Hostname: page-component-cd9895bd7-q99xh Total loading time: 0 Render date: 2024-12-25T07:18:44.512Z Has data issue: false hasContentIssue false
  • English
  • Français

Theurgy and its Relationship to Neoplatonism1

Published online by Cambridge University Press:  24 September 2012

E. R. Dodds
Get access Rights & Permissions [Opens in a new window]

Extract

The last half-century has seen a remarkable advance in our knowledge of the magical beliefs and practices of later antiquity. But in comparison with this general progress the special branch of magic known as theurgy has been relatively neglected and is still imperfectly understood. The first step towards understanding it was taken more than fifty years ago by Wilhelm Kroll, when he collected and discussed the fragments of the Chaldaean Oracles. Since then the late Professor Joseph Bidez has disinterred and explained a number of interesting Byzantine texts, mainly from Psellus, which appear to derive from Proclus' lost commentary on the Chaldaean Oracles, perhaps through the work of Proclus' Christian opponent, Procopius of Gaza; and Hopfner and Eitrem have made valuable contributions, especially in calling attention to the many common features linking theurgy with the Greco-Egyptian magic of the papyri. But much is still obscure, and is likely to remain so until the scattered texts bearing on theurgy have been collected and studied as a whole (a task which Bidez seems to have contemplated, but left unaccomplished at his death). The present paper does not aim at completeness, still less at finality, but only at (i) clarifying the relationship between Neoplatonism and theurgy in their historical development, and (ii) examining the actual modus operandi in what seem to have been the two main branches of theurgy.

Type
Papers Presented to N. H. Baynes
Information
The Journal of Roman Studies , Volume 37 , Issue 1-2 , November 1947 , pp. 55 - 69
Copyright
Copyright © E. R. Dodds 1947. Exclusive Licence to Publish: The Society for the Promotion of Roman Studies

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.)

Footnotes

1

I must express my gratitude to Professors M. P. Nilsson and A. D. Nock, who read this paper in manuscript and contributed valuable suggestions.

References

2 Kroll, W., de Oraculis Chaldaicis (Breslauer Philologische Abhandlungen VII, I, 1894)Google Scholar.

3 Catalogue des manuscrits alchimiques grecs (abbrev. CMAG), vol. VI; Mélanges Cumont 95 ff. Cf. his ‘Note sur les mystères neoplatoniciens’ in Rev. Belge de Phil. et d'Hist. 7 (1928), 1477 ffCrossRefGoogle Scholar. and his Vie de l'Emp. Julien 73 ff. On Procopius of Gaza as Psellus' proximate source, see Westerink, L. G. in Mnemosyne 10 (1942), 275 ffGoogle Scholar.

4 Griechisch-Aegyptische Offenbarungszauber (quoted as OZ); and in the introduction and commentary to his translation of the de mysteriis. Cf. also his articles ‘Mageia’ and ‘Theurgie’ in Pauly-Wissowa, and below, n. 115.

5 Especially ‘Die σύστασις und der Lichtzauber in der Magie’, Symb. Oslo. 8 (1929), 49 ff.Google Scholar; and ‘La Théurgie chez les Néo-Platoniciens et dans les papyrus magiques’, ibid., 22 (1942), 49 ff. I am indebted to the author's kindness in sending me a copy of the latter important paper. Theiler's, W. essay, Die chaldaischen Orakel und die Hymnen des Synesios (Halle, 1942Google Scholar), which reached this country too late for me to use it, deals learnedly with the influence of the Oracles on later Neoplatonism.

6 Papyri Graecae Magicae, ed. Preisendanz (abbrev. PGM).

7 Cf. Bidez-Cumont, , Les Mages Hellénisés I, 163Google Scholar.

8 τοῦ κληθέντος θεουργοῦ Ἰουλιανοῦ, Suidas s.v.

9 Suidas s.v., cf. Proclus, in Crat. 72, 10Google Scholar, Pasq., in Remp. II, 123, 12, etc. Psellus in one place (confusing him with his father?) puts him in Trajan's, time (Scripta Minora I, p. 241, 29Google Scholar, Kurtz-Drexl).

10 Vie de Julien 369, n. 8.

11 See Eitrem, , Symb. Oslo. 22, 49CrossRefGoogle Scholar. Psellus seems to have understood the word in the latter sense, PG 122, 721 D: θεοὺς τοὺς ἀνθρώπους ἐργάзεται. Cf. also the Hermetic ‘deorum fictor est homo’, quoted below, p. 63.

12 Proclus' expression οἱ ἐπὶ Μάρκου θεουργοί (in Crat. 72, 10, in Remp. II, 123, 12) perhaps refers to father and son jointly.

13 ad Philops. 12 (IV, 224 Jacobitz). On this scholion, see Westerink, o.c. 276.

14 Scripta Minora I, p. 241, 25 ff., cf. CMAG VI, p. 163, 19 ff. As Westerink points out, the source of these statements seems to be Procopius.

15 Marinus, vit. Procl., 26, cf. Procl. in Crat., c. 122. On such claims of divine origin, which are frequent in Hellenistic occult literature, see Festugière, , La Révélation d'Hermès Trismégiste I, 309 ffGoogle Scholar.

16 Bousset, , Arch. f. Rel. 18 (1915), 144Google Scholar, argued for an earlier date on the ground of coincidences in doctrine with Cornelius Labeo. But Labeo's own date is far from certain; and the coincidences may mean merely that the Juliani moved in Neopythagorean circles, which we know to have been interested in magic.

17 Scripta Min. I, 241, 29; cf. CMAG VI, p. 163, 20. On doctrinal oracles received in vision, see Festugière, o.c, I, 59 f.

18 Professional poets were kept for this purpose at Delphi (Strabo, 9, 3, 5; Plut. Pyth. Orac. 25, 407 B); and at Didyma the existence of a Χρησμογράφιον (Rev. de Phil. 44 [1920], 249, 251) suggests similar arrangements.

19 Kroll, o. c., 53 ff. The passages about the divine fire recall the ‘recipe for immortality’ in PGM IV, 475 ff., which is in many ways the closest analogue to the Chaldaean Oracles. Julian, , Or. V, 172Google Scholar D, attributes to Χαλδαῖος (i.e. Julianus) a cult of τὸν ἑπτάκτινα θεόν. This solar title has been disguised by corruption in two passages of Psellus, : Script. Min. I, 262, 19Google Scholar: Ἐρωτύχην ἤ Κασόθαν ἢ Ἔπτακις (read Ἑπτάκτις), ἤ εἴ τις ἄλλος δαίμων ἀπατηλός: ibid. I, 446, 26: τὸν Ἔπακτον (Ἑπτάκτιν Bidez) ὁ Ἀπουλήιος ὅρκοις καταναγκάσας μὴ προσομιλῆσαι τῷ θεουργῷ (sc. Juliano). Cf. also Procl. in Tim. I, 34, 20: Ἡλίῳ, παρ᾿ ᾦ…ὁ Ἑπτάκτις κατὰ τοὺς θεολόγους.

20 Περὶ τῆς χρυσῆς ἁλύσεως, Ann. Assoc. Ét. Gr. 1875, p. 216, 24 ff.

21 Proclus, in Tim. III, 120, 22Google Scholar: οἱ θεουργοί…ἀγωγὴν αὐτοῦ παρέδοσαν ἡμῖν, δι΄ ἧς εἰς αὐτοφάνειαν κινεῖν αὐτὸν δυνατόν: cf. Simpl. in Phys. 795, 4Google Scholar, and Damasc. Princ. II, 235, 22Google Scholar. Both σύστασις and ἀγωγή are ‘terms of art’, familiar to us from the magical papyri.

22 Proclus, in Remp. II, 123Google Scholar, 9 ff.

23 Suidas s.v. Ίουλιανός. The ascription of the credit to Julianus is perhaps implied also in Claudian, de vi cons. Honorii, 348 f., who speaks of ‘Chaldaean’ magic. For other versions of the tale, and a summary of the lengthy modern discussions, see Cook, A. B., Zeus III, 324Google Scholar ff. The attribution to Julianus may have been suggested by a confusion with the Julianus who commanded against the Dacians under Domitian (Dio Cass. 67, 10).

24 Scripta Minora I, 446, 28.

25 S. Anastasius of Sinai, Quaestiones (PG 89, col. 525 A). For Julianus' supposed rivalry with Apuleius, see also Psellus quoted above, n. 19.

26 Cf. Olympiodorus in Phaed. 123, 3 Norvin: οἱ μὲν τὴν φιλοσοφίαν προτιμῶσιν, ὡς Πορφύριος καὶ Πλωτῖνος καὶ ἄλλοι πολλοὶ φιλόσοφοι οἱ δὲ τὴν ἱερατεκήν (i.e. theurgy), ὡς Ἰάμβλιχος καὶ Συριανὸς καὶ Πρόκλος καὶ οἱ ἱερατικοὶ πάντες.

26a The prose injunction, μὴ ἐξάξῃς ἵνα μὴ ἐξίῃ ἔχουσά τι, which he quotes at Enn. I, 9 init., is called ‘Chaldaean’ by Psellus (Expos. Or. Chald. 1125C ff.) and in a late scholion ad loc., but cannot come from a hexameter poem. The doctrine is Pythagorean.

27 Porph. uit. Plot. 16. Cf. Kroll, , Rh. Mus. 71 (1916), 350Google Scholar; Puech in Mélanges Cumont 935 ff. In a similar list of bogus prophets, Arnob. adv. gentes I, 52Google Scholar, Julianus and Zoroaster figure side by side.

28 Cf. esp. c. 9, I 197, 8 ff. Volk.: τοὶς δ᾿ ἄλλοις (δεῖ) νομίзειν εἶναι χώραν παρὰ τῷ θεῷ καὶ μὴ αὑτὸν μόνον μετ᾿ ἐκεῖνον τάξαντα ὥσπερ ὀνείρασι πέτεσθαι…τὸ δὲ ὑπὲρ νοῦν ἤδη ἐστὶν ἔξω νοῦ πεσεῖν.

29 Enn. 4, 4, 37, 40. Observe that throughout this discussion he uses the contemptuous word γοητεία and introduces none of the theurgie terms of art. On the Stoic and Neoplatonic conception of συμπάθεια, see K. Reinhardt, Kosmos und Sympathie, and my remarks in Greek Poetry and Life 373 f. To theurgists such explanations appeared entirely inadequate (de myst. 164, 5 ff., Parthey).

30 Symb. Oslo. 22, 50. As Eitrem himself notes, Lobeck and Wilamowitz thought otherwise; and he might have added the names of Wilhelm Kroll (Rh. Mus. 71 [1916], 313), and Joseph Bidez (Vie de julien 67; CAH XII, 635 ff.).

31 See on this CQ 22 (1928), 129, n. 2.

32 Cochez, J., Rev. Néo-Scolastique 18 (1911), 328 ff.CrossRefGoogle Scholar, and Mélanges Ch. Moeller 1, 85 ff.; Cumont, , Mon. Piot 25, 77 ffCrossRefGoogle Scholar.

33 de abst. 4, 6, cf. de myst. 265, 16, 277, 4. See further E. Peterson's convincing reply to Cumont, , Theol. Literaturzeitung 50 (1925), 485 ffGoogle Scholar. I would add that the allusion in Enn. 5, 5, 11 to people who are excluded from certain ἱερά because of their γαστιυαργία probably refers to Eleusis,not Egypt: παραγγέλλεται γὰρ καὶ Ἐλευσῖνι ἀπέχεσθαι κατοικιδίων ὀρνίθων καὶ ἰχθύων καὶ κυάμων ῥοιᾶς τε καὶ μήλων, Porph. de abst. 4, 16Google Scholar.

34 Cf. CQ 22 (1928), 141 f., and Peterson, E., Philol. 88 (1933), 30 ffCrossRefGoogle Scholar. Conversely, as Eitrem has rightly pointed out (Symb. Oslo. 8, 50), the magical and theurgic term σύστασις has nothing to do with unio mystica.

35 See Bidez's sympathetic, elegant, and scholarly study, La Vie du Néoplatonicien Porphyre; and on the reasons for the decay of rationalism at this period, Festugière, op. cit., Introduction, ch. 1.

36 νέος δὲ ὤν ἴσως ταῦτα ἔγραφεν, ὡς ἔοικεν, Eun. vit. soph. 457, Boissonade; Bidez, op. cit., ch. III.

37 The fragments were edited by W. Wolff, Porphyrii de Philosophia ex Oraculis Haurienda (1856). On the general character of this collection, see Nock, A. D., ‘Oracles Theologiques,’ REA, 30 (1928), [280CrossRefGoogle Scholar ff.

38 The fragments as reconstructed (not very scientifically) by Gale are reprinted in Parthey's, edition of the de mysteriis. On the date, see Bidez, oc., 86.

39 apud Eus. Praep. Ev. 5, 10, 199 A ( = fr. 4 Gale): μάταιοι αἱ θεῶν κλήσεις ἔσονται…καὶ ἔτι μᾶλλον αἱ λεγόμεναι ἀνάγκαι θεῶν ἀκήλητον γὰρ καὶ ἀβίαστον καὶ ἀκατανάγκαστον τὸ ἀπαθές.

40 It is probable that the letter to Anebo did not quote Julianus or the Chaldaean Oracles, since Iamblichus' reply is silent about them. Whether the ‘theurgy’ of the de mysteriis is in fact independent of the Julianic tradition remains to be investigated. The writer certainly claims to be acquainted with the ‘Chaldaean’ (p. 4, 11) or ‘Assyrian’ (p. 5, 8) doctrines as well as the Egyptian, and says he will present both.

41 Marinus, vit. Procli 26; Lydus, mens. 4, 53; Suidas s.v. Πορφύριος.

42 Aug. Civ. Dei 10, 32 = de regressu fr. 1 Bidez (Vie de Porphyre, Appendix 11).

43 ibid., 10, 9 = fr. 2 Bidez. On the function of the πευματική Ψυχή in theurgy, see my edition of Proclus' Elements of Theology, p. 319.

44 Cf. Olympiodorus' judgment, supra, n. 26.

45 Julian, Epist. 12 Bidez; Marinus, vit. Procli 26; Damasc. 1, 86, 3 ff.

46 The de mysteriis, though issued under the name of ‘Abammon’, was attributed to Iamblichus by Proclus and Damascius; and since the publication of Rasche's dissertation in 1911 most scholars have accepted the ascription. Cf. Bidez in Mélanges Desrousseaux 11 ff.

47 Epist. 12 Bidez = 71 Hertlein = 2 Wright. The Loeb editor is clearly wrong in maintaining against Bidez that τὸν ὁμώνυμον in this passage means Iamblichus the younger: τὰ Ἰαμβλίχου εἰς τὸν ὁμώνυμον cannot mean ‘the writings of Iamblichus to his namesake‘; nor was the younger Iamblichus θεόσοφος.

48 Cf. what Eunapius says of one Antoninus, who died shortly before 391: ἐπεδείκνυτο οὐδὲν θεουργὸν καὶ παράλογον ἐς τὴν φαινομένην αἴσθησιν, τὰς βασιλικὰς ἴσως ὁρμὰς ὑφορώμενος ἑτέρωσε φερούσας (p. 471).

49 Thus Proclus learned from Asclepigeneia the θεουργικὴ ἀγωγή of ‘the great Nestorius’, of which she was, through her father Plutarchus, the sole inheritress (Marinus, vit. Procli 28). On this family transmission of magical secrets see Dieterich, Abraxas 160 ff.; Festugière, , La Révélation d'Hermès Trismégiste I, 332 ffGoogle Scholar. Diodorus calls it a Chaldaean practice, 2, 29, 4.

50 Marinus, vit. Procli 26, 28. The Περὶ is listed by Suidas s.v. Πρόκλος.

51 Scripta Minora I, 237 f.

52 Migne, , PG 149, 538Google Scholar B ff., 599 B; cf. Bidez, , CMAG VI, 104 f.Google Scholar, Westerink, o.c, 280.

53 Nauck's correction for φησιν, which has no possible subject.

54 Among later writers, Proclus (in Alc., p. 73, 4 Creuzer) and Ammianus Marcellinus (21, 14, 5) refer to the incident. But Proclus, who says ὁ Αἰγύπτιος τὸν Πλωτῖνον ἐθαύμασεν ὡς θεῖον ἔχοντα τὸν δαίμονα, is clearly dependent on Porphyry; and so, presumably, is Ammianus, whether directly or through a doxographic source.

55 Cf. J. Kroll, Lehren des Hermes Trismegistos 88 f.; Wilamowitz, Menanders Schiedsgericht 112 f.; Farquharson on M. Ant. 2, 13. Ammianus, l.c., says that while each man has his ‘genius’, such beings are ‘admodum paucissimis visa’.

56 Since the surviving part of the recipe is an invocation to the sun, Preisendanz and Hopfner think that ἰδίον is a mistake for ἡλίον. But loss of the remainder of the recipe (Eitrem) seems an equally possible explanation. On such losses, see Nock, , J. Eg. Arch. 15 (1929), 221Google Scholar. The ἴδιος δαίμων seems to have played a part in alchemy also; cf. Zosimus, Comm. in ω 2 (Scott, , Hermetica IV, 104Google Scholar).

57 e.g. PGM IV, 1927. Similarly IV, 28 requires a spot recently bared by the Nile flood and still untrodden, and II, 147, a τόπος ἁγνὸς ἀπὸ παντὸς μυσαροῦ. So Thessalus, , CCAG 8 (3), 136, 26Google Scholar (οἰκος καθαρός).

58 apud Porph. de abst. 4, 6Google Scholar (236, 21 Nauck). He goes on to speak of ἁγνευτήρια τοῖς μὴ καθαρεύουσιν ἄδυτα καὶ πρὸς ἱερουργίας ἅγια (237, 13). On magical practices in Egyptian temples see Cumont, L'Égypte des Astrologues 163 ff.

59 E.g. PGM IV, 814 ff. For φυλακή, cf. Proclus, in CMAG VI, 151Google Scholar, 6: ἀπόχρη γὰρ πρός … φυλακὴν δάφνη, ῥάμνος, σκύλλα, κτλ.; and for spirits turning nasty at séances, Pythagoras of Rhodes in Eus. Praep. Ev. 5, 8, 193BGoogle Scholar; Psellus, , Op. Daem. 22, 869BGoogle Scholar.

60 Aspersion with blood of a dove occurs in an ἀπόλυσις, PGM II, 178.

61 fr. 29 = de myst. 241, 4P. = Eus. Praep. Ev. 5, 10, 198AGoogle Scholar.

62 CRAI 1942, 284 ff. Doubt may be felt about the late date which Cumont assigns to the introduction of domestic fowl into Greece; but this does not affect the present argument.

63 ‘The cock has been created to combat demons and sorcerers along with the dog’, Darmesteter (quoted by Cumont, l.c.). The belief in its apotropaic virtues survives to this day in many countries. A cock-headed δαίμων is one of the commonest figures on Greco-Egyptian amulets.

64 Is. et Os. 46, 369F.

65 CMAG VI, 150, 1 ff., 15 ff. (partly based on the traditional antipathy of lion and cock, Pliny, , NH 8, 52Google Scholar, etc.). Cf. Bolos, Φυσικά fr. 9 Wellmann (Abh. Berl. Akad., Phil.-Hist. Kl., 1928, Nr. 7, p. 20).

66 Very similar ideas appear in the ‘recipe for immortality’, PGM IV, 475 ff., e.g. 511: Ἵνα θαυμάσω τὸ ἱερὸν πῦρ and 648: ἐκ τοσούτων μυριάδων ἀπαθανατισθεὶς ἐν ταύτῃ τῇ ὥρᾳ. It, too, culminates in luminous visions (634 ff., 694 ff.). But the theurgic ἀπαθανατισμός may have been connected with a ritual of burial and rebirth, Procl. Theol. Plat. 4, 9, p. 193Google Scholar: τῶν θεουργῶν θάπτειν τὸ σῶμα κελευόντων πλὴν τῆς ἐν τῇ μυστικωτάτῃ τῶν τελετῶν (cf. Dieterich, Eine Mithrasliturgie 163).

67 Psellus, though he, too, connects τελεστική with statues, explains the term otherwise: τελεστικὴ δὲ ἐπιστήμη ἐστὶν ἡ οἶον τελοῦσα (so MSS.) τὴν ψυχὴν διὰ τῆς τῶν ἐνταῦθ᾿ ὑλῶν δυνάμεως (Expos. Or. Chald. 1129D, in PG vol. 122). Hierocles, who represents a different tradition, makes τελεστική the art of purifying the pneuma (in aur. carm. 482 A Mullach).

68 Psellus says that ‘the Chaldaeans’ διαφόροις ὕλαις ἀνδρείκελα πλάττοντες ἀποτρόπαια νοσημάτων ἐργάзονται (Script. Min. I, 447, 8). For σύμβολα, cf. the line quoted by Proclus, , in. Crat. 21, 1Google Scholar: σύμβολα γὰρ πατρικὸς νόος ἔσπειρεν κατὰ κόσμον.

69 Epist. 187 Sathas, (Bibliotheca Graeca Medii Aevi v, p. 474Google Scholar).

70 CMAG VI, 151, 6; cf. also in Tim. I, III, 9 ff.

71 Cf. Proclus, in CMAG VI, 148 ff.Google Scholar, with Bidez' introduction, and Hopfner, , OZ I, 382 ffGoogle Scholar.

72 An identical practice is found in modern Tibet, where statues are consecrated by inserting in their hollow interiors written spells and other magically potent objects (Hastings, , Encycl. of Religion and Ethics VII, 144, 160Google Scholar).

73 Cf. R. Wünsch, Sethianische Verfluchungstafeln 98 f.; A. Audollent, Defixionum Tabellae, p. LXXIII; Dornseiff, Das Alphabet in Mystik u. Magie 35 ff.

74 Proclus, in Tim. II, 247, 25Google Scholar; cf. in Crat. 31, 27. Porphyry, too, included in his list of theurgic materia magica both ‘figurationes’ and ‘soni certi quidam ac voces’ (Aug. Civ. Dei 10, 11).

75 Marinus, vit. Procl. 28; Suidas, s.v. χαλδαῖκοῖς ἐπιτηδεύμασι. Cf. Psellus, Epist. 187, where we learn that certain formulae are inoperative εἰ μή τις ταῦτα ἐρεῖ ὑποψέλλῳ τῇ γλώσσῃ ἤ ἑτέρως ὡς ἡ τέχνη διατάττεται.

76 Psellus, in CMAG VI, 62, 4Google Scholar, tells us that Proclus advised invoking Artemis ( = Hecate) as ξιφηφόρος, σπειροδρακοντόзωνος, λεοντοῦχος, τρίμορφος τούτοις γὰρ αὐτήν φησι τοῖς ὀνόμασιν ἕλκεσθαι καὶ οἷον ἐξαπατᾶσθαι καὶ γοητεύεσθαι.

77 Proclus, in Crat. 72, 8Google Scholar. Cf. the divine name which ‘the prophet Bitys’ found carved in hieroglyphs in a temple at Sais and revealed to ‘King Ammon’, de myst. 267, 14.

78 Psellus, expos. or. chald. 1132 C; Nicephoros Gregoras, in Synes. de insomn. 541 A. Cf. Corp. Herm. XVI, 2.

79 Cf. the Greek translations of such magical names given by Clem. Alex. Strom. 5, 242Google Scholar, and Hesych. s.v. Ἐφέσια γράμματα.

80 See Wellmann, Abh. Berl. Akad., Phil.-Hist. Kl., 1928, nr. 7; Pfister, , Byz. Ztschr. 37 (1937), 381 ff.Google Scholar; Wirbelauer, K. W., Antike Lapidarien (Diss. Berl., 1937Google Scholar); Bidez-Cumont, , Les Mages Hellénistés I, 194Google Scholar; Festugière, , Révélation d'Hermès I, 137 ff., 195 ffGoogle Scholar.

81 PGM VII, 13; VII, 781. Cf. VII, 560: ἦκέ μοι, τὸ πνεῦμα τὸ ἀεροπετές, καλούμενον συμβόλοις καὶ ὀνόμασιν ἀφθέγκτοις, and IV, 2300 ff.; Hopfner, P-W s.v. ‘Mageia’, col. 311 ff.

82 Cf. J. Kroll, Lehren des Hermes Trismegistos 91 ff., 409; C. Clerc, Les théories relatives au culte des images chez les auteurs grecs du IIe siècle après J.-C.; Geffcken, J., Arch. f. Rel. 19 (1919), 286 ff.Google Scholar; Hopfner, P-W s.v. ‘Mageia’, col. 347 ff., and OZ I, 808–12; E. Bevan, Holy Images.

83 Cf. Plot. Enn. 4, 3, 11Google Scholar (11, 23, 21, Volk.): προσπαθὲς δὲ τὸ ὁπωσοῦν μιμηθέν, ὥσπερ κάτοπτρον ἁρπάσαι εἶδός τι δυνάμενον, where ὁπωσοῦν seems to involve denying any specific virtue to magical rites of consecration.

84 Erman, Die ägyptische Religion 55; Moret, A., Ann. Musée Guimet 14 (1902), 93Google Scholar f. Eusebius seems to know this: he lists ξοάνων ἱδρύσεις among the religious and magical practices borrowed by the Greeks from Egypt (Praep. Ev. 10, 4, 4). A simple ritual of dedication by offering χύτραι was in use in classical Greece (texts in G. Hock, Griech. Weihegebräuche 59 ff.); but there is no suggestion that this was thought to induce magical animation.

85 Asclep. III, 24a, 37a–38a (Corp. Herm. I, 338, 358, Scott). Cf. also Preisigke, Sammelbuch, no. 4127, ξοάνῳ (so Nock for αοανω) τε σῷ καὶ ναῷ ἔμπνοιαν παρέχων καὶ δύναμιν μεγάλην, of Mandulis-Helios; and Numenius apud Orig. c. Cels. 5, 38.

86 This is also the period when gems incised with magical figures or formulae begin to appear in large numbers (Bonner, C., ‘Magical Amulets,’ Harv. Theol. Rev. 39 [1946], 30 ff.CrossRefGoogle Scholar). The coincidence is not fortuitous: magic is becoming fashionable.

87 Legends about the miraculous behaviour of public cult statues were, of course, as common in the Hellenistic world as in the medieval: Pausanias and Dio Cassius are full of them. But such behaviour was ordinarily viewed as a spontaneous act of divine grace, not as the result of a magical ἵδρυσις or κατάκλησις. On the classical Greek attitude, see Nilsson, , Gesch. der Griech. Rel. I, 71 ff.Google Scholar; down to Alexander's time rationalism seems to have been in general strong enough to hold in check (at least in the educated class) the tendency to attribute divine powers to images, whether public or private.

88 Apul. Apol. 63. Cf. P. Valette, L'Apologie d'Apulée 310 ff.; Abt, Die Apologie des A. u. die antike Zauberei, 302. Such statuettes, which were permanent possessions, are, of course, somewhat different from the image constructed ad hoc for use in a particular πρᾶξις.

89 Philops. 42: ἐκ πηλοῦ Ἐρώτιόν τι ἀναπλάσας, Ἄπιθι, ἔφη, καὶ ἄγε χρυσίδα. Cf. ibid. 47, and PGM IV, 296 ff., 1840 ff.

90 vit. Apoll. 5, 20.

91 Animated statues may have played a part in the classical Greek Hecate-magic; see the curious notices in Suidas, s.vv. Θεαγένης and Ἑκάτειον, and cf. Diodorus 4, 51, where Medea makes a hollow statue of Artemis (Hecate) containing Φάρμακα, quite in the Egyptian manner.

92 Eus. Praep. Ev. 5, 12Google Scholar = phil. ex orac. p. 129 f. Wolff. So the maker of the image at PGM IV, 1841, asks it to send him dreams. This explains the reference to ‘somnia’ in the Asclepius passage.

93 See the fragments in Bidez, Vie de Porphyre, Appendix 1.

94 Photius, Bibl. 215. The report is secondhand, but may be accepted as showing the main drift of Iamblichus' argument. Cf. Julian, epist. 89b Bidez, 293 A B.

95 Eunap. vit. soph. 475. Cf. PGM XII, 12. The πῦρ αὐτόματον is an old piece of Iranian magic (Paus. 5, 27, 5 f.), of which Julianus may have preserved the tradition. But it was also known to profane conjurers (Athen. 19 E; Hipp. Ref. Haer. 4. 33; Julius Africanus, κεστοί, p. 62 Vieillefond). It reappears in medieval hagiology, e.g. Caesarius of Heisterbach, Dialogue on Miracles 7, 46.

96 Suidas s.v. His ‘psychic’ gifts were further shown by the fact that the mere physical neighbourhood of an impure woman always gave him a headache.

97 de Cauzons, Th., La magie et la sorcellerie en France II, 338Google Scholar (cf. also 331, 408).

98 Cf. Wolff's appendix III to his edition of Porphyry's de phil. ex orac.; H. Diels, Elementum 55 f.; Burckhardt, Civilisation of the Renaissance in Italy 282 f. (English edition); C. R. Beard, Lucks and Talismans 14 ff. Malalas attributed to a τελεσματοποιός the virtues even of the Trojan palladium (Dobschütz, Christusbilder 80* f.).

99 Olympiodorus of Thebes in Müller's, FHG IV, 60, 15Google Scholar (= Photius, , Bibl. 58, 22Google Scholar Bekker). The fire and water were doubtless symbolized by χαρακτῆρες. It may be a coincidence that they are the two elements used in theurgic purifications (Proclus, , in Crat. 100, 21Google Scholar).

100 Jâbir et la science grecque (= Mém. de l'Inst. d'Égypte 45, 1942). I am indebted to Dr. Richard Walzer for my knowledge of this interesting book.

101 Porphyry figures as an alchemist in Berthelot, Alchim. grecs 25, as well as in the Arabic tradition (Kraus, o.c. 122, n. 3). But no genuine works of his on alchemy are known to have existed. Olympiodorus, however, and other late Neo-platonists dabbled in alchemy.

102 References to the ad Aneb. in Arabic literature are quoted by Kraus, o.c. 128, n. 5.

103 I do not know on what ground Hopfner, (OZ II, 70Google Scholar ff.) excludes both these types of operation from his definition of ‘theurgic divination proper’. In defining a term like theurgy we should be guided, it seems to me, by the ancient evidence and not by a priori theory.

104 See Oesterreich's valuable book, Possession, and N. K. Chadwick's Poetry and Prophecy. The influence of this belief on ancient religious ideas has not yet been fully explored, though one aspect of it has been admirably studied by A. Delatte in his Les conceptions de l'enthousiasme chez les philosophes présocratiques, and Tambornino's de antiquorum daemonismo provides a good collection of material. For secondary personalities professing to be pagan gods and accepted as such by Christian exorcists, cf. Felix, Min., Oct. 27, 6 f.Google Scholar; Severus, Sulpicius, Dial. 2, 6Google Scholar (PL 20, 215C), etc.

105 in Remp. II, 123, 8 ff. To judge from the context, the aim of this τελετή was probably, like that of the real or imaginary experiment with the ψυχουλκός ῥάβδος which Proclus quotes at 122, 22 ff. from Clearchus, to procure a ‘psychic excursion’ rather than possession; but it must in any case have involved the induction of some sort of trance.

106 ‘Greek Oracles,’ in Abbott's Hellenica 478 ff.

107 Lines 216 ff. Wolff (= Eus. Praep. Ev. 5, 9). G. Hock, Griech. Weihegebräuche 68, takes the directions as referring to withdrawal of the divine presence from a statue. But such phrases as βροτός θεόν οủκέτι χωρεῖ, βροτόν αἰκίʒεσθε, ἀνάπανε δέ φῶτα, λῦσόν τε δοχῆα, ἄρατε φῶτα γέηθεν ἀναστήσαντες ἑταῑροι, can refer only to a human medium. (‘Controls’ at modern séances regularly speak of the medium in this way, in the third person.)

108 This is stated in several of Porphyry's oracles, e. g. I, 190, θειοδάμοις Έκάτην υε θεὴν ἐκάλεσσας ἀνάγκαις, and by Pythagoras of Rhodes whom Porph. quotes in this connection (Praep. Ev. 5, 8Google Scholar). Compulsion is denied in the de myst. (3, 18, 145, 4 ff.), which also denies that ‘the Chaldaeans’ use threats towards the gods, while admitting that the Egyptians do (6, 5–7). On the whole subject, cf. B. Olsson in ΔΡΑΓΜΑ Nilsson, 374 ff.

109 In CMAG VI, 151, 10 ff. he mentions purification by brimstone and sea water, both of which come from classical Greek tradition; for brimstone, cf. Horn. Od. 22, 481Google Scholar, Theocr. 24, 96, and Eitrem, Opferritus 247 ff.; for sea water, Dittenberger SIG 3 1218, 15, Eur. IT 1193, Theophr. Char. 16, 12Google Scholar. What is new is the purpose—to prepare the ‘anima spiritalis’ for the reception of a higher being (Porph. de regressu fr. 2). Cf. Hopfner, P-W s.v. ‘Mageia’, col. 359 ff.

110 Cf. λύσατέ μοι στεφάνους in the Porphyrian, oracle (Praep. Ev. 5, 9Google Scholar), and the boy Aedesius who ‘had only to put on the garland and look at the sun, when he immediately produced reliable oracles in the best inspirational style’ (Eun. vit. soph. 504).

111 Porphyry, l.c.

112 Proclus, in CMAG VI, 151Google Scholar, 6: ἀπόχρη γἀρ πρὸς μέν αὐτοφάνειαν τὸ κνέωρον.

113 in Remp. II, 117, 3; cf. 186, 12. Psellus rightly calls it an Egyptian practice (Ep. 187, p. 474 Sathas): cf. PGM va, and the Demotic Magical Papyrus of London and Leiden, verso col. XXII, 2 ff.

114 de myst. 157, 14. Olympiodorus, in Alc. p. 8 Cr., says that children and country people are more prone to ἐνθουσιασμός owing to their lack of imagination (!).

115 Cf. Hopfner's interesting paper, ‘Die Kindermedien in den Gr.-Aeg. Zauberpapyri,’ Festschrift N. P. Kondakov 65 ff. The reason usually alleged for preferring children is their sexual purity. The Pythia of Plutarch's day was a simple country girl (Plut. Pyth. Orac. 22, 405c)Google Scholar.

116 Cf. Lord Balfour in Proc. Soc. for Psychical Research, 43 (1935), 60Google Scholar: ‘Mrs. Piper and Mrs. Leonard when in trance seem to lose all sense of their personal identity, whereas, so far as the observer can judge, this is never the case with Mrs. Willett. Her trance sittings abound with remarks describing her own experiences, and occasionally she will make comments … on the messages she is asked to transmit.’

117 οủ φέρουσιν. This explains the line οủ φέρει με τοῦ δοχῆος ἡ τάλαινα καρδία, quoted by Proclus, in Remp. I, III, 28Google Scholar.

118 Proc. Soc. for Psychical Research 28 (1915): changes of voice, convulsive movements, grinding the teeth, pp. 206 ff.; partial anaesthesia, p. 16 f. Insensibility to fire was attributed to the medium D. D. Home, and is associated with abnormal psychological states in many parts of the world (Oesterreich, Possession 264, 270, Eng. trans.; R. Benedict, Patterns of Culture 176; Brunel, , Aissãoũa 109, 158Google Scholar).

119 Cf. PGM VII, 634, πέμψον τὸν ἀληθινόν Άσκληπιόν τινός ἀντιθέου πλανοδαίμονος: Arnob. adv. nat. 4, 12Google Scholar, ‘magi suis in accitionibus memorant antitheos saepius obrepere pro accitis’: Heliod. 4, 7' ἀντίθεός τις ἔοικεν ἐμποδίʒειν τὴν πρᾶξιν: Porph. de abst. 2, 41 f.Google Scholar; Psellus, , Op. Daem. 22, 869Google Scholar B. The source of the belief is thought to be Iranian (Cumont, Rel. Orient. 4 278 ff.; Bousset, , Arch. f. Rel. 18 [1915], 135 ff.Google Scholar).

120 Porphyry, l.c., quotes a ‘god's’ request in such circumstances that the sitting be closed: λῦε βίην κάρτος τε λόγων ψευδήγορα λέξω. Just so will a modern ‘communicator’ close the sitting with ‘I must stop now or I shall say something silly’ (Proc. SPR 38 [1928], 76).

121 According to Proclus, in Tim. I, 139, 23Google Scholar, and in Remp. 1, 40, 18, this involves, besides the presence of the appropriate αύνθημα, a favourable position of the heavenly bodies (cf. de myst. 173, 8), a favourable time and place (as often in papyri), and favourable climatic conditions. Cf. Hopfner, PW s.v. ‘Mageia’, col. 353 ff.

122 Proclus in Crat. 36, 2c ff. offers a theoretical explanation of what spiritualists would call ‘the direct voice’; it follows Poseidonian lines (cf. Greek Poetry and Life 372 f.). Hippolytus, knows how to fake this phenomenon (Ref. Haer. 4, 28Google Scholar).

123 ἐπαιρόμενον ὁρᾶται ἣ διογκούμενον. Cf. the alleged elongation of a sixteenth-century Italian nun, Laparelli, Veronica (Journ. SPR 19, 51 ff.Google Scholar), and of the modern mediums Home and Peters (ibid. 10, 104 ff., 238 ff.).

124 This is a traditional mark of magicians or holy men. It is attributed to Simon Magus (ps.-Clem. Hom. 2, 32); to Indian mystics (Philost. vit. Apoll. 3, 15Google Scholar); to several Christian saints and Jewish rabbis; and to the medium Home. A magician in a romance lists it in his repertoire (PGM XXXIV, 8), and Lucian satirizes such claims (Philops. 13, Asin. 4). Iamblichus' slaves bragged of their master being levitated at his devotions (Eunap. vit. soph. 458).

125 See the passages from Psellus and Nicetas of Serrae collected by Bidez, Mélanges Cumont 95 ff. Cf. also Eitrem, , Symb. Oslo. 8 (1929), 49 ffGoogle Scholar.

126 de myst. 166, 15, where τοὺς καλουμένους seems to be passive (sc. θεούς), not (as Parthey and Hopfner) middle (= τοὺς κλἡτορας): it is the ‘gods’, not the operators, who improve the character of the mediums (166, 18, cf. 176, 3). If so, the ‘stones and herbs’ will be σύμβολα carried by the ‘gods’ and left behind by them, like the ‘apports’ of the spiritualists.

127 Procl. in Remp. I, III, IGoogle Scholar; cf. in Crat: 34, 28, and Psellus, , PG 122, 1136Google Scholar B.

128 Nazianzus, Gregory of, Orat. 4, 55Google Scholar (PG 35, 577 C).

129 ‘Kindermedien,’ 73 f.

130 Cf. de myst. 3, 14, on various types of φωτός ἀγωγή.

131 Simpl. in phys. 613, 5Google Scholar, quoting Proclus, who spoke of a light τὰ αủτοπτικὰ θεάματα ἐν ἑαυτῷ τοῖς ἀξίοις ἐκφαῑνον ἐν τούτω άρ τὰ ἀτύπωτα τυποῦσθαί φησι κατὰ τό λόγιον. Simplicius, however, denies that the Oracles described the apparitions as arising ἐν τῷ φωτί (616, 18).

132 Greek Magical Papyri in the British Museum 14. Reitzenstein, Hell. Myst.-Rel. 31, translated it ‘damit sie sich forme nach’.

133 de myst. 133, 12: τοτὲ μὲν σκότος σύνεργον λαμβάνουσιν οί φωταγωγοῦντες, cf. Eus. Praep. Ev. 4, 1Google Scholar. Conjurors pretend for their convenience that darkness is necessary, Hipp. Ref. Haer. 4, 28Google Scholar.

134 de myst. 133, 13: τοτὲ δὲ ἡλίου φῶς ἥ σελήνης ἤ ὄλως τὴν ὑπαίθριον αủγήν συλλαυβανόμενα ἔχουσι πρός τὴν ἔλλαμψιν. Cf. Aedesius, supra n. 110, Psellus, Expos. or. Chald. 1133 B, and Eitrem, , Symb. Oslo. 22, 56Google Scholar ff.