A new AI tool can now identify the distinct handwriting of individual Hittite scribes on ancient clay tablets, giving researchers a major advantage in piecing together one of history’s earliest written records.
Scientists at the University of Würzburg and the Academy of Sciences and Literature Mainz developed “Palaeographicum,” which scans digitized photographs of clay tablets and detects individual variations in cuneiform signs.
The Hittites lived in modern-day Anatolia around 3,500 years ago and recorded their civilization using 375 cuneiform signs pressed into wet clay with a stylus. Thousands of those tablets broke apart over time, and fragments are now scattered across museums worldwide.
Reconstructing the original documents requires matching those fragments, a task that demands close analysis of handwriting patterns. Every scribe pressed the same wedge-shaped signs into clay, but each writer developed a personal style. Some pulled the stylus with a flourish, creating loops. Others placed signs with distinct spacing.
AI tool exposes the handwriting secrets of Hittite scribes
Daniel Schwemer, who leads the Chair of Ancient Near Eastern Studies at the University of Würzburg, said the tool fundamentally changes how researchers work. A comparison of handwriting across five tablet fragments once took three days.
The same work now takes five minutes. Schwemer said the tool is expected to save thousands of work hours overall.
The current version of Palaeographicum provides access to 70,000 photographs containing more than five million cuneiform signs. It searches the entire catalog for identical or similar signs, then extracts them from photographs and arranges them in visual tables for easy comparison.
Because Hittite tablets never carry dates, researchers have long struggled to place individual fragments in time. Palaeographicum also helps solve that problem by tracking how handwriting changed across centuries, which is how the tool got its name.
From a five-year research project to a global tool
Gerfrid Müller, a professor of ancient Near Eastern studies, said the team continues training the AI and plans to address user feedback, provided those requests are technically practical and benefit the broader research community.
The tool grew out of the DFG-funded CuKa project, which ran between 2018 and 2023 in cooperation with TU Dortmund. After the project ended, Müller, Christopher Rest, and Herbert Baier Saip from the University of Würzburg improved the system to handle the large volume of available photographs.
Looking ahead, researchers aim to train the AI to automatically identify individual scribes across different tablets. If achieved, Schwemer said, it could help build a detailed social history of the Hittite scribal profession, tracing what individual scribes produced throughout their careers.
See all the latest news from Greece and the world at Greekreporter.com. Contact our newsroom to report an update or send your story, photos and videos. Follow GR on Google News and subscribe here to our daily email!
