Dead Sea Scrolls Chronology Rewritten

The Dead Sea Scrolls are older than previously thought and in some cases were written at the same time as the anonymous authors of the Bible texts. This is what emerges from a new international study published in the journal "Plos One" in which the University of Pisa participated together with the universities of Groningen and Southern Denmark as part of the European ERC project "The hands that wrote the Bible". The research combined artificial intelligence with radiocarbon dating for the first time, allowing the chronology of the manuscripts to be refined. The Dead Sea Scrolls are a set of ancient Jewish manuscripts of religious content found in the Qumran caves in the Judean desert, near Ein Feshkha on the north-western shore of the Dead Sea in the West Bank, discovered between 1947 and 1956.

Tests on 135 scrolls revealed that many manuscripts are actually much older than previously thought. Manuscripts in the Hasmonean script, traditionally dated to 150-50 BC, sometimes even predate the mid-2nd century BC (i.e., around 175-150 BC or even the late 3rd century BC). The Herodian script was also found to be older than expected, indicating that the two styles coexisted as early as the late 2nd century BC, rather than the mid-1st century BC, as previously thought.

The study also identified two fragments of the Book of Daniel and Ecclesiastes - 4QDanielc and 4QQoheleta - that date back to the 2nd and 3rd centuries BC, respectively, the same period in which the anonymous authors of the respective biblical texts are presumed to have lived. This is the first time that fragments of the Bible can be associated with such a high probability to the time of their authors, offering tangible evidence for scholars who are questioning the origins of the Bible and the transmission of its texts.

The Pisa team, composed of professors Ilaria Degano and Maria Perla Colombini and Dr. Jacopo La Nasa of the Department of Chemistry and Industrial Chemistry, participated in the study by developing innovative protocols for the removal of contaminants present in the fragments, in particular residual substances from old restorations, which could have altered the accuracy of the radiocarbon dating.

"Our task was to ensure that the materials sent for dating were as clean as possible and free of residues that could alter the results - explains Professor Ilaria Degano - To do so, we developed and validated a specific protocol that allows us to remove contaminants without compromising the integrity of the sample and to verify it analytically. It was a challenge that required great precision, because we worked on unique and fragile finds. This highly innovative approach may be used in the future also on other ancient manuscripts, offering an advanced standard for the preparation of samples intended for radiometric dating".

The chemical-physical data obtained were essential to train Enoch, a new predictive model based on machine learning, capable of estimating the age of manuscripts starting from images of the writings. Developed at the University of Groningen, Enoch uses BiNet, a deep neural network designed to analyze ink traces and the morphology of handwritten characters. The algorithm, validated through radiocarbon dating, achieved unprecedented precision, with an uncertainty margin of approximately plus or minus 30 years, lower than that of the same radiometric techniques in the period considered.

(by Paolo Martini)