Mithra

The recognition that miθra is the common noun "contract" was won in the nineteenth century and codified by Christian Bartholomae in his Altiranisches Wörterbuch of 1904; but it was Antoine Meillet, in a short paper of 1907, who drew the decisive conclusion. As Schmidt summarises Meillet's case, the abstract meaning of the noun "largely agrees with the character and functions of the god", so that Mithra is "the personification and deification of the concept 'contract'".^[1][2] This reversed an older habit, Bartholomae's own included, of taking Mithra first as a sun-god or a god of light and deriving the contract-sense from that. The covenant is primary; the solar association is a later accretion. One technical point sits under the Persian evidence: the regular Old Persian development of the name would have been Miça, and the Mitra/Miθra found in the royal inscriptions is not the native Persian form but one carried in from the wider Iranian tradition.^[2]

The noun is broad, and just how broad is a real question. Albert de Jong states the plain reading: the name "means 'contract' and guarding or overseeing contracts is one of his oldest functions", a function "especially prominent in Yt. 10", the hymn to Mithra.^[3] Mary Boyce puts it the same way, that the common noun mithra demonstrably "means something like 'pact, contract, covenant'".^[4] The range is caught in two Avestan passages the philologists return to. In the Vidēvdād (4.2) there is a scale of miθras graded by how they are struck, a contract concluded by word, one by handshake, and then contracts reckoned in the value of a sheep, an ox, a slave, or a land, each with its penalty for breach.^[5] And the Mihr Yašt itself (10.116-17) grades the sanctity of contracts by the bond they seal: twenty-fold between friends, thirty-fold between fellow-citizens, fifty-fold between husband and wife, seventy-fold between pupil and teacher, a hundred-fold between father and son, a thousand-fold between two countries, and ten-thousand-fold the miθra of the Religion itself, whose breach would be apostasy.

Because that list folds in the natural ties of blood, the sense here widens past "contract" toward "alliance", and it is alliance, the moral tie a society at peace depends on, that predominates in the hymn. Some scholars press the point further. Paul Thieme argued that the Indo-Iranian mitra is not "contract" in the legal sense at all but "alliance", or more broadly the reconciling bond of friendship; Jan Gonda reached for "friendship" and something close to "loyalty" or "piety".^[6] The disagreement is narrow but genuine, and it is not idle: the whole ethical weight of the god turns on whether his sphere is the struck bargain, the wider alliance, or the disposition of good faith itself. The mainstream, from Meillet through Boyce, Gershevitch and Schmidt, keeps "contract/covenant" as the core and reads the broader senses as the outer reach of the one idea; but the exact width of miθra stays a live philological debate, and the hymn itself deploys the word across that whole span.

The god's own hymn is the tenth of the Yašts, the Mihr Yašt, the fullest and one of the finest of them, edited and translated in Ilya Gershevitch's classic study The Avestan Hymn to Mithra (1959).^[7] It opens with a sanction that measures his rank: the hymn starts, in Schmidt's summary, with the statement of Ahura Mazdā that he created Mithra and "made him as worthy of worship and prayer as himself" (Yt. 10.1).^[8] What follows is a portrait of a god of enormous reach and vigilance. He is a warrior who drives a chariot across the sky, his chief weapon the mace; he is attended by a retinue of lesser divine beings, Sraoša (Obedience), Rašnu (the Judge), the fravashis, and above all Vərəθraγna (Victory), who as a boar does the bloody work against the enemy. He brings the rain and makes the plants grow for the land whose ruler keeps faith (Yt. 10.61), and withholds them where the covenant is broken.

His most frequent epithet is vouru.gaoiiaoiti, in Schmidt's rendering "characterized by wide cattle-pastures", he of the wide pastures, and it is not idle scenery: cattle can graze the open range in safety only where neighbours keep the peace, that is, where the miθra between peoples holds, so the epithet names precisely the condition the covenant-god secures. He is "the lord of the country", indeed "the lord of the country of all countries" (Yt. 10.145), the drawer of the borderlines that keep neighbours from strife (Yt. 10.61), master of both peace and war. The welfare of a whole land is made to hang on the good faith of its ruler, which is why the hymn dwells so on the ruin that follows when a lord, a clansman, or a king breaks his word.

Because Mithra oversees every covenant, he must see everything, and the Mihr Yašt makes him the very type of the all-seeing god. He is "ever wakeful", sleepless (Yt. 10.7); he has "a thousand ears and ten thousand eyes", with watchers posted "on every height and every outlook" (Yt. 10.45); he watches even in the dark (Yt. 10.141); with his long arms he reaches out and seizes the liar whether the man is in the east or the west (Yt. 10.104); and he catches the very person who imagines that the god does not see his evil and deceitful deeds (Yt. 10.105). Against this watchfulness the oath-breaker has no cover.

The object of Mithra's wrath is the miθrō.druj, the deceiver of the contract, the man who lies to Mithra by breaking his sworn word, the covenant-Lie that is a special case of the general Lie. The Avestan idiom is exact: to break a contract is miθrəm druj, "to deceive a contract", or miθrəm jan, "to smash a contract", and it is a sin that does not stay with the sinner. A single treaty-breaker, the hymn says, destroys a whole country, killing the truthful as surely as would a hundred sorcerers (Yt. 10.2); Mithra robs the covenant-breakers of the vigour of their arms, the strength of their feet, the light of their eyes, the hearing of their ears (Yt. 10.23), and the arrows and spears and maces of those who enrage him are turned aside and made to miss (Yt. 10.39-40).^[8] The dwellings of the deceivers are destroyed (Yt. 10.38). It is one of the sharpest formulations in any religion of the idea that to break faith is to invite a cosmic, and a communal, catastrophe.

Mithra is universally associated with the sun and the light of day, and in the popular mind the two ran together, so that in the late Pahlavi texts the sun could simply be called "God Mihr". But the association is not, in the oldest layer, an identity, and the distinction is one to hold onto. In the Avesta Mithra is not the sun: the Mihr Niyāyišn and the Xoršēd Niyāyišn, the litanies of Mithra and of the sun, are recited together in daily prayer but remain two things, and the hymn is careful to place Mithra as the first of the divine beings to rise over the great mountain-range Harā in front of the swift-horsed, immortal sun (Yt. 10.13), his herald and forerunner, not the disc itself. Boyce and Gershevitch both insisted on this against Herman Lommel, who had argued the two were virtually one; the identification of Mithra with the sun, both held, is a secondary and later development, in Iran as in India.^[4]

When the identification does surface unmistakably, it is in a Greek source. The geographer Strabo, writing at the turn of the era, reports the Persian worship and equates their sun straightforwardly with the god:

Now the Persians do not erect statues or altars, but offer sacrifice on a high place, regarding the heavens as Zeus; and they also worship Helius, whom they call Mithras, and Selene and Aphrodite, and fire and earth and winds and water.

Strabo, Geography 15.3.13, trans. Jones^[9]

Strabo's "Helius, whom they call Mithras" is the first explicit statement that the Persians called the sun by the god's name. Yet even this was not a fixed dogma of the Achaemenid centuries: Curtius Rufus has Darius III, on the eve of battle, invoke "the Sun, and Mithra, and the sacred and eternal fire" as three distinct things, and Xenophon's report of the Persians sacrificing bulls to Zeus and horses to the Sun (Cyropaedia 8.3.11-12) leaves the Sun unnamed.^[10] Schmidt's conclusion is the safest ground: "in Achaemenid times there was no consistent identification of Miθra with the sun".^[2]

How early the two did fuse is itself disputed. Gershevitch put the conflation late, after the Avesta, perhaps only in the Parthian period; Gh. Gnoli argued the reverse, that Mithra had a solar character already in the early Achaemenid period, and pointed to a possible influence of the Babylonian sun-god Šamaš, whose role as judge would suit Mithra well.^[6] De Jong takes the mediating line the evidence best supports: whatever the date of the full identification, "some connection between Mithra and fire, light and the sun already existed in the Achaemenian period", so that the Greeks who knew Mithra chiefly as a solar god were catching a real thread of him, even as they simplified it into an equation the Iranian material never quite makes.^[3]

If Mithra oversees the covenant, he presides above all over the oath, the covenant sworn and put to proof, and here the Iranian material and the Greek notices agree. In the Avesta the oath and the ordeal (varah) take place before the fire and in the presence of Rašnu the Judge, Mithra's companion (Yt. 12.3); a suspect could be made to drink the "golden oath-water", which by causing jaundice was believed to reveal the man who had knowingly perjured himself by "contradicting Rašnu, the judge, and deceiving Miθra, the contract or promise" (Vidēvdād 4.54-55).^[5] The god of the covenant, the fire before which it is sworn, and the ordeal that tests it belong together.

The Greek witnesses catch the same practice from the outside: that the Persians, and above all their kings, swore by Mithra. Xenophon has Cyrus the Great swear by Mithra (Cyropaedia 7.5.53) and the younger Cyrus do the same (Oeconomicus 4.24); Plutarch reports the oath "by Mithra" in the mouth of Artaxerxes II. De Jong gathers the same notices and enters a sober caveat: in a Greek or Latin author the name of Mithra is often merely the oriental substitute for a Greek divinity in an exclamation, so that not every such oath is firm evidence for the Iranian cult, though the pattern as a whole is.^[3] It is that royal oath-formula, the covenant-god called to witness a king's word, that surfaces in a saying preserved by Aelian, where Artaxerxes, shown a great pomegranate a gardener had grafted by his own patient labour, exclaims:

By Mithra, this man by like care and diligence might also, in my opinion, make a little city great.

Aelian, Varia Historia 1.33, trans. Thomas Stanley, 1665^[11]

The oath is casual in the telling, but it is the god of the sworn word who is called to witness even a king's offhand judgement, exactly the everyday hold on Persian life that Mithra's cult implies.

Mithra belongs, by his whole nature, to the Achaemenid spine of Truth (Avestan aša) set against the Lie (drauga). To keep a covenant is to stand in the Truth; to break it, to become a miθrō.druj, is to fall to the Lie in its most concrete and social form, the betrayal of the given word. The royal ideology of Darius, carved at Behistun, turns on that axis, the king a friend to the right and no friend to the follower of the Lie, and the Persian horror of falsehood the god enforces is what the outsider Herodotus noticed as the moral centre of the whole culture:

The most disgraceful thing in the world, they think, is to tell a lie; the next worst, to owe a debt: because, among other reasons, the debtor is obliged to tell lies.

Herodotus 1.138, trans. Rawlinson^[12]

Herodotus does not name Mithra here, and the sentiment is broader than any one god; but the dread of the lie he records, and the training of Persian sons "in riding, in shooting, and in speaking the truth" (1.136), is the human face of the covenant-ethic that Mithra guards. In the Persian scheme the debtor is shameful because debt drives a man to break his word, and it is the word, the miθra, that the god watches over. The whole Persian culture of good faith, the oath before the fire, the bond that holds or is forsworn, sits at the meeting of Mithra and arta.

Here lies the fact about Achaemenid Mithra most easily got wrong: for all his antiquity and his hold on Persian life, he is absent from the royal inscriptions of the early kings. Darius and Xerxes invoke Ahura Mazdā on inscription after inscription, and give no other god a name, acknowledging only the anonymous "other gods who are" (the bagas) ranged behind him. Mithra is not among them. He is missing, too, from the level below the crown. The Persepolis Fortification archive is no help to him either: that administrative record logs sacrifices to a spread of Iranian and Elamite gods, yet Mithra is not among its named recipients, and "is conspicuously absent from the PFA texts edited thus far".^[13] A word once pressed into service as his, the tablet-form Mišebaka, was explained by Gershevitch as *Miça-baga, "the god Mithra"; the spellings exclude that reading on formal grounds, and Hinz's *Visaibagā, "All the Gods", has prevailed, a collective rite set beside Auramazdā, with Mithra nowhere invoked by name.^[13][14] The formally excluded Mithra reading is still occasionally repeated, and Gershevitch himself later offered the no more successful *Miçahya. So the individual Mithra does not appear even in the daily offering lists of the heartland.

Schmidt reads the royal silence as a matter of register rather than of belief: Darius's devotion stood close to that of the Gāthās of Zarathustra, which addressed the great Ahura and the powers about him but named none of the old gods, so a king in that manner would not carve Mithra's name however much his subjects prayed to him.^[2] On that reading the absence measures the inscriptions' idiom and says nothing about the reach of the cult; but it is a fixed point every reconstruction has to start from.

The silence also feeds the larger question of how far the Achaemenid kings were "Zoroastrian". The evidence pulls two ways at once. The names Persians gave their sons, the oaths they swore, and the everyday ethic Herodotus describes all show Mithra deep in Persian life from the start; the royal texts and the administrative tablets, the two bodies of contemporary Achaemenid evidence, do not name him at all before Artaxerxes II. A god can be at once central to a people and invisible in its official record, and Mithra is the clearest case of it: the gap is not between an early cult and a late one but between what was worshipped and what the crown chose to inscribe, and reading the one off the other is the standing trap of the subject.

The change comes with Artaxerxes II (r. 405/4-359/8 BCE), a century and more after Darius, and it is abrupt. In his inscriptions Mithra and the water-goddess Anāhitā appear beside Ahura Mazdā, the anonymous "other gods" replaced by a named triad, the king praying that Ahuramazdā, Anāhitā and Mithra protect him and his works.^[15] The two gods rose together, and the pairing is not random: they are the two great divinities of western Iran, and later Greek report treats them as consorts. But the innovation reached each of them differently. For the goddess it had a visible instrument, the image-cult that Berossus credits to this king (see Anāhitā); for Mithra it had none, and his elevation has to be read from the kind of god he was. He is the god of the sworn word, and the Persians, above all their kings, swore by him; he is a god of victory, with Vərəθraγna in his train; his autumn feast was a royal occasion. A king who wanted a named divine guarantor of his own good faith and his own victories had, in Mithra, exactly the god for it.

Whether the triad marks a real change in belief or only a change in what the kings chose to record is disputed, and the disagreement is old. One view makes it a genuine turn: Artaxerxes II set aside the reserve of his forefathers and lifted two gods long worshipped by his subjects into the official cult. The other makes it a change of wording, not of worship: the crown simply began naming the divine company it had always acknowledged, adopting the manner of the Yašts, which invoke each god by name, in place of the older reticence that left Ahura Mazdā's companions anonymous.^[2] The current reference synthesis declines to choose: "There is little scholarly consensus as to how these measures should be interpreted."^[13] Nor do the texts decide it. Ahura Mazdā keeps pride of place in all of them; the wording differs from inscription to inscription; and the one text that survives from Artaxerxes III at Persepolis (A³Pa) retains Mithra but drops Anāhitā, adding beside Ahuramazdā the disputed phrase read "Mithra Baga", in Kent's rendering "the god Mithras".^[15] Boyce took the second word instead for Apąm Napāt, the Child of the Waters, a reading Schmidt reports as controversial.^[2] Why the goddess should vanish from the record so soon after her debut has never been explained. What is clear is that the epigraphic Mithra of the Persian kings is a late Mithra, and any account of Achaemenid religion must reckon with his long silence in the royal record before it.

If the crown was slow to carve his name, ordinary Persian life was saturated with it. The clearest measure is the flood of personal names compounded with the god's, of which Mithradates (Old Iranian Miθradāta, "given by Mithra") is only the most famous, borne by satraps, generals, and later by whole dynasties of kings in Pontus and Parthia. The names reach into the Persepolis archives themselves, where *Miça/Miθra stands among the most popular elements in theophoric names, in the very corpus that never names the god as a recipient of offerings; and the divine symbols in the archive's own seal-imagery likewise "argue against the idea of a religious reform, taboo, or simple lack of popularity" as the cause of such absences.^[13] Henkelman still enters a caution worth keeping: "onomastic evidence is a treacherous source" for religious history, the element Miça/Miθra in a given name sometimes rather meaning "treaty", and the trend of Mithra names may reflect "the attitudes of a preceding generation", the fathers who chose the names rather than the sons who bore them.^[13] The other measure is the great autumn festival of Mithra, the Mithrakāna (Old Persian), later Middle Persian Mihragān, modern Mehregān, one of the two hinges of the Iranian year with the spring Nowruz. Later Greek writers tie royal and tribute occasions to it: Duris (through Athenaeus) preserves the report that the King drank hard and danced the Persian dance only at the feast of Mithra, and Strabo records the satrap of Armenia rendering up twenty thousand foals yearly "for the Mithrakāna" (Geography 11.14.9).

The festival's Achaemenid form has to be reconstructed from these later notices, and the reconstruction needs the caution Henkelman presses: Strabo does not make the Achaemenid date explicit, so it can "only be assumed on the basis of the Old Iranian form of the name", and a sixth-month sacrifice in the Fortification archive that might connect to it cannot be shown to.^[13] What is not in doubt is the broad shape: that Mithra was honoured in a great seasonal feast, and in the names Persians gave their sons, so that his cult is plainly visible below the crown even where the royal inscriptions stay silent.

One classical passage on Mithra needs careful handling, because it is wrong, and the way it is wrong is instructive. In his account of Persian worship, Herodotus reports that the Persians learned to sacrifice to a heavenly Aphrodite (Aphrodite Urania), a cult he thought borrowed from the Assyrians and Arabs, and he adds that the Persians call this goddess "Mitra" (1.131). Taken at face value the sentence makes Mithra a goddess of love, and it has sometimes been repeated in just that loose form. It is doubly mistaken. The male covenant-god has been given a female cult that is not his, and, one sentence earlier, the god himself has been left out of the list of Persian divinities where he belonged.

The celestial goddess the Greeks matched to Aphrodite Urania is Anāhitā, and it is her worship that Herodotus has labelled with Mithra's name. Working at second hand, and fitting the strange onto the familiar, he has crossed the empire's two most popular western divinities. That is de Jong's verdict from the side of the classical texts: he calls 1.131 "the earliest (but doubtful) occurrence of Mithra's name", one in which the name is "wrongly given as that of the Celestial Queen worshipped by the Persians".^[3] Robert Rollinger's survey of Herodotus's Persian ethnography is more reserved, recording only that the female Mitra "has caused debate" about how well Herodotus knew Persian custom, and referring the question of a Mitra/Anāhitā confusion onward to the specialist literature.^[16] The point to carry away is not that Mithra was a goddess, which is simply false, but that a careful outsider could garble the Persian pantheon badly enough to weld the covenant-god's name onto the water-goddess's cult, a caution about how Persian religion reached the Greeks, and how it reaches us through them.

The most persistent error about Iranian Mithra is to read him through the Roman Mithras, the god of the mystery cult that spread among soldiers and merchants across the Roman Empire in the first four centuries CE, with its underground sanctuaries and its central image of the god slaying a bull. The grand statement of a straight line between the two was Franz Cumont's. Around the turn of the twentieth century Cumont, in his Textes et monuments figurés relatifs aux mystères de Mithra and, later, with Joseph Bidez, in Les Mages hellénisés, reconstructed the Roman cult as Iranian Mazdaism carried west and reworked by the hellenised Magi, so that the Persian god's theology was thought to pass more or less whole into the Roman sanctuaries.

That reconstruction has collapsed. De Jong records that the basic idea of Les Mages hellénisés has been "shown ... to be untenable", doubted already by A. D. Nock and A. Momigliano and finally undone by the demonstration that there is no coherent "Magusaean" system behind the pseudo-Zoroastrian texts, only a scatter of works written for different purposes by different hands.^[17] The Roman iconography, and above all the bull-slaying (the tauroctony), has no attested Iranian original. Lommel tried to ground it in an Indo-Iranian myth in which the gods, Mithra reluctant among them, kill Soma in the form of a bull, and several scholars of the Roman mysteries welcomed the idea; but Gershevitch rejected it, and no bull-slaying Mithra has been found in the Iranian sources.^[18] The scholars who rebuilt Mithraic studies after Cumont, R. L. Gordon and Roger Beck among them, treat the Roman cult as a largely Western creation that took the old god's name and a deliberately "eastern", Persian colouring rather than his religion.^[17] De Jong, whose subject is exactly the Greek and Latin evidence, draws the line plainly: however one judges the relation, the mystery cult described by the classical authors "was not a religion practised by Iranians", nor one in which Iranians would have recognised much of their own heritage.^[3] The covenant-god of the Mihr Yašt and the Persian oath is not to be fitted out with the bull, the cave, and the mysteries of his Roman namesake.

Mithra outlived the empire that was slow to name him. In Plutarch's philosophical account of Persian religion he appears as Mithres, set as the mesitēs, the "mediator", between Horomazes (Ahura Mazdā) and Areimanios (Angra Mainyu), the good and the hostile principles, a role the Avestan Mithra does not clearly hold but which the later Iranian and Greek tradition gave him (De Iside et Osiride 46).^[10] In Sasanian and later Zoroastrianism he becomes Mihr, the yazata of the covenant still, judge of souls at the Činvat bridge, patron of the seventh month and the sixteenth day (both named for him), his festival kept as Mihragān; and his name survives in living Persian, where mehr means both "the sun" and, tenderly, "love, friendship", the last a softening of the old severe god of the sworn and binding word. Through all of it the core holds: Mithra is the covenant made divine, the god before whom oaths are sworn and by whom they are kept, the watcher of a thousand eyes who does not sleep and cannot be deceived, and the terror of every man who would break his given word.