Scivillage.com Casual Discussion Science Forum
Article 2,000 year old library buried by Pompeii’s eruption is finally revealing its secrets - Printable Version

+- Scivillage.com Casual Discussion Science Forum (https://www.scivillage.com)
+-- Forum: Culture (https://www.scivillage.com/forum-49.html)
+--- Forum: History (https://www.scivillage.com/forum-117.html)
+--- Thread: Article 2,000 year old library buried by Pompeii’s eruption is finally revealing its secrets (/thread-21719.html)



2,000 year old library buried by Pompeii’s eruption is finally revealing its secrets - C C - Oct 30, 2026

https://www.sciencefocus.com/future-technology/how-scientists-could-soon-unravel-pompeiis-lost-secrets

EXCERPTS: Nowhere is Herculaneum’s affluence more apparent than in the Villa of the Papyri, an opulent coastal estate that would not be out of place among the modern mansions of Malibu. In 1750, inside this remarkable estate, archaeologists unearthed the largest collection of statues ever recovered from a single classical building. But something far more valuable was hidden away: a vast library of ancient scrolls.

The excavators discovered more than a thousand papyrus scrolls, the only surviving library from the classical world – though, initially, they mistook the scrolls for logs as they were carbonised by the searing heat of Vesuvius’s eruption. Much of our knowledge of the ancient world comes not from direct sources but from copies passed down through centuries of transcription, such as Homer’s The Odyssey and Virgil’s The Aeneid.

To find even one original ancient text is extraordinary; to find hundreds is almost beyond comprehension. But there was one crucial problem: how to read them. The Herculaneum scrolls resembled lumps of charcoal rather than literary masterpieces. Any attempt to physically unravel them destroyed them.

Over the centuries, various methods were tried – some yielding faint glimpses of text – but always at the cost of further damage. From what little could be deciphered, scholars determined that many of the texts belonged to the philosopher Philodemus of Gadara.

[...] The first step involves capturing ultra-high-resolution scans of the scrolls using a particle accelerator. These work in much the same way as a CT scan in a hospital, only the details seen are thousands of times finer. This process, known as X-ray tomography, generates X-ray images from different angles, constructing a detailed three-dimensional model of the scroll’s interior.

[...] This is where AI comes into play. The carbonised papyrus and ink are virtually indistinguishable to the human eye – but not to an algorithm trained to detect microscopic differences in surface texture. Machine learning models are fed existing examples of inked regions. Then, as they’re given more and more examples, the models identify the parts of papyrus where the shape is distorted due to the ink, revealing letters invisible to the naked eye.

Once the ink is detected, human experts step in. Seales says that in the future the model might be able to help with word recognition, but as it stands AI will ‘hallucinate’ and make up meaning where there isn’t any. So, currently, the models simply identify where the ink is, not what it says. It’s up to papyrologists and classicists to interpret the meaning.

[...] “I was very sceptical whether this would ever work out,” says Dr Kilian Fleischer, a papyrologist at the University of Tübingen, in Germany, and board member of the Vesuvius Challenge. “But in late 2023, we had proof of concept.”

[...] “We think we’re going to read them all,” Seales says. “This could be the biggest cache of material from the ancient world in human history. And it’s all because of technology.” What’s more, a tantalising question remains to be answered: could more scrolls still be hidden within the buried villa?

The texts revealed so far seem to consist mostly of Epicurean philosophy (a system of thought based on the teachings of Greek philosopher Epicurus, who promoted happiness via moderation and tranquillity), suggesting that the library was a technical collection rather than a general one. Some speculate that a much larger library – the real library – lies yet undiscovered... (MORE - missing details)