Pre-registration of two robustness analyses and an external replication for "Astronomical Correspondence in the Textual Architecture of the Qurʾān"

Pre-registration of two robustness analyses, and of the design of an external replication, for the article Astronomical Correspondence in the Textual Architecture of the Qurʾān: Occurrence Ranks, Verse Coordinates and the Metonic Cycle (all versions: 10.5281/zenodo.20061565). The article reports a pattern that was found during exploration. Its multiverse analysis does not price two freedoms: which words were chosen, and the uniform null model. This record fixes, before they are run: Part A — the choice of words. The article's construction is applied to every configuration of three of the eleven nouns the article names: 495 configurations, the primary family. It is also applied to every configuration of two to four nouns: 2,530 configurations, the secondary family. The primary statistic is the family-wise probability that a search over the family finds at least one agreement. Part B — clustered null models. The article's configuration is evaluated under three null models that keep the clustering of the text: noun-verse, sūra-order and within-sūra permutation. Part C — external replication (design). The construction is to be applied to Arabic corpora with a fixed two-level numbering. The corpora will be fixed in an addendum, deposited as a new version of this record before any count is made. Parts A and B have decision rules, and every outcome will be reported. At registration: no statistic has been computed for any configuration other than the article's; no p-value has been computed under the new null models; no count has been made in any external corpus. For the article's own configuration, the deposited code reproduces the published values. Validation outputs, an independent implementation and a pre-planned sampler-equivalence test are included. Files. PREREGISTRATION.pdf and PREREGISTRATION.md are the registered document. LCI_preregistration_code_data.zip contains the code, the verse-by-noun table (checked against the Quranic Arabic Corpus), the validation outputs and SHA256SUMS.txt. SHA-256 of the files: PREREGISTRATION.md 4e4f5634b2e2df44d9756315dad5880c4e3740d35c3cfa57e710550be1f74146 PREREGISTRATION.pdf 77a64f941fe25af6a20cde63a89a15d7d264a34a9484b83cd94f00dcd423090f LCI_preregistration_code_data.zip 8953d5d6c8fd3eaedf90667a9786df07419c2150550440ef2b4b928e1d490be2