A digest keeps pointing at its dead parents

Claim (verbatim)

Al-Mas'udi's Muruj al-dhahab (completed 332/943, revised 345/956) presents itself as the portable tier of a cascade: behind it stand his vast Akhbar al-zaman and the middle abridgment al-Kitab al-awsat, both now lost. If the Muruj was built as a digest that deliberately offloads full treatment to its parents, the deferral formula - 'we have treated this fully in our book...' - should recur through the whole work as compositional discipline, not sit in the preface as a flourish; the surviving book is a running advertisement for two dead ones, which is the Inferome's subject in miniature. Prediction: in the OpenITI primary version of 0346Mascudi.MurujDhahab, occurrences of the alif-normalized string اخبار الزمان lying within 60 characters after an occurrence of كتاب (the pinned titular-use rule, excluding generic 'reports of the age' uses), plus all occurrences of the intrinsically titular string الكتاب الاوسط, will total at least 25, and at least one such occurrence will fall in each fifth of the text's milestone span (primary clause: the >=25 combined count; the verdict follows it); as a registered secondary, non-binding diagnostic, the KITAB one_to_all data will show fewer than 10 alignment instances between the Muruj and 0600PseudoMascudi.AkhbarZaman - the extant claimant to the lost parent's title shares essentially nothing with the book that claims to digest its namesake. Kill: the OpenITI primary-version text of 0346Mascudi.MurujDhahab, with marker positions taken from its native milestone tags, plus the KITAB one_to_all msdata file for the Muruj for the pseudo-Akhbar secondary.

Prediction clause (verbatim)

Prediction: in the OpenITI primary version of 0346Mascudi.MurujDhahab, occurrences of the alif-normalized string اخبار الزمان lying within 60 characters after an occurrence of كتاب (the pinned titular-use rule, excluding generic 'reports of the age' uses), plus all occurrences of the intrinsically titular string الكتاب الاوسط, will total at least 25, and at least one such occurrence will fall in each fifth of the text's milestone span (primary clause: the >=25 combined count; the verdict follows it); as a registered secondary, non-binding diagnostic, the KITAB one_to_all data will show fewer than 10 alignment instances between the Muruj and 0600PseudoMascudi.AkhbarZaman - the extant claimant to the lost parent's title shares essentially nothing with the book that claims to digest its namesake.

Kill-dataset (verbatim)

Kill: the OpenITI primary-version text of 0346Mascudi.MurujDhahab, with marker positions taken from its native milestone tags, plus the KITAB one_to_all msdata file for the Muruj for the pseudo-Akhbar secondary.

Novelty / leakage triage

anticipated in the literature — this exact test has never been run

Mas'udi's self-references to the lost Akhbar al-zaman and al-Kitab al-awsat are the published basis on which Khalidi and Shboul reconstructed the lost works' contents, so the deferral habit is thoroughly documented - Khalidi indeed argues the Muruj is an independent work rather than a mere digest - but no titular-string count with a positional-spread clause has been run on the text, and the pseudo-Akhbar reuse diagnostic is new.

• T. Khalidi, 'Mas'udi's Lost Works: A Reconstruction of Their Content', Journal of the American Oriental Society 94 (1974)
• A.M.H. Shboul, Al-Mas'udi and His World (London, 1979)

Predictions

No prediction registered yet.