“The problem is that it isn’t based on a great deal of training data. That’s not trivial because it means that the machine doesn’t just see gobbledygook,” says Camilla Di Biase-Dyson, a lecturer in Egyptology at Macquarie University who was involved in the project. “What the machine does well is to recognize where there are hieroglyphs and where there aren’t any. Researchers at Macquarie University in Australia teamed up with experts from Google to use artificial intelligence with the aim of speeding up the process of translating ancient Egyptian hieroglyphs into English and Arabic. There are efforts to make translating ancient languages a more modern pursuit. “This is the only famous example of success by using this method,” says Enmarch. From this deduction, they were able to work backwards to translate the whole text. This led the decoders to suppose that the sequences were therefore place names on the island, which turned out to be true. Certain sequences of symbols only appeared on tablets that were found on the island of Crete, but not on those discovered on the Greek mainland. They looked at patterns of variation within the script to see if that could offer clues.
However, decoders who had worked on cracking the German Enigma code during World War II were able to decipher linear B by assuming that it was in fact an ancestor of Greek.
The language is known as “linear B” and it predates the Greek alphabet and so most historians agreed it was probably a separate language altogether. that were dug up in Greece and for decades people tried to work out what the hell they said,” says Enmarch. “There are some clay tablets from about 1450-1200 B.C. The other way to resurrect a long dead language is through basic data processing. Sadly, the resulting translation of the Rosetta stone didn’t reveal anything particularly earth-shattering - it’s just a piece of administrative text to mark the anniversary of a king’s jubilee, but it did provide the means to understand other, more interesting texts. Using this technique, you can see where sections reoccur and translate those bits. “If you have a bilingual script and one of them is in a language that you understand then that’s really helpful as long as the content of the text repeats itself,” says Enmarch. Because ancient Greek was understood, this provided a route to decode ancient Egyptian. The first was written in hieroglyphs and the second in the demotic script, a cursive form of ancient Egyptian similar in style to written Arabic. The Rosetta Stone, a carving of a proclamation issued in 196 B.C., proved so useful in decoding Egyptian hieroglyphs because the decree was repeated three times over. “It’s the single most famous translational artifact.” “Next year marks the bicentenary of the Rosetta decipherment, which really was a watershed moment for Egyptology,” says Roland Enmarch, senior lecturer in Egyptology at the University of Liverpool in the United Kingdom. The fact that historians can now read and understand hieroglyphic inscriptions is down to an act of archaeological prowess involving a fairly banal, but ancient legal text chiseled onto a world-famous stone.
Until 200 years ago, no one in the modern era could understand Egyptian hieroglyphs ancient Egyptian was essentially a lost language.