Version classiqueVersion mobile

Mixanthrôpoi

 | 
Emma Aston

Section One : Cults and composition of mixanthropic deities

Chapter II

Terrestrial mixanthropes

Texte intégral

1. Horse mixanthropes: Cheiron and Demeter Melaina

1The juxtaposition of these two personalities is intended chiefly to demonstrate a cardinal truth about divine mixanthropy: that a single animal species cannot be considered to have an absolute and permanent symbolic meaning. A horse element (the example treated here) can carry completely different associations and significance, depending on context. We cannot use mixanthropic deities to build up a vocabulary of animal symbols in ancient thought.

  • 1 For a somewhat over-imaginative reconstruction of Cheiron’s religious significance in this region, (...)
  • 2 For the cave and its contents, see Giannopoulos (1912) and (1919); Levi (1923).
  • 3 On the inscription, see Comparetti (1921-2); Decourt (1995), 90-94, no. 73.
  • 4 On Pantalkes and other real-life nympholepts, see Larson (2001), 13-20; for nympholepsy as a mythol (...)

2There are various small traces of Cheiron-worship in the Greek world. For example, a boundary-marker discovered bearing the word ΧΙΡΩΝΟΣ, a genitive, suggests a temenos at Poseidonia, in S. Italy.1 Little can be done with such scraps, beyond noting that they corroborate the status of Cheiron as a cult-receiving deity. However, somewhat more substantial evidence comes from Thessaly, where there are two attested sites of worship. In a cave near Pharsalos in Phthia,2 on a spur of Mount Othrys, was found a long fourth-century inscription (still in situ but now rendered largely illegible by lichen)3 recording the favours granted by various deities to a person called Pantalkes who, roughly a century before, set up a dedication to the nymphs.4 Cheiron is included in the fourth-century text, which is long and metrical, alongside numerous others: the nymphs, Pan, Hermes, Apollo, Heracles and his companions (whoever they are), Cheiron, Asklepios and Hygieia. Cheiron, the inscription tells us, gave Pantalkes wisdom and musical ability, in line with his mythological qualities. This inscription is an interesting variation on the divine collective which so often features in cave-cult: some familiar faces are there (Pan, Hermes, the nymphs), but in addition some less typical ones, including Cheiron himself; besides, there is an unusual level of differentiation, by which each deity is described as having contributed his or her special property to the fortunate Pantalkes. This is more nuanced than the groupings discussed with regard to Acheloos in which the divine functions of the several deities seem substantially the same. However, although Cheiron is distinguished, he is not prioritised above the other recipients of worship in the cave.

  • 5 The text in question was originally attributed to Dikaiarchos, and appears under his name in FHistG (...)

3A second site in Thessaly is at once more promising and more problematic. The Hellenistic geographer Herakleides5 describes the following site and ritual on the summit of Mount Pelion:

  • 6 This spelling of the title is generally thought to be a mistake of Herakleides: inscriptions from t (...)
  • 7 Herakleides 2.8: Ἐπἄκρας δὲ τῆς τοῦ ὄρους κορυφῆς σπήλαιόν ἐστι τὸ καλούμενον Χειρώνιον καὶ Διὸς (...)

On the peaks of the mountain’s top there is the cave called the Cheironion and a hieron of Zeus Aktaios,6 to which, at the rising of the Dog Star, at the time of greatest heat, the most distinguished of the citizens and those in the prime of life ascend, having been chosen in the presence of the priest, wrapped in thick new fleeces. So great is the cold on the mountain.7

  • 8 See Arvanitopoulos (1911); Philippson (1944), 147; Chourmouziades (1982), 98. Unfortunately, the si (...)

4The Greek archaeologist Arvanitopoulos, whose work in the early years of the twentieth century added so greatly to our knowledge of Thessalian material remains, partially excavated a site on the Pliassidi peak of Pelion which may be identifiable as the one described by Herakleides.8 It does not, however, yield much further insight into the relevant ritual practices; in addition to a cave were found the remains of two small buildings which Arvanitopoulos interpreted as shrines, one of Cheiron and one of Zeus Akraios. However, this is tendentious, especially the notion that Cheiron would have had a built shrine in addition to his sacred cave.

  • 9 There has been a determination in the older scholarship to edge Cheiron into greater prominence wit (...)
  • 10 Buxton (1994), 93-4.

5The Cheironion mentioned in the text above must have been equated with the mythological cave-home of Cheiron on Mount Pelion. The word Cheironion may itself suggest worship, as does Thetideion, though this is not certain. The ritual described by Herakleides has as its chief recipient Zeus Akraios,9 whose cult was of great regional significance, but Buxton is surely right to argue that Cheiron’s cave, steeped in mythological significance, would have contributed to the function of the ritual as part of the symbolic ‘furniture’ of the site.10 So what kind of ritual was it, and how might the Cheironion have fitted into its performance?

6The main question is whether we can relate it to Cheiron’s mythological rôle as a nurse and educator of heroes. In myth, the centaur rears and tutors an impressive number of child heroes, feeding them strengthening foods and equipping them with aristocratic skills such as hunting and music. His care of Achilles is especially popular among vase-painters; an example can be seen at fig. 19. Can we see the Pelion ritual as forming a cultic counterpart to this rôle, demonstrating that Cheiron had a kourotrophic function as a deity, overseeing the development and the transitions of the young?

  • 11 IG XII.3, 360. See Vogel (1978), 218.
  • 12 Philippson (1944), 150-55. She posits a process whereby the cult travelled from Pelion in Thessaly (...)

7This possibility presents difficulties. First, it elides the rôle of Zeus, to whom the ritual is chiefly directed. Second, the men taking part in the ritual cannot be seen as historical counterparts of the young heroes whom Cheiron assisted to adulthood. They are described by Herakleides as akmazontes, that is, at the akmê of their strength: young, certainly, but already in adult­hood, not just about to enter it. This is not a male, Thes­salian equivalent of ‘playing the bear’ at Brauron. Like some rites of passage, the ritual does appear to involve a temporary stepping into the wild, through the donning of fleeces, primitive clothing associated with herders and huntsmen, and as Buxton points out, Cheiron’s cave would have assisted in this symbolic process, since caves, like fleeces, represent the realm of ‘Outside and Before’, in contrast with the sphere of human civilisation. However, the element of adolescent transition cannot be extrapolated from the scanty evidence. It is worth noting, however, that Cheiron (spelled Chiron) appears as one of several deities in a seventh-century BC dedicatory inscription11 found near the temple of Apollo Karneios on Thera, in the vicinity of which was also a sacred cave which was apparently used in ephebic rites.12 This provides a slight suggestion that Cheiron’s mythical rôle as nurse and educator found ritual expression, but it is far from substantial, and far from Pelion.

  • 13 Pind. Pyth. 4.102-3.

8That Thessalian Cheiron had a kourotrophic aspect as a deity is not really to be doubted. It is made clear by the juxtaposition of Cheiron with the nymphs in the Pharsalos cave inscription. This connection may be seen as referring to the nurturing function of both, dominant in myth. For example, Pindar makes the hero Jason say that he was nursed in infancy by Cheiron, his mother Philyra and his wife Chariklo, both nymphs, and by his nymph daughters, the kourai hagnai.13 However, vague kourotrophy is one thing, overseeing ritual transitions is another.

  • 14 Plut. Quaest Conv. 3.1.
  • 15 Herakleides 2.12.

9Cheiron’s divine function which, in Thessaly, seems most unambiguously attested is healing. Plutarch relates that the Magnetes brought offerings to Cheiron as a healer.14 Earlier, Herakleides tells us that a dynasty of physicians living at the foot of Mount Pelion, who practiced their craft strictly free of charge, traced their line back to the centaur himself, though no worship is actually mentioned.15 In his healing capacity, Cheiron was strongly associated with the natural forces of his mountain home. Cheiron, Pelion and the mountain’s native healing herbs form an inseparable triad in the works of many ancient authors. This is reflected especially strongly in a fragment of the Hellenistic author Nikandros’ Theriaka, which gives the following medicinal instruction:

  • 16 Nikandros, Theriaka 500-505: πρώτην μὲν Χείρωνος ἐπαλθέα ῥίζαν ἑλέσθαι, | Κενταύρου Κρονίδαο φερώνυ (...)

Choose first the medicinal root of Cheiron,
Which carries the name of the centaur, Kronos’ son; Cheiron once
Discovered and took note of it on a snowy ridge of Pelion.
It is encircled by waving leaves like sweet marjoram,
And its flowers are golden in appearance. Its root, at the
Surface and not deep, resides in the grove of Pelethronios.
16

  • 17 Theophrast. Peri Phytôn IX.11.1-7. It is clear from the description of the plant (golden flowers) t (...)

10Cheiron, then, is the mythical discoverer of a major natural resource, and a form of culture-hero. This plainly accords with his healing persona among the Magnesians. He gives his name to the plant called ‘Cheironeion’, whose properties are described by Theophrastos: it is used to cure the bites of snakes, spiders and other venomous creatures.17 So Cheiron in a sense embodies the healing properties of Mount Pelion itself, in the form of its native herbs. More generally, literary sources stress almost without exception his kindness, justness and goodwill towards humanity. In this regard, literary characterisation and divine function are in accord.

11How – if at all – do these benign characteristics connect with Cheiron’s physical form? It is meaningless to make vague assertions about the fertility-associations of the horse. Cheiron’s equine element has no absolute and intrinsic meaning. However, it does operate within a set of piquant contrasts in Greek thought, and these are illuminating. In fact, Cheiron’s centaur form cannot be understood at all without an examination of his relationship to the other centaurs, a relationship which mythological episodes often bring to the fore. However, this, in turn, cannot be understood without taking into account the two sides to the horse in Greek thought.

  • 18 As Hodkinson (1988, 64) puts it, ‘The horse was the useless animal par excellence in an era before (...)

12Horses had their place within the reality of Greek life; but they did not have the domestic ubiquity of the ox, or the pastoral ubiquity of the goat. Rather, they belonged to certain spheres – war, hunting – and to certain groups, chiefly the richer classes. They were not generally used for haulage; and mules and asses would have been more widely used as saddle-animals among the poorer elements of Greek societies.18 None the less, they had their place within the canon of domestic animals as equipment for human activity. Their rôle in myth, however, often goes against this fact. A significant number of stories, some well-known, some less so, show the horse as an aggressive, destructive and terrifying creature – to all intents and purposes a thêr – a wild beast – rather than a docile honorary member of the human community.

  • 19 Diod. 4.15; Hyg. Fab. 250. The animals also consume Abderos, Herakles’ servant, who has been set to (...)
  • 20 Aelian HA 15.25; Apollod. Bibl. 2.5.8. Domestic animals released into the wild do not always become (...)
  • 21 For some pertinent myths, see Ant. Lib. Met. 7 and 20.

13The best-known examples of horses in this rôle are the horses of Diomedes, which probably owe their fame, both ancient and modern, to their inclusion in the myth-complex of the labours of Herakles. Diomedes is a Thracian king, who feeds his guests to his murderous horses in drastic transgression of the laws of guest-friendship. Herakles defeats him, and pays him in kind, causing him to be devoured by his own horses.19 The myth has an interesting finishing touch: released by King Eurystheus (to whom Herakles has presented them), the horses wander on Mount Olympos and are themselves there eaten by wild beasts: clearly, for all their savagery, they are no match for the thêres of the mountain.20 Sometimes the horses are explicitly mares, sometimes stallions. There exist a number of variations of this theme of the man-eating horses who eventually turn against their owner: the very strength of this theme in myth reflects the tense symbolic potency of the horse.21

14A perennial theme in the accounts is the horse’s ambiguous and perilous quality. It is an animal which is meant to serve its owner, even though in some cases the use to which it is put is unnatural in itself: devouring of guests, abnegation of intercourse, failure to maintain pastoral land, a transgressive sacrifice. However, the horse turns on its owner, and the transgression involved in its use rebounds on the human perpetrator. The horse may be the instrument of a wider punishment; but there is still huge significance in the repeated choice of this particular animal to take this rôle. The domesticity of the horse is not, in myth, to be relied on.

  • 22 On the ancient characterisation of centaurs, and on the symbolic distinction between the group-cent (...)
  • 23 Ancient descriptions of the event in Pind. Nem. 3.74-5, Pyth. 4.181-4; Pyth. 3.83-95; Servius Verg. (...)
  • 24 See Pind. Pyth. 2.42-8; Diod. 4.69-70; Hyg. Fab. 62.
  • 25 Apollod. Bibl. 3.14.6. A parallel of sorts is the birth of Typhon to Hera, which is accomplished wi (...)

15Just as there are two sides to the character of the horse, so there are two kinds of centaur, the mixanthrope in which the horse-form is most famously and consistently incorporated. Both myth and representation are used to make and keep a very definite divide between, on the one hand, ‘the centaurs’ as a group, sometimes named but more often an anonymous rabble (called here, for convenience of expression, the group-centaurs), and on the other the virtuous individuals, Pholos and – far more prominently – Cheiron.22 Cheiron’s parentage sets him apart from the group-centaurs: he is the son of Kronos and the nymph Philyra, who are caught by Rhea in flagrante and are transformed by Kronos into horses.23 Whereas his parentage is not far removed from the god-nymph combination typical of heroes (on the face of it at least, before the metamorphosis element is taken into account), the way in which the group-centaurs come into being recalls much more strongly the ‘recipes’ for monsters to be found in myth. They are the products of an unnatural, a transgressive union. Ixion, legendary king of Thessaly, attempts to rape Hera but is thwarted by Zeus, who substitutes for his wife an eidôlon made of cloud; from the unwitting coupling of Ixion with this cloud comes Kentauros, the progenitor of the race.24 So the difference between the origin of Cheiron and of the group-centaurs is striking, though it should be noted that in the former case too, transgression and boundary-crossing are involved, in the forms of adultery and metamorphosis. Key to the latter, though, is the motif of the futile mating, which does not achieve its (violent) aim; this motif as recipe for monsters has a parallel in the birth of Erichthonios, created when Hephaistos fails to rape Athene and his sperm lands on, and fertil­ises, the earth.25

  • 26 It should be noted that one unusual and isolated variant on the myths elides the difference between (...)
  • 27 Jeanmaire’s (1949) discussion of the character of Cheiron in ancient poetry is still of great value (...)

16So the group-centaurs are monsters; Cheiron seems more the hero-with-a-difference.26 He might also be thought to embody the positive aspects of the horse, just as the group-centaurs typically display that species’ mythical potential for violence against humans. In fact, as was said in the Introduction, all centaurs may be accorded aspects of nobility. However, whereas the group-centaurs display them only occasionally and unreliably, Cheiron is undeviatingly good, never sharing in their ambiguity.27 His prime mythological rôle as nurse and educator of heroes places him at the heart of an enduring pattern of obsolete Homeric aristocracy just as powerful, as an idea, in democratic Athens as in Thessaly.

  • 28 Apollod. Bibl. 3.13.3.

17Occasionally, Cheiron is directly pitted against the group-centaurs in myth. An example occurs in the account by Apollodoros, who describes an attempt against the life of Peleus by his enemy Akastos. Peleus is marooned on Mount Pelion by Akastos who hides his sword, and is thus at the mercy of the group-centaurs who attack him while he is sleeping. Cheiron, however, comes to his aid, and finds his sword for him so that he is able to defend himself, and so escape to safety.28

  • 29 See e.g. Gantz (1993), 145 on the patterns at work in the representations of centaurs in the metope (...)
  • 30 An example of this composition is his depiction on the François Vase.

18This distinction between Cheiron and the group-centaurs is also expressed in artistic representation. Vase-paintings – mostly Attic, of course – are espe­cially important for this observation because the consistent popularity in this medium of both group and individual centaurs allows for comparison and for isolation of patterns and trends; but these trends are not limited to painted pottery.29 Put simply, whereas the group-centaurs have a whole horse’s body joined to the torso, arms and head of a man, Cheiron’s composition has a far more dominant human element, a man’s body, clothed and entire, to which the trunk and hindquarters are rather awkwardly attached.30 Whereas the group-centaurs are almost always shown in violent movement and in conflict with heroes, Cheiron is most often shown poised and solemn, usually receiving or caring for the infant hero Achilles.

  • 31 For examples of the early type of centaur with human front legs, see Padgett (2003), figs. 4-8
  • 32 Gantz (1993), 145; Mylonas Shear (2002), 151; Padgett (2003), 14.
  • 33 Why does Cheiron not move with the times? Is it simply to keep the dominant humanity, or is there a (...)

19In Cheiron’s composition the human element is the more powerful, whereas that of the group-centaurs is dominated by the horse. But there is another significance of their divergent forms which is perhaps even more telling. Cheiron preserves an older type of centaur-representation which is lost from his rowdy colleagues.31 Naturally, the centaur-type with whole human body does not yield abruptly to the new; there is considerable overlap. But from the end of the seventh century BC, the new type with equine forelegs becomes more frequent, and gradually becomes the standard, in Attic vase-painting at least.32 In the same medium, Cheiron keeps the human-forelegs type.33

20So although the horse does have certain recognisable associations in Greek thought, in Cheiron’s case it is significant not by and in itself, but in combination with the human part which dominates it. A different composition of human and horse is to be found in another case of equine mixanthropy in Greek cult: the figure of Demeter Melaina, ‘Black Demeter’, worshipped in Arkadia in the Peloponnese.

  • 34 Despite the infiltration of the Eleusinian form into Arkadia itself: see Paus. 8.25.2-3 for a sanct (...)

21The geographical specificity that we found in the cult of Cheiron is even more striking in that of Demeter Melaina. We are here dealing with an extreme example of the phenomenon of the local form of a deity who also has a pan-Hellenic dimension. Hugely pervasive as the figure of Demeter is in Greek religion, the goddess under the epiklesis Melaina appears only at a single site, and in this form is radically different from the widespread Eleusinian persona of Demeter.34

  • 35 For a discussion of landscape, settlement and sacred space in Arkadia, see Jost (1999).
  • 36 In fact, as Burkert has shown, various strands of mythology connect not only Phigalia and Thelpousa (...)
  • 37 The relationship between cult-type and topography is one of the consistent foci of Jost’s work, and (...)

22Demeter Melaina was worshipped in a sacred cave on the Arkadian Mount Elaion, in the territory of Phigalia.35 Not far away was Thelpousa, at which was worshipped Demeter Erinys, ‘Fury’, a near-identical form of the goddess whose name reflects the importance of anger and vengeance to both forms.36 As Jost has shown,37 both Phigalia and Thelpousa were located within the mountainous, largely uncultivated zone, and this location is far from coincidental. Neither Demeter Melaina nor Demeter Erinys is the goddess of straightforward agrarian abundance, as the myths, described below, make clear.

  • 38 Frazer (1898), 406-7. Autopsy is important to his discussion. (p. 407: ‘I visited the cave 2nd May (...)

23The cave of Demeter Melaina has never, alas, been found. Frazer thought he had discovered it in the course of his travels in the area; but it is almost certain that he was mistaken.38 The absence of any archaeological data mean that we rely solely on our literary informant for the site: the relevant section of Pausanias’ Periegesis (provided in full in the Appendix). This is a problematic state of affairs; Pausanias was by no means a straightforward eye-witness, to the extent that a later section of this book must be given over to his ideas and their possible impact on our perceptions. At this stage, however, we are justified in making what use we can of his account, dealing with specific issues as they arise.

24While recounting his visit to the sacred cave of Demeter Melaina, Pausanias describes a startling cult image:

  • 39 Paus. 8.42.4 (for the Greek text, see the Appendix below).

[The Phigalians say] the image was made by them in this way. It was seated on a rock, and resembled a woman in everything but the head. It had the head and hair of a horse, and images of serpents and other wild animals grew from its head. It wore a tunic reaching to the feet. On one hand was a dolphin, on the other a bird, a dove. As for why they made the effigy like this, that is clear to any intelligent man who is learned in traditions.39

  • 40 It seems fairly feasible to assume the existence of a cult image at the time of Pausanias’ visit; t (...)
  • 41 Unsurprisingly, the scholars who treat the xoanon without caution are those seeking traces of anima (...)
  • 42 Bruit (1986).

25Only rather later in the section does he reveal the important fact that at the time of his visit the xoanon was not actually in existence. It had in fact been succeeded by two replacement images: a bronze agalma crafted by the sculptor Onatas of Aegina, and the image which we may perhaps assume was in place when the author visited and made an offering, though of this he makes absolutely no mention.40 Most scholars have had no difficulty in taking on trust the existence of the xoanon in exactly the form described by the Periegete, and in adopting it as an important early instance of such mixanthropic representations.41 But in fact the description of the successive statue has far more of the mythological about it than the historical. The mare-headed xoanon has a mytho­logical aition; its loss is accompanied by a story of Demeter’s anger which forms a doublet with that of her earlier anger following her rape by Poseidon; the ways in which the xoanon and Onatas’ statue disappear have a strong element of the mysterious; all in all, as Bruit has demonstrated,42 there is much more to the whole narrative than religious history. Pausanias’ account by itself is not enough to prove that the mixanthropic statue ever existed.

  • 43 Coins of Thelpousa depicting Demeter and a horse, inscribed with the name Erion (a form of Areion), (...)
  • 44 Jost (1985), 304-7. See also Dietrich (1962), 130-134.
  • 45 Cf. a depiction of a gorgon-centaur fighting a lion on a sixth-century amethyst scarab from Byblos: (...)
  • 46 See Smith (1884), 239-40 and pl. XLIII; Gantz (1993), 144; Frontisi-Ducroux (2003), 210.

26That said, it is not implausible that Demeter Melaina should be represented in mare-headed form. Horse-imagery was plainly important in the cults at Phigalia and Thelpousa, not just in mythology but in coin imagery also.43 Coins do not display mixanthropy, but a striking parallel is provided by Medusa, who has many points of similarity with Demeter Melaina.44 For example, both produce equine offspring after mating with Poseidon in horse form (Medusa’s horse-child is Pegasos, Demeter’s is Areion). For Medusa we have no evidence of worship, so we are not dealing with cult imagery in her case; but on a small but noticeable number of vases, she is given mixanthropic, half-equine form. Most famously, she appears as a centaur on a Boiotian relief-pithos (fig. 20).45 But there seems to be an anatomical alternative. On an Archaic Rhodian kylix, the Gorgons are shown pursuing Perseus.46 To one side is Medusa, who is distinguished from her fellow Gorgons by having the head of a horse instead of the usual hideous gorgoneion-face which they possess (see fig. 21). It does not seem that the vase showing the mare’s-head Medusa was intending to depict her after decapitation, with Pegasos protruding from the wound; we must therefore assume that they were treating Medusa as a mixanthrope, and as one very similar to the xoanon of Demeter Melaina. Clearly the mare’s-head type never became in any way canoni­cal; but its appearance might encourage tentative belief in the mare-headed xoanon at Phigalia. And in any case the mixanthropic image is of great importance as an idea in Phigalian legend, regardless of its questionable reality.

  • 47 Paus. 8.25.7.
  • 48 For Areion (under the alternative name Erion) on the coins of Thelpousa – a reflection of his impor (...)
  • 49 Jost claims that Demeter Erinys at Thelpousa would have been depicted mixanthropically before Eleus (...)

27It is easy to imagine the sacred cave of Demeter as a remote and sequestered site, of limited religious importance, but this seems not to have been the case. In fact, it was connected with other locations in Arkadia in a significant configuration. The closest link existed between Phigalia and Thelpousa, as has been said. Both Phigalia and Thelpousa generated cult myths which told of Poseidon raping Demeter while both were in equine form; they differ on the resulting offspring, with the Thelpousans claiming that they were the miraculous horse Areion and a daughter whose name might not be spoken,47 and the Phigalians that it was Despoina. It seems almost certain that the unnameable daughter is in fact the same personage as Despoina. Despoina is a cult title of the goddess in question; her real name is nefas to the uninitiated, as Pausanias tells us at 8.37.9. So in fact the real divergence between the two stories is the figure of Areion.48 It may at first seem peculiar that the mare-headed Demeter of Phigalia was firmly dissociated by the Phigalians from an equine offspring. However, a very plausible explanation is that in Phigalian myth Areion was displaced by Despoina because of the latter’s increasingly overwhelming importance, an importance which derived chiefly from the relationship between Phigalia and the sanctuary at Lykosoura, discussed below.49

  • 50 In ancient myth, Lykaon kills and serves up a human infant. In Pausanias’ version this occurs as a (...)
  • 51 In several accounts, Lykaon is punished for his child murder by being turned into a wolf, though in (...)
  • 52 See Bruit (1986), esp. 95; Borgeaud (1988), 34-44 (on Lykaion).

28In myth, there are also significant resonances between the cult of Demeter Melaina and that of Zeus Lykaios on Mount Lykaion. First both are associated with cannibalism. At Phigalia cannibalism is divine punishment for a human failure to maintain – among other forms of tendance – appropriate sacrifices; on Lykaion, the cannibal, Lykaon, is himself the one who trespasses against the gods, not by failing to sacrifice, but by sacrificing transgressively.50 Second, both cults have the motif of animal metamorphosis in their surrounding mythology, though of very different kinds.51 It has been argued that both reflect persistent Arkadian (rather than local) religious concerns: with transgressive eating, a slide back into savagery, and the thin line between human and animal.52 But the discernible implications of this for cult are limited. They do not suggest a connection on the level of ritual and practice between Lykaion and Phigalia.

  • 53 For example, it was because of its famous Despoina-sanctuary that Lykosoura was allowed to keep the (...)
  • 54 Especially suggestive is the text of a sacred law from Lykosoura (IG V², 514), which details certai (...)
  • 55 For the interconnected cults of Phigalia, Lykosoura, Lykaion and Megalopolis, see Jost (1994).

29Such a relationship does, however, seem to have been in place between Phigalia and another site, Lykosoura, a place of particular religious significance on a pan-Arkadian level.53 Lykosoura was dominated in religious terms by the figure of Despoina, Demeter’s daughter according to Phigalian (though not Thelpousan) legend. Numerous aspects of ordinance54 and iconography recall the central elements of the Phigalian cult. Of the latter, the most famous example must be the various animal-headed human forms found in the material remains from the sanctuary of Despoina. These mixanthropic images are generally described as referring in some way to Demeter Melaina’s mixanthropic cult statue at Phigalia, though the correspondence is not exact; relatively few of the Lykosouran figures are mare-headed, and their relationship to Demeter remains troublingly vague. Detailed discussion of this issue must wait till a later section (see below, Chapter 6). Here the most important point to note is that the Phigalian cave was not a cult site in total isolation, unique and untouched. Demeter Melaina’s remote cave, then, had in fact a part to play within some of Arkadia’s most visible and prominent cult activity.55

30Her horse-form mating and her horse-headed image set Demeter Melaina apart, as has been said, from the manifestations of the goddess in other regions of Greece. What significance, however, does the horse have in this instance?

31We have seen in relation to Cheiron that, generally speaking, horses can have two potential strands of significance. They can be emblems of aristocratic status, with an associated military and hunting-related cachet. They can also, however (and this form dominates mythology) be aggressive, destructive, and with a taste for human flesh. It will be argued that Demeter Melaina encapsulates many important aspects of the negative side of the horse, but that in her case these are discernibly shaped by the local conditions of belief and folklore.

  • 56 At the start of Book 8 (8.1.5) we are told by Pausanias that in fact the acorn-eating phase was not (...)
  • 57 Paus. 8.1.3.
  • 58 There is also a third, which pervades Arkadian mythology particularly: that is the life, and the fo (...)

32Phigalia was at the centre of an interlocking series of myths concerned with the progress of human society from savageryff to civilisation, a progress which is articulated in the myths through food-symbolism. The details of this articulation have been examined very thoroughly and to great effect by Bruit (1986). Bruit, however, concentrates in her article on the practice of the bloodless sacrifice to Demeter, and does not really look at the position of the horse element within the resulting schema. In the Phigalian myths, the foodstuff which chiefly represents the primitive state is the acorn, which the earliest Arkadians are said to have lived on; this is a food which occurs naturally, and is not the product of human agricultural activity.56 Then they learn to cultivate the land, and to produce grain; this breakthrough stands for their development into a civilised society. However, in Arkadian myth as a whole, there is an alternative primitive food to the harmless acorn: human flesh. Lykaon, the son of Pelasgos brings this alternative into the frame in the opening section of Pausanias’ Book 8 when he sacrifices a human child to Zeus Lykaios;57 as we shall see, it also features instrumentally in the Phigalian narrative. So we have two main possi­bilities for primitive fare, two opposing ways of representing and viewing the primitive stage of human development.58

  • 59 9.215 ff. For the significance of food in connection with a primitive modus vivendi, see Kirk (1971 (...)
  • 60 Indeed, he does not even eat the sheep he herds, and his care of them is verging on the tender.
  • 61 Just as there are various possible versions of primitive food in ancient thought (acorns; leaves an (...)

33The opposing expressions of the primitive life via food are of course not limited to Arkadian myth. Perhaps the most famous example of their juxtaposition is the figure of Polyphemos the Cyclops in Homer’s Odyssey.59 Polyphemos is a herdsman, and his day-to-day diet is a harmless, dairy-based one.60 But given the chance, he naturally and automatically turns cannibal, with an extra layer of transgression as he is eating those to whom he should offer ritual hospitality. Polyphemos brings together two food-options which the Phigalian narrative is essentially trying to keep apart, as will be explained.61

34So how does Demeter Melaina fit into this theme as it is expressed in Phigalian myth? Not (and this is the most striking point) as the embodiment of grain and agriculture, the ‘finished’ phase of the human process. This despite the fact that we are told explicitly that the discovery of agriculture was facilitated by her. Rather, she embodies both of the two chief manifestations of the primitive, the pre-grain phase.

  • 62 Paus. 8.42.11.
  • 63 Paus. 8.42.5-6.

35In the first place, the bloodless sacrifices offered to her by long custom consist of the harmless primitive foods: various fruits and goods (grapes, honey, raw wool),62 from which grain is notably absent. But according to the myths, if these rites are neglected (as they were following the destruction of the mare-headed xoanon),63 primitive food-type number two comes into play. Demeter Melaina’s punishment for the lapse of the cult on that occasion was to cause akarpia – a lack of food, famine – but moreover, this was a famine that had a particularly extreme possible outcome: a resort to cannibalism and child-eating. The vital section is the text of the oracle given by the Pythia to the Phigalians when they send to Delphi for a cure for their suffering.

  • 64 Paus. 8.42.6 (for the Greek text, see the Appendix below).

Azanian Arkadians, acorn-eaters, who live
in Phigalia, the cave in which hid Deio horse-bearer,
you have come to learn a cure for painful famine,
who alone have twice been nomads, who alone have been eaters of wild fruits once more.
Deio made you cease from pasturing, Deio made you pasture
again, after being binders of corn and eaters of cake,
because she was deprived of the prize given by former men, and ancient honours.
And she will quickly make you eaters of each other, and eaters of children,
if you do not assuage her anger with public libations,
and adorn the recess of her cave with divine honours.64

  • 65 Jost (1992 b, 56) relates this in part to Demeter’s mountain location and affinity with the mountai (...)

36Demeter (here called Deio) led man from the stage of herding, from being agriodaitai, to the cultivation of grain; but if not accorded the proper offerings, she has the corresponding power, not just to undo that stage, but to push them into the unacceptable form of primitive life, cannibalism (especially the ultimate taboo, eating their own children). This is the outcome represented by the mythi­cal figure of Lykaon, just as the harmless life of herding and acorn-eating is embodied by Pelasgos. And Demeter is the deity who can undo the process of civilisation completely.65 If not placated by the food of harmless savagery, her punishment is to inflict savagery of the worst kind. Of course, the sacrifices to Demeter Melaina are also – as well as being grainless – bloodless, avoiding flesh-consumption altogether. We see from the myth of Lykaon, perhaps, that in Arkadian myth especially, any sacrifice, even when the victim is ostensibly animal, has the potential to turn out to be human. The Phigalian rite bypasses this possibility entirely.

  • 66 Bruit (1986), 85.

37So, Demeter Melaina is a kind of inverted Demeter, one whose chief potential rôle seems to be to blight crops rather than safeguarding them. Her shrine is in the wild, uncultivated land; she is, in the words of Bruit, a ‘force sauvage de la terre.’66 But what of the horse in her myths and in her cult image? How – if at all – does that animal fit in with her persona as described?

  • 67 Scheffer (1994) argues for a more specific association between horse-related deities and uncontroll (...)
  • 68 See Borgeaud (1988), 16-17.

38Various aspects of the associations borne by the horse in Greek myth have been outlined already with regard to Cheiron, and surface, I believe, with regard to Black Demeter also. The first, perhaps the most fundamental, is the ambiguity of the horse’s mythical relationship with, and position within, human culture and civilisation. The horse is a domesticated, vegetarian quadruped that works on man’s behalf; but this side of its nature can vanish in a flash, to be replaced with the qualities of a thêr, a wild beast. The horse can never be relied upon, in myth, to play its part consistently in the wider picture of man’s taming influence on nature.67 Therefore it is highly appropriate that Demeter Melaina, who represents above all the fragility of human civilisation, its potential for instant reversal,68 should have the horse as the chief component in her representation.

  • 69 See Ant. Lib. Met. 20.
  • 70 ibid. 7.

39Moreover, as we have seen, when a horse in myth does depart from its domestic aspect, it feasts on human flesh; that is the unwavering outcome. Once again, we may see the connection with Demeter. The horse goes to the extreme of savagery, its worst, most destructive aspect; it is with this very contingency that Demeter threatens her worshippers. The connection grows stronger when we note also that very often in myth a horse turning man-eater is the instrument of divine vengeance for a religious – particularly sacrificial – transgression on the part of the humans involved. We might think, for example, of the myth of the sons of Klinis, and their taboo ass-sacrifice to Apollo.69 More generally, the aggression of horses is vengeful: Glaukos refuses to let his mares breed, thus breaking a rule of nature and enraging Aphrodite; Anthos tries to drive his father’s horses out of their meadow, enraging them.70 Arkadian Demeter too is characterised by vengeful anger, both in a specific and a general way. The specific sense relates to religious failure, the lapsing of her cult; but themes of anger and revenge suffuse her myths more widely. She is angry at Hades for the rape of her daughter; angry at Poseidon for her own sexual misuse; angry at her worshippers for their neglect. Anger and the potential for subsequent punishment appear to dominate her persona. The horse is a fitting ingredient of the expression of this motif.

  • 71 Paus. 8.34.3: καὶ οὕτω ταῖς μὲν ἐνήγισεν ἀποτρέπων τὸ μήνιμα αὐτῶν, ταῖς δὲ ἔθυσε ταῖς λευκαῖς.

40This theme of anger and vengeance is shared equally between the two Demeters of the region, Melaina and Erinys. In fact, the latter raises a special point of interest with her name, which seems to be the most explicit reference we have to the central importance of the theme to her persona: after all, the pluralized Erinyes are the embodiment of the drive to exact punishment for a transgression. The motif may be reflected in the title of Melaina as well as that of Erinys. Pausanias tells us (8.42.4) that the Phigalians explained the name by reference to the myth that Demeter put on black clothing when angry at the snatching of Persephone and at her own rape. So, black is an indication of anger; and this is very reminiscent of another cult in Arkadia, that of the Eumenides, near Megalopolis. The Eumenides are the same as the Erinyes, and the cult in question is connected with a unique variant of the well-known myth of Orestes and his madness, a variant which introduces the idea of colour. When the Eumenides are pursuing Orestes in their vengeful fury, they appear to him black; after he has expiated his crime by biting off his own finger, they change to white. There follows in Pausanias’ account a revealing statement:71

So he made an offering (enêgisen) to the black goddesses to avert their wrath, and sacrificed (ethuse) to the white ones.

  • 72 For full analysis of the use of such terminology in ancient texts, see Ekroth (2002).
  • 73 Pirenne-Delforge (2008a), 187-241; see esp. 229-234 for the sacrifice by Orestes.
  • 74 Pace Scheffer (1994), 128-9.
  • 75 Paus. 8.34.2.
  • 76 Pirenne-Delforge (2008a), 233.

41The verbs used here to designate the two sacrifices, enagizein and thuein, are significant, in a way which no English translation can capture. Enagizein is especially conspicuous in this context. Although in ancient texts across the board it does not have a wholly consistent meaning,72 in Pausanias, as Pirenne-Delforge has shown, it almost always refers to sacrifices made to dead heroes (though its use is by no means simple and without variation), and, what is more, occurs in many instances where the contrast between heroes and gods is being explored.73 Therefore its use with regard to divinities like the Eumenides is striking. What it would appear to emphasise is their link with death and the underworld,74 a link reinforced by the proximity to the shrine of the ‘tomb of the finger’ (supposedly the final resting-place of Orestes’ severed digit).75 Enagizein also carries with it a sense of the deflection of a baneful and destructive power.76

  • 77 There is a connection between anger/grief and blackness in the characterisation of Thetis, too, as (...)

42Given the bundle of associations that Demeter and the Eumenides have in common (anger at a transgression; the need for ritual placation, the name Erinys; the importance of colour; the underworld connections), we are surely justified in regarding Black Demeter as a goddess in angry, vengeful mode.77 Offerings to her are rites of aversion, though they are very different in nature from the typical form of the enagismos (a holocaust). Demeter Melaina requires constant appeasement; but the means of doing so have to tally with the food-symbolism at work in the mythology discussed above.

  • 78 In the myths surrounding the cults at Phigalia and Thelpousa, metamorphosis into horse form is used (...)
  • 79 For the chthonic aspect of Poseidon Hippios, see Detienne and Werth (1971), 167-8.
  • 80 Though Scheffer (1994, 128-9) argues that it is less important than has often been claimed.

43So the horse-element of Demeter and her blackness combine to express her quality of anger and the potential hazards if suitable ritual placation is not maintained.78 We are left with the question of how the associations of the horse in her case fit in with those of Poseidon Hippios, which appear to be substantially different. For Poseidon Hippios, the horse-element is closely connected with the fertility-aspect of his nature, especially in association with water. As is usual in Greek religion, the fertility aspect has its underworld associations also.79 It would be perverse to argue that Demeter had no share in this quality.80 But whereas her horse-representation is underpinned by the motif of vengeful anger, in Poseidon’s case this feature is completely absent.

44Thus the horse’s significance in the case of Demeter Melaina is completely at odds with Cheiron, horse-bodied but man-headed, who, with his involvement in paideia and his concern for the welfare of humans, works to preserve human civilisation in the face of various aggressors, including his own fellow-centaurs. We learn from this that an animal ingredient does not by itself confer one simple and immutable significance. Rather, its import depends on the way in which it is anatomically disposed, and on regional religious and mythological context. The horse does have some consistent traits in ancient thought, but they are distinctly ambiguous, and how they are reflected in the character of the deity concerned is intensely, and fascinatingly, variable.

2. From Arkadia to Attica: Pan

45Pan’s is a cult whose diffusion, from the Classical period on, is so wide as to render a full survey of sites and practices impossible here. That said, two chief foci emerge from the wider picture: Arkadia and Attica. Arkadia is highly significant as being the place where Pan’s worship began, and Attica was the region which later adopted it most enthusiastically, and adapted it most revealingly.

  • 81 On the ‘reinstitutionalization’ of Pan in Arkadia after – and massively fuelled by – his diffusion (...)

46That said, one could at this point be accused of taking on and perpetuating a vicious dichotomy. There is no doubt that, as Borgeaud has shown, Athens and Arkadia were involved in a complex dialogue of identity and self-exploration, driven at first mainly by Athens but, from the fourth century at least, engaged in actively by the Arkadians also in their bid for increased cultural self-definition.81 Athens used Arkadia as a model for various concepts: her own past, the primitive, the Other, and so on; this model was then taken up by the Arkadians and applied to themselves. So by concentrating on these two areas, to the exclusion of countless other regions and their local Pan-cults, are we not simply perpetuating a dualistic fiction generated in antiquity and dominant ever since? Athens and non-Athens – this is in many ways a pernicious schema which must be challenged by a recognition of local diversity across Greece, for which (given that textual sources are almost always themselves the products of that schema) archaeological evidence is especially valuable.

  • 82 See Larson (2001), 97, on the rôle of Athens in disseminating the cult of Pan through Attica and th (...)
  • 83 For an example one might look at a case already discussed, the deities – among them Cheiron – inclu (...)

47In most cases, this would be an unassailable truth. In the case of Pan, however, local diversity is hard to find for the simple reason that Pan cults throughout Greece follow very closely the Attic pattern in terms of imagery and practice. It might plausibly be said that, as Athens was the first region to take Pan out of Arkadia, so she mediated Pan’s arrival on the wider Greek scene. Many other Greek states, at least outside the Peloponnese, received the god not directly from his Arkadian homeland but from Athens.82 Naturally, local adaptations of the Athenian way occurred.83 But the Pan they took on cannot be understood as an unmediated Arkadian phenomenon. For this reason, the focus on Arkadia as place of origin, and Attica, as the region in which the vital latter stage of development took place, appears inevitable and justified.

48When we look at the differences between his Arkadian and his Attic worship, Pan’s case starts to chime very strongly with others, particularly those of Acheloos and Cheiron. We shall summarize the worship in each region before analysing their divergences.

  • 84 Jost (1985), 458-9.
  • 85 Paus. 8.36.7; Jost (1985), 200, 458-9
  • 86 Paus. 8.30.2-3; Jost (1985), 221-2, 458-9.
  • 87 ibid. 458.
  • 88 Herbig (1949), 15. This notion also pervades Lamb (1926), who says of the kriophoros-type effigies (...)
  • 89 See e.g. Hübinger (1992, 203-6; 1993, 25-6) on the finds from Lykaion and what sort of people may h (...)

49The obscurity of evidence surrounding Pan’s cult in Arkadia is to a huge extent dispelled by Jost’s treatment, which brings out particularly strongly, as ever, the vital relationship between a cult and its topography. Pan was accorded, on the whole, small shrines in mountainous areas,84 though occasionally he is to be found in an urban setting, as in Peraitheis85 and Megalopolis.86 His cult generally links him with the same types of landscape as do his myths – the ‘paysage panique’ as Jost calls it.87 Though Herbig’s picture88 of his worshippers as all simple shepherds has now been questioned,89 there is no doubt that on the whole he received small private offerings and presided over the concerns of the herdsman and hunter.

  • 90 Initial excavation reports: Kourouniotis (1902), (1903) and (1904).
  • 91 Paus. 8.38.5. Pausanias says that in his day these games were no longer celebrated there, a stateme (...)
  • 92 Jost (1985), 183-5,267-8; id. (1994), 226.
  • 93 For a clear example, see Jost (1985), pl. 63, no. 4.
  • 94 The precise date of the synoecism is uncertain. For the event and its context of growing Arkadian s (...)
  • 95 Paus. 8.38.5.
  • 96 On Megalopolitan doublets, see e.g. Jost (1994), 227.
  • 97 Paus. 8.30.2. Also at Tegea: Paus. 8.53.11.

50However, there are sites which deviate from or go beyond this rôle, and the most striking is that on Mount Lykaion.90 The sanctuary of Zeus on Lykaion was an exceptionally important religious site in Arkadia, and wielded a pan-Arkadian scope and sway; it was the location, for example, of games – the Lykaia91 – which brought in competitors from the whole region. Its significance to Arkadian identity from the fourth century on92 is demonstrated by the use of Lykaian iconography on Arkadian federal coinage,93 and by the inclusion of Lykaian religious elements in the programmatic new centre of Megalopolis, created by synoecism in around 370/69 BC.94 This was therefore no unimportant zone for Pan to appear in, and by no means the haunt only of a small number of local peasants. The sanctuary of Pan has not been found, but Pausanias tells us95 that it was sited by the race-track used in the games (a prominent location) and had around it a grove of trees. In addition to the bare fact that they shared a cult location, there are other indications that Pan and Zeus Lykaios were in a dynamic religious relationship of which their worshippers were fully conscious. On the coinage cited above, they appear together, each on one side of the same coin. In Megalopolis, which to a large extent mirrored the religious arrangement on Lykaion,96 an image of Pan stood within the abaton of Zeus Lykaios.97

  • 98 Schol. Theok. 1.5.123c.
  • 99 Jost (1985), 474-5.
  • 100 Paus. 8.37.11: λέγεται δὲ ὡς τὰ ἔτι παλαιότερα καὶ μαντεύοιτο οὗτος θεός, προφῆτιν δὲ Ἐρατὼ νύμφη (...)

51There is also evidence that suggests that Pan had an oracular function on Lykaion. It is true that this evidence is by no means unproblematic. The sole source for a manteion of Pan on Lykaion is a scholion on Theokritos,98 though as Jost remarks, other texts make more general connections between Pan and prophecy and the link is indisputable.99 In his description of Lykosoura, when describing an effigy of Pan, Pausanias states: ‘It is said that at a yet earlier time, this god too gave oracles, and that the Nymph Erato became his prophetess, she who married Arkas the son of Kallisto.’100 With these words, which appear to apply not only to Lykosoura but more generally, Pausanias consigns any actual oracular function to the realm of myth and the distant past. It is therefore hard to assess at what stage prophecy was a significant part of Pan’s cultic function on Lykaion.

  • 101 Serv. Verg. Geor. 1.5.17. For discussion of the passage, see Jost (1985), 474-5; Larson (2007), 63.

52More generally, Pan’s position in Arkadian religion seems to share with that of Cheiron in Thessaly a certain power to keep at bay the malign natural forces of the landscape. For example, Servius says that Pan’s epiklesis was Lyceus because he keeps wolves – lykoi – away from the flocks.101 This has all the hallmarks of a bogus etymology, but it does echo some of the most potent names in Arkadian mythology, names which draw on the intense symbolic valency of the wolf. In this context, wolves are not merely real-life threats to young lambs. As has been said, Lykaon, with his wolf-metamorphosis, represents the transgressive past, a past into which humanity always risks lapsing; the cult site of Mount Lykaion, on which Pan had his shrine, was in essence a monument to this terrible possibility. Moreover, it is Pan who, in Phigalian myth, leads Demeter Melaina out of the condition of angry seclusion which has the potential, by crippling agriculture, to push man back into primitive savagery. There appears, therefore, to have been a loose bundle of religious ideas concerned with ‘keeping the wolf from the door’ – or sheep! – and Pan seems to have been a helpful figure in this regard. This is not dissimilar from Thessalian mythology which pits Cheiron against the destructive group-centaurs in the glens of Pelion. Both cases illustrate the apotropaic usefulness of the man-friendly monster.

  • 102 For this encounter, see Hdt. 6.105. See Parker (1996), 163-5 for the introduction and dissemination (...)
  • 103 See Borgeaud (1988), 134-5. Borgeaud argues convincingly that Pan’s key character in Athens was as (...)
  • 104 For the torch, its rôle in the Persephone myth and its fertility-associations more generally, see P (...)

53It was this deity, so firmly woven into the fabric of Arkadian mythology, who during the Persian Wars, in famous circumstances, made the jump to Attica and Athens. Philippides’ encounter with Pan appears really to have marked the inception of his Attic worship.102 The cult site which followed directly from the epiphany was a cave on the north-west skirt of the acropolis, right in the heart of the city. Here Pan was honoured with torch races, which clearly echo the scenario of his initial manifestation.103 It is possible that the torches also echo those connected in art with the return of Persephone and the restoration of natural abundance, a restoration over which Pan and Panes more than once preside.104

  • 105 Examined in great detail by Borgeaud (1988); the present section is merely a summary of some vital (...)
  • 106 The nymphs already had cult in Attica before the arrival of Pan, but Parker (1996, 163-5) argues th (...)
  • 107 Documents illustrating this are assembled and discussed by Borgeaud (1988), 140-43, where he also t (...)
  • 108 In fr. 96.2 Snell, he calls Pan the ‘dog of the Great Mother’.

54If so, this would certainly not be out of place with the wider picture of Pan’s typical character in Attic cult, outside the urban centre of Athens.105 The form which that cult usually takes is near-identical with that of Acheloos in Attica. A number of deities are worshipped on a single site, normally a cave; they are often depicted together, and named, in a votive relief. Their number includes the nymphs most frequently,106 Pan also very often, with Hermes, Acheloos and Dionysos sometimes present, and a number of other less frequent instances such as Herakles. Sometimes a connection is also made with the goddesses of Eleusis, and the fertility rôle of Pan, like that of the nymphs, is explicitly expressed.107 In addition, Pan is often linked with Cybele, the Mother of the Gods, and takes a place in her retinue, but this is as much a Boiotian as an Athenian allegiance, and goes back at least as far as Pindar.108

55Overall, Pan appears as one of a company, and as a component within a number of canonical groupings: Eleusinian, Dionysiac, korybantic. This is not a pattern unknown in Arkadia, but it is very different from the duality (albeit unequal) of Pan and Zeus on Lykaion. We might compare the difference between Cheiron in his cave on Pelion, and in his cave near Pharsalos: in the former his function and character are perhaps more mysterious, but he emerges as an important half of a divine double-act; in the Pharsalos cave, a later arrangement, he has joined a rustic collective. It is this collective persona that Attic Pan tends most often to have.

  • 109 Parker (1996, 165) remarks that ‘the new cult was not a simple re-creation in Attica of what had be (...)
  • 110 For discussion of all that Pan meant to the Athenians, wildness and animality, see Parker (1996), 1 (...)
  • 111 Borgeaud (1988), esp. 50-52. The function of the cave on the acropolis is particularly interesting; (...)

56Several scholars have noted the key difference between Pan before and after the diffusion of his cult: the importance of the cave. In Arkadia Pan is almost never worshipped in a sacred cave, yet as soon as his cult moves beyond his homeland this becomes the setting par excellence.109 Borgeaud’s famous thesis is surely right: that by placing Pan in a cave, other Greeks, Athenians especially, were placing him in a small segment of the Arkadian landscape with which he was so closely associated;110 he was thus always surrounded by his imagined Arkadian identity.111 However, this does not do justice to another, wider truth: the consistent and significant relationship between caves and mixanthropes. This relationship might at first appear to be the result merely of the cave’s character as rustic locale; after all, mixanthropic deities tend to be gods of the countryside. But, as I shall argue more fully later, there is more to it than that.

  • 112 For example, Herbig’s monograph covers almost the whole swathe of antiquity; for treatment of later (...)

57On the face of it seems superfluous to describe Pan’s physical form. Of all the mixanthropic deities included in this study, he is the one whose image was and is the most widespread, the most familiar, and the most influential in shaping both the ancient and the modern perception of mixanthropy. The long-lasting appeal of his representation is reflected in the tendency of the scholarship on the subject to take an extremely wide temporal range, encompassing at least Greek and Roman material if not Renaissance and modern as well.112 It is not, however, without controversy.

  • 113 In LIMC s.v. ‘Pan’.
  • 114 Fig. 29; LIMC s.v. ‘Pan’, cat. no. 1; see Herbig (1949), 51-3; Brommer (1949-50), 6-7.

58Chief among the uncertainties is the question as to when we find our first identifiable Pan-images. Both Herbig and Boardman113 tentatively include in their collations of Pan-images a number of very early figures which are essentially goats standing on their hind legs. (Jost, interestingly, does not; her catalogue starts with Classical period objects.) The first example of this form (dated as it is to somewhere in the seventh or sixth century BC) is fig. 22, the small bronze group of four animal figures standing in a circle with hands joined, from Methydrion in Arkadia.114 A small number of other early objects follow this type: the upright goat-man, or sometimes a group of such beings. Herbig suggests that we may see these as fore-runners of the being later designated Pan, but this is fraught with uncertainty: the Methydrion group demonstrates the difficulty even of identifying species with any accuracy. That said, a small number of Attic black figure vases of the early fifth century do provide examples of a Pan-form which may support the pedigree of such objects of the Methydrion group: in these, the god is shown essentially as a goat on his hind legs. (See figs. 23-24.)

  • 115 Herbig (1949), esp. 53-7.
  • 116 For a typical Classical Pan, see fig. 30.
  • 117 Hübinger (1992).

59In any case, the Classical period, as Herbig notes,115 sees the dramatic arrival of the first recognisable and identifiable mixanthropic Pans, and also with the clear partitioning of animal and human body-parts. The change from upright animal to mixanthrope is extremely important. As Herbig remarks (p. 51), the former type is ‘Nie Bock plus Mann, der halb Gott halb Tier, sondern immer göttlicher Bock oder bockhäfter Gott.’ By contrast, the Pan of the Classical period, of vase-painting first and foremost, belongs anatomically to the same basic class of creatures as other early Mischwesen and particularly the satyrs whose physical composition resembles his own fairly closely.116 There is no doubt that he becomes part of the artist’s popular mixanthropic palette. That means the Athenian artist first and foremost. This raises another fundamental difficulty attendant on Pan’s representation: the impossibility of giving it any secure anchor in Arkadia, the region which was undoubtedly the starting-place of his cult. The earliest known site of his worship in Arkadia, on Lykaion, has yielded no mixanthropic Pan-images.117 Only after Pan becomes popular as a character in art outside Arkadia (chiefly in Attic vase-painting) does his image also start appearing regularly inside Arkadia. We must certainly resign ourselves to the probability that the iconic mixanthropy so universally associated with the god is not an entirely accurate reflection of early Arkadian belief. That said, it swiftly attained iconic status within Arkadian cult and iconography; Athens can be said to have given Arkadia back her own god.

  • 118 An example of the ‘young and pretty’ type is LIMC s.v. ‘Pan’, cat. no. 36: an Apulian red figure kr (...)
  • 119 Numerous other examples come from Arkadia. Examples are given by Jost (1985), 464-6. Several of the (...)
  • 120 This latter medium tends to consist of the same nymph-reliefs in which Acheloos (see above) figured (...)

60Pan’s mixanthropic form, once it has emerged, dominates completely, though it continues to develop. One major development is the increasing popularity of young, attractive Pans who are, necessarily, more anthropomorphic, especially about the face.118 Cult images of Pan are generally less given to innovation than ‘secular’ ones, however, and a strongly consistent mixanthropic type was established, with goat or goatish head, usually a goat’s legs and feet, human arms and body, and very often holding a musical instrument or the lagobolon. This type is repeatedly employed across a large number of locations, in sculpture in the round (for example fig. 25)119 and in relief.120

  • 121 Herbig (1949), esp. 33-4.
  • 122 The conventional, pastoral etymology is discussed by Larson (2007, 63) as being the most resonant a (...)
  • 123 An example of the earlier, simpler view is that of Herbig (1949, esp. 15-18), who imagines Pan crea (...)

61So what does the god’s goat-element contribute to his character? In the most nebulous sense, the animal – like the bull and even perhaps the horse – has associations with fertility and sexual energy, both animal and human.121 In Pan’s case, however, these associations are given a more concrete form in his firm connection both with herdsmen and with their animal charges. His name, though its etymology is not unquestioned, would definitely have ‘spoken’ of the pastoral life and its concerns.122 Pan embodies both the human and the animal side of a particular way of life, that of animal-herding. Older scholarship has seen this embodiment as emerging from within the pastoral society itself; rather later views see it as a personification imposed on both Pan and Arkadia from outside.123 But whichever line one takes, there is no doubt that Pan’s goat element is inextricably linked with his character of patron god of shepherds and guardian of their concerns.

  • 124 His monstrous quality is expressed in the phrase τερατωπὸν ἰδέσθαι, ‘like a teras to behold’.
  • 125 Hom. Hymn 19.38-9. One might compare the terror of the nurses of Erichthonios when they see his hal (...)
  • 126 Hom. Hymn 19.45-7.

62The company that Pan keeps in Classical and post-Classical art reflects his fertility associations. He is worshipped and depicted alongside the nymphs, Hermes, Dionysos and Acheloos, especially in Attic votive reliefs which we have already examined with regard to the river-god. In other words, he joins an existing iconographic and cultic canon, in which his strange dual form and his goat elements are by no means incongruous. They are simply incorporated into an expanding collective of rustic curiosities. For our ancient sources are not unaware of the disturbing and grotesque aspects of Pan’s mixanthropy and animal parts. This is very much to the fore, for example, in the Homeric Hymn to Pan (lines 35-47), which makes much of the shocked reaction to the birth of the ‘monster’124: the nurse runs from him in fear,125 though his strangeness delights the gods on Olympos.126 This fear-inspiring quality of Pan’s mixanthropy is not explicitly connected in ancient sources with his ability to inspire the panic which bears his name, but one might suppose them to be related. At any rate, alongside the positive associations of Pan’s goat parts (fertility, playfulness) there existed an underlying awareness of their potential to disturb and disquiet.

  • 127 For example, the second line of the Homeric Hymn to Pan calls him aigipodes and dikeros, goat-foote (...)

63The composition of Pan’s mixanthropy reveals some interesting trends which are reminiscent of other instances already discussed. When they mention his mixanthropy at all, literary sources tend to focus on his horns, face and hooves, the same parts in which animality resides in his material depictions.127 This is also reflected to some extent in the material evidence; though his feet may be human depending on time and context, horns are always present even when all other theriomorphic components are not. The importance of and emphasis on animal horns and feet is to be discerned also in the case of the horned Dionysos, discussed below, even though there the animal species is different. The significance of feet is hard to pin down, but horns are important for a number of reasons. They are the non-human feature par excellence, and by themselves are enough to render a being mixanthropic. They suggest power, aggression, and sometimes fertility. They are laden with possible echoes, all of which may have been present in the case of Pan. His mixanthropy is neither chaotic nor random; it belongs to a type which is occupied also by the horned Dionysos and by his mixanthropic followers, the Silenoi and satyrs.

  • 128 Luc. Dial. Deor. 22.1: How can you be my son, asks Hermes of Pan, ‘κέρατα ἔχων καὶ ῥῖνα τοιαύτην κα (...)
  • 129 See Boardman (1997), 32-3. On the motif of the anodos of Persephone in vase-paintings, see Bérard ( (...)
  • 130 One could argue, though without the possibility of proof, that the fact that Pan’s face is so frequ (...)

64The final aspect which is repeatedly alluded to is the animal face of the god, a feature sometimes shared again by Dionysos. This can be made a subject of derision, as in Lucian, where Pan’s nose is one of a list of grotesque features;128 sometimes in art, too, there is the possibility that an artist is making use of its humorous or grotesque potential, as in the red-figure vase which depicts a pair of Panes reacting to the return of Persephone, with open animal mouths and staring eyes which it is hard not to view as comic, in conjunction with their wild leaping (fig. 26).129 But the grotesque or comic quality is not usually brought out. The animal shaping of the face varies in extent, but is a highly significant aspect of mixanthropy, having considerable implications for identity. In terms of human ritual, including theatre, masking the face, more than any other form of costume, is the quintessential way of effecting a transformation into another being, even across gender or species.130

  • 131 Epimenides fr. 16 DK.
  • 132 For scholarship on the topic of cannibalism on Lykaion, both mythical and ritual, see above, n. 51.
  • 133 Paus. 8.42.3. The emphasis in this passage is on Pan’s rôle as a denizen of the mountain realm: he (...)

65Dominant though the goat is in his character and physical form, Pan illustrates the possibility for mixanthropy to involve a multiplicity of species. First, Pan is enmeshed in a complex web of animal/god and animal/human relationships in Arkadian myth. According to one variant,131 his mother is Kallisto, which makes him the half-brother of Arkas, with two effects: first, to tie him into the foundation-mythology of Arkadia and place him at the start of its history, and second, to involve him indirectly in the famous story of Kallisto’s transgression and bear-metamorphosis, and the birth of an offspring called Arkas – ‘Bear’. His association with Mount Lykaion links him with Lykaon’s wolf-transformation, and the rumoured ritual reality of cannibalism and were­wolfism.132 Finally, he is involved in Demeter’s horse-metamorphosis, or rather its aftermath, coaxing her out of her anger.133 All these stories attempt to negotiate the states of human and non-human, or of divinity and beast, using episodes in which those states are dangerously transgressed or combined; and in every major instance, Pan is there on the fringes.

  • 134 Hyg. Fab. 206; Hyg. Astr. 2.28; Opp, Hal. 3.15; Luc. de Sacr. 14.
  • 135 For the title Haliplanktos, see Soph. Ai. 695; for Aktios, see Theok. Id. 5.14. Of course, we have (...)
  • 136 Hyginus for example (Astr. 2.28) derives Aigipan’s fish element from the fact that he hurled sea-sh (...)

66In addition to these various animal-metamorphosis associations, Pan’s own animal element is occasionally flexible. An example is his mysterious connection with the fish. This crystallises in the person of Aigipan, who is sometimes but not always treated as a separate entity from Pan; however, there is no doubt that the two have common roots. In our (rather late) sources, Aigipan turns himself into a goat-fish hybrid to escape the murderous attack of Typhon; it is in this form that he is catasterised as Capricorn.134 All actual hybridism and metamorphosis involving fish elements is restricted to Aigipan, but faint echoes are also discernible in Pan, through two epikleseis: Haliplanktos (‘Sea-roaming’) and Aktios (‘Of the shore’).135 These are marine rather than specifically fish-connected but connect significantly with the Aigipan material. They also seem as odd in the depths of Arkadia as the fish-tail of Eurynome, with which it is interesting to compare them. No explanation suggests itself, though ancient authors were not always so fatalistic and subjected the matter to rationalistic scrutiny.136 It is simply advisable to recognise that, with Pan as with so many divine mixanthropes, a dominant animal species in his composition co-exists with more shadowy alternatives which never achieve the same prominence.

3. Mixanthropy at the heart of Athens: Kekrops

  • 137 He is described as the first king of Attica by Apollodoros (Bibl. 3.14.1), and this is also strongl (...)
  • 138 See Kearns (1989), 110.
  • 139 It should also be noted that the daughters of Kekrops, Aglauros, Pandrosos and Herse, were themselv (...)

67As we have seen in the previous sub-chapter, it is a mistake to think of mixanthropes as occupying only an extra-urban cult-setting, and of being always divorced in location and function from the zone of civic life. A mixanthropic deity whose position on the Athenian acropolis is obviously significant for the understanding of his character is Kekrops, whose mixanthropic form was that of a man with the coiling body and tail of a snake from the waist down. Kekrops was considered an early king of Attica, normally the first, although ancient authors differ somewhat about chronology and succession.137 He is so closely bound up with Erichthonios and Erechtheus (who are themselves thoroughly intertwined138) that at times it seems perverse to separate them. However, he is selected for this study for two reasons: first, only Kekrops is consistently mixanthropic, and second, only for Kekrops do we have definite mention of worship. However, it should be recognised that in many ways he formed a close-knit group with Erichthonios and Erechtheus, especially on the level of mythology.139

  • 140 Paus. 9.33.1. The Kekrops Pausanias mentions is perhaps a rather different figure from the one wors (...)
  • 141 See Kearns (1989), 160. Gourmelen (2004) argues strongly that Kekrops’ rôle as eponymos constituted (...)

68The worship of Kekrops is an almost exclusively Attic phenomenon; the only other concrete mention is of a cult in Haliartos.140 The two chief branches of his cult were within the city of Athens itself: on the acropolis, and, as eponymos, in the agora. The latter is the less certain, resting almost entirely on a mention of a single dedication in the fourth century.141 It shows the use of a mythical figure to spearhead an aspect of political identity and loyalty, and reflects a particularly Athenian institution.

  • 142 Plut. Thes. 36.1-2.

69The cult on the acropolis is more varied and in some respects more interesting for the current study. It is another case of mixanthropic Totenkult, based on a supposed tomb; it has been suggested that the basis was a Mycenaean grave which came to be venerated as that of a dead hero. In any case, Kekrops undoubtedly belongs to the canon of heroes whose worship was an augmented form of regular grave-offerings, a canon widely associated with snakes, though not of course exclusively. The shrine of Kekrops on the acropolis clearly belongs to a type of hero tomb of civic importance discernible throughout Greece; the most closely analogous Athenian example is perhaps the tomb of Theseus. According to Plutarch, the remains of Theseus, brought back to Athens by Kimon, were established in a tomb near the gymnasium, another central and vital city location, where the hero’s cult is said to have had broad appeal.142 Though the akropolitan location had its own political and religious resonances, there is no detaching it from the workings of the Athenian polis.

70If Kekrops’ tomb earns him at least partial inclusion within a certain type of cult-receiving hero, there is another dimension of his Attic worship which may connect more closely with his mixanthropy: in Euripides’ Ion (l. 1400) we hear of a sacred cave, also on the acropolis. Was this a true cult site, or was Kekrops’ name simply applied to a topographical feature? We cannot say, for the evidence of offerings, dedications and so on is lacking; but we may draw a comparison with the periakropolitan cave of Pan, which placed the god at once in the civic heartland and inside a microcosm of his imagined native environment. A significant number of mixanthropic deities appear as residents of caves, both in myth and in cult, and it is not surprising that Kekrops should be one of them.

71So in some important ways Kekrops builds on patterns which recur among mixanthropic cult-recipients, and his contribution in this regard is valuable. However, these should not encourage us to ignore the locally specific manifestations of these patterns, and the particular symbolic rôles they place within a particular community. So often this relationship between mixanthrope and community is made obscure by lack of evidence, but the relative abundance of material from Athens allows Kekrops’ symbolic function in Athenian myth to emerge. Most strongly represented in the ancient literature is his link with the all-important theme of autochthony.

  • 143 See also Rosivach (1987) and Parker (1987, 193-207) for further analysis of the political applicati (...)
  • 144 Hall (1997), 51-6.

72The Athenians claimed to have been born from their very land, thereby pinning their projected identity on the idea of the entirely indigenous – or at least they did so in certain times and contexts. As Hall has noted, such claims gained prevalence in the fifth century, after the Persian invasions had reduced the appeal of Ionian origins (though the two traditions were to some degree reconciled).143 In this situation, the Athenians were distinguishing themselves from other Greek communities whose origins were ‘barbarian’; they were also choosing not to stake a claim within the wider Greek genealogy but rather to strike out on their own as uniquely uncontaminated products of their native land.144

  • 145 See e.g. Hyg. Fab. 48; Ant. Lib. Met. 6.
  • 146 Fascinatingly, a less strongly attested variant tradition makes Kekrops Egyptian (see Philochoros, (...)
  • 147 Apollod. Bibl. 3.14.6.
  • 148 As Loraux (2000, 41-2) notes, there is in fact an implicit underlying analogy between Pan and the A (...)

73Kekrops provides an embodiment of these claims through his own autochthony; he himself is literally gêgenês,145 even though mythology does not dwell on his birth from the earth as it does in the case of Erichthonios, and in this regard he is the ultimate native.146 He also, however, has a rôle to play in another episode which expresses the theme: the birth of Erichthonios. Erichthonios is gêgenês through special circumstances: the earth is fertilised by the seed of Hephaistos which he lets fall in his attempt to rape Athene. Kekrops presides at the birth, and his daughters, Aglauros, Pandrosos and Herse, undertake the care of the child.147 He, as one being who emerged from the earth, helps to facilitate the emergence of another. As autochthôn, Kekrops encapsulates the mythical past of the Athenians, and it is interesting in this regard to compare him with the case of Pan, discussed above. The incorporation of Pan into the periakropolitan fringe allows for a reflection of another past, someone else’s, that of the imaginary Arkadia, wild and primitive; Kekrops, on the other hand, ruler, civiliser, culture-hero, is an embodiment of self, not other.148

  • 149 For example, in Hdt. 1.78.3, snakes featuring in an omen are interpreted as representing native inh (...)
  • 150 Parker (1987), 193.
  • 151 See Ustinova (2002).
  • 152 On the connection between Kekrops’ mixanthropy and his autochthony, see Fourgous (1993), 331-2.

74Although it is constantly inadvisable to attach unchanging symbolic characteristics to an animal species in ancient thought, one can suggest some points of connection between the snake element of Kekrops and his character as gêgenês and autochthôn. We have occasional references in ancient literature to snakes being earth-born and residing in the earth,149 and this quality is plainly relevant to Kekrops; as Parker observes, ‘Having emerged from the earth he still in part resembled the creature that slips to and from between the upper and lower worlds.’150 In addition, the Greek pantheon provides numerous cult-heroes whose snake-associations coexist with an association with underground resi­dence, figures such as Asklepios and Trophonios.151 Plainly a persistent theme is at work.152 Yet again, however, more is revealed by the peculiarities of the individual case, and what is peculiar about Kekrops here is his mixanthropy, a highly specialised manifestation of snake imagery. Gods associated with, and represented as, snakes, form the largest and most coherent body of animal-related deities in ancient Greek culture. Indeed, the snake appears to be an especially popular animal for divine representation. However, within this group, mixanthropy is very rare. Only Kekrops is consistently portrayed as mixanthropic. To give his form context, however, something must be said about non-mixanthropic snake imagery and its relative abundance.

  • 153 The fullest treatment of them as a group is that of Mitropoulou (1977), who collates and catalogues (...)
  • 154 An example is a fourth-century votive relief from Mounychia, probably dedicated to Zeus Philios, sh (...)
  • 155 See Garland (2001), 135-7.
  • 156 See e.g. Aristoph. Plout. 730ff.; also the early-fourth-century marble relief from the Amphiaraon a (...)
  • 157 Divine metamorphosis into animal form often seems to be used as a means of achieving close communic (...)
  • 158 See e.g. Paus. 2.10.3: Asklepios turns up in Sikyon in the forms of a gigantic serpent in a chariot (...)

75Snake-related gods, separately and collectively, have received the extensive scholarly attention that their clear importance merits (another reason not to undertake a full discussion here).153 Two main sub-groups are discernible, the first composed of heroes, associated with underground chambers and with healing and prophecy, the second of various forms of Zeus, especially Zeus Meilichios, Philios and Ktesios. The latter are almost invariably depicted, in cult reliefs, in wholly snake form.154 A large proportion of these images derive from Piraeus, where snake-Zeus worship was concentrated.155 The snake element of the heroes is rather differently expressed. Snakes appear more often as attendants and attributes than as the form of the divinity itself; their rôle is often that of an instrument of healing, a divine emissary which carries out the deity’s miraculous work.156 The exception to this trend is the use of the snake as a temporary epiphanic form, especially by Asklepios. This is more in keeping with the representation of the snake Zeus; after all, when the latter is shown as a snake appearing before mortals, we have no way of knowing whether that form was considered permanent or simply adopted as a medium for communication.157 For Asklepios, the snake form is sometimes the one in which he travels and in which he appears in a new cult site. It is the medium of choice for movement and revelation.158

  • 159 The most famous instance of anguipede Giants is surely their depiction on the Hellenistic Great Alt (...)

76It is highly significant that in the midst of all this rich snake imagery, mixanthropy hardly features. Is mixanthropy avoided as in some way unsuitable? In any form, mixanthropy is the preserve of the monstrous; even ‘respectable’ figures like Cheiron have a strong element of the monstrous, and one can imagine a reluctance to depict a god of Zeus’ stature in that way. However, in the case of snake-mixanthropy, more specific objections existed. It is most commonly used in the depiction of the enemies of the gods, for example Typhon and (in some instances) the Giants,159 and for a number of lesser mythical villains such as Delphyne and Echidna. So its unpopularity in the iconography of most deities is not surprising. And yet it was consistently used for Kekrops – why? Why was snake-mixanthropy, the juxtaposition of snake and human parts, particularly appropriate and important to that deity in that context?

  • 160 This is very clear seen on an Attic red figure cup of c. 440 BC attributed to the Kodros Painter sh (...)
  • 161 As Fourgous notes (1993, 232-3), literary references to the part-serpentine form of Kekrops go back (...)
  • 162 e.g. LIMC s.v. ‘Kekrops’, cat. no. 37: a relief showing Kekrops and Athene from 410/9 BC (Louvre MA (...)
  • 163 Gourmelen (2004).
  • 164 This point is also made by Kearns, who remarks that ‘despite a few fully human representations, eve (...)

77First a word on the representation of Kekrops. This follows a pattern clear from many other mixanthropes: painted pottery favours a mixanthropic form, a bearded man with, in lieu of legs, a coiled serpent tail. The human upper half is almost always carefully and decorously clothed, the overall air one of solemnity and composure.160 This form appears to have been well-established by the fifth century BC, when it reached its peak of popularity.161 On the other hand, sculptural forms prefer wholly human anatomy.162 This means that the cult artefacts that we have (largely reliefs, not numerous) are less inclined to depict Kekrops as a mixanthrope than is the less explicitly cultic medium of ceramic. Does this suggest that his serpent element was relatively unimportant in cult? Gourmelen suggests otherwise.163 He demonstrates effectively that the concep­tion of Kekrops as a composite being with an animal part was absolutely central to his rôle, in Athenian thought, as cult-receiving hero, mythical figure and legendary early king. The fact that sculpture seems reticent about portraying his mixanthropy is interesting, but clearly does not eliminate the importance of the mixanthropic element which vase-painters repeatedly portray.164

  • 165 On the manifestations of this episode in Attic vase-painting and their political significance, see (...)
  • 166 Hyg. Fab. 166. A faint suggestion of wholly-snake form is perhaps to be found in Philostratus (Vit. (...)
  • 167 Apollod. Bibl. 3.14.6; Ovid, Met. 2.558-61.
  • 168 Pausanias (1.18.2) says that it was the sight of Erichthonios that drove them mad, but does not spe (...)
  • 169 Both traditions are recorded by Apollodoros (loc. cit.).

78The context in vase paintings is often connected to the birth of Erichthonios.165 Unlike Kekrops, Erichthonios as an adult is most often depicted as wholly human, in myth and art, though with snakes in the offing; at the same time, however, the snake aspects in his case are treated as far more alarming than those of Kekrops. The daughters of Kekrops open the chest which contains the infant, and receive a hideous snake-related revelation: the child in half-snake form,166 or with a snake coiled around it.167 They are then driven mad,168 or else are destroyed by the snake that accompanies the child.169 Erichthonios is rendered ‘respectable’ through his association with Athene and Kekrops, and he succeeds the latter as king of Attica; but it is significant that his animal elements carry associations of terror and shock, whereas those of Kekrops do not.

  • 170 Fourgous (1993), 234-6; Gourmelen (2004), esp. 97-112.
  • 171 Schol. Aristoph. Plout. 773.
  • 172 Philochoros, FGrHist 328 F 94.
  • 173 Cic. Leg. 2.63.
  • 174 Of course, the snake carries several positive associations as well, with fertility, knowledge, and (...)

79As so often with mixanthropic deities, we can remark on a species’ associations in Greek thought, but what really characterises individuals is the dynamic juxtaposition of animal and human. As a being with a double form and a double nature, Kekrops is ideally suited to his rôle as mediator between states: between wild and civilised, male and female, and so on.170 It is this mediating rôle which emerges most strongly from ancient literary descriptions of his character and deeds. Kekrops was said to have invented monogamous marriage,171 an invention considered to have contributed to Athens’ escape from savagery over which Kekrops presided; he also united the tribes in a single community172 and invented funeral rites.173 His characterisation is that of a culture-hero who assists his community through important boundaries of development and civilisation. In Kekrops’ case, the potentially dangerous power of the snake is harnessed, yoked to a controlling human component, and directed entirely for the good of his native Attica.174

4. Dionysos: a god on the borders

80The god Dionysos in his infinite flexibility challenges the definitions and the parameters within which scholarship attempts to function. His inclusion within this book is debatable. On the one hand, there is much to connect him to other deities discussed; like Proteus and Thetis, he is a shape-changer; his occasional mixanthropy exists within this varied palette of forms. However, in cases of attested cult the animal element of the god tends to be manifested not through mixanthropy but rather through full theriomorphism, rare, as already noted, in Greek religion. His mixanthropy, on the other hand, is a feature chiefly of literary accounts, and late ones at that. Whether or not he should be considered a mixanthropic god at all is frankly open to question. However, failing to include him creates an artificial distinction between ‘true’ or ‘proper’ mixanthropes and those who fall short of certain criteria. As in the case of Thetis, Dionysos, though divergent from the main trend, provides an invaluable model of variety without which the concept of mixanthropy and its meaning would seem unnaturally and unhelpfully rigid.

  • 175 Lyk. Al. 1237 and schol. ad loc.

81There is one suggestion of a mixanthropic Dionysos receiving cult in that form. This is a Macedonian ritual, in honour of Dionysos Laphystios (the name echoes Zeus Laphystios). Lykophron mentions, briefly and cryptically, the Λαφυστίας κερασφόρους γυναῖκας, about which the scholiast Tzetzes explains that they are the female participants in a rite, in honour of a bull-headed Dionysos, which involved the wearing of horned masks (presumably in a dance or proces­sion).175 In this cult Dionysos appears to have functioned as mixanthrope rather than as theriomorph, but the evidence of Tzetzes is, in isolation, extremely unsatisfactory.

  • 176 Pausanias (6.26.1) gives considerable information concerning the important cult of Dionysos at Elis (...)
  • 177 Plut. Quaest. Gr. 36. On the hymn, see Daraki (1985); Schlesier (2002). Further difficulty is added (...)
  • 178 The level of authorial uncertainty in this section is indeed high, as Schlesier (2002, 163) notes r (...)
  • 179 The nature, and indeed the very existence of coherent Orphic ideas are a notorious source of contro (...)

82Dionysos’ theriomorphic cult manifestation receives two disparate mentions. First, Plutarch records a ritual (whose date is essentially unknown)176 in which the women of Elis call upon Dionysos to come to them βοέῳ ποδὶ θύων, ‘rushing with ox-foot’, and address him as ἄξιε ταῦρε, ‘worthy bull’.177 Among various tentative explanations,178 Plutarch suggests that, since the bull’s foot is less destructive than his horn, this formula is asking the god to be mild and propitious. It is easy to dismiss this peculiar explanation out of hand. However, it is not insignificant that, for Plutarch, this Dionysos is one whose presence could be a violent, destructive one. This sense is conveyed in the participle θύων. Indeed, accounts of the taurine Dionysos are often characterised by depictions of his turbulence, violence, menace, ability to inspire terror. This is contained in, but by no means restricted to, texts which appear to be dealing with the so-called ‘Orphic’ Dionysos, sometimes called Zagreus, who was the son of Zeus and Persephone.179

  • 180 Ant. Lib. Met. 10.
  • 181 Nonn. Dion. 6.179-205. On the special rôle of metamorphosis within the Dionysiaka, see Buxton (2009 (...)

83Moreover, there is an explicit connection between the frightening aspect of Dionysos and his animal form(s), expressed via the motif of his shape-changing. The adoption of a series of animal forms is always intended to alarm and deter, even when it is used evasively and the user is at a distinct disadvantage. Antoninus Liberalis180 records the myth in which Dionysos terrifies and maddens the daughters of Minyas using animal metamorphoses, the first of which is a bull. Shape-changing is similarly (but defensively rather than aggressively) used by Dionysos when he is threatened by the murderous Titans in Nonnos’ account;181 to escape their clutches, he adopts several forms in succession, most of them ferocious animals. Among these forms, however, that of the bull is clearly distinguished. It is the final shape assumed; and it is the shape the god is in when he is finally slain and dismembered.

84With regard to this last point, the bull seems often to have a dual association with, on the one hand, active destruction, and, on the other, passive suffering and death. This leads us to our next rite, apparently practiced on Tenedos, and described by Aelian:

  • 182 Aelian, HA 12.34: Τενέδιοι δὲ τῷ ἀνθρωπορραίστῃ Διονύσῳ τρέφουσι κύουσαν βοῦν, τεκοῦσαν δὲ ἄρα αὐτὴ (...)

The people of Tenedos keep a cow in calf for Dionysos Anthroporraistos (Man-Slayer), and as soon as it has calved they tend to it as though it were a woman in child-bed. The new born calf they dress in buskins and then sacrifice. But the man who dealt it the blow with the axe is pelted with stones by the people and flees until he reaches the sea.182

  • 183 For an Argive ritual in which Dionysos is addressed as Bougenês, ‘Cow-born’, see Plut. de Isid. et (...)
  • 184 The equation of Dionysos with the sacrificial victim is famously strong. As Daraki (1985, 65) remar (...)
  • 185 This matter is treated in a fascinating discussion by Daraki (1985, 45-71), in which she questions (...)
  • 186 Interestingly, another of Dionysos’ bull-related cult titles is ‘Bull-eater’. The Suda entry (s.v. (...)

85This rite brings out the juxtaposition of violence and vulnerability. Dionysos is embodied, to some extent at least (this is the effect of the buskins), in a calf,183 which is then killed in guilt-laden circumstances.184 But this particular form of Dionysos is called Anthroporraistos, ‘Man-slayer’. That is not the title of a simple victim. The bull-Dionysos can mete out violence as well as suffering it, it would seem, and this duality appears to be focused on the emblem of the bull as both fierce animal and sacrificial victim.185 This duality underpins the Greek attitude to bulls more widely, and mythological settings sometimes place the two possibilities in at least potential opposition. An example is the myth of the bull of Poseidon which emerges from the sea onto the shore of Crete. The proper rôle of the bull is as a sacrificial victim; when that rôle is denied (by Minos keeping the bull to add to his own herds), the animal becomes a destructive force, engendering a monster – the Minotaur – and finding a narrative counterpart in the second bull sent by Poseidon, which kills Hippolytos between Athens and Troizen. Though this example is far removed from the case of Dionysos, one might argue that it illustrates the underlying purpose of the killing of the Dionysos-calf: to remove, by a sanctified method, an entity potentially destructive to the community.186

  • 187 Cf. lines 920-22: when Pentheus sees Dionysos as a bull, this is simultaneously delusional and trut (...)

86Dionysos’ theriomorphic form is less anomalous within Greek tradition than it may seem, for it is distinctly epiphanic and thus related to shape-changing, the adoption of temporary animal forms. The calf in the Tenedian ritual is only a temporary sacrificial stand-in. The Elean pleas to him to come ‘with ox-foot’ are, as Jaccottet notes, equivalent to the cry of the god’s followers in Euripides’ play: φάνηθι ταῦρος (line 1018) – not ‘come, bull’ but ‘come as a bull’. Throughout the Bacchai, the bull form is treated as guise rather than identity. This sense is reinforced if one looks at the whole sentence, lines 1018-9: Dionysos is implored to come as a bull, or as a serpent, or as a lion, and the words idein and horasthai – though not applied to the bull form, but only to the other two – make explicit the sense of epiphanic guise. The variety of species also connects with his shape-changing tendency.187

  • 188 Deipn. 11.476.
  • 189 de Isid. et Osir. 35.
  • 190 LIMC s.v. ‘Dionysos’, cat. no. 156 seems the only certain instance, a third-century coin thought to (...)
  • 191 LIMC s.v. ‘Dionysos’, cat. nos. 157-9.

87Theriomorphism and shape-changing: what of mixanthropy? We have no explicit references to mixanthropic cult statues; these seem to be wholly theriomorphic, although the terms used by the ancient authors tend to be ambiguous. Athenaios records a cult statue at Cyzicus, on the Propontis, which was tauromorphos;188 Plutarch says that many of the Greeks make tauromorphos statues of the god.189 Tauromorphos looks like a designation of full theriomorphism, but it is too vague for certainty. It is probable that Dionysos was shown as bull-horned on coinage of south Italy, but there are problems of identification; the images could show river gods, prevalent in that region.190 Horns are incon­veniently transferrable. The same uncertainty attends a small number of sculpted male heads with horns, which could be Dionysos, river gods, or Diadochoi.191

  • 192 For the text of the hymns, see Quandt (1962); the most recent edition, however, is that of Ricciard (...)
  • 193 This is a fifth-century AD text, which immediately necessitates some caution in the present study, (...)
  • 194 Rudhardt (2008), 171-2.
  • 195 For discussion of similarities between the texts, see Morand (2001), 83-6.
  • 196 Orphic Hymn to D. 30, line 6, calls him the son of Persephone. In Nonnos, this Dionysos is also cal (...)
  • 197 On the relationship in the Hymns between material particular to them and material known from elsewh (...)
  • 198 For example, echoes do appear to exist between the Hymns (especially their choices of deity and the (...)

88Mentions of Dionysos’ mixanthropy in literature are more substantial but are late; they are also from texts which cannot really be regarded as typical. The chief examples are the Orphic Hymns to Dionysos,192 and the sections of Nonnos’ Dionysiaka193 which appear to contain similar themes; though the dates of the Hymns are both uncertain and (probably) various,194 it is almost certain that they predate Nonnos and may have been among his influences.195 The Orphic Hymns and Nonnos contain several elements not found outside the Orphic tradition, such as the birth of Dionysos to Zeus and Persephone, and his murder by the Titans;196 however, they also incorporate ingredients familiar to us from (for example) Euripides’ Bacchai – the rôle of Semele is one of these – and should not be regarded as a wholly separate cultural, religious or literary phenomenon.197 Attempts to make them the product of specific and separate groups have tended eventually to founder.198 The concentration of mixanthropic motifs in the Orphic material is interesting, but cannot really support theories about the distinct theologies of specific groups of worshippers. So when the god is worshipped as a bull in Elis or as bull-formed in Cyzicus, the echoes with the mixanthropy in Orphic texts are interesting, but not indicative of any measurable convergence or influence. The overall impression one receives is that mixanthropy, theriomorphism and animal metamorphosis are pervasive themes in the depiction of Dionysos across the board, but that the story of his dismemberment by the Titans – a highly specific manifestation – gives them especial concentration and intensity.

  • 199 For example, in Orph. Hymn 45, line 1, Dionysos is called taurometôpe.
  • 200 This is the most common element, and appears in early sources: see Eur. Bacch. 100, 921; Soph. fr. (...)
  • 201 As in the Elean hymn cited in Plutarch’s 36th Greek Question. Another example occurs at Soph. Ant. (...)

89Even when mixanthropy is not explicitly defined, however, it is notable that particular emphasis is placed on certain animal parts. Very often, texts link bull form to specific parts of the body. These tend to be the head/face,199 horns200 and feet.201 These are the key ‘points’ of the body on which mixanthropy is often focused. They are particularly clearly seen in the case of Pan, whose extremities are – generally speaking – animal, and, crucially, remain so even when his torso becomes more human and more attractive. Certain parts of the body are critical in registering animality, and their distribution is not random. The head is clearly vital; it contains expression, thought and speech; if it is animal rather than human, this has profound implications for identity. And almost no mixanthrope has human feet. Put briefly, the animal parts brought out with regard to Dionysos are those which appear repeatedly in the trends of mixanthropic representation; this potentially brings the god into the animal-human discourse of which mixanthropy is a working part. However, Dionysos must be acknowledged as a borderline case, not unsuitable for such a perennial boundary-crosser as he is.

Notes

1 For a somewhat over-imaginative reconstruction of Cheiron’s religious significance in this region, see Picard (1951); see also Guarducci (1948).

2 For the cave and its contents, see Giannopoulos (1912) and (1919); Levi (1923).

3 On the inscription, see Comparetti (1921-2); Decourt (1995), 90-94, no. 73.

4 On Pantalkes and other real-life nympholepts, see Larson (2001), 13-20; for nympholepsy as a mythological motif, ibid. 66-71. See also Ustinova (2009), 61-4.

5 The text in question was originally attributed to Dikaiarchos, and appears under his name in FHistGr 2 F 60. For the attribution to Herakleides: Pfister (1951).

6 This spelling of the title is generally thought to be a mistake of Herakleides: inscriptions from the area (e.g. IG IX² 1103, 1105, 1008, 1009.54, 1110, 1128) support the variant Akraios.

7 Herakleides 2.8: Ἐπἄκρας δὲ τῆς τοῦ ὄρους κορυφῆς σπήλαιόν ἐστι τὸ καλούμενον Χειρώνιον καὶ Διὸς Ἀκταίου ἱερόν ἐφ κατὰ κυνὸς ἀνατολὴν κατὰ τὸ ἀκμαιότατον καῦμα ἀναβαίνουσι τῶν πολιτῶν οἱ ἐπιφανέστατοι καὶ ταῖς ἡλικίαις ἀκμάζοντες, ἐπιλεχθέντες ἐπὶ τοῦ ἱερέως, ἐνεζωσμένοι κῴδια τρίποκα καινά. τοιοῦτον συμβαίνει ἐπὶ τοῦ ὄρους τὸ ψῦχος εἶναι.

8 See Arvanitopoulos (1911); Philippson (1944), 147; Chourmouziades (1982), 98. Unfortunately, the site excavated by Arvanitopoulos is not now possible to locate; during my visit to Thessaly in November 2009 the Commander of the Pliassidi Air Force base very generously facilitated some exploration in the area, but no ancient remains were found.

9 There has been a determination in the older scholarship to edge Cheiron into greater prominence within the rite; for example F. Stählin (1924, 43), believed (on the basis of no evidence) that Zeus was a relative newcomer into the partnership.

10 Buxton (1994), 93-4.

11 IG XII.3, 360. See Vogel (1978), 218.

12 Philippson (1944), 150-55. She posits a process whereby the cult travelled from Pelion in Thessaly to the Peloponnese, and from there to Thera; it retained, she argues, its Thessalian character, including the cave-association and the kourotrophic element. The evidence adduced is somewhat questionable. (For example, Apollodoros’ account of Cheiron’s expulsion from Pelion to Malea – Bibl. 2.5.4. – is taken as referring, mythologically, to the first leg of the journey.) The theory is not, however, per se implausible.

13 Pind. Pyth. 4.102-3.

14 Plut. Quaest Conv. 3.1.

15 Herakleides 2.12.

16 Nikandros, Theriaka 500-505: πρώτην μὲν Χείρωνος ἐπαλθέα ῥίζαν ἑλέσθαι, | Κενταύρου Κρονίδαο φερώνυμον, ἥν ποτε Χείρων | Πηλίου ἐν νιφόεντα κιχὼν ἐφράσσατο δειρῇ. | τῆς μὲν ἀμαρακόεσσα χυτὴ περιδέδρομε χαίτη, | ἄνθεα δὲ χρύσεια φαείνεται· ἡ δ’ ὑπὲρ αἴης | ρἵζα καὶ οὐ βυθόωσα Πελεθρόνιον νάπος ἴσχει.

17 Theophrast. Peri Phytôn IX.11.1-7. It is clear from the description of the plant (golden flowers) that it is the same species as described in the Nicander passage.

18 As Hodkinson (1988, 64) puts it, ‘The horse was the useless animal par excellence in an era before the invention of harness enabled it to be used for traction.’ Its value was largely as an indicator of wealth and status. On cattle used for draught rather than horses, see Jameson (1988), 94-6.

19 Diod. 4.15; Hyg. Fab. 250. The animals also consume Abderos, Herakles’ servant, who has been set to guard them: Strabo 7, fr. 47.

20 Aelian HA 15.25; Apollod. Bibl. 2.5.8. Domestic animals released into the wild do not always become hapless victims. Those of Geryon’s cattle which giver Herakles the slip become thêres in the mountains of Thrace: see Apollod. Bibl. 2.5.10.

21 For some pertinent myths, see Ant. Lib. Met. 7 and 20.

22 On the ancient characterisation of centaurs, and on the symbolic distinction between the group-centaurs and Cheiron and Pholos, see Kirk (1971), 152-62.

23 Ancient descriptions of the event in Pind. Nem. 3.74-5, Pyth. 4.181-4; Pyth. 3.83-95; Servius Verg. Georg.. 3.93; Ap. Rhod. Arg. 2.1232-42 (this last unusually gives the location of mating as on the Black Sea); Hyg. Fab. 138. See Gantz (1993), 43 and Guillaume-Coirier (1995), esp. 114-120.

24 See Pind. Pyth. 2.42-8; Diod. 4.69-70; Hyg. Fab. 62.

25 Apollod. Bibl. 3.14.6. A parallel of sorts is the birth of Typhon to Hera, which is accomplished without the usual intercourse. In anger at Zeus, Hera decides to bear a child without his assistance, but the result is, inevitably given the circumstances of conception, a monster: see Hom. Hymn 3.331-55; Fontenrose (1959), 13-14.

26 It should be noted that one unusual and isolated variant on the myths elides the difference between Cheiron and the group-centaurs: Lucan (6.386-7) makes Cheiron a child of Ixion and Nephele and a brother of Peirithoös.

27 Jeanmaire’s (1949) discussion of the character of Cheiron in ancient poetry is still of great value despite its age.

28 Apollod. Bibl. 3.13.3.

29 See e.g. Gantz (1993), 145 on the patterns at work in the representations of centaurs in the metopes of Heraion I at Foce de Sele.

30 An example of this composition is his depiction on the François Vase.

31 For examples of the early type of centaur with human front legs, see Padgett (2003), figs. 4-8

32 Gantz (1993), 145; Mylonas Shear (2002), 151; Padgett (2003), 14.

33 Why does Cheiron not move with the times? Is it simply to keep the dominant humanity, or is there another reason behind his conservatism? The animal element of Cheiron is not generally played down in our sources; true, in Homer, he is named whereas the group-centaurs are called phêres; but Pindar calls him phêr theios. This question cannot be answered for certain, but there is a parallel example which suggests a very interesting possibility. In the representations of the river-god Acheloos (as has been described above) we find the preservation of an early type at the same time as the development and increasing popularity of a new version. The older type shows the god as a bull with a human neck and head; the new one attaches a human face to a taurine neck. The new, more popular image renders the mixanthrope more plausible as a ‘living’ creature, less stilted and stylised. But the stiffer image continues and, in the case of Acheloos, what ensures its survival is its rôle in the imagery of Acheloos’ cult. Cult representation is far more conservative than ‘secular’ forms such as painted pottery; there is an almost superstitious unwillingness to depart from a sanctified norm. Can this example inform us as to Cheiron’s representation? There are important divergences. First, though Cheiron received cult in Thessaly, no cult icons or images have been found. None is mentioned in literary sources; instead, his cave on Mount Pelion is his most important tangible monument. Thus we cannot claim for certain that the existence of a Kultbild of some sort affected his non-cultic depictions. And yet there are two objections to complete scepticism on this matter. The first is that we cannot know absolutely that no cult image of Cheiron ever existed; perhaps it was a wooden xoanon and was lost before our literary sources were operating. Secondly, and more importantly, it might very well be that the mere knowledge that Cheiron was a god, a recipient of cult, militated against changes to his representation. A third factor may well have been the fact that Cheiron was persistently regarded both as old himself and as a denizen of an earlier, Archaic age (for which see below).

34 Despite the infiltration of the Eleusinian form into Arkadia itself: see Paus. 8.25.2-3 for a sanctuary of Eleusinian Demeter in Thelpousa, the location of the radically un-Eleusinian Demeter Erinys. The more pan-Hellenic form did not cause the local one to be obscured or removed.

35 For a discussion of landscape, settlement and sacred space in Arkadia, see Jost (1999).

36 In fact, as Burkert has shown, various strands of mythology connect not only Phigalia and Thelpousa, but also the Boiotian sanctuary at the spring of Tilphoussa, near Haliartos: see Burkert (1979), 123-9.

37 The relationship between cult-type and topography is one of the consistent foci of Jost’s work, and one of its most important contributions to the understanding of Arkadian religion. See esp. Jost (1985), 82-3; (1994), 221-3; (1999), 208-9.

38 Frazer (1898), 406-7. Autopsy is important to his discussion. (p. 407: ‘I visited the cave 2nd May 1890, and have described it from personal observation.’) See also Voyatzis (1999), 149.

39 Paus. 8.42.4 (for the Greek text, see the Appendix below).

40 It seems fairly feasible to assume the existence of a cult image at the time of Pausanias’ visit; the total lack of any representation of the goddess on the site would surely have occasioned explicit comment. This conjecture cannot, however, support anything approaching certainty, especially since other mixanthropes in this study offer no evidence of a cult image (for example, Cheiron, the Sirens and Kekrops). In the case of Demeter Melaina, it is interesting to entertain the possibility of an element of aniconism in Pausanias’ day; but the matter cannot be settled one way or the other without further evidence.

41 Unsurprisingly, the scholars who treat the xoanon without caution are those seeking traces of animal-gods; an example is Lévêque (1961), 102; he quotes in full the relevant passage of Pausanias without identifying any problematic qualities. Farnell is rare among scholars who have treated this passage in that he urges caution, and he rightly draws attention to the fact that, in any case, ‘Some of the Phigalians were uncertain whether [the xoanon] had ever belonged to them.’. See Farnell, vol. 3 (1907), 51.

42 Bruit (1986).

43 Coins of Thelpousa depicting Demeter and a horse, inscribed with the name Erion (a form of Areion), see Head (1911), 382.

44 Jost (1985), 304-7. See also Dietrich (1962), 130-134.

45 Cf. a depiction of a gorgon-centaur fighting a lion on a sixth-century amethyst scarab from Byblos: LIMC s.v. ‘Gorgo, Gorgones’, cat. no. 285; London BM WA 103307.

46 See Smith (1884), 239-40 and pl. XLIII; Gantz (1993), 144; Frontisi-Ducroux (2003), 210.

47 Paus. 8.25.7.

48 For Areion (under the alternative name Erion) on the coins of Thelpousa – a reflection of his importance there – see Jost (1985), 64, and pl. 11, no. 2.

49 Jost claims that Demeter Erinys at Thelpousa would have been depicted mixanthropically before Eleusinian influence eradicated this form (Jost 1985, 302); however, this is an unprovable supposition.

50 In ancient myth, Lykaon kills and serves up a human infant. In Pausanias’ version this occurs as a sacrifice to Zeus (8.3.2), but the commoner setting is a banquet at which Zeus is present (see e.g. Apollod. Bibl. 3.8.1; Ovid, Met. 1.199-243; Hyg. Fab. 176). The two variants are closely linked, however; feast and sacrifice are elided in Greek culinary and religious practice, and both versions of the story strongly echo Hesiod’s description of Prometheus’ deceptive sacrifice to Zeus at Mekone (Hes. Theog. 521-616), in which the sacrifice takes place during a banquet shared by gods and men.

51 In several accounts, Lykaon is punished for his child murder by being turned into a wolf, though in another variant he is blasted by Zeus’ thunderbolt. See Paus. 8.1.3; Hes. fr. 163 MW; Apollod. Bibl. 3.8.1; Lyk. Al. 480-81 with schol.; Ovid, Met. 1.198-239. For discussion of the episode, see Burkert (1983), 84-93, which examines the connection between the myth and the Arkadian lycanthropy-ritual which was claimed to have existed in reality; a similar conjunction of myth and ritual is attempted by Buxton (1987), 67-74. Bynum (2005), 166-70, gives a treatment based largely on the Ovidian version but still interesting from a wider perspective. The fullest discussion of the myth remains Piccaluga (1968). For a discussion of the myth’s place in Pausanias’ work and its themes, see Pirenne-Delforge (2008a), 67-72.

52 See Bruit (1986), esp. 95; Borgeaud (1988), 34-44 (on Lykaion).

53 For example, it was because of its famous Despoina-sanctuary that Lykosoura was allowed to keep the status of city after the synoecism of Megalopolis: see Paus. 8.28.6 and Jost (1994), 226.

54 Especially suggestive is the text of a sacred law from Lykosoura (IG V², 514), which details certain requirements and prohibitions on visitors to the shrine of Despoina. Some of these seem to echo features of the Phigalian site: female visitors are forbidden to wear black, and the type of sacrifice demanded also, in the words of Jost, ‘rappelle le rite de Phigalie’ (Jost 1985, 330). On the sacred law, see Loukas and Loukas (1994), which focuses on the date (probably third century BC) and the historical context; for an early discussion of its significance, Leonardos (1896). For a recent analysis of the inscription and its links with the known rituals and beliefs of the site, see Jost (2008), 94-102. Cf. also SEG XXXVI 376 (no. 8 in Lupu [2005]): a second-century BC addition to the sacred ordinances of Lykosoura: the text is badly damaged and yields little information, but appears to concern ‘cathartic requirements’ (Lupu [2005], 216).

55 For the interconnected cults of Phigalia, Lykosoura, Lykaion and Megalopolis, see Jost (1994).

56 At the start of Book 8 (8.1.5) we are told by Pausanias that in fact the acorn-eating phase was not the first: before this, men ate leaves and roots, which were sometimes poisonous. The proto-king Pelasgos introduced them to the relatively nutritious acorn, and also to the construction of huts and the wearing of fleeces. In other words, he takes them out of the condition of undifferentiated animals, and places them on the first, and most primitive, step of human development.

57 Paus. 8.1.3.

58 There is also a third, which pervades Arkadian mythology particularly: that is the life, and the food, of the hunter and herdsman, both of which rôles are represented in the figure of Pan. Perhaps this third way stands somewhere between the other two, for it involves the consumption of flesh, but avoids the extreme of flesh-consumption, cannibalism. If we follow the central argument of Burkert’s Homo Necans, we may perhaps see that even the consumption of non-human flesh has its share of the guilt of murder; but in any case, the life of the hunter and herdsman is presented as another acceptable face of the primitive life. It does not play such an explicit rôle in the Phigalian narrative, for that is dominated by the antithesis between acorns and human flesh. But it pervades both the idea and the reality of Arkadia just as Pan pervades its mountains and caves.

59 9.215 ff. For the significance of food in connection with a primitive modus vivendi, see Kirk (1971), 162-8. He argues for the special intensity of the nature/culture theme in this myth, saying that ‘various aspects of nature and culture are being manipulated into proximity for the purposes of evaluation’ (p. 168).

60 Indeed, he does not even eat the sheep he herds, and his care of them is verging on the tender.

61 Just as there are various possible versions of primitive food in ancient thought (acorns; leaves and roots; cheese; human flesh), so there are different types of primitive sacrifice. There is the vegetarian variety described by Plato (Leg. 782c) in a passage which is strikingly reminiscent of the offerings to Demeter Melaina, including as it does fruits of the earth and honey, free from both meat and the products of agriculture. In opposition to this is the transgressive sacrifice of Lykaon, the killing and eating of a human child.

62 Paus. 8.42.11.

63 Paus. 8.42.5-6.

64 Paus. 8.42.6 (for the Greek text, see the Appendix below).

65 Jost (1992 b, 56) relates this in part to Demeter’s mountain location and affinity with the mountain zone, and sees her nature as a reflection of ‘la précarité de la civilisation dans les montagnes.’

66 Bruit (1986), 85.

67 Scheffer (1994) argues for a more specific association between horse-related deities and uncontrolled sexuality; but this may be seen as part of a wider theme of uncontrolled nature. There is no doubt, however, that horses had a particular association in Greek thought with female sexuality. See Scheffer (1994), esp. 129-33.

68 See Borgeaud (1988), 16-17.

69 See Ant. Lib. Met. 20.

70 ibid. 7.

71 Paus. 8.34.3: καὶ οὕτω ταῖς μὲν ἐνήγισεν ἀποτρέπων τὸ μήνιμα αὐτῶν, ταῖς δὲ ἔθυσε ταῖς λευκαῖς.

72 For full analysis of the use of such terminology in ancient texts, see Ekroth (2002).

73 Pirenne-Delforge (2008a), 187-241; see esp. 229-234 for the sacrifice by Orestes.

74 Pace Scheffer (1994), 128-9.

75 Paus. 8.34.2.

76 Pirenne-Delforge (2008a), 233.

77 There is a connection between anger/grief and blackness in the characterisation of Thetis, too, as Slatkin (1986) has shown, demonstrating that in fact Thetis and Demeter are linked in ancient texts by their assumption of dark clothing to indicate anger. It is interesting to recall that, as noted above, Thetis in the Heroikos of Philostratus also deals out, like Demeter Melaina, vengeance for the human neglect of cult rites (in that case, those of Achilles).

78 In the myths surrounding the cults at Phigalia and Thelpousa, metamorphosis into horse form is used by the goddess to escape Poseidon’s amorous attentions: that is, it is manifested rather earlier in the narrative sequence than the goddess’s anger. However, on this point we are justified in supposing that the myth has an explanatory function. The motif of evasion dominates Greek stories of metamorphosis; and though Forbes Irving (1990, 38-50) is right to deplore the tendency to search ‘behind’ and ‘beneath’ the stories themselves for traces of lost animal-gods, in the cults of Phigalia and Thelpousa we have enough complementary evidence to see the horse association as reaching beyond the single mythical episode of Demeter’s attempts to avoid rape. We may assume that the goddess’s horse-element, deeply lodged in her divine character and functions, was at some stage made to tally with the well-established motif of evasion.

79 For the chthonic aspect of Poseidon Hippios, see Detienne and Werth (1971), 167-8.

80 Though Scheffer (1994, 128-9) argues that it is less important than has often been claimed.

81 On the ‘reinstitutionalization’ of Pan in Arkadia after – and massively fuelled by – his diffusion elsewhere, see Borgeaud (1988), 52.

82 See Larson (2001), 97, on the rôle of Athens in disseminating the cult of Pan through Attica and then in other regions of Greece.

83 For an example one might look at a case already discussed, the deities – among them Cheiron – included in the long verse inscription in the cave near Pharsalos in Thessaly. Here Pan keeps exactly the type of company we find in Attic sites, including the nymphs and Hermes. Local associations cause the addition of unusual elements, most notably Cheiron, to what is otherwise an orthodox selection familiar from many Attic reliefs.

84 Jost (1985), 458-9.

85 Paus. 8.36.7; Jost (1985), 200, 458-9

86 Paus. 8.30.2-3; Jost (1985), 221-2, 458-9.

87 ibid. 458.

88 Herbig (1949), 15. This notion also pervades Lamb (1926), who says of the kriophoros-type effigies found in Pan’s shrine on Lykaion, ‘Evidently the figures are dedicated by pious Arcadians in their own likeness’ (p. 134).

89 See e.g. Hübinger (1992, 203-6; 1993, 25-6) on the finds from Lykaion and what sort of people may have dedicated them – not all simple shepherds. He makes the very important point that many of the votives found were too costly to have been dedicated by poor shepherds and hunters (Hübinger [1993], 29). He paints a slightly more complex picture than Jost (1985, 13, 467-8) who focuses on the kriophoros figures from the sanctuary in particular and sees them simply as generic self-portraits of the individuals who dedicated them. Parker (2005, 167) sums the matter up when he says that ‘the familiar notion that certain ‘country’ gods such as Pan were honoured only by countrymen, and countrymen honoured none but them, appears to be part of the pastoral dream.’ This is particularly true of Athens, which Parker is discussing, in which Pan held a decidedly civic status. (See Parker [1996], 166-7.)

90 Initial excavation reports: Kourouniotis (1902), (1903) and (1904).

91 Paus. 8.38.5. Pausanias says that in his day these games were no longer celebrated there, a statement which may be read in various ways; but their importance in the Classical period, especially the fourth century, is certain. See Jost (1985), 185.

92 Jost (1985), 183-5,267-8; id. (1994), 226.

93 For a clear example, see Jost (1985), pl. 63, no. 4.

94 The precise date of the synoecism is uncertain. For the event and its context of growing Arkadian self-assertion, see Rhodes (2006), 217-8. For the religious manifestations of the event, see Jost (1985), 184-5 and (1992), 224-38.

95 Paus. 8.38.5.

96 On Megalopolitan doublets, see e.g. Jost (1994), 227.

97 Paus. 8.30.2. Also at Tegea: Paus. 8.53.11.

98 Schol. Theok. 1.5.123c.

99 Jost (1985), 474-5.

100 Paus. 8.37.11: λέγεται δὲ ὡς τὰ ἔτι παλαιότερα καὶ μαντεύοιτο οὗτος θεός, προφῆτιν δὲ Ἐρατὼ νύμφην αὐτῷ γενέσθαι ταύτην Ἀρκάδι τῷ Καλλιστοῦς συνῴκησε.

101 Serv. Verg. Geor. 1.5.17. For discussion of the passage, see Jost (1985), 474-5; Larson (2007), 63.

102 For this encounter, see Hdt. 6.105. See Parker (1996), 163-5 for the introduction and dissemination of the cult in Attica following Philippides’ revelation.

103 See Borgeaud (1988), 134-5. Borgeaud argues convincingly that Pan’s key character in Athens was as a peaceful, unwarlike entity, the antithesis to war (135-8) rather than as a martial ally. Parker (2005, 477) remains doubtful that the races in his honour simply mirrored that of Philippides, and expresses uncertainty about their exact meaning.

104 For the torch, its rôle in the Persephone myth and its fertility-associations more generally, see Parisinou (2000), 81-99; plates 14, 15, 16 and 40 are of especially interest in this matter. In her study, Pan is connected with torches only via his participation in the torch-lit pannuchides of Cybele (pp. 160-61); but note the torch-racing Silenoi in plate 2; the Dionysiac context may be another important connection. To return to the theme of Persephone, two Panes attend her return in fig. 26 (p. 118); see below for discussion. We have already noted Pan’s rôle in bringing back Kore’s mother Demeter from her destructive sojourn in the Phigalian cave.

105 Examined in great detail by Borgeaud (1988); the present section is merely a summary of some vital aspects.

106 The nymphs already had cult in Attica before the arrival of Pan, but Parker (1996, 163-5) argues that the prominence of cave-locations is a fifth-century phenomenon. Larson (2001, 97) is less certain on this point, and contends that some caves may have been sacred to the nymphs before this time. The question is relevant because it affects whether we see Pan being grafted onto an existing pattern of worship (as Larson believes) or whether his arrival was in itself influential, perhaps bringing the cave to greater prominence. The latter would not be implausible given the strong thematic connection between mixanthropes and caves (see below).

107 Documents illustrating this are assembled and discussed by Borgeaud (1988), 140-43, where he also ties in Pan’s Arkadian relationship with Demeter. Pan’s fertility function is therefore both general and highly specific.

108 In fr. 96.2 Snell, he calls Pan the ‘dog of the Great Mother’.

109 Parker (1996, 165) remarks that ‘the new cult was not a simple re-creation in Attica of what had been practised in Arcadia hitherto.’

110 For discussion of all that Pan meant to the Athenians, wildness and animality, see Parker (1996), 167-8.

111 Borgeaud (1988), esp. 50-52. The function of the cave on the acropolis is particularly interesting; as Parker (2005, 52) remarks, it allowed the Athenians to ‘square the circle, both granting their honoured new divine guest a place near the heart of things, and also respecting that strange wildness which made him unsuitable to occupy a normal temple.’ This ‘strange wildness’, and the rarity of temple-occupation, is common to all mixanthropes. See further Loraux (2000), 39-46.

112 For example, Herbig’s monograph covers almost the whole swathe of antiquity; for treatment of later material up to and including the modern age, see Boardman (1997) and Merivale (1969). Both offer very interesting discussion of the modern reception and use of the Pan-image, which shed a retrospective light on the ancient sources but are outside the scope of this study. Pan’s form and nature have been extensively discussed, and Borgeaud’s 1988 work on the subject in particular took the subject forward by a great leap in terms of sophistication of thought. The current discussion will not try to encompass the huge wealth of material or ideas; instead, the focus will be on Pan’s mixanthropy, its significance and its development in the period under review. As ever, I shall be attempting to establish the contribution of the animal component and its relation, both physical and thematic, to the human part.

113 In LIMC s.v. ‘Pan’.

114 Fig. 29; LIMC s.v. ‘Pan’, cat. no. 1; see Herbig (1949), 51-3; Brommer (1949-50), 6-7.

115 Herbig (1949), esp. 53-7.

116 For a typical Classical Pan, see fig. 30.

117 Hübinger (1992).

118 An example of the ‘young and pretty’ type is LIMC s.v. ‘Pan’, cat. no. 36: an Apulian red figure krater of the fourth century (Vatican AA 2 18255), which shows Pan human-faced but for animal ears and horns, a composition very familiar from the river-gods of that type.

119 Numerous other examples come from Arkadia. Examples are given by Jost (1985), 464-6. Several of these have an explicitly cultic origin, being votives from known sanctuaries of the god. An example is a small fifth-century bronze which bears the words Κέος ἀνέθεκε and which shows Pan with bearded human face, horns and animal hooves (Mariemont Mus. G 57).

120 This latter medium tends to consist of the same nymph-reliefs in which Acheloos (see above) figured so persistently.

121 Herbig (1949), esp. 33-4.

122 The conventional, pastoral etymology is discussed by Larson (2007, 63) as being the most resonant and significant in antiquity. A number of possible etymologies are explored in Borgeaud (1988), 185-7 (appendix). A radically divergent view is presented by Brown (1977), who argues that the name’s starting-point is the Mycenaean form Opaon, meaning, in his words, ‘the Companion par excellence’ (see p. 59); this form is preserved in Cypriot cult, while Arkadia is the source of the truncated form Pan. Impossible as it is to verify, this certainly seems another possible strand in the god’s remote past. On the whole, it seems to me that pastoral associations would be more likely to be discerned by Pan’s worshippers in the historical period.

123 An example of the earlier, simpler view is that of Herbig (1949, esp. 15-18), who imagines Pan created by Arkadian pastoralists in their own image, as a reflection of their own concerns. cf. also Immerwahr (1891), 203-4; Lamb (1926). Borgeaud developed the more complex later argument that sees Pan and his character as largely the creation of Athenian thought, and part of their discourse on the primitive, the rustic, the pastoral, all of which both Pan and his native Arkadia represent in a symbolic function largely irrespective of the realities of Arkadian life.

124 His monstrous quality is expressed in the phrase τερατωπὸν ἰδέσθαι, ‘like a teras to behold’.

125 Hom. Hymn 19.38-9. One might compare the terror of the nurses of Erichthonios when they see his half-snake form (or, in other sources, the snakes in his cradle): see below, p. 126.

126 Hom. Hymn 19.45-7.

127 For example, the second line of the Homeric Hymn to Pan calls him aigipodes and dikeros, goat-footed and two-horned. The same two words in the same order are repeated on l. 37. In Aristoph. Fr. 230 he is called kerobatas; this appears to mean horn-footed, that is, with feet made of (the substance) horn, but which also surely serves to recall both the vital anatomical parts. Later references abound; see e.g. Nonn. Dion. 27.290 (feet) and Orphic Hymn 10, l. 5 (both horns and feet).

128 Luc. Dial. Deor. 22.1: How can you be my son, asks Hermes of Pan, ‘κέρατα ἔχων καὶ ῥῖνα τοιαύτην καῖ πώγωνα λάσιον καὶ σκέλη δίχηλα καὶ τραγικὰ καὶ οὐραν ὑπὲρ τὰς πυγάς;’

129 See Boardman (1997), 32-3. On the motif of the anodos of Persephone in vase-paintings, see Bérard (1974); examples of the participation of satyrs and Panes may be seen in his pll. 11-13, demonstrating the frequency of the theme. See also Brommer (1949-50), 22-7.

130 One could argue, though without the possibility of proof, that the fact that Pan’s face is so frequently animal corroborates the supposition that he started out as wholly animal, since it has such a profound effect on identity. Certainly, to refer again to the case of Dionysos, that god also can be treated as wholly bull as well as bull-faced or -horned. By contrast, mixanthropes whose wholly animal portions do not include the face (such as centaurs and Acheloos) are never referred to in this way, simply as a holy animal without mention of a human part. This is not an observation on which one should place too much weight; the number of cases on each side is too small to give reliable statistics. But I believe that Pan’s goat identity is rendered far stronger by his caprine facial features.

131 Epimenides fr. 16 DK.

132 For scholarship on the topic of cannibalism on Lykaion, both mythical and ritual, see above, n. 51.

133 Paus. 8.42.3. The emphasis in this passage is on Pan’s rôle as a denizen of the mountain realm: he is hunting and roaming from mountain to mountain when he finds Demeter.

134 Hyg. Fab. 206; Hyg. Astr. 2.28; Opp, Hal. 3.15; Luc. de Sacr. 14.

135 For the title Haliplanktos, see Soph. Ai. 695; for Aktios, see Theok. Id. 5.14. Of course, we have to ask whether these are in fact divine titles at all, or just poetic adjectives; the literary sources which contain them do not allow for an assumption of religious significance. That haliplanktos is an epiclesis seems, however, probable when we take into account the Suda entry (s.v. ‘Haliplanktos’) which says that Pan was called Haliplanktos because he hunted Typhon with nets; this has the feel of an aition to explain a rather mysterious cult title; but we know nothing about where and in what circumstances the title may have been used. Aktios is similarly mysterious. Apollo was titled Aktios, but this is almost always in connection with the site of Aktion, in Akarnania, and the epiclesis seems to have been a reference to the place-name; see Strabo 7.7.6, 10.2.1-7; Paus. 8.8.12. The Theokritos passage in which Pan is called Aktios sadly offers no evidence of cult practice; it merely testifies to some form of marine element in Pan’s nature.

136 Hyginus for example (Astr. 2.28) derives Aigipan’s fish element from the fact that he hurled sea-shells instead of stones at the Titans.

137 He is described as the first king of Attica by Apollodoros (Bibl. 3.14.1), and this is also strongly implied in Thuc. 2.15.1. For a divergent opinion, see Paus. 1.2.6.

138 See Kearns (1989), 110.

139 It should also be noted that the daughters of Kekrops, Aglauros, Pandrosos and Herse, were themselves of religious importance in Athens in conjunction with the figure of Kourotrophos: see Parker (2005), 216, 433-4; Shapiro (1995). They are non-mixanthropic and therefore do not merit inclusion in the main discussion, but it is interesting to bear this kourotrophic function in mind: could it be analogous to the same rôle played by the kourai hagnai, the daughters of mixanthropic Cheiron?

140 Paus. 9.33.1. The Kekrops Pausanias mentions is perhaps a rather different figure from the one worshipped on the Athenian acropolis. This latter figure comes later in the mythical succession of Attic kings, being the successor of Erechtheus; but as Kearns rightly says (1989, 110), this distinction is in fact ‘a somewhat desperate atthidographic hypothesis’ designed to give chronological order to a morass of mythological variants, and the two Kekropes are in fact ‘branches’ of the same basic entity.

141 See Kearns (1989), 160. Gourmelen (2004) argues strongly that Kekrops’ rôle as eponymos constituted cult: see pp. 295-309. There is no doubt that, whether or not we believe that in this rôle he received the mechanics of worship, Kekrops was an important figure within the conception and organisation of the tribes.

142 Plut. Thes. 36.1-2.

143 See also Rosivach (1987) and Parker (1987, 193-207) for further analysis of the political application of the autochthony theme in Athens; Loraux (2000) for a recent discussion of Athenian autochthony claims, esp. 29-31 on Erichthonios, Kekrops and the Kekropidai.

144 Hall (1997), 51-6.

145 See e.g. Hyg. Fab. 48; Ant. Lib. Met. 6.

146 Fascinatingly, a less strongly attested variant tradition makes Kekrops Egyptian (see Philochoros, FGrHist 328 F 93; Diod. 1.28; schol. Ar. Plout. 773; Suda s.v. ‘Kekrops’); Fourgous (1993, 233-46) argues convincingly that this represents an ancient critique of the Athenian construction of ideal Greek identity through autochthony. The curious dichotomy between the Athenian and the Egyptian Kekrops will be remarked on further in chapter 5.

147 Apollod. Bibl. 3.14.6.

148 As Loraux (2000, 41-2) notes, there is in fact an implicit underlying analogy between Pan and the Athenians since both are autochthones; she accepts, however, the impossibility of knowing for certain whether this ever really overrode the ‘otherness’ of Pan.

149 For example, in Hdt. 1.78.3, snakes featuring in an omen are interpreted as representing native inhabitants because the snake is a gês paida, a child of the earth. Writers of natural history give a scientific slant to this folkloric association by remarking on the tendency of snakes to reside in underground burrows: see e.g. Arist. HA 8 c 15. There is a striking convergence of the mythological and the biologically actual in this regard.

150 Parker (1987), 193.

151 See Ustinova (2002).

152 On the connection between Kekrops’ mixanthropy and his autochthony, see Fourgous (1993), 331-2.

153 The fullest treatment of them as a group is that of Mitropoulou (1977), who collates and catalogues the material evidence for snake-iconography of a number of cult figures, though with limited discussion.

154 An example is a fourth-century votive relief from Mounychia, probably dedicated to Zeus Philios, showing the god in snake-form (Athens NM 1434); for a drawing see Garland (1987), 136. Sometimes the serpent-gods dwarf human worshippers: for an example, see Mitropoulou (1977), 112, fig. 48a (stele dedicated to and depicting Zeus Philios as a gigantic snake with two small humans). For this type, and for the type consisting of a lone coiled serpent, see Mitropoulou (1977), 112-121.

155 See Garland (2001), 135-7.

156 See e.g. Aristoph. Plout. 730ff.; also the early-fourth-century marble relief from the Amphiaraon at Oropos (Athens NM 3369; LIMC s.v. ‘Amphiaraos’, cat. no. 63), which shows both Amphiaraos and his sacred snake ministering to sick people. On this sanctuary, see Schachter, vol. 1 (1981), 19-25.

157 Divine metamorphosis into animal form often seems to be used as a means of achieving close communication with mortals, and Zeus in myth employs this method especially frequently to perform seductions.

158 See e.g. Paus. 2.10.3: Asklepios turns up in Sikyon in the forms of a gigantic serpent in a chariot drawn by mules. For a similar snake-form arrival in Rome, see Ovid, Met. 15. 620-744. Alexander of Abonouteichos takes advantage of this trend when he chooses to introduce his fake Asklepios to the public in the form of a snake, here one which was first ‘discovered’ inside a miraculous egg (see Lucian, Alex. 14-15), though in this case the ‘deity’ remains a snake.

159 The most famous instance of anguipede Giants is surely their depiction on the Hellenistic Great Altar at Pergamon, on which see Radt (1999), 168-80; Queyrel (2005). It should be said that in this case the Giants have two snakes below the waist, rather than the single snake-tail of Kekrops’ type.

160 This is very clear seen on an Attic red figure cup of c. 440 BC attributed to the Kodros Painter showing a snake-tailed Kekrops watching placidly as Athene takes the new-born Erichthonios into her arms (LIMC s.v. ‘Kekrops’, cat. no. 7; Berlin Staatl. Mus. F 2537). Very similar physical form may be seen in fig. 27 (above, p. 124).

161 As Fourgous notes (1993, 232-3), literary references to the part-serpentine form of Kekrops go back to the fifth century also: see e.g. Ar. Wasps 438; Eur. Ion 1158-65.

162 e.g. LIMC s.v. ‘Kekrops’, cat. no. 37: a relief showing Kekrops and Athene from 410/9 BC (Louvre MA 831). In fact, as in most of the anthropomorphic cases, some doubt attends the identification of the relevant figure as Kekrops: for discussion of this, see Gourmelen (2004), esp. 304-6.

163 Gourmelen (2004).

164 This point is also made by Kearns, who remarks that ‘despite a few fully human representations, everyone knew how the hero should look’. Kearns (1989), 111.

165 On the manifestations of this episode in Attic vase-painting and their political significance, see Shapiro (1998).

166 Hyg. Fab. 166. A faint suggestion of wholly-snake form is perhaps to be found in Philostratus (Vit. Ap. 7.24) in which Apollonios jokes that Athene once bore a drakôn for the Athenians; this appears to contain a punning reference to the famous Athenian law-giver Drakon as well as to Erichthonios.

167 Apollod. Bibl. 3.14.6; Ovid, Met. 2.558-61.

168 Pausanias (1.18.2) says that it was the sight of Erichthonios that drove them mad, but does not specify exactly which aspect of the child wrought this effect: perhaps his mixanthropy? Compare the reaction of the nurse to the birth of Pan: Hom. Hymn 19.38-9.

169 Both traditions are recorded by Apollodoros (loc. cit.).

170 Fourgous (1993), 234-6; Gourmelen (2004), esp. 97-112.

171 Schol. Aristoph. Plout. 773.

172 Philochoros, FGrHist 328 F 94.

173 Cic. Leg. 2.63.

174 Of course, the snake carries several positive associations as well, with fertility, knowledge, and the provision of special protection to those it favours: on this last, see Kearns (1989), 111, which compares the example of the Elean Sosipolis, and makes the point that the protecting snake uses its power to terrify to the advantage of its chosen human community. She aptly describes this as ‘a belligerent protectiveness’.

175 Lyk. Al. 1237 and schol. ad loc.

176 Pausanias (6.26.1) gives considerable information concerning the important cult of Dionysos at Elis, and in particular a festival called the Thyia, but it is by no means certain that the hymn recounted by Plutarch is to be connected with this context. Exhaustive discussion may be found in Schlesier (2002).

177 Plut. Quaest. Gr. 36. On the hymn, see Daraki (1985); Schlesier (2002). Further difficulty is added to an already cryptic source by considerable uncertainty over the text; for example, an alternative reading of thuôn is duôn (‘plunging’, ‘descending’), which would convey a rather different set of associations. The reading thuôn is technically problematic (Schlesier 2002, 161-22 ns. 5 and 7), but the explanations of the alternative (that it refers to a sacred cave of Dionysos at Elis into which the god was being invited to plunge) feels awkward and finds no secure corroboration in archaeology or in ancient literature. The matter must perforce remain uncertain; the reading of ‘with bull-foot’ is not, however, in doubt.

178 The level of authorial uncertainty in this section is indeed high, as Schlesier (2002, 163) notes ruefully.

179 The nature, and indeed the very existence of coherent Orphic ideas are a notorious source of controversy. Here it is not necessary to become involved in the debate, since no claim is made that the mixanthropic Dionysos was a product of any one religious system or affiliation. On the relationship between Orpheus and Dionysos, see Guthrie (1952), 107-33; Linforth (1941), 307-64.

180 Ant. Lib. Met. 10.

181 Nonn. Dion. 6.179-205. On the special rôle of metamorphosis within the Dionysiaka, see Buxton (2009), 143-53.

182 Aelian, HA 12.34: Τενέδιοι δὲ τῷ ἀνθρωπορραίστῃ Διονύσῳ τρέφουσι κύουσαν βοῦν, τεκοῦσαν δὲ ἄρα αὐτὴν οἷα δήπου λεχὼ θεραπεύουσιν. τὸ δὲ ἀρτιγενὲς βρέφος καταθύουσιν ὑποδήσαντες κοθόρνους. γε μὴν πατάξας αὐτὸ τῷ πελέκει λίθοις βάλλεται δημοσίᾳ, καὶ ἔστε ἐπὶ τὴν θάλατταν φεύγει.

183 For an Argive ritual in which Dionysos is addressed as Bougenês, ‘Cow-born’, see Plut. de Isid. et Osir. 35. On this detail he is, however, citing one Sokrates, probably a third-century BC Argive historian (see RE s.v. ‘Sokrates’); if this identification is correct, it would make the source for this passage relatively early and potentially very well-informed.

184 The equation of Dionysos with the sacrificial victim is famously strong. As Daraki (1985, 65) remarks, ‘Le ‘sacrifice dionysiaque par excellence’ est une mise à mort du dieu.’ See also Seaford (2006), 84-6. Obbink (1993) is more sceptical: he casts doubt on the idea that the Greeks explicitly equated Dionysos with the sacrificial animal, and makes the point – admittedly important – that the myths of the Titans’ killing of Zagreus does not reflect the pattern of sacrificial ritual in Dionysos’ cult. The Tenedian ritual, he observes, is a very isolated instance. (Obbink [1993], 67-75.)

185 This matter is treated in a fascinating discussion by Daraki (1985, 45-71), in which she questions the relationship between the rite of ômophagia (the rending and eating of raw flesh), and the associations which Dionysos also had with the world of vegetation and with vegetarian offerings. In the former, she argues (68-71), his devotees mimic the animal savagery of the god. Dionysos is slain and slayer (see pp. 62-3), aggressor and victim. Interestingly, Dionysos appears to have received occasional human sacrifices, though these are not associated with his mixanthropic form: see Hughes (1991), 111-15.

186 Interestingly, another of Dionysos’ bull-related cult titles is ‘Bull-eater’. The Suda entry (s.v. ‘taurophagon’) explains this by reference to the sacrifice of an ox to Dionysos, but also mentions the custom of ômophagia, the eating of raw flesh. Cryptic though this is, it suggests that the sacrifice explanation is not enough, and that we are dealing with further divine violence, this time directed at the very animal in which the god is sometimes incarnate. The violence of maenads, the followers of Dionysos, against animals, seems to be reflected in this aspect of the god himself; or is it the other way round? Do the destructive tendencies of the maenads reflect the destructive potential of the deity they serve?

187 Cf. lines 920-22: when Pentheus sees Dionysos as a bull, this is simultaneously delusional and truthful. See Jaccottet, vol. 1 (2003), 103.

188 Deipn. 11.476.

189 de Isid. et Osir. 35.

190 LIMC s.v. ‘Dionysos’, cat. no. 156 seems the only certain instance, a third-century coin thought to be of the Brettioi, showing a nude, bearded and horned Dionysos. Numerous other items in the catalogue may or may not depict the god.

191 LIMC s.v. ‘Dionysos’, cat. nos. 157-9.

192 For the text of the hymns, see Quandt (1962); the most recent edition, however, is that of Ricciardelli (2000). On their date, nature and context, see Guthrie (1952), 257-61; Linforth (1973), 179-89; West (1983), 28-9. For a recent detailed discussion of their form and content, see Rudhardt (2008); see also Morand (2001).

193 This is a fifth-century AD text, which immediately necessitates some caution in the present study, although there is no doubt that the work makes use of many long-standing elements of Greek mythology. On date and context, see Vian (1976), XV-XVIII; Lindsay (1965), 359-65.

194 Rudhardt (2008), 171-2.

195 For discussion of similarities between the texts, see Morand (2001), 83-6.

196 Orphic Hymn to D. 30, line 6, calls him the son of Persephone. In Nonnos, this Dionysos is also called Zagreus; he is killed by the Titans, after which Zeus replaces him with a second Dionysos, this one the son of Semele. See Nonn. Dion. 5.562 ff. and 6.155 ff. (The murder of Dionysos/Zagreus ties in with another feature prevalent in these sources: the close identification of Dionysos with the bull killed in sacrifice, discussed above). On the death and rebirth of Dionysos in Orphic literature, and its wider significance within Greek thought, see West (1983), 140-75.

197 On the relationship in the Hymns between material particular to them and material known from elsewhere, see Rudhardt (2008), 266-74, discussing especially the dual maternity of Dionysos (Semele and Persephone).

198 For example, echoes do appear to exist between the Hymns (especially their choices of deity and their language) and inscriptions and other artefacts found in Pergamon, and it may be that the religious climate of Asia Minor at various periods had a relatively high concentration of the names and principles associated with the Orphic literature; but this does not allow for a definite claim that an Orphic community dwelling in Pergamon was the sole source of the Hymns and used them consistently in its ritual (pace Guthrie [1952, 258-61], who argues strongly for a Pergamene centre of cult).

199 For example, in Orph. Hymn 45, line 1, Dionysos is called taurometôpe.

200 This is the most common element, and appears in early sources: see Eur. Bacch. 100, 921; Soph. fr. 874 Nauck; also Nonn. Dion. 6.165, 9.14.

201 As in the Elean hymn cited in Plutarch’s 36th Greek Question. Another example occurs at Soph. Ant. 1143.

Table des illustrations

Titre Fig. 19
URL http://books.openedition.org/pulg/docannexe/image/1619/img-1.png
Fichier image/png, 105k
Titre Fig. 20
URL http://books.openedition.org/pulg/docannexe/image/1619/img-2.png
Fichier image/png, 210k
Titre Fig. 21
URL http://books.openedition.org/pulg/docannexe/image/1619/img-3.png
Fichier image/png, 73k
Titre Fig. 22
URL http://books.openedition.org/pulg/docannexe/image/1619/img-4.png
Fichier image/png, 117k
Titre Fig. 23
URL http://books.openedition.org/pulg/docannexe/image/1619/img-5.jpg
Fichier image/jpeg, 184k
Titre Fig 24
URL http://books.openedition.org/pulg/docannexe/image/1619/img-6.jpg
Fichier image/jpeg, 400k
Titre Fig. 25
URL http://books.openedition.org/pulg/docannexe/image/1619/img-7.png
Fichier image/png, 95k
Titre Fig. 26
URL http://books.openedition.org/pulg/docannexe/image/1619/img-8.png
Fichier image/png, 178k
Titre Fig. 27
URL http://books.openedition.org/pulg/docannexe/image/1619/img-9.png
Fichier image/png, 130k

© Presses universitaires de Liège, 2011

Conditions d’utilisation : http://www.openedition.org/6540

Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search