Hostname: page-component-797576ffbb-gvrqt Total loading time: 0 Render date: 2023-12-04T00:19:44.776Z Has data issue: false Feature Flags: { "corePageComponentGetUserInfoFromSharedSession": true, "coreDisableEcommerce": false, "useRatesEcommerce": true } hasContentIssue false


Published online by Cambridge University Press:  29 March 2016


The consultation of fixed oracles was a central and well-documented feature of ancient Greek life; Greece ought therefore to be able to contribute important evidence to anyone interested in the function of divination in different cultures. But, although no oracle is more famous than that of Apollo at Delphi, so many central questions about its operations are unanswerable that it provides a very shaky basis for comparison. A few points are secure: not only individuals but also states put enquiries to the oracle; and public enquiries related to matters of religious practice and cult, but also on occasion to colonizing projects, alliances, or declarations of war. Beyond these generalities, however, almost everything is contestable. Even the statement just made about public enquiries requires some hedging. On matters of cult they certainly never ceased: in a religion without revelation and specialized religious institutions, the direct access to divine will supposedly provided by the oracle was indispensable to authorize change, or to suggest ritual remedies in a time of crisis such as plague. But the extent to which public enquiries on military and political matters continued to be made after the fifth century is very uncertain.

Research Article
Copyright © The Classical Association 2016 

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)



Abbreviations of classical texts are taken from S. Hornblower and A. Spawforth (eds.), The Oxford Classical Dictionary, fourth edition (Oxford, 2012). The tablets cited in what follows are edited by E. Lhôte, Les Lamelles oraculaires de Dodone (Geneva, 2006) and S. Dakaris, J. Vokotopoulou, and A. P. Christidis, Τα Χρηστήρια Ἐλάσματα της Δωδώνης των ἀνασκαϕών Δ. Ευαγγελίδη, 2 volumes (Athens, 2013). I cite responses from these as Lhôte no. X and DVC X. All translations are my own.


1 Contrast R. Parker, ‘Greek States and Greek Oracles’, in R. Buxton (ed.), Oxford Readings in Greek Religion (Oxford, 2000), 101–5; H. Bowden, Classical Athens and the Delphic Oracle (Cambridge, 2005), 152–9.

2 See above all J. Fontenrose, The Delphic Oracle (Berkeley, CA, 1978); for a strong defence of the authenticity of the verse form as reported in Herodotus, see M. Flower, The Seer in Ancient Greece (Berkeley, CA 2008), 215–39.

3 R. Crahay, La Littérature oraculaire chez Hérodote (Paris, 1956).

4 For what can be done, see E. Eidinow, Oracles, Curses, and Risk Among the Ancient Greeks (Oxford, 2007), 45–9, 50–3.

5 See ibid., 56–71. Some Dodona responses also occur in literature, usually suffering from the same narrative elaboration that makes the Delphic material so problematic; for the few that are more reliable see H. W. Parke, The Oracles of Zeus (Oxford, 1967), 83–6, 137–43, 149.

6 Lhôte (acknowledgement note), 11–15, argues that no tablet postdates the Roman attack and destruction of 167; for Strabo 7.7.10 (327), writing a century and a half after that, the oracle was almost extinct. But SEG 28, no. 530, is an iron strigil inscribed with a flattering oracle in verse delivered to the pirate king Zeniketes (died 74 bc); the strigil was presumably dedicated by him at the sanctuary.

7 A few examples from early in the collection: ‘better for me acquiring a wife/woman?’ (wife or slave?), DVC 19; ‘will the guardianship which I Lykkidas now have be lucky for me?’ (guardianship of what?), DVC 31; ‘the iron pieces [coins?] of Philotis justly?’, DVC 36.

8 See Eidinow (n. 4), 72–138. Some of the contents of DVC were made available to Eidinow by Christidis in provisional form, but without the drawings that allow some control over readings. On the basis of these I have drawn back from some of Christidis’ suggestions that she accepts.

9 They are now in the Charlottenburg in Berlin, and their condition is so poor that they may be unusable (Lhôte [acknowledgement note], 7); only one has ever been published.

10 See acknowledgement note above.

11 Demotic: see E. Bresciani, L'archivio demotico del tempio di Soknopaiu Nesos (Milan, 1975), nos. 1–12; G. Martin, ‘Questions to the Gods: Demotic Oracle Texts from Dimê’, in F. Hoffmann and H. J. Thissen (eds.), Res severa verum gaudium (Leuven, 2004), 413–26. Greek: Papini, L., ‘Domande oracolari: elenco delle attestazioni in greco ed in copto’, Analecta Papyrologica 4 (1992), 21–7Google Scholar; Savorelli, G. Messeri and Pintaudi, R., ‘Due domande oracolari in greco’, ZPE 111 (1996), 183–7Google Scholar, with the corrigendum in ZPE 117 (1997), 211–12. Coptic: Papaconstantinou, A., ‘Oracles chrétiens dans l'Egypte byzantine: le témoignage des papyrus’, ZPE 104 (1994), 281–6Google Scholar. For an overview, see G. Tallet, ‘Oracles’, in C. Riggs (ed.), The Oxford Handbook of Roman Egypt (Oxford, 2012), 398–418.

12 Also, some published by Evangelidis from the earlier excavations seem not to reappear in DVC, e.g. Lhôte nos. 25 and 27. Dosuna, J. Méndez, ‘Nοvedades en el oráculo de Dodona: a propósito de une reciente monografía de Éric Lhôte’, Minerva 21 (2008), 53Google Scholar, speaks of a total of approximately 8,000.

13 Lhôte's rendering of a fragmentary tablet (his no. 31) ‘[est-il bon pour Untel] de rechercher ce qui est vrai’ (‘[is it good for such and such a person] to enquire about what is true’) will therefore be viewed with scepticism. For the question at Claros, see R. Lane Fox, Pagans and Christians (Harmondsworth, 1986), 168–77, 191–6.

14 For other clear cases, see Lhôte nos. 84, 103 (and his discussion, p. 354); DVC 35 with 37.

15 See also Lhôte nos. 53Bb(?), 135; DVC 123, 3661(?). Cf. Hdt. 5.43; Xen. Hell. 3.1.6.

16 More fully in De E apud Delphos 4, 386c: ‘Will they marry, will they win, if it's beneficial to sail, farm, go abroad’. In portions of the Theophaneia surviving in Syriac, Eusebius accuses pagans of asking only about marriage, travel, blindness, disease (Euseb. Theophaneia, 50.12–4, in Die Theophanie, ed. H. Gressmann [Leipzig, 1904], 103), run away slaves, lost pots, land deals, trade, marriage and such like (Euseb. Theophaneia 52.28–31 in ibid., 104).

17 So Lhôte (acknowledgements note), 336.

18 See the commentary on DVC 191.

19 Equally doubtful is the identification of the Tharyps of 2148 with a Molossian king.

20 Other instances are DVC 268(?), 295, 1089, 2425, 2519, 2952.

21 Kassopaioi(?) DVC 363; Genoaioi(?) DVC 1042; Thronion DVC 1184; Bulliones DVC 2364 = Lhôte no. 7; Chaones Lhôte no. 11; Phanote DVC 3822; the Epirotic league(?) DVC 3977; Athamanians DVC 4016; Molossians DVC 4195. There are also the –esmaioi (Lhôte no. 13) and perhaps an otherwise unknown community of Dexaireatai (DVC 1070).

22 Lhôte no. 8b, about lending a goddess’ money; Lhôte no. 11, about moving a temple.

23 DVC 268, 1089, 1834(?), 2519.

24 N. H. Demand, Urban Relocation in Archaic and Classical Greece. Flight and Consolidation (Bristol, 1990); Lhôte no. 12 is apparently similar. John Ma points out to me that the situation might be one in which the Dodonaians have already withdrawn with their property to a stronghold and are wondering about staying there: see also Müller, H., ‘Φυγῆς ἕνεκεν’, Chiron 5 (1975), 129–56Google Scholar.

25 Such fluctuations between first and third person are common in the tablets. What ‘it’ is here is unclear.

26 ἐπιγγ[ύ]ασις, a strange word in this context.

27 Lhôte no. 52 is another multi-purpose question.

28 See the commentaries to DVC 24 and 203 for lists. For the related word παγκλαρία (‘whole portion’, of property), see the note on DVC 45.

29 The note on DVC 1 lists more than a hundred possibly related questions.

30 Questions where the bride is named: e.g. Lhôte no. 22 Ba; DVC 165, 208, 449 (with splendid brevity: ‘Archedamos [to wed] Timokleia, better?’), 999, 1127, 2052 (the Epidamnian woman), 2470. Questions where the bride is not named: e.g. Lhote nos. 22 Bb, 25; DVC 328 (combined with a question about property), 1352, 1356, 2506, 3487 (combined with a question about ‘going home’). The gender of the enquirer of DVC 244 (‘Should I cohabit or not?’) is unclear.

31 Except in Lhôte no. 53 Bb, if rightly supplemented to give ‘the woman I have in mind’; cf. perhaps Lhôte nos. 26 (note the article) and 135.

32 DVC 2387; cf. DVC 4161: ‘If I marry, will I get a male child who will survive?’

33 DVC 2508 and Lhôte no. 39 (on which see Méndez Dosuna, ‘Nοvedades en el oráculo de Dodona’, 62). To these could possibly be added Lhôte no. 53 Ac and DVC 1302.

34 DVC 1208: ‘Pitthis. Would it be better if she [betrothed] Thebais her daughter…’; this is a probable example, though ‘Thebais’ is wrongly written in the nominative instead of the accusative case. See also perhaps DVC 2413, 3113.

35 Lhote nos. 29, 30 (= DVC 1), 33–34; DVC 1407 and 2322 are clear and apparently complete instances. DVC 465, 1033, and 2539 may be fragments: they do not look like labels, because they are introduced by an interrogative particle.

36 See also p. 89 below.

37 See also perhaps DVC 1463, and possibly Lhôte no. 36, where it is clear from the new evidence that the last word comes from the verb ἐFάω, ‘let go’ (Méndez Dosuna, ‘Nοvedades en el oráculo de Dodona’, 67–8).

38 Lhôte no. 41, ‘useful offspring in addition to what I have’; Lhôte no. 47, male offspring; Lhôte nos. 46A, 50 Aa; DVC 2050. DVC 1391 asks what god to pray to in order to secure the permanence of the ‘coming generation’.

39 See also Lhôte no. 44, DVC 497, 3034, 3554 (male offspring from the present wife).

40 See also e.g. 251, 1391, 2768 (‘profit of my children’), 4161. Lhôte no. 52 lays bare the issue: ‘children to rear me in old age’.

41 The fragmentary DVC 4115 is supplemented to give a less drastic alternative: ‘Will I get offspring by dedicating (feminine) to a different goddess?’ Juvenal satirically imagines a lower-class woman at Rome asking a fortune-teller whether she should ‘abandon the innkeeper and marry the clothes-seller’ (Satire 6.591).

42 See Eidinow (n. 4), 83 and 85 no. 10, reporting Christidis’ view that two questions are involved.

43 See DVC's note on 8.

44 I give a small selection; for a long list see the commentary to DVC 3.

45 The editors quote a law of Solon whereby anyone not finding water within ten fathoms should rely on a neighbour (Plut. Vit. Sol. 23.6).

46 There are several further references to θησαυροί (see DVC's note on 126), but the word is ambiguous (it can mean strong box or store room) and the force unclear.

47 Taken by Lhôte as put by a manumitted slave with the right of free movement: possible, but not certain.

48 Lhôte no. 54 = DVC 1380; Lhôte114 = DVC 24–25; Lhôte 131; possibly DVC 190, 524, 552.

49 Lhôte no. 130 and DVC 463: Pharos; Lhôte no. 133: Sybaris; DVC 3109 concerns an unnamed colony (ἐποικία).

50 The dating is problematic: its content would suggest Timoleon's panhellenic appeal for new colonists in the late 340s (Diod. Sic. 16.82.5), but the spelling appears to push it earlier (see Lhôte ad loc.).

51 Respectively Lhôte no. 55 = DVC 354 (DVC wonder whether Eurynous is the questioner and is thinking of exporting his children in a time of crisis); Lhôte no. 56; Lhôte no. 57 =DVC 3472 (DVC 993 is similar from a woman).

52 Lhôte no. 52 may be about resuming citizen rights at Athens.

53 Lhôte no. 127 (‘on land?’); Lhôte no. 128 (against Antiochos); DVC 471 (παρὰ τὸν βασιλῆ, ‘with the king[?]’); DVC 2981 (as cavalry); DVC 3648 (going to Pyrrhos); more vaguely DVC 2625, 3811.

54 See the commentary on DVC 57, a plain ‘Should I farm?’

55 Probably: but Eidinow (n. 4), 95, notes that the verb can also mean ‘treat medically’.

56 As Angelos Chaniotis has suggested to me.

57 See the note on DVC 206.

58 Lhôte nos. 71, 72, DVC 556 (apparently hereditary). Other health enquiries e.g. Lhôte nos. 46 Ba, 66 (individuals); 65, 68, DVC 2242 (whole families); Lhôte no. 73 (a son).

59 DVC 2549, 3174. On the healing touch, see Herodas 4.18 with the note of W. Headlam in his edition (Cambridge, 1922), ad loc. Lhôte no. 50 Ab is taken as a question about paying for the treatment of a son.

60 For a list, mostly very fragmentary, see the note on DVC 287.

61 See the note on DVC 73.

62 Lhôte no. 63, and see also 62 and 64, with the comments of Méndez Dosuna, ‘Nοvedades en el oráculo de Dodona’, 59. The isolated phrase ‘go away free’ in DVC 3356 does not help.

63 G.M. Browne (ed.), Sortes Astrampsychi. Vol. I (Leipzig, 1983), question 89.

64 For oaks, see DVC 1108, 2951; see also the public enquiry of the Dodonaians (DVC 2519), ‘is there a sign in the oak?’ On hero shrines, see DVC 80 (this, too, is an oak: the questioner asks if it should it be dedicated in a temple after removal), 2432 (possibly an answer), 3838.

65 A ruling on the subject forms part of the religious rules sought from Apollo of Delphi by Cyrene: P.J. Rhodes and R. Osborne, Greek Historical Inscriptions, 403–323 (Oxford, 2003), no. 97 A 8–10.

66 DVC 2128. There was no inhibition about this: Archephon in Lhôte no. 94 refers in a question to Zeus to a ship he has built ‘on the instructions of Apollo’.

67 See R. Parker, Miasma (Oxford, 1983), 254, on ‘turning the fair side outwards’.

68 See also DVC 224. Simple labels ‘about a/the court case’: DVC 423, 747, 1124.

69 Whether to litigate: DVC 142, 192, 1447(?), 1681, 2186, 3132; what god to pray to: DVC 436. DVC 2284 possibly asks whether to seek arbitration or to go to court.

70 See also DVC 825, 2089. DVC 1993 and 2036 apparently ask ‘how/by praying to whom?’ athletic victory can be achieved.

71 See the note on DVC 70. DVC 1397 reads ‘and about priesthood whether…’, but the question is lost.

72 The verb which this word is taken to come from and which is translated ‘die’ in fact means ‘kill’; it is possibly a noun, ‘death’, and the question is whether a person has died, as in DVC 2980.

73 On Lhôte no. 129, Méndez Dosuna, ‘Nοvedades en el oráculo de Dodona’, 76. I cannot classify DVC 1153: ‘Will there be any penalty?’

74 If one takes the new verb attested there, δατεύω, as synonymous with δατέομαι (‘I divide’).

75 See perhaps DVC 328, ‘about “whole property” (παγλαρία) whether it will occur/come to him and should he take a wife?’ All instances of παμπησία and παγκληρία in the literary sources can be taken as referring to an inheritance, inherited property, or the ancestral territory of a people (παγκληρία). But questions from Dodona about παμπασία, when not completely vague, seem to be about the acquisition of new wealth (by praying, trading, buying), not inheritance (DVC 261, 998, 1484, 2110, 2593, 2802); none of those about παγκλαρία is decisive either way.

76 Cf. now DVC 3550, ‘Is Tata child of Iphinoos?’ For a famous legitimacy question at Delphi, see Hdt. 6.66.

77 Cf. DVC 115 (= Lhôte no. 124) and perhaps DVC 1113.

78 The Greek word ϕάρμακον combines the senses of poison and bewitchment. Whether poisoning/bewitchment has occurred: DVC 272, 452 (= Lhôte no. 125b); naming of a suspect: DVC 1028 (= Lhôte no. 125) and perhaps ‘Lyson’ of DVC 452 (= Lhôte no. 125b); ‘about poisoning/bewitchment’: DVC 962. On charges of poisoning and magical attack in funerary epitaphs, see the studies cited in SEG 60, no. 2024.

79 See the references in the commentary on DVC 33; Lhôte nos. 119–22.

80 Cicero from Callisthenes: Cic. Div. 1.76, 2.54, 2.69 = FGrH 124 F 22(a) and (b).

81 DVC 1170, 1410, 2222. See also Parker, R., ‘The Lot Oracle at Dodona’, ZPE 194 (2015), 111–14Google Scholar. On the ticket oracles, see n. 11 above.

82 Cf. Lhôte (acknowledgements note), 355–7; Eidinow (n. 4), 123–4; the list attached to DVC 42: many even of the few cases the editors adduce are doubtful.

83 Lhôte nos. 141 Ba, 142; DVC 2393 and perhaps 1122; with single divine names: DVC 585, 1045, 1299; Lhôte no. 166c.

84 DVC 24–25 = Lhôte no. 114; DVC 107–8 = Lhôte no. 127. Other possibilities include Lhôte nos. 12B and 35Bb; DVC 2432.

85 Lhôte no. 95. There are problems of interpretation here, however. Lhôte gives the received interpretation (see especially F. Salviat, ‘Timodamos et son gaulos: oracles et marchands à Dodone’, in P. Cabanes (ed.), L'Illyrie méridionale et l'Epire dans l'antiquité II. Actes du IIe Colloque international de Clermont-Ferrand, 25–27 octobre 1990 [Paris, 1993], 61–4): according to this, the first side contains a question with a subsequent interlinear modification (clear, in fact, on the photo); the back, in the same hand, contains two responses, one to the first question and one to the modification, the second of which reverses the first. But such a passing of the tablet to and fro between consultant and oracle is very puzzling.

86 Wilhelm, A., ‘Orakelfragen und Orakelantworten’, APF 15 (1953), 76–7Google Scholar, likewise plausibly interpreted Lhôte no. 45 as ‘(Answer) to Kleanor about offspring to inherit. From your present wife’.

87 Eidinow (n. 4), 67–8. On the responses in literary sources, see n. 5 above.

88 See e.g. J. Černý, ‘Egyptian Oracles’, in R. A. Parker, A Saite Oracle Papyrus from Thebes in the Brooklyn Museum (Providence, RI, 1962), 35–48; Tallet, ‘Oracles’, see n. 11 above.

89 See e.g. K. Thomas, Religion and the Decline of Magic (London, 1971), index s.v. divination, and the works he cites.