Classifying the suras by their lexical semantics :an exploratory multivariate analysis approach to understanding the Qur'an

Abstract

The Qur'an is at the heart of Islamic culture. Careful, well-informed interpretation of it is fundamental both to the faith of millions of Muslims throughout the world, and also to the non-Islamic world's understanding of their religion. There is a long and venerable tradition of Qur'anic interpretation, and it has necessarily been based on literary-historical methods for exegesis of hand-written and printed text. Developments in electronic text representation and analysis since the second half of the twentieth century now offer the opportunity to supplement traditional techniques by applying the newly-emergent computational technology of exploratory multivariate analysis to interpretation of the Qur'an. The general aim of the present discussion is to take up that opportunity. Specifically, the discussion develops and applies a methodology for discovering the thematic structure of the Qur'an based on a fundamental idea in a range of computationally oriented disciplines: that, with respect to some collection of texts, the lexical frequency profiles of the individual texts are a good indicator of their semantic content, and thus provide a reliable criterion for their conceptual categorization relative to one another. This idea is applied to the discovery of thematic interrelationships among the suras that constitute the Qur'an by abstracting lexical frequency data from them and then analyzing that data using exploratory multivariate methods in the hope that this will generate hypotheses about the thematic structure of the Qur'an. The discussion is in eight main parts. The first part introduces the discussion. The second gives an overview of the structure and thematic content of the Qur'an and of the tradition of Qur'anic scholarship devoted to its interpretation. The third part xvi defines the research question to be addressed together with a methodology for doing so. The fourth reviews the existing literature on the research question. The fifth outlines general principles of data creation and applies them to creation of the data on which the analysis of the Qur'an in this study is based. The sixth outlines general principles of exploratory multivariate analysis, describes in detail the analytical methods selected for use, and applies them to the data created in part five. The seventh part interprets the results of the analyses conducted in part six with reference to the existing results in Qur'anic interpretation described in part two. And, finally, the eighth part draws conclusions relative to the research question and identifies directions along which the work presented in this study can be developed.