Ancient stone inscriptions offer some of the most powerful primary sources available to historians. However, as these stones are often many thousands of years old, and thus eroded, chipped or otherwise compromised, they can also be immensely difficult to translate, as well as to place in time and space. With this important challenge in mind, Thea Sommerschield, a historian, and Yannis Assael, an artificial intelligence research scientist, teamed up to create Ithaca – a deep neural network that helps to date, place and translate ancient text. Working from a database of 80,000 ancient Greek inscriptions, the neural network combines its sophisticated understanding of patterns to predict missing words and offer guesses as to when and where a text was written. Worth noting too that, while the program fills in the blanks much better than historians working independently of the technology, it was at its most accurate when collaborating with a human expert.
Video by Nature
Producer: Shamini Bundell
videoHistory
In Stalin’s home city in Georgia, generations clash over his legacy
20 minutes
videoArchitecture
Steep climbs lead to sacred spaces carved high into the cliffs of Ethiopia
9 minutes
videoHistory
In the face of denial, this film uncovers the hidden scars of Indonesia’s 1998 riots
21 minutes
videoArchaeology
At a prehistoric pigment mine, researchers glimpse our earliest moments in the Americas
25 minutes
videoComputing and artificial intelligence
A future in which ‘artificial scientists’ make discoveries may not be far away
9 minutes
videoHistory
Hags, seductresses, feminist icons – how gender dynamics manifest in witches
13 minutes
videoComputing and artificial intelligence
Why large language models are mysterious – even to their creators
8 minutes
videoFairness and equality
‘To my old master’ – a freed slave answers the request to return to his old plantation
7 minutes
videoInformation and communication
‘Astonished and somewhat terrified’ – Victorians’ reactions to the phonograph
36 minutes