¹ The following works are referred to by author's name: Rohde, E., Psyche: The cult of souls and belief in immortality among the Greeks, translated by Hillis, W.B. from the eighth edition (New York 1925)Google Scholar; Kramer, T., s.v. Prophetes in The theological dictionary of the New Testament, ed. Friedrich, G., trans. Bromley, G.W. (Grand Rapids 1968) viGoogle Scholar; Dodds, E.R., The Greeks and the irrational (Berkeley 1951)Google Scholar; Amandry, P., La mantique apollinienne à Delphes: Essai sur le fonctionnement de l'oracle (Paris 1950)Google Scholar; Parke, H.W. and Wormell, D.E.W., The Delphic oracle (Oxford 1956).Google ScholarWhittaker, C.R., ‘The Delphic oracle: belief and behaviour in ancient Greece—and Africa’, HTR lvii (1965) 21–48CrossRefGoogle Scholar, Roux, G., Delphes: son oracle et ses dieux (Paris 1976).Google ScholarFontenrose, J., The Delphic oracle (Berkeley 1978).Google ScholarPadel, R., ‘Women: model for possession by Greek daemons’ in Images of women in antiquity, ed. Cameron, A. and Kuhrt, A. (Detroit 1983) 3–19.Google ScholarPrice, S., ‘Delphi and divination’, in Greek religion and society, ed. Easterling, P. and Muir, J.V. (Cambridge 1985) 128–54.Google ScholarParker, R., ‘Greek oracles and Greek states’, in Crux: Essays presented to G.E.M. de Ste. Croix on his 75th birthday, ed. Cartledge, P.A. and Harvey, F.D. (Sidmouth 1985).Google ScholarSissa, G., Greek Virginity, trans. Goldhammer, A., (Cambridge MA 1990).Google Scholar

² Rohde's thesis that prophecy and possession originally belonged to Dionysus and not Apollo has been sensibly refuted by Kurt Latte, who has argued that in the East, Apollo is also associated with prophecy and, very often, he has female priestesses. Latte, K., ‘The coming of the Pythia’, HTR xxxiii (1940) 9–19.CrossRefGoogle Scholar Latte cites two inscriptions from Tralles, in Ramsay, W.M., Cities and bishoprics of Phrygia (Oxford 1895) i 95Google Scholar, no. 115 and Robert, L., Études anatoliennes (Paris 1937) 407Google Scholar as well as Herodotus' comments on the priestess of Apollo in Patara (i 182). See also Burkert, W., ‘Itinerant diviners and magicians: a neglected element in cultural contacts’, in The Greek renaissance of the eighth century BC: Tradition and innovation, ed. Hägg, Robin, (Stockholm 1983) 117 n. 24Google Scholar; Dietrich, B.C., ‘Reflections on the origins of the oracular Apollo’, BICS xxv (1978) 1–18.Google Scholar In a different vein, Dodds has rejected Rohde's thesis, arguing that it relies more on a Nietzschean dichotomy between ‘rational’ Apollo and ‘irrational’ Dionysus than on the evidence at hand. Dodds, E.R., The Greeks and the Irrational (Berkeley 1951) 69.Google Scholar

³ See, for example, Lucan's Pharsalia (v 123–224), Aeschylus' Agamemnon (1072–1330), and Vergil's Cumaean Sibyl (Aen. vi 33–15. Plut. Mor. 438).

⁴ Parke and Wormell i 37.

⁵ Price (n. 1) 137. Price also notes that the changes in the Pythia's voice during this episode are ‘bizarre and extreme’ and therefore can not illustrate her behaviour under normal divinatory conditions.

⁶ Ibid. i 39.

⁷ Parke and Wormell i chapters 7 and 11. For a convincing argument that Delphi did not have such political agendas, see Parker (n. 1).

⁸ P. Amandry is acutely aware of this contradiction and, like Parke and Wormell, posits that the rational prophets and not the possessed Pythia controlled the content and form of the oracle given to enquirers (n. 1). With slight variation, these roles are a commonplace in most literature on Delphi. Burkert, W., e.g., casually remarks, ‘The utterances of the Pythia are then fixed by the priests in normal Greek literary form, the Homeric hexameter.’ Greek religion, trans. Raffan, J. (Cambridge MA 1985) 116.Google Scholar

⁹ Nagy, G., ‘Ancient Greek poetry, prophecy, and the concepts of theory’, in Poetry and prophecy: the beginnings of a literary tradition, ed. Kugel, J.L. (Ithaca 1990) 56–64.Google Scholar

¹⁰ The most thorough treatment of the word prophetes is by Kramer (n. 1 ).

¹¹ Mantis: Aesch. Eum. 29 with 33.

¹² Prophetis: Eur. Ion. 42, 321, 1322; Plat. Phdr. 244b; IG xii (3) 863; Strabo, ix 3.5; D.S. xiv 13.3; xvi 26.4; Plut. Mor. 414b; Pollux x 81; Iamb. Myst. cxxvi 4.

¹³ Promantis: Hdt. vi 66, vii 111 and 141; Thuc. v 16.2; Plut. Alex. 14; Lucian Herm. 60; Heliod. iv 16. See also G. Radke, RE (1957) vol. 23 s.v. promantis.

¹⁴ In the Oresteia, he is the prophetes of Zeus as well as a mantis who does not lie (Eum. 19 and Ch. 559; Eu. 18). While the Pythia, like Apollo, may be called both a mantis and a prophetis, thus suggesting the synonymity of these words, in one oft-quoted passage, Pindar indicates that there may have been a slight distinction between these words. Pindar writes, ‘Be a mantis, Muse, and I will be a prophet.’ Pi. Fr. 150 (Snell). While Dodds assimilates the Pythia to the role of the Muse, when explicating this passage, Helmuth Flashar, on the other hand, likens the Muse to Apollo and the Pythia to Pindar. Neither analogy is convincing, however, since both the Pythia and Apollo are called by both titles. If a distinction is to be made between mantis and prophetes, perhaps, it is one of emphasis, rather than function. The word prophetes emphasizes the announcement of the divine message, while mantis emphasizes contact with the divine. Dodds (n. 2) 82, and Flashar, H., Der Dialog Ion als Zeugnis platonischer Philosophie (Berlin 1958) 64 n. 2.Google Scholar

¹⁵ Fontenrose and Roux (n. 1).

¹⁶ Roux 160. Here, Roux seems to follow Crahay who thought all the verse oracles in Herodotus were not authentic. Crahay, R., La littérature oraculaire chez Hérodote (Paris 1956)Google Scholar.

¹⁷ Fontenrose 42–47.

¹⁸ These discussions represent one feminist approach to women and religion, one which is ‘about the extent of male dominance in religious tradition or about the explanation for it’. Bynum, C.W., ‘Introduction: The complexity of symbols’, in Gender and religion: on the complexity of symbols, ed. Bynum, C.W., Harrell, S. and Richman, P. (Boston 1986) 4.Google Scholar

¹⁹ Sissa (n. 1).

²⁰ Sissa 51.

²¹ Sissa 53–70.

²² Sissa insists that representation of the Pythia can tell us nothing about the Pythia's actual service at Delphi (19).

²³ Padel (n. 1) 6.

²⁴ Padel 6.

²⁵ Most of Padel's evidence about male dominance of women pertains to fifth century Athens. The controls that men may exercise or may want to exercise over their (fertile) wives in Athens may not be an useful analogy for understanding the lives of virgin priestesses in temples, especially at Delphi whose political, social and economic structure differed greatly from that of Athens.

²⁶ Male prescriptions and descriptions about how women should and do act are not necessarily accurate indications of how women led their lives, on which see Cohen, D., ‘Seclusion, separation, and the status of women in classical Athens’, GRBS xxxvi (1989) 3–15.Google Scholar

²⁷ Price (n. 1) 141.

²⁸ Two rather distinct areas of interest, Europe and Africa, have recently received attention. The following is a representative sample of recent studies. Europe: Ginzburg, C., The night battles, trans. , J. and Tedeschi, A. (London 1983)Google Scholar; Walker, D.P., Unclean spirits (Philadelphia 1981)CrossRefGoogle Scholar; Kagan, R.L., Lucrecia's dream: politics and prophecy in sixteenth-century Spain (Berkeley 1990)Google Scholar; Niccoli, O., Prophecy and people in renaissance Italy, trans. Cochrane, L.G. (Princeton 1990).Google Scholar Africa: Peek, P.M., ed., African divination systems: ways of knowing (Bloomington 1991).Google Scholar The introductory essay in this collection contains an overview and a lengthy bibliography of recent studies.

²⁹ Whittaker (n. 1).

³⁰ Whittaker 30.

³¹ Parker (n. 1).

³² Bourguignon, E., ‘Introduction: a framework for the comparative study of altered states of consciousness’ in Religion, altered states of consciousness, and social change, ed. Bourguignon, E. (Columbus 1973).Google Scholar In Bourguignon's survey of 488 societies, 437 (90%) societies have institutionalized incidences of altered states of consciousness, 251 (52%) of which are associated with spirit possession. On the usefulness of the comparativist approach in the study of Old Testament prophecy, see A. Cooper, ‘Imagining prophecy’ in Kugel (n. 9) 31–3.

³³ Overholt, T.W., Channels of prophecy: the social dynamics of prophetic activity (Minneapolis 1989) 21 and 1–15.Google Scholar

³⁴ Bourguignon, E., ‘The self, the behavioral environment, and the theory of spirit possession’ in Context and meaning in cultural anthropology, ed. Spiro, M.E. (New York 1965) 39–60, 41.Google Scholar

³⁵ Wilson, R.R., ‘Prophecy and ecstasy: a re-examination’, JBL lxxxxviii (1979) 321–37.Google Scholar

³⁶ Bourguignon, 1965, 41.

³⁷ English, H.B. and English, A.C., A comprehensive dictionary of psychological and psychoanalytic terms (New York 1958).Google Scholar

³⁸ Lewis, I.M., Ecstatic religion: an anthropological study of spirit possession and shamanism (Baltimore 1971) 38.CrossRefGoogle Scholar

³⁹ Plutarch 438. See p. 2 above.

⁴⁰ Crapanzano, V. and Garrison, V., Case studies in spirit possession (New York 1977) 8.Google Scholar Emphasis on ‘indigenous’ is mine.

⁴¹ An altered state of consciousness is a mental state, ‘induced by various physiological, psychological, or pharmacological manuevers or agents, which can be recognized subjectively by the individual himself (or by the objective observer of the individual) as representing a sufficient deviation, in terms of subjective experience or psychological functioning, from certain general norms as determined by the subjective experience and psychological functioning ofthat individual during alert, waking consciousness.’ Crapanzano and Garrison 8.

⁴² Crapanzano and Garrison 9.

⁴³ Buss, M.J., ‘The social psychology of prophecy’ in Prophecy: essays presented to Georg Fohrer on his sixty-fifth birthday, ed. Emerton, J.A. (Berlin 1980).Google Scholar

⁴⁴ In many cultures, individuals must exhibit very specific behaviours in order to be diagnosed as ‘possessed’. If possessed persons function, as they often do, as healers and diviners, a long and arduous period as apprentice under an established practitioner of such arts must he fulfilled. The behaviours of the possessed apprentice are very often scrutinized and evaluated by the community in order to assess his/her suitability for such a prestigious career. See, for example, A. Morton, ‘Dawit: competition and integration in an Ethiopian Wuqabi cult group,’ in Crapanzano and Garrison (n. 63); R. Bier, ‘Diviners as alienists and annunciators among the Batmmaliba of Togo’ in Peek 1991 (n. 28); Beattie, J. and Middleton, J. (eds.), Spirit mediumship and society in Africa (New York 1969).Google Scholar

⁴⁵ Boddy, J., Wombs and alien spirits (Madison WI 1989).Google Scholar See also McCarthy-Brown, K., Mama Lola: a vodou priestess in Brooklyn (Berkeley 1991)Google Scholar: Davis, W., Dojo: magic and exorcism in modern Japan (Stanford 1980).Google Scholar

⁴⁶ Such views are critically reviewed by Spring, A., ‘Epidemiology of spirit possession among the Luvale of Zambia’, in Women in ritual and symbolic roles, Hoch-Smith, J. and Spring, A., eds., (New York 1978) 167.Google Scholar

⁴⁷ W., and Mischel, F., ‘Psychological aspects of spirit possession’, American Anthropologist lx (1958) 249–60Google Scholar; Messing, S.D., ‘Group therapy and social status in the Zar Cult of Ethiopia’ in Culture and mental health, ed. Opier, M. K. (New York 1959)Google Scholar; Ward, C., ed., Altered states of consciousness and mental health: a cross-cultural perspective (Newbury Park, CA 1989)Google Scholar; Bourguignon, E., Possession (San Francisco 1976).Google Scholar

⁴⁸ See Spring (n. 46) and Brown (n. 45).

⁴⁹ One of the most stunning examples of the considerable power accorded to such women is that of Hildegard of Bingen, the twelfth-century Rhineland mystic. Not only did Hildegard leave a large and impressive collection of manuscripts on medicine, sexuality, saints' lives, religious doctrines, several illuminations, symphonies, songs, and plays, she also left behind her letters, which give evidence of her great political prestige and power. She corresponded with the king of Germany, Frederick Barbarossa, Pope Alexander, various archbishops and other clergy and dignitaries. The considerable influence Hildegard had throughout her life was directly dependent on her visions and provides a detailed and compelling example of the surprising power women can wield within patriarchal structures by means of their spiritual lives. See Flanagan, S., Hildegard of Bingen: a visionary life (London 1989)Google Scholar; Newman, B., ‘Hildegard of Bingen: visions and validation’, Church History liv (1985) 163–75.CrossRefGoogle Scholar On medieval mystics, more generally, see Petroff, E.A., Medieval women's visionary literature (Oxford 1986)Google Scholar; Bynum, C.W., Jesus as mother (Berkeley 1982).Google Scholar

⁵⁰ Pollux Onom. i 15.

⁵¹ Dodds (n. 2) 94 n. 84; Burkeit (n. 8) 110. Interestingly, the Greek definition of the word ecstasy is remarkably similar to Crapanzano's and Garrison's definition of altered state.

⁵² Dodds (n. 2) 87 n. 41; Burkert (n. 8) 109; Plut. Mor. 438b. Smith, W.D., ‘The so-called possession in pre-Christian Greece’, TAPA lxxxxvi (1965) 403–36Google Scholar is cautious about claiming that the spirit was actually believed to enter into and possess a person.

⁵³ The concept of receives thorough treatment by Kleinklecht, s.v. in The theological dictionary of the New Testament (n. 1).

⁵⁴ Burkert (n. 8) 110; Padel (n. 1)13 n.20.

⁵⁵ On the ancient evidence about the imagined mode of Apollo's possession of the Pythia, see Sissa (n. 1).

⁵⁶ A definitive treatment of possession in Greek culture is beyond the scope of my project. I have relied upon Dodds' discussion, which remains, in my opinion, thorough and convincing. Dodds (n. 2) 64–102. See also Burkert (n. 8) 109–18.

⁵⁷ Pl. Apol. 22c. Translations are mine unless otherwise indicated.

⁵⁸ Pl. Ion 533e.

⁵⁹ Pl. Ion 534b-d. In a similar vein, Socrates assimilates statesmen to seers and poets in order to explain why they do not fully understand the words they seek. Meno 99c-e.

⁶⁰ Hesiod, Th. 31–32.

⁶¹ For a full list and discussion of passages on poetic inspiration such as these, see Tigerstedt, E.N., ‘Furor poeticus: poetic inspiration in Greek literature before Democritus and Plato’, JHI xxxi (1970) 163–78CrossRefGoogle Scholar, and Murray, P., ‘Poetic inspiration in early Greece’, JHS ci (1981) 87–100.CrossRefGoogle Scholar See also Homer's depiction of Demodocus, the bard, in Od. xxii 347–48. In the passage under discussion, Murray suggests that the divine song of the Muses may indicate the fluidity of speech necessary for oral ‘composition in performance’, for those whom the Muses love speak effortlessly (Th. 96–7 with Murray 95). Both Tigerstedt and Murray successfully demonstrate that earlier poets do not present themselves as witless or passive as Plato describes them.

⁶² Democritus also expresses similar ideas about possession, on which see Havelock, E.A., Preface to Plato (Cambridge MA 1963) 162Google Scholar, and Tigerstedt, 1970.

⁶³ Those afflicted with illness engage in telestic purifications such as those offered by the Corybantes and Bacchantes, so in effect we have, more or less, the same three groups as elsewhere. See Morgan, M.L., Platonic piety: philosophy and ritual in fourth-century Athens (New Haven and London 1990) 165–66.CrossRefGoogle Scholar

⁶⁴ In the Phaedrus, words for possession occur rarely. There is (245a). Most often the word mania is used, sometimes being qualified by ‘from the gods’ (apo theori). This, I think, indicates both the fluidity of the ideas surrounding possession, or contact with the divine, as well as the degree to which Plato's own purposes dictate his presentation of cultural phenomena.

⁶⁵ Pl. Phaedr. 244a. ‘Madness’ is an unfortunate translation for mania because the English word conjures up images of nineteenth century asylums and the like. As Amandry has pointed out, the word, though often associated with Bacchic revelries, does not necessarily imply frenzied behaviour. Amandry 43–8.

⁶⁶ Amandry 41–50; Dodds 64–102; 158–87. See Morgan (n. 63) 164 n. 23 for a bibliography of recent work on the Phaedrus.

⁶⁷ Pl. Laws 719c. ‘Whenever a poet sits on the Muses’ tripod, he is not in his senses, he is like a spring which readily allows its water to flow’ on which, see Murray (n. 61) 95–6. See also Connor, W.R., ‘Seized by the nymphs: nympholepsy and symbolic expression in classical Greece’ CA vii (1988) 155–89, esp. 156–62.Google Scholar

⁶⁸ Plato, Phaedr. 244 c-e. This distinction is discussed at length by Bouché-Leclerq, A., Histoire de la divination dans l'antiquité (1879–1889; reprint New York 1975).Google ScholarCaquot, A. and Leibovici, M., La divination: études recueillies (Paris 1969)Google Scholar rely on this distinction in their survey.

⁶⁹ Zeusse, E.M., ‘Divination’, in The encyclopedia of religion, ed. Eliade, M. (New York 1987) 375–82Google Scholar; Crawford, J.R., Witchcraft and sorcery in Rhodesia (Oxford 1967) 179–81.Google Scholar

⁷⁰ Amandry has argued that there were two distinct practices at Delphi, the lot oracle and the more prestigious prophetic oracle, (25–36). Fontenrose dismisses the notion that there was a lot oracle at all (219–223). The evidence for the use of cleromancy is scant—it is not directly referred to in any sources, but is merely suggested (i) by the verb ‘take up’ (anairein), used to introduce less than ten percent of all recorded oracles, namely 12, 17, 26, 43, 51, 57, 60, 80, 92, 102, 104, 107, 113, 123, 134, 136, 137, 155, 165, 172, 174, 178, 212, 218, 219, 255, 262, 281, 287, 316, 317, 320, 328, 358, 362, 375, 386, 398, 405, 412, 462, 482, 484, 509, 559, 565, 570, 575, 581, 605 (numbers refer to Parke and Wormell's edition of oracle; since the oracles introduced by anairein are evenly distributed throughout the corpus, are in both prose and verse, and pertain to a variety of matters, it is near impossible to draw any conclusions about the use of this verb; 2) by the phiale which the Pythia and/or Apollo holds in several vase paintings of Delphi (Printed in Amandry 66–77); 3) an inscription which may be restored in such a way so as to mention two beans and hence imply cleromancy was practised (on which see Fontenrose 222–23). Cleromancy may have been practised at Delphi; however, given the scanty evidence, I do not think it is possible to say much about it.

⁷¹ Ahern, E.M., Chinese ritual and politics (Cambridge 1981).CrossRefGoogle Scholar

⁷² Zeusse (n. 69) 376.

⁷³ Thus Zeusse describes ‘wisdom divination’, the category in his system of divination which corresponds to Ahern's ‘non-interpersonal category’ (Zeusse 376; Ahern 51–63).

⁷⁴ Ahern 53.

⁷⁵ Park, G.K., ‘Divination and its social contexts’, JRAI lxxxxiii (1963) 195–209.Google Scholar

⁷⁶ Ahern distinguishes three features which characterize the communication between spirits and humans: it follows constitutive rules instead of regulatory ones, its code can be highly restricted, and it often involves elaborate interpretation (53 ff.). Her useful distinctions apply to both human and non human randomizing devices and I have indirectly borrowed from them in order to derive my own understanding of the language of possessed diviners.

⁷⁷ S.R. Whyte, ‘Knowledge and power in Nyole divination’ in Peek (n. 28) 162–64.

⁷⁸ Jordan, D., Gods, ghosts and ancestors (Berkeley 1972) 75–6Google Scholar, quoted and discussed with a different perspective in Ahern 54.

⁷⁹ One is reminded of Cassandra in the Agamemnon.

⁸⁰ D. Parkin, ‘Simultaneity and sequencing in the oracular speech of Kenyan diviners’ in Peek (n. 28) 178. See also in the same volume J.W. Fernandez, ‘Afterword’.

⁸¹ Furthermore, these odd linguistic gestures mark the paradox of spirit possession—how can one represent the spiritual world in a familiar idiom such as language without diminishing its transcendence or otherness?

⁸² See for example, V. Garrison, ‘The Puerto Rican syndrome in Psychiatry and espiritismo’, in Crapanzano and Garrison (n. 40).

⁸³ On the preliminary rites, see Parke and Wormell i 30–3.

⁸⁴ Plut. Mor. 292d, 365a, 437a, and 438b. The title hosioi also appears in several Delphic inscriptions, for which see Fontenrose, 219, n.32.

⁸⁵ Fontenrose 218.

⁸⁶ Hdt. viii 36; Eur. Ion 369–372, 413–16, with Fontenrose, 216. The word prophetes also appears in Aelian, , NA x 26Google Scholar; Plut. Mor. 292d, 438b.

⁸⁷ Plut. Mor. 386. Parke, 1940, 87.

⁸⁸ A scholiast's note on Il. xvi 235 records, ‘they call those who sit around the oracle and carry out the prophecies made by the priests (hiereis) prophets (prophetai)’ only confuses matters further. But I do not think this late notice should be given too much weight. Fontenrose, 219, n. 31.

⁸⁹ Parke struggles to decide which of these officers may have versified the oracular response. No ancient source credits them with versifying the response. Both Strabo and Plutarch say that versifiers (not the prophetes or hiereus) put the Pythia's words into verse. Simultaneously, they both claim that the Pythia sometimes spoke in verse herself (Str. ix 3.5 and Plut. Mor. 405d and 407b). I find it difficult to draw any conclusions from these contradictory claims, both of which are quite late in Delphi's history. However, since there are no other notices of such versifiers in any of the literature on Delphi, nor in any inscriptions, and in addition, these same authors think there were versifiers at Delphi, contra McLeod, W., ‘Oral bards at Delphi?’, TAPA lxxxxii (1961) 317–25.Google Scholar See also Fontenrose 213–1

⁹⁰ In the cases where we have inscriptions which begin with the formula, ‘it is better for you…’ it appears that the consultants might come asking for the Pythia to simply sanction or forbid a course of action. Fontenrose 37–8 and 221.

⁹¹ Dewald, C., ‘Women and culture in Herodotus' Histories’, Women's Studies viii (1981) 93–127.CrossRefGoogle Scholar In Herodotus there are 62 notices of priestesses of which roughly half are the Pythia.

⁹² Hdt. ii 53.

⁹³ Hdt. ii 55.

⁹⁴ Hdt. v 63; vi 66; Thucydides also reports an instance where the Pythia is bribed (v 16.2).

⁹⁵ The most recent edition of the fragments from Euripides' Melanippe is that of Auffret, S., Mélanippe la philosophe (Paris 1987)Google Scholar (Fragment 13). Translation mine.

⁹⁶ Eur. Medea 410–430.

⁹⁷ At Dodona and Didyma, there is limited evidence which suggests that the prophetesses, not the prophets, issued oracular responses. At Didyma, Catherine Morgan has questioned the assumption that in archaic times the mantis was male, based as it is on Herodotus' account of Branchos, founder of the Branchidae, the family of priests in charge of the shrine. Morgan, C., ‘Divination and society at Delphi and Didyma’, Hermathena cxlvii (1989) 17–42 (27).Google Scholar More compellingly, in a recently discovered inscription from Didyma, a hydrophoros of Pythian Artemis refers to her great-grandmother as prophetis. The mention of a prophetis here seems to confirm the existence of prophetesses at Didyma, if not Iamblichus' account of their activity at the oracular shrine. The title prophetis also appears in two other inscriptions from Didyma. Gunther, W., Istanbuler Mitteilungen xxx (1980) 170–75Google Scholar, with Parke, H.W., Oracles of Apollo in Asia Minor (London 1985) 231 n. 12.Google Scholar At Dodona, the evidence for the existence and activity of prophetesses is equally obscure. Herodotus paints a vivid picture of his conversations with the Dodonaean priestesses (discussed above). Additionally there is frequent mention, in a variety of sources, of the priestesses at Dodona who are called doves, Peleiai or Peleiades, on which see Parke, H.W., The Oracles of Zeus (Oxford 1967).CrossRefGoogle Scholar In a sensible survey of this cult title and its implications for assessing the role of the priestesses at Dodona, Jebb argues that when Sophocles describes the prophetic oak as speaking he implies that the priestesses orally issue the oracular responses. R.C. Jebb, Sophocles: the plays and the fragments, V: Trachiniae. While the evidence from Dodona (and Didyma) is complex and requires a more thorough examination than possible here, it is significant that in both instances there are intimations that women were associated with these shrines and may have played an active part in the issuing of oracular responses. We may also evoke the comparative evidence from medieval Europe, which offers a compelling paradigm for women's lives in religious institutions.

⁹⁸ Cassandra's knowledge of the history of the house of Atreus is derived from a vision of Thyestes' feast. Since it is not possible to determine whether the Pythia had such visions, this aspect of her portrayal is a moot point. On such visions, see Dodds (n..2) 71.

⁹⁹ See n. 9 above.