Abstract
The Bible is the product of a complex process of oral and written transmissions that stretched across centuries and traditions. This implies ongoing revision of the "original" or oldest textual layers over the course of hundreds of years. Although critical scholarship recognizes this fact, debates abound regarding the reconstruction of the different layers, their date of composition and their historical backgrounds. Traditional methodologies have grappled with these challenges through textual and diachronic criticism, employing linguistic, stylistic, inner-biblical, archaeological and historical criteria. In this study, we use computer-assisted methods to address the question of authorship of biblical texts by employing statistical analysis that is particularly sensitive to deviations in word frequencies. Here, the term "word" may be generalized to "n-gram" (a sequence of words) or other countable text features. This paper consists of two parts. In the first part, we focus on differentiating between three distinct scribal corpora across numerous chapters in the Enneateuch, the first nine books of the Bible. Specifically, we examine 50 chapters labeled according to biblical exegesis considerations into three corpora: the old layer in Deuteronomy (D), texts belonging to the "Deuteronomistic History" in Joshua-to-Kings (DtrH), and the Priestly writings (P). For pragmatic reasons, we chose entire chapters, in which the number of verses potentially attributed to different authors or redactors is negligible. Without prior assumptions about author identity, our approach leverages subtle differences in word frequencies to distinguish among the three corpora and identify author-dependent linguistic properties. Our analysis indicated that the first two scribal corpora - (D, the oldest layers of Deuteronomy, and DtrH, the so-called Deuteronomistic History) - are much more closely related to each other than they are to the third, (P). This observation aligns with scholarly consensus. In addition, we attained high accuracy in attributing authorship by evaluating the similarity of each chapter to the reference corpora. In the second part of the paper, we report on our use of the three corpora as ground truth to examine other biblical texts whose authorship is disputed by biblical experts. Here, we demonstrate the potential contribution of insights achieved in the first part. Our paper sheds new light on the question of authorship of biblical texts by offering interpretable, statistically significant evidence of the existence of linguistic characteristics in the writing of biblical authors/redactors, that can be identified automatically. Our methodology thus provides a new tool to address disputed matters in biblical studies.