From deciphering burnt Roman scrolls to reading crumbling cuneiform tablets, neural networks could give researchers more data than they’ve had in centuries.
(30 Dec 2024) In October 2023, an e-mail pinged onto Federica Nicolardi’s phone with an image that would transform her research forever. It showed a fragment of a papyrus scroll that had been burnt in the eruption of Mount Vesuvius in ad 79. The scorched scroll was one of hundreds discovered in the remains of a luxury Roman villa in Herculeaneum, near Pompeii in Italy, in the eighteenth century. Attempts over the centuries to peel apart the scrolls’ fragile, carbonized layers left many in pieces, and scholars have been forced to accept that the rest can never be opened.
Nicolardi, a papyrologist at the University of Naples in Italy, had been enlisted in an effort to use artificial intelligence (AI) to read the unreadable. Now the latest results had arrived. The image showed a strip of papyrus packed with neat Greek lettering, glowing bright against a darker background. The writing was clearly legible, a few lines deep and stretched across nearly five columns.
“It was incredible,” says Nicolardi. “I thought, ‘So this is really happening.’” She knew right then that papyrology would never be the same. “In that moment, you really think ‘now I’m living something that will be a historical moment for my field.’” She was reading entire lines of a text that had been utterly inaccessible for 2,000 years.
That project, called the Vesuvius Challenge, is just one example of how sophisticated AI, which is already revolutionizing all areas of modern life, from banking to medical research, is poised to reshape how we see the ancient world. Artificial neural networks are being used to decipher ancient texts, from the classical stalwarts of Greek and Latin to China’s Oracle Bone Script, ancient divination texts written on cattle bones and turtle shells. They are making sense of archives too vast for humans to read, filling in missing and unreadable characters and decoding rare and lost languages of which hardly any traces survive.
Nature has the article in full.
