http://www.jstor.orgArtemis and IphigeneiaAuthor(s): Hugh Lloyd-JonesSource: The Journal of Hellenic Studies, Vol. 103, (1983), pp. 87-102Published by: The Society for the Promotion of Hellenic StudiesStable URL: http://www.jstor.org/stable/630530Accessed: 24/05/2008 14:40Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available athttp://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unlessyou have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and youmay use content in the JSTOR archive only for your personal, non-commercial use.Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained athttp://www.jstor.org/action/showPublisher?publisherCode=hellenic.Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printedpage of such transmission.JSTOR is a not-for-profit organization founded in 1995 to build trusted digital archives for scholarship. We enable thescholarly community to preserve their work and the materials they rely upon, and to build a common research platform thatpromotes the discovery and use of these resources. For more information about JSTOR, please contact support@jstor.org.Journal of HellenicStudies ciii (1983)87-102 ARTEMISANDIPHIGENEIA* (i) The sacrificeof Iphigeneia in Aeschylus: I FEW problems intheOresteia havebeenmoredebatedinrecenttimesthanthatof why Artemissends thecalmwhichdetainstheAchaeanfleetat Aulis.'Themain problem, a much vexedone',wroteEduard Fraenkel,'. .. arises fromthe fact that wearerenottold anywhere in the ode why the wrath ofArtemisis directed against the Atreidae'. By 'the ode'Fraenkel means, of course, the parodos ofthe Agamemnon; he might have written'in the play' or 'in the trilogy'. He believedthat Aeschylus had in mindthe story, toldin the Cypria and in the Electra of Sophocles, that Agamemnon had angered Artemis by boasting that he surpassed her as an archer, but that he madenoallusiontoitbecause itseemedto supply a motivetoo petty toaccord withhis great theme. Others have protested that in a trilogy concernedwith guilt and retributionit is strange that so important anactionshouldbeleftunmotivated.Sir DenysPage revivedtheview of Conington and othersthatArtemisis angry withtheAtreidaibecause they are symbolisedby the eagles which appear at Aulis, andthese eagles killa pregnant hare, ananimalunderher protection.2 Thatistoconfusetheworldofthe portent withtheworldof reality whichit symbolises, besides assigning tothe goddess a motiveofa still more objectionablepettiness. In pointing this out, twentyyears ago, I contendedthat since the eagles stand fortheAtreidaithe hare muststand for Troy, and deducedfromthis that Artemismust be angry at the prospective massacre ofthe Trojans. InHomerandinthelater tradition, Artemistakesthe Trojan side against the Greeks, and I argued that since the hare stoodfor Troy the partiality ofArtemisfor wildanimalsandtheir young muststandforher partiality fortheTrojans.3 Somecriticshave objected thatthis theory tooascribestothe goddess amotive incommensuratewiththe greatness ofthetheme.N.G.L.HammondwritesthatArtemis loathesthefeast ofthe eagles(Ag.137) 'becausesheloathesthebloodshedofthewarwhich Agamemnon and Menelaus are starting'; she is 'the goddess of the weakand helpless ...and she abominatesthe brutality ofthe impending war';4 several otherscholars have agreed withhim.5 * My Oxford colleagues C.W. Macleodt, R. C. T. Parkerand N.J. Richardsonall readan early draftof this paper and helped meto improve it.I am grateful to Prof. Joanne P. Waghorn for the information given in n. I7, whichis likely tobe supplemented inmost interesting fashion by her own publications; and I have profited by being allowed to read a so far unpublished dissertation onthecultofBrauron withwhichMiss Paula Perlman obtained the degree ofM.A.from the University ofCalifornia at Berkeley some years ago. Professor Albert Henrichs allowed me an early viewof thevaluable article ofhis mentioned below, and also supplied useful criticism; and I had help and encourage- ment from the scholar without whose work this article could not have been written, Prof. Walter Burkert.The following abbreviationsare used: Brelich: Angelo Brelich, Paides eParthenoi (Rome I969) Burkert HN: Walter Burkert, HomoNecans (Berlin etc. 1972) Burkert GR: id., Griechische Religion in der archaischen und classischen Epoche (Stuttgart1977) Calame: Claude Calame, Les choeursde jeunes filles en Grece archaique i (Rome1977) Deubner: Ludwig Deubner, AttischeFeste (Berlin 1932) Henrichs: Albert Henrichs, 'Human sacrifice inGreek religion: three case studies', Entr.Hardtxxvii (198i) 195 ff Honn: Karl H6nn, Artemis:die GestaltwandeleinerGattin (Zurich 1946) Meuli:Karl Meuli,Gesammelte Schriften (Basel etc. 1975) NilssonGF:M.P. Nilsson, GriechischeFestevon religioser Bedeutung (Leipzig1906) Nilsson GGR 3: id., Geschichteder griechischenReligion3 (Munich 1967) Aeschylus,Agamemnon (Oxford1950) ii p. 97. 2 Aeschylus,Agamemnon, ed. J. D.Denniston and D. L. Page (Oxford 1957) xxiii f.; cf. A. Lesky, Die tragische Dichtung der Hellenen3 (1972) I6. 3 CQxii (1962)187; cf. p.23 of mytranslation, Aeschylus:Oresteia-Agamemnon2 (London 1979). R.H. Klausen wasthefirst to point outthat Artemis was angry notwiththe birds but with the men they stood for; see his TheologumenaAeschyliTragici(Berlin 1829), and cf. P.M. Smith, 'Onthe Hymn toZeusin Aeschylus' Agamemnon', Amer. Cl. Stud. v (1980) 76 n. IOI. 4JHS lxxxv (1965)42 f.=StudiesinGreek History (Oxford1973)395 f. 5 E.g. J.J. Peradotto, Phoenixxxiii (1969)237f.; M. Ewans, Ramus iv (I975) I7f.; P.M.Smith (n.3) loc. cit. Inanarticle published just toolateformetotakenoteof it, WilliamWhallon argued that Artemiswas angry not only because ofthe impending massacre at Troy, but because ofthe guilt of Agamemnon's father Atreus, whohadservedhis brother Thyestes withhisownchildren's flesh: 'becauseofthe tecnophagy Artemisis impelledby remorselessvindictiveness'.6Hetoo stresses the role ofArtemisas the protectress ofthe weakand helpless; 'what shapes her part', he writes, 'in the vast schemeofdestructionwreaked upon the Atreidae is her lovefor any kind of helplessoffspring'. Butif as all these critics thinkArtemisis impelledby her hatred of bloodshedin general and her loveof young creatures in particular, why doesshe act ina way that causes thesacrifice of Iphigeneia? This consideration leadsJ.J. Peradottoto suggest that Artemisdoes not really intend that Iphigeneia shall be sacrificed; she stills the winds, he argues,simply to keep thefleetfrom sailing; so also HeinzNeitzel, who quotes Wilamowitzas denying that the protectress of the hare can possibly demandthebloodofthe virgin.7 Thosescholars whoassume that the kindlygoddess who protects the young of wildanimals is at all times averse from bloodshedare taking it for granted that the idea of Artemisentertained byAeschylus differeda good dealfromthecharacter assigned her by the religion ofarchaic Greece. In that religion the goddess cherishes the young of animals because they are her own; yet at the same time she is the huntress who destroys them.'In fact', writes Walter Burkert, 'Artemis is andremainsamistress of bloody sacrifices'.8 In recent years ithas beenestablishedthattheideaofsacrifice plays a notableroleinthe Oresteia.9 The eagles are said to'sacrifice' thehare (Ag.137), and Kalchas fears that in requital Artemis may hasten on'anothersacrifice' (Ag.150). Iphigeneia is sacrificed toArtemis;there is no mentionof the story, commonto the Cypria and the pseudo-HesiodicCatalogue,10 that at the last minute the sacrificewas averted; and years later Klytemnestra will assumeher daughter to have perished. Later the alastor is said to have sacrificed a grown man in requital for youngones; the plural seemstoindicatethatthechildrenof Thyestes as wellas herown daughter are in Klytemnestra's mind.1" Weare all agreed that the tragedians werefree tomake use ofancient legends without troubling themselvesabout their origin, or about their original significance. Butthe stress which Kalchas' accountofArtemis'concernfor young creatures hashas ledsomescholars to lay upon her gentle side must not lead us to forget the side ofthe goddess whichis not gentle. If wewish fully to explain the part in the action played by this deity, whose connection with sacrificeis so close, it may prove helpful to investigate the origin of the legend that Agememnon could not lead his armyagainstTroy without sacrificing his daughter. (ii) Human sacrifice inGreek myth Greek myth contains anumber of legends abouthumansacrifice, whichhaveoften encouraged scholarsto conjecture that at an early stage of their history the Greeks, or perhaps their ancestors,may have been in the habit of sacrificing human victims. In the masterly book HomoNecansinwhichhehasdemonstratedthecentral placeoccupiedby sacrifice inGreek religion, Burkerthas remindedus that legends of human sacrificeare associatedwith some of the principal Greek festivals celebratedat holy places. One notable example is the part played by Pelops at Olympia; another is that played byLykaon in Arcadia;at Delphi, Nemea and the 6 AJP lxxxii (1961)78 f. 150 f.; A. Lebeck, The Oresteia:a Study in Language and 7 Peradotto (n. 5); Neitzel,Hermescvii (1979) 10.Structure (Washington 1971) index s.v.'sacrifice'. 8 Burkert GR 237. 10See Proclus, Chrestomathiain Allen, Homeri Opera 9 See F. Zeitlin, TAPA xcvi (1965) 401 f. and xcviiv 104.12 f.A. Severyns, Recherchessur la Chrestomathie (1966)645f.;P. Vidal-Naquet, Par. del Pass. cxxixde Proclosiv (Paris1963)82,135f.;Hesiodfr.23a.17 (1969)401 f. =J. P.VernantandP. Vidal-Naquet, M.-W. Tragedy and Myth inAncientGreece (Brighton1981) 1 Ag.1504. (trans. of Mythe et tragedie en Greceancienne [Paris 1972]) 88HUGH LLOYD-JONES ARTEMISANDIPHIGENEIA Isthmus, Neoptolemos, Archemorosand Melikertesseem to occupy positions similarto that of Pelops. Several legends tell ofthe sacrificeof young persons, usually female, toensure successin war.12 Thebes cannot repel the Argive onslaught without the death of Kreon's son, whether Megareus or Menoikeus; Erechtheus cannot defeat Eumolpos withoutthesacrifice ofhis daughters;Demophon and lolaos cannot defeat Eurystheus without the sacrificeofMakaria. Thebes, even with the aid of Herakles, cannot defeat Orchomenos without the sacrificeof the daughters of Antipoinos; neither can Messenia repel the Spartans without the sacrificeofthe daughters of Aristodemos.The publication of the Sorbonne papyruscontaining new fragments of Euripides' Erechtheushas reminded us that such legends sometimes stood in close relation to religious observances practised in historicaltimes. Before an Athenian army took the field, its general sacrificedin the sanctuary of the Hyakinthides, who were identifiedwith the daughters of Erechtheus. 13 Can it be argued from these myths, in conjunction with such archaeological evidence as has from time totime been brought forward to try toshow that human sacrificewas practised during the Bronze Age, that at one time the Greeks, or at least their forebears, sacrificedhuman victims?With admirablecaution Albert Henrichshas lately surveyed the entire body of relevant material.14He rightly concludesthat the evidence for the practice of human sacrifice during the archaic period, not to mention the classicaland Hellenistic periods, is insufficient, and that even for the Bronze Age the evidence that hasbeen brought forwardleaves much room for doubt. He is right, also, to point out that the known fact that in historicaltimes a beastwas often sacrificed, in theory, as a replacement for a human victim does not by itself sufficeto prove that in earlier times a human victim had indeed been sacrificed.At the same time, the possibility can hardly be discounted.The evidence for both cannibalismand the ritual killing of childrenin palaeolithic times cannot be disputed; and even if the practice had become obsolete before the Bronze Age, the memory may have endured. Archaeologicalevidence, cited by Henrichs,has confirmed the statementsof Greek and Roman historiansthat the Carthaginians sacrificedchildrento Moloch at times of crisis.That meansthat the practice is attestedfor the Mediterraneanbasinas late as the fourth and thirdcenturies BC;but for our presentpurpose what mattersmost is that the Greeksof the historical age believed that it had once existed. (iii) The originsof sacrifice Keeping this in mind, let us consider what seems to me the most persuasivetheory of the origins ofsacrifice that has beenadvanced inmodern times.Karl Meulistarted fromthe observationthat during the infinitely long period of human history before the introduction of pasturage andthestilllater introduction of agriculture, human lifemusthave depended principallyupon the huntingexpeditions undertaken by themalemembersofeach community.15 Burials ofbones inSiberia and other places indicate that the hunters tried to placate the spirits of the animals they had slain by offering to returnto them parts of themselves; they offered, naturally, those parts which they themselves did not need for food and clothing. Very commonly such an offering was fixed to or suspended from a tree; for trees, like other plants, seem to perish in winter, but come alive again in spring. In earlyreligion, any tree may be atree of life;and Meulihas connected this withthe prevalence oftree worship and the importance oftrees in the worship of many deities. In the earliesttimes the fear of ghosts must have assumed an overwhelming importance. 12 See Burkert HN 77 f.; C.Austin, Nova fragmentaStengel, Die gr. Kultusaltertiimer (Munich 1920) I28f.; Euripideafr.50.16 f. and fr. 65.65-89; Eur. Hcld. 408f.;Nilsson,GGR 3,indexs.v. 'Menschenopfer'; Burkert Paus.ix 17.1;iv9.2,5. GR,indexs.v.id. 13 See Burkert HN 78. 15 See his 'Griechische Opferbriuche' in Phyllobolia 14 Henrichs;cf. F.Schwenn,Die Menschenopfer beifurP.vonder Miihll (Basel 1946)I85 f.,=GSii907f. den Griechen und Romern, RGVV xv.3(Giessen I9I 5); P. 89 Later, it was supplemented, and toa great extenteven displaced, by the fear of gods and spirits. The offering to a ghost becamean offering to a god,paradoxical as it may seemthat menshould offertoa god those parts ofthe slaughtered beastthat they donotthemselves require: thisis Meuli'ssolutionofthe problem whichtheHesiodic story ofthe deception ofZeus by Prometheuswas designed to solve, andwhichstill provided thecomic poets ofAthenswith materialfor jokes.16 (iv) The Mistress of Animals In the beginning, such offeringsmay wellhavebeenmadetono god or spirit in particular, but to a mysterious collective.Butlater they must have been made to a particular divinity. In the earliest Greek religion ofwhich anything isknownto us, theforestsandthewildswerethe domainofthe great goddess whois referred to by the nameofthe Mistress of Animals, 7rorvta OqpC6v. 17It is a reasonablesurmisethat the forebearsof the Greekstried to appease the Mistress ofAnimalsfor the killing ofbeasts which they believedtobe her property. Beforethe hunting group of males disappeared intoher domain, to risk death daily overa longperiod, someform of sacrifice is likely tohavebeenofferedinthe hope of propitiating theformidable goddess. IntheearliestGreek religion,goddesses werelittledifferentiatedfromoneanother: structuralists, withtheir tendency to neglect thehistorical dimension, woulddowellto rememberthatthe precisemarking offofone deity fromanotherintermsofattributesand function may not safely beattributedtotheremote past. Thusitisnot surprising thatthe Mistress of Animals has different heiresses in different places. She is recalled by certain features of theSamian and Argive Hera, the Tegean Athena Alea, the Cybele and AnahitaofAsia Minor and variouscults ofDemeterand Persephone; but her usual heiress in historictimes is, as weall know, Artemis.Artemis incorporated various local goddesses whohad inherited certain features of the Mistress of Animals, such as Aphaia in Aegina,Diktynna and Britomartis in Crete, Hekate in manyplaces. Fromthe comparatively latetimewhenshebecame Apollo'ssister, Artemis tookon Apollinecharacteristics;butthe virginhuntress,chasteand fair, whois already establishedin the Homeric epics, is a very different person from the Mistressof Animals as she appears inthemoreancientart,evenifonedoesnot go backso early asherCretan manifestations.8 (v) The worship of Artemis In the worship of Artemis in historical times, many characteristics ofthe Mistress of Animals persisted. That worship wascommoninthe Peloponnese,includingArcadia,that remoteand mountainous country whereas thedialectconfirms part ofthe pre-Dorianpopulation ofthe Peloponnese took refuge. Oneofthemost widely diffused cultswasthat ofArtemis Laphria, whichNilsson19 guessed tohavestarted inAetolia,spread to Naupaktos and across the Gulf of Corinthto Patrai, and then westwardsto Kephallenia and eastwards to Delphi and Hyampolis in Phokis.Atthe feast ofthe goddess celebrated in Patrai, not only fruits but living birds and beasts werehurledontotheflamesofthe great firethat blazedaroundthealtar.20 Thealliedcultof 16 Hes.Th. 535 with West ad loc.; Menander, Dysc. tiated withblood.Onelooksforward eagerly tothe 447 f.,with Handley and Sandbachad loc.publication ofProf. Waghorn's investigation ofthe 17 SeeNilsson,The Minoan-Mycenaean Religion2religion ofSouthernIndia, a subject thathasbeen (Lund 1950)503 f.; GGR 3 index s.v. 'rro'rva 0-qpcJv'; neglected by western scholars. Burkert GR 233 f.; B. C. Dietrich, The Originsof Greek 18 Whether or notthe name Artemis occurs inthe Religion(Berlin etc. I974)146 f.Prof. Joanne P. Pylos tabletsremainsuncertain; seeC.Sourvinou- Waghorn, of the University of Massachusettsat Boston,Inwood, Kadmosix (1970)42, and other literaturecited hasdrawn my attentiontothe startlingparallelby Burkert GR 85 n. 23. presentedby the goddess whoinSouthernIndia19 GF 218f. presides over the forest and the battle-field, causesdeath 20For a discussionof this rite, with full bibliography, inchildhood,receives offerings intrees and is propi- see Giulia Piccaluga, 'L'olocausto di Patrai', Entr.Hardt HUGH LLOYD-JONES 90 ARTEMISANDIPHIGENEIA Artemis Triklaria, closely linkedwiththatof DionysosAisymnetes, hadacult legend ofthe sacrifice ofa young man and a young woman.21In the famouscult ofArtemisOrthia at Sparta therewasabundantuseofmaskswhich representedmany differentanimals;22 just sointhe Syracusan feast ofArtemisdescribed by Theocritus'there were many beasts inthe procession aroundthe goddess, andonewasa lioness'.23 Artemisis oftenconnectedwithtrees and vegetation; oftenshe is connectedwith fertility. Asin myth shehas her entourage of nymphs,corresponding withthe satyrs whoattend on Dionysos, so in cultshe is oftenhonoured byyounggirls: they were virgins,though the dances which they performed in honourof the goddess were not always decorous.At Olympia, the title ArtemisKordax24indicatestheir character; similar things are reported ofthe Spartan cults of Artemis Korythalia and ArtemisDereatis.25Like cults ofArtemisas far off as Syracuse, Delos, Lesbos and Ephesos, thecultofOrthiahad its dances; their nature may lend plausibility tothe notionthat the cult title, susceptible as it is of several different explanationsmay not be altogether freeof phallicsignificance.26 Artemisisconnectedwith girls, butalsowith young men: in Sparta the ephebes were flogged at thealtar ofArtemis Orthia, and inAthens they swore their oathinthe temple ofArtemis Agrotera, thesame goddess whoreceivedasacrificeoffive hundred goats as a thank-offering forheraidatMarathon.27 (vi)Iphigeneia and the cult of Brauron In Attica, an importantplace inrelationtothecultofArtemis belonged to Iphigeneia. Athena in her speech fromthe machine near the end of Euripides' Iphigeneia in Tauris commands Orestesand Iphigeneia to convey thestatue ofArtemisfromthe TauricChersonesetotheeast coastofAttica. There, atHalai Araphenides, OrestesistoinstitutethecultofArtemis Tauropolos; at Brauron, only twomiles to the south, Iphigeneia is to institute the cult of Artemis Brauronia, whose priestess she is tobecome.Whenshe dies, she is tobe buried inthe precinct, andtheclothesofwomenwhodieinchildbirthare tobededicatedattheshrine.Theclear statementofthe Euripidean Athenathat thecults ofHalai and Brauronare quite distinctfrom oneanotherisconfirmed by the testimony of Strabo, sothattheonce prevalent habitof supposing that they werethesamewasa deplorable mistake.28 (a) ThecultatBrauron Unfortunately weknowlittleabouttheritual ofeitherofthetwo cults; butwhatwedo knowoftheactions performed andthedress worn by the participants intherituals ofboth suggests that theybelongtogether withotherknowninitiationrituals that are associated with xxvii (1981)243 f.She observes that the holocaust did notincludecornor wine, andis surelyright in connecting itwiththetransition froman economy based on hunting to one basedon cultivation. But likeJ. P. Vernant (283-4 of the same volume), I have doubts about the detailsof her structuralist interpretation of the connectionbetweentheholocaustandthe myth of Oeneus, and in particular about her contention that by choosing cultivation in preference to hunting Oeneus was choosing toremain mortal rather than tobecome immortal.Afterthefashionofstructuralists, Prof. Piccaluga combines differentforms of the myth that are attestedat differenttimes and places, as though we could be sure that each formed part of a unitary complex; and inher preoccupation with forming a neat pattern she resolutely averts her gaze from the unusual cruelty of therite and the effect whichitmust surely have had uponparticipants(252 n. I, and see G.S.Kirk on p. 280). She seems to think the holocaust was a deliberate defiance of Artemis; but surely itwasan attempt to placate her by offering her things that were her own, and naturally not adding corn and wine, in which she had no part. While conceding that the Artemisof Kalydonpreserved some elementsof the Mistressof Animals, Prof. Piccaluga is scornfulof thosewho have seen the Artemisofthis ritualas the heiressofthat divinity(245-6,250-I);yet her own theory contains nothing thatis inconsistentwith that supposition. 21 NilssonGF216 f.; Calame 73,245,273. 22 Nilsson GF I90;cf. R. M. Dawkins,JHS Suppl. v (1929); Calame 276 f. 23 ii66-8. 24 NilssonGF 187; BurkertHN II7. 25 NilssonGF 183; Calame 297f.; onArtemis Dereatis, see Calame 302. 26 NilssonGF I9I; on the complicatedquestion of the name's possibleimplications, see Calame 289 f. 27 Deubner 209. 28 Eur. IT 1449f.; Strabo 399; see Deubner 208. 9I puberty, ritualsin which an animalwas sacrificedas a surrogate for a human being. The festival of the Tauropolia, like many other festivalsof Artemis, involved dancing at night by chorusesof girls; this isthefestival at whichinMenander's Epitrepontes Charisios rapes his future wife Pamphile. According to the Euripidean Athena, the ritualinvolved the mock sacrificeof a man. 'When the people celebrate thefeast', she says (1458 f.), 'in memory of yournearly being sacrificed, let someone hold a sword to a man's neck and draw blood, for an appeasement and so that the goddess may be honoured.' The sanctuary ofBrauron was excavated by John Papadimitriou, Director ofthe Greek Archaeological Service, between 1948 and his suddendeath in 1963. Important discoverieswere made, and many interestingobjects may be seen, though not photographed, in the museum on the site; but lamentably the results ofthe excavation remain unpublished.29 If and when the inscriptions found during the dig are given to the world, we may learn more about the cult of Brauron and its ritual. In themeantime wewouldhave to depend solely onthescattered testimonies found in ancient authors, had notMme Lily Kahil published valuable studies of certainvasesfound at Brauronand at other places which are highly relevantto the problems now under discussion.30 Aristophanes Pax 872-6 seems to suggest that in the late fifth century the Brauroniatook place every four years; but this was not necessarily so in early times, and as weshall see the analogy of comparable festivals suggests that originally the riteshad been celebrated annually.31 They involved a sacrifice, the beast sacrificed, as in many other cults of Artemis, being a goat. The sacrificewas offered tothe goddess by younggirls known as 'bears'; they wore, at least during part ofthe ritual, the saffron-colouredrobe called the krokotos.Themost celebrated allusion to them is in Aristophanes'Lysistrata 638 f. The chorus of militant womenwhohave seized the Acropolis is about tooffer advice tothe citizens, as comic choruses often doin a parabasis. Inorder toestablish their credentials, theyparody thekindofrecital oftheir qualifications which men might have given in similar circumstances byenumerating various religious dutieswhich young girls might have performed in Athens, claiming to have performed all of them themselves.'We are setting out, all you citizens', they say, 'to say something usefulto the city, as wewell may, because it rearedme in splendid affluence.From the moment I was seven I served as arrhephoros; then at ten I was a baker for the Archegetis; then I had my krokotos and was a bear at the Brauronia; and I was once kanephoros, a lovely girl with a bunch of figs.' Historical specialists who rightly see Aristophanes as a priceless source of informationabout Athenian life sometimes forget that itis not his primary purpose to provide them with such information, he being not a historianor an antiquarian but a comic poet. Some of them have been incautiousin taking the women's word for it that they can really have performed all these duties, and that Athenianwomen regularlyperformed them all. But Wilamowitz in his note on the passage32rightly remindsus that in each year only two girls servedas arrhephoroi, though in 29 Something may be learned from Papadimitriou's article in Scientific Americanccviii (1963) 118 and from his situation reports in Praktikaand Ergon and those of G. Daux in BCH between I949 and I963. 30AKviii (I965) 20 f.; Beih.i (1968) 5f.; CRAI 1976, I26 f.; AKxx (1977) 86 f.A thorough, though rather literal-minded, discussion ofthe cult, witha precious collection oftestimonia, is in Brelich 241 f.; see alsoJ: D. Kondis, ADelt xxii (I967) I56 f.; Calame 186 f.; Henrichs n. 35. 31 Brelich 276 takes too literally the explanation of the passage of Aristophanes that is offered in the scholia and by the Suda s.v. 'Bpavpwv': ev Bpavpcv vL 8 8qrl,a T7S'AArTLKiSTroAAal Tropvat, EKELSEKalA ovvala fye-TO, KatKaGaE'Kaarov %8rjov, eV OlTS E,tEOvOV, p,EOVovTrs Se rToAAas 7rTopvas 7lp7Traov. Theinnocent rural demeofBrauron cannot havebeena red-light district:the excavations indicate that there had been no settlementtheresince Mycenaean times.Asforthe alleged Dionysia, they were surely invented in order to explain V7TO7TErrTW KOTEs (874), just asthe prostitutes and the absurdnotion that people got drunk and seized them was invented in a feeble effort to explain the text. 32 Wilamowitz, Aristophanes, Lysistrate (Berlin1927) 162.Line 643 appears in modern texts asKalT' E'ouv'a rov KpOKWTOVapKTrLos ? Bpavpwvtots. Ka' Exovaa is an interpretation ofr'sKarE'xovra usually ascribedto Bentley; IowetoProf.Henrichs theawareness that Bentley was anticipatedby the Flemish scholarNicasius Ellebodius (see F. Schreiber, TAPA cv [19751328). But theRavennashas KaTraxEovaa, andthis reading has beendefended by C.Sourvinou (-Inwood),CQ xxi (1971)339, whoinsertsa stop after dACrpts' j in 644, removesoneafter TaPX7ycE7T attheendofthat line HUGH LLOYD-JONES 92 ARTEMIS ANDIPHIGENEIA all likelihood they represented all girls born at the same time. About the 'bakers'weknow nothing; but at Athens 'the Archegetis' is likely to be Athena, and they presumably bakedfor her the special cakes required for some religiouspurpose. The kanephoroi referredto are presumably those whowalked in the procession at the Great Panathenaia, a duty which was certainly not performedby all Athenian girls. A scholion on the passage containsan aitionwhich implies that all Athenian women served as 'bears', but it would not be wise to believe this statementon no better authority thanthatofanancientscholarwho may wellhavebeen guessing irresponsibly.33 We do not know how many 'bears'there were, but all analogy suggests that a small numberof girls represented their entire age group. Neitherdo weknowwhetherit is true, as some have guessed, that they lived for some time in the sanctuary, as the arrhephoroi lived on the Acropolis, undergoing a 'reclusion'of a type familiarin initiation rites.34 Thevasesstudied by MmeKahilare likesmall mixing-bowls, andarethereforetermed krateriskoi. They havebeenfoundnot only at Brauronandat Halai, butalso intheshrines of Artemis Mounychia at Piraeus, ofArtemisAristoboulenearthe Agora, andofPanandthe Nymphs in the Agora and nearEleusis.It is not surprising that they should be found in the shrine of Artemis Mounychia, since two ancientauthorities say that 'bears'took part in her ritesas well as in those of Artemis Brauronia.The vasesshow young girls, sometimes wearing a short chiton, but sometimes naked. Sometimes they are running a race towards an altarwith a flame burning on it, or towards a palm-tree, an object often associatedwith Artemis and with Apollo; on some vases the runners carry torches. On other vases they are not running a race, but dancing what seems to be a slow and solemn dance, or executing rhythmic steps, near the altarand at the tree; sometimes they are carrying garlands. The words of Aristophanes have been taken to imply that the girls were over ten; but a scholion on the passagesays that they were between five and ten, and the very fact that this figure differsfrom what is in the text causesone to hesitatebefore concluding that it is mistaken.On the vases some girls seem very young, others a little older; Semni Karouzou35thinks they average seven or eight years old, which would suit the statementin the scholion. It may well be that at some time the age of the participants was altered;if the obvious explanation of the rites is the correctone, they were originally girls approachingpuberty. The girls represented in the statues dedicated in the sanctuary, which date from the fourth and third centuries, are teenage girls ratherthan childrenor young women; but we must rememberthat we have no positive reason to suppose that these statues represent 'bears'. We know that the animalsacrificedin the rite at Brauronwas a goat. Now according to the Cypria the animal substitutedfor Iphigeneia was a hind; but Phanodemos says it was a bear,36 and the statement ofTzetzes that Iphigeneia was turned into a bear seems to carry the same implication. But a goat was the beast commonly sacrificedto Artemis, and in historicaltimes goats were more easily procurable than either hinds or bears. The temple legend of the Mounychia cult37 was that a bearwas once killed, so that a plague (which she surprisingly thinksrefersto Artemisrather 34 Some people have supposed that the buildings thanto Athena), andfindsa referenceto the shedding ofroundthe stoacontaineda dormitory for thebears;but robescommonin initiationrites (seebelow) both hereforallweknow they contained dining-roomsor andat Aesch. Ag. 239f. Dr Sourvinou may be right, butguest-housessuchas therewerein many sanctuaries.Dr her viewinvolves an awkward punctuation and anRichardsonrefersme to J. J. Coulton,TheArchitectural insistence upon one particularstage of theritualwhichI Developmentof the Greek Stoa (Oxford 1976) 9 and 43. find surprising. T. C. W. Stinton,CQ xxvi (1976) I I f., To supplement the one inscription fromBrauronso far who readsKal X~Eovcra, avoids the former drawbackbutpublished (by Papadimitriou [n.29] loc.cit.; cf.L. not the latter.Robert, REG lxxv [I963]135, Brelich 260n.60and 33 P. Vidal-Naquet inComment fairel'histoire iiiKondis [n. 30]) inthissenseisincautious;other (1974)37 f .=Myth, Religion and Society, ed. R.L. supplements are possible. Gordon (Cambridge I 98I ) I 63f. treatsthe matterwith 35 Ap. Kahil,AK viii (1965) 25. admirablecommon sense. 'Les 6tapes', hewrites, 36 See Proclus,quoted in n. 10 above; Phanodemos speaking ofthe successive religious duties which theFGrH 325 F 14; Henrichsn. 14. women claim tohave performed, 'sont celles d'un 37 SeePausaniasAtticista 35 Erbse;Sudas.v. pseudo-cycle'. "'Elf3apo?'; App. Prov. 2.54(Paroem. Gr.i 402); 93 94HUGHLLOYD-JONES ensued, and an oracle pronounced that it would not cease until a girl was sacrificed.A certain Embaros promised to sacrificehis own daughter, on condition that the priesthood of the cult was made hereditary in his family; he then concealed his daughter, dressed up a goat to resemblea girl, and sacrificedit in her place. This aitionseems to have been designed to explain a ritual in which a goat was a surrogate for a human victim, presumably one of the 'bears'.That is what the legend of Ilphigeneia would lead us to expect, and we know that the cult of the Mounychia was closely related to that ofthe Brauronia. The vasesindicatethat the 'bears' danced, as we should expect of girls takingpart in a ritualin honour of Artemis; they alsoindicatethat they rana race.It is not surprising that they rana race,in view of parallels from Sparta and from Els; anldit is not surprising that a raceshouldbe connected witha sacrifice.38At Olympia themostancient part ofthe games wasthefootrace:after the sacrificeto Pelops, whose shrine layjust west of the great altarof Zeus, the priest with his torch gave the signal for the startof the footrace, and the winner with his torch set fire to the offering lying on the altar.39Like Iphigeneia,Pelops was sacrificed;we are remindedof Neoptolemos at Delphi, Archemorosat Nemea and Melikertesat the Isthmus; and at the Arcadianfestivalof the Lykaiagames were closely linked with a sacrificewhich had central importance in the cult ofZeus Lykaios, a cult closely linked with that Lykaon whosacrificedhis son. The girls shown running on the vases run naked or wearing a short chiton. Attempts have been made to identify this with the krokotos;but this is normally a long garment, more like the high-waisted, short-sleeved, ankle-length chitonworn by themaidens whosestatues were discovered in the sanctuary. Persons undergoing ritesde passage often start by wearing a long garment which at a certain stage ofthe ritual they throw off. At Phaistosthere wasera festival calledthe ekdysia, whose cult myth told of a gir whom at her mother's entreaty Leto turnedinto a young man who was named Leukippos. Cretan inscriptions refer to the participants in similar rites as outEKSVOIEVO.40WhenTheseusafter making his way toAthenscametothe temple of Apollo Delphinios, then the royal residence,he was wearing a long garment;being mocked for his girlish appearance he threw itoff, and then hurled the oxen which had drawn the builders' cart higherthan the roof.41The story of the 'twiceseven',the youths and maidens sent as tribute to the Minotaur,of whom Theseuswas one, has been convincingly explained as an aitionof rites de passage; so have the legends of Kaineusand of Achilles upon Skyros. In Corinth seven boys and seven girlsspent a year,wearing blackrobes,intheshrineofHeraAkraia:'theclimaxand conclusionoftheir service',Burkert writes,'was a sacrifice at the festival of Akraia, the sacrifice ofa black he-goat'.42 As Burkert says, this goat was clearly a substitute for the children: the cult legend wasthe story ofthe killing ofthe killing ofthechildrenofMedea.Allthebestauthorities say that Niobehad seven sons and seven daughters: the story oftheir destruction by Artemisand Apollo surely relatestoritesofthesamekind.Burkerthas argued thattherites performed onthe Acropolisby the arrhephoroi wereinitiationrites,andthesetooinvolveda sacrifice.43 Whatweknowoftherites performed attheBrauroniafitsinwell enough withother initiation rites carriedout bygirls just before reaching the age of puberty. Before marriage, every Athenian girl made toArtemis the sacrifice knownas proteleia.44 It would hardly be Apostolius 7.10(ib. ii 397); Bekker, Anecdotai 444. Alland Pyanopsia in connection with Theseus, see Calame these are in Brelich248-9;cf Deubner 206.225 f. Jacoby, FGrH III bi (1954) and III b2,193f., 38 See Burkert HN i io f., citing Plut. Qu. conv.675c;gives the evidence. Cf P. Vidal-Naquet, AnnalesESC for a possible parallel instance ofa race and a dance asxxiii (1968)947f.,=PCPScxciv (1968)57 n. 4,=Gor- parts ofthe same rite, see Calame 339.don (n. 33)i56 n. 24. 39 Philostr. Gymn. 5 (cited byBurkert, HN 112). 42GRBS vii (1966)117f.; cf. Calame 220f. 40 See Nilsson GF 370f;L. Gernet, L'Anthropologie 43 Hermesxciv (1966) if;he reminds me of the goat delaGrece Antique (Paris 1968)203(p.164ofthesacrifice probably connected withthe arrhephoroi(HN inadequate trans. The Anthropologyof ClassicalGreece 172). [Baltimore 1981]);Burkert,Structureand History in44SeeBurkert HN75n.20:itis for proteleia that Greek Mythology and Ritual (Berkeley 1979)29f. Iphigeneia is brought to Aulis at Eur. IA 433. On such 41 F.Graf,MusHelvxxxvi (1979)14,andtherites in general see BurkertGR 390-5, and theexcellent literature he cites p.15 n.iI8.Onthe Osc(h)ophoria brief account byVidal-Naquet (n. 33). HUGH LLOYD-JONES 94 ARTEMISANDIPHIGENEIA surprising if at Brauron girls about to reach the age of puberty propitiated the same goddess. (b)Iphigeneia Young women who sufferedfrom various maladieswere accustomedto dedicateclothes to Artemis, as weknowfrom the Hippocratic treatiseonthe maladiesof virgins.45 Wehave a number of inscriptionsrecording gifts ofclothes toArtemis Brauronia. Butthe Euripidean Athena says that the clothes of women who die in childbirthwill be dedicatedto Iphigeneia.46 At Megara, where the events after the assembly of Agamemnon's fleet usually set at Aulis were located by Megarian tradition, Iphigeneia had her ownheroon. More often she was linked in cult withArtemis. At Aigira inAchaea a very ancient statue ofher stood inthe temple of Artemis; and at Hermione in the Argolid there was a cult ofArtemis Iphigeneia.47 In the Iliad, Agamemnon has a daughter called Iphianassa, but not one called Iphigeneia; Lucretius'use of the name Iphianassamay well indicatethat some Greek poet gave that name to the daughter whowas sacrificed.In a list ofdivinities in a LinearB tablet, wefind the name I-pi-me-de-ja: the absenceof an initial digamma is surprising, but the fact is still significant.48 In the Odyssey,Iphimedeia is the name of the mother of the giants Otos and Ephialtes. The first appearance of Iphigeneia as daughter of Agamemnon isinthe Cypria. Inthe Catalogue Iphigeneia isnot killed, butisturned into Hekate, handmaid of Artemis; inthe Cypria, Iphigeneia is carriedoff to Tauris and there made immortal by the goddess. In Stesichorus, Iphigeneia is the daughter not of Klytemnestraby Agamemnon, but of her sister Helen by Theseus, born after Helen's kidnapping at an early age and handed over to her aunt to be brought up by her.49This looks like a compromise between the story that Iphigeneia was Agamemnon's daughterby his wife and a different story. Helen was originally a goddess, connectedwith vegetation and fertility. She was not always Leda's daughter; one story gives her as a mother Nemesis, worshipped as a goddess at Rhamnous, not very far from Brauron.50The story of a daughter of Nemesis being carriedoff by Theseus may well have startedas an Attic legend: the relations ofTheseus withAttic local divinities and local heroes were not always happy. Similarly, it seems that in the original version of Helen's kidnapping currentat Sparta, the aggressor was not Theseus, but a local character; Enarsphoros, one of the sons of her father's enemy Hippokoon, is known to have had designs on her.51 It has long been conjectured that Iphigeneia startedas a goddess, and was later subordinated to or identified with Artemis;52 and it seems distinctly possible that she had Helen as a mother before being transferredto Helen's sister. In historical times, Artemis was a goddess ofbirth.53 As such, she seems often tohave 45De virg. I7 f. (ed.Littre, viii 466-8). 46 SeeIGii2 1514-25 and I528-31 for fragments fromthe Acropolis and Hesp. xxxii (1963)I69-82 nos 7-I0 for fragments fromthe Agora;cf. T. Linders, Studies intheTreasureRecords of Artemis Brauronia, SkrifterSvenskaInst. iAtheniv.19 (Stockholm1972). RelevantextractsfromIGii2 I514 andIGii2 1388 and I400 are translated by M.R. Lefkowitzand M. Fant, Women's Life in Greece and Rome (London1982) 120nos 123,124. Dedicationsto Iphigeneia: Eur.IT I464. 47 Paus.ii 35.1;vii26.5; i 43.2. 48 II.ix 287, inalistof daughters whosenames describedifferent aspects of kingship; Lucr. i 85; Hes.fr. 23M.-W.; forthe tablet, seeM.Ventrisand J. Chadwick, Documentsin Mycenaean Greek (Cambridge 1956) no. 172 (Kn.O2). Astothe digamma, see H. Miihlestein, ColloquiumMycenaeum(Neuchatel1979) 235; MrE.L.Bowie suggests that thederivation of Iphigeneia's namefromtbti may beamere popular etymology, and that the first element may derive rather fromtherootof i'Trraual, a suggestion thatIfind attractive. 49 Fr. 191 in Page, PMG. 50 SeeM.L. West, 'ImmortalHelen', Inaugural Lecture, Bedford College, London (1975); Calame 333 f.; L. B. Ghali-Kahil, Les enlevementset le retourd'Hle'ne dans les textes etles documents figures(Paris1955). 51 SeePlut.Thes. 31; cf. D.L. Page, Alcman: the Partheneion (Oxford1951), who speaks of'afurther possibility that Alcman told a story ofthe Tyndarids' punishment of Enarsphoros (and his brothers) for their insolencetoHelen' (32 n. 2). 52See L.R. Farnell, Greek Hero Cults and Ideas of Immortality(OxfordI92I) for a useful, though to my mind over-cautious, discussion ofthe question. 53 See Farnell, The Cults of the GreekStatesii (Oxford I896) 444 and the nn. on 567 f. On sacrificesto Artemis in connection withbirth and pregnancy, see the sacral lawfrom Cyrene inSEGix1.72=F. Sokolowski, Les loissacrees descites grecques,Suppl.(1962) 15, where apKosmay wellbe equivalent to apKroS. 95 acquired much of the benignant characterof the ancient goddesses of birth, the Eileithyiai,54 but this quality is secondary,being foreign to the original natureof the goddess. In Homer Artemisis a lion to women;55 with her gentle arrows she sends them sudden death. As the heiressof the Mistressof Animals, she will initially have been not a kindly supporter, but a potential killerwho had to be propitiated. J. D. Kondis56in his useful articleon the cult of Brauronis unlikely to be right in making out Artemis as altogetherkindly and in taking Iphigeneia to be responsible for the sinisterelements:one may doubt whether their functionsare so easily to be separated.By the time Artemis is appealed to as a kindly helper by women in travail, as she is according to the chorusof Euripides'Hippolytus,57 she has travelledfar from her original function in connection with birth. (vii) Orestesand the cult of Halai Araphenides Let usnowturn tothecultofArtemis Tauropolos, located sometwomiles north of Brauron, but distinct from the Brauronian cult, and founded by a male hero, brother to the founder ofthe neighbouring cult ofBrauron. Like other festivals of Artemis, the Tauropolia involved dances for which girls wore beautiful costumes; but it was founded by a male, and males might be among the spectators.58 The cult claimed to possess the original image of Artemis which had been brought back by Orestes fromthe Crimea,59 theTauric Chersonese; sodidseveral other cults of Artemis, including that of Artemis Orthia at Sparta.Beyond doubt the story that the image came from the Tauric Chersoneseowed its invention to the name Tauropolos: the Greekswill have been glad to credit barbarianswith cruel sacrificesof a kind they were unwilling to attributeto their ownancestors. According to one story, the cult statue of Artemis was found in a clump of withies by the heroes Astrabakosand Alopekos, who thereupon went mad.60 Another hero whowent mad after the discovery of a cult statue is Eurypylos of Patrai; the statue, it is true, was not that of Artemis but that of Dionysos Aisymnetes, but his cult is closely linked withthat ofArtemis Triklaria,and the heroon of Eurypyloslay within the temenosof Artemis.61Artemis was expert in curing madness. According to one version of the story, Melampous cured the daughters of Proitos of the madnesssent by Hera by praying to ArtemisHemeraof Lousoiin Arcadia.62Gods who cause an afflictioncan also, like the spear of Achilles, cure it; and Orestes'connection with madnessis well known. In the version of his story best known to us his protector is Apollo, but Pherekydes63 tells howhe sought refuge from the Erinyes at Oresthasion, or Oresteion, in Arcadia, in a temple of Artemis, and was protected by the goddess. When the sailorswho form the chorus of Sophocles'Ajax64 are speculating about the causesof their master's madness, the 54 E.g., seeEur. Hipp. 145. 55I.xxi 483. 56Citedinn. 30. Calame 292 f.thinksArtemis protected the newborn child, Eileithyia the mother; but if so, and I see no reason for believing it, this will have been a comparatively late development. 57 I66f. 58 SeeFarnell (n.47) ii 451f.; Deubner208 f.; Nilsson GF 251f.; Jacoby on Phanodemos FGrH 325 F 14. 59 See Paus. iii I6.8; cf. F. Graf, Ant.Weltiv (I979)33 f.; for other cults making the same claims, see Brelich 244. 60Paus.iii 16.9; see Burkert, MusHelvxxii (1965) 172. 61 See Nilsson GF 294f.; also 217. 62 See Akousilaos FGrH 2F 28; Bacchyl. I I passim; Burkert HN 189 f.; and cf. Hippocr., De virg., cited in n. 45 above. At Eur. Hipp. 145 Diktynna is mentioned in this connection; for Euripides shewasidentical with Artemis, asBarrett's parallels indicate. Kybele is coupled with Diktynna here; both together with Hekate are mentioned by Hippocr., De morbosacro 4 in a similarconnection. Kybele isoftensaidtocause madness, sometimesin conjunction withthe Kory- bantes, as in Menander, Theophoroumene 27 and in the fragment containing hexameterswhich E. W. Handley, BICS xvi (I969) 96 with great probability assigned to that play (see Sandbach'sOxford text of Menander, pp. I45-6). It may wellberelevantthat Kybele, like Artemis, isan heiress ofthe Mistress of Animals; see Burkert GR 233-4 and E. R. Dodds, The Greeksandthe Irrational (Berkeley/L.A. 1951) 77 f.,96 f. 63 FGrH 3 F 135, with Jacoby ad loc. (I a 424). 64 172; for appeals to Artemis to send away diseases, cf. Philip, AP vi 240 Gow-Page, Garland ofPhilip 2648 f. and Hy.Orph. 36,I5(cf.O.Weinreich, Gebet und Wunder [Stuttgart 1929] 18 f.). 96 HUGH LLOYD-JONES ARTEMISANDIPHIGENEIA first divinity they imagine may have sent it is Artemis Tauropolos. Some have supposed that she is invoked becauseit is cattle that Ajax has destroyed, but the connection of Artemis Tauropolos with madnessis likelier to be relevant. The cult title Tauropolos occurs in several places in Asia Minor, besides Amphipolis, on the Strymon, where the Greeksseem to have identifieda local goddess with their own Artemis.65It is not unnaturalfor the heiressof the Mistressof Animalsto be connectedwith bulls, but the title seems tohave deeper implications. Fritz Graf66 has pointed outthat allcults ofArtemis Tauropolos seemtoinvolvethe suspension of everyday conditions by the importation of something strange or uncanny. He finds that they were linked with the incorporation of young persons into the adult world by means of ritesde passage, like the ordeal of the ephebes in the allied cult ofArtemis Orthia. Furtherevidence in support of Graf's opinion may be found in the evident connection of the Tauropolos withmale sexuality, not surprising ina goddess whose protection is sought for young males approachingpuberty. The bull is an obvious symbol of virility; thus Aeschylus and Aristophanes bothusewhatseemstobeanancientritualterm aTavpcoTos inthesenseof 'virginal'.67 The title Orthia or Orthosia may well reflectan ancientconnection of the goddess withthe sexuality ofmales.68 In the cults ofHera Akraia at Corinth, ApolloDelphinios at Athens, and Apollo and Artemis together nearthe river Sythas in the neighbourhood of Sicyon, male and female initiation rites seem to have been combined together.69 There is ground for suspecting that, atleast intheearlier stages ofits history, thecult ofthe Tauropolos was concernedwith the initiation of males, and was closely related to that of the Brauronia, which was concerned with that of females. (viii) Bears and the cult of Brauron Thecult legend of Mounychia70 indicates that the rite involved the sacrifice ofa goat, serving as surrogate for a girl, meant to appease the goddess for the death of a bear. The rites of Brauron also involveda sacrifice; the girl for whomthe sacrificed beast was surrogate was called a bear, and was in a certain sense identicalwiththe goddess, just as the sacrificed Iphigeneia was in a sense identicalwithArtemis.The analogy ofthe cult of Dionysos is easy to perceive.71 The ritual death of the bear marked the end of the girls' lives as children and their entry intothe adult world;so, I suspect, didtheritual deathofthemalevictimintherites ofthe Tauropolos. The appearance ofthe bear, ofall creatures, inthis connection, is of great interest; forno animal figures more prominently in the earliest material used by Meulito support his theory of the origins ofsacrifice as an attemptby huntersto assuage the guiltthey feltforthe killing of their animal victims.72 The strength and cunning ofthebear have always madea deep impression on its hunters; so has its resemblance, when standingupright on its feet, to a human being; and bears figure in countless stories, legends and beliefs of the hunting culturesof which 65 See Farnell (n. 53) loc. cit. and his nn. on 569 f. 66 Op. cit. (n. 59) 4I. 67 Aesch. Ag. 244 and Ar. Lys. 216; Fraenkelon the former passage agrees withWilamowitzonthe latter thatthewordmustbehieratic. ra6tpo = KoXWvq), Polluxii 173, Galen xiv 706; r6al'otovTroi dvSp6s, Suda s.v.; ravpivsa is the name of a phallic game played at Taras, Hsch. s.v.; cf. Nilsson GF 184. One would like toknow why thefemaleflatterers whowheedled Macedonian princesses werecalled ravp7TroAot or TploSL'rTtg: the latter name is suggestive of Hekate, a personage connected with Artemis (see Klearchosfr. 19, ed. Wehrli, p.15). Onthe rape of girls dancing in honour of Artemis, see Calame 176, and cf. 189 f.For attempts to connect the etymology of ravpuoroAos with raOpos, seePhanodemosFCrH 325 F I4, IstrosFGrH 334 F I8, Apollodoros FGrH 244 FII. 68 Seen.26. 69 See Burkert, GRBSvii (1966) I 17 f; Graf(n.41) I f., esp. 13 f.; on the Sicyonian cult, see Nilsson GF I7If., Calame 205 f., and Brelich 377 f. 70 See n. 37 above. 71 SeeE.R. Dodds, Euripides, Bacchae2(Oxford 1960) xviiif. 72SeeMeuli (n.15), esp. 225f.=949 f.and 242 f.=969f;seealsoGSindexs.v.'Bar' (ii1238). Paul Faure, BCHlxxxiv (1960) makes conjectures about bearsin the cult of Artemis at Akroteri in Crete which if correct are highly relevant: see R.F. Willetts, Cretan CultsandFestivals (London1962)275 f.andThe Civilization of Crete (London1977) I22; alsoA. J. Hallowell, American Anthropologist xxviii (1962) 87 f. (on bear burials). 97 HUGH LLOYD-JONES we have knowledge. Rhys Carpenter73 was gently rebuked by Dodds for showing 'an excessive preoccupation with bears', but he was right to stressthe importance of the bear in early Greek belief. The cult legend ofthe Lykaia in Arcadia told how Lykaon, whose name suggests the wolf, sacrificedhis son or grandson.74Lykaon'sdaughter is Kallisto,75 who was an attendantof Artemis and was changedby her into a bear for having broken her vowof chastity; her son was Arkas, the eponym of the Arcadians, whose name was often connected with the word for bear. Although punishedby Artemis, Kallisto, like Iphigeneia, was in a sense identicalwith Artemis. We seem to have no evidence of initiation rites for females parallel to those for malesknown to have existed inconnection withthe Lykaia; yet I strongly suspect that such rites originally existed, and that bears played in them a part corresponding with that played by wolves in the initiationof male persons. Since there is no recordof bearsin Greece during historical times, the origins of theseriteswould seem to have been extremely ancient, and the same is likely to be true ofthe origins of the cult ofBrauron. (ix) Male andfemale initiationritesand sacrifice Initiation rites for males marked the entry ofthe young male intohis full status as a warrior. Oftenhehad to begin witha probationary period, living inthe wilds likean animal and mastering the skills that huntersneed. In Arcadiamale initiateslived for a time a life like that of wolves; Burkerthas pointed out that this explainslegends about were-wolves, as it explains the stories of leopard-men that comefrommodernAfrica.The Spartan institutionofthe krypteia supplies an obvious parallel; here toothe skills learned by the ephebes during their period of probation were not those of the warrior, but those of the hunter.76Does that indicatethat such practices went back to the time when hunting, just as much as if not even more than war, was the main occupation ofthe group ofadultmalesonwhoseactivitiesthefateofthe community depended? At the centre of all initiation rites there was a sacrifice.77It was designed to secure divine protection for the male in hunting or in war, for the female in marriedlife and childbirth.In theory, as the legend of Embarosindicates,human blood had to be shed to atone for the shedding of the blood of an animal dear to or even identified with the divinity. In a sense that victim too was identified with the divinity; in historicaltimes the place of the victim was taken by another animal. It is so naturalto suppose that, in the early history of mankind, sacrificeof this sort should have been offered tothe power that ruled over the wilds and forests that one is tempted to conjecture that in the culturesfrom which that of the Greeksdescendedthe Mistressof Animals may have been the first divinity to receive such sacrifices. During the immensely long period of prehistorythroughout which human life depended on the hunting of animals by the group of maturemales, each man would have to venture his life daily during long sojourns in the domain of the Mistressof Animals;each man would be trying to kill the animalswho were her property, so that he would have causeto fear her wrath. At the startof its expedition, the group of hunters would need to propitiate her by the shedding of blood. Later,when the group of male fighters made war not merely upon animalsbut upon other men, the institutionsthat had developed to meet the needs of the hunting group must have been easily adapted tothe needs ofa community that from time totime engaged in war. In the initiationritesof Sparta and Arcadiaas we know them, the stage of the young man'slife at which he lived in the wilds like an animal and learnedthe skillsof hunterswas followed by the stage at 73 Rhys Carpenter, Folk Tale, Fictionand Saga in the II5 f.;cxvii (I975) 265f.;G.MaggiulliinMythos: Homeric Epics (Berkeley/L.A. 1946) ch.6 passim;cf.Scripta in honorem M.Untersteiner (Genova1970)179 f.; DoddsinM.Platnauer, Fifty Years (and Twelve) ofHenrichs 201n.27. Classical Scholarship(OxfordI968)32 n.19. 76 See Vidal-Naquet (n. 4),esp. 55 f. = 55 f. 74 See Burkert HN98f. 77 See Burkert HN 50 and 142. 75 See W.Sale, RhM cv (I962) I22 f.; cviii (I965) 98 ARTEMISANDIPHIGENEIA which he served as horsemanor as hoplite.78 At the point of transitionit was naturalto placate a potentially hostile divinity; so also when the group of malesset off on a campaign,during which as during the hunting expeditions of early times, they would try to kill others and would risk death themselves.The great wars of legendary Greecedid not lack preliminary sacrifices;neither did the wars of history. 'Marriage istoa girl whatwaristoa boy',Jean-Pierre Vernant79 has remarked in connection with initiation rites. Female puberty is accompaniedby the start of menstruation. The ancientsdid not always clearly distinguish blood from amniotic fluid; in any case, copious bleeding accompanies and follows birth, so that the author ofthe Hippocratic treatiseon the diseasesof women can remarkthat when a woman is in good health her blood gushes out 'like that of a sacrificialvictim'.80 Birth may be fatalto the mother or the child, or both: Artemisslew women, as she slew hunters, with her gentle arrows. Artemis came to be thought of as a kind protector of women in travail; but originally she had been a dangerousenemy, to be propitiated at great cost. Several myths record her fury against girls wholeft her sphere by surrendering their virginity. So femalesalso were in danger of having their blood shed, perhapsfatally,by the MistressofAnimals. Artemis was dangerous also to male hunters. Often this aspect of the goddess is linked in myth with sexuality. Burkert has written that her virginity is not asexual, like that of Athena, but 'a peculiarly erotic, challenging ideal'.81 Formidablemales who offer violence to her or to her mother Leto are struck down by her unerring arrows: such are Tityos82 and thetwo Aloadai, Otos and Ephialtes. She shot the hunter Bouphagos, who tried to rapeher;she sent mad the hunter Broteas, whofailed to do her honour, just as Astrabakosand Alopekos, Eurypylos and Orestes, all went mad. One story makes her responsible for the death of the hunter Adonis, lover of Aphrodite; there can benodoubt that this story plays a part inthe Hippolytus of Euripides, where the two goddesses function as polar opposites.83 There arevariousaccountsof the end ofthe great hunter Orion, but onemakes Artemis kill himfor having offered her violence. As for Aktaion, the hunter whois son of the divine Aristaios by the virtually divine Autonoe, the story that Artemis destroyed him for having seen her bathing is not attestedbefore the Hellenistic period, and probably derives from the legend ofthe blinding ofTeiresias by Athena for a similaroffence.84One older version, used by Euripides in his Bacchae, makesher kill him for having boasted that he surpassed her as a hunter; but that, like the similar tales about Orionand Agamemnon, mustbe secondary. BothStesichorus andthe pseudo-Hesiodic Catalogue make her get rid of him for being a suitor of Semele, thus furnishing anotherinstance ofa hunter destroyed by Artemis because he has desired a female. Cherished companions of Artemis, likethefemaleAtalante and Kallisto andthemale Hippolytos, hadtoobserve chastity:85 those whofailed to do so sufferedat her hands. We are reminded that in historical times athletes while training for the games had to abstainfrom all sexual activity, sometimes resorting tothe practice ofinfibulation inorder toremove temptation.86 The sacrifice and banquet that marked the moment of releasefrom the taboo after the games was the signal for renewed sexual indulgence, and myths like that of the visit of the Argonauts to Lemnos have beenconnectedwiththis fact.87In the earliest timesinitiation ceremonies,marking the passage 78 See Vidal-Naquet(n.41); Brelichii6 f.,citing the Hunterandthe Hunters,U.Cal. Publ. in Classicsxxiii older literature. (1981). 83 1416-22. 79 Myth and Society in AncientGreece (Brighton I980) 84 See Dodds on Eur. Ba. 337-40.The conjecture of (trans. of Mythe et Societeen GreceAncienne [Paris1974]) L. Malten, Kyrene(Berlin 191 i) i8that the story that 23. 80 Demorb.mul. 9(ed. Littre, viiip. 3o). Aktaion was killed for having pursued Semele occurred 81 Burkert GR 235. inthe pseudo-Hesiodic Catalogue has nowbeen con- 82 On Tityos, the Aloadai, Bouphagos,Orion,seefirmed by the papyrus published by T.Renner, HSCP Schreiberin Roscher's Lexikon i 578 f.; on Broteas, seelxxxii (1978) 283 f.; cf. Stes.fr.236 Page, PMG. Apollodorus, Epit. ii 2 (ed. Frazerii 154 f); on Aktaion, 85 See Burkert HN72,withn.I2. see Burkert HN 127f.; on such myths in general, see G.86See p. 544 of my appendix to H. W. Smyth's Loeb Piccaluga inII mito greco, ed. B.Gentili and G. Paioniedn of Aeschylus, ii. (Rome 1973)33 f. andJ. Fontenrose,Orion:the Myth of 87 See Burkert HN212f. 99 HUGH LLOYD-JONES of young men or women tofull adult status, may have been followed by such collective weddings as Louis Gernethas described.88It may be relevanttoremarkthat during the pre-agriculturalage the returnof the huntinggroup aftera long stay in the wilds may well have been signalisedby a period of licensed indulgencefollowing the enforcedabstinence. We know that the ritesof Artemis Orthiainvolved an initiation ceremony for males, and we have seen reason to accept Fritz Graf's suggestion that the same is true of the rites of Artemis Tauropolos.89 Ihave already mentioned that thecults ofHera Akraia at Corinth, Apollo Delphinios at Athens, and Apollo and Artemis by the river Sythas near Sikyon offer instancesof linked pairs of ceremoniesfor males and females.To these the linked cults of the Tauropolos and the Brauronia should, I believe, be added.90 In manyplaces the god concernedwithmaleinitiationriteswas Apollo. His origins are partly oriental, but Burkert has drawn attention tothe resemblanceofhis name to the word apella, used in the Peloponnese to denote the assembly of adult males.91As god of the apella he controlledthe processesby which young men qualified for membership; and Burkerthas argued with great probability that a native Peloponnesiandeity ofthis kind was amalgamated withone of eastern origin. AtDelosthetwo goddesses, motherand daughter, were present long before Apollo;92 the daughter, like the great goddesses of Ephesos and Magnesia onthe Maeander, became identified with Artemis. Conjecture in mattersso obscureis hazardous; but one is tempted to guess that the similarity offunctionbetweentwodeities,onemaleand one female, each ofwhom presided over initiation rites, ledtoa merger by which they became brother and sister, the brother henceforth usually presiding overinitiationrites for males, the sister overthosefor females.Of course, manyexceptions andabnormalitieswillhavesurvived:thusHeraAkraiaand Apollo Delphinios relatedto both sexes, Artemis Orthia dealt with male initiationsand the twin cults of Brauronand Halai Araphenides, both belonging to Artemis, dealt with females and with males respectively. Fromthetimewhenshe became Apollo'ssister, Artemis tookon Apolline characteristics,calculatedto obscure her original characterand disposition.93 In the chain by which, following Meuli and Burkert,I have tried to link palaeolithic hunter withwarriors ofhistorical times, primitive appeasement of ghosts with historically attested appeasement of gods, some links are a good deal more conjectural than others. Archaeological evidence indicatesthat palaeolithic hunterstried to appeaseby offerings the spirits of the beasts they slaughtered; here Meuli has found the origins of sacrificeslater made to gods. Warriorsin historical as wellas legendary times sacrificedat the outset ofa campaign in order toassure success;94that they inheritedthe practice from the hunterswho were their ancestorsis a matter of surmise.In historicaltimes, sacrificesmade before battlesor in thanksgiving for victory were frequently to Artemis. Artemisis also connectedwith initiationritesfor malesand females,some of which seem to date from the primitive period of the hunting cultures,not surprisingly in view of her statusas the main heiressof the Mistressof Animals. For the theory to be correct,it is not 88 Gernet (n. 40)39 f. =23 f.;henotestheirconnected with a rite of the same kind; see Nilsson GF connection withArtemis. 240 and Calame I85. 89 Graf (n. 59)41. 91 See Burkert, RhM cxviii (I975) If.;in his article 90 On the myth ofNiobe,see p. 94 above. Accord-inGrazer Beitrage iv (975) 5 f.hehas added tothe ing toeach ofthe great tragedians, as wellas Lasos ofevidence for an oriental element in Apollo's origins by Hermioneand Aristophanes, she hadsevensons and showing that the early Greek statuesof him belong to a seven daughters; forthenumbersofherchildren type which had represented Resep, the Egyptian lord of according tothe various authorities, see W.S. Barrettarrows and the plague, since the second millennium BC. ap. RichardCarden,The PapyrusFragmentsof Sophocles 92SeeH.GalletdeSanterre, Delosprimitive et (Berlin etc. I974) 227f.Their death atthehands of archaique(Paris I958). Apollo and Artemis may wellbeconnected withan 93 See H6nn76. initiation rite in which a sacrifice symbolised the end of 94 On Spartanpractice see Xen.Lac. Pol. 13.8; Hell. childish life for the participants. Herodotus' story of theiv 2.20; Plut. Lyc. 22.2; BurkertHN78and GR 107. On rescue by theSamians of Corcyreanboys sent by the general connection of Artemis with war, see Farnell Periander toPersia forcastration lookslikean aition (n. 53) ii 470-I. I00 necessary for the Greeksto have practised human sacrificeeven as early as the Bronze Age; but it is possible that they did so then, and it is likely that their ancestorsdid so earlier.If so, the Greek myths about human sacrifice may preserve memories of a time when the forebearsof the Greeks did as the Carthaginians did as late as the Hellenistic period. (x) The sacrifice of Iphigeneia in Aeschylus: II Inthe light ofall this, letus return tothe problem ofthewrath ofArtemis against Agamemnon as Aeschylus presents it. It seems clear that in the original legend the sacrificeof Agamemnon'sdaughter was, like the sacrificesof Kreon'sson and the daughters of Heraklesand of Erechtheus, a necessarypreliminary tothe war. Persephone, towhomMakaria and the daughters of Erechtheuswere sacrificed, is one heiressof the Mistressof Animals, and we recall Herodotus' statement95that Aeschylus (and Aeschylus alone) somewhere made Artemis the daughter of Demeter. But the chief heiressof the Mistressof Animals was Artemis, and it was natural that Iphigeneia should be sacrificedtoher. As the Aeschylean Kalchas says, Artemis delights in the young of beasts; yet at the same time she is herselfthe greatest hunter, and grudges toothers the right of killing that which is her own.So warriors, like hunters, may not kill without appeasing her; so both warriorsand hunters, on entering into the membership of the groups ofadult males which they belong to, need to propitiate her. Long before Aeschylus, and long before the Cypria, the sacrificethat was necessary before the Greekscould sailfor Troy will have figured in the legend of the war. In early times the reason for it will have been tooobvious to need an explanation; even in historical times, it is amply attested that before every battle the Spartangenerals sacrificedto Artemis. But for Aeschylus and his audience a reason for the sacrificeneeded to be given. Had he wished, Aeschylus could have given the reason given by the author of the Cypria and later by Sophocles in his Electra.Instead,Kalchas explains that Artemis feels pity for the victims of the eagles. He speaks of the killing ofthe pregnant hare as a sacrifice,and fears that Artemis may avenge it by bringing about another sacrifice, hinting directly at the sacrificeof Agamemnon's daughter. Artemis loathes the feast of the eagles, and the eagles, andthhetsons of Atreus;she feels pity for the hare and her young, and the hare and her young stand for Troy. In the Iliad, Agamemnon tells Menelausthat when Troy falls not even the unborn childrenin the womb are tobe spared. In the poetic tradition Artemis, like her brother Apollo, is a supporter ofthe Trojans, and since the hare and her young stand for the Trojans,this fact can hardly be called irrelevant.Yet at the same time the notionthatthe goddess pities the prospective victims of the expedition accordswith the ancientbelief, still to be reckonedwith in the time of Aeschylus, that those whoare setting out toshed the blood ofanimals or people have to propitiate a female divinity concernedfor their prospective victims by sheddintheir prospectiveblood of by shedding omethinebloodofrsomeone that belongs to them. So it is significant that Aeschylus chose this motive ratherthan the story of Agamemnon's boast. Artemis, kindly as she is towards the young ofbeasts, yet demands (Ag. 140f.) that the portent of the eagles be made valid: it seems to be agreed that her demand must be addressedto Zeus. Then Kalchasinvokes Artemis' brotherin his aspect as the Healer (146 f.), so that Artemis may notmake it impossible for the Greeks to sail, 'hastening on another sacrifice ... a maker of quarrels born inthe house, withoutfear ofthe man; for there abides a keeper ofthe house terrible,ever again arising, a treacherous,unforgetting Wrath, child-avenging'.The sacrificeof Iphigeneia is called a maker of quarrels, to quote Fraenkel's paraphrase, 'born in the house and grown onewithit';thatisto say, thesacrifice ofhis daughter will bring down upon Agamemnon the enmity of his wife, who will not fear her husband.For there waits 'a keeper of the house' whois identified with 'unforgetting Wrath, child-avenging'; these words can only refer to the Wrath that threatensthe son of Atreus, murdererof his brother'schildren,to what 95 ii I56. ARTEMISANDIPHIGENEIA IOI 102HUGH LLOYD-JONES Klytemnestra later (I 50 f.) calls 'the ancient savageavenger of Atreus, the cruel banqueter'. The sacrificeof Iphigeneia, caused by Artemis, will arousethe anger of Klytemnestra; and the anger of Klytemnestra will be employed by Zeus to punish Agamemnon for the crime committed by his father Atreus. As Zeus has used Agamemnon to punish the crime of the Trojans, so he uses Klytemnestra to punish thecrimeof Agamemnon's father Atreus: thisbecomesclear to Kassandra (see 1287f.,1327 f.) andtothe Argive elders (1335 f.).96As they laterremark, everything that is accomplished for mortals is the work of Zeus (1485 f.). HUGH LLOYD-JONES ChristChurch, Oxford 96 See my newintroduction totheOresteia (n.3) xviif.