Ardwīsūr Anāhīd, Middle Persian name of Arədvī Sūrā Anāhitā, a popular Zoroastrian yazatā; she is celebrated in Yašt 5 (known as the Ābān Yašt) which is one of the longest and best preserved of the Avestan hymns. Sūrā and anāhitā are common adjectives, meaning respectively “strong, mighty” and “undefiled, immaculate.”


ANĀHĪD (Old Pers. Anāhitā, New Pers. Nāhīd, Armenian Anahit, Greek Anaitis), Mid. Pers. form of the name of the Iranian goddess Anāhitā. The subject will be treated in four sections:

i. Ardwīsūr Anāhīd.

ii. The cult and its diffusion.

iii. Anaitis.

iv. Anāhitā in the arts.

i. Ardwīsūr Anāhīd

Ardwīsūr Anāhīd, Middle Persian name of Arədvī Sūrā Anāhitā, a popular Zoroastrian yazatā; she is celebrated in Yašt 5 (known as the Ābān Yašt) which is one of the longest and best preserved of the Avestan hymns. Sūrā and anāhitā are common adjectives, meaning respectively “strong, mighty” and “undefiled, immaculate.” Only arədvī (a word otherwise unknown) is special to this divinity, and on etymological grounds it too has been interpreted as a feminine adjective, meaning “moist, humid.” The proper name of the divinity in Indo-Iranian times, H. Lommel has argued, was Sarasvatī, “she who possesses waters” (“Anahita-Sarasvati,” Asiatica, Festschrift F. Weller, Leipzig, 1954, pp. 405-13). She was still worshiped in Vedic India by this name, which was also given there to a small but very holy river in Madhyadeśa. In its Iranian form (*Harahvatī), her name was given to the region, rich in rivers, whose modern capital is Kandahar (Av. Haraxᵛaitī-, OPers. Hara(h)uvati-, Greek Arachosia); originally *Harahvatī seems to have been the personification of a great mythical river which plunges down from Mt. Harā into the sea Vourukaša and is the source of all the waters of the world. It is thus that the yazatā is celebrated in Yašt 5 and in the Pahlavi books; but in time, it appears, her proper name fell into disuse in favor of her epithets arədvī and sūrā, which eventually coalesced to give her the Middle Iranian name of Ardwīsūr. In her hymn the river-goddess is described as a beautiful, strong maiden, clad in beaver-skins (5.129), who drives a chariot drawn by four horses: wind, rain, clouds, and sleet (5.120). As water-divinity she is worshiped as a bestower of fertility, who purifies the seed of all males, the wombs of all females, and makes the milk flow which nourishes their young (5.2). Like the Indian Sarasvatī, she nurtures crops and herds; and she is hailed both as a divinity and as the mythical river which she personifies, “as great in bigness as all these waters which flow forth upon the earth” (5.3). There is a mantic link in many ancient cultures between water and wisdom, and priests and their pupils pray to Arədvī Sūrā for knowledge (5.86); while in India Sarasvatī protects the study of the Vedas. As a water-divinity Arədvī Sūrā is linked with the Āpas (see Ābān), and verses from her hymn form the greater part of the Ābān Niyāyeš. She is also associated with Apąm Napāt (who, in the view of the present writer, represents the great ancient deity *Vouruna, see Apąm Napāt) and the rain-bringing Tištrya.

It seems less in the character of a river-goddess that Arədvī Sūrā is also held to bestow upon her worshipers possessions such as chariots, arms, and household goods (5.130), as well as victory in battle and the destruction of foes (5.34ff.). Some of the verses which indicate these aspects of her power correspond closely with others addressed to Aši, yazatā of Fortune; and there seems to have been some blurring of identity between these two beautiful, chariot-driving goddesses. Linguistically Arədvī Sūrā’s hymn appears older than Aši’s Yt. 17), which is short and badly preserved; and so it has been assumed that, where there are verses in common, it was Aši who was the borrower. In a fluid, oral literature, however, such criteria cannot be relied on. Once Arədvī Sūrā gained greater popularity, her hymn would have been more often recited and so would be better preserved; there would be a tendency, moreover, for priests to seek to extend it in her honor. “Great-gifted Aši” is a Gathic figure, worshiped of old; and it seems probable that, as she suffered gradual eclipse by Arədvī Sūrā, verses once addressed to her were transferred to her rival, so that gifts properly sought from the goddess of Fortune came to be asked of the river-goddess.

Arədvī Sūrā’s striking growth in popularity seems to have begun in Achaemenid times, through her identification with the Western Iranian divinity *Anāhiti, known from Greek sources as Anaitis (see below). The Achaemenids’ devotion to this goddess evidently survived their conversion to Zoroastrianism, and they appear to have used royal influence to have her adopted into the Zoroastrian pantheon. The problem of how to offer veneration to a divinity unknown to the Avesta was solved by assimilating *Anāhiti to *Harahvaitī Arədvī Sūrā Anāhitā, whose third epithet was very close to the western divinity’s proper name, and indeed may already in late Old Persian have become identical with it, through the dropping of the final vowel in ordinary speech.

The first Achaemenid king known publicly to have acknowledged “Anāhit(a)”—that is, the composite being born of the assimilation of Arədvī Sūrā Anāhitā and *Anāhiti—was Artaxerxes II (404-359 B.C.), who in inscriptions invoked her after Ahura Mazdā and Mithra, and who also set up cult-statues in her honor (see further under Anaitis); and it was presumably after this that verses were composed and incorporated in Yašt 5 which apparently describe a temple statue (see Ābān Yašt). In these Arədvī Sūrā Anāhitā is invoked, not as the personification of a rushing river, but as a magnificently static being, richly arrayed in high-girt robe and jewel-encrusted mantle, with golden shoes and earrings, necklace, and crown. There is no similar description of any other Avestan divinity; and the contrast between it and the concept of Arədvī Sūrā in bold motion, drawn swiftly on by her four elemental steeds, suggests how uneasy in some ways was the reconciliation of *Harahvatī and *Anaitis. In the Pahlavi books (some of which represent lost Avestan texts), the two are still sometimes treated as separate divinities, with Ardwīsūr as the personification of the mythical river, and Anāhīd, the fertility goddess, identified with the planet Venus. Thus the Greater Bundahišn, in describing the world’s lakes and seas, says they all have their origin with “Ardwīsūr” (10.2, 5); whereas, in a paragraph concerned with the stars and planets (5.4), there is mention of “Anāhīd ī Abāxtarī,” i.e., the planet Venus. In other chapters, however, the two divine beings are identified, e.g., 3.17, “Ardwīsūr who is Anāhīd, the father and mother of the Waters” (Ardwīsūr ī Anāhīd, pid ud mād ī Ābān). In the cult the two became indissolubly one. This is attested by her names in the Avesta; further, at a shrine in Asia Minor in Roman times “Anaïtis” was invoked with what seems to be an ancient epithet of *Harahvatī’s, namely, “of high Harā (barzochára; see R. Schmitt, “Ein neues Anahita-Epitheton aus Kappadokien,” ZVS 84, 1970, pp. 207-10; see, contra, S. Wikander, in Acta Orientalia 34, 1972, pp. 13-15), while in another Greek inscription there she is spoken of as “Anaïtis of the sacred water” (L. Robert, “Monnaies grecques de l’époque impériale,” Revue numismatique, 6th series, 18, 1976, pp. 45-46).

Nevertheless, there is some evidence to suggest that there were orthodox priests who put up what resistance they could to the royally favored syncretic cult, with its alien elements of temple- and image-worship. Thus, although Yašt 5 seems to have been adapted to incorporate the veneration of Anaïtis, and although “Anāhitā” seems to have displaced *Vouruna in the triad of high divinities worshipped by the Achaemenids, yet in the liturgy of the yasna (y.2.5 et passim) it is still *Vouruna, as Apąm Napāt, who is invoked with the Waters. Moreover, in the dedications of the days of the month (bestowed, it seems, in late Achaemenid times) a day is assigned to Aši (Middle Persian Ard), but none to her rival Arədvī Sūrā Anāhitā. It also seems probable that the characteristic Zoroastrian temple-cult of fire developed at this same period in opposition to the image-cult of Anaïtis (see further under ātaš). Despite this degree of priestly resistance, the cult of Arədvī Sūrā Anāhitā, uniting as it did those of water-goddess and mother-goddess, and being royally promoted, became widely popular. Worship was, in general, offered to the divinity under the name of Anāhīd (Anāhīt)/Anaïtis, which suggests the strength of Achaemenid influence. The Arsacids followed the example set by their predecessors in venerating Aramazd-Mihr-Anāhīd as their chief helpers; and the woman’s name Āb-Nāhīd (“Anāhīd of water”) is first attested in the Parthian period (see Faḵr-al-dīn Asʿad Gorgānī, Vīs o Rāmīn, ed. M. Mīnovī, Tehran, 1314 Š./1935, section 9.5). The temples to Anāhīt founded by Artaxerxes II probably all survived Alexander’s conquest and Seleucid domination, even though pillaged. Thus the one at Hamadān (Ecbatana) was twice plundered and was stripped of its gold and silver roof-tiles; but it was evidently restored, for Isidore of Charax (Parthian Stations 6) wrote of sacrifices being continually offered there in his day. A temple at Kangāvar was apparently also devoted to Anāhīd, if this place is indeed Isidore’s Concobar (loc. cit.); for he said that a temple there was dedicated to Artemis, which was one of the Greek identifications of Anāhīd (but see below, sec. iv). Hellenic influence having given a new impetus to the cult of images in Iran, it may safely be assumed that Anāhīd’s statues were still venerated during the Parthian period; and positive evidence for this comes from Armenia, then a Zoroastrian land. Here Anāhīd was much beloved, being invoked as “noble Lady... mother of all knowledge, daughter of the great and mighty Aramazd.” There are references to offerings at her altars; and in 36 B.C. one of Mark Antony’s soldiers carried off a famous statue to her in solid gold from the temple at Erez. A fine bronze head, like that of a Greek Aphrodite, has been found at Satala, which is thought to belong to a statue of Anāhīd. (All statues in Armenia, according to an old source, were made by Greek craftsmen.)

It is very likely that in the Parthian period, and probably even earlier, Ardwīsūr Anāhīd was also worshipped at many natural sanctuaries throughout the land, created by lake or mountain spring. One of these (which, to judge by its great sanctity, is probably old) was on a mountain with a spring at its foot, near the city of Ray. This shrine seems to have been devoted to Anāhīd as “the Lady of the Land” (Šahrbānū); and so great was the veneration in which it was held that, after the Arab conquest, it was rededicated to “Bībī Šahrbānū,” held to be a daughter of the last Sasanian king and the widow of Ḥosayn, son of ʿAlī b. Abī Ṭāleb (see M. Boyce, “Bibi Shahrbānū and the Lady of Pārs,” BSOAS 30, 1967, pp. 30-44). Muslim prayers and sacrifices are accordingly offered there to this day. Worship of the divine beings in the presence of natural objects is more consonant with orthodox Zoroastrianism than is the veneration of man-made images; and it is probable that an iconoclastic spirit sprang into being among some groups of Zoroastrians at the moment when Artaxerxes II set up the first statues to Anāhīd. There are slight indications that this spirit began to find active expression towards the end of the Parthian period, as Hellenistic influences waned; and it is possible that some destruction of statues, Anāhīd’s among them, began then. At the beginning of the 3rd century A.D. the Persian Sasanians were, it seems, hereditary guardians of a temple dedicated to Anāhīd at Eṣṭaḵr (probably one of the foundations of Artaxerxes II), which Ṭabarī describes as “the temple of the fire of Anāhīd”; the fact that Šāpūr I’s queen of queens (his daughter-wife) was called Ādur-Anāhīd (“Anāhīd of the fire”) suggests that a sacred fire, consecrated to Anāhīd, had replaced her image in this temple before the princess was born—-i.e., by the beginning of Sasanian ascendancy. Two reigns later, under Bahrām II, the high priest Kirdēr was honored with responsibility for two sacred fires at Eṣṭaḵr, one called “the Fire of Anāhīd the Lady,” the other “the Fire of Anāhīd-Ardašīr” (line eight of his Kaʿba-ye Zardošt inscription). The significance of the second dedication is uncertain.

Anāhīd was thus the patron divinity (under Ohrmazd) of the Sasanians, and her cult flourished during their rule as it had done during the two earlier empires, although she was now officially venerated, it seems, without statues. It has been suggested that a sunken temple made by Šāpūr I beside his palace at Bīšāpūr, whose stone-paved sanctuary could be flooded with water, was a temple to Anāhīd, where she could be worshiped in the presence of her natural icon, water (R. Ghirshman, Bīchāpour I, Paris, 1971; idem, Iran, Parthes et Sassanides, Paris, 1962, p. 149). Sasanian iconoclasm was evidently directed only, however, at free-standing cult-images, and representations of Anāhīd survive in Sasanian art. In an investiture scene carved at Naqš-e Rostam, Narseh had himself represented receiving the diadem of kingship from the hand of a female divinity generally recognized as Anāhīd; late in the epoch Ḵosrow Parvēz showed his fidelity to the family tradition by having Anāhīd present to support him at his investiture scene also, which was carved in the lake-side grotto of Ṭāq-e Bostān (for other interpretations see G. Hermann, The Iranian Revival, Oxford, 1977, p. 103). Here the divinity holds in one hand a tilted jug, from which water flows. There is little doubt that under the Sasanians Anāhīd overshadowed all other female divinities as far as private prayers and devotion were concerned, although in public worship the great Amešaspands, Spendārmad, Hordād and Amurdād, continued to be more honored, thanks evidently to the conservatism and orthodoxy of the priests. Even granted the widespread popularity of Anāhīd, however, it is doubtful whether the current tendency is justified whereby almost every isolated female figure in Sasanian art, whether sitting, standing or dancing, clothed or semi-naked, is hailed as her representation (see below).

The dedication of her fire at Eṣṭaḵr shows that, to Persians as to Parthians, Anāhīd was known as “the Lady.” In his inscription at Paikuli (Pahlavi text, line 10), Narseh invokes “Ohrmazd and all the yazads, and Anāhīd who is called the Lady”; and a Sasanian gem bearing what is thought to be a representation of her has beneath it simply the identification “the Lady” (bʾnwky). This usage influenced Zoroastrian priestly terminology in late Sasanian and Islamic times, and the yazata is spoken of then in religious works as “Ardwīsūr the Lady” and “Ardwīsūr, the Lady of the Waters” (for references see M. Boyce, “Bibi Shahrbānū,” p. 37, nn. 27, 28). One of the most beloved mountain shrines of the Zoroastrians of Yazd, set beside a living spring and a great confluence of water-courses, is devoted to Bānū-Pārs, “The Lady of Persia.” This sanctuary appears to have been devoted originally to Anāhīd “the Lady”; being rededicated in Islamic times, like the shrine of Bībī Šahrbānū near Ray, to a legendary Sasanian princess (see M. Boyce, “Bibi Shahrbānū”). This and other new dedications appear to have led to the partial eclipse of Anāhīd herself in living Zoroastrianism, although the veneration of the waters continues as an important part of the cult. It may be suggested that most of the many places in Iran, in mountains and by springs, which are named for “the Maiden” (Doḵtar) or “the Lady” (Bībī) were once sacred to Anāhīd. The Yazdi Zoroastrians still today often call their daughters by the name Āb-Nāhīd.


See also M. H. Ananikian, Armenian Mythology, Boston, 1925, pp. 20-21.

L. H. Gray, Foundations, pp. 55-62.

H. Lommel, Die Yäšt’s des Awesta, Göttingen and Leipzig, 1927, pp. 26-32.

S. Wikander, Feuerpriester in Kleinasien und Iran, Lund, 1946, chap. III, VII.

J. Šahīdī, Čerāḡ-e rowšan dar donyā-ye tārīk, Tehran, 1333 Š./1954.

M. E. Bāstānī Pārīzī, Ḵātūn-e haft qaḷʿa, Tehran, 1344 Š./1965.

M. Boyce, “Iconoclasm among the Zoroastrians,” Studies for Morton Smith at sixty, ed.

J. Neusner, Leiden, 1975, pp. 93-111.

Idem, Zoroastrianism I, pp. 71-74.

(M. Boyce)

ii. Anaïtis

Anaïtis is the Greek rendering of what appears to have been the name of the goddess of the planet Venus, who seems to have been worshiped by the Medes and Persians before they adopted Zoroastrianism. Her cult was apparently much influenced by that of Mesopotamian Ishtar, an enormously powerful divinity in the first millennium B.C., whose worship had by then been adopted in a number of pantheons (including that of the Elamites) outside the Semitic world. Ishtar was venerated as goddess both of love and war, and this is thought to be because in earlier times the morning and evening appearances of the planet Venus (with whom she was linked) had been regarded as those of two different though related stars, with the divinity of the evening star being held to be female, that of the morning star male. By the first millennium the identity of the planet as seen at dawn and twilight had come to be accepted by Babylonian astronomers. It cannot be supposed that this identity had been perceived earlier by the Iranians; but it appears probable that they had long been accustomed to venerate the brilliant planet, either at its morning or its evening appearance, as the goddess *Anāhiti, the “Pure One.” This name is represented in its Old Persian form only by Greek Anaïtis. The Middle and New Persian forms, Anāhīd, Nāhīd, have long internal “i”; and so it was assumed that Greek Anaïtis was written by itacism for *Anaeitis, with ei for OP ī, as elsewhere. It now seems more probable, however, that OP had an internal short “i” which became lengthened, regularly, in Mid. Pers. after the loss of the final syllable (cf. M. Back, Die sassanidischen Staatsinschriften, Acta Iranica 18, Leiden, 1978, p. 70). *Anāhiti can thus be understood as a regular feminine bahurvihi “having no stain, immaculate” (cf. AirWb. col. 125; Lommel, Die Yäšt’s des Awesta, Göttingen, 1927, p. 29).

Presumably the ancient Persians, having settled in the land of the Elamites, there learned to worship their goddess *Anāhiti in connection with both appearances of the planet Venus, and to associate her with the powerful Ishtar, called “the Lady.” (“Lady” is a characteristic Mesopotamian invocation of a goddess.) Her cult gained accordingly in popularity, and evidently presented a problem for Zoroastrian orthodoxy, once the western Iranians had embraced the eastern faith. The difficulty of how to incorporate the cult of *Anāhiti into Zoroastrian worship was probably not solved until the reign of Darius II, whose son Artaxerxes II publicly invoked “Anāhit(a)” in his inscriptions (A2Sa, A2Sd, A2Ha). Further, the Babylonian scholar-priest, Berossos (Book III, fragment 65) states that this king “was the first to set up statues of Aphrodite Anaitis, at Babylon, Susa, Ecbatana, Persepolis, Bactra, Damascus, and Sardis, thus suggesting to those communities the duty of worshiping them.” The way in which it was made doctrinally possible for a Zoroastrian king thus to impose the veneration of a partly alien divinity on the community at large was through assimilating her cult to that of the Zoroastrian Yazatā Arədvī Sūrā Anāhitā. Thereafter “the Lady” of the planet Venus was still popularly worshiped as Anāhīd ī Bānū, but was venerated in the Zoroastrian liturgies with the Avestan invocations proper to the river-yazatā. On the fusion, only imperfect, of the conception of the two divinities see further under Ardwīsūr Anāhīd and Ābān Yašt.

For a parallel instance of the influence of a Babylonian planetary cult on Iranian worship, see Tīr(i); and for the assimilation thereafter of Tīr(i)’s cult into Zoroastrianism see under Tištrya. A further complexity in the cult of “Anāhita” is the veneration also of the Mesopotamian goddess Nanāˊ (q.v.), which seems to have entered Zoroastrian worship in association with that of Ishtar-Anāhiti.


F. Windischmann, Die persische Anahita oder Anaïtis, Abh. der Königlichen Bayerischen Akademie der Wissenschaften 8, Munich, 1858.

E. Meyer, “Anaitis,” in W. H. Roscher, Ausführliches Lexicon der griechischen und römischen Mythologie, Leipzig, 1884-86, I pp. 330-34.

F. Cumont, Encyclopaedia of Religion and Ethics, ed.

J. Hastings, I, Edinburgh, 1908, pp. 474ff.

G. Gnoli, “Politique religieuse et conception de la royauté sous les Achéménides,” Commémoration Cyrus, Acta Iranica 2, Leiden, 1974, pp. 126ff.

M. Boyce, Zoroastrianism II, chaps. 2, 12, 13.

For further literature see the bibliography to Ardwīsūr Anāhīd.

(M. Boyce)

iii. The Cult and Its Diffusion

Although the Greeks sometimes assimilated Anāhitā to Aphrodite (e.g., Herodotus, Historiae 1.131-32, knows her by the name of Aphrodite Urania and compares her with the Assyrian Mylitta and the Arabian Alilat) or Athena, they most often viewed her as the Persian Artemis. She is not named in the Elamite texts found at Persepolis and dating from the reigns of Darius I and Xerxes. In 405 B.C., the year of the accession of Artaxerxes II Mnemon, there was a temple at Pasargadae in Persis dedicated to a warrior goddess who, according to Plutarch (Artaxerxes 3), could be compared with Athena; no doubt it was a temple of Anāhitā in one of her most important aspects. The fact that Artaxerxes II Mnemon (404-359 B.C.) received consecration in the kingship at this temple after he had donned the robe of Cyrus (Plutarch, loc. cit.), suggests that the Achaemenid monarchy had close links with Anāhitā, especially in her war-goddess aspect. Moreover Mnemon was the first Achaemenid to insert the names of Anāhitā and Mithra, after that of Ahura Mazdā, in official documents (Kent, Old Persian, p. 154). According to a passage from the Chaldaica of Berosus conserved by Clement of Alexandria (Protrepticus 5.63.5, ed. C. Mondésert and A. Plassard, Paris, 1949, p. 139), the same king caused statues of “Aphrodite Anaitis” to be erected in major cities of his empire such as Babylon, Susa, Ecbatana, Persepolis, Bactra, Damascus, and Sardis. Plutarch (Artaxerxes 27) also states that Artaxerxes Mnemon piously made his concubine Aspasia become a priestess of “Artemis whom they call Anaitis.” It was probably in his reign that the Anāhitā cult began to gain ground in Asia Minor and Syria before spreading to Armenia.

The cult long flourished in Lydia, which had temples of the Persian Artemis at Sardis, Philadelphia, Hierocaesarea, Hypaipa, Maeonia, and elsewhere; the temple at Hierocaesarea had reputedly been founded by Cyrus (Tacitus Annals 3.62). From the 2nd century A.D. there is an account of the ceremonies performed in accordance with ancient Mazdaean ritual at Hypaipa and Hierocaesarea, as personally witnessed by the geographer Pausanias (Description of Greece 7.27.5. For monuments and inscriptions in honor of Artemis Anaïtis in Lydia and Catacecaumene, see I. Diakonoff, in Babesch. Bulletin Antieke Beschaving, 1979, pp. 145f., 148f.). At Zela in Pontus the goddess was venerated together with two associate gods, Omanos and Anadates (Strabo Geography 11.8.4, 12.3.37); in Cappadocia likewise she and Omanos had common altars (ibid. 15.3.15). At Castabala she was named Artemis Perasia (ibid. 12.2.7). In connection with the cult of Artemis/Anaïtis, Hellenic-style games were held at places such as the Anaeiteia at Philadelphia, the Artemisia at Hypaipa, and the Sakaia at Zela. The fact that bulls were the animals sacrificed to Anaïtis probably explains why in Lydia, Cappadocia, and Armenia she was assimilated to Artemis Tauropola or Taurica. It was through this channel that the taurobolium (bull-sacrifice ritual) spread to Europe.

The Armenians, according to Strabo (Geography 11.14.16), shared in the religion of the Persians and the Medes and particularly honored Anaïtis (see Armenian religion). From the 1st century A.D. onward, a temple of this goddess at Eriza (Erez) in Acilisene enjoyed great fame; an unverifiable tradition ascribed its foundation to Tigranes the Great (Moses of Khoren 2.14 in Langlois, Historiens II, p. 88). This holy place, “the wealthiest and most venerable in Armenia” (Cicero Pro lege Manilia 9.23), was staffed with priests and priestesses; daughters of the most eminent families were required to serve as prostitutes in it before marrying (Strabo Geography 11.14.16), an element likely borrowed from the Semitic religious practices (E. Meyer, Lexikon der griechischen und römischen Mythologie, ed. W. H. Roscher, I, 1884, col. 333). Anaïtis was worshipped at Eriza in the guise of a huge gold image. In 34 B.C. (rather than 36 B.C.) this idol was taken away by Mark Antony’s soldiers who smashed it and shared the fragments among themselves (Pliny, Naturalis historia 33.82-83). Acilisene, being the cult’s main center in Armenia, came to be known as Anaetica, i.e., the land of Anaïtis (Dio Cassius 36.48.1; Pliny Naturalis historia 5.83). Another region lying on the Cyrus river, near the borders of Iberia and Albania, was also called “the land of Anaitis” (Dio Cassius 36.53.5); like Acilisene, it was doubtless the territory of a temple dedicated to Anāhitā but otherwise unknown. The kings of Armenia were steadfast supporters of the cult at Eriza, which seems to have been closely associated with the national monarchy. Tiridates III, before his conversion to Christianity, prayed officially to the triad Aramazd-Anahit-Vahagn but is said to have shown a special devotion to “the great lady Anahit . . . the benefactress of the whole human race, mother of all knowledge, daughter of the great Aramazd” (Agathangelos, section 22, in Langlois, Historiens, I, p. 127). According to Greek version of the Book of Agathangelos, tradition required the kings to travel annually to Eriza on the occasion of the goddess’s festival (G. Garitte, Documents pour l’étude du livre d’Agathange, Vatican City, 1946, p. 78). Tiridates accordingly made this pilgrimage and offered sacrifices as well as wreaths and boughs to Anāhīd (Anahit) in the first year of his reign (Agathangelos, section 21, in Langlois, I, pp. 125-26). Anāhīd was also worshiped at the capital, Artashat (Artaxata), at Astishat, in Taron (south of Armenia) and at many other places. At Artashat her temple was close to that of Tiur (Tīr), the oracular god assimilated to Apollo. At Ashtishat, one of the main centers of Armenian paganism associated with the god Vahagn (Verethragna) and the goddess Astlik, she was worshiped in the guise of a golden idol apparently known as oskimayr “the golden mother” (Agathangelos, section 141, in Langlois, Historiens I, p. 173). Another center of Anāhitā’s worship was the city of Tomisa on the Euphrates in Sophene (south-west Armenia) on the Cappadocian frontier. In 69 B.C., the soldiers of Lucullus could see in the territory of Tomisa plenty of sacrificial cows roaming around freely, which were consecrated to Persia Artemis and bore on the head the brand of her in the shape of a torch (Plutarch Lucullus 24.6). After the conversion of Tiridates, the images of Anahit throughout Armenia were smashed.

Regarding the Caucasian countries adjacent to Armenia, Strabo (Geography 11.2.17) states that there was a temple dedicated to Leucothea, obviously an analogue of the Iranian goddess (O. G. von Wesendonck, Caucasica I, 1924, p. 87) in the land of the Moschi in Colchis. The legendary and late-dated Life of the Apostle St. Andrew mentions a cult of Apollo and Artemis, that is, Mithra and Anāhitā, in the same region. On the other hand there is not evidence of Anāhitā worship in Iberia, a country that had close ties with Armenia and Iran. In Albania the moon stood highest among three popular deities and was worshiped in a famous temple with a large staff of priests at a place near the Iberian frontier (Strabo Geography 11.4.7); some have tried to identify her with Anāhitā (See K. V. Trever, Ocherki po istorii i kul’rure Kavkazskoĭ Albanii, Moscow and Leningrad, 1950, p. 151), but this is questionable.

In Parthian territory, Ecbatana, the greatest metropolis of Media, retained a temple of Anāhitā where sacrifices were regularly offered (Isidore of Charax Mansiones Parthicae, sec. 6; Polybius Histories 10.27.12). At Concobar (Kangāvar) in Lower Media a temple of “Artemis,” built about 200 B.C., was standing when Isidore of Charax (ibid.) wrote, and some vestiges of this Greek-style edifice survive today (L. Vanden Berghe, Archéologie de l’Iran ancien, Leiden, 1959, p. 108; R. Ghirshman, Iran: Parthes et Sassanides, Paris, 1962, p. 24 and fig. 30; V. G. Lukonin, “The Temple of Anāhitā at Kangāvar,” VDI, 1977, no. 2, pp. 105-11). Isidore mentions two more temples of this goddess, both on the right bank of the Euphrates in Mesopotamia, one at Basileia (OPers. apadāna), reputedly founded by Darius, the other at Beonan (see M. L. Chaumont, La route royale des Parthes de Zeugma à Séleucie du Tigre, forthcoming). Susa likewise had a place of worship that, in the words of Pliny (Naturalis historia 6.35), was Dianae templum augustissimum. Not far away was a temple in Elymais called Ta Azara, which was sacred to both Athena and Artemis (Strabo Geography 16.1.18); tame lions were to be seen in it. In Persis the cult was centered in Eṣṭaḵr (Persepolis), where Anāhitā was worshiped in her aspect of war-goddess, as she had been at Pasargadae in the Achaemenid period (see above). Around the end of the 2nd century A.D., the temple of Anāhitā at Eṣṭaḵr was in the custodianship of Sāsān, who was also a hunter and an intrepid warrior (T. Nöldeke, Geschichte der Perser, p. 4). His son Pāpak killed the king of Eṣṭaḵr and seized the throne. It seems probable that Pāpak continued to be the high priest after he made himself king, because a rock carving shows him making an officiant’s obeisance before a fire altar while wearing a crown and holding a sort of scepter in his left hand, with his son Šāpūr on horseback close by him (E. Herzfeld, Iran in the Ancient East, London, 1941, p. 307, figs. 401 and 402).

Pāpak’s son Ardašīr rebelled against his Arsacid suzerain and gradually conquered all the Parthian territories. He may perhaps have inherited the high priesthood of Anāhīd and have derived his spiritual authority mainly from it; in any case he showed great devotion to the goddess, to whom he sent heads of his slain enemies (Nöldeke, Geschichte der Perser, p. 17). As for Ardašīr’s son and successor, Šāpūr I, the fact that he named his daughter Ādur-Anāhīd (Anāhīd of the fire) is certainly significant. In the reign of Bahrām II (276-83), the ambitious Magian Kirdēr, who had been steadily rising in the religious hierarchy and tightening his political grip, reached the zenith of his power when the monarch appointed him chief mōbad and judge of the empire and at the same time “ēwēnbad (master of ceremonies) and supreme head of the fire [temple] of Anāhīd, [that] of Anāhīd-Ardašīr, and the lady Anāhīd” (inscription of Kirdēr, Kaʿba-ye Zardošt, line 8). Kirdēr’s promotion is to be seen as an overt encroachment on the spiritual authority of the descendants of Ardašīr. In the inscription at Paikuli carved for Narseh in 283, the king of the kings invokes Ōhrmazd, “the lady Anāhīd,” and all the gods. On a rock carving at Naqš-e Rostam, Narseh is shown receiving investiture from the hands of Anāhīd, who wears a serrated crown and a sleeveless cloak (pl. XXXVIII; see Christensen, Iran Sass., pp. 252-53). Šāpūr II, according to a Christian hagiographic text, caused the heads of twelve Christian martyrs to be exposed, in conformity with ancient custom, in the temple at Eṣṭaḵr (see J. Labourt, Le Christianisme dans l’empire perse, Paris, 1904, p. 71, n. 2). It has been suggested that a passage in the Pahlavi Dēnkard (p. 413) refers to the construction of a temple of Anāhīd by Šāpūr II, but the text has ābān-ḵāna (house of the waters), and the interpretation is questionable. No further evidence on the Sasanian monarchy’s association with Anāhīd comes until the reign of Ḵosrow II Parvēz. In an investiture scene carved in high relief in the grotto of Ṭāq-e Bostān, Ḵosrow II receives crowns from Ōhrmazd and Anāhīd; the goddess wears a crown similar to Ōhrmazd’s and holds a pitcher of flowing water in her left hand (pl. XXXIX; Christensen, Iran Sass., p. 460; Vanden Berghe, Archéologie, p. 104). She also appears, holding a garland of flowers in one hand and a lotus in the other, on one of the capitals of the two columns that once stood in front of the grotto (pl. XL; Vanden Berghe, Archéologie, p. 105; Ghirshman, Iran: Parthes et Sassanides, fig. 376). The statement of Ṭabarī (Nöldeke, Geschichte der Perser, p. 397) that the last Sasanian king, Yazdgerd III, was crowned in the “temple of Ardašīr” at Eṣṭaḵr is interesting because of the similarity of this name to that of the “temple of Anāhīd-Ardašīr” at Eṣṭaḵr mentioned in the inscription of Kirdēr (see above).

Because the temples of Anāhīd were built beside springs or watercourses, it has been suggested that they were quite different from ordinary fire temples and had a distinct style of architecture characterized by columns and arcades (hypothesis of I. A. Orbeli, revived in C. Trever, “À propos des temples de la déesse Anahita en Iran Sassanide,” Iranica Antiqua 7, 1967, p. 122). But there is no reason why, at least in Iran, Anāhīd worship should not have been performed in fire temples. The designation of the temple at Eṣṭaḵr by the term ādur (fire), which was applied to all Zoroastrian places of worship, seems significant in this respect. Equally rash are suggestions that particular temples, for example the one at Taḵt-e Solaymān, belonged to Anāhīd (L. I. Ringbom, “Zur Ikonographie der Göttin Ardvī Sūrā Anāhitā,” Acta Academiæ Aboensis (Humaniora) 23, 1957, pp. 24ff.).

Aside from the rock carvings of Naqš-e Rostam and Ṭāq-e Bostān, few figures unquestionably representing the goddess are known (see below). She is thought to appear on an Achaemenid cylinder seal (Duchesne-Guillemin, “Art et religion sous les Sassanides,” Atti del Convegno Internazionale sul Tema: La Persia nel Medioevo, Rome, 1971, p. 378 and pl. III, fig. 3), on some reliefs from the Parthian period (idem, La religion de l’Iran ancien, Paris, 1962, p. 333), and on two ossuaries, one found near Bīšāpūr (Ghirshman, Parthes et Sassanides, p. 106 and fig. 120), the other Sogdian (ibid, p. 313 and fig. 255). Anāhīd may be represented by figures to be seen on Sasanian silver utensils, which depict a nude or scantily clad woman standing in front of an arcade holding a flower or some fruit or sometimes a bird or a child (see, e.g., Trever, “À propos,” pls. XXVII-XXIX); the identification seems convincing. It has been suggested that the colonnaded or serrated crowns on Sasanian coins belong to Anāhīd (R. Göbl, Sasanidische Numismatik, Brunswick, 1968, pp. 7, 9).

It is difficult to tell whether Anāhitā had any connection with Nanā or Nanai, a goddess of Mesopotamian origin frequently mentioned in Babylonian and Assyrian texts. In Armenia, Nana was assimilated to Athena and worshiped in a temple at the small town of Thil, but in some countries she was identified with Artemis. At Dura-Europos she was worshiped as Artemis Nanaia. It is known from Kushan coins with effigies of Nana that her cult spread as far as the Kushan territories. In the Sasanian period she is said to have been worshiped in Mesene, Susiana, Babylonia, and Arbayestan (see G. Hoffmann, Auszüge aus den syrischen Akten persischer Märtyrer, Leipzig, 1880, pp. 130ff.). Nana, who is one with Ishtar in ancient texts, seems to have been both a war-goddess and a nature-goddess. There are grounds to suppose that at an early stage the Iranian Anāhitā acquired some of Nana’s attributes, in particular her warlike character. It would be rash, however, to see Anāhīd in Nanai, “the great goddess of the whole earth,” who, according to the Acts of Mār Moʿayn (Hoffmann, Auszüge, p. 29), was one of the principal deities worshiped by Šāpūr II. Also dubious is Hoffmann’s statement (p. 155) that the Nana on the Kushan coins denotes Anāhitā.

Bibliography: Given in the text.

(M. L. Chaumont)

iv. Anāhitā in the Arts

Anāhitā’s representation and identification pose one of the most complex iconographic problems in the study of architecture and the visual arts of Iran. In literature she is mentioned by name, but her identification in art remains tentative since it rests primarily on her form, attributes, and activities. Moreover speculative discussions have attempted to connect her visual representation in the arts of Parthian, Sasanian, and early Islamic Iran to verbal descriptions in classical and Iranian historical and religious texts.

Greek and Roman historians mention sanctuaries and statues connected with the cult and worship of Anaïtis in Anatolia and the Levant. According to his inscriptions (A2Sa, A2Sd, A2Ha), Artaxerxes II (404-359 B.C.) invoked Anāhitā, along with Mithra, and he encouraged her worship through images, which he had distributed throughout his empire. Anāhitā also figures prominently in Zoroastrian literature and was quite clearly venerated at various times in connection with both water and fire (see Ābān Yašt). From historical sources we know that several Sasanian kings performed as high priests in her cult, the main temple for which was located at Eṣṭaḵr. But neither the images in art nor the architectural monuments correspond precisely to descriptions in literature, and none of the numerous (contested) attributions to her of images and sanctuaries rests upon firm ground. What does seem fairly certain is her absence from the figural repertory of Islamic art, except for the earliest materials, which may be considered continuations of Sasanian style and iconography. But even this material, much of it yet problematic in its classification, is not clearly representative of Anāhitā, nor necessarily related to the practice of her cult. (For a rare and unusual allusion to Anāhitā in Persian literature, see W. Hanaway, Jr., “Anahita and Alexander,” JAOS 102, 1982, pp. 285-95; for the transformation of imagery of Anāhitā to that of Šīrīn in Islamic legend, see P. P. Soucek, “Farhād and Tāq-i Bustān: The Growth of a Legend,” in Studies in Art and Literature of the Near East in Honor af Richard Ettinghausen, 1974, pp. 27-52.)

For the period prior to the establishment of the Sasanian dynasty, mention of sanctuaries of Anāhitā suggests a widespread practice of her cult not only in Iran but as far west as Armenia and Babylon. Although archeological exploration has been relatively extensive in these areas, no temples to the cult of Anāhitā have been located with any degree of certainty outside of Iran. Within Iran today, tentative identification of her temple has been suggested for architectural remains excavated at Bard-e Nešānda (R. Ghirshman, “Les Terrasses sacrées de Bard-è Néchandeh et de Masjid-i Solaiman,” MDAFI, Paris, 1976). From additional references it may be surmised that temples of Parthian times or earlier were also constructed at Hamadān, Susa, and Persepolis.

Increased archeological excavation of Sasanian sites since World War II has contributed to the definition of many architectural monuments in Iran of both religious and secular nature, but without significantly clarifying either the form or function of sanctuaries to the goddess Anāhitā. In spite of numerous textual references to her temples, not a single Sasanian building can be attributed to her cult with certainty. Even the main sanctuary at Eṣṭaḵr, mentioned in local sources and known to have been revered by the many kings of the Sasanian dynasty who served there as high priests, has not been located in spite of both survey and excavation of the site. But the chance find of an architectural block carved in relief with the fragmentary remains of a female figure (pl. XXXVII) has led L. Bier to reconstruct an otherwise unique monumental investiture by Anāhitā (“A Sculpted Building Block from Istakhr,” AMI, 1983, forthcoming). Relying upon stylistic details in relation to images of Anāhitā on coins, Bier dates the sculpture to the reign of Bahrām II, whose concern for Anāhitā is well known. From the Kaʿba-ye Zardošt inscription, Bier proposes that the architectural relief once served to ornament the sanctuary itself; such an interpretation corresponds well with the eclectic series of rock reliefs commissioned by Bahrām II elsewhere (see G. Herrmann, “The Sculptures of Bahram II,” JRAS, 1970, pp. 165-71).

A. A. Sarfarāz has argued for a reappellation of the fire temple at Bīšāpūr, excavated by R. Ghirshman just prior to World War II, as a water temple dedicated to Anāhitā; its topographic placement and unique drainage facilities would have allowed for and encouraged the circulation of water within its structure (“Anāhitā, the great temple at Bīšāpūr,” in Proceedings of the IIIrd Annual Symposium on Archaeological Research in Iran, 2nd-7th November, 1974, Tehran, 1975, pp. 91-110 [in Persian]). A critique of views concerning the monumental columnar architectural remains at Kangāvar is presented by M. Azarnush (“Excavations at Kangavar,” AMI 14, 1981, pp. 69-94). Previously considered a Temple of Anāhitā of Parthian date, based upon reference in Isidore of Charax to a Temple of Artemis there, the ruins at Kangāvar are presently viewed by Azarnush and others as a palace from the time of Ḵosrow II; confirmation for the later dating is found in epigraphic details of certain masons’ marks. For earlier discussion of columnar sanctuaries associated with water that may have been dedicated to Anāhitā, see C. Trever, “À propos des temples de la déesse Anahita en Iran sassanide,” Iranica Antiqua 7, 1967, pp. 121-32, and E. Keall, “Archaeology and the Fire Temple,” in Iranian Civilization and Culture, Montreal, 1972, pp. 15-22. See also M. L. Chaumont, “Le culte de la déesse Anāhitā (Anahit) dans la religion des monarques d’Iran et d’Armenie au Ier siècle de notre ère,” JA 253, 1965, pp. 167-81, and “Le culte de Anāhitā à Stakhr et les premiers Sassanides,” RHR 153, 1958, pp. 154-75. For further earlier discussion of the architectural form of Anāhitā’s sanctuaries, see L. I. Ringbom, “Zur Ikonographie der Göttin Ardvi Sura Anahita,” Acta Academiae Aboensis Humaniora 23, 1957, and Paradisus terrestris, Acta Societatis Scientiarum Fennicae, N.S., 1958, the assumptions and conclusions of which are critically questioned by O. Grabar in his review, Ars Orientalis 5, 1963, pp. 286-89, but pursued by Ringbom subsequently in “Three Sasanian Bronze Salvers with Paridaeza Motifs,” in Survey of Persian Art XIV, 1967, pp. 3029-41, where he continues to argue for seminal Sasanian influence in the development of Western forms, such as baptismal fonts, associated with both water and symbols of paradise. Like Trever, he sees an Iranian origin for Roman nymphaia, as derived from hypothetical architectural forms of the cult of Arədvī Sūrā Anāhitā, but there is no archeological support for his view.

It is in the arts of Sasanian Iran, and in particular the royal arts, that we find the strongest evidence for the representation of Anāhitā, but even here, confirmation is lacking. There seems to be general agreement that the female figure who appears on the reverse of coins struck by several Sasanian kings is Anāhitā (see especially R. Göbl, Sasanian Numismatics, Brunswick, 1971). Göbl has argued persuasively that Anāhitā’s presence is related specifically to her role in the investiture of these kings (“Investitur im sasanidischen Iran und ihre numismatische Bezeugung,” WZKM 56, 1960, pp. 36-51). For eagle symbolism and its use on royal crowns to signify Anāhitā, possibly connoting investiture, see Göbl, op. cit., and L. Trümpelmann, “Šāpūr mit der Adlerkopfkappe,” AMI, N.S. 4, 1971, pp. 173-85. There is also consensus concerning the image of Anāhitā in the monumental rock carving of the investiture of Narseh at Naqš-e Rostam, similarly in the investiture of Ḵosrow II (?) at Ṭāq-e Bostān, and in the repeated image of an investing goddess on rock-carved capitals (pls. XXXIII-XL), now at Ṭāq-e Bostān (H. Luschey, “Zur Datierung der sasanidischen Kapitelle aus Bisutun und des Monuments von Taq-i-Bostan,” AMI, N.S. 1, 1968, pp. 129-42). But there is considerable lack of agreement with regard to the identification of female personages in other Sasanian rock reliefs (Sar Mašhad; the relief of Ardašīr I at Naqš-e Raǰab; Barm-e Delak; Tanq-e Qandīl). Harper argues against the identification of Anāhitā in such instances, citing the royal garb of the females represented (P. O. Harper, Sasanian Silver Vessels. Part One: Royal Imagery, MMA, 1981, pp. 34f., 38). But a similar argument is used by others to recognize Anāhitā by analogy with the image of Ohrmazd dressed in royal garb (V. G. Loukonin, “Monnaie d’Ardachir I et l’art officiel sassanide,” Iranica Antiqua 8, 1968 pp. 106-17). L. Vanden Berghe has offered an additional criterium of distinction, based upon his discovery of a much damaged fragmentary relief beneath the monumental rock relief of Ardašīr I (or Šāpūr I) at Dārabgerd (“La découverte d’une sculpture rupestre à Darabgird,” Iranica Antiqua 13, 1978, pp. 135-48). He identifies the female as Anāhitā depicted in profile with long wavy strands of hair and a mural crown with stepped crenellations. Through an analysis of the historical development of mural crowns, Vanden Berghe suggests that such crowns were reserved for use by goddesses and were not worn by women of the royal family. Whereas Anāhitā in the relief at Naqš-e Rostam wears a similar crown with stepped crenellations, the women who appear in the reliefs at Sar Mašhad, Naqš-e Raǰab, Barm-e Delak and Tang-e Qandīl do not. As for a date, he is circumspect, citing the reigns of Ardašīr I, Narseh, Hormizd II, or Šāpūr II as most likely, in view of these kings’ particular interest in and support of her cult.

With Sasanian silver, the attribution of images to Anāhitā is even more problematic. Numerous vessels thought to have been produced in Sasanian Iran bear representations of naked or scantily clad women in a variety of poses suggestive of dance. They are associated with many different attributes, such as birds, children, animals, flowers, bunches of grapes, branches of vine, vessels, etc.; sometimes they appear within an architectural setting of arcades and columns. D. Shepherd regards these all as hypostases of Anāhitā (“The Iconography of Anahita: Part 1,” Berytus 28, 1980, pp. 47-86), but her assertions have not met with unanimous approval. J. Duchesne-Guillemin, for example, accepts the identification of Anāhitā in the reliefs at Naqš-e Rostam and Ṭāq-e Bostān, and on the coins of Bahrām II, but he advises more caution with the representations on the silver, which are less certain. He suggests that in these instances it may no longer be the goddess herself shown in multiples, but rather her priestesses (“Art et religion sous les sassanides,” in La Persia nel Medioevo, Rome, 1971, pp. 377-88). Ettinghausen also objected to the identification of Anāhitā on the silver, because of repeated images on individual vessels; in his view, these dancers may be considered part of a syncretism in the cults of Anāhitā and Dionysos (“A Persian Treasure,” Arts in Virginia 8, 1967-68, pp. 28-41; idem, From Byzantium to Sasanian Iran and the Islamic World, Leiden, 1972, pp. 3-10).

If any of the identifications referred to above are correct, several conclusions might be drawn. It would seem that artists and builders made no attempt to correlate visual imagery with verbal traditions. Either they were unaware of textual descriptions in religious books, myth and legend, or such descriptions were not canonized; perhaps such correlation was irrelevant to their practical needs and concerns. Scholars have generally attempted to explain the notable discrepancies between art and text by hypothetically reconstructing a popular cult, written evidence for which, it is argued, has simply not survived; but this is fraught with difficulties and the status of research on Anāhitā in the visual arts reflects the risks involved. The topic requires critical reevaluation and increased caution in reaching speculative conclusions with regard to Anāhitā’s form, presence, nature, and roles as understood from visual sources.


Given in the text.

See also G. Azarpay, “The allegory of dēn in Persian art,” Artibus Asiae 38/1, 1976, p. 41 with ref., especially n. 34.

(C. Bier)

Plate XLI. Silver dirham. Bahrām II (276-293 A.D.) American Numismatic Society N.Y., 1944-100.30175

(M. Boyce, M. L. Chaumont, C. Bier)

Originally Published: December 15, 1989

Last Updated: August 3, 2011

This article is available in print.
Vol. I, Fasc. 9, pp. 1003-1011