S.M. Stern
In the original program of this meeting, I was to talk on the Qarmaṭians while the Ismā'īlīs had been allotted to another speaker; but as that speaker was prevented from coming, Ismā'īlism was dropped from the program, leaving me to deal with the Qarmaṭians in isolation. If the original distribution had put me in a quandary, I am even more confused now, since it is my view that the names "Ismā'īlī" and "Qarmaṭian" are, properly speaking, synonymous and do not refer to different entities. It is true that the movement to which both names are applied was at one moment in its history broken by a schism, and that the name "Qarmaṭian" was predominantly used in respect of the Qarmaṭians of Baḥrayn, who were at variance with the main body of the Ismā'īlī movement; yet even then the term "Qarmaṭian" was not exclusively reserved for them and was often used – usually in a derogatory sense –to denote any Ismā'īlī. It would not, therefore, have been wholly convenient to have these two terms as titles for two consecutive communications, as though they were correlative terms designing two different movements. I had originally decided to restrict myself mainly to the discussion of the relations between the Ismā'īlī movement in general and what one usually understands by Qarmaṭians, i.e. the Ismā'īlīs of 'Irāq and of Baḥrayn. I shall still keep to this plan on the whole; but as it is impossible to deal with the Qarmaṭians in isolation, I shall now, in the absence of a separate contribution on Ismā'īlism in general, also have to devote some time to the main problems of the history of early Ismā'īlism.
We may start with a remark about the name "Ismā'īlī" – a remark which will lead us straight to the first historical problem which I would like to discuss. The early Ismā'īlīs were seldom so denominated by their contemporaries, being called instead by such names as Qarmaṭians or Bāṭinīs, They themselves seem to have designated their movement simply by the name "the mission", al-da'wa, or more formally "the right-guided mission", al-da'wa al-hādiya; thus "to be converted to Ismā'īlism" would be rendered by them as "to enter the mission", dakhala'l-da'wa. The name "Ismā'īliyya" originally belonged to the language of the heresiographers who use it1 in connection with the Shī'ite sect which professed the Imāmate of Ismā'īl, the son of Ja'far al-Ṣādiq, or as a synonym of "Khaṭṭābbiyya" to design that extremist sect who believed in the Imāmate of Ismā'īl and his son Muḥammad but also professed in different forms, the divinity of the Imāms. All these sects belong to the lifetime of Ismā'īl himself, i.e. the middle of the second century of the Hijra (eighth century A.D.). Yet I think that there is very little connection between these sects and the Ismā'īlī movement in connection with which we use the word. I should like to make the point that the Ismā'īlī movement does not appear until about a hundred years afterwards, in the middle of the third/ninth century. This view, which is contrary to the generally accepted opinion according to which there is a direct connection between the early groups of the middle of the second/eighth century and the Ismā'īlī movement of the third/ninth century, can be supported by two considerations.
(1) About the year 260 A.H. (middle of the ninth century A.D.) there appear quite suddenly missionaries in the various provinces of the Islamic world preaching their revolutionary doctrines. In 261 we find an Ismā'īlī mission established in Southern 'Irāq, where the local leaders were Ḥamdān Qarmaṭ and 'Abdān; soon afterwards we find the Ismā'īlīs established in Baḥrayn under the leadership of Abū Sa'īd al-Jannābī, and in the Yemen under Manṣur al-Yaman and 'Alī b. al-Faḍl. The famous missionary Abū 'Abū Allāh ai-Shī'ī, to whom the Fāṭimids owed their throne, came to North Africa from the Yemen in 280. The establishment of an Ismā'īlī centre in Rayy in North-West Persia can also be dated to the second half of the third century A.H.
It is worth mentioning parenthetically that in the different areas reached by the Ismā'īlī propaganda this found acceptance in different social strata. In Southern 'Irāq, the missionaries appealed mainly to the rural population of the sawād, although they found response also among some of the bedouin tribes of the vicinity. In Baḥrayn, the mission was based on the bedouins; in North Africa, on some Berber clans. In North-West Persia, the first missionaries sought adherence among the villagers of the district of Rayy, and it was not till about 300 A.H. that they attempted to gain their ends by converting the rulers of Rayy. In Khurāsān, where the Ismā'īlī mission does not seem to have made its first appearance before the end of the third century A.H., this new policy seems to have been the one pursued from the beginning, and the missionaries seem to have concentrated their effort at the conversion of high dignitaries of the Sāmānid state and finally of the Sāmānid amīr himself.
Behind this sudden activity one can discern the directing hands of an energetic central organization which must have come into being shortly before the missionaries appeared in the field, i.e. about the middle of the third/ninth century. Thus the earlier Ismā'īliyya and the Ismā'īliyya in our sense of the word are clearly of a very different nature: on the one hand we see small sects of extremely limited appeal, on the other a large scale missionary movement of universal pretensions.
(2) The doctrines preached by the Ismā'īlī missionaries of the second half of the third/ninth century were, moreover, quite different from the doctrines of the earlier sects; for example, while the Khaṭṭabiyya and related sects believed in the divinity of the Imāms, the Ismā'īlī movement, as we shall presently see, did not profess a belief in a human manifestation of the divinity. To be sure, there must have been some connection between the early sects and the movement, since the movement, as we shall see, centres round the figure of Muḥammad b. Ismā'īl, and thus acknowledged the same lineage as the early Ismā'īliyya – but that in itself does not mean the existence of a historical continuity. We have to imagine that it was some persons who had belonged to one or the other Ismā'īlī conventicle which survived precariously till the middle of the third/ninth century who founded the new movement, giving it however a character entirely different from the sects as regards both its aim and organization as well as its doctrine.
One of the main provinces of this new Ismā'īlī propaganda was Southern 'Irāq, and the name "Qarmaṭian" is derived from the chief of the mission in that province, Ḥamdān the son of al-Ash'ath, surnamed Qarmaṭ or Qarmaṭūya. Qarmaṭ seems to be an Aramaic nickname, though it is uncertain what exactly it means; Qarmaṭūya is the same nickname with a Persian diminutive suffix added. According to our sources Ḥarndān Qarmaṭ had been won over to the Ismā'īlī cause about the year 260 A.H. – i.e. the earliest period of the movement –by a missionary who came to the countryside (sawād) of Kūfa, Qarmaṭ's native place; he then became the leader of the movement in Southern 'Irāq, with his brother-in-law 'Abdān as second in command. When the movement gained ground in 'Irāq and became notorious among the local population as well as the authorities in Baghdād, it did so under the name of the movement of the "Qarmaṭians", i.e. the followers of Qarmaṭ. The name also came to be used in a wider sense to include the followers of the movement in other parts of the Islamic world as well. One of our few contemporary witnesses, the author of the Firaq al-Shī'a, who has now turned out to be Sa'd b. 'Abd Allāh al-Ash'arī al-Qummī2 and who seems to have written this book in the eighth decade of the third century A.H., uses3 the term al-Qarāmiṭa to denote the members of the movement in 'Irāq and the Yemen, the number of whom he estimates to be about 100,000; this figure is obviously meant to cover the whole of the movement and so does the name al-Qarāmiṭa. The author says: "They are numerous, ... more especially in the sawād of Kūfa and the Yaman".
There is no reason whatsoever to assume that the beliefs of the followers of Ḥamdān Qarmaṭ in Southern 'Irāq differed in any respect from those held by the rest of the movement in its early days. The question arises what these beliefs were; and when we come to inquire into them, we must beware of the pitfall of attributing to earlier generations the doctrines professed by later ones. If we wish to find out the doctrines preached by the Ismā'īlī missionaries in the second half of the third century A.H., the only admissible method is to examine contemporary authorities. I think I have found a way to recover, by confronting certain contemporary evidence with some later Ismā'īlī texts, the original cosmological system taught by the Ismā'īliyya which had later to give way to more advanced doctrines. This original doctrine consists of a rather crude cosmological myth; but I have no time to discuss this aspect of the early Ismā'īlī doctrine in greater detail here, for I must turn to their central teaching, that about the Imāmate.
In this respect we are fortunate enough to possess the testimony of the author of the Firaq al-Shī'a, which is of primary importance just because it belongs to an early date, before subsequent political and doctrinal developments completely changed the face of the movement; in effect, it provides the key to the secret of the inner development of the Ismā'īlī movement. According to that author, the Ismā'īlīs believed that there existed seven law-giver prophets: Noah, Abraham, Moses, Jesus, Muḥammad, 'Alī, and lastly Muḥammad b. Ismā'īl. Muḥammad b. Ismā'īl had not died, but was to return as the expected Mahdī or Qā'im. That the doctrine about the Mahdīship of Muḥammad b. Ismā'īl was in fact the early doctrine of Ismā'īlism is also confirmed by other texts, and more especially by a remarkable, indeed a startling, text by Ja'far b. Manṣūr al-Yaman (son of the founder of the Ismā'īlī mission in the Yemen and himself one of the leading Fāṭimid theologians in the first half of the fourth/tenth century), which has been published only last year.4 I wish I had the time to discuss this document in greater detail, as it casts a curious light – or shall we say envelops in even thicker mist – some of the problems of Ismā'īlī history (it states e.g. that the Fāṭimid caliphs were descendants not of a non-existent Muḥammad b. Ismā'īl but of 'Abd Allāh b. Ja'far al-Ṣādiq); I must, however, confine myself to quoting the statement according to which some of the early dā'īs have thought "owing to a misunderstanding" that Muḥammad b. Ismā'īl was the Qā'im.
There was a related circle (belonging to the Khaṭṭābiyya) who acknowledged a line of Imāms who descended from Muḥammad b. Ismā'īl; but the author of the Firaq al-Shī'a, who mentions that circle,5 distinguishes it well from the Qarmaṭians and stresses that the Qarmaṭians did not believe in the existence of Imāms after Muḥammad b. Ismā'īl, but counted seven Imāms only, 'Alī (who was both an Imām and a prophet), al-Ḥasan, al-Ḥusayn, 'Alī b. al-Ḥusayn, Muḥammad b. 'Alī, Ja'far, Muḥammad b. Ja'far.
Several questions now arise: who were those who founded and organized the movement? Were they the ancestors of the Fāṭimid caliphs? And if so, what was their origin? And finally, what was their position in the movement? In the total absence of any contemporary documents about them and the tendentious character of later versions, we can hardly hope ever to be able to give definite answers to these questions.
As is well known, according to the official Fāṭimid version the line of Fāṭimid caliphs was preceded by a series of "Hidden Imāms" descended from Muḥammad b. Ismā'īl. The trouble is that there are in reality not one, but several contradictory accounts. The enemies of the Fāṭimids did not admit the descent of the Fāṭimid dynasty from Muḥammad b. Ismā'īl, but ascribed the foundation of Ismā'īlism to 'Abd Allāh b. Maymūn al-Qaddāḥ (who in fact lived about the end of the second/eighth century) and considered the "Hidden Imāms" and therefore also the Fāṭimids as his descendants who falsely claimed an 'Alid descent. One could dismiss this as a malicious invention were it not for the fact that some groups inside the Ismā'īlī movement itself agreed with the doctrine of the Qaddāḥid ancestry. Thus while we may when weighing the value of the different assertions incline this or the other way, it is hardly possible to reach any certainty, nor is there much hope that we will ever have decisive evidence in our hands.
Quite apart from the question of the actual descent of the ancestors of the Fāṭimids, whom we assume to have been the leaders of the movement, there is the additional question of their exact function. On the one hand if the movement preached the Mahdīship of Muḥammad b. Ismā'īl who was to return, there would hardly be a place left for actual Imāms; and in effect the author of the Firaq al-Shi'a says that the Qarmaṭians acknowledged no such Imāms. On the other hand we have seen that according to the same author there was a group who did believe in the existence of Imāms descended from Muḥammad b. Ismā'īl. What was the relation of that group to the movement? And what was the authority which the ancestors of the Fāṭimids claimed for themselves in the movement, the doctrine of which left no place for Imāms? In the absence of evidence, we cannot go beyond mere speculation.
We have now to examine the process by which the Ismā'īlī doctrine passed from its early form to that in which it appears in the middle of the fourth/tenth century. We have a detailed account of a complete change in the sect's doctrine about the Imāmate which is said to have occurred about the year 286 A.H., by Ibn Rizām, an anti-Ismā'īlī pamphleteer of the first half of the fourth/tenth century. According to this account the leader of the sect who lived in Salamiyya in northern Syria died about that time and was succeeded by his son. In the son's letters the local leaders of the movement in southern 'Irāq, Ḥamdān Qarmaṭ and 'Abdān, noticed a different tone; getting suspicious, they sent an envoy to the headquarters in Salamiyya. To this envoy the new chief declared – so Ibn Rizām asserts – that the idea of Muḥammad b. Ismā'īl's Mahdīship had been put forward as a mere ruse; that in fact the leaders of the movement, descendants of 'Abd Allāh b. Maymūn al-Qaddāḥ, should be considered as Imāms, and now he, the new chief, was Imām in their stead. The envoy reported back to the leaders in 'Irāq, who did not accept the new line and renounced their allegiance to the leadership in Salamiyya. Ḥamdān Qarmaṭ died at that time, while 'Abdān was murdered. One might regard this account, which derives after all from a pamphleteer whose aim was to blacken the reputation of the Fatimids,' with some suspicion. On the other hand, the main contents of the account are corroborated by a writer who was an adherent of the Fāṭimids, viz. the geographer Ibn Ḥawqal. He says (p. 295) that Ḥamdān Qarmaṭ and 'Abdān, who had been at first followers of the Fāṭimids, later renounced their allegiance, and that 'Abdān was killed. Thus the fact that a rupture occurred between the leaders in the headquarters and the provincial leaders in 'Irāq is confirmed by an adherent of the Fāṭimids. Moreover, Ibn Ḥawqal adds a very significant fact which is not contained in the other account, viz. that the leader of the movement in Baḥrayn, Abū Sa'īd al-Jannābī, took the part of Ḥamdān Qarmaṭ and 'Abdān and also renounced his allegiance. Thus the movement split in two.
On the one side were those who accepted the doctrine according to which there was a visible Imām at the head of the community. (We have seen that a Khaṭṭābī group believed in a series of Imāms descended from Muḥammad b. Ismā'īl.) When the first Fāṭimid caliph (commonly known as 'Ubayd Allāh, though his official name was 'Abd Allāh) ascended the throne in Ifrīqiya and founded the Fāṭimid state, he made this doctrine the official religion of that state. During the early days of the Fāṭimid rule in North Africa their doctrine had, however, an important feature which distinguished it from its form as we know it later: 'Ubayd Allāh's heir and successor had a special status and he, the second ruler of the dynasty, was considered as the Mahdī who would usher in the messianic era. This belief explains the choice of the regnal name of the second caliph (who bore the title al-Qā'im bi-Amr Allāh): Abū'l-Qāsim Muḥammad b. 'Abd Allāh corresponds exactly to the name of the Prophet Muḥammad and satisfies the requirements of the tradition according to which the Mahdī would share the same name with the Prophet. Only when the hopes set on al-Qā'im bi-Amr Allāh as the Mahdī were dispelled by his death were hopes deferred, and though even then one occasionally finds attempts to define the number of the Imāms who would reign before the advent of the Mahdī, the figure of the Imām assumed central importance in Ismā'īlī thought at the cost of that of the Mahdī-Qā'im which was pushed in the background.
Against this trend we find another wing of the movement which did not give up the original doctrine of Ismā'īlism and retained their devotion to the hidden Qā'im, Muḥammad b. Ismā'īl. The Qarmaṭians of Baḥrayn seem to have belonged to this group. It is true that we have hardly any direct evidence for their beliefs, but on the other hand there are some indications to guide us. We have seen that, according to Ibn Ḥawqal, Abū Sa'īd al-Jannābī made common cause with Ḥamdān Qarmaṭ and 'Abdān. Also the curious intermezzo under Abū Ṭāhir al-Jannābī, when in the year 319/931–2 the Qarmaṭians of Baḥrayn acknowledged an adventurer from Persia as their ruler, whose reign, however, did not last long and ended in his death, shows that the Qarmaṭians cannot have acknowledged the Imāmate of 'Ubayd Allāh or the Mahdīship of his heir; and while it is true that our authorities do not say that they considered the Persian as the manifestation of Muḥammad b. Ismā'īl, it is perhaps permitted to make this conjecture, as it is not easy to explain their attitude otherwise. We may assume that, apart from this short interlude, the Qarmaṭians of Baḥrayn, though not acknowledging the Fāṭimids as Imāms in the full sense of the word, came to some understanding with them and paid them allegiance as the political leaders of the Ismā'īlī movement, and as a kind of lieutenants of the expected Mahdī, Muḥammad b. Ismā'īl. This ambiguous relationship would account for some of the puzzling features in the history of the Qarmaṭians, and also explain the break between the Fāṭimids and the Qarmaṭians in the reign of al-Mu'izz. It seems that in addition to the Qarmaṭians of Baḥrayn and some groups in 'Irāq, the old doctrine of the Mahdī-Qā'im Muḥammad b. Ismā'īl also found adherents elsewhere: there is, for example, some reason to assume that the Musāfirid amīr Wahsūdān b. 'Alī belonged to this trend of the Ismā'īliyya.
We have now described the great ideological changes which together with the political events – most notably the foundation of an Ismā'īlī state in North Africa – changed the face of Ismā'īlism; we must not, however, forget at least to mention another great innovation, which consisted in the adaptation of Ismā'īlism to a form of the current Islamic Neoplatonism. I think we can name al-Nasafī, dā'ī of Khurāsān and Transoxania at the beginning of the fourth/tenth century, as the man who first effected this fusion between Ismā'īlism and Neoplatonism; and while it seems that in the circles near to the Fāṭimid court the primitive system continued to form, with certain modifications, the basis of the theological doctrine, in the more advanced section of the movement al-Nasafī's system prevailed, and it lay at the basis of, for example, the writings of Abū Ya'qūb al-Sijistānī.
Here, I must bring my paper to an end; and I feel that I owe you an apology for having confined myself to sketching the internal history of the Ismā'īlī movement at its early period instead of attempting to examine Ismā'īlism, or the phenomenon of the Qarmaṭians of Baḥrayn, under the more general point of view of Islamic history and trying to estimate their significance for it. Yet the establishment of the facts, as far as is possible, has precedence over their interpretation, and since I think that many misconceptions about the most elementary problems of Ismā'īlī history are rampant amongst us, I thought I could perform a useful task by discussing them. I am sorry that this has left me no time to discuss some of the points raised by Professor Cahen in his introductory paper. At least I have attempted to give a hint about one of these problems, viz. the social background of the movement. I may put in a short remark about another point, connected with the preceding one; the question whether the Ismā'īlīs were really mainly responsible for the emergence of the Islamic guilds. I only want to point out that as far as I can see there is not a shred of tangible evidence to show that they were, and the discussion is thus really in a vacuum. I also think that there can be no question of the Ismā'īlī movement being in any sense an interconfessional movement, or of, at least, their appealing also to non-Muslims as has often been said. It was their enemies who accused them of this, in order to make them disreputable in the eyes of orthodox Muslims, and it is in slanderous pamphlets that this, connected with the even more preposterous charge that in the last resort they were cynical nihilists, was falsely put into their mouths. The original doctrine of the Ismā'īlīs, it is true, contained the archheretical points that 'Alī and Muḥammad b. Ismā'īl were prophets, and that the latter, on his reappearance as the Mahdī, would bring a new sharī'a – but this has nothing to do with appealing to non-Muslims, and the whole system could have any meaning only to those accepting the premises of Shī'ism. It was only much later, in the middle of the fourth/tenth century, that a particular, non-official group of Ismā'īlīs, the authors of the "Epistles of the Sincere Brethren", developed a philosophy of religion according to which all religions were preparations for the ultimate truth. On the other hand, the official Fāṭīmid doctrine eliminated those peculiarly heretic points of the original Ismā'īlī doctrine and is in effect a very typical Shī'ite formation, and the particular hatred which it attracted is largely due to the success of their enemies in spreading their "black legend".6
* Originally published in L'Élaboration de l'Islam, Presses Universitaires de France, 1961, pp. 99–108.
1 See Firaq al-Shī'a (for its author, cf. below n. 2), p. 58.
4 See Hamdānī, Genealogy. I propose to comment upon this text in a special article.
5 P. 61. ll. 2–4.
6 The views expressed in this paper will be more fully developed in a book to be entitled Early Ismā'īlism. [See above, part I] [Some of my conclusions (most of which were reached long years ago) coincide with views expressed by W. Madelung in an article ("Fatimiden und Baḥrainqarmaṭen", Der Islam, 34, 1959, pp. 34–88) published some months after the reading of the present paper. On the other hand, in the light of the evidence produced by him, it is possible to think that, while some of these groups professing the Mahdīship of Muḥammad b. Ismā'īl paid some kind of political allegiance to the Fāṭimids, the Qarmaṭians of Baḥrain were entirely independent.]