“Authentic according to the criteria of the two Shaykhs.” — End of citation (abridged). This Ibn ‘Abbas report is recorded in Sunan (1/126) and in al-Mustadrak (1/231) through the chain Sufyan → ‘Amr → Sa‘id b. Jubayr → Ibn ‘Abbas with the first wording; al-Hakim, however, reads khatama (“concluded”) instead of fasala (“separated”). That chain is sound by his own standard, exactly as al-Hakim claimed, and al-Dhahabi endorsed his judgment. Al-Hakim then reproduced the hadith—and al-Bayhaqi (2/43) cites it from him—through the route Walid b. Muslim → Ibn Jurayj → ‘Amr with the wording: “The Muslims would not know that a surah had finished until Bism Allah al-Rahman al-Rahim (In the name of Allah, the Most Merciful, the Most Compassionate) was revealed; once Bism Allah al-Rahman al-Rahim came down they realised that the surah had ended.” In this version Sa‘id b. Jubayr is absent from the isnad. Al-Hakim then remarks, “Authentic according to both of them,” and that verdict is correct. He also narrates it with a third wording, but that version is weak. The sound view, therefore, is that the Basmalah is a Qur’anic verse in its own right and not part of al-Fatiha, for the Companions—may Allah be pleased with them—agreed unanimously to write it in the manuscripts at the head of every surah (apart from Bara’a) in the same script as the Qur’an, unlike the one-tenth markers and the surah headings, which were customarily written in red ink and the like. Al-Nawawi (3/336) observes: “Had it not been Qur’an they would never have dared to inscribe it in the mushaf without a distinguishing mark, for that would cause people to believe it is Qur’an and thus mislead the Muslims, inducing them to hold….” (*) The Shaykh later stated explicitly that it is part of al-Fatiha but should not be recited aloud in the prayer; see Talkhis Sifat Salat al-Nabi (p. 15, 1st ed., al-Ma‘arif) and al-Sahiha, no. 1183.