Artificial Intelligence Helps Crack the Code of the Dead Sea Scrolls

Handwriting Analysis of Dead Sea Scrolls

Revealed: The Secrets our Clients Used to Earn $3 Billion

Dead sea scroll text analysis collage. Credit: Mladen Popović

The Dead Sea Scrolls, found some seventy years earlier, are well-known for including the earliest manuscripts of the Hebrew Bible (Old Testament) and numerous hitherto unidentified ancient Jewish texts. But the specific individuals behind the scrolls have actually avoided researchers, due to the fact that the scribes are confidential. Now, by integrating the sciences and the liberal arts, University of Groningen scientists have actually split the code, which allows them to find the scribes behind the scrolls. They provided their lead to the journal PLOS ONE on April 21, 2021.

The scribes who developed the scrolls did not sign their work. Scholars recommended some manuscripts ought to be credited to a single scribe based upon handwriting. “They would try to find a “smoking gun” in the handwriting, for instance, an extremely particular characteristic in a letter which would determine a scribe,” describes Mladen Popović, teacher of Hebrew Bible and Ancient Judaism at the Faculty of Theology and Religious Studies at the University of Groningen. He is likewise director of the university’s Qumran Institute, committed to studying the Dead Sea Scrolls. However, these recognitions are rather subjective and frequently fiercely discussed.


That is why Popović, in his job The Hands that Wrote the Bible which was moneyed by the European Research Council, partnered with his coworker Lambert Schomaker, teacher of Computer Science and Artificial Intelligence at the Faculty of Science and Engineering. Schomaker has actually long dealt with methods to permit computer systems to check out handwriting, frequently from historic products. He likewise carried out research studies to examine how biomechanical qualities, like the method which somebody holds a pen or stylus, would impact handwriting.

Analyzing Letters Dead Sea Scrolls

Two 12×12 Kohonen maps (blue colourmaps) of complete character aleph and wager from the Dead Sea Scroll collection. Each of the characters in the Kohonen maps is formed from numerous circumstances of comparable characters (revealed with a zoomed box with red lines). These maps work for sequential design advancement analysis. In the existing research study of author recognition, Fraglets (fragmented character shapes) were utilized rather of complete character shapes to accomplish more exact (robust) results. Credit: Maruf A. Dhali, University of Groningen

In this research study, together with PhD prospect Maruf Dhali, they concentrated on one scroll in specific: the well-known Great Isaiah Scroll (1QIsaa) from Qumran Cave 1. The handwriting in this scroll appears near-uniform, yet it has actually been recommended it was made by 2 scribes sharing a comparable composing design. So how could this be chosen? Schomaker: “This scroll contains the letter aleph, or “a”, a minimum of 5 thousand times. It is difficult to compare them all simply by eye.” Computers are well fit to examine big datasets, like 5,000 handwritten a’s. Digital imaging makes all sorts of computer system computations possible, at the microlevel of characters, such as determining curvature (called textural), in addition to entire characters (called allographic).

Neural network

“The human eye is amazing and presumably takes these levels into account too. This allows experts to “see” the hands of various authors, however that choice is frequently not reached by a transparent procedure,” Popović states. “Furthermore, it is virtually impossible for these experts to process the large amounts of data the scrolls provide.” That is why their outcomes are frequently not definitive.

Binarization Dead Sea Scrolls

(From delegated right) Greyscale picture of column 15 of the Great Isaiah Scroll, the matching binarized image utilizing BiNet, and the cleaned-corrected image. From the red boxes of the last 2 images, one can see how the rotation and the geometric change is fixed to yield a much better image for more processing. Credit: Reprinted from Lim TH, Alexander PS. Volume 1. In: The Dead Sea Scrolls Electronic Library. Brill; 1995 under a CC BY license, with approval from Brill Publishers, initial copyright 1995.

The very first obstacle was to train an algorithm to separate the text (ink) from its background (the leather or the papyrus). For this separation, or “binarization,” Dhali established a modern synthetic neural network that can be trained utilizing deep knowing. This neural network keeps the initial ink traces made by the scribe more than 2,000 years ago undamaged as they appear on the digital images. “This is important because the ancient ink traces relate directly to a person’s muscle movement and are person-specific,” Schomaker describes.


Dhali carried out the very first analytical test of this research study. His analysis of textural and allographic functions revealed that the 54 columns of text in the Great Isaiah Scroll fell under 2 various groups that were not dispersed arbitrarily through the scroll, however were clustered, with a shift around the midway mark.

Aleph Dead Sea Scrolls

An illustration of how heatmaps of stabilized typical character shapes are created for specific letters (in this example: aleph). Credit: Maruf A. Dhali, University of Groningen

With the remark that there may be more than one author, Dhali then handed the information to Schomaker who then recomputed the resemblances in between the columns, now utilizing the patterns of letter pieces. This 2nd analytical action validated the existence of 2 various. Several more checks and controls were carried out. Schomaker: “When we added extra noise to the data, the result didn’t change. We also succeeded in demonstrating that the second scribe shows more variation within his writing than the first, although their writing is very similar.”


In the 3rd action, Popović, Dhali, and Schomaker have actually produced a visual analysis. They developed “heat maps” that include all the versions of a character throughout the scroll. Then they produced a balanced variation of this character for the very first 27 columns and the last 27 columns. Comparing these 2 typical letters by eye reveals that they are various. This connects the electronic and analytical analysis to human analysis of the information by approximation, due to the fact that the heatmaps are neither reliant nor produced from the main and secondary analyses.

Certain elements of the scroll and the positioning of the text had actually led some scholars to recommend that after column 27 a brand-new scribe had actually begun, however this was not typically accepted. Popović: “Now, we can confirm this with a quantitative analysis of the handwriting as well as with robust statistical analyses. Instead of basing judgment on more-or-less impressionistic evidence, with the intelligent assistance of the computer, we can demonstrate that the separation is statistically significant.”

New window

In addition to changing the paleography of the scrolls — and possibly other ancient manuscript corpora — this research study of the Great Isaiah Scroll opens an absolutely brand-new method to examine the Qumran texts based upon physical attributes. Now, scientists can access the microlevel of specific scribes and thoroughly observe how they dealt with these manuscripts.

Popović: “This is very exciting, because this opens a new window on the ancient world that can reveal much more intricate connections between the scribes that produced the scrolls. In this study, we found evidence for a very similar writing style shared by the two Great Isaiah Scroll scribes, which suggests a common training or origin. Our next step is to investigate other scrolls, where we may find different origins or training for the scribes.”

In in this manner, it will be possible for more information about the neighborhoods that produced the Dead Sea Scrolls. “We are now able to identify different scribes,” Popović concludes. “We will never know their names. But after seventy years of study, this feels as if we can finally shake hands with them through their handwriting.”

Reference: “Artificial intelligence based author recognition produces brand-new proof for the unidentified scribes of the Dead Sea Scrolls exhibited by the Great Isaiah Scroll (1QIsaa)” by Mladen Popović, Maruf A. Dhali and Lambert Schomaker, 21 April 2021, PLOS ONE.
DOI: 10.1371/journal.pone.0249769

Digital pictures of the Dead Sea Scrolls and of the Great Isaiah Scroll were kindly offered by Brill Publishers and the Israel Antiquities Authority (the Leon Levy Dead Sea Scrolls Digital Library).

This site uses Akismet to reduce spam. Learn how your comment data is processed.