Abu Bakr al-Khatib said, “Muslim used to champion al-Bukhari, and on account of that he estranged himself from Muhammad ibn Yahya. (1)
I say: Later on, however—because of a certain quick-temperedness in his nature—Muslim also turned away from al-Bukhari. He did not cite a single hadith from him, nor did he mention him by name in his Sahih. Rather, he opened the book by fault-finding with those who require an actual meeting (liqaʾ) between a narrator and the person from whom he reports when the transmission appears with the wording ʿan, claiming scholarly consensus that mere contemporaneity is sufficient and that verification of a meeting is unnecessary; he even rebuked anyone who imposed that condition.
Yet it is precisely Abu ʿAbd Allah al-Bukhari and his teacher ʿAli ibn al-Madini who hold that view, and it is the sounder and stronger position.
This is not the place to expand on that question. (2)
The hadith-master Abu al-Qasim Ibn ʿAsakir says at the beginning of his work Al-Atrāf (3) after
(1) Tarikh Baghdad, vol. 13, p. 103; Wafayat al-Ayan, vol. 5, p. 194; Tadhkirat al-Huffaz, vol. 2, pp. 589-590; al-Bidaya wa-l-Nihaya, vol. 11, p. 34.
(2) One of the points that gives Sahih al-Bukhari precedence over Sahih Muslim is the following: Muslim states in the introduction to his collection that a hadith transmitted with the particle *an* (“on the authority of”) is to be deemed uninterrupted (muttasil) as long as the narrator and the person from whom he reports are contemporaries, even if no meeting between them can be documented. Al-Bukhari, however, does not consider such a chain connected unless it is proven that the two met at least once.
In his introduction Muslim discusses the issue of *muʿanʿan* transmission and notes that al-Bukhari required the narrator to have actually met the person he cites with *an*. Muslim then devotes considerable space to refuting and disparaging this view—without naming al-Bukhari explicitly, although the scholars unanimously agree that he is the one intended. Muslim concludes: “Any hadith in which it is said, ‘So-and-so *an* so-and-so,’ and it is known that the two lived in the same era and it is possible that the narrator heard the report directly from him, even though we have no evidence of such hearing or of any meeting between them, remains valid and constitutes binding proof.” He adds that this is the prevalent, agreed-upon position among experts in reports, ancient and modern.
The disagreement between al-Bukhari and Muslim concerns only narrations employing the particle *an*. As for those introduced by *akhbarana*, *haddathana*, or *anbana* (“he informed us,” “he related to us”), both scholars treat them alike, for these formulas imply direct oral transmission.
(3) In his Atraf, the hadith master Ibn ʿAsakir indexed the four Sunan—Abu Dawud, al-Tirmidhi, al-Nasaʾi, and Ibn Majah. Before him, the atraf of the two Sahihs had been compiled by the scholars Abu Muhammad Khalaf ibn Muhammad al-Wasiti and Abu Masʿud Ibrahim ibn Muhammad al-Dimashqi. Thereafter, al-Hafiz…