adherents of a form of Iranian religion often identified as a survival or revival of the Zoroastrian heresy, Mazdakism.


ḴORRAMIS (Khorramis or Ḵorramdinis), adherents of a form of Iranian religion often identified as a survival or revival of the Zoroastrian heresy, Mazdakism.

Their name first appears in 118/736, when the Hāšemite missionary Ḵedāš was repudiated for having adopted din al-Ḵorramiya (Ṭabari, II, p. 1588). After the Hāšemite revolution the Ḵorramis are encountered as rebels under Sonbādh at Rayy in 137/755, under Moqannaʿ in Sogdia in ca. 158-163/775-80, under diverse other leaders in Gorgān in 162/778-9, 179/795-6, and 181/797-8, in the Jebāl in 162/778-9, 192/807-8, 212/827-8, and 218/833, and under Bābak in Azerbaijan in 201-22/816-37 (see Sadighi; Daniel). Other revolts are reported for the Jebāl and upper Mesopotamia under the caliph al-Wāṯeq (r. 842-47); and in 321/933 ʿAli b. Buya stormed some Ḵorrami fortresses in the Karaj region (Neẓām al-Molk, chap. 47, par. 13; Michael the Syrian, IV, p. 542, tr., III, p. 109; Meskawayh, I, p. 278, tr. , IV, p. 316; Ebn al-Aṯir, VIII, p. 269). There are also scattered reports on Ḵorrami communities down to the 12th and even the 13th centuries (ʿAwfi, p. 274).

They are invariably described as rural. They had no single overall organization or creed and seem to have differed from one locality to the next, but all are said to have believed in periodic and/or continuous incarnation or indwelling of the divine in man (ḥolul, tanāsoḵ), reincarnation of the human spirit (rajʿa, tanāsoḵ al-arwāḥ) in accordance with merit, and, at least in western Iran, in kindness to all living beings, sometimes coupled with abstention from meat-eating. The feature most commonly associated with them is a practice denigrated as “sharing their womenfolk” (ebāḥat al-nesāʾ), to which ebāḥat al-māl “sharing of property” is occasionally added. No writings by them are preserved, or even mentioned. The Ketāb al-ḵorramiya mentioned by Maqdesi (II, p. 20) was probably a book about them, perhaps, but hardly Abu Zayd Balḵi’s chapter on them in his ʿOyun al-masāʿel wa’l jawābāt (Ebn al-Nadim, 406.7, tr. p. 817). But they did have learned men, and Muslims sometimes engaged in disputation with them (Abu Tammām, p. 77, tr., p. 76; Maqdesi, III, p. 122; Masʿudi, Ketāb al-tanbih, pp. 353 f.; Moqaddasi, pp. 398 f.).


Though Ḵorramism and Mazdakism are undoubtedly related, the Ḵorramis are too widely attested to be the residue or a revival of a defeated sect. Their presence stretched from Isfahan northwards through the Zagros mountains to Qāšān, Qom, Rayy, Hamadān, Deylam, Azerbaijan, and upper Mesopotamia/Armenia, with particularly dense attestation in the Jebāl. Eastwards it ran through Gorgān to Khorasan, Ṭoḵārestān (including Balḵ and other parts of what is now Afghanistan), Sogdia (including the countryside around Bukhara, Samarqand, Keš and Nasaf), to Ṣāṣ, Ḵojand, Ilāq, Kāsān, and Farḡāna beyond the Jaxartes (Syr Darya). They were found from the mountain ranges of Anatolia to those of Tien Shan, far beyond the boundaries of the Sasanian empire. This suggests that we should see Ḵorramism as the religion of rural Iran, a Zoroastrian “low church” (Madelung, 1988, p. 3), from which the founders of Mazdakism emerged, rather than as a heresy which they founded. Pursued in local organizations such as that which Bābak took over in Azerbaijan, this “low church” will have functioned much like rural Sufism in later times and should not be envisaged as intrinsically rebellious. Its organized nature did however facilitate revolt when the Ḵorramis were politicized.

Unlike the founders of Mazdakism, the Ḵorramis do not seem to have subscribed to revolutionary ideas regarding women and property, but they certainly had practices offensive to the Muslims. Reconstructing these practices is mostly impossible. Some reports seem to relate women being lent out one way or the other, to guests, priests, or other men, in displays of generosity or with a view to obtaining a blessing or good offspring; others may refer to fraternal polyandry, documented for Ṭoḵārestān and other parts of Afghanistan (and well beyond, into India and Tibet), in Chinese sources, in Biruni (Hend, p. 52; tr., I, p. 108), and now also in Bactrian documents of the fourth and the eighth centuries CE (Sims-Williams, nos. A, X, Y). Fraternal polyandry is a system whereby brothers inherit the property of their parents without dividing it up, cultivate it in common, and share a wife, whose sons will jointly take over the family property in their turn. The system allows the property to pass intact from one generation to the next and is attested above all in mountainous areas where the land is poor. It is in some sense quite true that women and property are shared in polyandrous societies, but not in the sense that they are free for all to use as they please. An Indian high court judge who reported on fraternal polyandry in north India in the 1950s called it “a sort of family communism in wives … a joint family both in property and wives” (Peter, p. 83). It would seem to have been this family communism which Zarādošt of Fasā and Mazdak elevated into a utopian vision: all members of the Sasanian kingdom had to behave as if they were brothers. Explaining how communist ideas could have developed in Iran has long been a problem, and it has generated some far-fetched theories of influence from the Greek-speaking world (cf. Crone, 1991, p. 28), but they are hardly necessary.


The idea of the same divine being appearing in different incarnations is attested in the Bahrām Yašt (Yt. 14; cf. Yt. 8.13, 16, 20; see BAHRĀM [1]), but something close to the Ḵorrami conception is first attested in the Book of Elchasai, composed around 116 CE in “Parthia,” that is, Mesopotamia under Parthian rule, by a Jewish Baptist and perhaps a follower of Christianity. The book itself is lost, but the idea that the same divine being appears time and again, putting on different bodies, was shared by the Baptist followers of Elchasai in lower Iraq, the Elchasaite Baptists in Rome, where the book had been brought by a Greek-speaking Syrian, and by diverse Baptist readers of the book in Palestine, including the Ebionites and Nazoreans, from whom it went into the Pseudo-Clementines (Luttikhuizen, modified by Merkelbach and Cirillo). In Palestine, the doctrine was understood conservatively: only Adam and Christ, the first and the last, were genuine incarnations; in between, the divine being only appeared to the patriarchs (cf. Gieschen, pp. 208 f.). But the Elchasaite baptists, from whom Mani broke away, seem to have been understood all of them as incarnations, and Mani himself certainly did. To him, the Buddha, Zoroaster, and Jesus were divine beings who had come “without a body” to take up abode in human beings; he himself had become “one spirit in one body” with the Paraclete, and all apostles were incarnations of the pre-existing “Apostle of Light,” who “puts on the saints as his garments.” All were really a single spirit (Cirillo, pp. 50-52; Lieu, pp. 236, 242, 246; Gardner and Lieu, p. 75; Gardner, p. 132).

The Ḵorramis envisaged the incarnation of the divine now as periodic and now as continuous. Moqannaʿ held God’s spirit to have taken up abode in messengers (rosol) at long intervals, starting with Adam and running via the founders of religions, including Moḥammad, to Moqannaʿ himself, the Mahdi (q.v.). The Ḵorramis of the west similarly believed in “the change of the name and the body, claiming that all the messengers, with their diverse laws and religions, come into possession of a single spirit” (Maqdesi, IV, p. 30; cf. Clementine Homilies, III, p. 20). But more commonly we hear of the divinity as inherent in a continuous chain of community leaders (imams). A Rāvandi (on whom, see below) executed before the Hāšemite (also known as Abbasid) revolution, for example, held that the spirit which had been in Jesus had entered ʿAli and passed from him via other imams to the Abbasid Ebrāhim al-Emām, so that all the imams were gods (Ṭabari, III, p. 418). Either way, the divinity was envisaged now as light (e.g., Nawbaḵti, p. 29; Qomi, no. 80) and now as spirit, sometimes identified as the holy spirit (e.g., Ps.-Nāšeʾ, par. 56). These conceptions should perhaps be related to the Iranian concept of xᵛarəenah (see FARR[AH]), the divine light and spiritual force which is shared by Ahura Mazdā, Zoroaster, the legendary and historical kings, and the future saviors.

The sudden prominence of the idea of divine incarnation in 2nd/8th-century Iran reflects the fact that large numbers of Iranians had been recruited into Muslim armies, and thus into Muslim society, by the Hāšemiya in Khorasan and by ʿAbd-Allāh b. Moʿāwiya in western Iran. The recruits seem regularly to have cast the leaders of their new religious community as divine. Many of them seceded however, when, as repeatedly happened, the man to whom they owed their presence in Muslim society was killed. The first waves of secession came already before the revolution, triggered by the execution of Ḵedāš in 118/736 and by the killing of ʿAbd-Allāh b. Moʿāwiya in Abu Moslem’s jail in Herat in 131/748-9. When Abu Moslem was himself killed in 137/755, more extensive waves of secession followed, initially among his own by now unemployed troops, eventually further afield, among people uprooted by the revolution or adversely affected by the massive upheavals that followed. In all three cases the secessionists cast the victim as the true imam (and/or prophet), to reconstitute themselves as separate communities under leaders of their own. They continued to trace the imamate from Moḥammad: the leadership of the community had passed to ʿAli, Ebn al-Ḥanafiya, and Abu Hāšem, who bequeathed it to ʿAbd-Allāh b. Moʿāwiya (according to the Janāḥiya and Ḥāreṯiya/Ḥarbiya) or to a member of the Abbasid family (according to former members of the Hāšemiya), from whom it has passed to Ḵedāš (according to the Ḵedāšiya) or Abu Moslem (according to the Moslemiya). Their imams thereafter were usually Iranians and never Hāšemites (Ps.-Nāšeʾ, par. 52). By the 12th century, the Ḵorramis in Azerbaijan had extended the imamate chain back in time to include the Persian kings and taken to calling themselves Pārsis. They had also become ʿAlid Shiʿites: God had manifested himself in Moḥammad, ʿAli, and Salmān al-Fārsi, and their two current leaders were in the position of Moḥammad and ʿAli, the light manifesting itself now in three persons and now in one or two (Madelung, 1988, pp. 9-12).

In Iraq, too, the murder of Abu Moslem severely tested the loyalties of the Korāsānis, but here the so-called Rāvandiya reacted by casting al-Manṣur as the Mahdi, the full manifestation of God introducing the heavenly realm, justifying his killing of Abu Moslem (cast as his prophet) on the grounds that his will was inscrutable (he killed his prophets and messengers as he wished; Nawbaḵti, p. 47). If he wanted to make the mountains move, they would move, they said, and if he wanted them to pray with their backs to the qebla, they too would obey (Balāḏori, III, p. 235; cf. Ebn al-Moqaffaʿ, par. 12). The spirit of Adam had taken up abode in one of the caliph’s officers, Oṯmān b. Nahik; and another, Hayṯam b. Moʿāwiya, was a manifestation of Gabriel. The Rāvandiya sold their possessions and jumped naked or dressed in silken clothes (presumably those of the people of paradise; cf. Qorʾān 18:31) from walls and other high places in Iraq and Syria, thinking they would fly to heaven, or that they had become angels (Ṭabari, III, pp. 129 f., 418; Balāḏori, III, pp. 235 f.; Azdi, p. 173; Theophanes AM 6250). Like Moqannaʿ and some followers of Ebn Moʿāwiya, they thought in terms of seven eras (see dawr), but they construed the eras with reference to imams rather than seven messengers or seven Adams (Ebn al-Jawzi, VIII, pp. 29 f.; cf. Ps.-Nāšeʿ, par. 58). The episode is variously set in 136, 137, 139, 140, 141, or 142, but more than one may have been involved, for in 141-42 it was the caliph’s son, that is, al-Mahdi, rather than al-Manṣur himself that the Rāvandiya deified (Theophanes AM 6252; for the Basran episode mentioned there, cf. Dinavāri, p. 380). 

Moqannaʿ was among those who remained in Abbasid service after Abu Moslem’s murder, in Marv, where he turned rebellious some time after the downfall of his employer, ʿAbd al-Jabbār, in 141/758. The sources on his message do not mention the imamate, and it is probably by confusion with the Moslemiya that they include Abu Moslem in his sequence of messengers, making Moqannaʿ himself the Mahdi, the eighth rather than the seventh (Crone, forthcoming; cf. Ebn Dāʿi Rāzi, p. 179), though Moqannaʿ may well have deified Abu Moslem in another prophetic capacity, or as king (cf. Ṯaʿālebi, no. 14). There is no reference to the imamate in the earliest source on Bābak either, apparently because the Ḵorrami cult organization he led was still wholly non-Islamic. Bābak succeeded Jāvidān b. Ṣahrak as leader on the ground that the latter’s spirit had passed into him (Ebn al-Nadim, 407.11, 17; tr., II, pp. 820 f.; Ṭabari, III, p. 1015), and identified himself as “the spirit of the prophets” (Abu’l-Maʿāli, p. 62), presumably meaning that the holy spirit which had moved the prophets of the past, including Jāvidān, was now active in him. Unlike early Christian prophets, Bābak was not simply a passive instrument like a lyre or flute through which the holy spirit would speak from time to time (cf. Aune, pp. 204, 315 f.). Rather, it dwelled in him permanently, rendering him divine (Abu’l-Maʿāli, p. 62; Ebn al-Nadim, 406.10; tr., II, p. 818). He was not the Mahdi, however, and he did not wear a veil; rather he was a community leader. A full incarnation of the deity had appeared in the Mesopotamia/Armenia region around the same time. His followers are described among the Ḵorramis as pagans and “Magians in their cult,” suggesting that they venerated fire; but they were also Christians of sorts, for their long-awaited king called himself the holy spirit and Christ, as well as the Mahdi. He was held to be divine and, like Moqannaʿ, wore a veil. After his death and that of a successor, his followers accepted Bābak as their leader (Michael Syr., IV, pp. 508 f.; tr., III, pp. 50-52; Chron. 1234, pp. 25 ff.; tr., pp. 17-19; Bar Hebraeus, p. 131, all from Dionysius of Tell-Maḥré (a chronicle written in 774 CE under the name Dionysius, a monk from Tell-Maḥré in Mesopotamia).

The Ḵorramis are reported to have believed in continuous prophecy (Abu ʿIsā in Ebn al-Malāḥemi, p. 584; Biruni in Fück, p. 80; Maqdesi, III, p. 8, IV, p. 30). It is not easy to tell whether they held that there could be prophets such as Adam, Jesus, or Moḥammad in the future or that the sequence of divine community leaders (prophets, spirit-bearers, and the like in Christian parlance, imams in Muslim parlance) would continue forever. They may not have distinguished sharply between the two, the key difference being rather between such figures and the final, full incarnations of the divine, the Christ or Mahdi, with whom the material world, or the current cycle, would come to an end.

Some 50 years after Bābak’s death, Bābak had come to be identified as a descendant of Abu Moslem, implying that his followers now thought in terms of the imamate (Dinavari, p. 397). Thereafter the Moslemi nature of the Ḵorramis in Azerbaijan and elsewhere in western Iran is well attested (see, e.g., Masʿudi, Moruj al-ḏahab IV, par. 2398).

The sources regularly use the term tanāsoḵ for periodic and continuous incarnation of the divine, but no reincarnation was involved. The divine spirit was normally envisaged as taking up abode in adult human beings (for Jesus as an exception, see Maqdesi, III, p. 122). Moqannaʿ’s followers held God’s spirit to have entered Moḥammad in the encounter described in Qorʾān 53:3-10 (Abu Tammām, p. 76; tr., p. 75). Others claimed that the divinity had passed, in a form visible to ʿĀʾeša, from Moḥammad to ʿAli, who had ingested it (Ps.-Nāšeʾ, par. 56); and Bābak was an adult when Jāviḏān’s spirit passed into him. Another term for periodic incarnation is qalb, explained in connection with the Ḵedāšiya as the belief that God can change (yaqliba) Himself from one shape (ṣura) to another and clothe himself in different visible forms (manāżir). In support of this, the Ḵedāšiya would adduce the ability of a lesser being such as Gabriel to do the same, as recorded in Hadith. Here Gabriel merely appears in the guise of known or unknown people, however (Ps.-Nāšeʿ, par. 49). In the Ḵorrami conception the deity did not merely simulate a body or create one for itself; rather it entered a person with an identity of his own. We also hear of zanādeqa’l-naṣārā, “dualist/quasi-Manichaean Christians,” who held that the spirit in Jesus was the spirit of God, from the essence of God (ruḥ Allāh min ḏāt Allāh), explaining that God would enter a human being when He wished to convey His commands and speak human language (Ebn Ḥanbal, p. 19).


According to Euboulos (date uncertain), quoted by Porphyry (d. ca. 305), the Magi practiced various degrees of vegetarianism, because “it is the belief of them all that metempsychosis is of the first importance.” Porphyry also refers to a certain Pallas, who probably wrote under Hadrian (d. 138) and who explained the Mithraic habit of giving animal names to initiates as an allegory of human souls, which, they (the Magi) said, “put on all kinds of bodies” (Porphyry, IV, 16). Both reincarnation and non-violence to animals reappear in Manicheism. Mani is usually assumed to have picked up the doctrine of reincarnation from Buddhism in India or alternatively from the Greeks (Heinrichs, pp. 97 ff.; Bryder, pp. 488 f.; cf. Biruni, Hend, p. 27; tr., I, pp. 54 f.), but he only traveled to India after having formed his system, and his closely related doctrine of non-violence went far beyond anything found in Buddhism in that it extended to plants, trees, air, earth, and even stones. Thus, the Indian influence would have had to come from Jains. The possibility of Jain influence has in fact been aired (see Fynes; Gardner; Deeg and Gardner), but although the coincidences are striking, both Indian and Greek inspiration would seem unnecessary in that Mani’s views on reincarnation and non-violence were intimately connected with his conception of the world as a mixture of light and darkness. In this conception, light (divinity) was present in everything in this world and circulated thanks to natural processes, and everything endowed with light was live, sentient, and could feel pain. In the opinion of Šahrastāni (p. 133; tr., I, p. 511), all nations, including the Zoroastrians, had groups who believed in reincarnation. Abu Ḥātem Rāzi (Ketāb al-eṣlāḥ, p. 159) claims that Mazdak, a Zoroastrian priest, believed in it, and Abu ʿIsā al-Warrāq implies the same: according to him, Mazdak permitted the killing of opponents on the grounds that it would liberate their spirits from their harmful bodies, that is, save them from bad reincarnations (Ebn al-Malāḥemi, 584.4; cf. Šahrastāni, p. 193, tr., I, p. 663; ʿAbd-al-Jabbār, V, p. 65; tr. Monnot, p. 237). All the Ḵorramis are said to have believed in reincarnation (Ps.-Nāšeʾ, pars. 57 f.; Šahrastāni, p. 185, tr., I, p. 641), probably including the Rāvandiya (Ebn Qotayba, p. 227), though Baḡdādi did not think so (255.6).

Practically all further details come from an account relating to the followers of ʿAbd-Allāh b. Moʿāwiya, though it is also cited with reference to other groups (Ps-Nāšeʾ, pars. 57 f.; Nawbaḵti, pp. 32-34, 35-37; Qomi, nos. 93, 97 f.; Abu Ḥatem Rāzi, pp. 308-10; Ebn Dāʿi Rāzi, pp. 87 ff.; Freitag, pp. 9 ff.). The sectarians denied the resurrection, insisting that there was no world other than this one or, as they also put it, that the resurrection consisted in the spirit leaving the body for another body or form (qālab, ṣura). Bodies were like clothes that got worn out or houses one moved out of, and only the spirit was rewarded or punished. Obedient spirits would be moved into pure bodies of beautiful shapes for pleasurable lives and, according to some, continue to move up in the ranks of goodness, purity, and pleasure until they became angels and acquired pure bodies of light, while disobedient spirits would move into impure and ugly bodies of dogs, monkeys, pigs, serpents, dung beetles, scorpions, and the like, to be tormented forever. Some associated reincarnation with cycles: God had created seven Adams, corresponding to seven eras; each Adam would initiate an era of 50,000 years, at the end of which the righteous would be raised to the first heaven as angels while the rest would be placed below the earth. The ants, scarabs, and dung beetles that crawled around in people’s houses were nations that God had destroyed in the past. At the end of every era, those who had already been saved or damned would move further up or down to the next heaven or earth. When all the seven eras were over, religious worship would come to an end. They also had doctrines about the “shadows,” presumably along the lines known from the later Ketāb al-haft wa’l-aẓella attributed to al-Mofażżal b. ʿOmar al-Joʿfi (Halm, pp. 24 ff.), though this book differs in being Gnostic in character and focused on ʿAli. Some operated with different cycles, claiming that believers would assume human bodies for periods of 10,000 years followed by 1,000 years in animal bodies, in the best form in both cases, whereas unbelievers would spend 10,000 years in animal bodies of the worst kind, followed by 1,000 years as miserable humans such as tanners and sweepers. The alternations were meant as tests, and apparently these cycles would go on forever: there is no reference to release, whether individual or collective (cf. their eternalism in Ašʿari, pp. 6, 46).  Some believers in reincarnation claimed to recognize each other from one period to the next, typologically rather than individually, as the people who had been with Noah in the Ark, who had followed the other prophets in their time, and who had been the Companions of Moḥammad (ṣaḥāba). They would take their names, claiming that that their spirit was in them (similarly Ebn Qotayba, p. 227). Some called themselves al-ḥawāriyun among themselves (Baḡdādi, p. 236).

The Ḵorrami term for reincarnation was rajʿa (Nawbaḵti, pp. 33, 37; Qomi, no. 98) and they would adduce verses 6:38; 29:64; 32:26; 35:24; 82:8, 84:19 and 95:4-6 in support of it. The poet Koṯayyer ʿAzza believed in tanāsoḵ al-arwāḥ and rajʿa, probably meaning the same, and Sayyed al-Ḥemyari held that it was possible to be reincarnated as an animal (Abu’l-Faraj Eṣfahāni, VIII, p. 243; IX, pp. 4, 17-19; cf. Wellhausen, p. 93). Much later we learn that Mardāvij, the 4th/10th century military adventurer from Gilān, claimed to have the spirit of Solomon in him (Ebn al-Aṯir, VIII, p, 298, year 323). Like other Ḵorramis, he was both himself and somebody else.


As mentioned above, the Magi (according to Euboulos in Porphyry) who believed in reincarnation practiced varying degrees of vegetarianism. Vegetarianism and pacifism are also reported for Kawād I, the Sasanian king who adopted the utopia of Zarādošt of Fasā (Crone, 1991, p. 26). But though Bābak complained that the hands and breath of his prison-guard stank of meat (Ṭabari, III, p. 1228), he and his followers ate meat on a ritual occasion (Ebn al-Nadim, 407.19; tr., II, p. 821) and he would hunt, too (Ṭabari, III, p. 225f). Of the Ḵorramis of western Iran in general, however, we are told that they believed in “acts of charity (afʿāl al-ḵayr) and refraining from killing and inflicting harm on souls” (Ebn al-Nadim, 406.4; tr., II, p. 817), and that they took great care to avoid bloodshed, except when they rebelled (Maqdesi, IV, p. 31; wrongly attributed to the Mobayyeża along with Moslemi beliefs in Abu Tammām, p. 78; tr., p. 77). They also disapproved of speaking ill of the adherents of other religions as long as they did not seek to inflict harm on oneself: all messengers were really the same, and the followers of all religions were right as long as they believed in requital after death (Maqdesi, IV, pp. 30 f.). To the Ḵorramis (now Pārsis) of 6th/12th-century Azerbaijan, bloodshed was one of the five deadly sins, as was hurting other people or anything living; even hammering a peg into the earth was forbidden lest it be hurt by it (Madelung, 1988, p. 10). Their dislike of bloodshed was not linked with asceticism. They revered wine and insisted on the lawfulness of all pleasures as long as they did not harm others (Maqdesi, IV, p. 31); the “old Mazdak” (i.e., Zarādošt of Fasā) had ordered them to partake of all pleasures (Ebn al-Nadim, 406.2; tr., II, p. 817).

Their doctrine of non-violence was not sufficiently prominent for the Muslims to have a name for it. They were also notoriously violent as rebels, and anything but charitable in their visions of revenge. But Maqdesi (IV, p. 31) found those he met in Māsabaḏān and Mehrejānqaḏaq to be extremely kind people.


Abu ʿIsā had heard that the Ḵorramis believed light and darkness to have always existed and that the “Mazdaqiya” subscribed to a (Zoroastrian?) cosmological doctrine which had also influenced the Manicheans, to the effect that darkness was ignorant and blind and had swallowed some of the light by accident (Ebn al-Malāhemi, pp. 583 f., 598; cf. ʿAbd al-Jabbār, V, p. 16, cf. p. 64 f., tr. Monnot, pp. 163 f., 237; Šahrastāni, pp. 185, 192 f., tr., I, pp. 641, 663, turning the Mazdaqiya into Mazdak himself; and cosmology and cosmogeny i and iv). These Mazdaqiya seem to have been a subdivision of the Ḵorramis in (urban?) Iraq (Masʿudi, Ketāb al-tanbih, p. 353.ult.; Šahrastāni, p. 113, tr., I, p. 449).  Maqdesi (I, p. 143) credits the Ḵorramis with the doctrine that everything began as light, probably by misunderstanding Abu ʿIsā, though there must in fact have been many Ḵorrami cosmologies. Qomi (no. 127) more broadly says that “most of their doctrines are those of the Zoroastrians.”

Some Ḵorramis held that the separation would come about accidentally (Šahrastāni, p. 193; tr., I, p. 663, replacing Abu ʿIsā’s Mazdaqiya with Mazdak; cf. Ebn al-Malāḥemi, pp. 583 f.). Others seem to have been eternalists (cf. above), and there were also Ḵorramis inclined towards Manicheism in unidentified respects (Ebn al-Malāḥemi, 584.1). Like the Manicheans, though in a more positive vein, they seem to have thought of light, the divine element, as all-pervasive. It is this idea, variously called animism, pan-psychism, or pantheism in the modern literature, that lies behind their belief in reincarnation, non-violence, and divine incarnation alike (cf. Šahrastāni, p. 133; tr., I, pp. 511 f.; Malaṭi, p. 17). Ultimately, it was the same light or spirit which manifested itself again and again in different forms and strengths, in humans, animals, and inanimate things alike, rendering all of them live and sentient. There was no sharp distinction between the divine, the human, and the animal worlds, or between past and present: just as the same divine being incarnated itself time and again, so the same people lived on again and again, in human or animal form. The fact that there are Indian analogues to the key Ḵorrami and Manichean doctrines (avatāra, saṃsāra, ahiṃsā) should probably be credited to the shared roots of Indian and Iranian religion rather than Indian influence.


The Ḵorramis were notorious for not living by Islamic law. Their villages had no mosques, and if they did, only for outsiders, and although they (or some of them) would teach their children the Qorʾān, they did not pray, observe the dietary taboos of Islam, or perform ablutions according to Islamic precepts; and they married only among themselves (Moqaddasi, pp. 398 f.; Eṣṭaḵri, p. 203; Baḡdādi, p. 252; Abu Tammām, p. 77, tr., p. 76). But they had their own norms, for which they would consult their imams, and purity was of the utmost importance to them (Maqdesi, IV, p. 31). They argued their way out of Islamic law by interpreting it allegorically, holding the commands and prohibitions to stand for persons or activities one should seek out or avoid; or they said that the (literal meaning of) the law did not apply to those who knew the imam (Abu Tammām, p. 77, tr., p. 76; Ps-Nāšeʾ, pars. 48 f., 59). Since the entire community knew the imam, this was presumably a way of legitimating an ancestral way of life rather than antinomian behavior by perfected individuals, though the tone in which it is reported often suggests the latter. The Ḵorramis undoubtedly saw themselves as the only saved, however. As the only people of Paradise they were free to take the women, children, and property of other Muslims when the apocalypse came (and they rebelled), but it does not follow that they, or the perfected among them, were free to take what they wanted at all times, let alone among themselves (it is misrepresented as a doctrine of Mazdakite sharing in Biruni’s reporting on Moqannaʿ). The Ḵedāšiya are credited with extreme hostility to outsiders even when they were unable to rebel; they interpreted jihād (see ISLAM IN IRAN xi. JIHAD IN ISLAM) as meaning killing opponents by assassination, strangling, crushing, or poisoning, probably meaning that even such methods were allowed rather than that such methods were ritually prescribed; their property could be taken, and a fifth had to be given to the imam, as if it were booty taken in war (Ps.-Nāšeʾ, par. 49). How far the sense of being the only elect spilled over in a sense that individuals could reach a state of such perfections that they were above the law, even their own, under normal conditions is hard to say. Some may have held that they could become angels, divine beings, or people of Paradise in the here and now, and the possibility that “transgressive sacrality” was a feature of some communities cannot be ruled out. One account of Rāvandi practice in the east (Ṭabari, III, p. 418) could refer to sexual rituals used in Buddhist (or Šaivite, cf. Škoda) Tantrism. Maqdesi (IV, p. 31) confirms that some Ḵorramis endorsed the sharing of womenfolk (ebāḥat al-nesāʾ), with the women’s consent, but does not tell us what it consisted in. The Pārsis of 6th/12th-century Azerbaijan were strictly monogamous and forbade both divorce and the purchase of slaves. Yet they held that women were like a well that anyone could drink from (Madelung, 1988, p. 10). Did they mean that adultery was not a sin, or that women could be given to holy men for blessing, or, on the contrary, that since women were common to all, they had to be distributed equally, so that nobody could have more than one? It is impossible to tell. 


We do not know what the Ḵorramis were called before the coming of Islam, but Ḵorramdin “adherent of the joyous religion” is a term coined on the model of Behdin “adherent of the good religion,” that is, Zoroastrianism, and could be a self-designation. The heresiographers usually relate the name to the scandalous sexual practices of the Ḵorramis, but if the name had been coined with reference to them, a more offensive term than “joyous” would surely have been chosen. However this may be, when the sources speak of Ḵorramis, it is usually the Moslemiya that they have in mind, usually those of the Zāgros and Alborz mountains or Azerbaijan, but occasionally also those of eastern Iran (see, e.g., Balḵi in Ebn al-Nadim, 408.13; tr., p. 824). Ps.-Nāšeʾ (probably Jaʿfar b. Ḥarb) identifies the Ḵedāšiya as the Ḵorramis in Khorasan, the Moslemiya as those of the Jebāl, clearly over-simplifying. In Khorasan they came to be known as the Bāṭeniya (Masʿudi, Moruj al-ḏahab VI, par. 2399). Ṯaʿālebi mentions “Ḵosrawiya and Ḵorramiya” in, perhaps, the Bādḡis region (Houtsma, p. 35; tr., p. 33). But in the east we more commonly hear of “Whiteclothed ones” (Mobayyeża, Sepid-jāmagān), often identified as Moqannaʿ’s followers, though they existed before him and are found in areas far beyond those involved in his revolt (Gardizi, p. 273; Moqaddasi, p. 323; Neẓām-al-Molk, chap. 47.22; cf. Šahrastāni, p. 115; tr., I, pp. 454 f.). In Gorgān we hear of “Redclothed ones” (Moḥammera, Sorḵ-jāmagān; Daniel, p. 147), a term also used to refer to the Ḵorramis of the west (e.g., Ebn al-Nadim, 405.ult.; tr., II, p. 817). We are not told what the differences between them were. Masʿudi uses the term Moḥammera to refer to a smaller group, distinguishing them from the “Mazdaqiyya, Māhāniya, and others” (Ketāb al-tanbih, 353.ult.). The Māhāniya were Iranianized Marcionites, that is, Christians of sorts (Ebn al-Malāḥemi, p. 589), as was clearly true of many Ḵorramis before they were Islamized. Masʿudi also divided the Ḵorramis into Kuḏakiya (or Luḏakiya), Kuḏšahiya (or Luḏšahiya), and others, placing them in the mountainous regions of western Iran with a wealth of place names, and more cursorily mentioning that they were also found in Khorasan and the rest of Iran (Ketāb al-tanbih, p. 353; cf. idem, Moruj al-ḏahab VI, par. 2399). Abu Ḥātem Rāzi (pp. 305 f.; Madelung, 1986, p. 65), on the other hand, says that they were known as Kuḏakiya and Ḵorramiya in the Isfahan region, as Mazdakiya and Sonbādiya in Rayy and elsewhere in the Jebāl, as Moḥammera in Dinavar and Nehāvand (al-Māhayn), and as Ḏāqufiya (or Ḏafuliya) in Azerbaijan (repeated, slightly differently, by Šahrastāni, p. 132; tr., I, p. 508). Again, we are not told about the differences between them. According to Abu Ḥātem Rāzi (p. 298), Ebn Moʿāwiya’s followers, or at least the Ḥāreṭiya, were known as Ḵorramdinis, while Šahrastāni (p. 113; tr., I, p. 449) held all Ḵorramis to spring from them. They secured toleration through heavy payments to the ruler (Moqaddasi, p. 399; Ṭaʿālebi, p. 38).


Ḵorramism does not seem to have survived the Mongol invasions. In western Iran and Anatolia, however, belief in divine incarnation and human reincarnation reappeared along with varying degrees of pantheism in the heresies of the Ḥorufis, Noqṭavis, Bektašis, Qezelbaš, Yazidis, Ahl-e Ḥaqq and others, filtered through Sufism (cf. Pirouzdjou; Babayan; Šafiʿi-Kadkani, pp. 55 ff.). All three beliefs had appeared in Sufism already among the 3rd/9th century Muʿtazilite Sufis, including Aḥmad b. Ḥāʾeṭ (or Ḵābet), who shared the pan-psychism of the Ḵorramis and Manicheans: they held all animate beings and inanimate things, including stones, to be endowed with rationality, and they also believed in reincarnation (Jāḥez, IV, p. 288; van Ess, III, pp. 430-45). Other Sufis believed in ḥolul, claiming that God could dwell in humans and wild animals, especially beautiful ones (Ašʿari, pp. 13 f.; Baḡdādi, p. 245; Maqdesi, V, p. 148), or that he might dwell in the entire world, animate or inanimate, which they called “the universal manifestation” (al-ẓohur al-kolli). To those who subscribed to this view, the idea of divine indwelling of the spirits by reincarnation (ḥolul al-arwāḥ be’l-taraddod) was unproblematic, as stated by Biruni (Hend, p. 29; tr., I, pp. 57 f.). The idea that women and property could be freely taken also reappears, both in Sufism and elsewhere (e.g. Malaṭi, p. 73f; Nasafi, p. 359; Haftād o seh mellat, nos. 27, 72). Evaluating such reports is notoriously problematic, but we do at least know from Ḥorufi sources that some Ḥorufis held themselves entitled to take everything in existence, believing that they were already in a Paradisical state and thus freed from the constraints of the law: the ʿāref could take whatever was within reach and should endeavor to obtain what was in the hands of others (Browne, pp. 75 f.). This is close to Moqannaʿ’s preaching in that the concern is with taking the property (and women?) of others, not with sharing within the community. All in all, the legacy of Ḵorramism and other pan-psychist forms of Iranian religion in Islam seems to be much greater than that of Zoroastrianism as known from the Pahlavi books.



Primary sources.

ʿAbd al-Jabbār, Moḡni fi abwāb al-tawḥid wa’l-ʿadl, Cairo, 1959-65.

P. Bedjan, ed., Acta Martyrum et Sanctorum, 7 vols., Paris, 1890-97.

O. Braun, tr., Ausgewählte. Akten persischer Märtyrer, Kempten, 1915.

Abu’l-Faraj Eṣfahāni, Ketāb al-aḡāni, Cairo, 1927-74. 

Abu Ḥātem Rāzi, Ketāb al-eṣlāḥ, eds. Ḥ. Minučehr and M. Moḥaqqeq, Tehran, 1998.

Idem, Ketāb al-zina, ed. ʿA. S. al-Sāmarāʾi in al-Ḡoloww wa’l-feraq al-ḡāliya fi’l-ḥaḏāra al-eslāmiya, Baghdad, 1972, pp. 227-312.

Abu’l-Maʿāli, Bayān al-adyān, ed. H. Reżā, Tehran, 1964.

Abu Tammām, in W. Madelung and P. E. Walker, eds. and trs., An Ismaili Heresiography, Leiden, 1998.

Aḥsan al-taqāsim fi maʿrefat al-aqālim, ed. M. J. de Goeje, BGA, Leiden, 1877.

Ašʿari, Maqālāt al-eslāmiyin, ed. H. Ritter, Istanbul, 1929-33.

Awfi, Jawāmeʿ al-ḥekāyāt wa lawāmeʿ al-rewāyāt, ed. J. Šeʿār, Tehran, 1995. Azdi, Taʾriḵ al-Mawṣel, ed. ʿA. Ḥabiba, Cairo, 1967. 

Baḡdādī, al-Farq bayn al-feraq, ed. Moḥammad Badr, Cairo, 1328/1910. Balāḏori, Ketāb ansāb al-ašrāf, ed. ʿA.-ʿA. al-Duri, Wiesbaden, 1978.

Bar Hebraeus, Chronography, tr. and ed. E. A. W. Budge, 2 vols, London, 1932. Abu Rayḥān Biruni, Ketāb fi taḥqiq mā le’l-Hend, ed. E. [C.] Sachau, London, 1887; tr. E. C. Sachau, Alberuni’s India, London, 1910.

Idem in J. Fück, “Sechs Ergänzungen zu Sachaus Ausgabe von al-Biruni’s ‘Chronologie Orientalischer Völker’,” Documenta Islamica Inedita, Berlin, 1952, pp. 69-98. 

Chronicon ad A. C. 1234 pertinens, ii, ed. I. B. Chabot, Louvain 1916; tr. A. Abouna, Louvain 1974. 

Clementine Homilies, ed. A. Roberts and J. Donaldson, 25 vols., Edinburgh, 1870.

Dinavari, al-Aḵbār al-ṭewāl, ed. V. Guirgass, Leiden, 1888.

Ebn al-Aṯir, al-Kāmel fi’l-taʾriḵ, ed. C. J. Tornberg, 12 vols., Leiden, 1851-76; repr., 13 vols., Beirut, 1965-67.

Ebn Dāʿi Rāzi, Tabṣerat al-ʿawāmm fi maʿrefat maqālāt al-anām, ed. ʿA. Eqbāl Āštiāni, Tehran, 1934.

Ebn Ḥanbal, al-Radd ʿalā‘l-Zanādeqa wa’l-Jahmiya, Cairo, 1973. 

Ebn al-Jawzi, al-Montaẓam fī taʾrīḵ al-molūk wa’l-omam, ed. M. A. ʿAṭāʾ and M. A. ʿAṭāʾ, Beirut, 1992.

Ebn al-Malāḥemi, Ketāb al-moʿtamad fi oṣul al-din, ed. M. McDermott and W. Madelung, London, 1991.

Ebn al-Moqaffaʿ, al-Resāla fi’l-ṣaḥāba, ed. and tr. Ch. Pellat, Paris, 1976.

Ebn al-Nadim, Ketāb al-fehrest, ed. M. R. Tajaddod, Tehran, 1971; tr. B. Dodge, as The Fihrist of al-Nadīm, 2 vols., New York, 1970.

Ebn Qotayba, Ketāb al-ʿoyun wa’l-ḥadāʾeq fi aḵbār al-ḥaqāʾeq, part III, ed. M. J. de Goeje, Leiden, 1871.

Eṣṭaḵri, Ketāb Masālek al-mamālek, ed. M. J. de Goeje, Leiden, 1870; 2nd ed., Leiden, 1927; 3rd ed., Leiden, 1967.

Gardizi, Zayn al-aḵbār, ed. ʿA.-Ḥ. Ḥabibi, Tehran, 1984.

Jāḥez, Ketāb al-ḥayawān, ed. ʿA-S. M. Hārun, 7 vols., Cairo, 1938-58.

Haftād o seh mellat, ed. M. J. Maškur, Tehran, 1962. 

Malaṭi, Ketāb al-tanbih wa’l-radd ʿalā ahl al-ahwāʾ wa’l-bidaʿ , ed. S. Dedering, Istanbul, 1936.

Maqdesi, Ketāb al-badʾ wa’l-taʾrīḵ, ed. and tr. Clément Huart as Le livre de la création et de l’histoire, 6 vols., Paris, 1899-1919.

Masʿudi, Moruj al-ḏahab wa maʿāden al-jawhar, rev. ed. Charles Pellat, 7 vols., Beirut, 1962-79.

Idem, Ketāb al-tanbih wa’l-ešrāf, ed. M. J. de Goeje, Leiden, 1894.

Meskawayh, Tajāreb al-omam, in D. S. Margoliouth and H. F. Amedroz, eds. and trs., The Eclipse of the Abbasid Caliphate: Original Chronicles of the Fourth Islamic Century, 7 vols., Oxford, 1921-22.

Michael the Syrian, Chronicle, tr. and ed., J.-B. Chabot, as Chronique de Michel le Syrien, patriarche jacobite d’Antioche 1166-119, 4 vols., Paris, 1899-1910; new ed., Brussels, 1964.

Nasafi, Baḥr al-kalām, Damascus, 1997.

Ps.-Nāšeʾ, “K. oṣul al-neḥal,” in J. van Ess, ed., Frühe Muʿtazilitische Häresiographie, Beirut, 1971, pp. 31-42 (Arabic; for the authorship, see W. Madelung, “Frühe muʿtazilitische Häresiographie: das Kitāb al-Uṣul des aʿfar b. Ḥarb?”  Der Islam 57, 1980, pp. 220-36).

Nawbaḵti, Feraq al-šiʿa, ed. H. Ritter, Istanbul, 1931.   

Neẓām-al-Molk, Siāsat-nāma, ed. H. Darke, 2nd ed., Tehran, 1985; tr. H. Darke as The Book of Government  or Rules for Kings, London, 1960.

Porphyry, On Abstinence from Killing Animals, tr. G. Clark. Ithaca, 2000.

Saʿd b. ʿAbd-Allāh Qomi, Ketāb al-maqālāt wa’l-feraq, ed. M. J. Maškur, Tehran, 1963.

Šahrastānī, Ketāb al-melal wa’l-neḥal, ed. W. Cureton, London, 1846; tr. D. Gimaret and G. Monnot, as Livre des religions et des sectes, Paris, 1986.

N. Sims-Williams, ed. and tr., Bactrian Documents from northern Afghanistan I: Legal and Economic Documents, Oxford, 2000.

Ṯaʿālebi, Ādāb al-moluk, ed. J. al-ʿAṭiyya, Beirut, 1990.

Ṭabari, Ketāb taʾriḵ al-rosol wa’l-moluk, ed. M. J. de Goeje et al., 15 vols., Leiden, 1879-1901; repr. Leiden, 1964.

Theophanes, Chronicle, ed. C. de Boor, Leipzig, 1883-85; English tr. C. Mango and R. Scott, with the assistance of G. Greatrex, as The Chronicle of Theophanes Confessor: Byzantine and Near Eastern History AD 284-813, Oxford, 1997. 


D. E Aune, Prophecy in Early Christianity and the Ancient Mediterranean World, Grand Rapids, Mich., 1983.

K. Babayan, Mystics, Monarchs, and Messiahs, Cambridge, Mass. and London, 2002.

E. G. Browne, “Some Notes on the Literature and Doctrines of the Ḥurufi Sect,” Journal of the Royal Asiatic Society, 1898, pp. 61-94.

P. Bryder, “Buddhist Elements in Manichaeism,” in U. Bianchi, ed., The Notion of “Religion” in Comparative Research, Rome, 1994, pp. 487-90.

L. Cirillo, “From the Elchasaite Christology to the Manichaean Apostle of Light,” in A. van Tongerloo and L. Cirillo, eds., Il Manicheismo: nuove prospettive della ricerca, Turnhout, 2005, pp. 47-54.

P. Crone, “Kavād’s Heresy and Mazdak’s Revolt,” Iran 29, 1991, pp. 21-42.

Eadem, “Zoroastrian Communism,” Comparative Studies in Society and History 36/3, 1994, pp. 447-62.

Eadem, “Abu Tammām on the Mubayyiża,” forthcoming in the festschrift for F. Daftary.

E. Daniel, The Social and Political History of Khurasan under Abbasid Rule 747-820, Minneapolis and Chicago, 1979.

M. Deeg and I. Gardner, “Indian Influence on Mani Reconsidered: the Case of Jainism,” International Journal of Jaina Studies 5, 2009, pp. 1-30.

J. van Ess, Theologie und Gesellschaft im 2. und 3. Jahrhundert Hidschra, Berlin and New York, 1991-97. R. Freitag, Seelenwandering in der islamischen Häresie, Berlin, 1985.

R. C. C. Fynes, “Plant Souls in Jainism and Manichaeism: the Case for Cultural Transmission,” East and West 46, 1996, pp. 21-44.

I. Gardner, “Some Comments on Mani and Indian Religions according to the Coptic Kephalaia,” in A. van Tongerloo and L. Cirillo, eds., Il Manicheismo: Nuove prospettive della richerca, Turnhout, 2005, pp. 123-35.

I. Gardner and S. N. Lieu, Manichaean Texts from the Roman Empire, Cambridge 2004.

C. A. Gieschen, Angelomorphic Christianity, Leiden, 1998.

H. Halm, “Das ‘Buch der Schatten.’ II,” Der Islam 58, 1981, pp. 15-86.

A. Heinrichs, “‘Thou Shalt not Kill a Tree’: Greek, Manichaean and Indian Tales,” Bulletin of the American Society of Papyrologists 16, 1979, pp. 85-108.

G. Hoffmann, Auszüge aus syrischen Akten persischer Märtyrer, Leipzig, 1880.

M. Th. Houtsma, “Bihʾafrid,” Vienna Oriental Journal 3, 1899, pp. 31-37.

S. N. C. Lieu, “An Early Byzantine Formula for the Renunciation of Manichaeism,” in idem, Manichaeism in Mesopotamia and the Roman East, Leiden, 1994, pp. 203-51.

G. P. Luttikhuizen, The Revelation of Elchasai, Tübingen, 1985. 

W. Madelung, “Khurramiyya,” EI ² V, Leiden, 1986.

Idem, Religious Trends in Early Islamic Iran, New York, 1988.

R. Merkelbach, “Die Täufer, bei denen Mani aufwuchs,” in P. Bryder, ed., Manichaean Studies, Lund, 1988, pp. 273-82.

G. Monnot, Penseurs musulmans et religions iraniennes, Paris, 1974. 

Peter, Prince of Greece and Denmark, A Study of Polyandry, the Hague, 1963.

H. Pirouzdjou, Mithraisme et Émancipation, Paris, 1999.

G. H. Sadighi, Les mouvements religieux iraniens au IIe et au IIIe siècle, Paris, 1938; Persian tr. Ḡ. Ṣadiqi, as Jonbešhā-ye dini-e irāni dar qarnhā-ye dovvom va sevvom-e hejri, Tehran, 1996 (see esp. chapter 6).

M. Šafiʿi-Kadkani, Qalandariya dar taʾriḵ, Tehran, 2007.

V. Škoda, “Ein Śiva-Heiligentum in Pendžikent,” Archaeologische Mitteilungen aus Iran 25, 1992, pp. 319-27.

J. Wellhausen, Die religiös-politischen Oppositionsparteien im alten Islam, Berlin, 1901.

E. Yarshater, “Mazdakism,” in idem, ed., The Cambridge History of Iran, III/2, Cambridge, 1983, pp. 991-1024.


(Patricia Crone)

Last Updated: June 29, 2011