The Significance of Three Methods of Grouping Biblical Hebrew Text Portions

Abstract

The paper first indicates the implications of the mixed results obtained by using three disparate analytical methods to infer relationships among biblical text portions based upon their spelling practices. Next, a sketch is provided of matres lectionis (“mothers of reading”) in Biblical Hebrew and of the Andersen-Forbes classification system. Vowel features are specified, and examples presented. The notion of transmissional textual change is introduced. The criticality of comparing the results provided by different analytical methods is emphasised. Next, three complementary analytical methods are introduced in turn, and their results are appraised. Clustering is a heuristic data exploration method, its prime result being that the spelling of the Torah sets it well apart from the other portions of the Hebrew Bible. Clustering, however, produces many other provocative portion groupings inviting investigation. While multidimensional scaling also gathers the Torah portions, it also yields its own tantalising juxtapositions. Seriation orders the portions along a timeline. It results in an expected horseshoe-shaped band of portions, albeit rather “puffy.” Also, some of its text-portion orders are suspicious. While many results produced by the three methods are encouraging, many are perplexing. Envisioned future application of evolving methods to our BH text-portion data may well enhance the trustworthiness of our inferences.