¹ ‘One thing at least cannot be denied the ultimate author of Acts… and that is the language and style in which the speeches are written’, Cadbury, H. J., ‘The Speeches in Acts’, Beginnings of Christianity (ed. Jackson, F. and Lake, K.; London: Macmillan, 1933) 5. 407;Google Scholar‘Now, no one (as far as I know) denies that the language of all the speeches is, generally speaking, Luke's,’ Gasque, W.W., The History of the Criticism of the Acts of the Apostles (Tüubingen: J.C.B. Mohr, 1975) 229.Google Scholar

² Using ‘the general sense of the things actually said’, Thucydides, 1. 22. Cf. Gasque, W.W., ‘The Speeches of Acts: Dibelius Reconsidered’, New Dimensions in New Testament Study (ed. Longenecker, R. and Tenney, M.; Grand Rapids: Zondervan, 1974) 232–50, esp. 242–7.Google Scholar

³ Dibelius, , ‘The Speeches in Acts and Ancient Historiography’, Studies in the Acts of the Apostles (London: SCM, 1956) 179.Google Scholar

⁴ For full bibliography and discussion see Gasque, History of the Criticism, esp. Ch. 7.Google Scholar

⁵ The reading: τήν θεάν of Acts 19. 37 by the original hand of Codex Bezae-corrected to τήν θεόν-and its family shows unfamiliarity with the usage in Ephesus and is probably a later harmonization with θες in v. 27.Google Scholar

⁶ Robertson, A. T., A Grammar of the Greek New Testament in the Light of Historical Research (Nashville: Broadman, 1934) 252. Cf. p. 257 where he takes the presence of θεά and θεός so close together in Acts 19 as merely stylistic variation.Google Scholar

⁷ Aristophanes, for instance, uses ή θεός 20 times while ή θεά occurs only in one line preserved in two fragments (Meineke, ed., Fragmenta Comicorum Graecorum II. 2, p. 1030 sec. 12).Google ScholarLSJ gives several examples of ή θεός from as far back as Homer; Goodwin, W., A Greek Grammar (2nd ed.; London: 1894) sec. 158;Google ScholarSmyth, H., Greek Grammar (Cambridge, Mass.: Harvard, 1956) sec. 198.Google Scholar

⁸ LSJ.Google Scholar

⁹ Moulton, Proleg., 244 n. Gignac (A Grammar of the Greek Papyri of the Roman and Byzantine Periods [Milano: Cisalpino-Goliardica, 1981] 2. 97–8) states that ή θεό is usually found in the papyri;Google Scholarcf. BDF, sec. 44(2). Also Bruce, F. F., The Acts of the Apostles (Grand Rapids: Eerdmans, 1984; repr. of 2nd ed. London, 1954) 364;Google ScholarConzelmann, H., Acts of the Apostles (Herm., ; Philadelphia: Fortress, 1987) 165. The formulaic character of the phrase is not discussed nor conclusions drawn from it in the grammars or commentaries that I checked.Google Scholar

¹⁰ I would like to thank the people of the Thesaurus Linguae Graecae Project at the University of California, Irvine for free access to the computer data bank and for their kind assistance.Google Scholar

¹¹ E.g. Men of Cappadocia (Geog. 12. 3); Atargatis of Syria (Geog. 16.1) and the Phrygian Great Mother (Geog. 10. 3).Google Scholar

¹² Aphrodite (Geog. 9. 5; 11. 2) and Nemesis (Geog. 13. 1).Google Scholar

¹³ As cited by Moulton, n. 9 above. The same use of ή θεός for both state goddesses and other deities is found in Polybius (the Great Mother, Hist. 21. 37. 7 and the ‘goddess’ Truth, Hist. 13. 5. 4) and Dio Chrysostom (Artemis Ephesia, Or. 31. 55 and the ‘goddess’ Pleasure, Or. 4. 101).Google Scholar

¹⁴ The inscriptions have recently been collected and made more accessible through Wankel, et al. eds., Die Inschriften von Ephesos (IEph.) (8 vols.; Bonn: Rudolf Habelt, 1979–81).Google Scholar

¹⁵ The dates for Xenophon the novelist — not the historian with the same name — are unknown, yet he may have written his Ephesiaca as early as 125 AD; Gareth, Schmeling, Xenophon of Ephesus (Boston: Twayne, 1980) 18–19. Xenophon uses ή θεός 14 times, 9 times for Artemis. He uses ή θεά twice; once in the formula discussed below and once in direct address to Isis: ‘ὠμεγίστη θεό’, Eph. 5. 13. 4 which conforms to the classical use of Ged in prayers.Google Scholar

¹⁶ The Greek translation of this Roman decree is also worth noting because it tones down the divinity of the emperor: quae a divo Augusto deo deae restitul[ta sun]t (IEph. 19b. 5–6); τν ύπό το Σεβαστο άποκατασταθεισν τ τε(IEph. 18b. 5–6). See Kreitzer, L. (‘A Numismatic Clue to Acts 19. 23–41’, JSNT 30 [1987] 59–70) for the suggestion that Claudius was not very popular with the Ephesians about this time.Google Scholar

¹⁷ Other divinities are called θεά by the Ephesians: the empress Sabina (Σαβîναν τήν θεάν Σεβστήν, IEph. 278.1) and Hygia (IEph. 1212. 2).Google Scholar

¹⁸ Cf. Xenophon of Ephesus when he refers to Artemis without the formula: τήν πάτριονήμîν θεόν, τήν μεγάλην Εφεσίων Αρτεμτν(Eph.1.11.5). Also: μεγάλην θεόν… τήν Ιστν (Eph. 5. 13. 3); Aristophanes, τ⋯ς άρουραîαζ θεοû (Ran. 840); Dio Chrysostom, γυναικείαν θεόν (Or. 4. 101); Strabo, ἔστι δέ τις Νέμεσις μεγάλην θεός, ⋯… (Geog. 13. 1), Фρυγίαν θεόν μεγάλην(Geog. 10. 3),τ⋯ν Συρíαν θεόν τ⋯ν Άταγράτιν(Geog. 16. 1), τ αύτ θε (Geog. 12. 3).Google Scholar

¹⁹ SEG 34 (1984) 1124. 9; τ θε occurs on the next line.Google Scholar

²⁰ These statistics do not include conjectural readings where the stone is broken which would probably double the amount of data if the original readings were known.Google Scholar

²¹ Both the Dalmeyda, (1926) and Peerlkamp (1818) editions give θεός; as a secondary reading here; however, this reading should be regarded as a later harmonization to the predominant use of ή θεός, outside the formula.Google Scholar

²² Strabo (Geog. 14.1) and Dio Chrysostom (Or. 31. 55) both refer to Artemis of Ephesus as ή θεός. AS a native of Asia Minor, Dio may have been familiar with the usage at Ephesus, yet Strabo refers to all individual goddesses in this fashion.Google Scholar

²³ See n. 18.Google Scholar