“The words ‘when he retires to his bed’ do not occur in Shuʿbah’s hadith.”

I say: Sufyān had a stronger memory than Shuʿbah, especially since an addition from a trustworthy narrator is accepted. Moreover, Misʿar also attributed it to the Prophet. I then found that al-Aʿmash narrated it from Ḥabīb as a report attributed to the Companion, in Ibn Abī Shaybah (9/73–74 and 10/250). Thus, the additional wording is to be given precedence—particularly since this is not something that could be said merely on the basis of personal opinion, as is evident. The chain is authentic: its narrators are trustworthy and are among Muslim’s narrators, were it not that Ḥabīb ibn Abī Thābit practiced concealment in transmission, as Ibn Ḥibbān himself stated, following his teacher Ibn Khuzaymah. However, I have previously explained what indicates that his concealment in transmission was infrequent, and that scholars accept the hadith of one like him unless it becomes clear that it contains a decisive hidden defect. This is why Ibn Ḥibbān included reports transmitted by him with the ambiguous “from” formula in his Ṣaḥīḥ, and this is one of them. See the discussion under the preceding hadith.

Furthermore, there is something here that reassures us against concern over his concealment in transmission: Shuʿbah’s narration from him, as al-Nasāʾī previously related. Shuʿbah was well known for his scrupulousness in not narrating from someone known for concealment in transmission unless he explicitly stated that he had heard it directly, as is recorded in Taqdimat al-Jarḥ wa-al-Taʿdīl (161 and 169):

“Shuʿbah said: ‘I used to watch Qatādah’s mouth closely: if he said, “I heard,” or, “He related to us,” I retained it; but if he said, “So-and-so related,” I left it.’”

Consult 2/35 of that work, and your certainty will increase.

Note: The annotator of al-Iḥsān contradicted himself regarding this hadith. In his annotation to al-Iḥsān (12/338), he declared it defective because Ḥabīb used the ambiguous “from” formula; yet in his edition of Mawārid al-Ẓamʾān (2/1066), he said that it is “good”! He offered no explanation for this after having declared it defective, even though he referred to that discussion. The three annotators likewise blindly followed him in grading it good, without any explanation—as is their habit!

As for the two annotators of the Syrian edition of al-Mawārid, they followed the proper course and judged its chain weak because of the defect of ambiguous transmission. *