The Kushan Empire was an ancient state that stretched from modern-day Uzbekistan into what is now northern India, and helped spread Buddhism to East Asia between the first and the third centuries. Starting in the 1950s, archeologists began identifying artifacts in the empire’s region bearing an “unknown Kushan script” that they couldn’t understand. But researchers in Germany and Tajikistan announced in July 2023 that they had finally deciphered part of the ancient writing system.

The researchers accomplished this by comparing a trilingual text that French archaeologists located in Afghanistan in the 1960s with a bilingual text that Tajik archaeologists identified in Tajikistan in 2022.

So-called bilingual or trilingual texts are parallel examples of writing that present approximately the same meaning, but in two to three different languages. Similar to the Rosetta Stone, these multi-script texts allowed researchers to identify characters and phrases in the ancient Kushan writing system.

Archaeologists first identified the unknown Kushan script in the 1950s, when they found its characters written in ink on a stairway at Surkh Kotal in Afghanistan. In the 1960s, French archaeologists found the script again on a boulder at Dašt-i Nāwur, also in Afghanistan. The text on the boulder contained the Kushan script alongside two others scripts representing the Bactrian language and the Gāndhārī language.

However, the trilingual text didn’t immediately lead to a breakthrough in understanding the unknown Kushan script. This was at least partly because French archaeologists took low-quality photos of the trilingual text and used an inaccurate drawing of it to discuss the finding, says Svenja Bonmann, a linguistics researcher at the University of Cologne in Germany, and lead author of the recent article about the unknown Kushan script published in Transactions of the Philological Society in July 2023.

Without clear, accurate images of the trilingual text, researchers couldn’t translate the unknown Kushan script. Over the next several decades, archaeologists located many more examples of the script, including an inscribed silver bowl in Kazakhstan, but its meaning continued to elude them. Then, in 2022, Tajik archaeologists identified a bilingual text at the Almosi Gorge in Tajikistan containing the unknown Kushan script alongside the Bactrian language.

“[The archaeologists] were following information given by a local man,” says Jakob Halfmann, a linguistics researcher at the University of Cologne and co-author of the recent article. “He had seen inscriptions on the side of the mountains close to where he lived, and he had been trying to tell the archaeologists about this for some time, but they only believed him and listened to him last year. And then they started an expedition and actually found that there really were inscriptions there.”

After locating the bilingual script, the Tajik archaeologist Bobomullo Bobomulloev worked with Bonmann, Halfmann and Natalie Korobzow, another linguistics researcher at the University of Cologne, to decipher the text (Bobomulloev and Korobzow are also co-authors on the recent article).

One of their key findings was that the name of the ruler Vema Takhtu appeared in the Bactrian portions of both the bilingual and the trilingual texts, allowing them to identify his name in both portions of the Kushan script.

This strategy is similar to the procedure that French philologist Jean-François Champollion used when translating the Rosetta Stone, which famously bears a passage written in Egyptian hieroglyphs, Egyptian Demotic script and Greek script. Champollion started by identifying royal names like Cleopatra in the Greek script and then finding them in the ancient Egyptian portions of the stone.

Most archaeological examples of the unknown Kushan script date from the second century B.C. to the third century, and were found in the modern-day countries of Kazakhstan, Uzbekistan, Tajikistan and Afghanistan. The script corresponded to an unknown Iranian language that was likely one of the official languages of the Kushan Empire, along with languages like Bactrian, Gāndhārī and Sanskrit.

One reason Bonmann and her colleague’s research is significant is that it provides further evidence of the existence of Vema Takhtu, who ruled the Kushan Empire during the first century and expanded its rule into northwest India. Previously, some scholars have debated whether he actually existed. The identification of his name in two different examples of Kushan script adds to the historical evidence that he was an important ruler of the Kushan Empire.

In addition to identifying Vema Takhtu’s name, the authors of the recent article also identified the phrases “king of kings” and “great savior”—both descriptors of Vema Takhtu. The authors estimate that they have deciphered about half of the Kushan characters, and hope to decipher more in the future. Further research may provide more information about the empire and the people who lived in it roughly two millennia ago.