universal object of worship for Cretan, Anatolian, and the various invading races—Thracian, Phrygian, Hittite, Semite, and finally Greek. Whether she is called Kybele, or Allilat, or Ishtar, or Artemis, or Zemelo, or Aphrodite, or Ariadne, or has no name that we can recover-perhaps none at all—is a minor matter; and the question whether her cult spread from one centre or arose independently in a score of places seems insoluble in our present state of knowledge. Certainly she appears to have nothing to do with either race or social organization. Of this also we can be fairly sure-that in every form of her cult which we know anything about she is provided with a male partner, normally far inferior to herself, and having for his function in many cases nothing but the task of enabling her to conceive. That he is an afterthought, and that the goddess dates back to a far remote time when the part played by the male in procreation was quite unknown, is indeed a plausible hypothesis, but cannot be said to rest on cogent evidence. It thus is natural enough that he is often not quite a god. Attis is probably divine in origin, and made into a human youth by later fancy, chiefly that of the Greeks, who were puzzled by a god dying; but Endymion has no divine attributes as we know him,' and Anchises has just a faint trace of worship. Aphrodite is called Anchisias in an inscription from Ilium Nouum, and there was a cult or two of him at some of his supposed graves.2 But the most curious example of the hesitancy as to the nature of the goddess' consort comes from Naxos. That Ariadne was originally a goddess and no heroine is, I think, pretty clear both from her occurrence in the territory of the great Cretan goddess, of whom she, like Britomartis, would seem to be a 'faded' form, and by her extraordinary ritual at Amathus in Crete, where she is called Aphrodite Ariadne, and in her rites a pantomime of child-birth is gone through. The Naxians, who also worshipped her, went so far as to declare that there must be two Ariadnes, one the bride of Dionysos and the other of Theseus; obviously a desperate attempt to reconcile the, to them, contradictory indications furnished by her ritual and legend. What lies behind all this is no doubt in all cases a god of the general type of Attis, who is a vegetation daimon, dying and rising again, and therefore appears to the Greeks now as a god, now as a hero, one character or the other predominating in different regions or different legends. But as we enter definitely Oriental territory we find a characteristic besides those mentioned already, which is common to many of the goddess' lovers-they are eunuchs. This is too well known to need demonstration in the case of Attis. Adonis is never actually represented as castrated, but it is noteworthy that he dies of a wound in the thigh, which, considering the euphemistic meaning of unpoi and the fact that his death is represented as quite rapid, suggests that we have not exactly the original tale as it was told of Tammuz. If we go north-east we S.u. 1 See the article of von Sybel in Roscher, 2 Particulars in Wörner, op. cit., col. 339. 3 Plut. Theseus, 20. See Aesch., frags. 134, 135, Nauck. find the great goddess of the Iranian Scythians served by eunuchs, much as Kybele was.1 Now it is a curious fact that Anchises is definitely afraid of something which is not death, but some kind of life-in-death, and that the doom which he says overtakes anyone in his position is that he is not Biolámos. With this we may couple the fact that very little is said of him as husband or father apart from Aphrodite and Aineias. He has one other son, Lyros, also by Aphrodite, according to Apollodoros, l.c. It is noteworthy that the Great Mother sometimes bears two children (thus Ariadne is the mother of Oinopion and Staphylos, Plut. l.c.); he is just mentioned as having a wife and a daughter in a single passage of Homer (N. 428 sqq.), and Homer may generally be trusted to make a sacral hero into an ordinary, fully human one. Save that the scholiasts find or invent a name, Eriopis, for his wife, and Naevius casually mentions her (ap. Serv. on Aen. III. 10), we know nothing about her. I suggest that she has no existence in the original story, which represented Anchises as being as cold (to all but Aphrodite) as Adonis himself. Now it is a curious and paradoxical thing that a goddess of fertility should be worshipped by the unfertile, and it seems to me that only two explanations are possible. One, which for a time I adopted, is that because the goddess is female her worshippers make themselves as nearly female as possible. This is quite in keeping with the general tendency, in mystic cults, to seek assimilation with the deity, but it is open to the serious objection that the Great Mother had no need to be worshipped by artificial females when real ones were to hand; the Galli could, one would think, have been replaced with advantage by women or girls, such, for instance, as worshipped Artemis of Brauron. The other explanation, derived from ritual facts, on which Graillot rightly lays some stress,2 is, I think, the true one: her worshippers are eunuchs because they have given her all their fertility. Macte esto is a Roman formula, but not a merely Roman idea; the worshipper is indeed benefited by the god, but so also is the god by the worshipper (see the excellent discussion by van Leeuw of Die do-ut-des-Formel' in Arch. f. Relig. for 1921, p. 241 sqq., and the classical treatment of the matter by Warde Fowler in Religious Experience of the Roman People). As at the Fordicidia the reproductive power which the cow is not allowed to use (in the shape of her unborn calf) is given to Tellus (Telluri plenae uictima plena datur, says Ovid, Fast. IV. 634, with substantial accuracy), so in a much more horrible rite the adorers of the Great Mother give her their reproductive powers. It is from the East, it would seem, that the idea, afterwards prominent in Christian theology, of entire and unreserved consecration to 1 See Rostovtzeff, Iranians and Greeks in South Russia, p. 105, and Plate XXIII., and his whole discussion of the cults of that region, ibid., and in Rév. ét. gr., t. XXXII., p. 462 sqq.; Herod. IV. 67; Hippokr. de aere aquis locis, p. 561, Kühn. 2 Op. cit., p. 156. Cf. the legends recorded by Clement of Alexandria, Arnobius, and Pausanias, conveniently collected by Hepding, Attis, P. 31 sqq. Legend and ritual alike show the uires to be magically potent and desired by the goddess. a deity comes; it is not surprising to find a limited and savage form of this mystic doctrine in the cult of the Great Mother, that extraordinary mixture of debasing superstition and transcendentalism, as it was in later days. The husbands of the goddess do not simply play their part in the process of fertility, they expend upon that union all that they have or ever will have of reproductive power. It is parallel to that well-known magical idea underlying much human sacrifice and the like, according to which the life of the victim is absorbed into the greater life of some superhuman power.1 I am inclined to think that in one of the most curious and most unhellenic of the myths which have come down to us we have a confused recollection of another story growing out of this practice-I refer to the mutilation of Ouranos. I would see in the familiar Hesiodic account a conflation of two stories: (1) The widespread myth (it reaches as far as New Zealand) of how Heaven and Earth were separated to make room for other beings to walk about between them. Kronos in this part of the tale is the Greek Maui. (2) A quite different story of the type of the legend of Kybele and Attis, in which Earth has some other and much inferior husband, whose virility is wholly sacrificed in her service. To a Greek, familiar throughout his development with the concept of the sky-father mated to the earth-mother, the former story was intelligible, the latter only imperfectly so. Hence conflation and confusion. There remains now the other curious circumstance in the legend of Anchises-the incident of the thunderbolt which does not kill him, but leaves him to live to a good old age, and receive honourable burial and some modicum at least of cult. The general conception of the thunderbolt in European belief is that it is a destroying missile. Examples will be found in abundance in Blinkenberg's useful compilation, The Thunderweapon. We may recollect that, to take prominent myths from the north and the south of the continent, Thor and Zeus use it to overcome hostile giants. But there exists another idea, and it is noteworthy that it is to be found in a district where a cult of a Great Mother flourished—namely, Naxos, whose worship of Ariadne has already been mentioned. The Naxians say, according to Diodorus Siculus (V. 32), that the reason why Semele was smitten with the thunderbolt was that Dionysos might be the child, not of a god and a mortal, but of two immortals. To this legend Ovid alludes in the Fasti, as I have briefly pointed out elsewhere.2 It is worth noting also that one people, geographically European, but almost certainly of Asiatic origin, the Etruscans, not only divined from thunder, but had a ritual for evoking it, and that the Romans, 1 For example, the notorious sacrifice of infants to Melqart at Carthage seems to result from the assumption that the god needs a supply of fresh, unworn life to sustain his own immortal life. A simpler form of this is the Alkestissacrifice in its many ramifications in folklore: A, in order to live longer, appropriates the life of B. Deme meis annis et demptos adde parenti, suggests Jason to Medea (Ov. Met. VII. 168). In view of the orgiastic character of the rites under discussion, it is worth noting that lunatics not infrequently have, besides the tendency to suicide, a perverse desire for self-mutilation. See Hastings, Enc. Rel. Eth., art. SUICIDE. 2 See Class. Rev. XXXVI. (1922), p. 116, note on Fast. III. 715; Rohde, Psyche', I., p. 324. their pupils in this matter, regarded the victim of lightning as sacer, too full of mana to be lightly touched or approached. I may also mention that Demeter and Isis, two of the most prominent representatives of the Great Mother, have attached to them in Greek mythology the notion that, if an arrangement could be made by which someone could survive burning by ordinary fire, he would become immortal. On this, however, I would not lay too much stress, for a variant of the same tale attaches to Thetis, and another has come by way of a local rite to be a famous, if relatively late, episode in the Herakles-saga.2 Apart from public cult and from legend, we have an interesting survival of this conception of thunder in the inscription of Thyateira in Lydia (Kaibel, 320), which runs as follows: Αὐτὸς Ζεὺς Κρονίδης ὑψίζυγος αἰθέρι ναίων Here the dead woman, having been killed by lightning, is definitely claimed as divine, and it is stated in so many words that the manner of her death rendered her so. If we go from the east to the west of the Greek world we get the curious tale of Empedokles' attempt at self-deification by jumping into Aetna. It is a commonplace of popular Greek belief that the fires of Aetna, and of volcanoes in general, are divine, whether because of their connexion with the workshop of Hephaistos, or because they come from the flame of the thunderbolt wherewith the giant who lies beneath the mountain was struck down. Is this what was in the mind of Empedokles, if he really ended himself in such a fashion, or if he did not, then in the imagination of whoever invented the story ?3 I therefore suggest that originally the thunderbolt was not meant to do Anchises any harm at all, and was no punishment for vain babbling or any other offence, but that he was touched with celestial fire, perhaps before, not after, his meeting with Aphrodite, to make him more than mortal and a more fitting consort for her. UNIVERSITY COLLEGE OF Wales, 1 Hom. Hymn. Cer. 233 sqq.; Plut. de Is. et Os., ch. 16. 2 This has now been finally cleared up by Nilsson; see his articles Der Flammentod des Herakles auf dem Oite, Arch. f. Rel., 1922, p. 310 sqq., and Herakles in Nordisk tidskrift, 1923, p. 119 sqq. The notion is a latecomer into the H. J. ROSE. story of Herakles, but early in itself. I am not sure that the practice of cremation may not have owed something to the desire to give the corpse the heat (i.e. life) which it obviously lacked. 3 The suicide in historical times of Peregrinus is clearly a deliberate imitation of Herakles. PHILIP V. AND PHTHIA. CONTEMPORARY inscriptions prove practically beyond doubt that Philip V. was the son of Demetrius II. and of the Epeirot princess Phthia.1 But historians have always started from Eusebius' statement (1,237 Schoene) that he was the son of Chryseis, a Thessalian captive whom Demetrius married and who afterwards married Doson, and have tried to fit other things in with Eusebius. Now it does not much matter to us which of two unknown women was Philip's mother; but it does matter how we approach our evidence, and we must start from the contemporary evidence and work downward, not vice versa. I need not here go through all the different theories based on the late evidence; what I am going to do is first to examine the inscriptions, which has not yet been done in this connexion, and then explain how Eusebius' blunder arose;3 for even if a definite statement be certainly wrong, we want to know how it got there. I.G. 112, 1299-Syll.3 485 (first decree for Aristophanes). In common with almost all scholars now, I regard Kolbe's attribution of this decree to the reign of Demetrius II., 239-229, as certain; the references to the war cannot be made to fit Demetrius I. The war has continued for at least four yearsLysias, Cimon, [Ecphantus], and the year of the decree; but the decree may be two or three years later than [Ecphantus]. I use the Ferguson-Kirchner dating of I.G. 112: Lysias, 238/7; Cimon, 237/6 (so also Kolbe); [Ecphantus], 236/5. The only alternative now, which I do not agree with, is Johnson's, who puts them all one year earlier (which makes no difference to my argument). The decree consequently cannot be earlier than 235/4—i.e. in the middle of Demetrius' reign. (We shall see that it is probably rather later.) In the January last before the decree-i.e. January 235 at the earliest, some four years after Demetrius' accession-Aristophanes has sacrificed for king Demetrius, queen Phthia, and their children, the formula (1. 10) being ὑπὲρ . . . τοῦ βασιλέως [Δημητρίου κ]αὶ τῆς βασιλίσσης [Φθίας] καὶ τῶν 1 Strictly speaking, of a queen of Demetrius' whose name has five letters in the genitive. But the name is not in doubt. 2 Droysen alone, III. (1)2, p. 400, n. 1, does say that Philip was son of Demetrius and an Epeirot princess. But his later definite statement that Chryseis was his mother [III. (2)2, p. 52] shows that this may only be a slip. The modern evidence was, of course, not available to Droysen. 3 It would be worth compiling a complete list of the confusions in late writers as to names and relationships. It would show once for all how unscientific it is to use them as evidence for details. 4 Beloch alone stood out for Demetrius I. He has been followed by Costanzi, Δημητριακός Tóλeμos, in Beloch's Festschrift, Saggi di storia antica, 1910; but Costanzi has no new arguments, and does not even consider the war. I know of no other dissentient; it remains to be seen what view Beloch's forthcoming second edition will take. 5 The Creation of the Tribe Ptolemais, A.J. Phil., 34, 1913, 381 sqq., to which I refer throughout. B |