Scholars use data to untangle Bible’s origins
Graphical representation of the team’s results identifying biblical authors by writing style. Courtesy Faigenbaum-Golovin et al., Duke University.
The Hebrew Bible is the product of centuries of oral and written transmission, combining numerous sources and multiple redactions. Although this has been a long-recognized feature of the biblical text, scholars still fiercely debate the precise details, including what parts of the text originated from the same authors or scribal schools. Publishing in the journal PLOS One, a team of computer scientists and biblical scholars has proposed a new method to untangle the web of authorial pieces that make up the Hebrew Bible.
FREE ebook: Cyber-Archaeology in the Holy Land — The Future of the Past. Modern archaeological methods help create a new and objective future of the past.
Biblical scholars generally agree that individual books of the Hebrew Bible, as well as many chapters and verses, were the product of multiple authors and redactors. Yet, despite hundreds of years of biblical text criticism, scholars continue to argue about seemingly every facet of this theory, most notably, who wrote what. In hopes of providing a new tool to answer this question, an international team of computer scientists and biblical scholars developed a statistical analysis model for the biblical text. The model is designed to analyze language patterns across large sections of texts and identify patterns that can be used to differentiate one set of texts from another. Such a model was used over an artificial intelligence (AI) model because it allowed the team to track and analyze the model’s reasoning in making conclusions and predictions.
In applying this model to the Hebrew Bible, the team took 50 chapters from the first nine books that are widely accepted to belong to one of three different scribal collections: Deuteronomy, the Deuteronomistic History, and the Priestly material. The model then analyzed the chapters and identified patterns that set each collection apart from the others. The specific patterns investigated were individual Hebrew words and phrases, with the model identifying which words and phrases each collection used and at what frequency.
If you'd like to help make it possible for us to continue Bible History Daily, BiblicalArchaeology.org, and our email newsletter please donate. Even $5 helps:
Once the model had identified noticeable patterns in the three collections, the team applied it to an assortment of other chapters and verses in the Hebrew Bible, whose attribution to a particular scribal tradition has been more debated. Among the most notable results, the model suggested that the Book of Esther does not belong to any of the three scribal collections. Similarly, while the later part of the so-called ark narrative (2 Samuel 6) belonged to the Deuteronomistic History, the earlier part (1 Samuel 1–6) did not. According to the team, this can be interpreted as evidence that the earlier part of the narrative predates the Deuteronomistic History.
Although the team suggests that further work is needed to perfect their method, they hope that the research “paves the way for scholars to address additional questions in biblical studies in statistical methods, thus improving our understanding of the formation of the Hebrew Bible.”
Not a BAS Library or All-Access Member yet? Join today.
Sign up to receive our email newsletter and never miss an update.
Become an All-Access Member to explore the Bible's rich history. Get Biblical Archaeology Review in print, full online access, and FREE online talks. Plus, enjoy special Travel/Study discounts. Don't miss out—begin your journey today!