Hostname: page-component-76fb5796d-9pm4c Total loading time: 0 Render date: 2024-04-29T01:20:15.989Z Has data issue: false hasContentIssue false
  • English
  • Français

Gods and emperors: the Greek language of the Roman imperial cult

Published online by Cambridge University Press:  11 October 2013

S. R. F. Price
S. R. F. Price
Affiliation:
Lady Margaret Hall, Oxford
Get access Rights & Permissions [Opens in a new window]

Extract

The Greeks under Roman rule suffer from a double prejudice. On the one hand, Hellenists lose interest in the Greeks after the classical period; on the other, Roman historians find it hard to avoid a Romanocentric perspective. This double prejudice becomes particularly acute when the issue is the religious language used by the Greeks to refer to the Roman emperor. For example, the Greeks called the living emperor both theou huios (‘son of god’) and also theos (‘god’). The language looks odd from the perspectives both of classical Athens and of imperial Rome. One way to make sense of it is to treat it as a translation out of Latin. Thus the bizarre practice of calling the emperor theou huios is seen as perfectly natural because it is simply the translation of divi filius. Why natural? Because, as the heirs of Rome, we can attempt to ignore the cultural differences between us and the ancient world. But the tactic of treating Greek as a translation out of Latin does not always work. Calling the living emperor theos cannot be seen as a translation of divus, a term which applies only to dead emperors. Modern scholars are therefore forced to treat the usage as ‘deviant’, the product of either folly or flattery. In fact the failure of theos to translate divus undermines the first assumption that theou huios is a translation of divi filius.

Type
Research Article
Information
The Journal of Hellenic Studies , Volume 104 , November 1984 , pp. 79 - 95
Copyright
Copyright © The Society for the Promotion of Hellenic Studies 1984

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.)

References

1 I should like to thank Mary Beard and Lucia Nixon who have greatly improved this article. The analysis runs parallel to that of my book Rituals and power: the Roman imperial cult in Asia Minor (Cambridge 1984)Google Scholar. I refer to corpora of inscriptions by the standard abbreviations; those not listed by Hondius, J. J. E., Saxa loquuntur (Leiden 1938)Google Scholar are mainly to be found in the series Inschriften griechischer Städte aus Kleinasien (Bonn 1972–)Google Scholar.

2 See, however, Pötscher, W., Theos. Studien zur älteren griechischen Gottesvorstellung (Diss. Wien 1953)Google Scholar; Burkert, W., Griechische Religion der arch. u. kl. Epoche (1977) 406–8Google Scholar.

3 von Wilamowitz, U., Der Glaube der Hellenen i (Berlin 1931) 17–18Google Scholar; Habicht, C., Gottmenschentum und griechische Städte 2 (Munich 1970) 157Google Scholar; cf. Nilsson, M. P., Geschichte der griechischen Religion ii 2 (1961) 197–8Google Scholar. See contra Pötscher (n. 2) 187–218.

4 For this distinction see, for example, Strawson, P. F., Subject and predicate in logic and grammar (London 1974)Google Scholar. Of course it is possible to convert theos into a name by adding the definite article. ‘This is the theos’ is comparable to ‘This is Tiberius’.

5 Molinari, P., New Catholic Encyclopedia iii (1967) 55–9Google Scholar.

6 Sextus Empiricus, contra mathematicos ix 182–90Google Scholar and Cicero, , de natura deorum iii 43–52Google Scholar, quoted from 43: ‘si di sunt, suntne etiam Nymphae deae? si Nymphae, Panisci etiam et Satyri? hi autem non sunt; ne Nymphae [deae] quidem igitur. at earum templa sunt publice vota et dedicata. ne ceteri quidem ergo di, quorum templa sunt dedicata.’ See on this type of argument Couissin, P., ‘Les sorites de Carnéade contre le polythéisme’, REG liv (1941) 43–57CrossRefGoogle Scholar; Barnes, J., ‘Medicine, experience and logic’, in Science and Speculation, ed. Barnes, J. et al. (Cambridge 1982) 24–68Google Scholar; and esp. Burnyeat, M. F., ‘Gods and heaps’, in Language and logos, ed. Schofield, M., Nussbaum, M. C. (Cambridge 1982) 315–38CrossRefGoogle Scholar.

7 Cicero's use of the argument shows that deus is comparable to theos.

8 Fishwick, D., The Imperial Cult in the Latin West i (Leiden forthcoming)Google Scholar, following Habicht (n. 3).

9 Isoc. Ep. 3.5.

10 Habicht (n. 3) 156–9; Badian, E., in Ancient Macedonian Studies in Honor of Charles F. Edson (Thessaloniki 1981) 27–71, esp. 54–9Google Scholar.

11 Fraser, P. M., Ptolemaic Alexandria (Oxford 1972) 215–20Google Scholar.

12 Mørkholm, O., Studies in the Coinage of Antiochus IV of Syria (Copenhagen 1963) 68–74Google Scholar.

13 Mellor, R., ΘΕΑ ῬΩΜΗ (Göttingen 1975) 112Google Scholar.

14 Habicht indeed argues that Augustus explicitly prohibited the Greeks from using it themselves: in Le culte des souverains dans l'empire romain, Entr. Hardt xix (Vandoeuvres 1973) 41–88Google Scholar.

15 Philo, Leg. 353.

16 Josephus, AJ xix 284.

17 Taeger, F., Charisma ii (Stuttgart 1960) 187 n. 3Google Scholar; Habicht (n. 14) 84.

18 E.g. Tiberius, , Ann. Ep. 1934 89Google Scholar; Claudius, , IGR iii 328Google Scholar; Nero, Le Bas, P., Waddington, W. H., Inscriptions grecques et latines … iii (1870) 600aGoogle Scholar; Domitian, , I.Priene 229Google Scholar; Hadrian, , IGR iii 286Google Scholar; Pius, Antoninus, IGR iv 594Google Scholar; Commodus, , TAM ii 829Google Scholar. Cf. also Veyne, P., Latomus xxi (1962) 57Google Scholar; Le Glay, M., BCH c (1976) 351–3Google Scholar.

19 See, however, Plautilla, , Imhoof-Blumer, F., Kleinasiatische Münzen (Vienna 1901) 106 no. 11, 107 no. 6Google Scholar; SNG von Aulock 2412–15, 2694–6; Domna, Julia, IGBulg ii 623Google Scholar; Mammaea, Julia, IGBulg ii 640Google Scholar; Gordiani, , Mionnet, T. E., Description de médailles antiques … iii 545 no. 45Google Scholar; also Julian, , Side Agorasi ve Civanndaki Binalar (Ankara 1956) 81 no. 48Google Scholar.

20 Price (n. 1) ch. 9.

21 Josephus, AJ xix 345, 347; Acts of Apostles xii 22Google Scholar.

22 IG xii suppl. 124 (Eresus).

23 Sherk, R. K., Roman documents from the Greek East (Baltimore 1969)Google Scholar no. 65 lines 41 and 43 (a slightly superior text is in Laffi, U., SCO xvi [1967] at p. 22)Google Scholar. Cf. also I.Olympia 53.8 and 37 and OGIS 456.17.

24 References in François, G., Le polythéisme et l'emploi an singulier des mots THEOS, DAIMON (Paris 1957) 317–23Google Scholar.

25 SNG Copenhagen Aeolis 139–43 (Cyme), Phrygia 567–8 (Laodicea), 702 (Synaos).

26 Cerfaux, L., Tondriau, J., Le culte des souverains dans la civilisation gréco-romaine (Tournai 1957) 191Google Scholar.

27 Robert, L., Hellenica xiii (1965) 176Google Scholar analyses the data from Aphrodisias; e.g. ἱερεὺς διὰ βίου θεοῦ Διονύσου (MAMA viii 454)Google Scholar, τοῦ ἱεροῦ θεᾶς ᾿Αφροδείτης (MAMA viii 521)Google Scholar.

28 Syll. 3 867 = I. Ephesos ia 24.

29 IGR iii 583–4 = TAM ii 188–9.

30 As Robert notes, the development of this usage has not received detailed study. It seems to be largely a phenomenon of the later Hellenistic and Roman periods, but it has roots in Homer (Od. xiii 189–90; xix 396–7). I can discern no significance in the variations in the titles of joint priesthoods of a traditional god and the emperor (e.g. IGR iv 229, 984; TAM ii 1200).

31 IGR iv 180 = I.Lampsakos 11; Hist. Zeits. xxix (1921–1922) 217 n. 1Google Scholar; BCH ix (1885) 79 no. 10Google Scholar.

32 SEG xxiii 450 (Demetrias).

33 SEG xix 760 (Ilyas); IGR iv 808 (Hierapolis); IGR iii 664 = TAM ii 408 (Patara); I.Labraunda ii 65Google Scholar.

34 Namely, , IGR iv 256Google Scholar = I.Assos 15; IGR iv 1302 = I.Kyme 19; Altertümer von Pergamon viii. 3 (1969) 164–5Google Scholar.

35 BCH v (1881) 191 no. 14Google Scholar with BCH xviii (1894) 25 no. 21Google Scholar and Fayer, C., Il culto della dea Roma (Pescara 1976) 142–3Google Scholar. It is possible that the use of thea with Roma influenced the imperial usage in the first priesthood.

36 See, however, the (obscure) debate in Athens about Alexander: Badian loc. cit. (n. 10).

37 ILS 8794 = Syll. 3 814: ὁ ἀρχιερεὺς τῶν Σεβ-

This assimilation had extra resonance because of the ancient cult of Zeus Eleutherios celebrated by the League of Greeks at the neighbouring Plataea. Cf. Etienne, R., Piérart, M., BCH xcix (1975) 63–7Google Scholar; West, W. C., GRBS xviii (1977) 315–16Google Scholar. Note that Sebastos is the Greek equivalent of Augustus (see n. 45).

38 Weinstock, S., Divus Julius (Oxford 1971) 386–92Google Scholar; Bickerman, E., ‘Consecratio’, in Culte des souverains (n. 14) 3–25Google Scholar and ‘Diva Augusta Marciana’, AJP xcv (1974) 362–76Google Scholar. See further Schwering, W., ‘Deus und divus. Eine semasiologische Studie als Ergänzung zum Artikel divus in Thesaurus linguae latinae’, Indogermanische Forschungen xxxiv (1914–1915) 1–44CrossRefGoogle Scholar.

39 Servius on Aen. v 45 ( = Varro fr. 424 Grammaticae Romanae fragmenta, ed. Funaioli). The Ateius is either Ateius Praetextatus, a contemporary of Varro (fr. 12 Funaioli), or Ateius Capito, an Augustan writer (fr. 15 Funaioli). See also Servius on Aen. xii 139 ( = Varro, de lingua lat. fr. 2, Goetz-Schoell).

40 Deus was used ‘unofficially’ of emperors (and others) (Du Quesnay, I. M. Le M., Papers of the Liverpool Latin Seminar iii (1981) 104Google Scholar, on the late Republic; TLL s.v. ‘deus’, col. 891. 10–78), but this shows only that divus might be seen as a subset of deus; there remains a contrast with the undivided category of theos.

41 For a list see Cagnat, R., Cours d'épigraphie latine 4 (1914) 170–2Google Scholar with Stein, E., Hermes lii (1917) 571–8Google Scholar.

42 E.g. IGR iii 82, 138, 145; iv 267, 599–601. IGR iv 924–6 are peculiar in beginning τοῖς θεῶν ἐπι- φανεστάτοις (reign of Septimius Severus).

43 IGR iv 1124 = Syll. 3 810: [Νέρων] Κλαύδιος, θεοῦ Κλαυδίου υἱός, Τιβεριου Καίσ[α]ρος Σεβαστοῦ καὶ Γερμανικοῦ Καίσαρος ἔγγονος, θεοῦ Σεβαστοῦ ἀπόγονος , Καῖσαρ Σ[εβ]αστὸς Γερμανικός , ἀρχ- ιερεύς , δημαρχικῆς ἐξουσίας , αὐτοκράτωρ, ῾Ροδίων ἄρχουσι βουλῇ [δή]μῳ χαίρειν.

44 IGR iii 83 = ILS 5883 = ÖJh xxviii (1933) Beib. 64 no. 13 (Amastris)Google Scholar: divi Aug. perpetuus sacerdos / ὁ τοῦ ἐπουρανίου θεοῦ Σ(ε) βαστοῦ ἀρχ[ιερεὺς διὰ βίου?]; Reynolds, J., Aphrodisias and Rome (London 1982) no. 22. 6Google Scholar; IGR iv 1150 = ASAA xxvii–xxix (1949–1951) 284 no. 38Google Scholar (Camirus): τετειμημένος ἐς τὸ διενεκὲς ὑπὸ τῶν ἐν θεοῖς Αὐτοκρατόρων; SEG xi 492–3 (Laconia)Google Scholar: ὁ ἐν θεοῖς ῾Αδριανός.

45 Note also how the Greek ‘sebastophantes’ used wrongly to be equated with the flamen Augusti, the standard term for an imperial priest in the western empire; in fact the sebastophantes probably displayed an imperial image in the imperial mysteries, which are unique to the Greek world (Pleket, H. W., HThR lviii [1965] 338–41Google Scholar). It is also wrong to imagine that Sebastos is an exact translation of the Latin Augustus. It did indeed become the standard equivalent almost instantaneously (e.g. SEG xxvi 1243 = I.Ephesos iii 902), but its semantic motivation is more strongly religious than Augustus. It also functioned differently; ‘the Sebastoi’ may refer to two joint Augusti, but it may equally refer to a sole emperor and his collective ancestors: e.g. IGR iv 1676 (Apollonia, Mysia; AD 40–1); IGR iv 1509 = Sardis vii.1 45Google Scholar (c. AD 80); IGR iii 493 (Oenoanda; Trajanic); IG v.1 380 (Sparta; AD 115).

46 Thus the question whether theou huios was used of someone who was not divi filius is irrelevant to my case.

47 IGR iv 201 = I.Ilion 81: Αὐτοκράτορα Καίσαρα θεοῦ υἱὸν θεὸν Σε[βα]στὸν, ἀνυπερβλήτοις πράξεσιν κεχ[ρη]μένον καὶ εὐεργεδίαις ταῖς εἰς ἅπ[αν]τας ἀνθρώπους. Cf. IGR iv 309–11, 314 (Augustus, Pergamum); Robert, L., RPh xiii 3 (1939) 181–3Google Scholar = Opera Minora Selecta ii 1334–6 (Tiberius, Myra)Google Scholar; IGR iii 286 (Hadrian, Isaura).

48 IGR iv 1302 = I.Kyme 19, lines 54–7: ἐπἰ ἰερέος τᾶς ῾Ρώμας καὶ Αὐτοκράτορος Καίσαρος θέω ὐίω θέω Σεβάστω ἀρχιέρεος μεγίστω καὶ πατρὸς τᾶς πατρίδος. Cf. IGR iv 594 (Synaos).

49 Robert, L., Hellenica ii (1946) 37–42Google Scholar.

50 E.g. IGR iv 454 (Pergamum; AD 16): τὸν νεωκόρον θεᾶς ῾Ρώμης καὶ θεοῦ Σεβαστοῦ Καίσαρος IGR iii 360 = Robert, L., Les gladiateurs dans l'Orient grec (Paris 1940) no. 97Google Scholar (Sagalassos; 2nd cent.): ἀρχιερέα τοῦ Σεβαστοῦ. The provincial assembly of Asia, however, attempted to follow the lead of Rome; its high priest was ‘of theos Sebastos’ only after Augustus' death (Buckler, W. H., RPh ix 3 [1935] 177–88Google Scholar).

51 Mason, H. J., Greek terms for Roman institutions (Toronto 1974)Google Scholar. Note the review by Crawford, M., JRS lxix (1979) 249–50Google Scholar.

52 Quine, W. V. O., Word and Object (Cambridge, Mass. 1960)Google Scholar.

53 Bowersock, G. W., in Culte des souverains (n. 14) 199Google Scholar: ‘It is evident that cultivated Greeks at least were fully conscious of the difference between divus and deus, even if they were obliged to render both by the same word’.

54 See e.g. ZifF, P., Semantic Analysis (Ithaca 1960)Google Scholar. An obvious area of contrast, which I do not discuss here, is between ‘god’ and ‘hero’: see Price (n. 1) esp. ch. 2. I note here that the pagan use of ‘son of god’ probably has no bearing on early Christian usage: see Hengel, M., The Son of God (London 1976) 30Google Scholar.

55 Riewald, P., De imperatorum romanorum cum certis dis et comparatione et aequatione (Diss. phil. Halle xix. 3 1912)Google Scholar has much of the evidence.

56 Maiuri, A., Nuova silloge epigrafica di Rodi e Cos (Florence 1925) no. 468Google Scholar.

57 SEG xv 530 (Chios)Google Scholar.

58 Nock, A. D., JHS xlviii (1928) 30–8Google Scholar = Essays on Religion and the Ancient World (Oxford 1972) 144–52Google Scholar. For a poet as ‘new Homer’ see Raubitschek, A. E., Hesp. xxiii (1954) 317–19CrossRefGoogle Scholar and I.Side 107.

59 Hist. Zeits. xxix (1921–1922) 217 n. 1 (Apollo)Google Scholar; AA 1903 193, IGR iv 1053, 1061 (Asklepios)Google Scholar.

60 Robert, L., BCH cii (1978) 437–52Google Scholar = SEG xxvii 809.

61 TAM ii 760c (Arneae): Τιβέριον Κλαύδιον Καίσαρα Σεβαστὸν Γερμανικόν, θͺεὸν ἐπιφανῆ, σωτῆρα καὶ το[ῦ] ἡμετέρου δήμου, ᾿Αρνεατῶν ἡ βουλὴ καὶ ὁ δῆμος ἑτείμησεν ταῖς πρώτ[αις τειμαῖς ]. Also e.g. IGR iii 680 = TAM ii 420 (Patara), IGR iv 986 (Samos). I discuss ἐμϕανής and ἐπιϕανής together as I cannot detect any significant differences between them (cf. LSJ s.vv.).

62 Haspels, C. H. E., The Highlands of Phrygia (Princeton 1971) 333 no. 93Google Scholar: αὐτοκράτορα (Μᾶρκον) Αὐρήλιοͺ[ν]᾿Αντωνεῖνον, θεῶν ἐνφανέστατον, γῆς καὶ θαλάσσης δεσπότην, καὶ αὐτοκράτορα Κόμοδον Καίσαρα, τῶν θεῶν ἐνφανέστατον, γῆς θαλάσσης δεσπότην . Also e.g. IGR iii 704 iii. 14–16 (Cyaneae), SEG xvi 758 (Derbe), IGR iv 341 (Pergamum).

63 Mellor (n. 13) 114; similarly Weinstock (n. 38) 296–7 and Le Glay, M., BCH c (1976) 353, 365Google Scholar. See generally Pfister, F., RE suppl. iv (1924) 277–323Google Scholar and Pax, E., RAC v (1962) 832–909Google Scholar.

64 Syll. 3 796 B 10; IGR iii 628= TAM ii 288.

65 LSJ s.v. ἐπιϕάνεια.

66 Nock (n. 58) 38–41 = Essays 152–6.

67 Steinleitner, F., Die Beicht im Zusammenhange mit der sakralen Rechtspflege in der Antike (Leipzig 1913) 15–21Google Scholar; Altertümer von Pergamon viii. 3 no. 101Google Scholar.

68 I.Didyma 496 with Robert, L., Hellenica xi–xii (1960) 543–6Google Scholar: ἐπεί, ἐξότε τὴν ἱερατείαν ἀνείληφεν, οὐδέποτε οὔτως οἱ θεοὶ ἐνφανεῖς δι ᾿ ἐπιστάσεων γεγένηται, τοῦτο μὲν διὰ παρθένων καὶ γυναικῶν, τοῦτο δὲ καὶ δι ᾿ ἀρρένων καὶ νηπίων, τί τὸ τοιοῦτο καὶ εἰ ἐπὶ αἰσίῳ; Cf. POxy 1381 (2nd cent. AD).

69 Sherk (n. 23) 65, lines 40–1: ἧρξεν δὲ τῶι κόσμωι τῶν δι ᾿ αὐτὸν εὐανγελί[ων ἡ γενέθλιος ἡμέ]ρα τοῦ θεοῦ. ‘Epiphaneia’ was not often used explicitly. For imperial visits epidemia, which lacks religious overtones, was more frequent. Cf. also parousia (Robert, L., Hellenica xiii (1965) 129–31Google Scholar).

70 ICR iv 251 = I.Assos 26: ἐπεὶ ἡ κατ᾿ εὐχὴν πᾶσιν ἀνθρώποις ἐλπισθεῖσα Γαίου Καίσαρος Γερμανικοῦ Σεβαστοῦ ἡγεμονία κατήνγελται, οὐδὲν δὲ μέτρον χαρᾶς εὕρηκε ὁ κόσμος , πᾶσα δὲ πόλις καὶ πᾶν ἕθνος ἐπὶ τὴν τοῦ θεοῦ ὄφιν ἔσπευκεν, ὡς ἄν τοῦ ἡδίστου ἀνθρώποις αἰῶνος νῦν ἐνεστῶτος, … Cf. IGR iv 1102, a dating formula.

71 IGR iv 145 = Syll. 3 798 (Cyzicus): ἐπεὶ ὁ νέος ῾῾Ηλιος Γἀϊος Καῖσαρ Σεβαστὸς Γερμανικὸς συναναλάμψαι ταῖς ἰδίαις αὐγαῖς καί τὰς δορυφόρους τῆς ἡγεμονίας ἠθέλησεν βασιλήας, ἵνα αὐτοῦ τὸ μεγαλεῖον τῆς ἀθανασίας καὶ ἐν τούτωι σεμνότερον ᾖ . . . Cf. the reference by a Roman governor to the immortality of Livia (I.Ephesos i a 17.65).

72 IGR iv I44 = SEG iv 707 (Cyzicus). See generally Instinsky, H. U., ‘Kaiser und Ewigkeit’, Hermes lxxvii (1942) 313–55Google Scholar.

73 SEG xvii 750.

74 SEG xxviii 1227 (Tlos): συνεσταμένη δὲ καὶ Σεβαστῶν γένος κατὰ διαδοχὴν ἰερωτάτην θεῶν ἐπιφανῶν οἶκον ἅφθαρτον καὶ ἀθάνατον εἰς τὸν ἀεὶ χρόνον.

75 Nilsson (n. 3) ii 140–1; Habicht (n. 3) 196. Also I.Side 101 = Ann. Ep. 1966 462Google Scholar (Pompey) and IG v.1 435 (procurator of Augustus).

76 Rudhardt, J., Notions fondamentales de la pensée religieuse … (Geneva 1958) 57Google Scholar; Habicht (n. 3) 211–12.

77 So too Is-Olympic games were modelled on (isos) the traditional games at Olympia, down to the details of organisation, but they were not held at Olympia.

78 Bowersock, G. W., in Cube des souverains (n. 14) 182–4Google Scholar.

79 E.g. IGR iii 473 (Balbura); IGR iv 98 (Mytilene); IGR iii 493 (Oenoanda); IGR iv 1155 (Sandaina); Ann. Ep. 1972 626 (Side)Google Scholar; IGR iii 1507 (Termessus minor).

80 IG xii suppl. 124. 25–6 (Eresus): οὐ μόνον τ[ὰν πόλι]ν ἀλ[λὰ καὶ τὰν] [λο]ίπαν ἐπαρχήαν μάρτυρα ποήμενος τᾶς εἰς τὸν θέον εὐσε[βεία]ς. Cf. also IGR iv 1608c = I.Ephesos vii.2 3801 iiGoogle Scholar; I.Ephesos ii 236Google Scholar.

81 JRS l (1960) 75–9Google Scholar = SEG xviii 578 = Herrmann, P., Der römische Kaisereid (Göttingen 1968) 124 no. 5Google Scholar: αὐτο[ί] τε καὶ οἱ ἔκγονοι ἡμῶν ὑπακͺούσεσθαι πειθαρχήσειν κατά τε γῆν καὶ κατὰ θάλαττ[αν] εὐνοήσειν σεβάσ<e> σθαι vacat Τιβέριον Καίσαρα Σεβαστοῦ ὑὸν Σεβαστὸν σὺν τῶι ἃπαντι αὐτοῦ οἴκωι. Cf. Weinstock, S., ‘Treueid und Kaiserkult’, Ath. Mitt. lxxvii (1962) 306–27Google Scholar. Note also the importance of asebeia (‘impiety’) to the emperor: Thasos ii 185Google Scholar = SEG xviii 350; Reynolds (n. 44) no. 62; Hist. Zeits. xxix (1921–1922) 222 n. 2Google Scholar (Cos). The oaths of Paphlagonia (n. 88) and Assos (n. 70) call only for the subjects to eunoēsein the emperor.

82 IGR iv 981: εὐσεβείας μὲν [ἕν]εκεν τῆς [εἴ]ς [τ]ε τὴν [ἀρχ]ηγέτιν ῾῾Ηραν καὶ Καίσαρα Γερμανικοῦ υἱὸν Γερμανικὸν Σεβαστὸν καὶ τὸν οἶκον αὐτοῦ, εὐνοίας δὲ καὶ φιλοδόξου διαθέσεως εἰς τὴν πατρίδα καὶ τὴν γερουσίαν. Cf. Milet i.3 134Google Scholar.

83 Rudhardt (n. 76) 11–17; Kaufmann-Bühler, D., RAC vi (1966) 985–1052Google Scholar; Burkert (n. 2) 408–12. A new term thrēskeia, also meaning ‘piety’, but with a more specifically religious connotation, appeared in the Roman period (Robert, L., Etudes épigraphiques et philologiques [Paris 1938] 226–35Google Scholar; Hellenica ii [1946] 132–3Google Scholar) and is attested twice of the emperor: Sherk (n. 23) no. 65 lines 25 and B 5; REG xix (1906) 100 no. 14Google Scholar.

84 See the texts on the aid of Hekate and Zeus to Stratonicea: I.Stratonikeia i 10, 14, 20Google Scholar; ii.i 512, 1101 (with Robert, L., Etudes Anatoliennes [Paris 1937] 29, 516–23Google Scholar).

85 Eusebeia was also shown to family and friends: Dio Or. xxxi 12–15; Ath. Mitt. lxxv (1960) 162, no. 60Google Scholar; Syll. 3 1107; IG ii2 1275.

86 IGR iv 145 = Syll. 3 798: εὔξασθαι μὲν ὑπὲρ τῆς Γαῖου Καίσαρος αἰωνίου διαμονῆς καὶ τούτων σωτηρίας.

87 Note that ‘Menander Rhetor’ ends his recommendations for a speech to the emperor by stressing the importance of prayers to the gods for the emperor (ed. D. A. Russell, N. G. Wilson [Oxford 1981] 92).

88 IGR iii I37 = OGIS 532 = ILS 8781 = Studia Pontica iii 66Google Scholar: ὀμνύω, Δία, Γῆν, ῾῾Ηλιον θεοὺς πάντα[ς καὶ πά]σας καὶ αὐτὸν τὸν Σεβασ[τ]ὸν εὐνοή[σειν Καί]σαρι Σεβαστῶι καὶ τοῖς τ[έκ]νοις ἐγγό[νοις τε] αὐτοῦ πάν[τ]α [τὸ]ν τοῦ [βίου] χρόνον κ[αὶ λό] γωι[κ]αὶ ἔργωι καὶ γνώμη[ι] . . .

89 Knopf, R., Krüger, G., Ruhbach, G., Ausgewählte Märtyrerakten 4 (Tübingen 1965) 3–4Google Scholar.

90 I.Ephesos i a 17.56–61; IGR iv 353: ὑμνῳδοὶ θεοῦ Σεβαστοῦ καὶ θεᾶς ῾Ρώμης.

91 Thesmodoi: Forschungen in Ephesos ii 27Google Scholar = I.Ephesos i a 27. 457–8, 533Google Scholar; ‘composer and reciter of poems for the god Hadrian’ μελοποιοῦ καὶ ῥαψῳδοῦ θεοῦ : BCH ix (1885) 124–8Google Scholar = I. Ephesos i a 22. 3–4, 63–4Google Scholar.

92 Ziehen, L., RE v A (1934) 2031–3Google Scholar; Robert, L., RPh xvii 3 (1943) 184–6Google Scholar.

93 I.Didyma 148 with Robert, L., Hellenica vii (1949) 210Google Scholar.

94 I.Erythrae 205 (paean to Antiochus); IG vii 417.68 (for the Romans at Oropus).

95 Nilsson (n. 3) ii 377–81.

96 Robert, , Etudes (n. 83) 23Google Scholar, who notes that the text needs republishing: [νικ]άσαντα ἐ[γκ]ωμίοις [ἐν]ταῖς ἐπισημοτάταις τͺᾶς ᾿Ασί̣ας πόλεσι ἔς τε τὸν κτίσταν τᾶς πόλιος Σεβαστὸν Καίσαρα καὶ τὸς εὐεργέτας Τεβέριον Καίσαρα καὶ Γερμανικὸν Καίσαρα καὶ τὸν ὅλον οἶκον αὐτῶν καὶ [ἐς τὸς ἄλ]λος τὸς ἐν ἐ[κάσταις ταῖς πόλεσ]ι θεός -καὶ Παναθήναια- ᾿᾿Ισθμια.

97 Frei, J., De certaminibus thymelicis (Diss. Bâle 1900) 34–41Google Scholar; Robert, , Etudes (n. 83) 21–30Google Scholar. Add Ann.Ep. 1974 602Google Scholar and SEG xxix 452 (Thespiae); Corinth viii. 1 14.87 and 19.1–3, 5–7Google Scholar; viii.3 153 with Robert, L., REG lxxix (1966) 743Google Scholar; Hesp. xxxix (1970) 79–83Google Scholar = Ann.Ep. 1969–1970 587Google Scholar with J. and Robert, L., REG lxxxiv (1971) 434 no. 307Google Scholar (Caesarea, Corinth); Hesp. suppl. xii (1967) 189Google Scholar line 22 (Athens, fragmentary). Note also the epideictic speeches delivered by Aelius Aristides at meetings of the provincial assembly (Orr. xxiii, xxvii, xxviii, xxxiv K).

98 E.g. Apollo at Didyma: Robert, L., Hellenica xi–xii (1960) 446–9Google Scholar. Robert suggests that the praise earlier given to the Romans acted as a spur.

99 Nock, A. D., Gnomon viii (1932) 517–18Google Scholar; CAH x (1934) 481Google Scholar. He later collected a number of exceptions to his rule: ‘Deification and Julian’, JRS xlvii (1957) 115–23Google Scholar = Essays (n. 58) 833–46.

100 den Boer, W.; ‘Heerserscultus en ex-voto's in het Romeinse Keizerrijk’, Mededelingen der Koninklijke Nederlandse Akademie van Wetenschappen, Afd. Letter-kunde xxxvi.4 (1973)Google Scholar; he repeated the case in Entr. Hardt xxvi (1980) 36–41Google Scholar.

101 There has been a similarly misconceived discussion about miracles. Many have felt that their extreme rarity is crucial (Nock, ‘Deification and Julian’ [n. 99]), though Morenz, S. (Würzburger Jb. f. d. Altertumswissenschaft iv (1949–1950] 370–8)Google Scholar attempted to use one miracle to show that the imperial cult was not an empty shell. But it is arbitrary and ethnocentric to use the ‘royal touch’ as the criterion of significance; the practice is found almost exclusively in France and England and for peculiar historical reasons.

102 Heiler, F., Prayer (trans. 1932) xiiiGoogle Scholar. Versnel, H. S., ‘Religious mentality in ancient prayer’, in Faith, hope and worship, ed. Versnel, (Leiden 1981) 1–64CrossRefGoogle Scholar accepts that ‘it is possible to describe a phenomenology of prayer in general’ (p. 3).

103 Lienhardt, G., Divinity and experience (Oxford 1961) 219Google Scholar.

104 Votives were made only by individuals; Kötting, B., RAC ix (1976) 1069–70Google Scholar gives the exceptions.

105 SEG ii 718 (?Pednelissus): Αὐτοκράτορι Καίσαρι Τραϊανῶι Ἀδριανῶι Σεβαστῶι καὶ τῷ δήμῳ τὴν εὐχὴν Σάλμων Θ[έ]ονος? ἱερεὺς ⊿ιὸς καὶ προθύτης τ[ῶν Σεβ]αστῶν γενόμενος [ἀνέθη]κεν σὺν γυναικὶ(δηνάρια) σ´.

106 MAMA i 23 (Laodicea Combusta). See also IGR iv 363 (Pergamum, τύχη ἐπήκοος of emperor); MAMA vi 370 (Synnada) is perhaps a votive jointly to Zeus and the emperor; Denk.Öst.Ak. Wiss. lxxv. 1 (1952) 40 no. 78 (Claudiopolis, Antinous)Google Scholar; IGR iv 93 (Mytilene) may be a prayer to Zeus and Augustus, but the text is uncertain (IG xii suppl. p. 23); IG iv 584 (Argos), votive? to Titus; PSI 1261, a private letter from Egypt (212–17), talks of the ‘fortune’ (τύχη) of the emperor saving someone, presumably following a prayer.

107 Div. Jul. 85: ‘apud earn (sc. columnam) longo tempore sacrificare, vota suscipere, controversias quasdam interposito per Caesarem iure iurando distrahere perseveravit (sc. plebs).’

108 Philostorgius, , Kirchengeschichte (Die griechischen christlichen Schriftsteller xxi 3 [1981] 28 no. 17)Google Scholar. Cf. Weinstock, S., RE xxiii (1957) 824–5Google Scholar on propitius.

109 Cf. the disparity between the numbers of votives for Apollo and Dionysos (Taşlıklıoǧlu, Z., Anadolu'da Apollon kült'ü ile ilgili Kaynaklar [Istanbul 1963]Google Scholar; Quandt, W., De Baccho ah Alexandri aetate in Asia Minore culto [Diss. phil. Halle xxi.2 1913])Google Scholar or in the usage of epēkoos (Weinreich, O., ‘ΘΕΟΙ ΕΠΗΚΟΟΙ’, Ath.Mitt. xxxvii [1912] 1–68Google Scholar = Ausgewählte Schriften i [Amsterdam 1969] 131–95Google Scholar).

110 V. G. i 40–2, Aen. i 286–90. Cf. earlier Plut. Dion 29; hymn and prayer to Demetrios Poliorcetes (Ath. 253e); and the Jewish response to Antiochos Epiphanes (Jud. iii 8; Dan. vi 6–14). Such prayers had a long history: A. Supp. 980–2, Eupolis, , Demoi (fr. 117 K)Google Scholar, Xenophon of Ephesus i 12. Similarly Prudentius, looking back from a Christian perspective at the beginning of the fifth century, assumed that prayers formed apart of the ruler cult: c. Symm. i 245–8, 271–7.

111 Odes iv 5Google Scholar:

hinc ad vina redit laetus et alteris

te mensis adhibet deum;

te multa prece, te prosequitur mero

defuso pateris et Laribus tuum

miscet numen, uti Graecia Castoris

et magni manor Herculis.

112 Epist. ex Pont. iv 9Google Scholar:

nee pietas ignota mea est: videt hospita terra

in nostra sacrum Caesaris esse domo.

stant pariter natusque pius coniunxque sacerdos,

numina iam facto non leviora deo.

neu desit pars ulla domus, stat uterque nepotum,

hie aviae lateri proximus, ille patris.

his ego do totiens cum ture precantia verba,

Eoo quotiens surgit ab orbe dies.

See also the earlier Trist. iii 8. 13–14 and v 2. 43–78.

Livy i 16 talks of prayers to the apotheosized Romulus.

113 Philostr. VA viii 5 (p. 299 Kayser), 7 (p. 310).

114 Aristides, Or. xxvi (K) 32 (with comments of Oliver, J. H., The Ruling Power, Trans. Amer. Philos. Soc. xliii. 4 [Philadelphia 1953] 918)Google Scholar: . I assume the double prayer to be more significant than the change of preposition in the two limbs. The speech dates to AD 143: Klein, R., Die Romrede des Aelius Aristides (Darmstadt 1981) 77Google Scholar.

115 Libanius, Or. xviii 304: . Nock, ‘Deification and Julian’ (n. 99) argued for Christian influence on this passage, but my other parallels show that this is not a necessary assumption.

116 Panegyr. Lat. xii 6.4 (Budé): ‘talem esse debere qui gentibus adoratur, cui toto orbc terrarum privata vel publica vota redduntur, a quo petit navigaturus serenum, peregrinaturus reditum, pugnaturus auspicium.’ Some scholars believe that language of this sort is merely metaphorical, a rhetorical commonplace found in Pliny, Pan. iv 4. I do not believe that the distinction between the literal and the metaphorical is clear or useful; still less do I accept that only the ‘literal’ is significant. A more helpful position, which I cannot argue here, is presented by Sperber, D., Rethinking Symbolism (Cambridge 1975)Google Scholar.

117 Aquinas, , Summa Theologiae Ia 13. 9–10Google Scholar; Geach, P., God and the soul (London 1969) 57–8, 108Google Scholar; also Durrant, M., The logical status of ‘God’ (London 1973) ch. 1CrossRefGoogle Scholar. Note the debate in the early church inspired by the challenge of Arius: Williams, R. D., ‘The logic of Arianism’, JThS xxxiv (1983) 56–81, esp. 81Google Scholar (though note the very different emphasis in Gregg, R. C., Groh, D. E., Early Arianism—a view of salvation [London 1981])Google Scholar.

118 Butterworth, G. W., ‘The deification of man in Clement of Alexandria’, JThS xvii (1915–1916) 157–69Google Scholar; Gross, J., La divinisation du chrétien d'après les pères grecs (Paris 1938)Google Scholar. Note also the Indians' predication of deva (‘god’) of their kings: Gonda, J., Ancient Indian kingship from the religious point of view (Leiden 1966) 24–33Google Scholar.

119 O'Neill, C., New Catholic Encyclopedia xii (1967) 962–3Google Scholar. Anglicans, for whom the practice of invocation was proscribed by one of the 39 Articles, tend to be unhappy and ignorant about the invocation of saints. Note the interesting Anglican controversy reflected in Stone, D., The Invocation of Saints 2 (London/N.Y./Bombay 1909)Google Scholar.

120 Chantraine, P., La notion du divin depuis Homère jusqu'à Platon, Entr. Hardt i (1954) 60Google Scholar: ‘L'aspect humain des dieux est un trait essentiel. L'anthropomorphisme des dieux n'est pas seulement plastique, il est fondamental.’

121 Griffin, J., Homer on life and death (Oxford 1980) ch. 5Google Scholar.

122 Gordon, R. L., ‘The real and the imaginary: production and religion in the Graeco-Roman world’, Art History ii (1979) 5–34CrossRefGoogle Scholar.

123 Vernant, J. P., Myth and thought among the Greeks (London 1983) 328Google Scholar. Cf. Rudhardt (n. 76) 55–111, and ‘Considérations sur le polythéisme’, Revue de théologie et de philosophie xvi 3 (1966) 353–64Google Scholar, = Du mythe, de la religion grecque et de la compréhension de l'autrui (Geneva 1981) 71–82Google Scholar. For a helpful survey of other literature on the naming of gods see Gladigow, B., RAC xi (1981) 1202–38Google Scholar.

124 Philol. lxxx (1924–1925) 339Google Scholar with RhM cxii (1969) 48–53Google Scholar.

I quote these two maxims from a longer list of maxims preserved on a second-century AD papyrus, but the ideas they express were commonplace. Cf. Artemidorus, , Oneirocr. ii 36, 69Google Scholar, with Dölger, F. J., Antike und Christentum iii (1932) 128–31Google Scholar.