The Tukarian language remains one of history's most fascinating puzzles, discovered in western China centuries after it fell silent.
This analysis examines the linguistic remnants of the Tukarian language, a unique tongue that challenges our understanding of migration and cultural exchange. By comparing fragments of script and structural patterns, we see how this isolated culture fits into the broader history of Indo-European languages.
We move beyond the surface to apply ancient linguistics to the specific charts that map its characteristics against its neighbors. Understanding these extinct languages allows us to trace the complex evolution of speech across Eurasia and reconstruct the social landscape of a civilization that eventually faded into silence.
Subscribe for weekly historical breakdowns, and let us know in the comments which ancient language you want us to map next.
AI Generated·May contain errors
Video Summary
AI Generated
This video examines the mystery of Tocharian, an Indo-European language discovered in China's Tarim Basin. It explores the linguistic puzzle of its \"centum\" classification, the debate between genetic evidence and migration theories involving the Afanasievo culture, and the language's eventual disappearance following the rise of the Uyghur Empire.
Peser, meser, proser, father, mother, brother. If you know any Latin, you've already heard it.
Pater, mater, frater. But these three words don't come from Rome, or from Greece,
or from anywhere in Europe. They were written down more than a thousand years ago,
in a Buddhist monastery on the edge of a desert in western China. The language is called Turkarian,
and it's one of the strangest puzzles in the history of language.
It belongs to the Indo-European family, the same family as English, Latin, Greek and Sanskrit.
Yet it was spoken in the oasis cities of the Tarim Basin, on the Silk Road, thousands of kilometers
east of its closest relatives. So, here's what we know for sure. Turkarian was a real,
living language, with two main forms, spoken in what is now Xinjiang in China until around
a thousand years ago. What we don't know is how it got there. And the latest evidence,
from ancient DNA, has made that question even harder. In this video, we'll hear Turkarian words
side by side with Latin and Sanskrit. We'll see why its discovery overturned what linguists
thought they knew. We'll meet the people who wrote it, including an unknown love poet,
and follow the trail back to the mountains of Siberia. And stay to the end for the twist,
the famous mummies of the Tarim Basin and what their DNA revealed in 2021.
This is Language Atlas.
In the far west of China lies one of the driest places on Earth, the Taklamakan Desert. It fills
almost the whole of the Tarim Basin, a vast bowl ringed by some of the highest mountains in the
world. Rain almost never falls here. But around the edges of the desert, where rivers run down
from the snowy peaks, there are green oases. And for more than a thousand years, those oases were
stops on the Silk Road. Towns like Kucha, Karachar and Turfan grew rich on the caravan trade between
China, India, Persia and the West. And with the merchants came Buddhism. Monks built monasteries
and carved cave temples into the cliffs, filled with colourful wall paintings and libraries of
sacred texts. Near Kucha, at a place called Kizil, hundreds of caves were cut into a cliff above a
river valley. Their walls were covered in paintings of the Buddha, of musicians and dancers, and of the
local people who paid for them, in deep blues and greens. For centuries, these caves were one of the
great centres of Buddhist art in Central Asia. Then the caravan stopped, the towns emptied and
the sand moved in. At the start of the 20th century, explorers from Europe, Russia and Japan began to
arrive. They dug through the ruins and the desert gave up something extraordinary. The dry air had
preserved tens of thousands of manuscript fragments, on paper, on palm leaves, on wooden tablets. Some
expeditions took far more than manuscripts. Whole wall paintings were cut from the cave walls and
shipped back to museums in Europe. Many of those taken to Berlin were destroyed in the bombing of
the Second World War. But it was the manuscripts that held the real puzzle. Most of them were in
languages scholars already knew. Sanskrit, Chinese, Old Turkic. But some were different. They were
written in Brahmi, a well-known script from India. So, scholars could sound out every single letter.
They just had no idea what the words meant. Imagine being able to read every word on a page
and understand none of it. The breakthrough came from the Buddhist texts. Some manuscripts were
bilingual, with Sanskrit next to the unknown language. Scholars could line them up, word by
word, like a codebook. And as they did, they began to notice something they did not expect at all.
In 1907 and 1908, German scholars working on the manuscripts, among them Emile Sieg, announced
their conclusion. This unknown language of the Chinese desert was not related to Chinese. It was
not Turkic. It was Indo-European. They named it Tukarian. As we'll see, that name was a mistake.
But it stuck. So, how Indo-European is Tukarian? Let's put it to the test. If a Roman could somehow
travel to a monastery in Kucha in the 7th century, would he recognize anything? Let's start with the
family. Father. In Tukarian, pesa. In Latin, pater. In Sanskrit, pita. Mother. Tukarian,
macer. Latin, mater. Sanskrit, mata. Brother. Tukarian, procer. Latin, frater. Sanskrit,
brata. And in English, brother. You can hear the pattern. The consonants shift,
the vowels move, but the skeleton of each word is the same. That's exactly what you'd expect from
languages that share a common ancestor, spoken thousands of years earlier. Now let's try some
numbers. Two. In Tukarian, we. In Latin, duo. Not so obvious. Tukarian changed a lot over the
centuries, and some words wandered a long way from their cousins. Eight. Tukarian, oct. Latin,
octo. That one is almost identical. You can still hear it in English words like octopus and
October. And now my favorite. Horse. In Latin, equus. Which gives us the English word equine.
In Tukarian, yaquay. Say them together. Equus. Yaquay. Once you hear it, it's hard to unhear it.
Both come from the same ancient word for horse, reconstructed as something like equos,
the word used by the first speakers of the Indo-European family more than 5,000 years ago.
So would our Roman traveler understand Tukarian? No. It's far too different. He couldn't follow
a sentence, let alone a sermon. But he would keep catching words that felt strangely familiar,
like hearing a distant cousin speak with an accent he'd never heard before.
And that's the heart of the mystery. These family words tell us, without any doubt,
where Tukarian belongs. The problem is where it was found. Because one word in particular
should never have turned up in China. That word is hundred.
In the 19th century, linguists noticed something neat about the Indo-European languages.
They seemed to fall into two groups, and you could tell them apart with a single word,
hundred. In one group, hundred starts with a hard K sound. Latin says centum, which the Romans
pronounced kentum. Ancient Greek, the Celtic languages, and the Germanic languages all belong
here too. Linguists called them the centum languages. In the other group, that same sound
turned into an S or a sh sound. In Avestan, the ancient language of Iran, hundred is satum. In
Sanskrit, it's shata. The Slavic and Baltic languages follow the same pattern. These are
satum languages. And here's the neat part. On a map, the split looked geographical. The centum
languages were in the west, Italy, Greece, Gaul, and Germany. The satum languages were in the east,
Russia, Iran, and India. It looked as if the Indo-European family had divided into a western
half and an eastern half. Then Tukarian was found, the easternmost Indo-European language
ever discovered, thousands of kilometers beyond India and Iran. And its word for hundred? Kante,
a hard K, a centum language, in the far east. It gets stranger. Tukarian's closest Indo-European
neighbors were the Sakha of Khotan, just across the desert on the southern edge of the basin.
They spoke an Iranian language, a satum language. So Tukarian wasn't just far from its relatives,
it wasn't even closely related to the people next door. The discovery broke the simple picture
of west versus east. Today, most linguists think the centum-satum split isn't a clean
family division at all. Instead, the satum change probably spread through a group of
languages that lived close together, while languages that had already left, like Tukarian,
kept the old K. In other words, Tukarian's ancestors must have left the Indo-European
homeland very early, before the satum change swept through their eastern neighbors. It's as
if you found a village in Mongolia where people spoke something closer to Welsh than to anything
around them. As scholars worked through the manuscripts, they found that Tukarian wasn't
one language, it was two. Today, we call them Tukarian A and Tukarian B. Tukarian A comes from
the eastern oases, around Turfan and Karachar. Tukarian B was used across the whole region,
but its heartland was Kucha in the west. They're related, but different enough that most linguists
treat them as separate languages, roughly as different as Spanish and Italian, or perhaps
more so. Listen to the word for horse again. In Tukarian B, yakwe, in Tukarian A, yuk, and they
seem to have been used in different ways. Almost everything we have in Tukarian A is religious,
Buddhist texts, hymns, and stories. It may already have been a church language,
the way Latin was in medieval Europe. Tukarian B was used for everything. Religion, yes, but also
letters, monastery accounts, medical recipes, and travel documents. Now, about that name.
When the language was first identified, scholars borrowed the name Tukarian from the Tokaroi,
a people mentioned by ancient Greek writers. The trouble is that the real Tokaroi lived in Bactria,
what is now Afghanistan, and spoke an Iranian language. They almost certainly had nothing to
do with this language at all. The name is a mistake. But after more than a century,
it has stuck. And what the speakers of Tukarian actually called themselves,
we still don't know for certain. What we do have are glimpses of their lives. Monastery accounts
listing grain and payments. Travel passes for caravans checked at posts along the road,
listing the people and animals allowed through. And one tiny fragment of a love poem in Tukarian
B. In one translation, a line reads, there will be no one dearer to me than thee. We don't know
who wrote it or to whom, but more than a thousand years later, it still sounds like something a
person would say. And one son of Kucha became famous far beyond the Tarim Basin. In the fourth
century, a monk named Kumarajiva was born there. He became one of the most important translators
of Buddhist texts into Chinese in history, and his translations are still read today.
So how did an Indo-European language end up on the edge of a Chinese desert?
To answer that, we have to go back to the very beginning of the Indo-European family.
Most linguists and archaeologists now think the ancestor of all these languages,
Proto-Indo-European, was spoken on the grasslands north of the Black Sea and the Caspian Sea
more than 5,000 years ago. If you want the full story of that lost language,
we've told it in our video on the Lost Language of the Great Steppe.
From that homeland, groups of herders spread out in every direction, taking their language
with them. Some went west into Europe, some went south into Anatolia, others headed east,
eventually reaching Iran and India. But one of the very earliest groups to leave
headed somewhere else entirely. Around 3300 BC, a culture appears in
the Altai Mountains of southern Siberia, far to the east of the steppe homeland.
Archaeologists call it the Afanasievo culture. Its people buried their dead in the same way as
the steppe herders in the west, kept the same animals, and made similar pottery.
Genetically, they were close relatives of those western herders.
It's as if a piece of the western steppe had been picked up and set down 2,000 kilometers to the
east. For many scholars, Afanasievo is the most likely starting point for Tocharian.
It fits the linguistic evidence well. Tocharians split away from the rest of the family very early,
before the Satum change, and Afanasievo is one of the earliest known offshoots of the steppe cultures.
From the Altai, the theory goes, their descendants moved slowly south over many
centuries until they reached the oases of the Tarim Basin.
If you're enjoying this journey, subscribe to Language Atlas. Every video in this series
traces a language across the map, just like this one. There may also be clues in Chinese history.
Chinese records describe peoples living in the lands to the northwest of China,
including a group called the Yuezhi. Some scholars think the Yuezhi were linked to
the Tocharian speakers. Others disagree. It's one of the many open questions in this story.
So, the theory is a long road east, from the steppe to the Altai,
and from the Altai to the Tarim Basin. A journey of thousands of kilometers,
over thousands of years. But a theory like this makes a prediction. If it's right,
the ancient people buried in the Tarim Basin should carry the ancestry of those Afanasievo herders.
And the Tarim Basin, as it happens, is home to some of the best-preserved ancient bodies in the
world. In the salty, bone-dry sands of the Tarim Basin, archaeologists have found hundreds of
naturally preserved bodies. Some are around 4,000 years old. They were buried in boat-shaped wooden
coffins, wrapped in woolen cloaks and felt hats, with baskets, feathers, and small bundles
of plants beside them. When the Tarim mummies became famous in the 1980s and 90s, many people
connected them straight away with the Tocharians. The mummies were ancient, they were in exactly the
right place, and some had features that newspapers of the time loosely described as European.
The story seemed to write itself. Here, at last, were the ancestors of the Tocharian speakers.
In 2021, a team of geneticists put that story to the test. They analyzed the DNA of 13 of the
oldest mummies, from around 4,000 years ago, and published the results in the journal Nature.
The answer surprised almost everyone. The mummies had no ancestry at all from the Afanasievo herders.
None. Instead, they descended from a very ancient local population, a remnant of a group that
geneticists call the Ancient North Eurasians. They were genetically isolated. And yet culturally,
they weren't isolated at all. They grew wheat, kept cattle, sheep and goats, and made dairy products,
ideas that had traveled from far away. So what does that mean for Tocharian?
It's important to be careful here. The study looked at the earliest mummies, from about 2,000 years
before the first Tocharian texts. It doesn't tell us what language those people spoke,
and it doesn't rule out Tocharian speakers arriving later. But it does remind us of
something linguists have always known. Genes, appearance, and language are three different
things. People can adopt a new language without any change in their ancestry. And ideas, like
farming or words, can travel along trade routes without large migrations of people. The one piece
evidence everyone expected to solve the mystery ended up making it deeper. The Afanasievo theory
is still the leading explanation for how Tocharian reached China. But the full story is still being
written. However Tocharian arrived, it survived in the Tarim Basin for a very long time.
So what finally happened to it? In the year 840, far to the north on the Mongolian steppe,
the Great Uyghur Empire collapsed. Its people, who spoke a Turkic language, moved south and west,
and many settled in the oasis towns of the Tarim Basin, including Turfan. They founded new kingdoms
and became the new ruling class. Over the following generations, the people of the oases
began to speak the language of their rulers. Tocharian didn't vanish overnight. For a while,
the two languages lived side by side. Some Buddhist texts were even translated from Tocharian
into Old Uyghur. But slowly, parents stopped passing Tocharian on to their children. Within
a few centuries, it was gone. The language that had survived an almost unimaginable journey across
Asia disappeared within a few generations of its speakers switching to another tongue.
It survived only in the fragments left behind in ruined monasteries, waiting under the sand for
a thousand years. It may have left one small trace behind though. Some scholars think the
Chinese word for honey, 蜜, was borrowed long ago from Tocharian. If they're right, every time
someone in China talks about honey, they may be using a word from a lost Indo-European language.
Three words from a desert at the far end of the Silk Road that still sound like home to anyone
who speaks a European language. So why was an Indo-European language spoken in ancient China?
Because, more than 5,000 years ago, one of the very first groups to break away from the
Indo-European family headed east instead of west. Their language survived for thousands of years in
the oases of the Tarim Basin, until its speakers switched to Turkic and it faded into the sand.
Exactly how it got there is still one of the great open questions of history.
So what do you think? Did Tocharian arrive with migrating herders,
or was it passed from people to people along the trade routes? Tell us in the comments.
And if you want to see where this whole family began,
watch our video on the lost language of the Great Steppe next. Thanks for watching Language Atlas.