Determining the Significance of Lexical Features as Indicative of CBH and LBH: Insights from the Tiberias Stylistic Classifier for the Hebrew Bible

Abstract

Since the time of Gesenius, scholars have rightly grounded their determinations of linguistic dating of the biblical texts on the comparison of pairs of features that can be contrasted as early and late. However, at times Hebraists also identify terms as indicative of either Classical Biblical Hebrew (CBH) or Late Biblical Hebrew (LBH) on account of their relatively exclusive occurrence in either one corpus or the other. In this study I demonstrate the propensity of scholars to identify such terms in an impressionistic manner and unwittingly fall victim to a probability illusion long known by cognitive psychologists as the small sample fallacy. Conversely, I will show that in seeking lexical terms that are indicative of CBH and LBH respectively, they overlook other terms that are far more indicative and significant from a statistical perspective. To arrive at these conclusions, I employ data generated by the recently launched Tiberias Stylistic Classifier for the Hebrew Bible. Tiberias marshals cutting edge advances in the field of machine learning and computational linguistics to empower users to easily conduct their own experiments analysing and classifying the texts of the Hebrew Bible through the measurable features of linguistic data, and providing them with verifiable results. As an illustration of what is at stake, I reference the debate surrounding the linguistic profile of Genesis 24.