Published online by Cambridge University Press: 15 March 2018
The question of the exact nature of the Pythia's expertise has been the subject of academic debate for a very long time. It would indeed not be an exaggeration to say that this has been, and continues to be, one of the most controversial questions in the study of ancient Greek religion. Modern scholars are sharply divided over whether any inspired female oracles, and especially the Pythia at Delphi, had the ability to prophesy in hexameter verse without male assistance. During the classical period the two most famous oracles were those of Zeus at Dodona in Epirus in north-western Greece and of Apollo at Delphi, which was located on the south-western spur of Mount Parnassus. According to Plato (Phaedrus 244), the Delphic priestess, as well as the priestesses at Dodona, prophesied in a state of altered consciousness (which he calls mania), and were practitioners of ‘inspired prophecy’ (mantikē entheos).
I would like to thank Harriet Flower, Esther Eidinow, Matthew King, and especially Kathleen Cruz for their help and suggestions. An earlier version of this article was given at the Friedrich-Alexander University of Erlangen-Nüremberg in 2015. All translations are my own, unless otherwise indicated.
1 For diametrically opposed views, see Bowden, H., Classical Athens and the Delphic Oracle. Divination and Democracy (Cambridge, 2005)Google Scholar, and Flower, M. A., The Seer in Ancient Greece (Berkeley, CA, 2008), 211–39CrossRefGoogle Scholar.
2 For Dodona, see Parke, H. W., The Oracles of Zeus (Oxford, 1967), 1–93 CrossRefGoogle Scholar; Eidinow, E., Oracles, Curses, and Risk Among the Ancient Greeks (Oxford, 2007), 56–71 Google Scholar; Dieterle, M., Dodona. Religionsgeschichtliche und historische Untersuchungen zur Entstehung und Entwicklung des Zeus-Heiligtums (Hildesheim and New York, 2007), 25–102 Google Scholar.
3 For Plato's attitude to divination, see Flower (n. 1), 29, 84–8; Flower, M. A., ‘Religious Expertise’, in Eidinow, E. and Kindt, J. (eds.), Oxford Handbook of Ancient Greek Religion (Oxford, 2015), 301–2Google Scholar; Morgan, K., ‘The Voice of Authority: Divination and Plato's Phaedo ’, CQ 60 (2010), 63–81 Google Scholar; Struck, P., ‘Plato and Divination’, Archiv für Religionsgeschichte 15.1 (2014), 17–34 Google Scholar. L. Maurizio, ‘Questioning the Divide between Technical and Non-Technical Divination: Sortition and Possession at Delphi’, in E. Eidinow and L. Driediger-Murphy (eds.), Negotiating, Communicating, Relating. Approaches to Ancient Divination (forthcoming), argues conclusively that lots were never used by the Pythia. For the use of the lot at Dodona, see Eidinow (n. 2); Parker, R., ‘Seeking Advice from Zeus at Dodona’, G&R 63 (2016), 69–90 Google Scholar; Larson, J., Understanding Greek Religion. A Cognitive Approach (Abingdon and New York, 2016), 97–102 Google Scholar. Larson argues that the priestesses must have communicated directly with the consultants, but she denies (98, and 116, n. 111) that they could have been in a state of altered consciousness because it would have been impossible to tell which deity was possessing them, Zeus or his ritual partner, Dione (many questions are addressed to both). In the Tibetan tradition, however, the possessing deity usually identifies himself or herself before giving a response (see the quotation from Diemberger below), and either this or some other device could have been in use at Dodona.
4 Johnston, S. I., Ancient Greek Divination (Malden, MA, 2008), 63–72 Google Scholar, plausibly reconstructs how the priestesses at Dodona might listen to and interpret the sounds made by doves, ringing cauldrons, rustling leaves, or a murmuring spring, while simultaneously being in an altered state of consciousness.
5 So Dodds, E. R., The Greeks and the Irrational (Berkeley, CA, 1951), 70Google Scholar.
6 For the panoptic vision of the gods, see Manetti, G., Theories of the Sign in Classical Antiquity, trans. Richardson, C. (Bloomington, IN, 1993), 15Google Scholar.
7 Arnott, W. G., ‘Nechung: A Modern Parallel to the Delphic Oracle?’, G&R 36 (1989), 152–7Google Scholar; Flower (n. 1), 227–8, 239.
8 See Arnott (n. 7). For a general treatment of Tibetan oracles, see de Nebesky-Wojkowitz, R., Oracles and Demons of Tibet.The Cult and Iconography of the Tibetan Protective Deities (London, 1956), 409–54Google Scholar.
9 Note, for example, Bowden (n. 1), 16, 33–4, dismissing the conclusions of L. Maurizio's seminal article, ‘Anthropology and Spirit Possession: A Reconsideration of the Pythia's Role at Delphi’, JHS 115 (1995), 69–86 Google Scholar, and Maurizio, L., ‘The Voice at the Centre of the World: The Pythia's Ambiguity and Authority’, in Lardinois, A. and McClure, L. (eds.), Making Silence Speak. Women's Voices in Greek Literature and Society (Princeton, NJ, 2001), 38–54 Google Scholar. Yet even Fontenrose, J., The Delphic Oracle (Berkeley, CA, 1978), 223–4Google Scholar, while rejecting the authenticity of the verse oracles in our literary sources, leaves open the possibility that some Pythias might have had the requisite skill to compose oracles spontaneously in hexameter verse. As Lloyd-Jones, H., ‘The Delphic Oracle’, G&R 23 (1976), 67Google Scholar, points out, ‘The rapid improvisation of hexameters is less difficult than some people imagine; it is helped by practice.’ Luraghi, N., ‘Oracoli esametrici nelle Storie di Erodoto: appunti per un bilancio provvisorio’, Seminari Romani di cultura greca 3.2 (2014), 237–9Google Scholar, however, argues that the total absence of verse oracles from Dodona makes it highly unlikely that Delphic oracles were delivered in verse.
10 Maurizio, M., ‘Shared Meters and Shared Meanings: Delphic Oracles and Women's Laments’, in Dillon, M., Eidinow, E., and Maurizio, L. (eds.), Women's Ritual Competence in the Greco-Roman Mediterranean (Abingdon and New York, 2017), 97–114 Google Scholar. For elegiac couplets as the probable metre of funerary lament, see Nagy, G., ‘Ancient Greek Elegy’, in Weisman, K. (ed.), The Oxford Handbook of the Elegy (Oxford, 2010), 32–3Google Scholar; and Budelmann, F. and Power, T., ‘The Inbetweenness of Sympotic Elegy’, JHS 133 (2013), 13Google Scholar.
11 For example, Amandry, P., La Mantique apollinienne à Delphes. Essai sur le fonctionnement de l'oracle (Paris, 1950), 155–9Google Scholar; Fontenrose (n. 9), 212–24; Morgan, C., Athletes and Oracles. The Transformation of Olympia and Delphi in the Eighth Century b.c. (Cambridge, 1990), 155–6; Bowden (n. 1), 19, 22–4, 33–9Google Scholar; Johnston (n. 4), 49; Hall, J., Artifact and Artifice. Classical Archaeology and the Ancient Historian (Chicago, IL, 2013), 30–1Google Scholar; Scott, M., Delphi. A History of the Center of the Ancient World (Princeton, NJ, 2014), 27–8Google Scholar. However, Stoneman, R., The Ancient Oracles. Making the Gods Speak (New Haven, CT, 2011), 37–9Google Scholar, believes that the Pythia spoke verses herself (‘perhaps not very perfect ones’ [39]); but he leaves open the possibility (following Bowden [n. 1], 36–3) that these verses were improved by freelance oracle-collectors who frequented the sanctuary. This theory is based on late sources (Strabo 9.3.5 and Plut. Mor. 407b–c), and Fontenrose (n. 9), 212–15, is quite right to dismiss their evidence as a later invention intended to explain the tradition of verse oracles from archaic and classical Greece. Luraghi (n. 9), 233–55, by contrast, argues that the verse oracles recorded by Herodotus are the creation of a sophisticated tradition of oral narrative. Similarly, Kindt, J., ‘Delphic Oracle Stories and the Beginning of Historiography: Herodotus' Croesus Logos ’, CPh 101 (2006), 34–51 Google Scholar, and Kindt, J., Revisiting Delphi. Religion and Storytelling in Ancient Greece (Cambridge, 2016)Google Scholar, view oracles as storytelling devices.
12 See Flower, M. A., Xenophon's Anabasis, or The Expedition of Cyrus (Oxford, 2012), 123–5CrossRefGoogle Scholar; and Bruit-Zaidman, L., ‘Xénophon, l'oracle de Delphes et la divination’, Kernos 26 (2013), 59–72 Google Scholar.
13 The widespread use of the lot at Delphi was first proposed by Amandry (n. 11), 29–36. See further Johnston (n. 4), 51–6 (on Delphi) and 68–71 (on Dodona). Johnston, however, assumes that the questions on the lead tablets from Dodona were answered when the priestesses (not knowing what the question was) drew lots marked to signify ‘yes’ or ‘no’; but this reconstruction fails to account for the existence of articulated answers that are written on some of the tablets. For these written responses, see Lhôte, É., Les Lamelles oraculaires de Dodona (Geneva, 2006), 355–7Google Scholar; Eidinow (n. 2), 123–4; and Parker (n. 3), 88–90 (who, however, thinks that almost all of the responses could have been created by a variant of the lot).
14 Maurizio (n. 3).
15 Two striking examples are Eidinow (n. 2), 98, no. 13 = Lhôte (n. 13), 205–8, no. 96 (accepting Eidinow's text rather than Lhôte's), and Eidinow (n. 2), 105–6, no. 6 = Lhôte (n. 13), 156, no. 68. I discuss these more fully in M. A. Flower, ‘Divination and the “Real Presence” of the Divine in Ancient Greece’, in Eidinow and Driediger-Murphy (n. 3).
16 Parke (n. 2), 271, no. 25 = Eidinow (n. 2), 99, no. 14 = Lhôte (n. 13), 227–9, no. 107.
17 See Parker, R., ‘Greek States and Greek Oracles’, in Buxton, R. (ed.), Oxford Readings in Greek Religion (Oxford, 2000), 99–100 Google Scholar; Powell, A., ‘Divination, Royalty and Insecurity in Classical Sparta’, Kernos 22 (2009), 57–60 Google Scholar; Trampedach, K., Politische Mantik. Die Kommunikation über Götterzeichen und Orakel im klassischen Griechenland (Heidelberg, 2015), 305–8Google Scholar. Other alleged examples of bribery are found at Hdt. 5.63.1, 6.123.2, 6.66.3; Plut. Vit. Lys. 24–6, 30; Diod. Sic. 14.13.
18 For text and commentary, see Furley, W. D. and Bremer, J. M., Greek Hymns. Selected Cult Songs from the Archaic to the Hellenistic Period, 2 vols. (Tübingen, 2001), ii.445–52Google Scholar. Note also the translation and discussion in LeVen, P. A., The Many-Headed Muse. Tradition and Innovation in Late Classical Lyric Poetry (Cambridge, 2014), 299–304 Google Scholar.
19 Spirit possession (also called spirit mediumship) is a common phenomenon cross-culturally. In early China we have some evidence for spirit mediums (called wu), but the nature of their role in society is controversial; in contemporary south-western China, Malaysia, Singapore, and Taiwan, freelance spirit mediums, many of whom are women, play an important role in religious life. See Raphals, L., Divination and Prediction in Early China and Ancient Greece (Cambridge, 2013), 2, 98–9Google Scholar, 363; and Cline, E. M., ‘Female Spirit Mediums and Religious Authority in Contemporary Southeastern China’, Modern China 36.5 (2010), 520–55CrossRefGoogle Scholar. Smith, F. M., The Self Possessed. Deity and Spirit Possession in South Asian Literature and Civilization (New York, 2006)Google Scholar, is a magisterial study that emphasizes the pervasiveness of spirit possession in all strata of South Asian society.
20 I would like to thank Professor Matthew King, an expert on Tibetan Buddhism, for assistance with the formulation of this paragraph. A good introduction to this topic is Van Schaik, S., Tibet. A History (New Haven, CT, 2011)Google Scholar.
21 The different types of possession, as well as various modern theories about them, are well surveyed by Smith (n. 19), 33–94.
22 Havnevik, H., ‘A Tibetan Female State Oracle,’ in Blezer, H. (ed.), Religion and Secular Culture in Tibet. Tibetan Studies II. Proceedings of the Ninth Seminar of the International Association for Tibetan Studies (Leiden, 2002), 270Google Scholar.
23 Plut. Mor. 405c–d. See Flower (n. 1), 222–3, 230–2. During the fifth century, the priestesses at Dodona were apparently well educated (Hdt. 2.53, 55).
24 Flower (n. 1), 218–19. The best evidence is for Sparta, where the two kings each appointed two officials called Pythioi, whose job it was to consult Delphi; the texts of the oracles were then kept in the possession of the kings, although the Pythioi also had knowledge of them (Hdt. 6.57). According to the sixth-century bc poet Theognis, it was impious to alter an oracle (lines 805–10).
25 Diemberger, H., ‘Female Oracles in Modern Tibet’, in Havnevik, H. and Gyatso, J. (eds.), Women in Tibet (London and New York, 2005), 136–7Google Scholar.
26 We know the names of three types of male attendants at Delphi: hosios, hiereus, and prophētēs, although only the last of these appears in texts of the classical period. See Maurizio (n. 9 [1995]), 70, 83–4.
27 By Lipsey, R., Have You Been to Delphi? Tales of the Ancient Oracle for Modern Minds (Albany, NY, 2001), 259–76Google Scholar.
28 Ibid ., 262–3.
29 For a fuller discussion, see Flower (n. 1), 235–9.
30 Lama, Dalai, Freedom in Exile. The Autobiography of His Holiness, the Dalai Lama of Tibet (London, 1990), 236Google Scholar.
31 Ibid ., 148.
32 Ibid ., 149.
33 Among modern studies of spirit possession, see especially Sargant, W., The Mind Possessed. A Physiology of Possession, Mysticism and Faith Healing (London, 1973)Google Scholar; Lewis, I. M., Religion in Context. Cults and Charisma (Cambridge, 1986)Google Scholar; Lewis, I. M., Ecstatic Religion. A Study of Shamanism and Spirit Possession, second edition (London, 1989)Google Scholar; Klass, M., Mind Over Mind. The Anthropology and Psychology of Spirit Possession (New York, 2003)Google Scholar; Smith (n. 19).
34 A somewhat different procedure for consultation is reported by the journalist Iyer, P., The Open Road. The Global Journey of the Fourteenth Dalai Lama (New York, 2008), 111–14Google Scholar, who describes a group of Tibetan monks standing around the Nechung oracle, while one of them ‘scribbled furiously, covering page after small page’ (114) for ten minutes or more, apparently in the absence of the Dalai Lama himself.
35 A point well made by Arnott (n. 7), 152: ‘Analogy admittedly is not argument, and the individual reader must judge for himself the applicability of the evidence.’
36 De Boer, J. Z., Hale, J. R., and Chanton, J., ‘New Evidence for the Geological Origins of the Ancient Delphic Oracle (Greece)’, Geology 29.8 (2001), 707–10Google Scholar, sparked a renewed interest in the once common theory that the Pythia prophesied while intoxicated by hydrocarbon gases (they argue for ethylene in particular). The scientific basis for their claims (which were widely accepted in non-academic venues) has been convincingly refuted by Lehoux, D. R., ‘Drugs and the Delphic Oracle’, CW 101.1 (2007) 41–56 Google Scholar, and Foster, J. and Lehoux, D. R., ‘The Delphic Oracle and the Ethylene-Intoxication Hypothesis’, Clinical Toxicology 45 (2007), 85–9Google Scholar.
37 A point nicely made by Prince Peter of Greece and Denmark, ‘Tibetan Oracles in Dharamsala’, in Ligeti, L. (ed.), Proceedings of the Csoma de Körös Memorial Symposium (Budapest, 1978), 328Google Scholar.
38 As astutely argued by Badian, E., ‘Thucydides and the Outbreak of the Peloponnesian War’, in From Plataea to Potidaea. Studies in the History and Historiography of the Pentekontaetia (Baltimore, MD, 1993), 145–52Google Scholar. Contra Cawkwell, G. L., Thucydides and the Peloponnesian War (London, 1997), 34–7Google Scholar.
39 For discussion, see Rubel, A., Die Stadt in Angst. Religion und Politik in Athen während des Peloponnesischen Krieges (Darmstadt, 2000), 123–34Google Scholar; Powell (n. 17), 55–7; Flower, M. A., ‘Athenian Religion and the Peloponnesian War’, in Palagia, O. (ed.), The Timeless and Temporal. The Political Implications of Athenian Art (Cambridge, 2009), 4–9, 16–18 Google Scholar; Kallet, L., ‘Thucydides, Apollo, the Plague, and the War’, AJPh 134 (2013), 362–4Google Scholar; Trampedach (n. 17), 296–7. Thucydides’ attitude to oracles in general is discussed by Oost, S. I., ‘Thucydides and the Irrational: Sundry Passages’, CPh 70 (1975), 186–96Google Scholar; Marinatos, N., ‘Thucydides and Oracles’, JHS 101 (1981), 138–40Google Scholar; Dover, K. J., ‘Thucydides on Oracles’, in The Greeks and Their Legacy. Collected Papers. Volume II. Prose, Literature, History, Society, Transmission, Influence (Oxford, 1988), 65–73 Google Scholar.
40 For example, Westlake, H. D., ‘λεγεται in Thucydides’, Mnemosyne 30 (1977), 349–50Google Scholar, 354; Powell (n. 17), 47 and 55, n. 42; Bonnchere, P., ‘The Religious Management of the Polis: Oracles and Political Decision-Making’, in Beck, H. (ed.), A Companion to Greek Government (Malden, MA, and Oxford, 2013), 372Google Scholar. On the other hand, Fontenrose (n. 9), 33 and 246, includes this oracle (his H5) among the only seventy-five in his collection that he considers to be ‘historical’ (in the sense of deriving from contemporary sources), and his discussion assumes that he considers it to be genuine as well. Parke, H. W. and Wormell, D. E. W., The Delphic Oracle, 2 vols. (Oxford, 1956)Google Scholar, i.189, while acknowledging Thucydides’ reservations, conclude that this was the version in circulation and that ‘there is no reason to doubt that Delphi came down wholeheartedly on the side of Sparta’.
41 Demont, P., ‘Les oracles delphiques relatifs aux pestilences et Thucydide’, Kernos 3 (1990), 150Google Scholar, while noting Thucydides’ apparent hesitation, concludes that he presents it in the narrative as being authentic.
42 On this formula, see Fontenrose (n. 9), 37–8.
43 Heraclitus: Diels-Kranz F 93 = Plut. Mor. 404d; Arist. Rh. 1407a32–7. Theopompus: F. Jacoby, FGrHist 115 F 336 = Plut. Mor. 403e–f. For arguments against the authenticity of ambiguous and riddling oracles, see Fontenrose (n. 9), 80; for their authenticity, see Parker (n. 17), 80; Maurizio (n. 9 [1995]), 79–83; and Larson (n. 3), 101–2.
44 Maurizio (n. 9 [1995]), 79–83; note also Flower (n. 1), 221.
45 For a different view, see Kallet (n. 39), 363, n. 18.
46 For texts and translation, see Parpola, S., Assyrian Prophecies (Helsinki, 1997)Google Scholar, and Nissinen, M. (ed.), Prophets and Prophecy in the Ancient Near East (Atlanta, GA, 2003), 97–132 Google Scholar. For a comparison with Delphic oracles, see Flower (n. 1), 228–30. The translation given here is Parpola no. 1.1 (it is translated somewhat differently as Nissinen no. 68).
47 It is commonly asserted, based on Thuc. 5.87, that as much as a third of Athens’ adult male citizen population was killed by the plague between 430/429 and 427/426. Hornblower, S., A Commentary on Thucydides (Oxford, 1991)Google Scholar, i.494, rightly expresses caution about such estimates.
48 Mikalson, J. D., ‘Religion and the Plague in Athens, 431–423 bc ’, in Studies Presented to Sterling Dow on His Eightieth Birthday, Greek, Roman, and Byzantine Monographs 10 (Durham, NC, 1984), 220Google Scholar, incorrectly claims that it was a convention of Athenian popular religion not to attribute diseases to a specific deity but to a ‘nameless, formless, and cultless δαίμων’. See rather H. W. Parke and D. E. W. Wormell (n. 40), i.189–90; Parker, R., Miasma (Oxford, 1983), 275Google Scholar.
49 Kallet (n. 39), 364.
50 Parker (n. 3), 70, assumes that it was in prose and considers it a straightforward response.
51 The bibliography on the social function of divination is extensive. For an anthropological perspective, see Park, G. K., ‘Divination and Its Social Contexts’, Journal of the Royal Anthropological Institute 93 (1963), 195–209 Google Scholar (a classic study); Fortes, M., Religion, Morality and the Person. Essays on Tallensi Religion (Cambridge, 1987), 1–21 CrossRefGoogle Scholar, esp. 11; Peek, P. M. (ed.), African Divination Systems. Ways of Knowing (Bloomington, IN, 1991)Google Scholar; Abbink, J., ‘Reading the Entrails: Analysis of an African Divination Discourse’, Man 28 (1993), 705–26CrossRefGoogle Scholar. For ancient Greece specifically, see Parker (n. 17); Morgan (n. 11), 153–7; Eidinow (n. 2), 1–55, 125–38; Flower (n. 1), 1–21, 104–31.
52 Prince Peter of Greece and Denmark, ‘Tibetan Oracles’, in Fisher, J. F. (ed.), Himalayan Anthropology. The Indo-Tibetan Interface (The Hague, 1978), 295Google Scholar.