Sacred Texts  Bible  Bible Critical Views  Index  Previous  Next 


Pagan Christs, by John M. Robertson, [1911], at sacred-texts.com


§ 9. Mithraism and Christianity.

Of course, we are told that the Mithraic rites and mysteries were borrowed and imitated from Christianity. 8 English scholars of good standing are still found to say that the Mithraic and other mysteries "furnish a strange and hardly accidental parody of the most sacred mysteries of Christianity." 9 The refutation of this notion, as has been pointed out by M. Havet, 10 lies in the language of those Christian fathers who spoke of Mithraism. Three of them, as we have seen, speak of the Mithraic resemblances to Christian

p. 316

rites as being the work of devils. Now, if the Mithraists had simply imitated the historic Christians, the obvious course for the latter would be simply to say so. But Justin Martyr expressly argues that the demons anticipated the Christian mysteries and prepared parodies of them beforehand. "When I hear," he says, 1 "that Perseus was begotten of a virgin, I understand that the deceiving serpent counterfeited also this." Nobody now pretends that the Perseus myth, or the Pagan virgin myth in general, is later than Christianity. Justin Martyr, indeed, is perhaps the most foolish of the Christian fathers; but what he says about the anticipatory action of the demon or demons plainly underlies the argumentation also of Tertullian and Julius Firmicus. 2

When, again, Justin asserts 3 that the Mithraists in their initiation imitate not only Daniel's utterance "that a stone without hands was cut out of a great mountain," but "the whole of [Isaiah's] words" (Isa. xxxiii, 13-19), he merely helps us to realise how much older than Christianity is that particular element of Christian symbolism which connects alike Jesus and Peter with the mystic Rock. That Mazdeism or Mithraism borrowed this symbol from Judaism, where it is either an excrescence or a totemistic survival, 4 is as unlikely as it is likely that the Hebrews borrowed it from Babylonia or Persia. 5 In Polynesian mythology, where (as also in the rites of human sacrifice) there are so many close coincidences with Asiatic ideas, it was told that the God Taaroa "embraced a rock, the imagined foundation of all things, which afterwards brought forth the earth and sea." 6 Here again we are in touch with the Græcised but probably Semitic myth of the rock-born Agdestis, son of Jupiter. 7 Even the remarkable parallel between the myth of Moses striking the rock for water and a scene on one of the Mithraic monuments suggests rather a common source for both myths than a Persian borrowing from the Bible. In the monument,

p. 317

[paragraph continues] Mithra shoots an arrow at a rock, and water gushes forth where the arrow strikes. As the story of the babe Moses is found long before in that of Sargon, 1 so probably does the rock-story come from Central Asia. 2

The passage in Isaiah, which strongly suggests the Mithraic initiation, seems to have been tampered with by the Jewish scribes; and corruption is similarly suspected in the passage Gen. xlix, 24, where "the Shepherd, the Stone of Israel," points to some credence latterly thrust out of Judaism. Above all, the so-called Song of Moses 3 (in which both Israel and his enemies figure as putting their faith in a divine "Rock," and the hostile "Rock" is associated with a wine-sacrament) points to the presence of such a God-symbol in Hebrew religion long before our era. There is a clear Mazdean element, finally, in the allusion to the mystic stone in Zechariah, 4 the "seven eyes" being certainly connected with the Seven Amesha-Spentas, of whom Mithra on one view, and Ormazd on another, was chief. 5 And when we find in the epistles 6 phrases as to Jesus being a "living stone" and a "spiritual rock," and read in the gospels 7 how Jesus said, "Thou art Peter, and upon this rock I will build my church," we turn from the latter utterance, so obviously unhistorical, back to the Mithraic rite, and see in the mystic rock of Mithra, the rock from which the God comes—be it the earth or the cloud—the probable source alike of the Roman legend and the doctrine of the pseudo-Petrine and Pauline epistles.

The Mithraic mysteries, then, of the burial and resurrection of the Lord, the Mediator and Saviour; burial in a rock tomb and resurrection from that tomb; the sacrament of bread and water, the marking on the forehead with a mystic mark—all these were in

p. 318

practice, like the Egyptian search for the lost corpse of Osiris, and the representation of his entombment and resurrection, before the publication of the Christian Gospel of a Lord who was buried in a rock tomb, and rose from that tomb on the day of the sun, or of the Christian mystery of Divine communion, with bread and water or bread and wine, which last were before employed also in the mysteries of Dionysos, Sun-God and Wine-God, doubtless as representing his body and blood. 1 But even the eucharist of bread-and-wine, as well as a bread-and-meat banquet, was inferribly present in the Mithraic cultus, 2 for the Zoroastrian Hom or Haoma, identical with the Vedic Soma, 3 was a species of liquor, and figured largely in the old cult as in itself a sacred thing, and ultimately as a deity = the Moon = a king. 4 Indeed, this deification of a drink is held to be the true origin of the God Dionysos, 5 even as Agni is a deification of the sacrificial fire. And whereas the Mazdean lore associated the Haoma-Tree with the Tree of Life in Paradise, 6 so do we find the Catholic theologians making that predication concerning the Christian Eucharist. 7 The "cup" of Mithra had in itself a mystical significance: in the monuments we see drinking from it the sacred serpent, the symbol of wisdom and healing. 8 Again, as there is record of an actual eating of a lamb in early Christian mysteries 9—a detail still partly preserved in the Italian usage of blessing both a lamb and the baked figure of a lamb at the Easter season, 10 but officially superseded by the wafer of the Mass—so in the old Persian cult the sacrificed flesh was mixed with bread and baked in a round cake called Myazd or Myazda11 and sacramentally eaten by the worshippers.

Nor was this all. Firmicus 12 informs us that the devil, in order to leave nothing undone for the destruction of souls, had beforehand resorted to deceptive imitations of the cross of Christ. Not only did they in Phrygia fix the image of a young man to a tree 13 in the

p. 319

worship of the Mother of the Gods, and in other cults imitate the crucifixion 1 in similar ways, but in one mystery in particular the Pagans were wont to consecrate a tree and, towards midnight, to slay a ram at the foot of it. This cult may or may not have been the Mithraic, 2 but there is a strong presumption that Mithraism included such a rite. We have seen 3 that a ram-lamb was sacrificed in the Mithraic mysteries; and not only are there sacred trees on all the typical Mithraic monuments, but the God himself is represented as either re-born of or placed within a tree—here directly assimilating to Osiris and Dionysos and Adonis, 4 and pointing to the origins of the Christian Holy-Cross myth. The Christian assimilation of Mithraism is, however, still more clearly seen in the familiar Christian symbol in which Christ is represented as a lamb or ram, carrying by one forefoot a cross. We know from Porphyry 5 that in the mysteries "a place near the equinoctial circle was assigned to Mithra as an appropriate seat; and on this account he bears the sword of the Ram [Aries], which is a sign of Mars [Ares]." 6 The sword of the Ram, we may take it, was simply figured as the cross, since a sword is a cross. 7 Again, as we have seen, Porphyry

p. 320

explains 1 that "Mithra is the Bull Demiurgos and lord of generation." Here then would be, as we have already seen, a symbolical slaying, in which the deity is sacrificed by the deity; 2 and we may fairly infer that the symbolic ram in turn would be sacrificed by the Mithraists on the same principle. Now, it appears to be, as we have said, the historic fact that among the early Christians a ram or lamb was sacrificed in the Paschal mystery. It is disputed between Greeks and Latins whether at one time the slain lamb was offered on the altar, together with the mystical body of Christ; but it is admitted by Catholic writers—and this, by the way, is the origin of a certain dispute about singing the Agnus Dei in church—that in the old Ordo Romanus a lamb was consecrated, slain, and eaten, on Easter Day, by way of a religious rite. 3 Of this lamb, too, the blood was received in a cup. 4 Everything thus goes to show not only that the Lamb in the early Christian cultus was a God-symbol from remote antiquity, but that it was regarded in exactly the same way as the symbolical lamb in the Mithraic cult. 5 In the Apocalypse, one of the earliest quasi-Christian documents, and one that exhibits to us the stage in which Jesuism and the Lamb-God-symbol were still held parts of Judaism, the Gentile differentiation being repudiated, 6 we have the Slain Lamb-God described as having seven horns and seven eyes, "which are the seven spirits of God, sent forth unto all the earth," and as holding in his right hand seven stars 7—that is to say, the seven planetary Mazdean "Amshaspands" or Amesha-Spentas, before mentioned, of which Mithra was the chief and as it were the embodiment.


Footnotes

315:8 So Sainte-Croix, Recherches, ii, 147; and Beugnot, Hist. de la Destr. du Paganisme, i, 157, 158.

315:9 G. H. Rendall, The Emperor Julian 1879, Introd. p. 15. Cp. Elton, Origins of English History, 2nd ed. 1890, p. 337.

315:10 Le Christianisme et ses Origines, iv. 133.

316:1 Dial. with Trypho, c. 70.

316:2 Paul, as M. Havet remarks, would be in the way of knowing the cults of Cilicia. Tarsus, indeed, was a Mithraic centre. (Preller, Röm. Mythol. p. 758; Cumont, i, 19, 240.) This connects with the vogue of the cult among the Cilician pirates (below, p. 325). In Asia Minor and Syria it seems to have been confined to the seaports they frequented. It is highly probable that it is Mithra who was represented by several of the figures identified with Apollo and other deities in the Lares and Penates of Messrs. Barker and Ainsworth (1853), which deals with antiquities discovered at Tarsus, and with the cults of Cilicia, without once mentioning Mithra or Mithraism. Cp. Creuzer, Symbolik, 3te Ausg. i, 342. We know that on the coins of Kanerki, an Indo-Scythian king of the first century of our era, the same aureoled figure is alternately represented as Helios and Mithra. Windischmann, p. 60, citing Lassen, Indische Alterthumskunde, ii, 837.

316:3 Last cit.

316:4 Cp. Jevons, Introd. to Hist. of Religion, ch. 11.

316:5 Cp. Cumont, i, 165-6; Haug, Essays, p. 5. Haug rightly suggests that both Jews and Persians may have drawn from a central source.

316:6 Ellis, Polynesian Researches, 2nd ed. i, 324-5.

316:7 Arnobius, Adv. Gentes, v, 5. 8 That found at Neuenheim. See Cumont, i, 165.

317:1 Jastrow, Religion of Babylonia and Assyria, p. 562; Maspero, Hist. ancienne des peuples de l’orient, 4e édit. D. 157; Sayce, Hibbert Lectures, pp. 26-6.

317:2 Prof. Cumont is satisfied that the rock is here, as in Vedic mythology, the symbol of the cloud, which the Sun-God transfixes with his spear or shaft. Oh this view, the shooting at the rock may be simply a myth-duplicate of the stabbing of the bull. See above, p. 300, note. It is certain that the sky was very commonly conceived in the ancient East as solid. Cp. Yasna, xxx, 5, b, as trans. by Mills (Zendav. iii, p. 31), and by Haug from the Pahlavi (Essays, 3rd ed. p. 346). So also among the Tongans (Mariner, Tonga Islands, ii, 99). There is something to be said also for Dr. Jevons’s theory that rude rock-altars came to be regarded as Gods through being drenched with the blood of sacrifices which the Gods were supposed to enter the stone to consume (though it is not clear that he had the "Rock of Israel" in view). But this theory takes a stronger form in the argument of Mr. Grant Allen (Evolution of the Idea of God, ch. v) that the altar-stone was originally a tomb-stone, erected over an ancestor, and that he was the spirit identified with the stone. That all altars, and all temples, are evolved from grave-stones and grave mounds is well proved by Mr. Spencer, Principles of Sociology, §§ 137-9. On this basis, myths of the origination of men and Gods from rocks become newly intelligible. See Mr. Allen again (p. 248, sq., and p. 389) for the suggestion that the divine "corner-stone" may signify a victim slain as foundation-spirit.

317:3 Deut. xxxii.

317:4 Zech. iii, 9. Cp. Dan. ii, 34.

317:5 Windischmann, p. 62; Seel, p. 215; Darmesteter, Ormazd et Ahriman, p. 38.

317:6 1 Peter ii, 4, 5; 1 Cor. x, 4. In the first case the Greek word is lithos; in the second petra.

317:7 Matt. xvi, 18.

318:1 Cp. Frazer, Golden Bough, 2nd ed. i, 359; ii, 366.

318:2 Cp. Cumont, i, 146, 197, 320.

318:3 Spiegel, Avesta, i, 8, citing Windischmann, Ueber den Somakultus der Arier; Max Müller, Physical Religion, p. 101; Psychological Religion, p. 65.

318:4 Max Müller, as cited, and in Psych. Rel. pp. 121, 139-140, 147. Cp. in the Zendavesta, Yasna iii, iv, vii, viii, ix. In Yasna ix, Haoma becomes house-lord, clan-lord, tribe-lord, and chieftain of the land. Cp. Mills on Yasna ix (S. B. E. xxxi, 230) as to the antiquity of the idea; and see Spencer, Principles of Sociology, vol. i, ch. 23, as to its causation. Mr. Spencer makes a striking suggestion in this connection as to the origin of the idea of the tree of knowledge in Genesis.

318:5 Roscher, Ausführliches Lexikon, 3045; Max Müller, Anthropological Religion, p. 355. As above noted, p. 53, Miss Harrison has newly proved the point, tracing a number of the obscurer epithets of Dionysos to names of grains used to make beer.

318:6 Cp. Bundahish xviii, 2, 3; xxvii, 4; xxx. 25 (S. B. E. vol. v); Yasna I (S. B. E. xxxi); and Mrs. Philpot's monograph, The Sacred Tree, 1897, pp. 13, 123, 130-1.

318:7 Fischer, Heidenthum and Offenbarung, p. 150.

318:8 Creuzer, Das Mithrēum von Neuenheim, p. 37.

318:9 Below, p. 320.

318:10 See refs. on p. 143.

318:11 Haug, Essays on the Parsis, 3rd ed. pp. 112, 139, 368.

318:12 De Errore, xxviii.

318:13 See Julian (In deorum matrem, c. 5) on the tree of Attis, which was "cut down at the moment when the sun arrives at the extreme point of the equinoctial arc."

319:1 Horos, it should be remembered, was by the Valentinian Gnostics called "The Cross" and the Redeemer (Tertullian, Contra Valentin. c. 9). Suggestions of the crucifix appear in the Mazdean monuments. See the development from the winged figure, in Lajard's "Atlas"; and compare the plates in Bryant, i, 294; R. K. Porter, Travels in Georgia, etc., 1821-2, i, 668; ii, 154; and Texier, Descrip. de l’Arménie, etc., pl. 111—the two latter reproduced by Justi, Geschichte des alten Persiens, pp. 52, 69. See there also, p. 13. the tomb of Midas, covered with ornamentation of crosses. That the "crown of thorns" is a variation on a nimbus has long been surmised. Mithra, of course, had a nimbus, and this appears from the monuments (Cumont, ii, 336) to be the kind of crown given in the mysteries to the initiate. In the older Persian form of the cult, again, the Sun-God rode "with his hands lifted up towards immortality" or heaven (Mihir Yasht, xxxi; in Darmesteter, ii, 152). He would further be associated with some form of the cross which stood for the four-spoked sun-wheel, as in the myth of Ixion. See Böttger's Sonnencult der Indogermanen, 1891, p. 160, citing E. Rapp's essay, Das Labarum and der Sonnencultus; and compare the Assyrian sculpture of the Sun-God with the solar-wheel in presence as his symbol.

319:2 This tree-cult is assumed by Dr. Frazer (Golden Bough, 2nd ed. ii, 132, note) to have been that of Attis, in which the tree figured so prominently; but that is one of the points at which the cults were likely to converge, both being associated with that of the Magna Mater. Firmicus, in the chapter cited, seems in separate passages to point to two tree cults, mentioning the ram in the second reference only and the simulacrum juvenis in the first. See above (p. 306) as to Dr. Frazer's similar ascription to the Attisian cult of the rock-tomb, which presumptively belongs to the Mithraic.

319:3 Above, p. 301.

319:4 On the Adonis myth see Frazer, Golden Bough, 2nd ed. ii, 115 sq. And see in Guigniaut's edition of Creuzer (figure 139 b, vol. iv) the representation of Osiris as the Sun-God emerging from a tree. Dionysos was similarly figured. Cp. Frazer, ii, 160, and refs.

319:5 De Antro, xxiv.

319:6 The later Persians specially celebrated the entrance of the sun into Aries as the "new day" (Nùrùz). "The public Nùrùz [as distinguished from that of the nobles] falls on the first day of the month Ferwardin [March], which happens as the sun enters the first point of Aries; and when it arrives at this first point it is the Spring. They say that Almighty God on this day created the world, and that all the seven planets revolved towards the ascending nodes of their orbit, and all these ascending nodes were is the first degree of Aries, on which day it is firmly believed that they enter on their march and circle. He also created on this day Adam (on whom be peace!)—on this account likewise they call it Nùrùz." Berhan-ĭ Katteā, cited by Wait, Antiquities, p. 187. The Nùrùz of the courtiers was six days later (another parallel to the Christian system); and "the Khosrùs every year, from the public Nùrùz to that of the courtiers, which was a space of six days, were in the constant habit of relieving the poor, of liberating the prisoners, of granting pardon to the malefactors, and of entirely devoting themselves to mirth and gladness" (ib. p. 190).

319:7 Note, on this, the astronomical "crossing" of lines at the "first point of Aries" (see p. 320 English or Chambers’ Encyclopædia, art. Zodiac); and see it imaged in the old figure in Brown's ed. of Aratos.

320:1 Last cit.

320:2 Firmicus tells (vi) that the people of Crete destroyed a bull to represent the destruction of Dionysos; and in the Egyptian slaying of the ram for Amun the ram was mourned for by the worshippers, and was put on the image of Amun, an image of "Herakles" (presumably = Khonsu) being then placed beside it (Herodotus, ii, 42). "We may conjecture," says Dr. Frazer (Golden Bough, 2nd ed. ii, 167), "that wherever a God is described as the eater of a particular animal, the animal in question was originally nothing but the God himself." Cp. Lang, Myth, Ritual, and Religion, 2nd ed. ii, 251-4.

320:3 Bingham, Christian Antiquities, B. xv, c. 2, § 3; Hatch, Hibbert Lectures, p. 300.

320:4 Casalius, De Veterib. Christ. Ritib. ii, 4, cited by Dupuis.

320:5 A sacramental quality attached to the lamb also in the worship of Apollo, whose oracle at Larissa was given by a priestess who once a month tasted of the blood of a sacrificed lamb, and so became possessed by the God. Pausanias, ii, 24.

320:6 See above, p. 142.

320:7 Rev. i, 16; v, 6; iii, 1; v, 6; etc.


Next: § 10. Further Christian Parallels