk inscriptions of the eighth century only kagan occurs, and it is used of the supreme ruler alone.[1] The distinction between a khagan and lesser khans beneath him is clear in the later steppe empires, and the literature commonly projects it back; this work does not, because the sixth-century sources do not require it, and writes Khan for every holder.

The origin of the word is unknown. The usual view is that it was coined among the Xianbei or the Rouran, and its predecessor among the Xiongnu was a different word. It is older than the Rouran khaganate: the title is recorded among the Xianbei in 261 and again in 278, more than a century before Shelun, so the tradition that makes Bumin’s Rouran predecessors the first to bear it cannot stand.[2] It has been derived from an “old Hunnic” kehan, with the Rouran adoption as the taking over of a title already in use; that is the position of this work, and it is recorded as such.[3] The two datings that have been given for the word are no longer in conflict once the Xianbei evidence is counted: the Rouran founder Shelun took the title about 400, and the word itself is third-century.

Doerfer’s Rouran derivation, and why it fails

Gerhard Doerfer proposed that khan, khagan and khatun are Rouran words. The argument has two steps: none of the three has a reliable etymology in any known language, and all three are first attested among the Rouran — from whom, on this reading, the Hunnic peoples, the Avars and everyone else in the steppe and beyond took them over.[4] It is a widely repeated derivation, and it appears in reference works that do not mention the objection.

The second step is wrong, and the first step depends on it. The title is recorded among the Xianbei in 261 and again in 278.[2] The Rouran khaganate did not exist then. Its founder Shelun took the title about 400, more than a century later. The Rouran therefore cannot be the people among whom the word is first attested, and a word first written down among the Xianbei in the third century is not a Rouran coinage. What Shelun did in about 400 was to adopt a title already in use, which is what this work holds and what the Xianbei dates require.[3]

The argument from the missing etymology falls with it. It was never an argument that the word is Rouran; it was an argument that the word is not Turkic, not Mongolic and not Chinese, and that the earliest attestation should decide the rest. Move the earliest attestation back a century and a half and the same reasoning would make the word Xianbei — which is not a conclusion anyone has wanted to draw, and which shows what the reasoning was worth. What remains standing is the negative finding: the origin of the word is unknown, and no language yet claims it convincingly.

Two of the three words are on better footing than khan itself. For the khatun a different derivation is on offer and is the best supported of the three: Turkic khatun is taken for a word of ultimate Eastern Iranian origin, borrowed from an Early Saka form and matched by the attested Sogdian xwt’w, ruler, and xwt’yn, wife of the ruler.[4] For khagan itself a Middle Iranian hva-kama, self-ruler, has been proposed, which would put the highest title of the steppe in the same borrowed company.[4] An attempt to read khan and khagan as Yeniseian roots inflected with a Tabghach suffix has been made and rejected, on the ground that a hybrid formation of that kind, unattested as a whole, would let one explain any word in any language.[4]

Before the Göktürk

The ruler of the Xiongnu had been the chanyu, and his full title, as the Han history transcribes it, has been read as containing the Turkic words teŋri and Kut: this work, following Grousset, reads the whole as Teŋri Kut Tanyu, “the ruler endowed by Heaven with fortune”, and holds that the Xiongnu ruling title was Turkic where the khaganal title was not. The article on the Xiongnu and the article on the ethnogenesis debate set out that reading and its standing. What is not in doubt is the sequence: chanyu fell out of use when the Xiongnu fell, and khagan took its place with the Rouran. What did not change with the word was the claim it carried. The Xiongnu title contains tengri, and that word is not the Chinese one; whoever formulated a heaven-given sovereignty in the steppe did it in a steppe language, before the khaganal title existed.[5]

One structural inheritance runs with the title. The Xiongnu paired every office into a left and a right, and the pairing was not, as in China, a matter of where a man stood at court: it was a territorial order, two branches of government administering two halves of the empire, and its nearest parallel anywhere is the division of the Göktürk realm into an eastern and a western half under a senior and a junior ruler.[5] The article on the making of the first steppe empire carries the argument in full. The Chinese, oddly, went on calling the Göktürk rulers chanyu in some contexts long after the Göktürk themselves used khagan, which this work reads as a memory of continuity and which may equally be a habit of the chancery.

Title among the Göktürk

Bumin took the title in 552, after the destruction of the Rouran, in the form the Zhou Shu renders 伊利可汗, conventionally read Illig Khan.[6] Before the war he appears without a title of any kind.[6] Modern literature commonly styles him yabgu for those years, which is plausible given the rank the Ashina leaders are said to have held, but the sources do not use it of him. The meaning of the first element is contested. This work, with much of the literature, glosses it “ruler of the land”, from Turkic el, “realm”; the translations offered in modern accounts are interpretations rather than readings, and the element is not securely explained.

The sons who followed him bore titles of the same construction, an epithet before the word khan, and in every case the epithet rather than the personal name is what the Chinese recorded: Issik, 乙息記可汗; Mukan, 木汗可汗; Taspar, 他鉢可汗.[6] This work reads Issik as a title meaning ruler of the state, as distinct from Illig as ruler of the land, and reports the personal name of its holder as Kulo, reconstructed as Kara; the reading rests on the Chinese transcription alone.

The Göktürk’s own understanding of the title is in the inscriptions. Bilge Khan opens his monument as “Bilge Khan of the Türk, like Heaven and born of Heaven”; his father was given the realm because Heaven willed it and because he had Kut; and the khan who lost the realm did so because he was ignorant and bad and his ministers likewise.[1] The khan held the realm, el, and set the law, törü, in order, and the two are named together every time the founders are named.[1] The articles on Kut, Töre and Tengrism set out what that meant. It did not mean autocracy. A khan could be deposed, and the nobility assembled in a toy chose a successor when the succession was disputed, as it did in 581.[7]

Ranks below the khan

The khan stood at the head of a hierarchy the Göktürk had taken over from the Rouran and filled with their own kin. Beneath him the western wing was held by a yabgu: Istemi held that rank for twenty-three years, and the Chinese histories treat it as a subordinate office of the eastern court, while the Göktürk inscriptions give him the same title as his brother.[7] This work reads Istemi’s title as Sır Yabgu, an elevated yabgu above a mere tribal chief and below a khan, and reads the inscriptions’ “Istemi Khan” as the founders’ descendants legitimating the autonomy the western ruler had in fact enjoyed; both readings are this work’s. The Byzantine envoys, who met Istemi in person, called him the ruler of the Göktürk without qualification.[8]

Under Taspar the hierarchy grew a further layer. He set his nephew Ishbara over the east as Erfu Khan, 爾伏可汗, and the son of his brother Rudan Khan over the west as Buli Khan, 步離可汗, so that the khan had deputies who bore the title; this work reads it as the point at which the title began to be held by more than one man at a time.[7] The other ranks the sources name are the šad, a prince holding a military command over a wing or a people; the tegin, a prince of the blood; the elteber, the tributary king of a subject people; the tarkhan, a commander freed from tax; the buyruk, a minister; and the beg, the lord of a tribe or a following, whose consent the khan needed and whose withdrawal of it undid him. Most of these appear in the Kül Tegin inscription; the yabgu and the šad are named there as offices the founders gave to the Tölis and the Tarduš, and the begs and the people are named as the two parties whose straightness with each other let the realm be held.[1]

The khan’s consort was the khatun, in Chinese 可賀敦, and the Western Wei princess Changle became khatun when Bumin took the title: 號其妻為可賀敦，亦猶古之閼氏也, he called his wife khatun, as of old the yanzhi of the Xiongnu.[6]

Later history

The title outlived the polity. When the Chinese courts of the early seventh century wanted a Göktürk client they conferred the title on him, and when a Göktürk khan wanted a Chinese client he did the same: Shibi Khan gave the title to the warlord Liang Shidu and a wolf’s-head standard with it.[9] A Göktürk khan who received his title from a Chinese emperor was, from the standpoint of the inscriptions, a khan who had given up his realm and his law to the Chinese khan; that is how the Kül Tegin inscription describes the fifty years between 630 and 682, and the reproach is aimed at the title’s holders, not at the title.[1]

Khagan passed from the Göktürk to the Uyghurs, the Khazars, the Avars of Europe, the Bulgars and the Mongols, and khan, in the shorter form, to every ruler of the Turkic and Mongol world thereafter. This work writes the shorter form for all of them, and this article is where the longer one lives.

References

  1. ^ a b c d e f Kül Tegin inscription
  2. ^ a b Mau-tsai 1958, Turkish edition
  3. ^ a b Yavuz 2025
  4. ^ a b c d Savelyev 2020
  5. ^ a b Cosmo 2011
  6. ^ a b c d Zhou Shu (Book of Zhou), juan 50
  7. ^ a b c Sui Shu (Book of Sui), juan 84
  8. ^ Menander Protector
  9. ^ Zizhi tongjian (Comprehensive Mirror in Aid of Governance), juan 185

Sources

Literature

  • Emre-E. Yavuz, The Gokturk Empire: 2nd Edition (2025)
  • Alexander Savelyev; Choongwon Jeong, Early Nomads of the Eastern Steppe and Their Tentative Connections in the West (2020)
  • Nicola Di Cosmo, Ethnogenesis, Coevolution and Political Morphology of the Earliest Steppe Empire: The Xiongnu Question Revisited (2011)
  • Liu Mau-tsai, Çin Kaynaklarına Göre Doğu Türkleri (1958, Turkish edition)

Related

What links here16

Persons
Bumin Khan · Ilig Khan · Ishbara Khan · Issik Khan · Mugulü · Taspar Khan
Polities
Khazar Khaganate · Rouran Khaganate · Second Göktürk Khaganate
Peoples
Avars
Institutions and Titles
Atabeg · Beg · Ilteber · Shad · Tuğ · Yabgu

Cite this article

Yavuz, Emre-E. “Khan (Khagan).” Turkic Atlas, revised 13 September 2026. https://turkicatlas.business-5d4.workers.dev/institutions/khan-khagan/

Last revised: 13 September 2026