Kathrine Frich
Centuries-old debates about who wrote the Bible have taken a new turn with the help of modern technology.
Researchers say artificial intelligence has revealed patterns long invisible to human readers.
The findings suggest the origins of some of the Bible’s earliest books may be more complex than previously thought.
Patterns in scripture
An international research team led by Duke University used artificial intelligence to analyze the first nine books of the Hebrew Bible, known as the Enneateuch.
By examining word usage and sentence structure, the AI identified three distinct writing styles within the text.
According to the researchers, these differences point to separate authors or scribal traditions rather than a single unified voice.
The team said even common words such as “no,” “which,” or “king” varied consistently between styles.
“We found that each group of authors has a different style, surprisingly, even regarding simple and common words,” said Thomas Römer, a professor at the Collège de France.
How the model works
The project was spearheaded by mathematician Shira Faigenbaum-Golovin, who began developing the method in 2010 while studying handwriting on ancient pottery shards dating back to around 600 BC.
She later assembled a team of mathematicians, archaeologists, linguists, and computer scientists to apply similar statistical techniques to biblical texts.
The AI model was designed to identify linguistic fingerprints rather than theological themes.
“Our method accurately identifies these differences,” Römer said, describing the results as statistically robust.
Unexpected gaps
The analysis linked most passages to one of three traditions: the Priestly source, the Deuteronomistic History, and the Book of Deuteronomy itself.
But parts of the Ark Narrative in 1 Samuel did not match any of the identified styles.
Researchers said this anomaly could indicate the presence of another, as yet unidentified, author or editorial layer.
They described the finding as an open question rather than a definitive conclusion.
Beyond the Bible
The team believes the technique could be used far beyond biblical studies. Faigenbaum-Golovin said the same approach could help authenticate historical documents or identify forgeries.
“If you’re looking at document fragments to find out if they were written by Abraham Lincoln, for example, this method can help determine if they are real or just a forgery,” she said.
The researchers concluded that their work offers “interpretable, statistically significant evidence” that authorship patterns in ancient texts can be detected automatically.
Scholars involved in the study say the project is an example of the growing collaboration between science and the humanities. T
hey are now exploring how the method could be applied to other ancient manuscripts.
“This is a unique collaboration,” Faigenbaum-Golovin said. “It’s a surprising symbiosis.”
Sources: Duke University, Collège de France, study authors, Express.
