We apply a uniform computational pipeline (TF-IDF normalized cosine similarity, Fibonacci additive growth metric, Hurst R/S persistence analysis, and lag-20 autocorrelation) to five corpora across two language families: the Quran (114 surahs), three major hadith collections (Sahih al-Bukhari, Sahih Muslim, and Muwatta Malik), and the Hebrew Bible (39 books in the Protestant canonical division). Three layers of structural difference emerge. First, Fibonacci additive semantic growth is a Classical Arabic compiled-text property shared between the Quran and all three hadith collections (all p = 0.19) within this sample and is a candidate Quran-specific structural property. Third, the lag-20 autocorrelation sign-flip is a robust cross-text finding: the Quran shows positive thematic continuity (+0.53), the Hebrew Bible shows negative thematic alternation (-0.42), and hadith collections show no directional pattern (near zero). The paper reports five honest caveats including a methodological reconciliation between two independent pipelines that produced contradictory Hebrew Bible results (p = 0.44 vs. p = 0.0000), resolved by identifying TF-IDF normalization as the differentiating factor. The Hurst persistence finding was discovered post-hoc and is reported as a candidate finding. Related software: Bayyinah Integrity Scanner (https://github.com/BayyinahEnterprise/Bayyinah-Integrity-Scanner, https://bayyinah.dev) and Furqan programming language (https://github.com/BayyinahEnterprise/furqan-programming-language).
Arfeen et al. (Fri,) studied this question.