Audience Architecture Across 114 Surahs:<b>A Reproducibility Dataset of Addressee Fields, Vocative Signals, Pronoun Shifts, and Discourse Transitions in the Received Qur'an</b>

Clear lament and wish uses are retained as methodological controls rather than being misclassified as audience signals. The remaining surface-address layer records addressive and address-like occurrences with Qur'anic reference, Arabic text, local context, lexical target class, addressivity class, and confidence.</p><p dir="ltr">A separate 114-row surah registry summarizes explicit address density, target diversity, first and last address references, and whether an explicit address occurs within the opening or closing three-ayah windows.

Deeper fields such as dominant discourse addressee, speaker architecture, addressee-shift count, pronoun-shift profile, relation to ʿamūd, and structural function remain UNCODED unless contextual analysis has actually been completed.</p><p dir="ltr">The package also includes a full audience-architecture coding template, controlled vocabulary, codebook, source register, summary statistics, and five bounded pilot studies covering Al-Baqarah, Al-Ahzab, Al-Hujurat, Al-Muzzammil, and Al-Muddaththir.</p><p dir="ltr">The dataset is intended for Qur'anic studies, Islamic studies, Arabic discourse analysis, rhetoric, iltifāt research, naẓm and munāsabah studies, digital humanities, corpus-assisted close reading, and reproducible structural research.</p><p dir="ltr">Its wider purpose is to make audience claims inspectable, testable, and versionable by keeping textual observation, referential resolution, structural inference, and theological interpretation in separate fields.</p>