SoylentNews

upstart [soylentnews.org] writes:

████ # This file was generated bot-o-matically! Edit at your own risk. ████

Are Dead Sea Scrolls older than we thought? [arstechnica.com]:

Text settings

Over the years, scholars of the Dead Sea Scrolls have analyzed the ancient parchments with various methods: for example, X-rays, multispectral imaging, "virtual unfolding," and paleography [wikipedia.org], i.e., studying elements in their writing styles. The scrolls are believed to date back to between the third century BCE and the first century CE, but those dates rely largely on paleography, since only a handful of the scrolls have calendar dates written on them.

However, the traditional paleographic method is inherently subjective and based on a given scholar's experience. A team of scientists has combined radiocarbon dating from 24 scroll samples and machine-learning-based handwriting analysis to create their own AI program—dubbed Enoch. The objective was to achieve more accurate date estimates, according to a new paper [plos.io] published in the journal PLoS ONE. Among the findings: Many of the scrolls are older than previously thought.

As reported earlier [arstechnica.com], these ancient Hebrew texts—roughly 900 full and partial scrolls [eurekalert.org] in all, stored in clay jars—were first discovered scattered in various caves near what was once the settlement of Qumran [wikipedia.org], just north of the Dead Sea, by Bedouin shepherds in 1946–1947. (Apparently, a shepherd threw a rock while searching for a lost member of his flock and accidentally shattered one of the clay jars, leading to the discovery.) Qumran was destroyed by the Romans [wikipedia.org], circa 73 CE, and historians believe the scrolls were hidden in the caves by a sect called the Essenes to protect them from being destroyed. The natural limestone and conditions within the caves helped preserve the scrolls for millennia.

This isn't the first time that AI tools have been applied to analyzing the handwriting on the Dead Sea Scrolls. Back in 2020, we reported [arstechnica.com] on an AI-aided handwriting analysis of the Great Isaiah Scroll [wikipedia.org]. Most scholars believed that the Isaiah Scroll was copied by a single scribe due to the seemingly uniform handwriting style. But others have suggested that it may be the work of two scribes writing in a similar style, each copying one of the scroll's two distinct halves. The 2020 analysis revealed [plos.org] that the text was indeed likely written by two scribes. It also showed that the second scribe's handwriting was more variable than the first's, although the two styles were quite similar, indicating a possible common training.

Enter Enoch

The authors of this latest paper wanted to find a better means than paleography alone for determining the chronology of the scrolls, with an eye toward reconstructing the evolution of ideas. They thought that given the small data set in the case of the scrolls, it was wisest not to rely on a pre-trained model and instead "let the available data speak."

The development of Enoch grew out of the team's earlier deep neural network for ferreting out handwritten ink-trace patterns in digitized manuscripts, involving micro-level geometric shape analysis. "Enoch emphasizes shared characteristics and similarity matching between trained and test manuscripts, where traditional paleography focuses on subtle differences that are assumed to be indicative for style development," the authors wrote. "Combining dissimilarity matching and adaptive reinforcement learning can uncover hidden patterns."

They tested Enoch by having paleographic experts evaluate the AI program's age estimate for several scrolls. The results: About 79 percent of Enoch's estimates were deemed "realistic," while its age estimates for the remaining 21 percent were either too young, too old, or just indecisive.

This new model revealed that many of the Dead Sea Scrolls are older than previous estimates based solely on paleography. That should be relevant for the question of when two ancient Jewish script styles—"Hasmonean" and "Herodian"—developed, for example. The former script was thought to have emerged between 150–50 BCE, but the authors believe Hasmonean could have emerged much earlier; ditto for the Herodian script. So both scripts may have coexisted since the late second century, challenging the prevailing view that they preexisted by the mid-first century BCE.

Enoch may even be able to shed light on biblical authorship. For instance, the authors concluded that two of the scrolls are the first known fragments of the Book of Daniel, believed to have been finished by an anonymous author around 160 BCE. And Ecclesiastes was likely completed by an anonymous author in the third century BCE, rather than by King Solomon in the 10th century BCE.

“With the Enoch tool we have opened a new door into the ancient world, like a time machine, that allows us to study the hands that wrote the Bible, especially now that we have established, for the first time, that two biblical scroll fragments come from the time of their presumed authors,” the authors wrote. “It is very exciting to set a significant step into solving the dating problem of the Dead Sea Scrolls and also creating a new tool that could be used to study other partially dated manuscript collections from history."

PLoS ONE, 2025. DOI: 10.1371/journal.pone.0323185 [doi.org]  (About DOIs [arstechnica.com]).

Journal Reference:
Mladen Popović, Maruf A. Dhali, Lambert Schomaker, et al. Dating ancient manuscripts using radiocarbon and AI-based writing style analysis, PLOS ONE (DOI: 10.1371/journal.pone.0323185 [doi.org])

Original Submission