the tier that contains narrators about whom there is consensus that nothing of their ḥadith may be adduced as evidence unless they explicitly state having heard it, for they habitually practice tadlīs from weak or unknown transmitters—Baqiyyah ibn al-Walīd, for example, as he notes in the introduction. These two passages from al-Ḥāfiẓ himself therefore prove that he slipped when he weakened ʿAṭiyyah on the ground of tadlīs in the sentence cited a moment ago. This is one point of conflict between those passages and what appears in al-Taqrīb.
A second point is that, in that very sentence, he failed to describe ʿAṭiyyah with the criticism he himself deems disqualifying—as we saw from Sharḥ al-Nukhbah—namely his statement in al-Taqrīb: “he frequently errs.”
All of this shows that al-Ḥāfiẓ—may Allah have mercy on him—was not fully assisted by his memory when he traced this ḥadith, and so he fell into the shortcoming that his remarks in his other writings attest to. Those other works deserve greater reliance than his Takhrīj, for in them he quotes directly from the primary sources and produces abridgements, whereas his method in the Takhrīj is different.
Because of al-ʿAwfī’s condition, more than one ḥadith master declared the report weak: al-Mundhirī in al-Targhīb¹, al-Nawawī, Shaykh al-Islām Ibn Taymiyyah in al-Qāʿidah al-Jalīlah, and likewise al-Būṣīrī, who said in Miṣbāḥ al-Zujājah 52/2:
“This isnād is a continuous chain of weak narrators: ʿAṭiyyah, Fuḍayl ibn Marzūq, and al-Faḍl ibn al-Muwaffaq—every one of them is weak.”
Ṣiddīq Khān likewise said in Nuzul al-Abrār