For half a century after Arthur Evans took the first clay tablets out of Knossos, the position with authority behind it was that Linear B was not Greek. Evans treated the script as an administrative revision of the Minoan writing he called Linear A, built to record the same non-Greek Aegean language. When the same script turned up on the mainland — Carl Blegen at Pylos in 1939, later Alan Wace at Mycenae — the explanation that saved the denial was Minoan rule or Minoan bookkeeping planted in Greece, not Greek speakers writing Greek. Other candidates were tried, including an Etruscan-related reading that Michael Ventris himself explored. The denial mattered because it kept the earliest Greek off the mainland palaces and out of the Knossos magazines. In 1952 Ventris broke it. John Chadwick, at Cambridge, was the first philologist to accept the break and to help test it. The tablets are Mycenaean Greek. They are not a script still waiting to be read, and they are not Linear A.
The method was public, which is why the old denial could not simply ignore it. Emmett L. Bennett’s classification of the signs, and Alice Kober’s work on inflection before her death in 1950, gave Ventris a grid instead of a guess. In the summer of 1952 he circulated the working note that asked, in effect, whether the Knossos and Pylos tablets were Greek. Within weeks he was in correspondence with Chadwick, newly appointed at Cambridge. Chadwick’s first letter did not applaud from a distance. It offered further suggestions that tightened the Greek. Their joint paper, “Evidence for Greek Dialect in the Mycenaean Archives,” appeared in the Journal of Hellenic Studies in 1953. It states Evans’s non-Greek premise and then sets out the opposite: the archives show a Greek dialect. The four years after that letter produced Documents in Mycenaean Greek, published by Cambridge in 1956, with a foreword by Wace. Ventris died that same year, weeks after the book, at thirty-four. Chadwick carried the philology on. The decipherment did not die with the architect who had forced the question.
The confirmation that outsiders remember is a tablet, not a manifesto. In 1952 Blegen’s season at Pylos produced new texts, among them the tripod tablet, later numbered Ta 641. Ideograms of tripod cauldrons sit beside syllabic groups that the Greek values already read as a word for tripod. A grid built without that tablet predicted the spelling the tablet then showed. Chadwick’s join on a Knossos horse tablet worked the same way: the Mycenaean spelling i-qo lines up with later Greek hippos, “horse,” next to horse ideograms. The Cambridge archive of the Ventris–Chadwick letters holds that collaboration in manuscript, including furniture lists, the horse join, and the postcard Chadwick wrote in Linear B to say the book manuscript had gone to press. Agreement of that kind is why “still unread” is a romance. The syllabary is a clumsy tool for Greek — clusters are awkward, some distinctions are missing — and the tablets are office records, not epics. Clumsy is not the same as sealed. Mycenaean Greek is read as an archaic dialect. It is not classical Athenian prose, and it does not have to be.