After the Prophet Muhammad, the most contested figure in Islamic history would be his son-in-law, ʿAlī b. Abī Ṭālib (d. 40/661). Umayyads, Khārijīs and early ʿUthmāniyya (Muslims who revered the first three caliphs, but were hostile to ʿAlī) disavowed themselves of ʿAlī. According to these factions, ʿAlī wrongfully waged war against other Muslims in pursuit of power and betrayed the values of Islam. He and his partisans were no longer Muslims. For those who revered him, ʿAlī personified justice and righteousness. He was the ideal imam. Through the efforts of his admirers, ʿAlī became a respected authority in both Sunnī and Shīʿī Islam within a few centuries of his death. His near-universal portrayal in Muslim literature as a pious authority, however, rather obscures a centuries-long process of contestation and rehabilitation.Footnote 1 The Umayyad state apparatus (40–132 ah/661–750ce) had facilitated the proliferation of ʿUthmānī and Umayyad portrayals of ʿAlī for close to a century. With the fall of the Umayyads, ʿAlī’s subsequent transformation from heretic to saint was neither immediate nor complete.
Beginning in the third/ninth century, the compilers of Sunni ḥadīth literature faced a great challenge in sifting through conflicting narratives regarding the legacy of ʿAlī. On one hand, transmitters, some described as Shīʿī and some not, narrated ḥadīth about his merits (faḍāʾil, manāqib) and the aid he provided to the Prophet and many others as a pious member of the Muslim community. On the other hand, ʿUthmānī and pro-Umayyad scholars transmitted accounts that usually portrayed him as irreligious and immoral.Footnote 2 This study considers the methods that ḥadīth transmitters and scholars employed to reconcile expectations regarding ʿAlī’s character and image in Sunnism with the vast and heterogeneous body of accounts about him. As a sect, Sunnism encompasses Muslims who differ from one another considerably on the subject of ʿAlī and the Prophet's Household (ahl al-bayt). Key studies have already documented the role that the ʿUthmāniyya played in shaping early hostile views on ʿAlī.Footnote 3 Others have noted the popularity of the cult of ʿAlī and his admiration among poets, mystics and soldiers in later periods.Footnote 4 Elsewhere, I have considered the stark contrast between Ibn Taymiyya (d. 728/1328) and his pro-ʿAlid interlocutors who denounced his views.Footnote 5 Although Salafism in the late twentieth century has greatly enhanced Ibn Taymiyya's reputation and prestige, the antagonism that existed between him and his interlocutors seem at times to mark the boundaries of Sunnī Islam. Ibn Taymiyya's antipathy to Shīʿism led him to reject reports about ʿAlī and Fāṭima (d. 11/632) that previous Sunnīs had accepted as authentic.Footnote 6 Over the centuries, various non-Shīʿī scholars have expressed support for the doctrine of tafḍīl ʿAlī (the superiority of ʿAlī to his contemporaries) and for granting the ahl al-bayt a unique status in the community.Footnote 7 But the most influential scholars of ḥadīth in Sunnī Islam tended to maintain a position somewhere in the middle. They were fonder of the ahl al-bayt than Ibn Taymiyya, but their love for ʿAlī did not mean support for tafḍīl ʿAlī. Sunnīs with competing theological commitments, whether to pro-ʿAlid sentiment or anti-Shī‘ī polemics, clearly dealt with the early source material differently.
Pro-ʿAlids consistently accepted and transmitted ḥadīth that exalted ʿAlī, whereas early ʿUthmāniyya and pro-Umayyads viewed him and his followers as a scourge of the community and as the source of sedition. These anti-ʿAlids transmitted ḥadīth that extolled the merits of ʿAlī’s rivals. The narratives of the Kūfan storyteller Sayf b. ʿUmar (d. c. 180/796) reflect a slightly more moderate ʿUthmānī sentiment compared to that which was popular under the Umayyads. In Sayf's stories, ʿAlī is surrounded by criminals, and it is these criminal associates, not ʿAlī himself, who cause civil unrest and misguidance in the community. Sayf does not seem to recognise ʿAlī as a rightly guided caliph, instead portraying him as only one contender among many in a time of social turmoil.Footnote 8 The literary contributions of Sayf and other more temperate ʿUthmānīs nonetheless represent an important shift in the legacy of ʿUthmānī sentiment. In their reports ʿAlī no longer appears as an arch-heretic; rather he is a Companion who found himself in the company of heretics who venerated him, and fell victim to their machinations on numerous occasions. Both early Shīʿī and ʿUthmānī accounts portray ʿAlī as someone who disagreed with his predecessors and rivals on a number of issues.Footnote 9 The more moderate ʿUthmānīs, by contrast, circulated counter reports in which ʿAlī appeared as a loyal partisan of the first three caliphs.Footnote 10 The ʿUthmānīs of the third/ninth century may have appropriated this image of ʿAlī from quietists, centrists and ʿAlī’s partisans who revered him and the first two caliphs together. Some, such as Abū al-Qāsim al-Saqaṭī (d. 406/1015), went further by claiming that ʿAlī and his family members in fact loved Muʿāwiya (d. 60/680). In one report transmitted by al-Saqaṭī, al-Ḥusayn b. ʿAlī (d. 61/680) ascribes to Muʿāwiya the honorifics “scribe of the Qurʼān” and “uncle of the believers” (khāl al-muʼminīn) and asserts that the angel Gabriel had declared Muʿāwiya to be so pious that no true devotee of the Prophet's family could ever speak ill of him.Footnote 11
E.I. Petersen previously examined historiography regarding ʿAlī by comparing the interests and methods of ʿAbbāsid-era storytellers, like Sayf b. ʿUmar, who composed historical chronicles.Footnote 12 This study considers the work of influential scholars in the genre of ḥadīth to complement Petersen's work. In terms of prestige, the most venerated work of ḥadīth in Sunnī Islam would be the The Authentic Collection (Jāmiʿ al-musnad al-ṣaḥīḥ) of Muḥammad b. Ismāʿīl al-Bukhārī (d. 256/869).Footnote 13 Al-Bukhārī’s Ṣaḥīḥ not only provides readers with the doctrines of its compiler, but also the views championed by an emerging group of ḥadīth scholars actively engaged in the formation and maintenance of orthodoxy. While al-Bukhārī is more circumspect in transmitting controversial material regarding ʿAlī, I occasionally contrast him with one of the most celebrated ḥadīth scholars of Baghdad, Aḥmad b. Ḥanbal (d. 241/855). Aḥmad b. Ḥanbal consistently transmits material that provides the audience with greater context and additional commentary from transmitters. Although these two scholars and their students dedicated their lives to the collection and transmission of prophetic ḥadīth, the agency and predilections of each author becomes apparent in a comparative study.
Unlike their Muʿtazilī, Shīʿī and Khārijī interlocutors, these proto-Sunnī scholars of ḥadīth optimistically hoped that all of the Prophet's Companions could be recognised as righteous figures in the literature that they produced.Footnote 14 To achieve this objective, Aḥmad b. Ḥanbal transmitted many reports about the merits of the Companions embroiled in the early conflicts from their partisans.Footnote 15 ʿUthmānī, pro-Umayyad, and pro-ʿAlid ḥadīth all appear in Aḥmad b. Ḥanbal's Musnad. Although each of these factions contribute ḥadīth to al-Bukhārī’s Ṣaḥīḥ, the latter generally refrains from transmitting ḥadīth about ʿAlī’s merits and the history of his caliphate. To explain his position, al-Bukhārī cites the opinion of Ibn Sīrīn (d. 110/729) who considered most reports about ʿAlī to be fabricated.Footnote 16 Al-Bukhārī limits himself to reporting only three unique merits of ʿAlī.Footnote 17 By contrast, Aḥmad b. Ḥanbal preserves and transmits hundreds of ḥadīth in praise of ʿAlī.Footnote 18
Despite their differences in terms of methodology and receptiveness to pro-ʿAlid reports, the two scholars shared a concern for articulating orthodoxy through ḥadīth and their assessments of ḥadīth transmitters. Consequently, they sought to (1) condemn and suppress the legacy of anti-ʿAlid sentiment (naṣb), (2) discredit ḥadīth that undermined the superiority of Abū Bakr and ʿUmar (or explicitly upheld tafḍīl ʿAlī), and (3) appropriate ʿAlī as an innocuous member of the early community. The third objective resulted in these authors’ acceptance of ḥadīth that depicted ʿAlī making mistakes and upsetting the Prophet or other Companions. In one case, ʿAlī refuses the Prophet's invitation to join him in worship,Footnote 19 and, in another, he leads a congregation in prayer while intoxicated.Footnote 20
The compilers of Sunnī ḥadīth literature faced great challenges in sifting through the plethora of conflicting narratives about ʿAlī and reconciling them with their own vision of early Islamic history and what constituted orthodoxy. Although the scholars sought to portray this process of selection as an objective one by relying solely on narrators who were trustworthy and avoiding those who were not, the reality was much more complex. Ḥadīth scholars clearly judged reports by their contents even when they cited problems in the chain of transmission as the principal reason for any negative assessment.Footnote 21 When confronting anti-ʿAlid ḥadīth, they responded in one of at least seven different ways.
Rejection
In a number of cases, ḥadīth scholars rejected an anti-ʿAlid report outright, declaring it a fabrication. For example, claims that ʿAlī tried to physically injure or kill the Prophet or that the Prophet referred to him as the Korah (Qārūn) rather than the Aaron (Hārūn) of the community were systematically excluded from well-known ḥadīth collections.Footnote 22 The transmitter of these claims, Ḥarīz b. ʿUthmān (d. 163/779), was nevertheless considered trustworthy, so other reports that he transmitted appear in the collections of Aḥmad b. Ḥanbal, al-Bukhārī and many others.Footnote 23 The prevalence of anti-ʿAlids like Ḥarīz in the chains of transmission in Sunnī ḥadīth literature is unknown, since biographers usually do not specify a transmitter's exact views on ʿAlī when the transmitter was pro-Umayyad or ʿUthmānī. Geographically, contempt for ʿAlī seems to have been common among ḥadīth transmitters active in the pro-Umayyad Levant and ʿUthmānī Baṣra.Footnote 24 Scholars from these regions generally believed there was no caliph during the tumultuous years in which ʿAlī ruled. It was Muʿāwiya who eventually followed ʿUthmān as the fourth caliph of the community.Footnote 25
Deflection
Scholars deflected accusations that ʿAlī committed serious crimes by acknowledging his culpability for minor sins, including that of keeping bad company. For example, the Marwānids accused ʿAlī of leading the hypocrites (munāfiqūn) in the slander of ʿĀʾisha in the Ifk incident. In narratives circulated on the authority of al-Zuhrī (d. 124/742) and found in the canonical collections of al-Bukhārī and Muslim, ʿAlī appears as an antagonist who does not assume ʿĀʾisha's innocence and encourages the Prophet to divorce her. However, he is not depicted as one of her slanderers.Footnote 26 When a Marwānid asked al-Zuhrī whether ʿAlī was a slanderer, he reportedly answered, “No … but ʿĀʾisha said, ‘He behaved badly in my affair [kāna musīʾan fī amrī]’”.Footnote 27
While the Umayyads claimed that ʿAlī bore direct blame for the assassination of ʿUthmān,Footnote 28 Sunnī scholars tended to shift responsibility to ʿAlī’s close associates.Footnote 29 Some Sunnīs portrayed ʿAlī as unwilling to surrender ʿUthmān's murderers because he was in need of their military and political support.Footnote 30
Likewise, the heretical belief in ʿAlī’s superiority to his predecessors was deflected away from ʿAlī to Ibn Sabaʾ, the legendary heretic in his army. According to this narrative, Ibn Sabaʾ was the real source of tafḍīl ʿAlī. ʿAlī himself strongly condemned this doctrine and punished Ibn Sabaʾ for holding it.Footnote 31 Ibn Sabaʾ came to serve as a figure to whom Sunnīs could attribute all crimes and heresies related to the memory of ʿAlī and the first civil war.Footnote 32 Ibn Sabaʾ was responsible not only for the death of ʿUthmān but also for the Battle of the Camel and the birth of Shī‘ism. Abbas Barzegar explains the significance of such historiography: “Through reliance on stories such as the infiltration of the community by the subversive Jew ʿAbd Allāh b. Sabaʾ, the responsibility for the events of the fitna in Sunni historical traditions are externalised, placed outside the space of the ‘community’ …”.Footnote 33
Instead of accepting narratives in which Companions were responsible for discord and bloodshed, Sunnī heresiography and historiography mostly opted for a conspiracy theory that identified an outsider, a legendary black, Jewish scapegoat as the cause for everything that went wrong in the community.
Recasting: The Curious Case of “Abū Turāb”
In at least one case, ḥadīth transmitters attempted to recast a derisive epithet frequently used by the Umayyads to refer to ʿAlī into an honorific nickname and a sign of distinction. ʿAlī possessed the unique distinction of having fathered the Prophet's descendants and was thus entitled to use the agnomen Abū al-Ḥasan in honour of al-Ḥasan, his eldest son, whose mother was Fāṭima, the daughter of the Prophet. However, it was not in the interests of the Umayyads to remind their audiences of ʿAlī’s close relationship to the Prophet every time that they publicly disparaged or ritually cursed him. Consequently, according to abundant literary evidence in the Sunnī tradition, the Umayyads opted to refer to him as Abū Turāb, ‘the father of dust’.Footnote 34 In letters between ʿUmar II (d. 101/720) and the Byzantine Emperor Leo III (r. 717–741), preserved in Arabic as well as (non-Muslim) Armenian, Aljamiado and Latin sources, Leo only knows ʿAlī by this epithet.Footnote 35 The Byzantine assumption that Abū Turāb was the name of ʿAlī was the result of a practice among leading Umayyads like Muʿāwiya,Footnote 36 Marwān b. al-Ḥakam (d. 65/685),Footnote 37 and al-Ḥajjāj b. Yūsuf (d. 95/714)Footnote 38 to only mention ʿAlī by this nickname in public lectures. Although anti-ʿAlids intended to disparage ʿAlī with this name, by the third/ninth century, Sunnī ḥadīth literature had firmly established a pious narrative in which the Prophet gave ʿAlī the nickname Abū Turāb. Some believed that ʿAlī received the name in the course of a battle,Footnote 39 whereas others said that he obtained it after a disagreement with his wife.Footnote 40 According to the reports that mention the marital dispute, ʿAlī himself considered Abū Turāb to be his most cherished nickname. Shīʿīs followed their Sunnī coreligionists in circulating many ḥadīth that recast Abū Turāb in positive terms.Footnote 41 The apparent agreement between the Sunnī and Shīʿī traditions leaves little room for challenging the shared narrative regarding the origins of the epithet. However, there is reason to believe that the epithet was neither honorific nor commonly used by those who knew or venerated ʿAlī.
Linguistic evidence
According to some lexicographers, variations of an invocation based on the verb ta-ri-ba were used in classical Arabic to damn someone. Examples include taribat yadāk (‘may your hands be soiled’), taribat yamīnuk (‘may your right hand be soiled’), and taribat jabīnuk (‘may your forehead be soiled’).Footnote 42 The invocation taribat yadāh was understood to mean lā aṣāba khayran, ‘may he not find any bounty!’.Footnote 43 Scholars also argued that like other curses, these invocations were used to express condemnation of someone, usually in response to words or deeds that the invoker considered objectionable, but they did not entail a wish for a literal outcome.Footnote 44 The phrases’ literal meaning—‘your hands have become soiled’ or ‘your forehead has become soiled’—conveys the figurative message ‘you have become impoverished’, ‘your mind has become impoverished (and in need of knowledge)’, or ‘you have lost everything (and become impoverished)’.Footnote 45
Evidence from ḥadīth
As some lexicographers noted, taribat yadāk and its variants were commonly used in classical Arabic and even appear in ḥadīth. Sometimes the Prophet is depicted chiding a Companion for saying something wrong or rude.Footnote 46 On another occasion, he gives advice and concludes with a cautionary taribat yadāk.Footnote 47 The commentators understood the Prophet's use of the phrase to mean that ignoring his advice would lead to disastrous consequences.Footnote 48 Finally, when the Prophet reportedly said to someone, “Your forehead has become soiled”, the phrase was interpreted to convey his desire that the addressee repent for his error with abundant prayers and prostration on the ground.Footnote 49
Evidence from the Qurʼān and its exegesis
The Qurʼān refers to turāb (earth, soil, dust) as the fundamental origin of humankind in a number of verses.Footnote 50 The most relevant verse to this discussion is Q90:16, aw miskīnan dhā matraba, or ‘a poor person in dire need’ (lit. covered in dust). Exegetes understood dhū matraba literally as being covered in dust but also figuratively as being in abject poverty and in dire need of assistance.Footnote 51 Al-Ṭabarī provides a long discussion about the various possible interpretations of the phrase.Footnote 52 This verse of the Qurʾān may have referred to a person who had too many children and lived in poverty with them.Footnote 53 Others said the phrase referred to a homeless person who slept outside subjected to the elements and “possessed nothing but the dust that adhered to him.”Footnote 54
The reception of the epithet among ʿAlī’s disciples
A few reports indicate that those who personally knew ʿAlī or lived in Iraq and respected his legacy refrained from using the name Abū Turāb to refer to ʿAlī. In a number of cases, the Umayyads are depicted as calling ʿAlī by this name to the confusion of ʿAlī’s associates. In these cases, when the non-Umayyad interlocutor realises that the Umayyads are referring to ʿAlī, he frequently interprets the epithet as demeaning to ʿAlī. For example, in reports about the execution of ʿAlī’s companion Ṣayfī b. Faṣīl (d. 51/671) one finds the following exchange involving Ṣayfī and the Umayyad governor of Kūfa, Ziyād b. Abīh:
Another report links the epithet to the Umayyad practice of cursing ʿAlī. In it, a man comes to Sahl b. Sa‘d and says, “So-and-so, the governor of Medina, yadʿu ʿAlīyan from the pulpit”. Sahl asks, “What does he say?” The man replies, “He says Abū Turāb …”.Footnote 56 And in a third report, a Kūfan is brought before the Umayyad prince Muḥammad b. Hishām, who asks him whether or not he is a follower of Abū Turāb. The man responds, “Who is Abū Turāb?” The prince says, “ʿAlī b. Abī Ṭālib”. The man responds, “Do you mean the cousin of God's messenger and the husband of his daughter Fāṭima? The father of al-Ḥasan and al-Ḥusayn?”.Footnote 57 Likewise, when al-Ḥajjāj requested that al-Ḥasan al-Baṣrī share his opinion of Abū Turāb, al-Ḥasan asked for clarification: “Do you mean ʿAlī?”.Footnote 58 All of these anecdotes suggest that Abū Turāb was an Umayyad epithet that Muslims who venerated ʿAlī never used. It is true, according to both Sunnī and Shī‘ī ḥadīth, that the Prophet gave ʿAlī the nickname Abū Turāb. In Sunnī ḥadīth, the Prophet called him by this name jokingly upon finding him sleeping on the ground and covered in dust.
However, the Umayyads applied the epithet disparagingly throughout their reign. It is unclear why they chose this particular nickname. Perhaps they were aware of the version of the name's origin story that describes ʿAlī and Fāṭima experiencing marital strife, as a result of which ʿAlī leaves their home and sleeps in the mosque. It is there that the Prophet finds him covered in dust and calls him Abū Turāb.Footnote 59 The Umayyads may thus have used the name to highlight alleged unhappiness in ʿAlī’s and Fāṭima's marriage. The story could also be read to show the Prophet as giving ʿAlī the name Abū Turāb in dismay. In this case, the story would fall under a genre of anti-ʿAlid ḥadīth that were used to portray ʿAlī as a bad husband to Fāṭima. Another example of this genre is the famous report, narrated by al-Bukhārī and others, in which the Prophet allegedly censures ʿAlī for upsetting him and Fāṭima by considering the daughter of Abū Jahl as a second wife.Footnote 60 According to some accounts, the Prophet goes on to praise the fidelity of another son-in-law, Abū ’l-ʿĀṣ b. al-Rabīʿ, who shared close kinship ties with the Umayyads.Footnote 61 In contrast to ʿAlī, this cousin of the Umayyads is described as a devoted husband. Thus, the topos of ʿAlī as a bad son-in-law that appears elsewhere in the ḥadīth literature may have something to do with the Umayyad use of Abū Turāb.
If Abū Turāb is tied to the phrases taribat yadāh or dhā matraba, then the Umayyads used the epithet to deride ʿAlī’s appearance and to imply that he looked dirty and homeless. Whereas the Umayyads possessed great wealth and distributed it to their partisans, the figure of Abū Turāb was one of a pretender to the caliphate who commanded no such wealth. Pro-ʿAlid texts interpreted ʿAlī’s modest means as a consequence of his principled refusal to use public funds to enrich himself or the aristocracy of his society,Footnote 62 but the Umayyads may have cast his poverty as a sign of failure or weakness. The epithet may also have referred to the fact that ʿAlī had many children but remained extremely poor, or it may have alluded to the wrathful invocation taribat yadāh as appropriate for someone who, in the Umayyads’ view, had caused great misfortune with his many errors. Therefore, it is possible that the epithet Abū Turāb began as an anti-ʿAlid aspersion on ʿAlī but was subsequently recast and accepted as a merit.
Erasure
Scholars of ḥadīth occasionally felt compelled to delete components of a report that were offensive to their sensibilities. In particular, ḥadīth explicitly denigrating ʿAlī could not continue to circulate intact after the Umayyad period, since the ʿUthmāniyya gradually came to accept him as the fourth caliph. This development rendered problematic ḥadīth such as that transmitted by Abū Bakr b. al-ʿArabī (d. 543/1148) and Ibn Abī al-Ḥadīd (d. 656/1258) on the authority of al-Bukhārī, in which the Prophet declares, “The family of Abū Ṭālib are no allies [awliyāʾ] of mine”.Footnote 63 By the Mamluk period, extant copies of al-Bukhārī’s Ṣaḥīḥ no longer identified the family of Abū Ṭālib as the rejected clan mentioned in the report,Footnote 64 but Ibn Ḥajar al-ʿAsqalānī (d. 852/1449), in his assessment of the report, conceded that the report had indeed originally named Abū Ṭālib's family: he had found a variant of the report in Abū Nuʿaym al-Iṣbahānī’s (d. 430/1038) Mustakhraj of al-Bukhārī’s text that did not omit the family's name.Footnote 65 The following reports reflect the transmission of this report in the canonical ḥadīth collections and their commentaries:
Aḥmad b. Ḥanbal, al-Bukhārī, and Muslim all narrate from Muḥammad b. Ja‘far Ghundar [active in Baṣra, d. 193/809], who narrates from Shu‘ba [Kūfa and Baṣra, d. 160/777], who narrates from Ismā‘īl b. Abī Khālid [Kūfa, d. 146/763], who narrates from Qays b. Abī Ḥāzim al-Aḥmasī [Kūfa, d. ca. 98/717], who narrates from ʿAmr b. al-ʿĀṣ [d. ca. 43/663] that the Prophet announced openly, not privately, “The family of Abū so-and-so are no allies of mine. Rather, God and the righteous among the faithful are my allies …”.Footnote 66
Al-Bukhārī’s direct informant ʿAmr b. ʿAbbās (active in Baṣra, d. 235/849) notes that “there is a blank space [bayāḍ] in the book of Muḥammad b. Ja‘far [Ghundar]”.Footnote 67
Al-Bukhārī adds, on the authority of the Umayyad ʿAnbasa b. ʿAbd al-Wāḥid [active in Kūfa, fl. early third/ninth century], who narrates from Bayān b. Bishr al-Aḥmasī [Kūfa, fl. second/eighth century], who narrates from Qays b. Abī Ḥāzim al-Aḥmasī, who narrates from ʿAmr that the Prophet continued, “but they have kinship ties that I will honour”.Footnote 68
Ibn Ḥajar transmits a report from al-Bukhārī as “The descendants of Abū _____ are no allies of mine …”.Footnote 69
Al-Bukhārī’s first report of this statement, transmitted by ʿAmr b. ʿAbbās, seems to have circulated in anti-ʿAlid Baṣra from at least the middle of the second/eighth century. Al-Bukhārī’s second report comes from an Umayyad informant who narrates the ḥadīth on the authority of two transmitters belonging to the Aḥmasī clan in Kūfa. The chain of transmission seems incomplete since only one person in it, Bayān, was active in the second/eighth century. Ibn Ḥajar and Badr al-Dīn al-ʿAynī (d. 855/1451) note in their respective commentaries on the Baṣran text that some copyists mistook a note about the deletion or blank space (bayāḍ) in the manuscript to stand for the name of a tribe, incorrectly reading the text to refer to the family of an “Abū Bayāḍ”.Footnote 70
Al-Bukhārī’s ḥadīth appeared in three different forms, reflecting the varying sensibilities of its narrators. First, the earliest narrators transmitted the report in its complete form, explicitly naming the family of Abū Ṭālib (Text A). Sunnī ḥadīth scholars pointed to Qays b. Abī Ḥāzim and the Umayyad ʿAnbasa b. ʿAbd al-Wāḥid, both part of the report's chain of transmission, as anti-ʿAlids who might have fabricated the report.Footnote 71 Pro-ʿAlids, meanwhile, identified ʿAmr b. al-ʿĀṣ, a close confidant of Muʿāwiya, as the culprit.Footnote 72 ʿAmr is depicted as instrumental to Muʿāwiya's political victories first as a rebel against ʿAlī and al-Ḥasan b. ʿAlī and finally as an Umayyad governor. ʿAlī reportedly denounced ʿAmr as sinful on repeated occasions and prayed for his punishment in supplications (qunūt) that he made in daily worship.Footnote 73 Most Sunnīs did not follow suit in censuring ʿAmr, since he was a Companion of the Prophet. However, some prominent Sunnīs, such as al-Nasāʾī (d. 303/915) and Abū al-Fidāʾ (d. 732/1331), refrained from venerating him because of his opposition to ʿAlī.Footnote 74 At least in the Umayyad period, transmitters generally identified Abū Ṭālib's family as the subject of the ḥadīth. However, scholars who read al-Bukhārī’s Ṣaḥīḥ report that they frequently found Abū Ṭālib's name deleted from the report. Since al-Bukhārī himself states that his informant found the clan's name omitted in his source, it is clear that deletions began to occur at least one generation before al-Bukhārī, though the precise point in time when copies of al-Bukhārī’s Ṣaḥīḥ gained or lost the name cannot be pinned down. Extant copies of the work no longer contain Abū Ṭālib's name in full.
The testimonyFootnote 75 of al-Bukhārī’s informant suggests that Ghundar's book of ḥadīth once carried Abū Ṭālib's name in full, but either Ghundar or a copyist of his book deleted the second part of the name, leaving the ‘Abū’ intact (Text B). The person responsible for the deletion probably considered the report anti-ʿAlid in tone and offensive to the Ṭālibids (the descendants of ʿAlī, ʿAqīl and Ja‘far b. Abī Ṭālib). As the Prophet's kinsfolk, the Ṭālibids possessed great social capital in early Islamic history, to the point that they threatened ʿAbbāsid claims to power.Footnote 76 Transmitters who sought to convey the report's lesson that allegiance to the faith should trump family ties, but had qualms about its anti-Ṭālibid tone, transmitted the text with either a lacuna or the anonymous ‘Abū so-and-so’ (fulān). Neither Abū Bakr b. al-ʿArabī nor Ibn Ḥajar had problems in accepting the authenticity of the report. The latter reasoned that the ḥadīth cut ties only between the Prophet and non-Muslim Ṭālibids.Footnote 77 As previously mentioned, pro-ʿAlids such as Ibn Abī al-Ḥadīd considered the report an Umayyad fabrication.
Finally, in its third form (Text C), the report refers to the family of Abū Bayāḍ (‘Father of blank space’). This version arose, as described earlier, from copyists’ misreading of notes left in the text to indicate a lacuna (bayāḍ) after the word ‘Abū’ and their conclusion that the Prophet spoke of a clan named Abū Bayāḍ. The three forms of the ḥadīth appeared can be summarised thus:
Emendation
Copyists and scholars emended ḥadīth that they considered objectionable in at least three ways: by obscuring the identity of a Companion, by omitting reported speech or by emending key words. As for the first method, if a ḥadīth seemed to depict a Companion in a negative light, his identity might be obscured. For example, in the ḥadīth discussed in the previous section, the clan of Abū Ṭālib became “Abū so-and-so”. In another case, ʿUmar b. al-Khaṭṭāb (d. 23/644) curses Samura b. Jundab (d. 60/680) for selling intoxicants.Footnote 79 In the recension of al-Bukhārī, Samura's name is omitted and the report consistently refers to him as “so-and-so” (fulān), giving no indication that the person selling intoxicants had been a Companion.Footnote 80 In a few anti-Umayyad ḥadīth in which the Prophet allegedly condemns Muʿāwiya as evil, Muʿāwiya's name is also replaced with “so-and-so”.Footnote 81
In another case, Aḥmad b. Ḥanbal and Ibn ʿAsākir (d. 571/1176) note that when Marwān b. al-Ḥakam became the governor of Medina, he would censure and ritually curse ʿAlī every Friday.Footnote 82 The name of this Umayyad governor (and future caliph) was omitted in other recensions where he commands others to join him in cursing ʿAlī.Footnote 83 Al-Bukhārī included a heavily redacted version of the report in his Ṣaḥīḥ.Footnote 84 In al-Bukhārī’s report, the name of the governor and his demand that others curse ʿAlī are omitted. However, when a witness complains of the incident to a revered Companion, Sahl b. Saʿd al-Anṣārī (d. c. 91/710), Sahl consoles the person by explaining why one should not be offended by the governor's custom of referring to ʿAlī as Abū Turāb. Al-Bukhārī’s redacted report is indicative of how denigrating statements in one version may be omitted elsewhere. Unlike the reports of Aḥmad b. Ḥanbal and Ibn ʿAsākir, which quoted the words of the governor directly, the governor's speech is fully excised in al-Bukhārī’s account. Instead, a witness mentions only that the governor referred to ʿAlī as Abū Turāb. The report is sanitised of its anti-ʿAlid historical context.
There are reports in which Muʿāwiya and Marwān censure and curse ʿAlī from the pulpits and in the presence of ʿAlī’s own sons.Footnote 85 This led ʿAlids to avoid attending certain worship services at the mosque. Al-Ḥasan b. ʿAlī, for example, would wait until Marwān was finished with his Friday sermon before entering the mosque to join the community in congregational worship.Footnote 86 Marwān also alienated the family of Abū Bakr when he publicly censured ʿAbd al-Raḥmān b. Abī Bakr (d. 53/673) for refusing to recognise Yazīd as Muʿāwiya's heir apparent. As Marwān delivered his speech on the merits of Muʿāwiya's son, ʿAbd al-Raḥmān angrily replied that the caliphate was not an institution governed by dynastic succession like the kingdoms of Rome and Persia.Footnote 87 Marwān responded by claiming that Q46:17 was revealed as a critique of ʿAbd al-Raḥmān's insolence and unruliness as a young man. When ʿAbd al-Raḥmān angrily responded that such a claim was slander, ʿĀʾisha comes to the aid of her brother and corroborates him, arguing that the verse had nothing to do with him. When ʿAbd al-Raḥmān reminds the congregation that the Prophet once cursed (and exiled) Marwān's father, Marwān orders his arrest and ʿAbd al-Raḥmān flees from the mosque to avoid capture.Footnote 88 Al-Bukhārī omits all of ʿAbd al-Raḥmān's fiery words to Marwān. In al-Bukhārī’s version of the ḥadīth, his likening the Umayyads to Roman and Persian rulers and reference to the Prophet's curse is replaced with a vague note that ʿAbd al-Raḥmān “said some things to him”.Footnote 89
Al-Bukhārī’s proclivity for transmitting reports in which objectionable material is omitted can also be seen in cases where ʿAlī appears too Shīʿī for a Sunnī audience. Multiple sources, including Muslim's (d. 261/875) Ṣaḥīḥ and ʿAbd al-Razzāq al-Ṣanʿānī’s (d. 211/827) Muṣannaf, narrate ʿUmar's dismay that ʿAlī and ʿAbbās considered Abū Bakr and himself unjust (ẓālim) or sinful (āthim) in their decision to disinherit Hāshimids from the Prophet and convert the latter's estates into public endowments.Footnote 90 These two prominent Hāshimids are portrayed as holding opinions of the first two caliphs that would be considered quite offensive, Shīʿī and incendiary to a Sunnī audience. Al-Bukhārī transmits versions of the report in which the views of ʿAlī and ʿAbbās are not explicitly stated; rather, it is vaguely noted that they used to claim “this and that” (kadhā wa-kadhā) about Abū Bakr.Footnote 91 In another recension of al-Bukhārī, neither ʿAlī nor ʿAbbās voice any objection to the ruling of Abū Bakr and ʿUmar. Their offensive views regarding the caliphs are completely omitted.Footnote 92
In some cases, ḥadīth may have been emended so that negative words about a Companion were transformed into positive ones. ʿAlī benefited from this third type of emendation when early transmitters reported that ʿĀʾisha criticised ʿAlī’s conduct in the Ifk incident. She reportedly said, “He behaved badly in my affair”.Footnote 93 Some transmitters changed kāna musīʾan to kāna musallaman, with the effect that ʿĀʾisha now praised ʿAlī as free (musallam) of any wrongdoing in the matter.Footnote 94 Consequently, depending on the version they received and their own sensibilities, scholars taught al-Bukhārī’s Ṣaḥīḥ with either version of the text. Published editions of al-Bukhārī’s work contain the positive musallam, but many scholars in the Mamlūk period still possessed copies in which ʿAlī was censured as musīʾ. The rehabilitation of ʿAlī played an important role in the gradual shift in the interpretation of this report. In the Umayyad period, an ʿUthmānī such as al-Zuhrī had no qualms in saying that ʿAlī had treated ʿĀʾisha unfairly in the Ifk incident, but centuries later, after ʿAlī’s retroactive acceptance as an ʿUthmānī, it was unthinkable to acknowledge that he had ever been portrayed as an antagonist of Abū Bakr, ʿUmar, or ʿĀʾisha. Thus, later Sunnīs took for granted that ʿĀʾisha described ʿAlī as musallam, not musīʾ, in the Ifk incident.
Circulation of Counter reports
Sunnī ḥadīth collections included contributions from ʿAlī’s partisans and detractors alike in their construction of an image of ʿAlī that was neither entirely evil nor fully pure. Rather, ʿAlī appeared as a normal human being, subject to the same challenges and temptations as everyone else. This balancing effect may not have been coincidental: the content of certain reports suggests that when ʿAlī’s detractors encountered a ḥadīth about his merits, they would narrate a counter report to contradict it.
As noted earlier, some anti-ʿAlids (nawāṣib), such as Ḥarīz b. ʿUthmān, cited Marwānids as their authorities for emending a famous ḥadīth that described ʿAlī as the Hārūn (Aaron) of the community; in the emended version, he became its Qārūn (Korah, who rebelled against Moses). In this case, it is clear that anti-ʿAlids were engaged in circulating a report that contradicted a well-known merit of ʿAlī. Other examples of counter reports are slightly more subtle.Footnote 95 ʿAlī’s partisans often portrayed him as a saint who worshipped God abundantly and greatly resembled the Prophet in his habits of worship.Footnote 96 But according to other reports, ʿAlī led prayers while intoxicated in the lifetime of the Prophet and in a state of major ritual impurity as caliph.Footnote 97 Al-Bukhārī and others narrate a report in which ʿAlī annoys the Prophet by declining his invitation to join him in prayer.Footnote 98 These reports appear to contradict the image of ʿAlī as a devout worshipper and support the Umayyad image of ʿAlī as someone who did not engage in daily worship.Footnote 99
When ʿAlī married Fāṭima, the Prophet reportedly congratulated him for having been selected by God to marry the Prophet's daughter.Footnote 100 Marriage to the Prophet's daughter was undoubtedly a great honour and an indication of ʿAlī’s stature in the Prophet's eyes. Since Fāṭima was considered a woman of great piety, uniquely honoured by God and her father, she required a spouse of equal calibre. Thus, some pro-ʿAlid ḥadīth assert that had it not been for ʿAlī, Fāṭima would never have found a suitable partner.Footnote 101 However, as the examples discussed earlier in this article indicate, there were counter reports that depicted ʿAlī as a bad husband to Fāṭima.
According to some ḥadīth, the Prophet commanded everyone in his community to close their private entrances to his mosque.Footnote 102 The only exception was granted to ʿAlī, Fāṭima and their two sons, who could enter the mosque through their private entrance at any time, even in a state of major ritual impurity (janāba).Footnote 103 Pro-ʿAlids and Shī‘īs understood these reports as further confirmation of the exceptional purity of the Prophet's household. The dispensation also offered a practical benefit: it allowed the family easy access to the Prophet's home. They could pass through the mosque even in a state of major ritual impurity without angering God or His Prophet.
As Hossein Modarressi has pointed out, the same merits that were ascribed to ʿAlī in pro-ʿAlid circles were also ascribed to the first three caliphs in ʿUthmānī circles.Footnote 104 Thus, in the Ṣaḥīḥ collections of al-Bukhārī and Muslim, the permission given to ʿAlī and Fāṭima to keep their entrance to the Prophet's mosque open was given to Abū Bakr instead.Footnote 105 The ḥadīth granting Abū Bakr the same privilege may thus be viewed as a counter report to the ḥadīth about ʿAlī. In addition, ʿUthmānīs further narrated ḥadīth that portrayed ʿAlī as afflicted with frequent seminal discharge (madhy).Footnote 106 Reports about this malady may be understood as ʿUthmānī explanations for the dispensation that he received to enter the Prophet's mosque even in a state of ritual impurity.
The Principle of Charity
An ideological commitment to belief in the righteousness of all Companions led many scholars either to reject or to charitably interpret texts that seemed to present Companions in a negative light. Ibn Ḥazm (d. 456/1064), for example, argues that the man who killed the Prophet's revered Companion ʿAmmār b. Yāsir should receive a reward from God for his deed.Footnote 107 Ibn Ḥazm explains that the killer Abū ’l-Ghādiya had also been a Companion and, thus, his deed should be charitably understood as the error of an expert (mujtahid) engaged in religious hermeneutics. Elsewhere, I have discussed the reception of canonical reports that depict ʿAlī as delaying his pledge of allegiance to Abū Bakr.Footnote 108 These texts were reinterpreted to deny that ʿAlī ever questioned the first caliph's preeminence or challenged his candidacy. Accounts in which ʿAlī explicitly complains about the succession of his predecessors never entered the canon and were largely rejected as forged.Footnote 109
Both Muʿāwiya and ʿAlī benefitted from the principle of charity and the tendency to defend all Companions as righteous. Influential scholars such as al-Nawawī (d. 676/1277) proposed charitable reinterpretations for canonical ḥadīth that appear to show Muʿāwiya cursing ʿAlī or encouraging others to do so, while other scholars rejected such texts altogether.Footnote 110 The ḥadīth discussed above, “The family of Abū Ṭālib are no allies of mine”, is another example: the text was read to refer hypothetically to non-Muslims in ʿAlī’s family.Footnote 111 Such generous interpretations were irrelevant to early ʿUthmānīs and pro-Umayyads who never recognised ʿAlī and his descendants as Muslims but rather condemned them as apostates and evil criminals. Consequently, charitable interpretations of the ḥadīth came to play a key role in safeguarding the honour of ʿAlī and his sons after their rehabilitation in Sunnism.
From Three Caliphs to Four
The early ʿUthmāniyya supported not only the caliphate of the first three caliphs but also the insurrection of ʿĀʾisha, Ṭalḥa and Zubayr against ʿAlī. ʿUthmānīs such as Wurayza b. Muḥammad al-Ḥimṣī (d. 281/894) reportedly refused to recognise ʿAlī as a legitimate caliph because they believed that such recognition would necessarily entail opposition to and censure of the leaders who fought ʿAlī at the Battle of the Camel.Footnote 112 The ʿUthmānī shift to accepting ʿAlī as a legitimate caliph probably began in Kūfa and Baghdad. Scott Lucas has argued for the possibility that early theologians who were Zaydīs or Baghdādī Mu‘tazilīs “contributed to the profound respect for ʿAlī and his family found in the Musnad of Ibn Ḥanbal, Muṣannaf of Ibn Abī Shayba, and Ṣaḥīḥ of Muslim that seems stronger than the fourth-place status accorded him by [later] Sunnī doctrine”.Footnote 113 It should be added that al-Maʾmūn (r. 198–218/813–33) played a key role in initiating public debates about the place of ʿAlī in Islamic history by proclaiming tafḍīl ʿAlī to be orthodoxy in 211/826 and once more the following year.Footnote 114 The caliph invited ḥadīth scholars and Muʿtazilīs who opposed tafḍīl ʿAlī to debate the issue with him in his court.Footnote 115 Al-Maʾmūn undoubtedly encouraged al-Jāḥiẓ (d. 255/869), Abū Ja‘far al-Iskāfī (d. 240/854) and other theologians to discuss the issue of tafḍīl ʿAlī in their literary work. During the reign of al-Maʾmūn and in the years that followed, these scholars carefully considered evidence indicating ʿAlī’s distinguished status. The same can be said about Aḥmad b. Ḥanbal who made the conscious decision to locate and preserve hundreds of Kūfan ḥadīth about the merits of ʿAlī. All of these figures also resided in Baghdad, where they encountered each other's opinions. Aḥmad b. Ḥanbal may have accepted ḥadīth about ʿAlī’s merits from pro-ʿAlid transmitters in Baghdad after conceding to the arguments of pro-ʿAlid theologians in the city. For example, probably to the dismay of the city's ʿUthmānīs, he reportedly agreed with proponents of tafḍīl ʿAlī that no Companion possessed more merits than ʿAlī.Footnote 116 Aḥmad b. Ḥanbal's decision to transmit hundreds of anecdotes in which the Prophet singles out ʿAlī for praise bears witness to his assessment.Footnote 117 He also reportedly began arguing for the need to accept ʿAlī as a legitimate fourth caliph among his ʿUthmānī peers.Footnote 118 Such advocacy would have involved some acceptance of the historical narratives of ʿAlī’s partisans. Although Aḥmad b. Ḥanbal was not an outright proponent of tafḍīl ʿAlī, his acceptance of pro-ʿAlid ḥadīth led him to transmit reports associated with this doctrine.Footnote 119
Sunnī scholars transmitted reports that explicitly articulated the merits of Companions both generally and specifically, but Lucas suggests that the most enduring achievement of Aḥmad b. Ḥanbal was an implicit polemic: the vindication of all Companions who participated in the civil wars that engulfed the community after the Prophet's death. By including them as important sources of ḥadīth in his Musnad, Aḥmad b. Ḥanbal acquitted these personalities of charges of impiety.Footnote 120 The inclusion of Companions who fought against ʿAlī in the Musnad indicated that despite reports that cast their political careers in a negative light and despite the criticisms levied against them by pro-ʿAlid theologians, these Companions were nonetheless trustworthy sources of information about the life of the Prophet and his teachings. By the middle of the third/ninth century, ʿAlī had also come to benefit from an emerging Sunnī orthodoxy that used the hermeneutical tools described in this article to delegitimise hostile depictions of him and appropriate him as the fourth caliph, extending the three-caliph model of the early ʿUthmāniyya. As others have noted, this acceptance of ʿAlī’s fourth place (tarbī‘ ʿAlī) was an innovation for the ʿUthmāniyya of the third/ninth century.Footnote 121
Conclusions
The image of ʿAlī b. Abī Ṭālib that appeared in Sunnī ḥadīth collections produced after the beginning of the third/ninth century was as complex and composite as the compilers’ sources. Anti-ʿAlids viewed ʿAlī and his family with contempt, whereas to many pro-ʿAlids he was the most meritorious Muslim after the Prophet. A third group consisted of those who were ambivalent about ʿAlī’s personality and considered him a Companion no different from his peers. For example, Ibn Taymiyya argued that ʿAlī possessed merits but also many shortcomings.Footnote 122 He forcefully argued that ʿAlī upset the Prophet and later waged war unnecessarily against his rivals.Footnote 123 Thus, ʿAlī was responsible for civil strife in the community, though he was not evil.
Whereas pro-ʿAlids remembered ʿAlī as someone who exercised independent judgment after the Prophet, later orthodoxy frequently portrayed him as agreeing with the positions of other authorities. ʿAlī’s variant opinions on political and religious questions were gradually replaced in reports about him with answers that affirmed Sunnī orthodoxy.
Various caliphs, from Muʿāwiya to al-Maʾmūn, were clearly invested in shaping public perceptions about ʿAlī. While the Umayyads supported the circulation of tales that maligned him, al-Maʾmūn appears to have spurred ʿAlī’s rehabilitation in the community. The case studies in this article indicate the ways in which Sunnī scholars made use of their editorial privilege to reshape ʿAlī’s image: they transmitted selected versions of reports that omitted what they saw as controversial material and obfuscated certain sensitive elements of the narratives that they transmitted. In some cases, individuals resorted to outright deletion of particularly inflammatory words or passages when they were obliged to transmit such material. It is unclear to what extent copyists contributed to this revisionary process.
