1. However, it is important to note that even a casual use may be occasionally pressed for its literal meaning: when at Clouds 817ff. Pheidippides exclaims ‘You're crazy, by Olympian Zeus!’, Strepsiades expressly recognizes the phrase as an oath (825), and assumes that it implies belief in the god named (819).

2. J. Mikalson remarks that ‘in oath-taking an individual most often faced the choice between pious and impious action, deciding between what was pious and what might bring social or financial gain. For this reason the maintenance of oaths, at the popular level, was often treated as the key element of personal piety’ (Honor Thy Gods, Chapel Hill, 1991, p. 80Google Scholar). The basic source book on oaths is still Hirzel's, R.Der Eid (Leipzig, 1902, repr. Arno PressGoogle Scholar, 1979), which is much more detailed than Plescia's, J.The Oath and Perjury in Ancient Greece (Tallahassee, 1970Google Scholar). See also Bonner, R. J. and Smith, G., The Administration of Justice from Homer to Aristotle (Chicago, 1938), vol. 2, pp. 145–91Google Scholar, Dover, K. J., Greek Popular Morality in the Time of Plato and Aristotle (Oxford, 1974), pp. 248–51Google Scholar, Burkert, W., Greek Religion (Harvard, 1985), pp. 250–4Google Scholar, and Mikalson's, previous book, Athenian Popular Religion (Chapel Hill, 1983), pp. 31–8Google Scholar.

3. . The locus classicus is Thuk. 2.52f.: ‘Overwhelmed by the disaster, the people did not know what would happen to them and took little account of secular and sacred things alike (2.52.3)... Neither fear of the gods nor human law restrained them; they judged it all the same whether to act piously or not, seeing that everyone was dying in equal measure (2.53.4).’ (All translations are my own.) Mikalson, J., ‘Religion and the Plague in Athens, 431–423 B.C.,’ GRBS Monograph 10, (1984), 217–25Google Scholar examines the evidence for religious activity during the plague years and finds Thukydides' pessimism extreme, though he does not doubt that the effects of the plague were profound.

4. Sokrates is of course the outstanding victim of this backlash; the charge of impiety, whatever lay behind it, must have had emotional appeal. Recall too the public outrage over the blasphemies of 415 (the mutilation of the herms and profanation of the Eleusinian Mysteries), to which may possibly be added a number of more dubious incidents involving charges of asebeia (impiety) and the suppression of intellectual freedom. See further below, p. 146.

5. Op. cit. (n. 2), p. 250.

6. Oaths were the basis of international treaties (Mikalson, , Athenian Popular Religion, p. 34Google Scholar). Within the polis, , Lykourgos, (Leokrates 79Google Scholar) mentions the oaths of the archons, jurors, and private citizens, and earlier (77) invoked the ephebic oath taken by all young men on entering military service. In the fifth century all citizens swore loyalty to the restored democracy in 411 and again in 403 (Hirzel, , op. cit., p.31Google Scholar; Bonner, and Smith, , op. cit., p. 150Google Scholar). Business contracts were based on oaths: Burkert, , op. cit., p. 253Google Scholar, Rauh, N., The Sacred Bonds of Commerce, (Gieben, 1994Google Scholar). On the question of oaths taken by witnesses in court, see below, n. 9.

7. Athenian Popular Religion, p. 31.

8. . Cf. Thalheim, T., RE s.v. ‘Epiorkia’, vol. 6, col. 191Google Scholar: ‘Von einer gesetzlichen Strafe hören wir nirgends’ (‘we hear nowhere of a statutory penalty’), and Burkert, , op. cit., p. 252Google Scholar: ‘the conviction exists that only fear of the gods provides a guarantee that oaths will be kept.’ A separate issue is the matter of false witness (pseudomartyria or pseudomartyrion) which was an actionable offence only within the context of an ongoing trial. Aristotle claims (Polit. 1274b5) the procedure was instituted in Sicily by Charondas (6th c.?), but the evidence for Athens begins, as often, only with the orators in the late fifth and fourth centuries, and much remains unclear. While pseudomartyria is often translated as ‘perjury’, I can find no example where oaths are explicitly said to have been violated, and I suspect that oaths were not in fact involved. This interpretation depends on the vexed question whether Athenian witnesses had to take an oath before giving testimony in all cases, as they did in cases of homicide. See further Lipsius, J. H., Das Attische Recht und Rechtsverfahren (Leipzig, 1908), vol. 2, pp. 777–83Google Scholar; Calhoun, G., CP 10 (1915), 1–7Google Scholar; Bonner, and Smith, , op. cit., pp. 190, 261–9Google Scholar; and Plescia, above a 2, pp. 88–90.

9. The author of the Rhetorica adAlexandrum (17.1) expressly mentions ‘disgrace among men’ as an inhibiting factor, but lays greater stress on divine vengeance–all in the context of rhetorical strategy, to be sure, which says little about religious belief but much about what was publicly acceptable.

10. Cf. Antiphon 5.11, Demosthenes 23.68; Mikalson, Athenian Popular Religion, p. 36Google Scholar.

11. Zeus Horkios is first attested in tragedy: Eur, . Hipp. 1025Google Scholar, Soph, . Phil. 1324Google Scholar, and the subject of an impressive statue described by Pausanias 5.24.9ff., but Zeus is already associated with oaths in Homer, II. 3.279, 7.411, 19.260: Od. 19.303.

12. Cf. Hesiod, , W&D 283ffGoogle Scholar; Mikalson, , Athenian Popular Religion, p. 36. This is of course a particular application of the general notion of sons paying for the sins of the fathers, articulated by, among others, Solon 1.25–32. Alternatively, perjurers might themselves pay in the afterlife, as indicated by Frogs 145–50, but this belief seems less prevalentGoogle Scholar.

13. One passage in Aristophanes suggests the gods could use some help: Peisthetairos thinks the birds would be quicker to punish perjurers who swore falsely in their name (Birds 1608 ff.).

14. Hirzel, , op. cit, pp. 81ff., 208ff.Google Scholar, and Plescia, , op. cit., p. 87 make a plausible case for a decline in the trustworthiness of the oath during the fifth century. That there was a ‘crisis of faith’ in this period cannot be denied, even if the dimensions are disputable, and that the use of oaths reflected this crisis will be shown below. But belief in the gods and the practice of oaths certainly survived, and many of the staunchest endorsements of the oath, such as the oft-quoted passage from Lykourgos (Leok. 79), date from the fourth century and later. Moreover, Hirzel himself admits (p. 22) that oaths were often treated lightly ‘schon in ältester Zeit’ (‘already in earliest times’). It seems safe to say that the oath was used and abused at all stages of Greek historyGoogle Scholar.

15. Hirzel, , op. cit, pp. 41ffGoogle Scholar. is right to draw a distinction between outright perjury and subtle sophistries that might literally be true (and so fulfil the obligations of the oath, but wholly misleading. Experts in the latter, such as Autolykos, would likely be admired for their cleverness rather than condemned.

16. Athenian Popular Religion, pp. 7–12.

17. . For detailed commentary, see Henderson, J., Lysistrata (Oxford, 1967), ad loccGoogle Scholar.

18. . Hirzel, , op. cit, pp. 84Google Scholarf. discusses this ‘meaningless’ use (esp. in the orators) as evidence for his thesis of chronological decline, which, as has already been mentioned above, is dubious, or at least difficult to prove; again, he notes that Homeric men and gods are already very quick to use oaths. The evidence from the orators is ambivalent: according to Blass, F., Die Attische Beredsamkeit (Leipzig, 1893), vol. 3, pt. 1, p. 82Google Scholar, earlier orators almost never used colloquial oaths like nē Dia, but these became almost clichés in Demosthenes and his contemporaries. Yet Demosthenes' famous oath by the casualties of the Persian Wars (On the Crozvn 208), one of the triumphs of ancient rhetoric according to Blass, p. 177, drew eloquent praise from ‘Longinus’ (On the Sublime 16) as an outstanding example of the omotikon schēma, the ‘oath figure’. Effective use of oaths in court cases was a recognized feature of oratorical technique: cf. Aristotle, , Rhet. 1377a8ff.Google Scholar, Rhetorica ad Alexandrum 1431b 7ff.

19. Cf. Ziebarth, E., RE, s.v. ‘Eid’, vol. 5, coll. 2076f. (cf. note 8)Google Scholar.

20. Honor Thy Gods, pp. 82–6.

21. Id., p. 82.

22. Apollo urges the jurors to consider their oath of less value than the will of Zeus; Athena on the contrary reiterates the importance of the oath (Eum. 708—10). Podlecki, A. in Aeschylus' Eumenides (Warminster, 1989Google Scholar), ad loc, points out that Apollo claimed that the marriage bond is ‘greater than any oath’ (Eum. 217f.). In the Elektra Orestes urges the paidagōgos to add a false oath to the fictional report of his death, but the servant does not in fact do so.

23. Honor Thy Gods, p. 85.

24. Ibid.

25. Ibid.

26. Id., p. 283 n. 88.

27. Dover, K. J., Aristophanes' Frogs: a Commentary (Oxford, 1993), pp. 40fGoogle Scholar.

28. The phenomenon of ‘discontinuity of characterization’ is discussed (without reference to this passage) at length elsewhere by Dover, K. J., Aristophanic Comedy (Berkeley, 1972), pp. 59–65Google Scholar. Such discontinuity is usually for the sake of a joke, of which three follow here in rapid succession. In addition, the naval context suggested itself due to the poet's (and surely the public's too) preoccupation with the recent battle of Arginousai, which is referred to three times in the play (33f., 190ff., 693ff.).

29. Mikalson, , Honor Thy Gods, p. 85Google Scholar.

30. TAPA 99 (1968), 19–35Google Scholar; quote p. 35.

31. Id., 25.

32. Id., 35.

33. Honour Thy Gods, p. 86.

34. Dover, , Frogs, pp. 16ff, with examplesGoogle Scholar.

35. Of course the former's charge was unofficial, in the context of a civil suit. Dover, K. J., Talanta 7 (1976), 29Google Scholar, evaluates the sole source for the latter's prosecution (Satyros' Vita Euripidis, col. 10) and finds it wanting, as it lacks corroboration and comes from an author who considered the plot of Ekklesiazousai to be a historical event (p. 29). We can agree with Dover, that ‘the extent to which [Euripides] laid himself open to attack for impiety should not be exaggerated’ (p. 43Google Scholar), but neither should it be minimized: I would not say Euripides was trying to be impious, but he was certainly trying to be controversial, and this was risky in the late fifth century.

36. The fragment comes from Diogenes Laertius 2.18, which includes similar claims from other comic poets. The two were likewise connected in later tradition; see passages cited in Kassel-Austin, PCG III. 2, 2l7Google Scholar. As noted above, Sokrates is obliquely associated with Euripides at Frogs 1491, again in the context of ‘babbling’ (lalein).

37. Strepsiades' Byzantine witticism puns on the word nomisma: coinage, but also custom, usage-recall that Sokrates was charged with ou theous nomizein, not acknowledging the gods.

38. That Strepsiades immediately swears by Zeus after being told it is no longer valid is perhaps another small joke, although earlier his student guide had also done so (217), as does Sokrates himself (330).

39. It must be noted that Sokrates' argument here is not ridiculous, but rather strikes to the heart of the problem: how could people believe the gods punished perjury in the face of obvious evidence to the contrary? This is potentially very serious ground, perhaps too serious—and so the discussion quickly turns to the nature of lightning (403).

40. E.g., Air and Ether at 264f. (recall Ether in the quotations from Euripides, at Thesm. 272Google Scholar, Frogs 100); a trinity of Chaos, Clouds, and Tongue at 424 (recall too the emphasis on ‘tongue’ in Hipp. 612).

41. It is interesting that the opening words of Pythagoras' On Nature (ou ma ton aera, ton anapneō, Diog. Laert. 8.6) seem to echo this oath of Sokrates (ma tēn anapnoēn, ma to Chaos, ma ton Aera), although the Pythagorean context does not seem controversial: he swears ‘by the air I breathe and the water I drink’ that he will bear no blame for his treatise. Cf. Hirzel, , op. cit., p. 99Google Scholar n. 2, and for the latest attempt to trace Pythagorean elements in the Clouds, Marianetti, M., Religion and Politics in Aristophanes' Clouds (Hildesheim, 1992), pp. 63ffGoogle Scholar.

42. Variations of this joke were good enough to serve later philosophers whose belief in the gods was called into question, at least according to Diogenes Laertius 2.102 (attributed to Theodoros of Kyrene) and 6.42 (Diogenes the Cynic), cited by Neil, R., The Knights of Aristophanes (Cambridge, 1901), ad locGoogle Scholar.

43. Not including Sokrates, Dover, n. 35, 24 f., counts 14 separate instances involving nine individuals, including Diagoras, Aspasia, Anaxagoras, Protagoras, Euripides, and Prodikos. He is inclined to dismiss all but the case of Diagoras. Ostwald, M., From Popular Sovereignty to the Sovereignty of Law (Los Angeles, 1986), pp. 274ff., 528ff.Google Scholar, admits the trials vs. Anaxagoras and Protagoras as well. The case for an extensive witch hunt was made by Dodds, E. R., The Greeks and the Irrational (Berkeley, 1951), pp. 189ffGoogle Scholar.

44 . Cf. Hirzel, , op. cit., pp. 100f. a 1Google Scholar.

45. Numerous variations of this scholion are found elsewhere; see passages cited by Kassel-Austin ad Kratinos fr. 249. Cf. Hirzel, , op. cit., pp. 96 n. 1 and 100 n. 3Google Scholar.

46. ‘Euripidean oath: perhaps the one by the dog or goose’ in Paroemiographi Graeci, edd. Leutsch, E. and Schneidewin, F. (Göttingen, 1839), vol. 1, 413Google Scholar, cited by Hirzel, , op. cit., p. 101Google Scholar. The ‘perhaps’ shows that the author did not really know what a ‘Euripidean oath’ was, though his guess reflects negatively on both Sokrates and his peculiar oath by linking them to Euripides. Could the term have had anything to do with Hipp. 612?.

47. Plato's, Gorgias (Oxford, 1959)Google Scholar, ad loc. Cf. Hoerber, R., CJ 58 (1963), 268fGoogle Scholar.

48. Neither Plato nor Xenophon have Sokrates swear by the goose, or any of the ‘Rhadamanthine’ oaths supposedly used by him.

49. Hirzel, , op. cit., p. 102Google Scholar n. 21 cites passages which specifically emphasize Sokrates' respect for oaths: Plato, , Apol. 32bGoogle Scholar; Xen, . Mem. 1.1.18Google Scholar, Hell. 1.7.15.