ZURVANISM, a hypothetical religious movement in the history of Zoroastrianism. 

The first notices of a movement called zurvānīyya among the Iranians derive from the rich tradition of Muslim writings on religions and religious movements in the early Islamic world. These sources reflect, first and foremost, Muslim concerns (Monnot, 1986, pp. 97-125). These Muslim writers had a keen interest in elaborating their own concept of the oneness of God (Ar. towḥīd) by contrasting it with the views of their competitors. Where they found what to them must have looked like substantive differences in the phrasing of God’s nature (both in the developing Muslim theologies and among the non-Muslim religions of their time), they intuitively translated these differences into differences of schools or sects. In this, they were the inheritors of a similar style of classifying (and solving) philosophical and religious difference that originated in the ancient world, both among historians of philosophy (known as doxographers) and among early Christian authors who intended to defend their version of the religion against the opinions of others (so-called heresiographers). 

Although much valuable information has been preserved in all these sources, this information is almost irredeemably corrupted both on an empirical and on a structural level. Empirically, the evidence is often contaminated by the intrusion of imaginary persons, schools, and opinions (or by the suggestion that groups that had long been extinct were still to be found). More importantly, on a structural level, differences of opinion are automatically translated into social units (“sects,” “movements,” “schools”) that may not correspond to any recognizable socio-historical reality. The mechanisms of this sanitized view of the world are well-known: it contrasts the unity of the “orthodox” with the divisions among those seen as “heretics” (who fight against the majority opinion as well as among themselves), and this acts as a sign that the opinions of these heretics must be wrong.

Western scholars who first began to write about Zoroastrianism were likewise indebted to this tradition in the history of ideas and of religions. It had acquired for them, moreover, the same type of urgency that it had had for the early Muslim writers, since the Reformation had engendered a very similar (and ongoing) fragmentation of the Christian churches that manifested itself in the realms of ideas (theology), of ritual, and of modes of organization. Since their sources, apart from the writings of Classical authors (who did not distinguish between different types of religion among the Persians at all), were the same early Muslim writers on Zoroastrianism, it is not surprising that the notion of a Zurvanite movement entered Western views of Zoroastrianism at an early stage. The earliest authors, however, resemble the Classical attitude more, in that they simply described or listed the story of Zurvan as progenitor of the two spirits, without relegating it to a separate movement. Thus, the twenty-second chapter of Thomas Hyde’s (q.v.) influential Veterum Persarum et Parthorum et Medorum Religionis Historia (Oxford 1760), which deals with the Zoroastrian ideas about the two spirits and their origin, drew its information from Damascius, Plutarch, and Šahrastānī (Hyde 1760, pp. 292-303) and faithfully discusses Zurvan and his role in the origin of the two spirits. Likewise, Anquetil-Duperron mentions the fact that Zoroaster taught the origin of the two spirits as creatures of a first principle, “le tem[p]s sans bornes” (“infinite time”; Anquetil-Duperron, 1771, I/2, p. 68).

It was Friedrich Spiegel who revived the notion of a Zurvanite movement (or, as he called it, “sect”), first in his Érân. Das Land zwischen dem Indus und Tigris (Spiegel 1863, pp. 64, 366), and then, more extensively (on the basis of Šahrastānī again) in his interpretation of Zoroastrian sects in his monumental Erânische Alterthumskunde (Spiegel, 1871-78, II, pp. 175-87). It is characteristic of Spiegel’s often underestimated importance for the history of Iranian studies that most of the “information” and interpretations used in later discussions of Zurvanism, or the Zurvanite myth, were already given by him, even though not all later scholars cared to refer to his contributions (see Rezania, 2010, pp. 12-24, for a much fuller discussion of the history of the study of Zurvanism than can be given here).

After Spiegel, Zurvanism became a standard subject in discussions of Zoroastrianism. Some scholars thought of it as a pre-Zoroastrian (Median) religion that was brought into Zoroastrianism with the conversion of the Medes to that religion (Nyberg, 1938, p. 388; Widengren, 1938, pp. 271-74) or as the religion of the Parthians (Widengren 1955, pp. 88-89). Others rightly stressed the fact that almost all sources are from the Sasanian period and saw a special role for the Sasanian kings as supporters of a Zurvanite version of Zoroastrianism (Christensen, 1944, pp. 150-55). Two discoveries considerably widened the discussion in the early twentieth century. The first was the discovery of Manichean Middle Persian texts from Central Asia, in which the Father of Greatness, the supreme God of the Manichean pantheon, was (sometimes) called Zurvān, suggesting a similar role for this god in third century Persian Zoroastrianism (see below). The second was the discovery of the inscriptions of Antiochus I of Commagene, which presented a Graeco-Iranian dynastic cult with gods who bore Greek and Iranian names, and with an important reference also to the concept of “boundless time” (Gk. chronos apeiros) as a reality in which human destiny unfolds (Boyce and Grenet 1991, pp. 332-34). These discoveries led to a mass of publications, many of them of an extremely speculative nature, that gave Zurvanism (and Iranian religion generally) a very important role in the development of Hellenistic religions. For several scholars saw in Zurvanism a solution to the problem of the rise of the concept of aiōn in Hellenistic thought. This led them to exaggerate that concept in Hellenistic religions themselves, but especially to misrepresent the nature of Zurvanism (Eisler, 1910; Reitzenstein, 1921, pp. 151-250; Junker, 1921-1922).

The problem was (and continues to be) the fact that, while the myth of Zurvan is fairly well known from Armenian, Syriac, Greek, and Arabic sources, it is not to be found in any Zoroastrian source. This led scholars to speculate that it must have existed in Zoroastrian sources themselves, but that all references to it had deliberately been purged from the surviving literature (Widengren, 1967). This, in turn, led them to detect Zurvanite ideas in passages that, although they had thus been purged, would still reflect a Zurvanite coloring. The procedure was completely circular: all speculations on time and space, all speculation on the workings of fate, and all passages in which women were seen as ambiguous beings were declared “Zurvanite,” because “Zurvanism” was simply assumed to have been characterized by an interest in time and space, by an accommodation of “fatalism,” and by a negative view of women (De Jong,  1995). This led to a very rich dossier of texts, none of which mentioned anything called Zurvanism or gave any hint of the possibly controversial nature of the information provided (Zaehner, 1955). With this development, Zurvanism became a convenient receptacle for everything that clashed with generally held notions about “real” Zoroastrianism, and since the dominant approach to the subject was that Zurvanite ideas originated in a confrontation of Iranian with Babylonian or Greek ideas, the Zurvanite hypothesis came to function as a shield to protect “proper” (Iranian) Zoroastrianism from a hybridized version of that religion that was contact-induced. This in turn led to the interpretation of Zurvanism as a “heresy” or even a “betrayal” of “true” Zoroastrianism, even though it clearly represented a dominant interpretation of Zoroastrian theology at the Sasanian court (Boyce, 1996).

With this, the study of Zurvanism and of the history of Zoroastrianism generally had reached an impasse, and protest became inevitable. It was Shaul Shaked who contributed most to the demolition of the edifice that had thus been erected. In a series of important articles and a monograph, he not only showed the feeble source-base in which most speculation had been grounded, but also the pernicious side-effects of the whole construction of a “Zurvanite” sect/heresy/movement/church (Shaked, 1992; 1994a; 1994b). He showed that the myth of Zurvan, securely attested (see below), was one among many variants of the Zoroastrian cosmogony and that there are no signs to indicate that it was seen as somehow offensive. This left the importance of that particular myth for an understanding of the history of Sasanian Zoroastrianism somehow unexplained, so that there is still room for discussion here. It is necessary, therefore, to present the main evidence and to attempt to give it some context.

Zurvan as deity. Zurvān “Time” is known as a minor god in Avestan texts (see  ZURVĀN). He is not mentioned very often, but he is clearly present as one among many abstract deities, in this case representing the notion of “time” in the sense of a “period of time.” The Yasna ceremony ends with an invocation of this deity, in the company of the related gods Rāman (“peace”), Thwasha (“firmament”), and “Vayu” (representing the “void” between the worlds of light and darkness), suggesting speculation on the structure of the cosmos in terms of space and time as the precondition for the genesis of our present world of mixture (Rezania, 2010). It is this theme that is consistently taken up in (much) later sources, which give systematic discussions of the (crucial) narrative of the origin of creation in the battle between the two spirits. 

“Time” and “place” were bound to come up in the development of this narrative, which took its current form in the Achaemenid period (De Jong, 2005). It chiefly represented the two spirits as pre-existent, in a context of (equally pre-existent) “time” and “space,” which were further delimited (“cut”) as part of the pact sealed between them (this is the most likely background of the much-discussed passage from Eudemus of Rhodes (4th cent. BCE) in the work of the Neoplatonic philosopher Damascius (De Jong, 1997, pp. 336-37). These speculations are standard elements of all retellings of the story of creation, and there is nothing to suggest a special role for a deity Zurvan in it. 

This is also true for the development of the Zoroastrian calendar; earlier scholars had found a “Zurvanite” element in the day-dedications, in the fact that four days of the month were allotted to Ahura Mazda, which was interpreted in the light of (very late) Manichean sources that suggested that the Father of Greatness was a fourfold deity (Nyberg, 1931). Since this Father of Greatness was (sometimes) called Zurvan in Manichean Middle Persian texts, his fourfold nature, in combination with the dating of the development of the calendar to the Achaemenid period, was used to support the idea of a very early development of Zurvanite theology. This reconstruction has since been shown to be untenable (Buyaner, 2006), and the whole notion of early Zurvanism must therefore be abandoned.

The Manichean Zurvan. The only element that remains, apart from the (Sasanian) myth of Zurvan, is the Manichean evidence, which is important for two reasons: it shows that the notion of Zurvan as supreme deity is older than the earliest sources we have for the myth of Zurvan (see below), and it shows that this role for Zurvan was more typical of Pars and speakers of Middle Persian than it was of the other Iranian lands. The evidence is conclusive, in the sense that the interpretation of the Father of Greatness as Zurvan occurs already in the Šābuhragān, an early Manichean composition in Middle Persian (and the only text written by Mani himself of which large portions survive). This allows us to date its occurrence to the early Sasanian period. The identification of the Father of Greatness with Zurvan is not found in Manichean Parthian, which has usually been taken to indicate that this role of Zurvan did not correspond to a Parthian version of Zoroastrianism. The name Zurvan occurs, it is true, in Sogdian and other Central Asian languages, but since these texts derive from Sogdian Manicheans who used Middle Persian and Parthian as their church languages, this in itself does not contradict the impression that Zurvan was chiefly recognized by Persian Zoroastrians. Generally, moreover, the “interpretation” of Buddhist and Manichean deities with Zoroastrian ones is an open-ended affair with uncertainties on both sides: even though one can speculate about reasons why a certain Zoroastrian deity was chosen to represent a Manichean or Buddhist spiritual being, it is almost impossible to present the data from Buddhist or Manichean texts as evidence for developments in Zoroastrian theology.

The myth of Zurvan. We are left, therefore, with a very specific and amply attested fact: the myth of Zurvan, which is an alternative version of the Zoroastrian myth of creation known from Greek, Syriac, Armenian, and Arabic sources (texts given synoptically in Zaehner, 1955 and Rezania, 2010). In this myth, which is surprisingly uniform in the various sources, the two spirits (Ohrmazd [see AHURA MAZDĀ] and Ahreman [see AHRIMAN]) are presented as the twin offspring of a pre-existing god Zurvan. The difference among versions of the myth is largely restricted to the origin of the two spirits: the myth presents Zurvan as sacrificing for a period of a thousand years, in order to beget a son. Zurvan experienced a moment of doubt; as a result, Ohrmazd came into being because of the sacrifices and Ahreman out of Zurvan’s doubt. When Zurvan realized that two children had been formed in his womb, he promised to give dominion to his first-born, intending it to be Ohrmazd, but Ahreman pierced the womb and presented himself as the first-born. It is here that the stories vary, but in general it is from the birth of the two spirits onwards that the narration of the cosmogony follows the customary lines known from “standard” Zoroastrianism. The myth of Zurvan is thus some sort of a “prequel” to the ordinary story of creation, and there are very few (if any) indications that this prequel was considered as imposing by any contemporary Zoroastrian, as it has seemed to modern Western scholars.

It is important to stress here that the notions of “monotheism,” “dualism,” and “polytheism” belong to pre-modern Europe (with evident precursors in Jewish, Christian, and Muslim texts, which differ sharply among themselves) and do not correspond in any meaningful way to self-identifications of Zoroastrianism before the Islamic period. It has been possible, as a consequence, to present “standard” Zoroastrianism with each of these labels simultaneously, which is a sure indication that the labels do not fit. Much ink has been spilled on sorting out if the myth of Zurvan was a “reworking” (or even “betrayal”) of classical Zoroastrianism in the face of a growing “monotheism” in the Near East in Late Antiquity, but since there is nothing to indicate that this was even noticed by any Zoroastrian of the period, most of this debate has been pointless.

As to the origin of the myth, several suggestions have been made, most of which trace it to contact with non-Iranian religions, either Babylonian (Zaehner, 1955, pp. 19-20) or Greek (De Blois, 2000, p. 6). None of these suggestions is very convincing. There are two good suggestions that have been made from an internal Iranian perspective. The first of these was already hinted at above: ideas about “space” and “time” were a necessary part of the development of the Zoroastrian story of creation, and since “time” in particular came to play a dominant role in these speculations, it is no surprise that a pre-existent time(-god) could play a role in ideas about reality before the creation of the cosmos (Rezania, 2010). More importantly, a convincing case has been made for the possibility that scholar-priests interpreted the famous line from Y. 30.3 of the Avesta, according to which (in conventional interpretation) the two spirits are presented as “twins,” as indicating that, if they were twins, they needed a father (Boyce, 1982, p. 232). It is indeed the idea that Ohrmazd and Ahreman are brothers that has been the target of Manichean and Zoroastrian polemics, which suggests that the verse was indeed interpreted in this way. It is an example, moreover, of a type of reasoning (or theology) that comes into being with the development of a class of learned priests or professional theologians who begin to “study” the text of the revelation for its meaning. Since there are many indications that the early Sasanians used such a class of priests for the reformulation and streamlining of the religion (to suit an imperial agenda; De Jong, forthcoming), the early Sasanian period seems a likely time-frame for the origin of the myth. It would be confined, then, to circles of court priests, which is firmly supported by the surviving references to the myth.

The social background of the myth. Although the earliest attestation of the myth goes back to the late fourth century (Theodore of Mopsuestia as quoted in the Library of the ninth-century Byzantine patriarch Photius), the best indications for the social and political importance of the myth come from Armenia. The Sasanians were aware, it seems, of the fact the Armenians had been Zoroastrians before they converted to Christianity (Russell, 1987), and in the fifth century they attempted to re-impose Zoroastrianism on the Armenians. This led, eventually, to the notorious battle of Avarayr (q.v.; in 451 CE), in which the Armenians were defeated by the Sasanians. This defeat is seen by the Armenians, not without justification, as a victory, since it led to the informal recognition of the fact that the imposition of Zoroastrianism on Christian Armenia was no longer a viable option. It is thus the subject of important Armenian historical works, with pride of place going to the History of Vardan of Ełišē (q.v.) Vardapet (Thomson, 1982), probably written within one or two generations of the actual events. In Ełišē’s narration of the events, the myth of Zurvan is very important; he includes in his History a letter written by the wuzurg-framādār Mihr-Narseh (see MEHR-NARSEH) to the Armenians, urging them to give up their religion and accept the truth of Zoroastrianism, which he summarizes by retelling the myth of Zurvan. 

This myth is also the target of the polemical works of the Armenian theologian Eznik of Kołb (Blanchard and Young 1998), which can be dated to the fifth century. The Armenians, it is well known, did not recognize this religion as the religion from which their forefathers had been converted to Christianity, and they summarily rejected its relevance to them. Some of the reports in Syriac also go back to the fifth century, although most of them (and all of the Arabic ones) are much later; the early ones also reproduce encounters between lapsed Zoroastrians and representatives of “official,” imperial Sasanian Zoroastrianism.

This strengthens the case for the fact that the Zurvanite myth was not just one among many alternative versions of the cosmogony, but that it was—at least in the fifth century – the version of the court. This gives it a social importance that has occasionally been downplayed and makes it urgent to try and reconstruct how it, eventually, disappeared.

The end of Zurvanism. Since no Zoroastrian text refers to anything resembling “Zurvanism” as an organized movement (while showing no hesitation in mentioning Zurvan’s name or in speculating about “time”), this question is evidently difficult to answer. In sixth-century Iran, two important developments changed Zoroastrianism drastically. The first was the destruction of the Mazdakite movement (after a very brief period of royal approval), which led to a tightening grip of the priesthood on the instruction of the laity. The second, possibly even more momentous, development was the writing down of the Avesta (with its Zand), which led to a scriptural movement among the Zoroastrians (De Jong, 2009). Kings and priests had acted together several times in imperial Iranian history to streamline the Zoroastrian religion. These two developments were to transform Zoroastrianism into a version of that religion that has survived to the present. They coincide, for example, with a huge elaboration of the purity rules, possibly with the development of the Vendidad ceremony as a night office, and very likely with the disappearance of all funerary traditions except excarnation. These have in common the fact that they very much rely on Avestan precedent, which supports the idea that they originate in such a scriptural movement, in which the Avesta began to be thought of as a text that should be consulted, explained, and applied. 

If that is the case, this must have spelled the end of a number of developments in “lived” Zoroastrianism for which scriptural authority could not be found. The most obvious example of this are the non-Iranian deities who were very popular in (regions of) Iran, such as the goddess Nanaia or the god Sasan. The story of Zurvan as progenitor of the two spirits likewise cannot be located in any part of the extant Avesta, and it may have fallen victim to the same movement. To this must be added the fact that, if the myth of Zurvan was indeed a version of Zoroastrian theology that was characteristic of the Sasanian court (in the fifth century), the disappearance of that court—and its priests—must have accelerated the erosion of the myth of Zurvan.

As is well known, it did not disappear, for it was recorded in early Islamic times by a large number of Syriac and Arabic writers, not all of whom should be considered to have found it in earlier sources that were no longer current in their own times. This final piece of the puzzle, which should include the survival of “Zurvanite” ideas in the Persian treatise ʿOlamā-ye Eslām be dīgar raveš (Rezania 2010, p. 6; see OLAMĀ-YE ESLĀM) is difficult to crack, but the fact that the myth persisted in some Zoroastrian writings suitably underlines the conclusion reached earlier by Shaul Shaked, and by now broadly accepted, that up to the ninth century a wide array of stories of creation was still current among the Zoroastrians—and that this meant less than modern scholars have been able to accept.



A. H. Anquetil-Duperron, Zend-Avesta, ouvrage de Zoroastre, Paris, 1771.

M. J. Blanchard and R. D. Young, A Treatise on God written in Armenian by Eznik of Kołb, Leuven, 1998.

F. de Blois, “Dualism in Iranian and Christian Traditions,” JRAS, 2000, pp. 1-19.

M. Boyce, “Some Reflections on Zurvanism,” BSOAS 19, 1957, pp. 304-16.

Idem, A History of Zoroastrianism II. Under the Achaemenians, Leiden, 1982.

Idem, “Some Further Reflections on Zurvanism,” in  Iranica Varia. Papers in Honor of Professor Ehsan Yarshater (Acta Iranica 30), Leiden 1990, pp. 20-29

Idem, “On the Orthodoxy of Sasanian Zoroastrianism,” BSOAS 59, 1996, pp. 11-28

M. Boyce and F. Grenet, A History of Zoroastrianism III: Zoroastrianism under Macedonian and Roman Rule, Leiden, 1991.

D. Buyaner, “On the Structure of the Zoroastrian Month,” IIJ 48, 2006, pp. 199-206.

A. Christensen, L’Iran sous les sassanides, Copenhagen, 1944.

R. Eisler, Weltenmantel und Himmelszelt. Religionsgeschichtliche Untersuchungen zur Urgeschichte des antiken Weltbildes, München, 1910.

T. Hyde, Veterum Persarum et Parthorum et Medorum religionis historia, Oxford, 1760.

A. de Jong, “Jeh the Primal Whore? Observations on Zoroastrian Misogyny,” in  R. Kloppenborg and W. J. Hanegraaff, eds., Female Stereotypes in Religious Traditions (NUMEN Book Series 66), Leiden, 1995, pp. 15-41

Idem, Traditions of the Magi. Zoroastrianism in Greek and Latin Literature (Religions in the Graeco-Roman World 133), Leiden, 1997.

Idem, “The Contribution of the Magi,” in  V. S. Curtis and S. Stewart, eds., Birth of the Persian Empire (The Idea of Iran 1), London, 2005, pp. 85-97.

Idem, “The Use of Writing and the Idea of the Avesta in Sasanian Iran,” in  E. Pirart and X. Tremblay, eds., Zarathushtra entre l’Inde et l’Iran. Études indo-iraniennes et indo-européennes offertes à Jean Kellens à l’occasion de son 65e anniversaire (Beiträge zur Iranistik 30), Wiesbaden, 2009, pp. 27-41.

Idem, “Religion and Politics in pre-Islamic Iran,” in  M. Stausberg and Y. Vevaina, eds., The Blackwell Companion to Zoroastrianism, forthcoming.

H. F. Junker, “Über iranische Quellen der hellenistischen Aion-Vorstellung,” Vorträge der Bibliothek Warburg, 1921-1922, pp. 125-78.

G. Monnot, Islam et religions, Paris, 1986.

H. S. Nyberg, “Questions de cosmogonie et de cosmologie mazdéennes,” JA 219, 1931, pp. 1-134; 193-244.

Idem, Die Religionen des alten Iran, Leipzig, 1938.

R. Reitzenstein, Das iranische Erlösungsmysterium. Religionsgeschichtliche Untersuchungen, Bonn, 1921.

K. Rezania, Die zoroastrische Zeitvorstellung. Eine Untersuchung über Zeit- und Ewigkeitskonzepte und die Frage des Zurvanismus (Göttinger Orientforschungen, Iranica, N.F. 7), Wiesbaden, 2010.

J. R. Russell, Zoroastrianism in Armenia, Cambridge, Mass., 1987.

S. Shaked, “The Myth of Zurvan: Cosmogony and Eschatology,” in  I. Gruenwald, S. Shaked, and G. G. Stroumsa, eds., Messiah and Christos: Studies in the Jewish Origins of Christianity, presented to David Flusser (Texte und Studien zum antiken Judentum 32), Tübingen 1992, pp. 219-40.

Idem, “Some Islamic Reports concerning Zoroastrianism,” Jerusalem studies in Arabic and Islam 17, 1994 , pp. 43-84.

Idem, Dualism in Transformation. Varieties of Religion in Sasanian Iran, London, 1994.

F. Spiegel, Érân. Das Land zwischen dem Indus und Tigris. Beiträge zur Kenntniss des Landes und seiner Geschichte, Berlin, 1863.

Idem, Erânische Alterthumskunde, 3 vols., Leipzig, 1871-1878.

R. W. Thomson, Ełishē. History of Vardan and the Armenian War, Cambridge, Mass., 1982.

G. Widengren, Hochgottglaube im alten Iran: eine religionsphänomenologische Untersuchung, Uppsala and Leipzig, 1938.

Idem, “Stand und Aufgaben der iranischen Religionsgeschichte II: Geschichte der iranischen Religionen und ihre Nachwirkung,” Numen 2, 1955, pp. 47-134.

Idem, “Primordial Man and Prostitute: A Zervanite Myth in the Sassanid Avesta,” in  Studies in Mysticism and Religion presented to Gershom G. Scholem, Jerusalem, 1967, pp. 227-34.

R. C. Zaehner, Zurvan. A Zoroastrian Dilemma, Oxford 1955.

(Albert de Jong)

Originally Published: March 28, 2014

Last Updated: March 28, 2014