FRAVAŠI (OP *fravarti, Pahl. fraward, frawahr, frōhar, frawaš, frawaxš), the Avestan word for a powerful supernatural being whose concept at an early stage in Zoroastrianism became blended with that of the urvan. The urvan is the human soul, which in pre-Zoroastrian times was believed after death to pass a shadowy existence in the underworld, from where it returned once a year to its former home at the feast of H amaspaθmaēdaya (see FRAWARDAGĀN). It then received prayers and offerings from its descendants, whom it rewarded with blessings (see DEATH among Zoroastrians). Close parallels with the Indian pitaras establish that belief and cult go back to proto-Indo-Iranian times. The concept of the fravašis appears in contrast unique to the Iranians. The principal source for knowledge of them is Yašt 13 (see FRAWARDĪN YAŠT). In this they appear as beings who inhabit the upper air and are powerful to aid and protect those who worship them. Yet they too are associated with the cult of the dead and are sometimes directly identified as urvans. References to the fravašis occur in the text of the Yasna and in the Pahlavi books. Abū Rayḥān Bīrūnī describes their annual worship (Āṯār, p. 224), and there are abundant materials for their veneration in living Zoroastrianism.
The fravašis were conceived as a vast host of “many hundreds, many thousands, many tens of thousands” (Yt. 13.65). They are presented in their yašt as aiding Ahura Mazdā in the beginning to order the world, and as powerful now in maintaining it, in particular in nurturing waters and plants, and protecting sons in the womb (vv. 1-11 and passim). Every year when rain comes, they strive among themselves to obtain water for their own “family, settlement, tribe, and country” (vv. 65-68); but though many verses celebrate their prowess and power to help in battle, in two sections of the Yašt (vv. 49-52, 96-144) they are fully identified with the relatively helpless urvan, although the term fravaši is used throughout.
Much thought has been given by Western academics to trying to explain these perplexing anomalies and arrive at the original concept of the fravaši. There is a wealth of Indo-European and Indo-Iranian verbal roots from which to derive their name (Gray, Foundations, pp. 75-79); and in choosing from among these, scholars have divided into two groups: Those who consider the concept to be pre-Zoroastrian and assimilated into the faith, and those who think that it originated with Zoroastrianism and subsequently absorbed more archaic elements. Of the first group Nathan Söderblom defined the fravašis as essentially spirits of the dead. He derived their name (pp. 389-90) from √vart- “turn” with preverb fra- and abstract suffix -i, meaning “turning away, departing, death”; but this explanation offered little distinction between the original fravaši and the urvan, and could not readily be reconciled with the cosmic functions assigned to the fravašis in their yašt. James Moulton (1913a, pp. 270-72) derived “fravaši” from √var- “make/be pregnant,” with abstract suffix -ti. He considered the fravašis to be ancestor-spirits, whose essential concern was the conceiving of sons (Yt. 13.11, 15) so that their line might continue—a concern which he saw then extended to the fertility of the whole natural world, and hence to cosmic activities. He rendered their name accordingly as “birth-promotion; promoter of birth, birth-spirit.” Harold Bailey (Zoroastrian Problems, p. 109) saw the basic element in the concept to be the fravašis’ warlike activities, and suggested that the term was originally used in a hero-cult, “fravaši” meaning “protective valor (vərəti),” from √var- “cover, protect,” with fra- “ward off.” The fravaši would thus be the spirit of a valiant warrior, greater than the ordinary urvan, with a different fate after death, and invoked for help by his family and tribe. This interpretation gained wide support (although against an etymological link with NPers. gord “hero,” see Narten, 1985, p. 86). It was enlarged on by Georges Dumézil (against whose theory of the fravašis being the Iranian counterparts of the Indian Mārūts, see Malandra, 1971), while Gherardo Gnoli argued for a double origin in a hero- and an ancestor-cult. Bernfried Schlerath accepted the hero-cult as basic, but suggested deriving “fravaši” from √var- “choose,” with *fravarti being the public acclamation—a “choosing forth—”in this life of an outstanding warrior whose spirit was venerated thereafter.
The Iranian Heroic Age probably developed from about 1500 B.C.E.; so, if the fravašis were originally the objects of a hero-cult, this could explain why their concept cannot be traced back to proto-Indo-Iranian times. This fact is no problem for scholars of the second group, who see the concept as original to Zoroastrianism. At the beginning of this century (Moulton, 1913a, p. 268), the prevailing opinion was, like Schlerath’s, that the term fravaši came from √var- “choose,” but stress was laid on the fact that in Avestan fravar- means “choose/profess a faith.” Hence the fravaši was seen as the spirit of confessional faith. This interpretation was put forward again decades later by Karl Hoffmann (p. 91). Meantime the same derivation had been used by Herman Lommel (pp. 102-5) to yield the meaning “decisive choice.” He saw the spirit of decisive choice as a typical Gathic “abstraction,” and hence a part of Zoroaster’s own doctrinal system. Certain similarities between the daēnā (q.v.) and fravaši had been commented on by previous scholars, and Lommel now suggested that the two were identical, and that the prophet had simply preferred the former term in his Gāthās. This part of his analysis proved unacceptable, but otherwise his interpretation was adopted by several scholars.
Johanna Narten, however (1985, 1986), comparing (as Schlerath had done) Vedic pravara-, translated OAv. “fravaši” simply as “choice.” The focus of her study was the word’s unique occurrence in the liturgical text Yasna Haptaŋhāiti, which she produced strong reasons for attributing to Zoroaster himself (Narten, 1986; cf. Gershevitch, p. 18; Boyce, 1992, pp. 87-94). In this Yasna 37.3 ends with the words: “Him (Ahura Mazdā) we worship, the fravašis of the righteous, of men and women” (tə̄m ašāunąm fravašīš narąmcā nāirinąmcā yazamaidē). There are two awkwardnesses in this. One is that verses 1-3 are otherwise devoted to Mazdā, so that naming the fravašis here appears out of place. The other is the “formidable grammatical difficulty” (Kellens, 1989, p. 101) of yaz- governing two accusatives, both apparently its direct objects, without a conjunction. Previously scholars in translating had simply supplied an “and,” but Narten, confronting both awkwardnesses, argued (1985, p. 35; idem, 1986, pp. 180-81) that fravašīš here was a rare instance of the accusative used in an instrumental sense, allowing one to translate: “Him we worship through the choices of the righteous.” This was strongly endorsed by Jean Kellens (1989, pp. 101-4). Yet not only does the linguistic argument require special pleading, but interpreting Old Avestan “fravaši” as a common noun meaning “choice” appears highly dubious, because of the huge gulf between such a rendering and the well-attested Young Avestan use of the word. Nor has any scholar upholding Lommel’s subtler rendering been able to suggest why thereafter the fravašis should have become associated with the urvan and absorbed into the cult of the dead.
In this vital respect interpreting the fravašis as by origin spirits of the heroic dead appears to offer the most satisfactory solution. The cult of the dead was of great importance to the Indo-Iranians; and it is not difficult to imagine why during the dangerous Iranian Heroic Age a special cult of dead heroes, able to fight invisibly beside the living, should have become popular. There is no reason to suppose that Zoroaster, conservative in a number of ways, frowned on offering due respect to the souls of the departed; but his thoughts as a prophet were with the living (cf. Y. 39.1-2) and the future, and the actual cult of urvans, with offerings of gifts to ease their lot, had clearly no place in his doctrinal system; for he taught that the soul’s fate, in bliss or torment, was fixed immutably at death according to the individual’s conduct here on earth. No human intercession thereafter could enhance the one or mitigate the other. It is likely, moreover, that he would have looked still more coldly on the cult of the fravašis, if these beings were indeed warrior-spirits, linked with the destructiveness of battle. The strong probability is, therefore, that he required his first followers to abandon both cults; but that as generations passed and the faith spread, more and more converts were reluctant to do so, and that debate over this matter grew intense. The results show that those seeking to maintain them within Zoroastrianism triumphed, despite the doctrinal difficulties, but because of these difficulties, the contest cannot have been easily won. This almost certainly explains the awkward reference to the fravašis in Yasna 37.3, which in all likelihood is a clumsy and audacious interpolation, made so as to gain for their controversial cult the ultimate sanction of endorsement by the prophet himself (Boyce, 1992, pp. 91-2). The earnestness of the endeavors of the cult’s supporters to assimilate it to the faith is demonstrated in Yašt 13, which contains more Zoroastrian doctrine than any other yašt, but with this doctrine modified, sometimes dramatically, to serve this overriding aim.
Moreover, since the cult of the urvan was also probably initially proscribed, it would have been practical to identify it with that of the fravaši and make a single issue of the two, and this may well have helped the cause of the fravaši cult, since that of the urvan, maintained in every home, would have had even more supporters, and was probably a little less controversial. Hence, presumably, the annexation of Hamaspaθmaēdaya for the fravašis; and hence too recurrent declarations, such as the following, that fravaši and urvan are one: “We worship the sunny abodes of Aṧa, in which dwell the souls [urvan-] of the dead, which are the fravašis of the righteous” (xᵛanuuaitīš ašahe vərəzō yazamaide, yāhu iristinąm urvąnō šāyente yå ašaonąm fravašayō, Y. 16.7). The regular qualification of the fravašis as “of the righteous” (ašavan-, q.v.), was presumably part of their reconciliation with Zoroastrianism; and being of the righteous, they and the righteous urvan, identified as one, have both abandoned the places where they dwelt in pre-Zoroastrian thought, one under ground, the other in the air, and are conceived of as being on high in the paradise of the blessed. Nevertheless, the urvan retains some chthonic associations, and this may be why in their yašt the fravašis are so much concerned with the growth of plants and flowing of springs (Yt. 13.10, 14, 29, passim; so Söderblom). Further, the souls of the dead are numberless, and this is probably why the fravašis came to be thought of in vast throngs.
If the original concept of the fravaši was indeed that of the soul of a famed warrior, then there was some justification for stating that fravaši and urvan are one; and some ancient heroes whose souls are venerated in Yašt 13 (131-38) may earlier have been honored in a pre-Zoroastrian fravaši cult. But fusing this cult with that of the urvan, while at the same time aggrandizing the fravašis and weaving belief in them into the fabric of the faith’s mythology, led to insoluble contradictions. These are more likely, however, to have troubled theologians and schoolmen than the mass of believers. Because of them it often seems possible to distinguish in texts between fravašis as urvans and fravašis as heroic protectors. Thus in the story of Zoroaster’s miraculous conception (Dēnkard, 7.2.15-47; Molé, pp. 16-23; Boyce, 1984, pp. 72-73) the prophet’s frawahr, i.e., urvan, is sent down from Paradise within a unique hōm-plant to be united on earth with his bodily substance (tan-gōhr) and xwarrah; and in a work on creation (Bundahišn 3.23-24) the collectivity of fravašis is drawn deep into Zoroastrian myth with the story that in the beginning Ahura Mazdā gave them the choice either to remain in his presence, protected, or to go down into the world to be born into men, there to suffer and help defeat Aŋra Mainyu (see AHRIMAN). But in another work on creation (Zādspram, 3.2-3) the fravašis appear in the role of divine warriors, for there it is told that after Aŋra Mainyu had broken into the world they drew together in close-packed ranks along the sky’s rim, mounted with spear in hand, to imprison him. Giving the fravašis parts to play in the story of creation (as here and in their yašt) meant supposing them to have pre-existed the material world; but whereas the lesser yazatas, of whom also this is true, are said to have been brought then into being by Ahura Mazdā to help him in the struggle against evil, no such theological statement is known about the fravašis. As far as the existing texts go, they are simply there, assimilated into Zoroaster’s dualistic system as beneficent spirits but not, it seems, fundamentally part of it.
To make matters still more complex, already in Yašt 13 (149, 155) urvan and fravaši are recognized as separate components (along with ahu, daēna and baoḏah) of man’s immaterial being; and similar analyses are found in the Pahlavi books (Bailey, Zoroastrian Problems, pp. 97-118). One explanation offered there of the relation of urvan to fravaši on this plane is that during this life the urvan dwells in the body, the fravaši in the presence of Ohrmazd (which does not accord with the story of Zoroaster’s conception), but that at death they are united (Bundahišn 3.13). Elsewhere it is said that the fravaši is united only with the blessed soul and cannot go to hell (Dēnkard, ed. Madan, p. 123, ll. 10-13; Zaehner, p. 270); but popular belief (which appears the main force in the persistence of the urvan-fravaši cult, among priests as well as laity) maintained that at Hamaspaθmaēdaya all “fravašis” return to their former homes, some from the joys of heaven, others from the torments of hell (Bīrūnī, Āṯār, p. 224; Boyce, Stronghold, pp. 213-14).
The persistence of irreconcilable ideas about the fravaši is shown by uses to which the term was put by Manichean and Christian proselytizers in the Sasanian period. The Manicheans used Middle Persian frāwahr (prʾwhr), Parthian ardāw frawardīn (ʾrdʾw frwrdyn, in Sogd. ʾrtʾw frwrṯyy) as a term for air or ether, the purest of the five Manichean light-elements (Andreas and Henning, p. 185, a I R I 5; Sundermann, pp. 147, 155, svv.). This was plainly because the frawahr was regarded by Zoroastrian schoolmen as the highest of the five immaterial components (in Manichean thought, light-elements) of the microcosm, man. But Christian missionaries in Sogdia, working perhaps among humbler groups of Zoroastrians, adopted frwrt for “corpse,” frwrṯqty for “corpse-house, tomb” (Sims-Williams, p. 211, svv.); and this appears to reflect a local usage parallel to the common Zoroastrian one of today of speaking at funerals of the corpse as “rovān,” a euphemism deprecated by some but partly justified by the belief that the soul, lingering on earth, accompanies the body to the place of disposal (see corpse).
Despite the confusions which it produced, the postulated struggle in the Young Avestan period to have the fravaši-urvan cult accepted was overwhelmingly successful. Every liturgical text contains their veneration of these beings with the words of Sīrōza 1 or 2 (see FRAWARDĪN YAŠT) and they are regularly invoked in the Middle Persian dedications of religious services, in which Av. ašāonąm fravašayō is rendered by “artā” (from OPers. *artāvanąm) frawahr or frawaš. No act of worship takes place, therefore, without their invocation, which is understood to be that of the souls of the departed (cf. Boyce, 1968, pp. 270-71 with no. 4). Further, the fourth watch (gāh) of the twenty-four hours, from sunset to midnight, is believed to be under their protection, and in the Zoroastrian calendar a day and a month are dedicated to them (see calendars i). This means that they have an annual name-day feast on the day Farvardīn (“of the fravaršīs”) of the month Farvardīn; and still today on Farvardīn Jašan the devout, and especially those who have lost a relative during the year, go to the daḵmas (q.v.) and have religious ceremonies performed for the souls of departed family members (Seervai and Patel, p. 216). Nevertheless, despite the predominant association in living Zoroastrianism of fravašis with souls of the dead, there is still a sense of their power as protective spirits, who are invoked among the yazatas.
There is no certain plastic representation of the fravašis. The Avestan term, an abstract, is grammatically feminine, but there is no indication that the spirits themselves were conceived as Valkyrie-like beings. In Yašt 13 (v. 70) they are said to “swoop down like an eagle,” and wings are still popularly attributed to them, locally at least, among Zoroastrians (Boyce, Stronghold, p. 161, n. 47). Moulton accordingly (p. 260), rejecting the then prevailing Western interpretation of the Achaemenid winged symbol as that of Ahura Mazdā, identified it as the king’s fravaši, and it is still generally regarded as a fravaši symbol by Zoroastrians. Strong arguments, however, have by now been put forward by Western scholars for its representing xᵛarənah (for references see Boyce, Zoroastrianism II, p. 103 n. 82; Boyce and Grenet, Zoroastrianism III, p. 104 and nn. 197-98).
F. C. Andreas and W. B. Henning, Mitteliranische Manichaica aus Chinesisch-Turkestan I, Sitzungsberichte der Preussischen Akademie der Wissenschaften 10, 1932, Berlin, 1932.
Boyce, Zoroastrianism I, pp. 117-29; II, pp. 55-56.
Idem, “The Pious Foundations of the Zoroastrians,” BSOAS 31/2, 1968, pp. 270-89.
Idem, Textual Sources for the Study of Zoroastrianism, Manchester, 1984.
Idem, Zoroastrianism, Its Antiquity and Constant Vigour, Columbia Lectures on Iranian Studies 7, Costa Mesa, Calif., 1992.
J. Darmesteter, Le Zend Avesta, Annales du Musée Guimet, 3 vols., Paris, 1892-93; repr., Paris, 1960.
J. Duchesne-Guillemin, “L’homme dans la religion iranienne,” in C. J. Bleeker, ed., Anthropologie religieuse, Leiden, 1955, pp. 93-107.
G. Dumézil, “Víṣṇu et les Mārūt à travers la réforme zoroastrienne,” JA 241, 1953, pp. 1-25.
I. Gershevitch, “Old Iranian Literature,” in HO IV, 2/1, pp. 1-30.
G. Gnoli, “Le ‘Fravaši’ et l’immortalità” in La mort, les morts dans les sociétés anciennes, Cambridge and Paris, 1986, pp. 339-47.
K. Hoffmann, “Das Avesta in der Persis,” in J. Harmatta, ed., Prolegomena to the Sources on the History of Pre-Islamic Central Asia, Budapest, 1979, pp. 89-93.
J. Kellens, “Les frauuaṧis dans l’art sassanide,” Iranica antiqua 10, 1973, pp. 133-38.
Idem, “Les fravašis” in Homo Religiosus 14: Anges et Démons, Actes du Colloque de Liège et Louvain-al-Neuve, 1987, pp. 99-114.
H. Lommel, Die Yäšt’s des Awesta, Göttingen and Leipzig, 1927, pp. 101-30.
W. W. Malandra, The Fravaši Yašt, University Microfilms, Ann Arbor, 1971.
Idem, An Introduction to Ancient Iranian Religion, Minneapolis, 1983, pp. 102-17.
M. Molé, La légende de Zoroastre selon les textes pehlevis, Paris, 1967.
J. H. Moulton, Early Zoroastrianism, London, 1913a, repr. 1972, pp. 254-85.
Idem, “Fravashi,” in J. Hastings, ed., Encyclopaedia of Religion and Ethics, Edinburgh, 1913b, pp. 116-18.
J. Narten, “Avestisch frauuaṧi,” Indo-Iranian Journal 28, 1985, pp. 35-48.
Idem, Der Yasna Haptŋhāiti, Wiesbaden, 1986.
B. Schlerath, “Indo-iranisch *var ‘wählen,’” Studien zur Indologie und Iranistik 5/6, 1980, pp. 199-208.
K. N. Seervai and B. B. Patel, “Gujarāt Pārsis,” Gazetteer of the Bombay Presidency IX/2, Bombay, 1899, pp. 183-288.
N. Sims-Williams, The Christian Sogdian Manuscript C2, Berliner Turfantexte 12, Berlin, 1985.
N. S. Söderblom, “Les fravashis,” RHR 39, 1899, pp. 229-60, 373-418.
W. Sundermann, Der Sermon vom Licht-Nous, Berliner Turfantexte 17, Berlin, 1992.
J. C. Tavadia, “Jaina Parallelen zu den awestischen Begriffe von Leib und Seele (daēnā-, fravaši- und kəhrp-),” ZII 10, 1935, pp. 192-200.
R. C. Zaehner, The Dawn and Twilight of Zoroastrianism, London, 1961; repr. London, 1975.
Originally Published: December 15, 2000
Last Updated: January 31, 2012
This article is available in print.
Vol. X, Fasc. 2, pp. 195-199