Version classiqueVersion mobile

Démons iraniens

 | 
Philippe Swennen

Going, going, gone! The fate of the demons in late Sasanian and early Islamic Zoroastrianism

Albert de Jong

Résumé

Les démons dénommés ne jouent pas de rôle significatif dans les formes modernes et contemporaines du zoroastrianisme. Cet article tente d’expliquer leur disparition de la documentation en posant la question de savoir d’où ils vinrent d’abord. Ceci débouche sur un développement double, car il y a deux types de démons : ceux qui sont nommés dans l’Avesta, et qui survivent dans les textes classiques sous leur propre nom, et les démons qui représentent les (mauvais) affects humains. L’importance des démons en tant que classe d’êtres, plutôt qu’en tant que groupe de pouvoirs individuellement nommés, est mise en exergue d’un point de vue comparatif.

Texte intégral

  • 1 Dk. 7.4.42–46 (with a quotation from the [otherwise lost] Zand of Yt. 19.78–81). See Molé 1967, ad (...)
  • 2 SDB 3 (ed. Dhabhar 1909; trl. Dhabhar 1932, 506).
  • 3 Y. 9.15; Yt. 19.81; see Skjaervø 1996, 600–601, for the topos.
  • 4 An interesting parallel, in this respect, is found in the equally traditional stories that tell tha (...)
  • 5 It has often been assumed (especially since Nöldeke 1920, 10, n. 9) that Akvān is a garbled form of (...)
  • 6 See the classical study of Nöldeke 1915.
  • 7 Eilers 1979.
  • 8 Benveniste 1960.
  • 9 SDB 3: “But when Spitama Zarathushtra brought the Religion to the earth and revealed it, he immedia (...)
  • 10 PhlY 9.15: tō andar zamīg nigān kard hēnd harwisp dēw Zardušt kē pēš az ān wīr-ārōyišn padīd hēnd a (...)

1When Zarathushtra brought the revelation to the world, in order to instruct mankind and open a path to salvation, he broke the shape of the demons. This is what Pahlavi1 and Persian2 Zoroastrian texts tell us. The Avesta, it is well-known, tells us that he made the demons disappear from the surface of the earth, forcing them to go underground.3 Sources are united in telling us that before Zarathushtra spoke the Ahuna Vairya prayer, the demons roamed the earth, taking on human shapes whenever they wanted, something that became impossible for them after the words of the Ahuna Vairya, the most potent words of the religion, had first sounded on the earth. This particular aspect of Zoroastrian belief is also nicely illustrated, it seems, by the Shahnameh, which has very many stories on demons taking on human shapes and interacting with humans, but only in the section of the Shahnameh that comes before the chapter on Gushtasp and Zoroaster’s bringing of the Revelation to mankind. After that, the demons as humans seem to disappear from the Persian national epic.4 These demons in the Shahnameh are either unnamed, carry names that are unknown from Zoroastrian traditions — Akvan5 — or are known only by titles: the White Div is the best known example.6 Similarly, the most important demons of Persian folk tradition, those that were still feared in recent memory, are unknown from Zoroastrian sources. These are Al, immortalized as ein persisches Kindbettgespenst by Wilhelm Eilers,7 and the probably related Albasti, with variants, both famously studied by Benveniste more than fifty years ago.8 We shall meet some more unknown demonic beings in a moment, but first I must finish the discussion of the physical appearance of the demons. For disconcertingly, the Sad dar-i bundahiš, a medieval compendium explaining priestly traditions to a lay audience, while repeating the statement that the demons can no longer appear to us in human shape, adds that they can do so — if they wish — in the shape of an ass or of a cow, both animals deeply loved by Zoroastrians and decidedly belonging to the realm of good.9 The animals themselves, however, appear to be a free interpretation of the ultimate source of this statement, a Middle Persian gloss on Y 9.15, the very passage which states that Zoroaster caused the demons to hide under the earth.10 The gloss is difficult, but it appears to say, indeed, that demons can still take on the shape of large cattle, stōr, also belonging as a class to the good creation. The passage has rarely drawn attention, perhaps because the Sad dar is a late and much maligned text and the gloss is, as I already mentioned, quite difficult. But it serves our purpose here today, for it disturbs the picture and this disturbed picture is what I would like to discuss with you.

2The basic question is this: where did the demons, perceived as individual beings, go in Zoroastrian traditions? We shall see that, in order to answer that question, we also need to ask where they came from in the first place. Both questions are fairly difficult to answer, but perhaps I should begin by showing that they did, indeed, disappear from Zoroastrian tradition. This is generally known for modern times, but the disappearance of the individual demons goes back to a much earlier stage of the history of Zoroastrianism. We shall, however, begin at the end.

  • 11 Kreyenbroek 2001, 299: “Parsis whose views on religion have not been shaped by years of study tend (...)
  • 12 See de Jong 2004, 62–64.

3Among living Zoroastrians, belief in Ahriman himself as an independent spiritual being active in this world and of a totally evil nature has almost completely vanished.11 The fate of his demonic hosts, the Iranian demons, is even worse: no one knows their names and they play no role whatever in Zoroastrian observances or beliefs. In some Parsi homes, one can find protective spells written by a priest fixed to the wall near the main door of the house, but these do not mention specific demons, but speak of “evil” either in general terms or in terms of classes of evil beings, something to which we shall return. Zoroastrian rituals in general do not deal with the forces of evil, apart from an occasional curse formula. Generations of scholars have attempted to find evidence for ritual translations of Zoroastrian “dualism”, but they did not really find anything significant, with the exception of the very extensive code of purity laws and these have therefore somehow become the mainstay of “dualist” interpretations of Zoroastrian rituals.12 It is interesting to observe differences, for example, in the way the minor ablution rituals before saying prayers in Zoroastrianism and in Islam are usually interpreted. Whereas the Islamic ablutions — certainly wuḍū but also the fuller ghusl — are generally interpreted as a preparatory rite to establish the right conditions for the performance of the prayers, the Zoroastrian washings are frequently interpreted in terms of exorcism or apotropaic rites. It is clear that there are spoken elements that support such an interpretation, but it is equally clear that for modern Zoroastrians, such an interpretation is very wide off the mark; for them, it seems, it is just a way of being well prepared for an important religious ritual.

  • 13 Pirart 2007a; Jackson 1928, 67–109; Gray 1929.

4In earlier texts, it is clear that the notion of an active fight against evil, and against Ahreman and his demons who represent the forces of evil, is more on the surface of Zoroastrian beliefs and the interpretation of their rituals. These are, of course, strictly priestly documents, but even in them there are two distinct patterns of “thinking about” the demons. There are a few more or less technical treatments of the demons, in which the personality of a series of named demons — most often in strictly functional terms — is discussed. These have recently been collected and discussed by Éric Pirart in his book Georges Dumézil face aux démons iraniens, which is a real step forward compared to the earlier rather perfunctory treatment of the Iranian demons in the classical works of Jackson and Gray.13 Evidently, attempts were made by some priests to isolate and discuss the largest possible number of individual demons and this was usually done by pairing them with individual deities. This way, we end up with the series of Akoman versus Wahman, Endar versus Ardibehesht, Sawul versus Shahrewar etc.etc. I think we can refer to this system as a scholastic one and I shall come back to that label shortly.

  • 14 SDN 81: “Take care not to postpone today’s work until tomorrow. For the evil Ahreman, the Foul Spir (...)
  • 15 The largest collection of such texts is the sixth book of the Dēnkard, for which see Shaked 1979.
  • 16 See especially Boyce 1992a, 53–54 and see the index, s.v. mainyu.
  • 17 Boyce – de Jong forthc., ch. 9.

5The second pattern of speaking of the demons, and interpreting their behaviour and activities, is different and frequently uses demons who are not, or at least not directly, taken from the Avesta, whose names are clearly understandable and who function chiefly as personifications of undesirable emotions, attitudes and activities among men. They are seen, therefore, most often as indwelling in the human person. Unlike the exotic names of Sawul and Zaric etc., their names are usually ordinary nouns or, sometimes, nouns taken from the religious vocabulary — hence, ultimately going back to the texts of the Avesta. This list is quite long, but I do not think that a genuine list of them can be found anywhere in Zoroastrian literature. They include Āz, “greed”, Waran, “Lust”, Drō, “deceit”, Mihōxt, “Falsehood”, Kēn, “hate”, Arešk, “Envy”, Spazg, ‘Slander”, etc. In principle, this list is almost endless and in later texts we meet such interesting demons as Dīr and Pas, “Later” and “Afterwards”, evil beings who, like the better known Bušyąstā or “Sloth”, continually try to convince humans that they can postpone doing good works.14 This system, I think, can be labeled the affective one, in the sense that it consists chiefly of affects. It is massively present in those Pahlavi texts that belong to Zoroastrian wisdom literature, in which the theme of welcoming spirits representing good affects into one’s body is omnipresent, as is the theme of banishing from the body those demons who represent evil affects.15 This has a long history in Zoroastrianism,16 in the sense that it presupposes the notion of spiritual beings representing qualities, emotions and actions. Let me quote for this from the as yet unpublished remarks Mary Boyce wrote, perceptively, on the way this is evident from the romance of Vīs o Rāmīn by Gurgani:17

Another aspect of Zoroastrianism which was partly smoothed over by Gurgani, and which has been largely lost in modern translations of his work, is the instinctive dualism which shapes his characters’ thoughts. They look on the world, and above all on their own conduct in it, as constantly affected by the struggle between the active forces of good and evil. This is exemplified in Ramin’s words […] that when Love (that is, passion) enters, Wisdom leaves the heart. By itself this can easily be understood as a simple psychological statement, or as a poetic personification of abstractions; but as the story unfolds it becomes plain that it is neither, but exemplifies the traditional Zoroastrian apprehension that all things, whether animate, inanimate or “abstract”, possess an inner cognitive force or spirit, their mainyu, Middle Persian menog. In the ethical sphere different spirits, good and bad, were thought to seek entry into the heart, where if one became dominant the way for the other was barred; and it was the duty of everyone striving to be righteous to let only good spirits in. In course of time the term mainyu/ menog became restricted to beneficent spirits, and among the forces of wickedness only the “Evil Spirit”, Ahriman himself, was still called a mainyu, this having become a fossilized part of his proper name […]. Members of his cohorts were instead termed dev […], which had come to mean simply “devil”; and as such they appear abundantly in Vis u Ramin, together with their master, Ahriman, the Dev. Prominent among good spirits is, characteristically for Zoroastrianism, Wisdom (Khirad), whose presence should give moral guidance and so keep all bad ones at bay; but (as Ramin’s words show) the bad can be powerful against even it. This concept of a world of spirits, the forces within things, presents a problem for translating the Parthian romance, as it does for translating Avestan and Pahlavi texts; and this can be met, in English renderings, by using an initial capital letter for the menog or dev, to distinguish the cognitive force from the thing, for example Wisdom or Wrath from wisdom, wrath; but a clear-cut distinction is not always possible, probably in thought and so in word; and once this way of apprehending the world is understood and entered into, drawing such distinctions becomes of little importance.

6It is obvious that such a system is easily transformed into one in which these negative or positive affects are seen as just that and lose the “personal” quality attaching to them. This is evidently what happened, for with the exception of the Sad dar texts, there is little evidence in later Zoroastrian writings that the world was still perceived this way and that these negative emotions were still perceived, if they have ever been, as demons with a personality.

  • 18 Most perceptively, for example, in Boyce 1992b, 697–698.
  • 19 See de Jong 2009.
  • 20 On the contrary: the evidence from classical sources (de Jong 1997) shows that they did, at the ver (...)
  • 21 The main example would be the myth of the “holy immortals”, heroes of the religion who did not die, (...)

7Why, then, did the other demons, those named in the Avesta and mentioned in some Pahlavi texts, disappear? The answer to that question lies, I believe, in an answer to the question how they entered Middle Persian literature to begin with. This, I submit, is part of a “scholastic” transformation of Zoroastrianism in late Sasanian times. This transformation itself has often been noticed18 and it is intimately tied up with the momentous process of the writing down of the Avesta, which transformed the Avesta to a collection of texts that could be used — and were, indeed, used — as a source to be read and studied.19 I am not, of course, arguing that Zoroastrians did not have any theology before they wrote down their ritual texts,20 but I have been gathering evidence to show that the fact of a written Avesta — or, more broadly, of the notion of the Avesta as a text to be studied — has produced a number of myths, stories, beliefs and rituals that were not there before.21

  • 22 These points are elaborated in de Jong forthc., troisième leçon.
  • 23 See Cantera 2004, 320–328.
  • 24 Kreyenbroek 1991.
  • 25 As argued in de Jong 2004.
  • 26 Especially Shaked 1967.

8At the same time, this new status of the Avesta has enforced a number of traditional aspects of Zoroastrianism as it had developed to be abandoned: the worship of deities whose names could not be found in the Avesta — Nane, Sasan and others — the exalted position of Zurvan in some formulations of Zoroastrian theology, and the use of temples dedicated to named deities.22 It is clear even from the way their names were written in Pahlavi that the demons with the exotic names were lifted from the Avesta and it is also clear that the process by which they were paired with Zoroastrian deities was artificial and wholly dependent on the existence of lists of divine beings.23 The most prominent of these lists was obviously the calendar. So, deities from the calendar, especially the seven Amahraspands, were given demonic pairs, just as they were given co-workers, hamkār,24 the whole model being based on the systematic pairing of Ahura Mazda and Angra Mainyu. It is to be expected that this way of pairing deities with demons led to a personality of these demons being described, most often again in mirror-images of the personality of their divine counterparts: what God A is, Demon –A is not. This, incidentally, I believe to be the origin of the statement that Ahreman does not exist: since Ahura Mazda exists and Ahreman is his exact opposite, you could say that he does not exist.25 I know other scholars have suggested much more intricate explanations,26 but for the moment I shall stick to this simple one as being more satisfactory.

  • 27 See Shaked 2009, with references.
  • 28 Furlani 1954.

9The point of all this is, of course, that such a system was easily confined to learned circles only. They were the ones who produced it and it meant most to them. There would be one other category for whom knowledge of the names of the demons would be useful, and these are writers of amulets and similar protective spells. In living tradition, and this goes back some time, these have been priests and the amulets one finds in Pahlavi and Pazand examples are frequently brief formulae from the Avesta, which needed to be spoken while they were written to exert their power. Other amulets, in Middle Persian, protect their bearers from various classes of evil beings, also taken from the tradition ( “kek and karb”, “jadug and parig”, etc.).27 This is strongly reminiscent of the Mandaean tradition, if we follow the important research done on Mandaean demonology by Giuseppe Furlani,28 most of which is confirmed by the recent enormous amount of incantation bowls that are currently being studied. These bowls, especially the Jewish ones, very often mention demons by name, in order to send them away, but the Mandaean ones more often send every evil being away by listing it:

  • 29 BM 81761 = text 076M in Segal 2000.

Oh, you gods; temple spirits and idol spirits and goddesses, sorcery spirits, demons and devils and amulet spirits and liliths and angels and evil designs, take wing, flee and go forth from my house […].29

10I think for most ordinary Zoroastrian users, too, such a binding spell en bloc would be satisfactory. This makes the need to know the names of the demons even smaller. Zoroastrians were taught to interact with evil as little as possible and it seems they learnt their lessons well; so well, in fact, that the demons as individuals came to be forgotten. For most Zoroastrians, therefore, it seems, the named demons of the learned tradition were mere figures of speech.

Notes

1 Dk. 7.4.42–46 (with a quotation from the [otherwise lost] Zand of Yt. 19.78–81). See Molé 1967, ad locum.

2 SDB 3 (ed. Dhabhar 1909; trl. Dhabhar 1932, 506).

3 Y. 9.15; Yt. 19.81; see Skjaervø 1996, 600–601, for the topos.

4 An interesting parallel, in this respect, is found in the equally traditional stories that tell that Ahura Mazda used to speak directly to humans before Zarathushtra brought the revelation to the earth, but not after this turning-point in history had taken place. Since then, it seems, the text of the Avesta was thought to contain everything worth knowing; in exceptional cases, moreover, humans could be allowed to ascend to heaven while alive, there to speak with Ahura Mazda directly. See for this theme de Jong 2005a; de Jong 2008, 102–104.

5 It has often been assumed (especially since Nöldeke 1920, 10, n. 9) that Akvān is a garbled form of the name of the famous demon Akōman. I see little merit in this suggestion. The phonetic similarity between the two names is rather limited (restricted to the first syllable /ak/ which by itself means “evil”) and, in Middle Persian, the two would be graphically very distinct. Nöldeke had to assume, therefore, that the confusion was created by the Arabic source underlying Firdausi’s text, which strikes me as contrived. One could, of course, reconstruct a Middle Persian word+ak-bān, ‘evil-keeper’, but such a word is not attested. Scholars who consider the identification of Akvān with Akōman likely make much of the “personality” of the two, but as I shall try to show below, Akōman has no personality. See, however, Khaleghi-Motlagh 1985.

6 See the classical study of Nöldeke 1915.

7 Eilers 1979.

8 Benveniste 1960.

9 SDB 3: “But when Spitama Zarathushtra brought the Religion to the earth and revealed it, he immediately broke the shape of the demons: they disappeared under the earth. And now, when they want to bring about sins, they cannot act as men and impersonate them, but they can only appear in the likeness of an ass, a cow and similar (animals).”

10 PhlY 9.15: tō andar zamīg nigān kard hēnd harwisp dēw Zardušt kē pēš az ān wīr-ārōyišn padīd hēnd abar pad ēn zamīg [pad dēw-kirbīh. Hād, harw ān kē tan mēnōg tuwān būd kardan ā-š kālbod be škast; ān kē nē tuwān būd kardan xwad be škast. Kālbod be škastan ēd kū az ān frāz pad dēw-kirbīh wināh nē tuwān būd kardan tā pad stōr-kirbīh ud mardōm-kirbīh nūn-iz ōh kunēnd. “You buried all the demons in the earth, Zarathushtra, who roamed the earth before this time in human shape [in the body of a demon. Now, every (demon) who could take on a spiritual body, he broke his shape; the one who could not do so, broke his own. ‘Breaking the shape’ means that henceforth they could no longer do evil in the body of a demon, but they can still do so in the bodies of large cattle and of men]”. Both grammatically and with regard to content, the passage is obscure.

11 Kreyenbroek 2001, 299: “Parsis whose views on religion have not been shaped by years of study tend to ignore Ahriman altogether”. Michael Stausberg (2009, 244–249) has, however, drawn attention to the fact that the belief in (heaven and) hell has been more resilient, especially among Parsi priests.

12 See de Jong 2004, 62–64.

13 Pirart 2007a; Jackson 1928, 67–109; Gray 1929.

14 SDN 81: “Take care not to postpone today’s work until tomorrow. For the evil Ahreman, the Foul Spirit, has appointed two evil spirits (druj) for this work; one is called ‘Later’, the other ‘Afterwards’. These two evil spirits constantly attempt to convince people to postpone their work. For every good deed and every virtue that comes up, the evil spirit called ‘Later’ says: ‘You are going to live long; you can do this any time’, and the evil spirit called ‘Afterwards’ says: ‘Leave it now, for you can do it afterwards’. These two evil spirits constantly hold that soul back from performing its own duties until the day comes that the time for everything has passed. Then, regret and repentance will be of no use.”

15 The largest collection of such texts is the sixth book of the Dēnkard, for which see Shaked 1979.

16 See especially Boyce 1992a, 53–54 and see the index, s.v. mainyu.

17 Boyce – de Jong forthc., ch. 9.

18 Most perceptively, for example, in Boyce 1992b, 697–698.

19 See de Jong 2009.

20 On the contrary: the evidence from classical sources (de Jong 1997) shows that they did, at the very latest in Achaemenid times. I have tried to argue that this was part of an Achaemenid project to produce a (more) systematic and unified interpretation of the religion, suitable for all Iranian inhabitants of the empire. See de Jong 2005b; 2010a; 2010b. This momentous development is distinct from the development made here, which focuses on a further transformation of the religion after the codification of the Avesta with its Zand.

21 The main example would be the myth of the “holy immortals”, heroes of the religion who did not die, but are thought to be present in this world and will reemerge at the end of time. Their names are taken from various parts of the Avesta, with a particular cluster owing its existence, it seems, to the rich inventory of names in Yt. 13 and Yt. 5 (Ašavazdah son of Pourudāxšti; Fraδāxšti the son of the jar, the Tree-opposing-harm, etc.).

22 These points are elaborated in de Jong forthc., troisième leçon.

23 See Cantera 2004, 320–328.

24 Kreyenbroek 1991.

25 As argued in de Jong 2004.

26 Especially Shaked 1967.

27 See Shaked 2009, with references.

28 Furlani 1954.

29 BM 81761 = text 076M in Segal 2000.

Auteur

Le texte et les autres éléments (illustrations, fichiers annexes importés) sont sous Licence OpenEdition Books, sauf mention contraire.

Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search