On the Iranian Origins of Mithras

Mithra was one of several eastern deities adopted by the Romans, as has long been widely accepted. No clearly Mithraic monument can be dated earlier than the late 1st century AD, and the earliest mention of the deity is by Cassius Dio in 63 AD, who in documenting the visit of Armenia’s Parthian kind Tiridates I, refers to him worshipping Mithras, though the context would suggest that the Persian Mitra is intended¹. After this, circa 80 AD, Statius made a vague reference, possibly to the tauroctony at the centre of the Mithraic mysteries – “Mithras twists the unruly horns beneath the rocks of a Persean cave”, or in another translation “Mithras, that beneath the rocky Persean cave strains at the reluctant-following horns.”². Again, the connection to Persian Mithra is seen. Investigations at Pompei, buried by the ashes of Vesuvius in 79 AD, found no images of the deity³.

Below is the earliest known Roman tauroctony, circa 98-99 A.D:


Plutarch claimed that the pirates of Cilicia, south-east Anatolia, who were defeated by Pompey, introduced the Mithraic rituals practised in Rome in his time⁴, and Servius claimed in the 4th century that Pompey settled these pirates in Calabria⁵. Plutarch (46–119 AD) commented that they practised “strange sacrifices of their own … and celebrated certain secret rites, amongst which those of Mithra continue to the present time, having been first instituted by them”⁶. The claim that the Persian cult and rituals of Mithra could have been introduced into the Roman empire via Anatolia is otherwise absent, however.

References to Mithra or Mitra (from Proto-Indo-Iranian mitrás, Miça (𐎷𐎰𐎼) in Old Persian) go as far back as the bronze age, with the Mitanni documents from Northern Syria and Iraq, circa 1450-1350 BC, mentioning Mitra along with other Vedic deities such as Varuna and Indra⁷, and the Indo-Aryan Rigveda, likely composed around the same time, perhaps closer to 1400 BC⁸. Zoroaster appears to have had his own reasons for excluding Mithra from the earliest Zoroastrian texts, but the Achemaenid elite appears to have seen the importance of Mithra as part of a central triad in the ancient Iranian pantheon when introducing Zoroastrianism in practice, placing him at the centre of the religion along with Ahura Mazda and Anahita.

Herodotus claimed that the Persians “have no images of the gods, no temples nor altars, and consider the use of them a sign of folly”⁹. Depictions of Mithra appear to be absent in the Achaemenid period, but his name appears in inscriptions from the reign of Artaxerxes II (404 – 358 BC) at Susa, asking that “Ahuramazda, Anahita, and Mithra protect me against all evil” and protect what he has built. Artaxerxes II was the first to name any deity other than Ahura Mazda in such a way, so whether he chose them as his patron deities or it reflects a widespread reverence of a central Persian triad is unknown. Artaxerxes III also invoked Mithra in an inscription at Persepolis, which reads “Ahuramazda and the God Mithra preserve me, my country, and what has been built by me”. According to the Zoroastrian texts Denkard and the Book of Arda Viraf, many texts were lost when Alexander’s army burned the library at Persepolis – these texts could well have contained references to Mithra and even information regarding the cult practices. This legend appears to be supported by Siculus’s Bibliotheca historica, which was completed circa 60 BC.

Detail from the Golden Bowl of Hasanlu, pre-Median, likely related to those referred to by Herodotus:


One might speculate that the solar deity depicted here is an older incarnation of Mithra, or a parallel solar deity at least, and it is notable that his chariot is drawn by a donkey, as the average gestation period of a donkey is 365 days. This suggest the deity’s purpose in representing rebirth in the yearly solar cycle.

Herodotus also referenced the ancient Persian sky gods¹°, who may well have included a primitive form or equivalent of Mithra as a solar deity, before describing later Persian customs under Zoroastrianism, stating that “at a later period they began the worship of… the Persian Mitra”¹¹. George Rawlinson in his translation notes that:

“The Persians, like their Vedic brethren, worshipped the sun under the name Mithra. This was a portion of the religion which they brought with them from the Indus, and was not adopted from any foreign nation. The name of Mithra does not indeed occur in the Achaemenian inscriptions until the time of Artaxerxes Mnemon, but there is no reason to question the antiquity of his worship in Persia, Xenophon is right in making it a part of the religion of Cyrus”¹²

That Xenophon noted its significance in Cyrus’ religion is significant, as is the clear status of Mithra as a solar deity. Herodotus also describes sacrificial rites relating to Zoroastrianism that involved wearing a myrtle wreath and cooking the meat with clover¹³, but nothing is indicative of any sacrificial rite, or the sacrifice of any animal in particular, in relation to Mithra in his account.

James Frazer recognised Mithra as an “Old Persian Deity” in the Golden Bough¹⁴ , and the scholarship on Mithraism begins with Franz Cumont. Cumont’s theory was that Roman Mithraism was simply “the Roman form of Mazdaism”, or in other words Zoroastrianism, filtered through from the East. This is in any case an oversimplification, as Mithraism was most likely a merging of the original Iranic deity with native European – particularly Greek and Anatolian, and perhaps Roman – symbols and iconography instead. Cumont largely focused on the notion that the Mithraic mysteries were themselves lifted from Iranian tradition, but with lack of sources on either the similar iconography or on anything properly resembling what we can ascertain about Mithraic mystery rites, we have no way of verifying this. Cumont’s theories were widely discredited in the 1970s. Much of the debunking appears to have been mainly concerned with the questionable claims that the Mithraic rituals themselves in Rome were clearly Iranian in origin, rather than addressing the issue of whether or not the deity and his name were lifted from Persian tradition.

Mary Boyce argues that “no satisfactory evidence has yet been adduced to show that, before Zoroaster, the concept of a supreme god existed among the Iranians, or that among them Mithra – or any other divinity – ever enjoyed a separate cult of his or her own outside either their ancient or their Zoroastrian pantheons”¹⁵. Here again, we see how much focus of the debunking of Cumont’s theories relate to his on his claims that the original Persian deity resembles the Roman one in terms of a separate cult and the “mysteries”, and a claim that he was worshipped as a separate deity rather than as part of an Iranian pantheon.

John R. Hinnells, another leading critic, was dismissive of these theories overall, but still acknowledged likely Iranian origins, stating that “Nevertheless we would not be justified in swinging to the opposite extreme from Cumont and Campbell and denying all connection between Mithraism and Iran”¹⁶, arguing further that studies of the Mithraic mysteries should be limited to the Roman sources, rather than based on speculation about their origins in Iranian customs¹.

Roger Beck has since suggested that recent studies relating to the state of Zoroastrianism during the BC period do now add more credibility to Cumont’s claims of a spread from east to west, arguing that “Apart from the name of the god himself, in other words, Mithraism seems to have developed largely in and is, therefore, best understood from the context of Roman culture”¹⁸. Beck argues that a “single founder” first introduced Mithraism in the Roman empire by incorporating knowledge of both Greek and Eastern religion, suggesting that some ideas may have passed through the Hellenic kingdoms, and argues that a Mithras identified with the Greek solar deity Helios was one of several Greco-Iranian deities from the cult founded in the middle of the first century BC by Antiochus I of the state of Commagene¹⁹, who was of half Sogdian Iranian descent. A similar theory has been proposed by Reinhold Merkelbach, who also argues that the mysteries were created by one person and in a specific place, in this case Rome, and that this person came from an eastern province and had extensive knowledge of the Iranian myths, which he wove into these myths along with aspects of Hellenic Platonism²°. M.J. Vermaseren, in relation to inscriptions of Mithradates Kallinikos and his son Antiochus of Commagene, stressed the predominately Hellenic nature of this early version of what would become Roman Mithraism²¹.

Below is a relief featuring Mithras-Helios, with solar rays in Iranian dress, with Antiochus I, 1st century BC:


It is therefore, unclear how much of the content of the mysteries were taken from Iranian rituals entirely, from a mixture of Hellenic and Iranian rituals or entirely from Greek rituals, but given the lack of evidence of surviving early Zoroastrian texts and of well-documented ancient Iranian rituals, such origins are possible, if unproven. Cumont focused in the presence of the bull-slaying iconography in Iranian myth, but it must be noted that the myth in question involves Ahriman, not Mithras, slaying the primordial bovine creature Gavaevodata²². Equivalent myths relating to Mithras slaying such a creature have not been found. Ehsan Yaghmaei observed links between Mehr or Mithra and iconography involving birds and bulls, arguing that the crow kills the cow through the sun, and that Mithra takes the bull into the cave and kills it there²³. This is very reminiscent of Statius’ reference to Mithras “twisting the unruly horns beneath the rocks of a Persian cave”.

Parthian relief of a bird on the back of a bull, Zahhak Castle, East Azerbaijan, Iran:


A similar design on a Parthian belt bucket:


This bas relief from Taq-e Bostan, Iran featuring Mithra along with Ahura Mazda and Ardashir II from the 3rd century AD highlights his status as an Iranian solar deity:

Taq-e Bostan

On this Sassanian relief the status of Mithra as a solar deity is clear, but despite there being some apparent Hellenic influence on Persian art in this period, there is nothing linking this image to the Phrygian camp and bull-slaying image seen in the Roman sculptures of Mithras. This would appear to indicate roots in the older native Iranian tradition and iconography, with the sword perhaps relating to the “oath” associated with his name, as seen in the translation of the name Mithra and clearly stated in the Avestas. One must note how scarce any depictions of deities under Zoroastrianism are, which apart from the apparent Zoroastrian tendency to refrain from depicting deities, likely has much to do with the iconoclasm that occurred during the Sassanian era, which saw the conversion of shrines into fire temples, due to the monotheistic Zoroastrian religion no longer being a religion of the elite as it was when introduced by the Achaemenid dynasty.

There are therefore several questions to address in determining the origin of Mithras, and Mithraism: whether Mithras as a deity and as a name was adopted from Persian tradition by the Romans, whether the symbolism and iconography were likewise adopted from Persian tradition and whether the mysteries and cult practices of Roman Mithraism had any roots in Persian customs. That the name and deity of Mithras is taken from Iranian tradition is clear, as even those critical of Cumont’s pioneering study have agreed, and reference to the deity vastly predating any Roman depiction or documentation of Mithraism. As revisionist studies in the 1970s widely agreed, there is little if anything to prove that the rites of the mysteries of Mithraism were lifted from Iranian practices, though they hypothetically could have been, and as noted it is perfectly feasible, perhaps likely, that the introduction of Mithraism into the Roman empire via Greece with Mithras-Helios involves some level of syncretism of Hellenic and Iranian customs.

Regarding the iconography, again, though it is possible that the iconography of the Roman tauroctony, like the rites of the mysteries, were largely Roman in nature, there are nonetheless striking parallels between the three forms – Persian Mithra, the Greco-Iranian Mithras-Helios and the Roman-Mithras – in their clear depiction as solar deities, with the Sassanian relief being clearly Persian rather than Roman-derived in style and iconography, with the bull-slaying motif being absent. As noted, there is also strong argument for the same tauroctony being depicted earlier in more animist fashion in Iran with the bird slaying the bull, especially as the bird was commonly used as a solar symbol throughout ancient Iranian art, both during and preceding the Zoroastrian period. It is perhaps plausible that this animistic motif influenced the later Roman fashion to some extent. Regardless, what is beyond doubt, though the nature of the Mithraic mysteries should perhaps be addressed within a Roman context and origin, the origin of the deity himself, his purpose as a solar and rebirth deity, and it would appear at least some aspects of the iconography and symbolism as well, originate in Iranian tradition.


¹ – Dio, Cassius, Epitome of Book 63, 5:2

² – Statius, Thebaid, Book i. 719,720

³ – Vermaseren, M.J., Mithras: the Secret God, p. 29

⁴ – Plutarch, Pompey, 24, 7

⁵ – Servilius, Georgics, 4, 127

⁶ – Daniels, C.M., “The role of the Roman army in the spread and practice of Mithraism” in John R. Hinnells (ed) Mithraic Studies: proceedings of the first International congress of Mithraic Studies, Manchester University Press (1975), vol. 2, p. 250

⁷ – Witzel, M., Early Sanskritization – Origin and development of the Kuru state, November 2011 at the Wayback Machine

⁸ – Kochar, Rajesh, The Vedic People: Their History and Geography, 2000, Orient Longman

⁹ – Herodotus, Rawlinson, George (transl.), The Histories, Book I, Chapter 131Everyman’s Library, p. 75

¹° – Herodotus, Rawlinson, George (transl.), The Histories, Book I, Chapter 131, Everyman’s Library, p. 75

¹¹ – Herodotus, Rawlinson, George (transl.), The Histories, Book I, Chapter 131, Everyman’s Library, p. 75

¹² – Rawlinson G., in Herodotus, Rawlinson, George (transl.), The Histories, Everyman’s Library, p. 75

¹³ – Rawlinson G., in Herodotus, Rawlinson, George (transl.), The Histories, Everyman’s Library, pp. 75-76

¹⁴ – Frazer, James George, The Golden Bough, Oxford University Press, p. 360

¹⁵ – Boyce, Mary , “Mithra the King and Varuna the Master”, in: Festschrift für Helmut Humbach zum 80, Trier, 2001, pp. 239-257; p.243, n.18

¹⁶ – Hinnells, John R. , “Reflections on the bull-slaying scene” in Mithraic Studies, vol. 2, p. 303-4

¹⁷ – Hinnells, John R. , “Reflections on the bull-slaying scene” in Mithraic Studies, vol. 2, p. 303-4

¹⁸ – Martin, Luther H. , foreword in: Roger Beck, “Beck on Mithraism: Collected Works With New Essays”, Ashgate, 2004, p.xiv.

¹⁹ – Beck, Roger , “Mithraism”, Encyclopaedia Iranica, 2002, Costa Mesa:Mazda Pub.

²° – Merkelbach, Reinhold , Mithras, Konigstein, 1984, ch. 75-7

²¹ – Vermaseren, M.J., “Nuove indagini nell’area della basilica di S. Prisca in Roma”, in Mededelingen van het Nederlands Instituut te Rome. Antiquity, n.s., 37, 2 (1975), pp. 87-96, p.93: (transl.)

²² – Cumont, Franz, McCormack, Thomas J.  (transl.), The Mysteries of Mithra. (2nd ed.). 1903. Chicago: Open Court; London: Kegan Paul, Trench, Trübner, fasc. repr. New York: Dover, 1956, p. 135

²³ – Yaghmaei, Ehsan,  A Study of Mehr and Mitra in Ancient Iranian Sources and “Masalak wa Mamalak”, Iran Book News Agency, February 22, 2011