The problematic portrayal of the Qarmaṭīs of Iran

The dearth of sources on the Qarmaṭīs has raised key conceptual problems, such as how to determine whether individuals or groups may be described as Qarmaṭī, as well as how to differentiate between the Qarmaṭī or the Fāṭimids in the sources, since both were Ismāʿīlī. Whether individuals or groups may or may not be defined as Qarmaṭī in this article is contingent on whether they believed in Muḥammad b. Ismāʿīl as the Mahdī and followed Ḥamdān Qarmaṭ, the leader of the dāʿīs. The Ismāʿīlī community in Iraq acknowledged Ḥamdān Qarmaṭ as “dāʿī al-dāʿīs” and Muḥammad b. Ismāʿīl as the true Mahdī, and the Yemeni and Iranian Ismāʿīlī followers acknowledged and followed the Qarmaṭīs of Baḥrayn. We may also consider that some of the Ismāʿīlīs in Iran were affiliated to the Qarmaṭīs on account of their sharing reciprocal concepts as well as the fact that there was a direct correspondence between Abū Ḥātim al-Rāzī (d. 322/934), an outstanding Iranian Ismāʿīlī scholar, and Abū Ṭāhir al-Jannābī (r. 311–332/924–944), the Qarmaṭīs’ leader in Baḥrayn.Footnote ¹⁰

Distinguishing between what is Qarmaṭī or Fāṭimid in the different Ismāʿīlī societies of Iraq, Syria, Baḥrayn, Yemen or Iran is complicated. Ismāʿīlīsm was a single system of belief that remained unchanged during the period before the Fāṭimid state (297–567/909–1172). Nevertheless, this problem of the difference between what is Qarmaṭī or Fāṭimid can be traced to the split of the daʿwa c. 286/899. It is here that we first see the emergence of two different Ismāʿīlī beliefs as represented by the Qarmaṭīs and the Fāṭimids.Footnote ¹¹ Being a Qarmaṭī meant denying the legitimacy of the Fāṭimid imams-caliphs and advocating Muḥammad b. Ismāʿīl as the Mahdī, while other Ismāʿīlīs who acknowledged the Fāṭimid imamate were described as Fāṭimids. This is according to al-Qummī (d. 301/914) and al-Nawbakhtī (d. c. 311/924), who referred to Qarmaṭīs as those in Iraq and Syria who believed in Muḥammad b. Ismāʿīl as the Mahdī and followed Ḥamdān Qarmaṭ’s daʿwa.Footnote ¹² This raises an important question: to what extent were the sources able to recognize what is genuinely Qarmaṭī within the community of north-west Iran?

The most detailed sources that informed the accounts of the Ismāʿīlīs’ (including the Qarmaṭīs) presence in areas of north-west Iran were written by Ibn al-Nadīm (d. 380/990), al-Thaʿālibī (d. 429/1038), Niẓām al-Mulk (d. 485/1092) and others. It is from the most detailed of these authors’ accounts that we have the first written evidence for Qarmaṭīs in this region. Furthermore, all of these sources were able to distinguish between the different identities of Qarmaṭī and Fāṭimid.

The Ismāʿīlīs (including the Qarmaṭīs) in north-west Iran were first described in Ibn al-Nadīm’s al-Fihrist, which was written in approximately 377/987, according to its editor Ayman Fuʾād Sayyid.Footnote ¹³ He provides a valuable selection of material relating to the history of their settlement, including a number of accounts not recorded elsewhere. Ibn al-Nadīm also mentions the titles of early Ismāʿīlī works not known through other sources. One reason Ibn al-Nadīm was able to do this, I would argue, is that, on the one hand, he had direct access to eminent members of the Ismāʿīlī daʿwa, for he provides specific details on the Ismāʿīlīs operating in Iraq and Iran during his own lifetime. Ibn al-Nadīm’s prominent source for the Ismāʿīlī daʿwa was anti-Ismāʿīlī writer Ibn Rizām (d. late fourth/tenth century), a Sunnī polemicist who lived in Baghdad during the first half of the fourth/tenth century.Footnote ¹⁴ Ibn Rizām’s Kitāb radd ʿalāʾl-Ismāʿīliyya, which has not survived, is quoted in al-Fihrist, giving a variety of details relating to the Qarmaṭīs and their dāʿīs.Footnote ¹⁵

Al-Fihrist has been studied by Samuel Stern, Wilferd Madelung and other scholars as an essential source for the early history of the Ismāʿīlī daʿwa. It provides valuable information on the dāʿīs’s activities in Iraq and western Iran.Footnote ¹⁶ Moreover, Ibn al-Nadīm describes in some detail the works of ʿAbdān (Ḥamdān Qarmaṭ’s brother-in-law) (d. c. 286/899), a leader of the Ismāʿīlī movement in Iraq, namely al-Nīrān (the Fires), al-Lāmiʿ (the Illustrious) and others, all of which have been lost, but their titles suggest they contained aspects of philosophy and metaphysics.Footnote ¹⁷ Both Stern and Madelung observe that Ibn al-Nadīm had significant access to information about the Ismāʿīlīs. However, they did not examine al-Fihrist critically in terms of scrutinizing the people from whom Ibn al-Nadīm drew his information; rather, they commented only that he knew Ismāʿīlīs, including several dāʿīs, and had read Ibn Rizām’s Kitāb radd ʿalāʾl-Ismāʿīliyya.Footnote ¹⁸ The few studies that directly address Ibn al-Nadīm’s Ismāʿīlī contacts misinterpret the relevant passages of al-Fihrist and are limited in their scope. Devin Stewart has revised these earlier studies, including Stern and Madelung’s discussions of early Ismāʿīlīs, by examining other prominent dāʿīs who had possibly been Ibn al-Nadīm’s sources for the Ismāʿīlīs in Iran, namely Ibn Ḥamdān, Ibn Nafīs and al-Ḥasanābādhī, who were active dāʿīs during the mid-fourth/tenth century.Footnote ¹⁹

Second, Ādāb al-mulūk, by the prominent writer al-Thaʿālibī (d. 429/1038), contains many less-discussed texts relating to the Ismāʿīlīs in Iran during the third/ninth and fourth/tenth centuries.Footnote ²⁰ Julia Bray remarks on Ādāb al-mulūk as follows: “Ādāb al-mulūk was written for an Iranian prince and his entourage, on the eastern edge of the Muslim world, but in Arabic, even though Arabic was not the addressees’ mother tongue [there].”Footnote ²¹ Ādāb al-mulūk was dedicated to the Sāmānid dynasty (204–395/819–1005), an Iranian dynasty centred in Khurāsān (in north-east Iran) and Transoxiana (today Central Asia) that held control of Iran and some of the eastern lands towards India, where they initially acted as a vassal to the ʿAbbāsid Caliphate (132–656/750–1258).Footnote ²² Nevertheless, they later used the title of amīr (commander), since they were independent rulers whose regime was legitimized but not constrained by the ʿAbbāsids.Footnote ²³

During the reign of the Sāmānids, the Ismāʿīlīs were active in Iranian lands.Footnote ²⁴ The Sāmānid state initially accommodated the daʿwa but this policy eventually changed to hostility and ultimately to the persecution of all Ismāʿīlīs.Footnote ²⁵ Al-Thaʿālibī’s account tells of the corruption of the Sāmānid amīr, Naṣr b. Aḥmad (r. 301–331/914–943), by Ismāʿīlī dāʿīs, highly trained intellectuals peddling doctrines designed to undermine allegiance to the ʿAbbāsids.Footnote ²⁶ One aspect of the historical content of Ādāb al-mulūk relating to the Ismāʿīlīs has been discussed by Patricia Crone and Luke Treadwell.Footnote ²⁷ Crone and Treadwell do not, however, explain why al-Thaʿālibī includes this substantial chunk of recent history in the literary style of Naṣīḥat al-mulūk (Advice for Kings), otherwise referred to as “mirrors for princes”. The Ādāb al-mulūk was written in a style that is generally characterized by an assimilation of Islamic political and ethical norms to traditions of kingship drawn from ancient, mostly Persian, sources.Footnote ²⁸ Bray remarks that the Sāmānid state had been “the greatest power in the east” and this is one reason why al-Thaʿālibī often points to their example (both good and bad) as rulers.Footnote ²⁹ Further, according to Bray, another reason why al-Thaʿālibī used the Sāmānids as an example in his writings was that al-Thaʿālibī’s grandfather, named only as Abū ʿAlī, seems to have had some link with the Sāmānid court.Footnote ³⁰

Third, a near-contemporary source, the Seljuk vizier Niẓām al-Mulk (d. 485/1092), categorized the Qarmaṭīs (specifically those in Iran) as bāṭinīs (a term referring to the esoteric nature of their belief, which was synonymous with being heretics, according to Sunnīsm), thereby accusing them of heresy.Footnote ³¹ According to Niẓām al-Mulk, those regarded as bāṭinīs in every city and province have been recognized by different titles. For example, those in Baghdad, Transoxiana and Ghaznin are identified with the name Qarmaṭīs and those in Baḥrayn are called Jannābīs, which refers to their founder Abū Saʿīd al-Jannābī.Footnote ³² Niẓām al-Mulk’s accusation was used later by al-Ghazālī (d. 505/1111) and Ibn al-Jawzī (d. 597/1200), who defined the Qarmaṭīs in general as bāṭinīs.

Niẓām al-Mulk, who was the most detailed source for the Ismāʿīlīs in Iran and eastern lands, observes that a major conspiracy appeared 15 years into the Sāmānid ruler’s reign. Niẓām al-Mulk attempts to convince his readers of the existence in Iran of a well-organized and long-lasting conspiracy against the Muslim social order (principally the Sunnīs), the intention of which was to weaken Islam in favour of a revival of Zoroastrianism and other Iranian dualist groups.Footnote ³³ Niẓām al-Mulk used several sources related to the Qarmaṭīs in Iran, among them the Muḥammad al-Malaṭī’s (d. 377/987) al-Tanbīh wa-l-Radd ʿalá Ahl al-Ahwāʼ wa-l-Bidaʿ and the Muʿtazilī theologian ʿAbd al-Jabbār’s (d. 416/1025) Tathbīt dalāʾil al-nubūwa.Footnote ³⁴ Further, Niẓām al-Mulk presents Rayy as one of the chief centres of this conspiracy. It is worthy of note that the daʿwa missionaries in Iraq, Baḥrayn, Yemen and North Africa had attempted to appeal to the public and relied on the diversity of society so they could appeal to a variety of people’s requests and ambitions.Footnote ³⁵ However, several recent scholars, such as Crone and Treadwell, refute Niẓām al-Mulk’s account as inaccurate.Footnote ³⁶ Both Crone and Treadwell emphasized that the version of Niẓām al-Mulk’s account of the Ismāʿīlīs during the Sāmānid period was different from the two contemporary sources (i.e. Ibn al-Nadīm and al-Thaʿālibī).Footnote ³⁷

These are the most comprehensive accounts that we have – the first written evidence for Qarmaṭīs in the north-west areas of Iran. Most modern scholars have built their hypotheses on the Qarmaṭīs in the region from these sources. However, there are some pieces of evidence in the primary sources that these modern writers have missed and others that have been misinterpreted – a theme that will be explored in the following discussion.

Rayy

It is important for the purpose of this article to first understand how the Ismāʿīlī daʿwa spread from Iraq to Iran during its formative stage. In the 260s/870s, the central leadership of the Ismāʿīlīs in Salamiyya recruited Ibn Ḥawshab and ʿAlī b. al-Faḍl as dāʿīs to Yemen, where they in turn recruited and sent dāʿīs to Sind (modern-day Pakistan) and other remote provinces in the region.Footnote ³⁸ It was also around 260/873 that the Ismāʿīlīs reached the region of Jibāl in north-west Iran. During this period, the Sāmānids held control of the area. After its preliminary success in the Jibāl region, the daʿwa was also applied in Khurāsān. Indeed, it was in following this method that the dāʿī Ghiyāth was able to convert al-Ḥusayn b. ʿAlī al-Marwazī, an amīr in the service of the Sāmānids in Khurāsān and other prominent figures, such as the ḥājib (chamberlain) Manṣūr b. Bāyqarā and Abū Manṣūr al-Razzāq, the governor of Ṭūs (near Mashhad in modern-day Iran).Footnote ³⁹ The dāʿīs formed their local headquarters at Rayy and, about three decades later, the daʿwa extended to the eastern regions. According to Ibn al-Nadīm, the first dāʿī to travel to Iran was Khalaf, who began his mission in Rayy during the 260s/870s, where he successfully attracted many followers.Footnote ⁴⁰ In addition, ʿAbdān’s brother, al-Maʾmūn, was selected as a dāʿī in Iran, where the Ismāʿīlīs were known as the Maʾmūniyya.Footnote ⁴¹

It was the Qarmaṭīs leaders of Iraq, Ḥamdān and ʿAbdān, who started the daʿwa in southern Iran. They recruited Abū Saʿīd al-Jannābī for his first mission to southern Iran, and he settled in Shīrāz (south-central Iran) with his family. His family remained there until the middle of the fourth/tenth century.Footnote ⁴² Indeed, Abū Saʿīd, during his later leadership of the Qarmaṭīs in Baḥrayn, sent Ibrāhīm al-Ṣāʾigh, a prominent member of the community in Baḥrayn, to Ahwāz (a town in southern Iran) to organize the daʿwa.Footnote ⁴³ In this regard, al-Ṣūlī (d. 335/948) indicated in his writing that there was a connection between the Qarmaṭīs of Baḥrayn and Mardavij (d. 323/935), who was a Gilaki (an Iranian ethnic group) prince who established the Ziyarid dynasty in Isfahan (central Iran).Footnote ⁴⁴ According to al-Ṣūlī, he stated:

Such connections between the Qarmaṭīs of Baḥrayn and local Iranian powers illustrate the broader spread of Qarmaṭī influence into Iran, paving the way for the activities of later Iranian dāʿīs.

These Iranian dāʿīs – Abū Ḥātim al-Rāzī (d. 322/934), Muḥammad al-Nasafī (d. 332/943) and Abū Yaʿqūb al-Sijistānī (d. after 360/971) – left us significant works in philosophy and religion, reflecting the intellectual culture during that time. In this respect, Daftary characterized the writings of these Iranian dāʿīs as follows: “al-Nasafī and other early dāʿīs of the Iranian lands, such as Abū Ḥātim al-Rāzī (d. 322/934) and Abū Yaʿqūb al-Sijistānī (d. after 360/971) wrote for the ruling elite and the educated classes of society, aiming to attract them intellectually and win their support for the daʿwa”.Footnote ⁴⁶ This may be due to the realization that their initial mission – to appeal to people in north-west Iran – had ultimately failed.Footnote ⁴⁷ On the other hand, it may be understood as a reflection of the relatively cultured milieu in this area at that time. Paul Walker has argued that due to these regions’ social and intellectual environment, a more elaborate form of doctrinal discourse was required and thus catalysed a heightened interest in philosophical knowledge.Footnote ⁴⁸ For instance, kalām (theology) in north-west Iran during the mid-fourth/tenth century was popular in circles of cultural debate, notably among the Muʿtazilite school, which influenced the development of Ismāʿīlī theology in the region.Footnote ⁴⁹ Building on this growing intellectual engagement, individual Iranian dāʿīs such as Abū Ḥātim al-Rāzī, Muḥammad al-Nasafī and Abū Yaʿqūb al-Sijistānī emerged as prominent figures, each contributing in distinctive ways not only to the Qarmaṭī movement in north-west Iran but possibly affecting the Qarmaṭīs in Baḥrayn as well.

Abū Ḥātim al-Rāzī’s first mission as an Ismāʿīlī dāʿī took place in Rayy. There, he worked to establish an underground network of followers by spreading Ismāʿīlī teachings, showing enthusiasm for Qarmaṭī ideas.Footnote ⁵⁰ Abū Ḥātim al-Rāzī corresponded with the Qarmaṭīs’ leader Abū Ṭāhir al-Jannābī (r. 311–332/924–944), son of Abū Saʿīd al-Jannābī.Footnote ⁵¹ Abū Ḥātim al-Rāzī, in his Kitāb al-Iṣlāḥ, criticizes a variety of themes in al-Nasafī’s Kitāb al-Maḥṣūl, namely the prophets’ law and the prophets’ followers,Footnote ⁵² while also dealing with metaphysical matters. According to Madelung, Abū Ḥātim broadcast the news of the advent of a messianic figure in the year 316/928, but the aftermath of the exposure of the false young Iranian Mahdī in Baḥrayn in 319/931 (a subject that will be analysed below) obliged him to abandon his supporters in Ṭabarestān (northern Iran).Footnote ⁵³ Madelung further argues – relying on various sources, among them the Muʿtazilī theologian ʿAbd al-Jabbār and the Seljuk vizier Niẓām al-Mulk – that Abū Ḥātim al-Rāzī spread the prediction that the advent of the Mahdī would take place in 316/928, astrologically calculated as the year of the conjunction of Saturn and Jupiter.Footnote ⁵⁴ Yet, the fallout of the exposure of a pseudo-Mahdī in Baḥrayn in the year 319/931 forced him to withdraw from Ṭabarestān and sever ties with his earlier base of support. According to Niẓām al-Mulk:

However, a few details have yet to be clarified. For example, can we link the prediction of the advent of the Mahdī of 316/928 and Abū Ḥātim al-Rāzī’s subsequent disappointment over the pseudo-Mahdī with Niẓām al-Mulk’s narrative of Abū Ḥātim al-Rāzī? Certain corresponding events might substantiate Madelung’s observation. Thus, the connection between Abū Ḥātim al-Rāzī and the Qarmaṭīs that Madelung suggests seems reasonable.

Shin Nomoto, in her study on Abū Ḥātim al-Rāzī, described it as a probability that Abū Ḥātim had participated in the Qarmaṭīs’ propaganda for the pseudo-Mahdī.Footnote ⁵⁶ Moreover, Nomoto examined Madelung’s survey on Abū Ḥātim al-Rāzī’s Qarmaṭ affiliation, stating:

Nomoto concluded her argument by stating: “We can only make reasoned guesses as to the answers to these questions because of the paucity of historical sources available at this moment. But although al-Rāzī’s involvement in the Qarmaṭīs movement seems plausible, there still remain questions concerning its nature and extent.”Footnote ⁵⁸ Does Abū Ḥātim al-Rāzī broadcast the news of the advent of the Mahdī that appeared in Baḥrayn? Indeed, we cannot provide a decisive answer to this question, since we cannot determine exactly when Abū Ḥātim al-Rāzī began his daʿwa activity in Rayy. Still, both the sources and modern scholarship suggest a convincing possibility that Abū Ḥātim al-Rāzī circulated or conveyed this news of the Mahdī in Baḥrayn.

The dāʿī al-Nasafī, who is credited with succeeding Ḥusayn b. ʿAlī al-Marwazī as leader of the daʿwa in Nīshāpūr (a city in north-east Iran), was successful in converting the most prominent leaders in the Sāmānid court.Footnote ⁵⁹ As a dāʿī, he travelled to Transoxiana (a historical region of Central Asia) and succeeded in converting the Sāmānid amīr, Naṣr b. Aḥmad (r. 301–331/914–943), and several dignitaries of the court to the Ismāʿīlī faith, as mentioned above. According to Ibn al-Nadīm,

This triumph, however, was short term and the fortunes of the Ismāʿīlīs were soon reversed when the Turkish army revolted and effected a complete reversal of policy under Naṣr’s successor Nūḥ b. Naṣr (r. 331–343/943–954).Footnote ⁶¹ Al-Nasafī, along with many Ismāʿīlīs and their sympathizers, was massacred in 332/943 – a year referred to by Niẓām al-Mulk as “the persecution of the Iranian Ismāʿīlīs”.Footnote ⁶²

Al-Nasafī reasserted the Mahdīship of Muḥammad b. Ismāʿīl in his Kitāb al-Maḥṣūl, which was written around the year 300/912Footnote ⁶³ and was devoted to an exposition of the Neoplatonic metaphysical system that had been introduced into Ismāʿīlīsm, namely the theory of an emanational chain of creation and its hierarchic chain of beings.Footnote ⁶⁴ Although most of al-Nasafī’s texts are now missing, some have been preserved in the writings of other Ismāʿīlī dāʿīs such as Abū Ḥātim al-Rāzī’s Kitāb al-Iṣlāḥ and al-Kirmānī (d. 411/1021) in his Kitāb al-Riyāḍ fī al-ḥukm bayna al-ṣādayn ṣāḥibay al-iṣlāḥ wa-l-nuṣra.Footnote ⁶⁵ Al-Nasafī conceived of history as a chain of the eras of the seven Ismāʿīlī prophets (referred to as “speakers”).Footnote ⁶⁶ Cosmologically speaking, this signified a drastic change from the Gnostic mythological theory of the creation of the early Ismāʿīlīs.Footnote ⁶⁷ Kitāb al-Maḥṣūl claims that the importance of philosophy with the Ismāʿīlīs should not be undervalued and that it was also practised by the Qarmaṭīs of Baḥrayn, with reference to Adam’s Paradise as a sign of the Iranian Ismāʿīlī school’s influence on the Qarmaṭīs. It is possible that Kitāb al-Maḥṣūl was widely accepted within the Qarmaṭīs of Baḥrayn and played a vital role in unifying them with the Qarmaṭīs in Iranian lands who otherwise left behind a scattered textual history of their existence and, because they had no leader, were in danger of disappearing or becoming subsumed under other ideologies like that of the Fāṭimids.Footnote ⁶⁸

The idea of Adam’s Paradise, as interpreted by the Qarmaṭīs, embodied the divine macrocosm and the stages of Neoplatonic emanation, a vision portrayed in al-Nasafī’s Kitāb al-Maḥṣūl. The Qarmaṭīs believed that everything created by God (the prohibited and the permitted alike) would all be granted to Muḥammad b. Ismāʿīl in Adam’s Paradise.Footnote ⁶⁹ They refer to this verse from the Quran: “Adam dwell thou, and thy wife, in the Garden, and eat thereof easefully where you desire; but draw not nigh this tree” (2:33).Footnote ⁷⁰ The Qarmaṭīs relate the first part of the verse to Muḥammad b. Ismāʿīl and his father, Ismāʿīl, but they relate the second part to the opposing group represented by the line of Mūsā b. Jaʿfar (d. 183/799) and his ancestors who declared the imamate for themselves (i.e. the Twelver Shīʿa).Footnote ⁷¹ According to Qarmaṭī belief, the Mahdī, Muḥammad b. Ismāʿīl, would appear in his final manifestation to sit in judgement over humanity, overturn all laws and conclude the work of all previous prophets – thus raising the world soul, from which he would rule over a Neoplatonist version of Adam’s Paradise.Footnote ⁷²

In Kitāb al-Maḥṣūl, al-Nasafī explained that Adam (the prophet) did not carry a law; instead, he taught the unity of God without recourse to preaching.Footnote ⁷³ According to al-Nasafī,

Further, al-Nasafī states:

Adam thus had no asās (fundament): Seth (Adam’s son) was the access to the revelation of his identity. In the same manner, the seventh speaker, Muḥammad b. Ismāʿīl, is expected to restore the original state of humans without a system of law and restore the doctrine of God’s unity without recourse to advocating.Footnote ⁷⁶ Al-Nasafī elucidates that Muḥammad b. Ismāʿīl was gone and would later return, as happened with the Prophet Jesus. However, what al-Nasafī had to say about the intervening period until his return cannot be understood with certainty.Footnote ⁷⁷

The ideas of al-Nasafī’s Kitāb al-Maḥṣūl (such as the notion of Adam’s Paradise) appear to have been applied and practised among the Qarmaṭīs community of Baḥrayn. In around 319/931, Abū Ṭāhir al-Jannābī introduced to his community a young Iranian who was presented to him as the Mahdī. Accounts of the young Iranian Mahdī’s arrival are contradictory. Historians al-Masʿūdī (d. 346/956) and Thābit b. Sinān (d. c. 363–365/973–975), whose accounts were used by Miskawayh (d. 421/1030) and others, described that Abū Ṭāhir was tricked by his cousin Ibn Sanbar into giving up power. Al-Masʿūdī argued that Ibn Sanbar (from the influential Sanbar family whose daughter was married to Abū Saʿīd al-Jannābī) sought to use this young Iranian to throw out his competitor, Abū Ḥafṣ al-Sharīk, a prominent leader in Baḥrayn, in order to control the “ʿiqdāniyya council”.Footnote ⁷⁸ Other accounts of the young Iranian Mahdī were written by the Muʿtazilī theologian ʿAbd al-Jabbār (d. 416/1025), who cited Ibn Rizām (d. late fourth/tenth century) as saying that when Abū Ṭāhir summoned his brothers and the community he declared: “You people! Know: We and you also were all donkeys! But now God has sent us this man [the young Iranian] – this is my Lord and your Lord and my God and your God. We are all his slaves! The rule has now been transferred to him, and he will be king of us all” and Abū Ṭāhir and his community then circled around the young Iranian Mahdī, completely naked and yelling: “Our God, he is mighty and exalted!”Footnote ⁷⁹

Abū Ṭāhir also declared all previous religions to be false, claiming that the true religion – “the religion of our father Adam”, that is, “the religion of Paradise” without laws – had now been detected.Footnote ⁸⁰ The prophets Moses, Jesus and Muḥammad had shown themselves to be frauds.Footnote ⁸¹ Indeed, the Qarmaṭīs assumed that the religions of the holy law – Judaism, Christianity and Islam – were only alternative superficial shells for the “True Religion”, the original Religion of Paradise free from ritual obligations, mandates and prohibitions. This was the pure worship of God that Adam and the Angels practised before their fall to Earth.Footnote ⁸²

These ritual ceremonies have been recorded in modern scholarship. For example, Heinz Halm described this ritual as a Qarmaṭī endeavour to restore Adam’s religion, a way of “returning to Paradise”.Footnote ⁸³ Halm also demonstrates that this was not an idea held only by the Qarmaṭīs.Footnote ⁸⁴ Similar notions regarding the original religion of Adam were present in late antique Gnosticism, which was one of the roots of Ismāʿīlīsm. Halm remarks that the distinctive features of these “communities of Adam” (that is, the Qarmaṭīs in Baḥrayn) were their general sense of ibāḥiyya (“antinomianism”, implying that the beliefs were sexually perverse) and that applicants were always naked during rituals. Halm also portrayed such societies as pagan,Footnote ⁸⁵ since no sooner had they entered the condition of prelapsarian innocence than their bodies had become incapable of sinning.Footnote ⁸⁶ François de Blois considers the young Iranian Mahdī as a puppet invented by Abū Ṭāhir, who arranged the whole interlude as a scheme to gain more power.Footnote ⁸⁷ Wilferd Madelung and István Hajnal observed the idea of Adam’s Paradise alongside antinomian trends in ʿAbd al-Jabbār’s account and in his source (Ibn Rizām).Footnote ⁸⁸ According to the Qarmaṭīs, the Mahdī will restore this original religion for mankind, announce the end of time, eliminate the laws of the preceding religions and serve as the connection between the upper and lower worlds.Footnote ⁸⁹ No matter how much of this account is true, it signifies the existence of syncretistic Gnostic-Ismāʿīlī ideas during this time – ideas that were not entirely Islamicized and remained sectarian. This formed the framework for initiating the Qarmaṭīs’ radical symbolic discourse, followed by a tribalism that shaped the bulk of the daʿwa. Alternatively, it may be that the whole passage was invented by ʿAbd al-Jabbār or his source Ibn Rizām to discredit the Qarmaṭīs.

According to Paul Walker, this abrogation of religion was a process whereby the original religion would improve the preceding laws rather than replace them entirely.Footnote ⁹⁰ Indeed, Walker concluded that the prophets would make their mark on their period by naming the successive periods in such a way that they would all point to the restoration of a final era without law when the truth would no longer be concealed in its physical form.Footnote ⁹¹ In this regard, the Iranian traveller and poet Nāṣir-i Khusraw (d. 481/1088), who visited the Qarmaṭīs’ capital al-Ḥasā (in eastern Arabia) in 443/1051 during the late Qarmaṭīs’ period, reported: “They [the Qarmaṭīs] do not prevent anyone from performing prayers [any kind of prayer, possibly including non-Muslim prayer].”Footnote ⁹² This account indicates the liberty and receptivity of the law, which seems to be more spiritual than physical. In this regard, Nadia El-Cheikh demonstrates that Nāṣir-i Khusraw was sympathetic towards the Fāṭimid outlook, which viewed the Qarmaṭīs as heretical extremists. Consequently, he related accounts that were a strange mixture of truth and fiction.Footnote ⁹³ Here, Nāṣir-i Khusraw may well be prejudiced against the Bedouin, or he was implying that they are not Muslims because they do not acknowledge the Fāṭimid Imamate/Caliphate.Footnote ⁹⁴

Incidentally, these ceremonies of the pseudo-Mahdī bolstered the Sunnī polemicists’ accusations of animosity against the Qarmaṭīs as a confirmation of the Qarmaṭīs’ heresy. Chief among these sources were ʿAbd al-Qāhir al-Baghdādī (d. 429/1038), Niẓām al-Mulk (d. 485/1092), al-Ghazālī (d. 505/1111) and Ibn al-Jawzī (d. 597/1200), who all categorized the Qarmaṭīs (particularly those in Iran) as bāṭinīs.Footnote ⁹⁵

Another Qarmaṭī dāʿī was Abū Yaʿqūb al-Sijistānī who, according to Ibn al-Nadīm, served as the lieutenant of the imam in Rayy. The Banū Ḥammād of Mosul, the leaders of the daʿwa in the Jazīra (northern Mesopotamia), operated on his behalf.Footnote ⁹⁶ It is quite likely that al-Sijistānī was, on the former occasion, already a fully active senior member of the Ismāʿīlī daʿwa and based in Rayy. Later, he moved to Khurāsān, where he may have been a disciple of the dāʿī Muḥammad al-Nasafī.Footnote ⁹⁷ Al-Sijistānī followed what Walker described as an “old belief” (relating to the proto-Ismāʿīlī daʿwa before the Fāṭimids), and wrote his Nuṣra in defence of al-Nasafī’s Kitāb al-Maḥṣūl and in response to al-Rāzī’s Kitāb al-Iṣlāḥ. Nonetheless, al-Sijistānī seems to have changed his standpoint over time, or later copyists may have deliberately altered parts of the older texts.Footnote ⁹⁸ For instance, al-Sijistānī primarily believed in tanāsukh (metempsychosis), but al-Kirmānī, who dealt with al-Sijistānī’s works in his Kitāb al-Riyāḍ, found this problematic and noted that al-Sijistānī’s doctrine of the soul contained offensive implications. Al-Kirmānī considered al-Sijistānī’s doctrine to be the same as Plato’s (d. 347 bce) as it implied metempsychosis: “What the author of the Nuṣra said in the forty-fourth chapter of his book entitled al-Maqālīd … shows the contrary of what he said in his Nuṣra.”Footnote ⁹⁹ One of the chapters in al-Sijistānī’s al-Maqālīd, titled “That Metempsychosis is False”, implies the change in his doctrine.Footnote ¹⁰⁰ After al-Nasafī’s execution in 332/943 at the hands of the Sāmānid, al-Sijistānī went to Sijistān (south-east Iran), and he later officially accepted and followed the policy conversion of the Fāṭimids (which will be discussed below), revising his dogma on the Mahdīship of Muḥammad b. Ismāʿīl to accept the notion of the Fāṭimid deputies.Footnote ¹⁰¹ Evidently, al-Sijistānī put forward this argument in his works, such as in Ithbāt al-nubūwāt and al-Yanābīʿ.Footnote ¹⁰²

Overall, it is difficult to be certain whether affiliation to al-Rāzī, al-Nasafī and al-Sijistānī or their ideas were present within the Qarmaṭīs of Baḥrayn’s practices. Regardless, the above discussion has sought to illustrate a possible link between the philosophical, theoretical practice of the intellectual school of Iranian Ismāʿīlīsm and the actual praxis of the Qarmaṭīs of Baḥrayn. Although these dāʿīs are presumably associated with the Qarmaṭī movement, they do not fully represent a unified group. Even though al-Sijistānī in his early life followed his teacher al-Nasafī and defended him on many occasions, there were some notable divergences among the Iranian dāʿīs. For example, both al-Rāzī and al-Nasafī differed significantly on the nature of the Mahdī and the degree to which philosophy influenced cosmology. These doctrinal divergences not only illustrate the internal heterogeneity of the Iranian Ismāʿīlī-Qarmaṭī daʿwa, they also reflect the intellectual nature of the debate of north-west Iran during that period. Yet, despite these differences among the Iranian dāʿīs, the Ismāʿīlī daʿwa retained a remarkable capacity for regional adaptation, as evidenced by its subsequent manifestations in areas such as Azerbaijan, which was shaped by earlier Iranian daʿwa traditions, was responsive to new political circumstances and continued to develop distinct organizational features.

Azerbaijan

The Qarmaṭīs survived in areas of north-west Iran later under the Musāfirids (or Sallārids), a dynasty of Daylam (a mountainous region of Gilan, Iran), which originated in Ṭārum (modern-day Azerbaijan) and ruled during the fourth–fifth/tenth–eleventh centuries in Azerbaijan and Armenia.Footnote ¹⁰³ The Musāfirids coming to power was an indication of the important movement of Iranian liberation, which represented an interlude between the end of Arab domination and the first Turkish conquests.Footnote ¹⁰⁴ The most prominent ruler of the Musāfirids was al-Marzubān b. Muḥammad (r. 330–346/941–957), who held control of Azerbaijan. During this period, Azerbaijan was a place of internal conflict between the Khārijites (an Islamic group), Daysam al-Kurdī and Lashkarī b. Mardī, a native of Jīlān (a village in northern Iran), all of whom were vying for authority over this region.Footnote ¹⁰⁵ Both Daysam and Lashkarī’s upheavals ultimately failed; hence, al-Marzubān was able to extend his rule and increase his authority, conquering Ardabīl and Tabrīz (cities in north-west Iran). Eventually, Daysam yielded to al-Marzubān, who gave him a castle in Ṭārum from where he then dominated most of the northern Iranian lands.Footnote ¹⁰⁶ Later, the Musāfirids lasted for a certain time under the Seljuk dominion and eventually were defeated by the Persian Nīzārī Ismāʿīlīs, who came to rule Shamīrān and other strongholds of the region.Footnote ¹⁰⁷

Several sources have commented on the period of al-Marzubān and his people at that time. The traveller Ibn Ḥawqal (d. c. 380/990), who visited Azerbaijan in the 340s/950s, reported a large presence of Qarmaṭīs: “And among them [the Azerbaijan community] at this time are those of the schools of Ahl al-Ḥadīth [‘the People of Ḥadīth’, that is, a Sunnī school that believed the Quran and the record of the sayings and actions of the Prophet Muḥammad to be the only authority in Islam], and they speak in tautologies and many of them are bāṭinīs of the Baqliyya.”Footnote ¹⁰⁸ The Baqliyya were a group founded by Abū Ḥātim “al-Ẓuṭṭī”,Footnote ¹⁰⁹ who was active as a dāʿī among the Qarmaṭīs in southern Iraq. Al-Ẓuṭṭī prohibited the consumption of certain vegetables and the slaughtering of animals, because of which his followers were called the Baqliyya, a name derived from the term baql, meaning “plants”; this was subsequently applied to all the Qarmaṭīs of southern Iraq, who for the most part had retained their earlier belief in the Mahdīship of Muḥammad b. Ismāʿīl.Footnote ¹¹⁰ According to al-Masʿūdī, the Baqliyya group in southern Iraq later joined the Qarmaṭīs of Baḥrayn.Footnote ¹¹¹

This Ismāʿīlī presence has also been confirmed by Miskawayh, who mentioned al-Marzubān’s conversion to Ismāʿīlīsm:

Al-Marzubān was not alone in his family in having Ismāʿīlī sympathies. His brother and successor, Washsūdān, also followed the Ismāʿīlī (Qarmaṭī) dogma. Washsūdān and his nephews (al-Marzubān’s sons), while bequeathing power to his brother, al-Marzubān, neglected to revoke an earlier will that named al-Marzubān’s sons – Justān, Ibrāhīm and Nāṣir – as his successors.Footnote ¹¹³ Washsūdān began to conspire against his nephews and sent Nāṣir from Justān, but the quarrel was short. Under assurances from Washsūdān, Justān, with his mother and Nāṣir, came to Ṭārum only to be thrown into prison. Ibrāhīm, who was ruling Armenia, made a move to attack his uncle in 349/960, which gave Washsūdān a justification to execute his prisoners, after which he ruled the kingdom again. Washsūdān certainly seems to have been influenced by the Ismāʿīlīs since he named his son Ismāʿīl, though this does not reveal anything deeper about the doctrine he followed.

Material evidence implies a clear affinity between the Qarmaṭīs and the north-west Iran Ismāʿīlīs. The evidence of a coin from the year 343/955 verifies that the dāʿīs in Azerbaijan adhered to the Qarmaṭī doctrine and acknowledged the Mahdīship of Muḥammad b. Ismāʿīl rather than the Fāṭimid imam-caliph, al-Muʿizz (r. 341–365/953–975).Footnote ¹¹⁴ The coin, which bears the Musāfirid ruler Washsūdān’s Ismāʿīlī affiliation, was first mentioned by Miskawayh.Footnote ¹¹⁵

This coin has been analysed by Vladimir Minorsky, who remarked that it clearly demonstrates Washsūdān’s Ismāʿīlī affiliation (see Figure 1). The coin reflects the territory ruled by Washsūdān, in which Jalālābād (a town most probably in Azerbaijan) must have been situated.Footnote ¹¹⁷ Washsūdān’s Ismāʿīlī affiliations are evident on this coin in three ways. First, to the two parts of the Muslim creed “There is no god but God, Muḥammad is the Messenger of God”, he adds the Shīʿī expression “And ʿAlī is the caliph of God”.Footnote ¹¹⁸ Second, Washsūdān gives himself the Shīʿī title of “Sword of the family of Muḥammad [the Prophet]”, a designation that recurs in Washsūdān’s coin minted c. 366/976 in Ardabīl (a city in north-west Iran).Footnote ¹¹⁹ Third, the outer obverse margin contains a list of the Ismāʿīlī imams. Ismāʿīlī doctrine believes in a hierarchy of seven imams beginning from the Prophet Muḥammad’s son-in-law ʿAlī b. Abī Ṭālib (d. 40/661) and his descendants Ḥasan b. ʿAlī b. Abī Ṭālib (d. 50/670), Ḥusayn b. ʿAlī (d. 61/680), ʿAlī b. Ḥusayn (known as Zayn al-ʿĀbidīn) (d. 95/713), Muḥammad al-Bāqir (d. 117/733) and Jaʿfar al-Ṣādiq (d. 148/765) down to Ismāʿīl b. Jaʿfar al-Ṣādiq (c. d. 142/759), the father of the seventh and final imam, Muḥammad b. Ismāʿīl (d. 193/809), who would return as the Mahdī and bring about an age of justice.Footnote ¹²⁰ The coin included the names of the Prophet Muḥammad and Ḥasan b. ʿAlī b. Abī Ṭālib, who were not actually part of the chain of imams, although the Shīʿa would occasionally include these names with their seven imams to emphasize their link to the Prophet’s family.

Figure 1.

Jalālābād, AU Dirham 343/943.Footnote ¹¹⁶

obv.: area: لاإله الا الله/محمد رسول الله obv.: outer margin: محمد علي الحسن الحسين علي محمد جعفر اسمعيل محمد rev.: علي خليفه عدل الله/سيف ال محمد/وهسودان بن محمد

Most studies present the Musāfirids’ Ismāʿīlī affiliation as being related to the Fāṭimids. However, I would refute this and argue that their link was with the Qarmaṭīs. The Ismāʿīlī affiliation of the Musāfirids is an area that has not received significant attention in modern scholarship. Only the studies of Samuel Stern and Farhad Daftary have dealt with this point. Stern’s analysis included this subject, yet I would argue that he erroneously links the Musāfirids with the Fāṭimids, whom he deems to have been head of the Ismāʿīlī movement at that time.Footnote ¹²¹ In contrast, Daftary’s examination was superficial – he barely indicates the Musāfirids’ Ismāʿīlī affiliation to the Qarmaṭī line without considering the context of their affiliation.Footnote ¹²²

However, during this period (before the mid-fourth/tenth century), it is difficult to determine who the head of the Ismāʿīlīs was. Indeed, as discussed above, after the split of the Ismāʿīlī daʿwa in c. 286/899, two different Ismāʿīlī beliefs emerged, as represented by the Qarmaṭīs and the Fāṭimids. Being a Qarmaṭī meant refusing the Fāṭimid imams-caliphs and advocating only Muḥammad b. Ismāʿīl as the Mahdī. Other Ismāʿīlīs who recognized the imamate of the Fāṭimid imams-caliphs were described as Fāṭimids, and their followers usually minted their coins with the names of these rulers. For example, the Fāṭimid creed usually required the allegiance of the Fāṭimid imams-caliph al-Muʿizz, the so-called “Master of the Era” in that period, as can be seen through the coins of the Fāṭimids in Iran and eastern lands.Footnote ¹²³ In this regard, Stern questioned why the coin of Washsūdān did not bear the name of (or at least allude to) the Fāṭimid imam-caliph al-Muʿizz. Stern answered this by arguing that it would have been risky for Washsūdān to come out so openly on the side of the heretic Fāṭimids, whereas the remarkable Shīʿī mottos could pass as less offensive and the somewhat vague list of imams might not easily be recognized for what it was, thus avoiding criticism.Footnote ¹²⁴ Stern concluded that “Washsūdān’s Ismāʿīlīsm was of the ordinary type and he was a faithful adherent to al-Muʿizz, and [we must] leave the apparent difficulty unresolved – reflecting that it is not always possible to explain to the last puzzling feature the events of the past.”Footnote ¹²⁵ However, I would argue that Stern is mistaken in linking the Musāfirids to the Fāṭimids rather than the Qarmaṭīs; he disregards the sources (Ibn Ḥawqal and Miskawayh) that show (as I have demonstrated) that the Qarmaṭīs were active during this period in this region. Indeed, the list of imams ends with Muḥammad b. Ismāʿīl, whom the Qarmaṭīs believed would return as the Mahdī, in complete contrast to what the Fāṭimids believed. When Stern says that “Washsūdān’s Ismāʿīlīsm was of the ordinary type”, he hints that “ordinary” means “not Fāṭimid”, disregarding that this “ordinary” may possibly refer to the pure/proto Ismāʿīlīs, that is, the Qarmaṭīs, according to many modern scholars, such as Madelung and Daftary.Footnote ¹²⁶

An alternative view of the Musāfirids’ Ismāʿīlī affiliation is given by Daftary:

Daftary’s examination was based on Ibn Ḥawqal’s account related to Azerbaijan, which reported the existence of many Ismāʿīlīs there.Footnote ¹²⁸ Further, Daftary stated the numismatic evidence dated 343/955, mentioned above, confirming that Washsūdān and his brother al-Marzubān followed the Qarmaṭī form of Ismāʿīlīsm, recognizing the Mahdīship of Muḥammad b. Ismāʿīl rather than the imamate of the Fāṭimid imam-caliph al-Muʿizz.Footnote ¹²⁹ Nevertheless, Daftary did not analyse the coin’s content or the texts of the sources relating to the Musāfirids’ Ismāʿīlī affiliation.

I believe, however, that this lack of knowledge about the Qarmaṭīs in these regions of Iran, Azerbaijan and eastern lands is due to two reasons. First, a radical split occurred in Ismāʿīlī doctrine in 286/899, before which the figure of Muḥammad b. Ismāʿīl had been at the forefront of Ismāʿīlī belief. The second reason is the policy enforced by the Fāṭimid imam-caliph al-Muʿizz in around 354/965 as part of the Ismāʿīlī daʿwa to convert the Qarmaṭīs in Iran, Azerbaijan and the eastern regions to Fāṭimid Ismāʿīlīsm.Footnote ¹³⁰ Thus, al-Muʿizz’s policy of reorganization to influence certain segments of society ensued from a confrontation with the eastern Ismāʿīlī communities. There, too, the Ismāʿīlī creed had not simply been preserved in its former shape; it had gained a new appearance at the creative hands of the early dāʿīs. Because they had no unified leadership, these communities had probably drifted further and further apart in questions of doctrine as well, but nothing much can be determined about the details of this development due to a paucity of sources. It is, however, possible to discern at least the outlines of the Qarmaṭīs’ relationship with the Ismāʿīlīs in Iran, Azerbaijan and eastern lands through their belief in the Mahdīship of Muḥammad b. Ismāʿīl. Indeed, most of the north-west Iranian Ismāʿīlī communities were affiliated with the Qarmaṭīs until the time of al-Muʿizz’s conversion policy.