In order to understand ancient scrolls that were carbonized by a volcanic eruption 2,000 years ago, scientists expect that a combination of artificial intelligence and human expertise will be useful.
In the 1750s, hundreds of papyrus scrolls were discovered in the ruins of an opulent villa in the Roman town of Herculaneum. Pompeii and Herculaneum were destroyed when Mount Vesuvius erupted in A.D. 79.
If the scrolls, which are rolled up into the size of a candy bar, could be read, the library of what is known as the Villa of the Papyri could significantly advance our understanding of ancient thought.
The town was destroyed by the heat and volcanic ash of Vesuvius, and the scrolls were saved but rendered illegible by the charred, fragile blocks that crumble when manually unrolled.
More than 250 years have been spent by researchers and academics trying to figure out how to translate the scrolls, most of which are kept in the National Library of Naples.
A number of tech leaders sponsored the “Vesuvius Challenge” competition in 2023, giving financial incentives for attempts to use geometry, computer vision, and machine learning to read the scrolls.
Researchers have created the first image of the interior of one of the three scrolls housed at Oxford University’s Bodleian Library, according to a “historic breakthrough” declared by the challenge on Wednesday.
Co-founder of the Vesuvius Challenge and computer scientist Brent Seales of the University of Kentucky stated the organizers were ecstatic that the scroll had been successfully imaged. According to him, it has more text that can be recovered than any Herculaneum scroll that has ever been scanned.
The scroll was scanned by Diamond Light Source, a lab located in Harwell, close to Oxford, that produces an extremely potent X-ray using a particle accelerator called a synchrotron.
The images were then put together by scientists using AI, which also improved the text’s clarity and looked for ink that indicated where writing was. Experts were able to visually unroll the scroll by utilizing a technique known as segmentation, which produced a 3D image of the scroll.
The current state of AI has limitations. So far, hardly much of the text has been decoded. The word for “disgust” in ancient Greek is one of the few that has been deciphered.
Academics are being urged to participate in the endeavor to finish the text.
The Associated Press was informed by Peter Toth, the Cornelia Starks Curator of Greek Collections at the Bodleian, that they are only beginning a lengthy process. They need better images, but they are optimistic and firmly believe that they can still make the text more readable and the images better.
In order to avoid having to ship the other two delicate scrolls to Diamond’s headquarters, Toth also expressed his desire that the technology may be made available locally.
He responded, “Perhaps there will be something that can be moved.” Not to mention that Naples has around a thousand more scrolls.
