¹ , Peek, Ath. Mitt. 57 (1932), p. 142Google Scholar; Hermes 68 (1933), p. 353Google Scholar; Ath. Mitt. 59 (1934), p. 252Google Scholar. , Bowra, C. Q. 32 (1938), p. 80Google Scholar. , Reinhardt, Hermes 73 (1938), p. 234Google Scholar. Reference to these articles is understood throughout. I am indebted to Professor A. D. Nock for many helpful criticisms and suggestions.

² The use of the vocative is too common to need illustration; the οἷον clause follows naturally as in Soph., Ajax 641 and Trach. 1112. The idea of endurance is, of course, expressed more than once in classical epigram and may be felt here in τελέσαντες: cf. , Wade-Gery, Classical Epigrams and Epitaphs, J. H. S. 53 (1933), p. 74Google Scholar (E, ταλακάρδιοι), p. 95 (V, οἵτινες ἔτλαν); Τλήμονες in the sense postulated here occurs in F (p. 75), where the idea of endurance is already expressed by μενεγχέας.

³ Eur., fragm. (Nauck) 100 and 550.

⁴ Liddell-Scott-Jones, s.v. κατὰ δαίμονα persists later; e.g. Parth, ., Erot. Path. IX, 1.Google Scholar

⁵ See Wilamowitz, on Eur., Her. 1228Google Scholar, quoted by Pearson on Soph., fragm. 650.

⁶ Lobeck, , Paralip. Gramm. Gr., p. 361Google Scholar; θεσπέσιος coincides in other usages with δαιμόνιος.

⁷ The same point arises with ἀγλαός; Wade-Gery, p. 80, n. 35. It is doubtful if much weight can be given to Peek's point that the position after the noun favors the adverb.

⁸ Bowra seems to be relying on the very inadequate section of Liddell-Scott-Jones when he rejects δαιμονίους on the ground that it does not seem to be used of human beings with a suggestion of “a rising above the mortal state.”

⁹ Aeneid VI, 660; cf. Norden, p. 34. Tyrtaeus IX, 31 ff. (Diehl); Heracl. 24 and perhaps 53 (Diels).

¹⁰ Cf. also more generally Plato, Politicus 309c: τὴν τῶν καλῶν καὶ δικαίων πέρι καὶ ἀγαθῶν καὶ τῶν τούτοις ἐναντίων ὄντως οὖσαν ἀληθῆ δόξαν μετὰ βεβαιώσεως, ὁπόταν ἐν ψυχαῖς ἐγγίγνηται, θείαν ϕημὶ ἐν δαιμονίῳ γίγνεσθαι γένει. The idea of the ψυχή as οἰκητήριον δαίμονος (Democr. 171) is rather different.

¹¹ Lysias, Epit. 80; Dem. LX, 33; Isocr., Pan. 84. Cf. , Aristeides, ὑπὲρ τῶν τεττάρων, p. 284 (δσιμονίους)Google Scholar.

¹² In view of the structure of the couplet, I do not regard the use of two adverbs as a serious objection to the reading adopted above.

¹³ The place of battle is omitted also in the ‘Tanagra’ epitaph; Wade-Gery, p. 79. Cf. Paus. I, 29, 4: ἑστᾶσιν ἐπὶ τοῖς τάϕοις στῆλαι τὰ ὀνόματα καὶ τὸν δῆμον ἑκάστου λέγουσαι.

¹⁴ Schwyzer, Dial. Gr. Ex. 17, 47.

¹⁵ Cf. Eur., Bacch. 1127:

οὐχ ὑπὸ σθένομς

ἀλλ᾽ ὁ θεὸς εὐμάρειαν ἐπεδίδου χεροῖν.

Cf. Virgil, Aen. V, 466: non viris alias conversaque numina sentis?

¹⁶ Plut., Nic. 17, quoted by Bowra, p. 86.

¹⁷ Kaibel, Ep. Gr. 529.

¹⁸ Il. XVI, 847 ff.; cf. Finsler, , Homer I², pp. 176 fGoogle Scholar.

¹⁹ Cf. VIII, 109, 3, for attribution of victory to the gods.

²⁰ Symp. 43; cf. de re equ. XI, 13 (ἢν μή δαιμόνιον κωλύῃ) and Oec. V, 13 (ἢν μὴ θεὸς ἀποκωλύῃ).

²¹ Dem., De Cor. 289, quoted by Bowra, p. 86.

²² I am unable to find a parallel for the use of ἡμίθεος here; the ἡμίθεοι are as a rule not a class of supernatural beings, they are “a species of men not of spirits or daimones” (Rohde, Psyche c. iv, n. 23). I have assumed that our writer is using the word as equivalent to ἥρως, though the heroes were not necessarily ἡμίθεοι; the terms seem to be identified in Plato, Crat. 398c: οὐκ οἶσθα ὅτι ἡμίθεοι οἱ ἤρωες; The identification may have been assisted by Hesiod, , Op., 159 ffGoogle Scholar. It is possible also that the word is being used as equivalent to δαίμων; see Plato, Ap., 27d–e with the significant play on ἡμίονος. I have used the word ‘god’ without prejudice to the interpretation of ἡμίθεος.

²³ Cf. Denniston on Eur., El. 1233–7: “When gods appear at the end of a tragedy, their divinity is always recognized at once by chorus or actor, their identity never.” It may be added that the identification of a hostile ‘hero,’ either immediately or afterwards, is less likely than that of a friendly and native one.

²⁴ Bowra chooses Orion on the ground that he “stood somewhat in the same relation to the Boeotians as Heracles did to the Peloponnesians,” but Heracles might well help the Thebans; see the amusing fable (Babrius 15) in which the Theban concludes a discussion with an Athenian on the merits of their national heroes with the words:

πέπανσο νικᾷς τοιγαροῦν χολωθείη

Θησεὺς μὲν ἡμῖν, Ἡρακλῆς δ᾽ Ἀθηναίοις.

²⁵ Ath. Mitt. 59 (1934), p. 253; he does not regard violation of the stoichedon arrangement as an objection to εἴσοδον (Bowra, p. 81), which he assumes might be a later orthography for ἔσοδον.

²⁶ Cf. th e fine distinction in Theocritus XXV, 10: ἱερὸν θείοιο παρὰ ῥόον Ἀλϕειοῖο.

²⁷ Cf. Kern, Rel. d. Gr. I, pp. 99 f.; hence the sacral term κελευθοποιοί (Pfister, Rel. d. Gr. u. R., Bursians Jahresb. 1930, p. 384).

²⁸ Θεῖος is, of course, more general in its application than θεός.

²⁹ For the abstract uses of εἴσοδος the following points may be added to the article in Liddell-Scott-Jones. (1) Entrance on magistracy, O. G. I. 458, 15. (2) Entrance of a spirit in demoniac possession, Aretaeus (Hude) III, 4, 2; for this sense ἔϕοδος is familiar from Hipp., περὶ ἱερ. νόσ. IV (ἡρώων ἐϕόδους). (3) For the doctor's visit, add Hipp., περὶ εὐσχημ. (Heiberg) 12, 13; Menander 1112 (C. A. F; III, 269). (4) For the legal sense, add Dem. XLII, 2. (5) In Aesch., Eumen. 30 it may have a semi-technical ritual sense.

³⁰ Cf. the use of πρόοδος in Julian, contra Christ. 200A (Neumann). Professor Nock points out that the use of εἴσοδος in Acts may be derived from Malachi 3, 2. See Suicerus, Thes. Eccles., s.v.

³¹ See also Jebb on Trach. 159 and Kaibel on El. 418 (δευτέραν ὀμιλίαν έλθόντος, of an apparition).

³² Cf. Pyth. V, 116 and Soph., El. 685, εἰσέρχεται.

³³ This use of eir εἰς ὀδόν is to be distinguished from that in which the idea of meeting is involved. The lion in the path is a familiar image in Homer (Il. XV, 271) and the gods like highwaymen often meet men in the way. This is perhaps the meaning in Ajax 36, rather than “entering on the route by which Ajax must pass” (Jebb). See Becker, O., Das Bild des Weges (Hermes, Einzelschriften 4), pp. 16, 46Google Scholar.

³⁴ We might add Pindar, Pyth. X, 29:

ναυσὶ δ᾽ οὔτε πεζὸς ἰών κεν εὔροις

ἐς Ὑπερβορέων ἀγῶνα θαυματὰν ὁδόν.

³⁵ Gundel, R.-E. VII, 1, col. 560 ff., s.v. Γαλαξίας. For the topic see also Dio Chrys. I, 65 (ἄγει λαβὼν αὐτὸν ἄϕραστον καὶ ἄβατου ἀνθρώποις ὁδόν) and Hermae Pastor, Vis. I, 1, 3 (καὶ πνεῦμά με ἔλαβεν καὶ ἀπήνεγκέν με δι᾽ ἀνοδίας τινός, δι᾽ ἦς ἄνθρωπος οὐκ ἑδύνατο ὁδεῦσαι). Cf. perhaps Hdwtb. des deutschen Abergl. I, 726 (Jesu Pfad).

³⁶ Pfister, R.-E. Suppl. IV, col. 280.

³⁷ This use of ἀντιάσας would agree well also with Bowra's view of θείαν εἰς ὁδόν; cf. Hdt. IX, 6 (ἠντίασαν ἐς τὴν Βοιωτίην) and I, 166, 2. Σύμβολοι and συναντήσεις are ἐνόδιοι (Aesch., P. V. 502) and a local reference is natural; cf. Apollod. Bibl. I, 9, 11 (περὶ τὸν Ἀλϕειὸν συντυχών) Diod. XV, 52, 3 (περὶ τὰς πύλας ἀπήντησε), Pindar, Pyth. VIII, 57 ff. In this matter the gods are like robbers (above, n. 33); see Sehulze, Kl. Schr., p. 167, n. 6, for ἐν ὁδῷ, and compare the similar use of ἀπαντάω, e.g. Dem. LIV, 37; Aristophanes plays on the similarity, Birds, 1490 ff.

³⁸ Od. XIV, 178 (τις ἀθανάτων); Trag. Adesp. 296, 455; Jebb on Antig. 622; cf. βλαψίϕρων, ϕρενοβλαβής, θεοβλαβής.

³⁹ Il. VII, 142; Aesch., P. V. 229; Aristotle quoted by Bowra, p. 88; Trag. Adesp. 540; Aristotle, Diehl, Anth. Lyr. I, p. 100.

⁴⁰ See Jebb ad loc.

⁴¹ Mor. 880b; cf. Eur., Cycl. 524. Cf. Scholia in Theocr. vet. (Wendel), p. 102 (πταρμοί).

⁴² , Kern, Orph. Fragm., p. 289Google Scholar (περὶ ἐπεμβ.), 11. 5, 20, 23, 32.

⁴³ Artemid. Onir. II, 36. The use of βλάπτειν dealt with by Liddell-Scott-Jones, s.v., III, 2, may be related to that in Eur., Hel. 868, where it means to mar or pollute in a ritual sense.

⁴⁴ Rep. 364b.

⁴⁵ For example, Xen. Symp. 43 (quoted above); Soph., El. 696 (τις θεῶν); Ajax 456 (τις θεῶν).

⁴⁶ Cf. note 20 above.

⁴⁷ Cf. Becker, O., op. cit., p. 8, n. 6Google Scholar; p. 159, n. 22.

⁴⁸ The exact length of the lacuna appears to be indeterminable and the only certain point is that there is room for one more letter in line 5 than in line 6. I cannot see in the photograph any trace of the Δ which Peek shows after the lacuna in line 5.

⁴⁹ We miss something like the τὸ πρῶτον with ϕιλόϕρων in Persae 111.

⁵⁰ Heracl. 405; delet Wilamowitz.

⁵¹ Professor Nock suggests a number of passages in Diod. Sic. where λόγιον may be equivalent to χρησμός. For example, IV, 73, 5; V, 54, 4: κατά τι λόγιον (perhaps a Delphic response; cf. V, 59, 4: κατά τινα χρησμόν); VII, 3, 4. The words are synonymously of a prophecy παραδεδομένον ἐκ προγόνων (II, 26, 9; 27, 2).

⁵² Ἄγρη and ἐκθηρεύειν are used metaphorically by Herodotus (IX, 39, 2; 90, 2; VI, 31, 2), but the symbolism is specially appropriate to religious language. Cf. ἀγρεῖ in Agam. 126, used to recall the ἄγρα of the eagles; θηραθεῖσαι in P. V. 1072; θηρῶσιν τὸν ἄσεπτον, Eur., Bacch. 890; θήρενσε, Pindar, Pyth. IV, 90; εὔαγρος in prayer, Soph., O. C. 1089; the symbolism of the net belongs to the same sphere, P. V. 1078; Persae 113. If, as is possible, one particular λόγιον was referred to, it may have been similar in its symbolism to that parodied in Knights 197 ff. (cf. Neil, ad loc). Δύσμαχος ἄγρα suggests the serpent (Antig. 126, δυσχείρωμα δράκοντος; Aesch., Ch. 246) or a bird of prey attacked by another (Persae, 205 ff.), and θηρεύειν seems to have been technical of the eagle as a bird of omen. Homer twice mentions it as θηρητήρ (Il. XXI, 252; XXIV, 315); cf. perhaps Artemid. Onir. III, 64, 191 and IV, 56. In Xen. Anab. VI, 1, 23, λαμβάνειν τὰ ἐπιτήδεια is equivalent to θηρεύειν. When it was desired to take the omens without waiting for casual observation, pieces of meat seem to have been cast to birds as a kind of symbolic ἅγρα (Paus. IX, 3, 4; Apollod. I, 9, 12). Our poet clearly drew his language from this religious sphere.

⁵³ Op. 665; cf. Wilamowitz ad loc.

⁵⁴ Πρόϕρων is common in prayer, Soph., El. 1380 (γενοῦ πρόϕρων ἡμῖν ἀρωγός); Aristoph., Birds 930; Ariphron, Diehl II, p. 130. It is correspondingly appropriate of a friendly epiphany, as in Apoll. Rhod. IV, 919.

⁵⁵ Other restorations are possible, but in default of anything more convincing I take Fraenkel's reading as an approximation to the original. Reinhardt's τοισίδε is awkward. Apart from Peek's πέϕραδε, which we have rejected, the only verb which readily suggests itself is ἴδε; it might be possible to read πρόϕρων γὰρ ὅπως ἴδε (that is, “when he saw that the prey was too difficult for the enemy to take”), if a suitable object could be found for ἐξετέλεσσε. The general interpretation of the poem proposed in this paper does not depend on any particular restoration of these lines.

⁵⁶ Cf. Agam. 178 (θέντα).

⁵⁷ Cf., for example, Hdt. I, 159, 3; 11, 2; 189, 2; VII, 237, 3; VIII, 143, 2.

⁵⁸ For secularized use in verse see Philiscus (Diehl I, p. 97, 1. 7) and I.G. II, 2, 1078 (public notice).

⁵⁹ Ovid, Ex Ponto I, 1, 55:

talia caelestes fieri praeconia gaudent

ut sua quid valeant numina teste probent.

⁶⁰ As far as I have noted, Pettazoni, La confessione dei peccati, does not treat the passage.

⁶¹ III, 6 (ἀγγέλλειν πᾶσιν); 298 (κατὰ πᾶσαν γαῖαν); 811 (πᾶσι θνητοῖς); IV, 19; VIII, 3 (πᾶσι ἀνθρώποις). Similarly in the morality of fable, Babrius 22, 96.

⁶² The connection of this phraseology with the missionary idea is noted by , Pfister, op. cit., p. 346 (cf. 150, 212)Google Scholar; cf. Björck, G., Der Fluch des Christen Sabinus, pp. 65 fGoogle Scholar. The element of publicity and propaganda is a regular element in miracle report; see for example the references to the presence of the ‘congregation’ in the miracle records of the Tiber island, S.I.G. III, 1173 (δημοσίᾳ etc.). Cf. Kerényi, Die Griechisch-Orientalische Roman literatur, pp. 9 f. with notes 39 to 42.

⁶³ Becker, O., op. cit., p. 205, n. 23Google Scholar, cites Soph., Ajax 648; Antig. 388; Eur., fragm. 761. Add perhaps Eur., Ion 1510. I am unable to consult Ernest Dutoit, Le Thème de l'Adynaton dans la poésie antique. The power to perform what is ἀδύνατον is an attribute of the gods. Thus the Meter Leto ἐξ ἀδυνάτων δυνατά πυεῖ (Steinleitner, op. cit., p. 59). It should be noted that this typical antithetical form of expression is reflected in the contrast between ἄελπτον and ἐπίελπτα in Archilochus. See this journal, vol. XXXII (1939), p. 8, with notes 40 and 41; add also Cleanthes (Powell, Coll. Alex. p. 227, 1. 18) and Aristophanes, Lys. 772 (τὰ δ᾽ ὑπέρτερα νέρτερα θήσει Ζεὺς ὑψιβρεμέτης).

⁶⁴ , Kerényi, op. cit., p. 63, n. 76 (cf. p. 104, n. 39)Google Scholar.

⁶⁵ See, for example, Isocr. πρὸς Νικ. 23a (παράδοξον, ἅπιστον); Plut., Camillus 3 (ἀπίστων θαυμάτων); Palaephatus (Festa), p. 2 (ἀπιστότερον, θαυμασιώτερον); p. 11 (ἅπιστος, ἀδύνατος); p. 63 (ἀπίθανον); Pseudo-Eratosthenes, Catasterismi (Olivieri), pp. 22, 34 (ἀπίθανον). On παράδοξον see , Weinreieh, op. cit., pp. 198 fGoogle Scholar. and , Kerényi, op. cit., p. 12, n. 55 and p. 18, n. 93Google Scholar. For miracle literature in general, see , Nock, Conversion, pp. 90 f. and p. 287Google Scholar.

⁶⁶ Reflected in literature of the ἰαμβικὴ ἰδέα in the ψόγος of seers and prophets; see F. J. Brecht, Motiv- und Typengesch. des gr. Spottepigr. (Philol. Suppl. XXII, 2), pp. 41 ff.

⁶⁷ , Kerényi, op. cit., p. 2Google Scholar.

⁶⁸ I. G. IV, 951, 23 ff.; cf. 951, 34: τῶν ἰαμάτων διεγέλα, ὡς ἀπίθανα καὶ ἀδύνατα ἐόντα χωλοὺς καὶ τυϕλοὺς ὑγιεῖς γίνεσθαι ἐνύπνιον ἰδόντας μόνον.

⁶⁹ Φράζεσθαι is in keeping with the style; see Hendess, Oracula Graeca, 64, 69b, 108, 131.

⁷⁰ Dornseiff, , Literarische Verwendungen des Beispiels, Vorträge der Bibl. Warburg, 1924-1925, pp. 206 ffGoogle Scholar. Nock, , Conversion, p. 233Google Scholar.

⁷¹ , Dornseiff, op. cit., has treated this topic in an illuminating fashion in relation both to literary form and to the history of ideas. For the πρόνοια literature seeGoogle Scholar, Weinreich, op. cit., pp. 133 ff.Google Scholar; cf. the collocation of παράδοξον and πρόνοια in Diod. Sic. XVII, 103.

⁷² Op. cit., p. 233.

⁷³ Greek Art and Literature, pp. 97 f.; pp. 121 f.

⁷⁴ , Webster, op. cit., pp. 127 and 148Google Scholar.

⁷⁵ Cf. on this aspect, Heliodorus IX, 24 (, Kerényi, op. cit., p. 21Google Scholar): τὰ μεγάλα τῶν πραγμάτων μεγάλων δεῖται κατασκευῶν.

⁷⁶ Cf. Eur., fragm. 62:

Ἑκάβη, τὸ θεῖον ὡς ἄελπτον ἔρχεται

θνητοῖσιν.

For the popular view note the significant passage in Aristophanes, Wasps 733:

σοὶ δὲ νῦν τις θεῶν παρὼν ἐμϕανὴς

ξυλλαμβάνει τοῦ πράγματος.

There is, of course, no question of a real epiphany. See too the almost proverbial phrase in Wasps 799: ὅρα τὸ χρῆμα᾽ τὰ λόγια ὡς περαίνεται. The belief is reflected in the use of this motive in folk-tale (, Rose, Handbook of Greek Mythology, p. 294)Google Scholar.

⁷⁷ At the end of Alcestis, Andromache, Bacchae, Helena, Medea (varied). This passage and that which follows are brought into relation by Mueller, E., De Graecorum Deorum partibus tragicis, R. V. V., 1910, p. 124Google Scholar.

⁷⁸ Schol. in Luc., ed. Rabe, H. (Lipsiae, 1906), p. 164, 2 ffGoogle Scholar.

⁷⁹ Apul. XI, 2; , Kerényi, op. cit., p. 23, n. 112Google Scholar.

⁸⁰ Thuc., VII, 77, 3.

⁸¹ , Webster, op. cit., pp. 53 fGoogle Scholar.

⁸² , Kerényi, op. cit., p. 5, n. 23Google Scholar.

⁸³ Vita Aesch. 9.

⁸⁴ In ‘ecplectic’ effect the supernatural has yielded place in comparatively recent times to crime.

⁸⁵ This is not the place to enter upon a discussion of the obscure topic of Peloponnesian lament (Page, , Greek Poetry and Life, pp. 206–217Google Scholar; Bowra, p. 86). I do not attach any importance to mention of defeat as a significant element in the poem. The direct address, on which Bowra lays some stress, is appropriate to epitaph; it occurs in Homer in the three laments for Hector (Il. XXIV, 725, 748, 762) and persists into modern Greek (, Passow, Popularia Carmina, pp. 257 ff.)Google Scholar. It occurs early in epitaph (, Schwyzer, op. cit., 165Google Scholar) and in three Simonidean epigrams (Diehl, 79, 101, 117; cf. Wilhelm, Öst. Jahresh. II, 1899, p. 225). Other influences may have been at work in furthering its use; for example, the style of the hymn (Meyer, H., Hymn. Stilel. in d. frühgr. Dicht., pp. 69 ff.Google Scholar; he notes it at the beginning of the Καθαρμοί of Empedocles) and the method of varying narrative and increasing emotional effect by direct address in lyric. It may be noted that the Skolia offer some parallels in this feature and others to our poem. They deal on occasion with defeat and death and have the elements of direct address (Diehl, 11, 13, 15, 24), short narrative (5, 12, 24) and γνώμη (7, 9, 14, 20 ϕράζευ, 25, 29, 32 ff.). For the formal connection of narrative and γνώμη in this branch of literature, see Wüst, E., Philologus 77, 1921, pp. 36 fGoogle Scholar.

⁸⁶ To illustrate the situation which we assume, reference may be made to the accounts of the disaster of Drabescus. The Athenians were massacred by the Edonoi (ἀνέλπιστοι ἐπιθέμενοι) and according to the tradition in Pausanias (I, 29, 4) there was a supernatural intervention in the battle: λέγεται δὲ καὶ ὡς κεραυνοὶ πέσοιεν ἐπ᾽ αὐτούς. The victims were the first to receive a public tomb κατὰ τὴν ὁδὸν τὴν εἰς Ἀκαδημίαν. From the scholia to Aeschines, de F. L. 34, we learn further of a belief that the Athenians were doomed to suffer disaster nine times at Ennea Hodoi by the curse of Phyllis, the deserted lover of the Athenian Demophon. It is possible that these traditions go back to a contemporary source and that the curse of Phyllis was represented by λόγια foretelling disaster to expeditions to Thrace. There is no need, of course, to resort to the hypothesis of a later commemoration in our text of men who had fallen earlier in Thrace; the motive, as we have seen is a recurrent one. Professor Nock adds a reference to Soph., O. C. 92; according to tradition Attica was protected by the grave of Oedipus (Jebb, pp. xxviii f.).

⁸⁷ Though religious propaganda was no doubt employed by all parties, a case might be made out for maintaining that it was directed rather to the lower classes. Radermacher (Euripides, und Mantik, die, Rh. Mus. 53, 1898, pp. 497 ff.Google Scholar) has argued (p. 504) that in the Peloponnesian war it was used by the democrats more than by their opponents. See also Staehelin, R., Das Motiv der Mantik im antiken Drama, R. V. V. XII, 1912–1913, pp. 1 ffGoogle Scholar.