In 1936, a fourteen-year-old schoolboy visited an exhibition at Burlington House in London marking fifty years of the British School at Athens, and listened to Sir Arthur Evans — the octogenarian excavator of Knossos — describe the clay tablets from Bronze Age Crete that nobody on earth could read. The story goes that the boy piped up to ask whether the tablets really remained undeciphered. Sixteen years later that same boy, by then an architect with no university training in ancient languages, announced on BBC radio that he had read them — and that, against all expert opinion, they were Greek.
Evans had begun digging at Knossos in 1900 and unearthed thousands of clay tablets, accidentally preserved when the fires that destroyed the palace baked them hard. He distinguished the island's scripts, naming the two linear ones Linear A and Linear B. Evans was certain the tablets recorded the language of his sophisticated pre-Greek 'Minoans', and emphatically not Greek, which he believed reached Crete only much later. He guarded the material jealously and published only a fraction of it before his death in 1941, so for half a century the decipherment went almost nowhere.
The unglamorous groundwork was laid by Alice Kober, a classics lecturer at Brooklyn College, who through the 1940s analysed the script's behaviour with monastic rigour, filing her statistics on tens of thousands of hand-cut paper slips stored in cigarette cartons — wartime shortages left her nothing better. Her famous 'triplets', sets of words identical but for their endings, proved that the hidden language was inflected, like Latin or Greek, and showed how sign groups could be anchored to one another. She died in 1950, aged just forty-three, largely unrecognised. The American scholar Emmett Bennett Jr, meanwhile, produced the definitive inventory of the script's roughly ninety syllabic signs — because no code can be cracked until you know exactly which symbols you are counting.
Michael Ventris, born in 1922, had absorbed languages effortlessly since childhood — schooled partly in Switzerland, he was said to pick up new tongues on holiday the way other travellers pick up souvenirs. Educated at Stowe and trained at the Architectural Association, he served as an RAF navigator during the war and practised as an architect afterwards; decipherment was the obsession he kept in the margins of his life. He circulated numbered 'Work Notes' to scholars around the world and built grids arranging the signs by shared consonants and vowels, long before he could pronounce a single one. His own theory, held almost to the very end, was that the tablets concealed a relative of Etruscan. He was wrong — and his greatness lay in noticing that he was wrong.
The breakthrough
In 1952, testing his grid, Ventris tried something he had long resisted: reading certain recurring sign groups as the names of Cretan places. Out came ko-no-so and a-mi-ni-so — Knossos and Amnisos, its harbour town. With those values fixed, Greek words began cascading from the tablets, in strange archaic dress but unmistakable. In July 1952 he announced on the BBC Third Programme that the language of Linear B was Greek. John Chadwick, a Cambridge philologist who had spent the war as a codebreaker, heard the broadcast, checked the results and joined him; their 1953 paper, 'Evidence for Greek Dialect in the Mycenaean Archives', made the scholarly case.
Confirmation arrived almost at once, from evidence Ventris could not have rigged. The American archaeologist Carl Blegen applied the proposed sound values to a newly excavated tablet from Pylos, on the Greek mainland, and it read ti-ri-po-de — tripods — directly alongside little pictograms of three-legged cauldrons. The tablets turned out to contain no poetry, no myth and no history: they are palace bureaucracy — rations, land records, chariot wheels, rowers, and sheep in their tens of thousands. Yet names later sung by Homer appear in the ledgers, Poseidon prominent among the gods, and the decipherment pushed written Greek back to roughly 1400–1200 BC, about five centuries before the Greek alphabet existed. At a stroke, Greek became one of the oldest continuously recorded living languages on earth.
Ventris was appointed OBE, and with Chadwick completed Documents in Mycenaean Greek, the foundation stone of a whole new discipline. He never saw it in print. On 6 September 1956, driving at night, he collided with a lorry near Hatfield and was killed instantly, aged thirty-four, weeks before the book appeared. Linear A, the older Cretan script, remains undeciphered to this day — waiting, perhaps, for another outsider who cannot leave a locked door alone.
Quiz nuggets
- Arthur Evans excavated Knossos from 1900, found the clay tablets and coined the names Linear A and Linear B.
- Alice Kober's 'triplets' proved the hidden language was inflected; she died in 1950, aged forty-three, before the decipherment.
- Michael Ventris, an architect by profession, announced on BBC radio in July 1952 that Linear B was Greek.
- Carl Blegen's 'tripod tablet' from Pylos confirmed the decipherment: the word for tripods sat beside pictures of tripods.
- Ventris died in a car crash near Hatfield on 6 September 1956, aged thirty-four; Linear A remains undeciphered.