Some 2,000-year-old Roman scrolls are stubbornly hanging onto their ancient secrets, defying the best efforts of computer scientists at the University of Kentucky to unlock them.
The researchers have learned much about the scrolls, which were reduced to lumps of carbon in the heat of an eruption by Italy’s Mount Vesuvius in 79 A.D. But they can’t read what’s written on them.
“What we’ve found is that the problem is even more challenging than we thought going in,” said Brent Seales, Gill professor of engineering in UK’s computer science department and leader of the team working on the scrolls.
The UK team spent a month last summer making numerous X-ray scans of two of the scrolls that are stored at the French National Academy in Paris. They hoped that computer processing would convert the scans into digital images showing the interiors of the scrolls and revealing the ancient writing. The main fear, however, was that the Roman writers might have used carbon-based inks, which would be essentially invisible to the scans.
That fear has turned out to be fact.
“We hoped that we could look for calcium or other trace compounds in the ink that might help us tease out the writing,” Seales said. “But that hasn’t worked.”
Seales says he now hopes that re-scanning the scrolls with more powerful X-ray equipment will reveal the text, which scholars are anxious to read.
The effort is part of UK’s EDUCE project — Enhanced Digital Unwrapping for Conservation and Exploration — which has drawn international attention for using computer technology to peek inside fragile and faded books and manuscripts from antiquity, and produce exact digital copies for study. EDUCE, which Seales launched several years ago, is best known for producing stunning digital images of the oldest known copy of Homer’s Iliad, which is stored in Venice.
The Roman scrolls, however, have been a harder nut to crack.
Hundreds of the scrolls were stored in a Roman villa that was buried under tons of hot ash when Mount Vesuvius destroyed the Roman towns of Pompeii and Herculaneum in one of history’s most famous volcanic eruptions. The scrolls lay hidden for 1,600 years, until excavators stumbled upon them at Herculaneum in 1709.
What they found was a mystery. Volcanic heat had carbonized the scrolls — they resemble lumps of charcoal ready for a barbecue grill — which crumbled when anyone tried to unroll them. Scholars think the scrolls contain writings in Latin by the Roman philosopher Philodemus. But that’s only a guess until someone figures out how to read the scrolls without destroying them.
The UK team hoped to do that with computer magic last year.
Seales says that, in addition to the carbon-ink problem, the sheer volume of computer data produced from the X-ray scans overwhelmed UK’s interactive software. That slowed the system to the point that technicians were typing in commands and waiting half an hour or more for a response, he said.
“We’re not ready to say yet that we’re definitely not going to see the ink,” Seales said. “But we haven’t found a way yet to get at what we want.”
According to Seales, UK is looking at possibly rescanning the scrolls, in partnership with a group in Belgium that built the X-ray scanner used last year. A meeting with the group had to be canceled in April when the eruption of a volcano in Iceland interrupted flights to Europe.
“We’ve been talking with the engineers over there on how we could go back and scan the scrolls again, knowing what we know now, and do a better job of capturing the data we need,” Seales said. He has said that it ultimately might take the creation of new computer technology to unlock the scrolls.
“Of course, we want to be the ones to do that,” he said. “We’ve solved every other part of the problem. This is the missing link.”
UK’s computer imaging has confirmed that the rolled up papyrus scrolls are 30 to 40 feet long, which seems to suggest writing must be present. Why store a 40-foot scroll with no writing on it?
“The scholars are really excited by that,” Seales said. “If the scrolls are that large, think how much text there could be.”
Another item on the project mentions a couple more Homer manuscripts on the ‘scanning list':