An AI engineer working outside academia says he has deciphered Linear A, a Bronze Age writing system from Crete. Tom Di Mino made the claim in June, and his research is now being examined by linguistics scholars at Rutgers University and the University of Cambridge. Linear A has not been read by professional linguists in more than 100 years, despite decades of study.
Di Mino says the discovery centers on a prayer text repeated at five sanctuary sites on Crete. Most of the words in the prayer were already understood, but one verb had no known meaning.
Di Mino traced that verb to a root shared with Hebrew and Akkadian, tied to the idea of living in or settling a place. Once translated, he says the passage reads much like a later Hebrew prayer, but directed at a goddess rather than a god.
A researcher named Cyrus Gordon proposed a similar Semitic link for Linear A in 1957, but few linguists accepted it. Di Mino says his work goes further by connecting specific inscriptions directly to the structure of Hebrew prayer.
How an AI engineer put the Linear A theory to the test
To test his theory, Di Mino, an AI engineer, wrote a set of Python scripts using a coding tool called Claude Code. The scripts checked his proposed word against two digital collections of Linear A writing, known as GORILA and SigLA, covering nearly every surviving inscription.
He says the results include proposed sounds for 40 previously unclear signs, a list of 408 translated Linear A words, and a nine-page draft describing the grammar of the prayer text.
Linear A was used from about 1800 BC until 1450 BC, when Mycenaean Greeks took control of Crete and adapted the script into a related writing system called Linear B. Linear B was read in 1952 and identified as an early form of Greek.
Linear A has remained unread partly because it has never been tied with confidence to any other known language. Its full record amounts to about 7,500 characters, a small sample that leaves room for competing readings.
Linguists say Di Mino’s claim will need to pass peer review before it can be treated as a confirmed decipherment.
Self-taught AI engineer behind the Linear A decoding
Di Mino has no formal training in linguistics. He lives in New York’s Hudson Valley and has taught himself classical history and ancient languages since he was 18, gaining some ability in eight of them, including Attic Greek, Latin, Sanskrit, Arabic and Ugaritic.
He first became interested in Linear A seven years ago and has traveled to Crete twice. His concentrated work on the script began in January, and he says the breakthrough came on May 22.
See all the latest news from Greece and the world at Greekreporter.com. Contact our newsroom to report an update or send your story, photos and videos. Follow GR on Google News and subscribe here to our daily email!
