The Kushan Empire and its languages
« previous post | next post »
Jean Nota Bene, the biggest French YouTuber (millions of followers) on historical subjects, recently focused on the Kushans. He follows many of the same themes that we do on Language Log and Sino-Platonic Papers (including Greek-Indian-Chinese associations), so many readers of this post will be interested in what he has to say about the Kushan Empire (ca. 30–ca. 375 AD). Although Nota Bene speaks in French, I think readers will be able to glean a lot of valuable information on this subject. Plus his presentation is richly illustrated, so watch carefully and pause the video if you want to take a closer look.
Nota Bene touches on the history, religion, art, architecture, ethnicity, locale, numismatics, and virtually all other major aspects of their political and cultural identity.
The Kushans migrated vast distances in response to pressure from enemies and the attraction of new lands to occupy
Here are the common languages of the Kushan Empire:
| Greek (official until c. 127) Bactrian (official from c. 127) Gandhari Prakrit Hybrid Sanskrit |
The Kushan Empire (c. 30–c. 375 CE) was a syncretic empire formed by the Yuezhi in the Bactrian territories in the early 1st century. It spread to encompass much of what is now Tajikistan, Uzbekistan, Afghanistan, Pakistan, Eastern Iran and Northern India, at least as far as Saketa and Sarnath, near Varanasi, where inscriptions have been found dating to the era of the Kushan emperor Kanishka the Great.
In addition, the people bearing the demonym Yuezhi were a key, formative component of the confederation of Indo-European nomadic tribes who ultimately formed the Kushan Empire.
Under the rule of the Kushans, northwest India and adjoining regions participated both in seagoing trade and in commerce along the Silk Road to China. The name Kushan derives from the Chinese term Guishang, used in historical writings to describe one branch of the Yuezhi—a loose confederation of Indo-European people who had been living in northwestern China until they were driven west by another group, the Xiongnu, in 176–160 B.C. The Yuezhi reached Bactria (northwest Afghanistan and Tajikistan) around 135 B.C. Kujula Kadphises united the disparate tribes in the first century B.C. Gradually wresting control of the area from the Scytho-Parthians, the Yuezhi moved south into the northwest Indian region traditionally known as Gandhara (now parts of Pakistan and Afghanistan) and established a capital near Kabul. They had learned to use a form of the Greek alphabet, and Kujula’s son was the first Indian ruler to strike gold coins in imitation of the Roman aureus exchanged along the caravan routes.
(The Met)
There is a strong likelihood that the Yuezhi were linked to the Tocharians, the second oldest (after the Hittites) Indo-European people, who had affinities with the centum group to the west and early loans into Sinitic, after passing through Central Asia. See Victor H. Mair, “Die Sprachamöbe: An archeolinguistic parable,” The Bronze Age and Early Iron Age Peoples of Eastern Central Asia, 2 vols. (Washington D.C. and Philadelphia: Institute for the Study of Man and the University of Pennsylvania Museum, 1998).
The Kushans were most probably one of five branches of the Yuezhi confederation, an Indo-European nomadic people of possible Tocharian origin, who migrated from northwestern China (Xinjiang and Gansu) and settled in ancient Bactria. The founder of the dynasty, Kujula Kadphises, followed Iranian and Greek cultural ideas and iconography after the Greco-Bactrian tradition and was a follower of the Shaivite sect of Hinduism. Two later Kushan kings, Vima Kadphises and Vasudeva II, were also patrons of Hinduism. The Kushans in general were also great patrons of Buddhism, and, starting with Emperor Kanishka, they employed elements of Zoroastrianism in their pantheon. They played an important role in the spread of Buddhism to Central Asia and China, ushering in a period of relative peace for 200 years, sometimes described as "Pax Kushana".
Because of the recent craze for ancient DNA studies, there has been much speculation about how genetic analysis can help us understand where the homeland of the PIEs may have been (around 4000 BC, the Pontic-Caspian steppe region [between the Black Sea and the southern Ural Mountains]), and the pattern of their dispersal during the following two to three thousand years. In contrast, I have come to rely more heavily on hard archeological evidence (horses, chariots, textiles, bronze weapons and other implements, pottery, botany, zoology, biology, etc., and, of course, the development and relationships of languages.
That is to say, I prefer empirical, direct, precisely measured evidence over speculative, statistically derived results from what is essentially chemical, statistical research.
Pots, bones, stones, and artifacts over ciphers!
Selected readings
- J. P. Mallory, The Problem of Tocharian Origins: An Archaeological Perspective (Sino-Platonic Papers, 259 [Nov. 2015]; free pdf, 63 pp.)
- "The geographical, archeological, genetic, and linguistic origins of Tocharian" (7/14/20)
- "Of chains and Old Sinitic reconstructions" (1/27/21)
David Marjanović said,
March 16, 2025 @ 11:36 am
These two statements contradict each other if by "second oldest" you mean the usual tree hypothesis – (Anatolian (Tocharian (rest of IE))) – and if by "affinities" you mean that Tocharian is nested inside the "rest of IE" branch.
Prof. Mair, first of all, let me state once again that I have great respect for you. There are people who have one narrow specialty about which they "know more and more about less and less till they know everything about nothing". You are not such a person. Far from it! You have several specialties, all of them are quite large fields of knowledge and research, and as best I can tell you know the whole breadth and depth of each of them as well as anyone could.
This makes it all the more baffling when you make statements like the one I just quoted – that packs at least three fundamental misunderstandings into a single confidently delivered sentence.
First of all, there isn't necessarily any contradiction between archeology and genetics. Genetics tracks the biological relationships of people. Archeology tracks culture. Culture is often inherited along with genes, but by no means always! To take a concrete example from a DNA-centric paper, the spread of the Bell Beaker culture throughout western Europe north of the Pyrenees was accompanied by a movement of people from the northeast; the spread of the same – archeologically identical – culture south of the Pyrenees was not accompanied by any change in the ancestry percentages of the local population. "Pots, not people", as the saying goes: archeology tracks the pots, genetics tracks the people, and now that we can track both we can derive interesting conclusions from when the two move together and when they don't.
Second, phylogenetics from molecular data is anything but speculative. It gives you error bars – it tells you how large the 95% confidence interval on any particular hypothesis is. The size of each of these intervals may change if you add new data – it's science after all.
Third, calling it "statistical" like that's a bad thing – archeology works best when it's statistical, too: when you can tell that one pot in one grave isn't a fluke – say, a single person that traveled around while everyone else actually stayed put – but part of a well-supported pattern. That's when it, too, can and does give you 95% confidence intervals on specific hypotheses. In a Disney-franchise comic I once came across the following statement by a character in the story: "This sword is the proof that the Trojan War happened!" You and I both know this is not how archeology works.
Fourth, how is ancient DNA not "precisely measured"?
Fifth, what, if anything, is wrong with "chemical", and why?
Pots, protein and starch residues in the pots, bones, genes in the bones, stones, artifacts – everything together all at once. A "total-evidence approach".
Rodger C said,
March 16, 2025 @ 12:07 pm
What's the evidence that the name "Kushan" is derived from Chinese, and relatedly, what did that "Guishang" sound like 2000 years ago?
Jongseong Park said,
March 17, 2025 @ 12:22 am
Disappointingly, the video itself doesn't go much into the languages of the Kushan Empire, only briefly discussing the transition from Greek to Bactrian and not mentioning Gandhari Prakrit or Hybrid Sanskrit at all.
As posted on Language Log in 2023, the so-called Kushan script has recently been partially deciphered to reveal a Middle Iranian variety that was similar to but distinct from Bactrian. Here is a video that goes into the decipherment in detail. Although the decipherers seem to think the language of the script might have been the original language of the Kushans, it could well have been the language of the Saka people who came under their rule.
@Rodger C: I think the Met just got confused. "Kushan" is generally equated with 貴霜 as recorded in Chinese sources, which is likely the earliest attestation of that name, but I don't think anyone seriously claims that it actually derives from Chinese. The name is reconstructed as *kuj-s [s]raŋ in Old Chinese.
Lucas Christopoulos said,
March 17, 2025 @ 6:02 am
Euthydemos of Bactria (Greek: Εὐθύδημος, c. 260 BC-195 BC) used mainly Greek on his coins. His son Demetrios “The Invincible” (Greek: Δημήτριος Ἀνίκητος 222-180 BC), who conquered India, was the first to strike coins with a bilingual inscription in Greek and Prakrit, suggesting that he pursued a policy of treating the Indian peoples and the Bactrian people (including Greeks) as equals. Agathocles “The Just” (Greek: Ἀγαθοκλῆς Δικαῖος, c. 180 BC) incorporated the Brahmi script on his coins and several deities from India, which have been variously interpreted as Vishnu, Shiva, Vāsudeva, Balarama or the Buddha. The local Sogdians, Sakas, and other Bactrian tribes must have started to use Greek scripts at the time of Diodotes and Euthydemos; thus, the Kushana adopted the same Greek scripts as an official writing system.
According to Indian sources, the Kushanas are known as the Kambodjas. The Udyoga Parva records that Yavanas, Kambojas, and Sakas supported the Kauravas (Sanskrit: कौरव) at war under the leadership of the Kamboja (Yuezhi) King Saddakshina or Sadashkana (Saddakshina the son of Kujula Kadphises, is a Kushana).
Something interesting to observe about the Yuezhis, Heraios (23-30) and his son Kujula Kadphisès (Κοζουλου Καδφιζου, or Κοζολα Καδαφες 30-80), the bone plate from Orlat descriptions of warriors or the later statues of Maitreya from the Kushana kingdom, is a description of the Kushana people with a mustache, a Caucasian appearance but East-Asian like eyes, a bit like the northern Finns Sami people.
David Marjanović said,
March 17, 2025 @ 8:52 am
…so it probably sounded a lot like "Guishang" soon afterwards.
Interesting, if not surprising.
Robin Métral said,
March 20, 2025 @ 10:19 am
I don't have anything interesting to say about the Kushans, but Nota Bene is definitely not the biggest French YouTuber. They currently have 2.5 million subscribers, other popular French YouTubers on top of my head (@cyprien, @NormanFaitDesVideos) have over 10 million.
Victor Mair said,
March 20, 2025 @ 11:51 am
The o.p. begins:
"Jean Nota Bene, the biggest French YouTuber (millions of followers) on historical subjects…".