The Qur’anic Style in Descriptions of Natural Phenomena: Corpus-Based Feature Annotation and Decision-Tree Analysis (مقاله علمی وزارت علوم)

چکیده

The present study seeks to extract stylistic features of Qur’anic verses that address natural phenomena, some of which commentators have attempted to explain in scientific terms. To this end, seventy verses containing references to natural phenomena that had been claimed to constitute scientific miracles were selected. In addition, thirty verses were chosen which describe natural phenomena but from which no falsifiable propositions can be derived. As a third category, fifty verses whose central theme is supernatural were chosen for comparison. Consequently, three categories of verses were formed and, for brevity, labeled scientific verses, natural verses, and supernatural verses. Through an examination of existing scholarship on scientific verses, features potentially serving as stylistic markers of this category were identified. A corpus of 150 verses representing the three groups was then compiled and their features annotated. To analyze this corpus and extract distinctive stylistic traits of the first category, computational methods were applied, specifically a decision-tree analysis. The findings highlight linguistic characteristics of the Qur’an in its description of natural phenomena. The features with the highest probability of occurrence in scientific verses are, in descending order: multiple possible syntactic analyses, greater verse length, the presence of thought-provoking expressions, the use of augmented verb forms ( al-abwāb al-mazīd ), greater variety of letters, a higher number of derivational forms, and polysemy. A common underlying implication of these traits is that such verses require further inquiry and interpretive effort to construct a meaningful understanding. The study demonstrates that verses providing precise and falsifiable descriptions of natural phenomena display distinctive stylistic features that deserve further attention, ideally through interpretive approaches that are specifically sensitive to these linguistic markers.