A famous Dead Sea Scroll manuscript was written by not only one but two scribes, consistent with a latest new study that used AI and statistics to detect subtle differences in handwriting on the ancient-document.
The two scribes wrote in such an identical manner that the differences between the 2 aren’t visible to the eye , the analysis revealed — a detail that means the scribes may received similar training, perhaps at a faculty or in close social setting, the researchers wrote in study.
“This is simply the 1st step,” study PI (principle investigator) Mladen Popović, a professor of the Hebrew Bible and ancient Judaism at the University of Groningen in Netherlands. “We have opened the door to the microlevel of individual scribes; this may open new possibilities to review all the scribes behind the Dead Sea Scrolls and put us in new and potentially better position to know with what quite collection, or collections of manuscripts we’re handling here.”
The Dead Sea Scrolls were first discovered in late 1940s, when a young shepherd trying to find a stray goat found several manuscripts in-cave-in Qumran, in West Bank . Over subsequent decade, researchers and native Bedouins found quite 900 manuscripts in 11 caves. These manuscripts are the oldest remaining texts of the Hebrew Bible, dating from the fourth century B.C. to the second century A.D. But it’s unclear who or maybe what percentage people wrote them, because the scribes didn’t sign their names, the researchers of the new study said.
That hasn’t stopped biblical scholars from guessing what percentage scribes were involved in penning the varied Dead Sea Scroll manuscripts. “They would attempt to find a ‘smoking gun‘ in handwriting, for instance , a really specific trait during a letter which might identify a scribe,’ Popović, who is additionally the director of the University of Groningen’s Qumran Institute, said in statement. But these “smoking gun” analyses were often subjective and, as a result, hotly debated, he said.
So, Popović and his colleagues used another approach — AI and statistics — to research the Great Isaiah Scroll, one among the seven scrolls originally found by the Bedouin shepherd. This well-preserved scroll, which dates to about 125 B.C., is lengthy — it measures 24 feet (7.3 meters) long and 10 inches (26 centimeters) high — and contains 54 columns of Hebrew text. One spot, especially , caught Popović’s eye; between columns 27 and 28, there’s alittle break within the text and a new “page,” where two sheets are sewn together. Other researchers had already debated whether this scroll was written by one or two scribes, and Popović’s team wanted to ascertain if they might solve the mystery.
In effect, the team try to find out “whether subtle differences in writing should be considered normal variations in handwriting of 1 scribe or as similar scripts of two different scribes,” they wrote in a study.
The researchers’ methods detected “subtle and nuanced differences in handwriting that we cannot discern with the human eye only,”the invention that two scribes collaborated on the Great Isaiah Scroll reveals that ancient scribes “worked in teams,” he said. And, unlike the “smoking gun” analyses, this research “is not just a conjecture, but supported evidence now,” Popović added.
How they did it
When designing the algorithm, the researchers had to coach it to differentiate the text, or the ink, from the background — the animal skin or papyrus. This distinction, referred to as binarization, was designed by study co-researcher Maruf Dhali, a doctoral student within the AI department at the University of Groningen, who created a man-made neural network that would be trained using deep learning. This neural network recorded the first ink traces on the manuscript, even when these ancient letters were transformed into digital images.
“This is vital because the ancient-ink traces relate on to a person’s muscle movement and are person-specific,” study senior researcher Lambert Schomaker, a professor of computing and AI at the University of Groningen, said within the statement.
The neural network analysis revealed that the 54 columns of text in Great Isaiah Scroll fell into two distinct groups, which had a transition about halfway through the manuscript. Dhali told Schomaker that there could be quite one writer, so Schomaker did a separate analysis but got same result. In second analysis, Schomaker checked out fraglets, or parts of the letters that “can be more precise, distinctive and informative find significant shape differences than the complete characters,” the researchers wrote in-a study.
To be extra cautious, the team added checks and controls to the text. “When we added extra noise to the info , the result didn’t change,” Schomaker said. “We also succeeded in demonstrating that the second scribe shows more variation within his writing than the 1st, although their writing is extremely similar.”
Next, the team performed a visible analysis by creating “heat maps.” These maps incorporated all of the variants of a given letter, like the Hebrew letter aleph (א), found in scroll. Then, they made a mean version of the letter from the 1st 27 columns and another from the last 27 columns. then , they compared these averaged letters, and located that they might easily spot differences between the 2 . Moreover, the differences were statistically significant, Popović said.
Popović and his colleagues decide to investigate other scrolls, which can reveal different origins or training for various scribes, he said. These analyses can also shed light on the communities that wrote the Dead Sea Scrolls. “Understanding the scribes of the Dead Sea Scrolls makes it possible to better-understand what I call the cultural evolution of the Hebrew Bible,” Popović said.
The new research “is the 1st time that automatic procedure was applied to spot the transition of favor in Great Isaiah Scroll,” Shira Faigenbaum-Golovin, a researcher in Department of applied math at Tel-Aviv University who focuses on biblical-era handwriting analyses. Faigenbaum-Golovin wasn’t involved in study. “The method utilized in this study handles well the challenges raised by the poor state of preservation of the scroll via robust binarization.”
The study was published in the journal PLOS One.