¹ There is much indebtedness to acknowledge. This inscription has been discussed in common by our Dura group, and many suggestions have been made by various per sons, not all of which are specifically acknowledged below. The text was studied in Professor Rostovtzeff's Seminar, where Mr. C. M. Dawson made it the subject of a special report. Advice was furnished also through letters by Monsieur Franz Cumont. I am especially indebted to Professor Rostovtzeff and to Mr. F. E. Brown, former director of the excavations, both of whom have read the manuscript to its great profit, and to Professor Nock, for much valuable advice on the subject of Hellenistic religion.

² Section G; cf. the key map published within the rear cover of The Excavations at Dura-Europos, Preliminary Report of the Seventh and Eighth Seasons (Rep. VII–VIII), New Haven, 1939.

³ Cf. the report of Hopkins, C., Rep. V (1934), pp. 73–97Google Scholar.

⁴ Cf. Rostovtzeff, Dura-Europos and its Art (1938), plan on p. 35, and Index s.v. Agora. A full account by Mr. Brown will appear in Rep. IX, now in preparation.

⁵ Professor Rostovtzeff informs me of the similar handling of a marble plaque of irregular shape from Samaria now in the Palestine Museum in Jerusalem, containing a painted inscription in honor of Kore, as yet unpublished.

⁶ The alcove behind the epistyle contained a stove, and like the alcoves in the courts of Dura houses generally, doubtless had served as a kitchen.

⁷ Inscriptions at Dura more commonly than not were embedded in the plaster of wall surfaces, and building inscriptions were commonly placed over doorways. One was found in situ in the Temple of Azzanathkona, Rep. V, p. 151, no. 468, pl. XXVIII, 1. No. 875 (Rep. VII/VIII, p. 171), from the Temple of Adonis, had been set in an epistyle.

⁸ The type is common. Cf. for example the instances collected by me at Jerash, Kraeling, C. H., Gerasa, City of the Decapolis (1938), p. 363, fig. 12Google Scholar.

⁹ The same spelling in Robert, L., Les Gladiateurs dans l'Orient Grec (1940), pp. 145 f., no. 106 (Xanthus)Google Scholar.

¹⁰ This style of composition has been often noted in the Latin epitaphs; cf. Cagnat, R., “Sur les Manuels Professionnels des Graveurs d'Inscriptions Romaines,” Rev. de Phil. N.S. XIII (1889), pp. 51–65Google Scholar; Tolkiehn, J., Neue Jahrb. f. d. kl. Altertum VII (1901), p. 184Google Scholar; Focillon, H. in Plessis, F., Poésie Latine, Epitaphes (1905), pp. xvii, xxixGoogle Scholar; Galletier, E., Étude sur la Poésie Funéraire Romaine d'après les Inscriptions (1922), 225–235Google Scholar.

¹¹ Cf. Danoff, C. M., Jahresh. oesterr. arch. Inst., XXX (1936), Beibl. 84Google Scholar.

¹² Cf. also Lanckoroński, K. Graf, Städte Pamphyliens und Pisidiens, vol. II (1892), p. 228Google Scholar, no. 208 (Sagalassus, 5th cent.): τεύχεσιν ἀστράπτει Ζήνων θρασυκάαδιος ἀνήρ, χάλκεος ἐν πολέμῳ, χρύσεος ἐν γράϕισιν.

¹³ Galletier, op. cit., p. 293; on the irregularity of meters cf. ibid., pp. 288–304.

¹⁴ It seems less likely that ΔEΞAΣΘAI and KAΛϒΨAI are aorist indicatives without the augment (= ἐδέξασθε and ἐκάλυψε), because in this case the change in person is less accountable, and the meaning of the second cliché becomes quite distorted. It would, in addition, be possible also to take both ΔEΞAΣΘAI and KAΛϒΨAI as aorist infinitives used as imperatives, or to combine KAΛϒΨAI and TE as the aorist optative with the subject θεαί, EΛAΦPA ΓAIA being then an instrumental dative with the iota adscriptum omitted. That would, however, involve the sacrifice of the connective, which is necessary even if misplaced metri causa.

¹⁵ For this cliché it is enough to cite Cumont, After Life in Roman Paganism (1922), p. 46.

¹⁶ Interesting in this connection is the study of Strauberg, K., “De Titulis Graecis Sepulcralibus” (Acta Univ. Latviensis, III, 9, 1937, pp. 313–352b)Google Scholar where all kinds of formulae are assembled. The work suffers, however, from using a syntactical rather than an ideological approach.

¹⁷ Santoro, op. cit. (note 27, below), p. 17; Schwarzlose, W., De Titulis Sepulcralibus Latinis Quaestionum Capita Quattuor (Diss. Halle, 1913), pp. 20 f.Google Scholar; Cumont, op. cit., pp. 86 f.; Steuding in Roscher's Lexicon II, 2, 2320. Lists of these texts are given by Steuding and Schwarzlose.

¹⁸ The same expression as Hesiod, Opera 122/3; later Greek writers use δαίμονες commonly of human souls, and the term Di Manes is rendered into Greek, not only as Θεοὶ Καταχθόνιοι, but also as Θεοὶ Δαίμονες; cf. now Robert, L., Rev. de Phil. LXV (1939), p. 207Google Scholar.

¹⁹ Late Jewish sources similarly refer to “reception” of the dead by “the fathers” (cf. LXX, IV Macc. 5, 37 and 18, 23).

²⁰ A fate to which βιοθάνατοι and ἀωροθάνατοι were especially liable, whether or not ἄταϕοι (Rohde, E., Psyche (8th ed., transl. W. B. Hillis, 1925), p. 533Google Scholar; Cumont, After Life, pp. 128–147; L'Egypte des Astrologues [1937], pp. 199 f.; Nock, Sallustius (cf. below, n. 27), p. xcii). Soldiers were only with difficulty removed from this group (Cumont, After Life, p. 142; cf. however Virgil, Aen. VI, 660), and if Terentius had died in action (see below), his widow may have had reason to be concerned on this point.

²¹ There is no need of analyzing completely its range of use, for which it is sufficient to refer to the works of Santoro, Jacobsen, and Otto, cited below.

²² As quoted by Cicero, Leg. II, 9, 22; 25, 62; cf. Hild in Daremberg-Saglio, III, 2, 1572; Jacobsen, J. P., Les Manes (transl. E. Philipot, 1924), vol. I, pp. 33–52Google Scholar.

²³ In this sense Manes and Lares approach each other in meaning. Cf. the definition of Festus (in Paul. Diac., 121): Lares … animae putabantur esse hominum redactae in numerum deorum (cited in the Thesaurus, loc. cit.). Jacobsen also remarks that the Manes who rule the dead could not be the individual souls (op. cit., vol. I, p. 49). Cf. also Otto, W. F., Die Manen (1925), pp. 52 fGoogle Scholar.

²⁴ There is no need of citing examples, as they are familiar. Plato makes Socrates imply in the Apology (40/41) that the idea was a part of the general Greek thinking about the after life. The same conception was adopted by the Romans together with an interest in individual immortality about the time of the Empire (Nock, H.T.R. XXV (1932), p. 345), and it appears commonly after Virgil.

^24a The passive of the verb νεμεσάω comes to be used in later Greek in the sense of “to be grudged,” i.e. success, “to fail.” This is the category I, 2 in the new Liddell and Scott. Robert in his commentary adds a parallel from another Tomi epitaph, Benndorf, O., Arch.-Epigr. Mitth. aus Oesterreich, VI (1882), pp. 29 f., no. 59Google Scholar. In the present instance, the success which the Pontarch was grudged by Fate was continued life, and the verb is practically a euphemism for ἀποθνήσκω.

²⁵ For this use of the imperfect cf. Kühner-Gerth, , Ausführliche Grammatik der griechischen Sprache, I (1898), pp. 145 fGoogle Scholar. A similar case occurs in a Rhodian epitaph published by von Gaertringen, F. Hiller, B.C.H. XXXVI (1912), pp. 230–239Google Scholar (cf. P. Boyancé Le Culte des Muses chez les Philosophes Grecs, Bibl. d. Écoles Franç. d'Athènes et de Rome, 141, 1937, p. 278, n. 1): οὔ τί σε νώνυμνον κρύπτεν τόδε Δωρίδος αἴης σῆμα.

The passive of μαραίνω is used frequently enough in the sense of “to be killed,” “to die”; cf. e.g. Benndorf, op. cit., p. 30, no. 60: ἰς τοῦτ᾽ <αὖτ᾽> ἐλύθη σῶμα μαραινόμενον. The active is inexplicable, unless we were to discover an object of the “quenching” in the EKXON of the stone (possibly better ἔσχον, “they got” or “took me” than as εἶχον). But as the text is an epitaph, there must be some reference to the death of the person concerned. Unless we are to assume a mistake on the part of the lapidary, we must, I think, regard this active aorist as intransitive in meaning, a usage possibly influenced by the metrically equivalent θανόντα.

²⁶ More parallels could doubtless be produced by a more complete survey of the field. I am inclined to see another instance in the second century inscription from Pergamum, Kaibel, 243 (Inschr. v. Perg., vol. II, no. 576; cf. Boyancé, Culte des Muses, p. 286, n. 3): ψυχὴ δ᾽ ἐκ ῥεθέων πταμένη μ[ε]τὰ δαίμονας ἄλλους ἤλ[υ]θε σή, ναίεις δ᾽ ἐν μακάρω[ν] δαπέδῳ, where the δαίμονες may be the Di Inferi, the “Other Gods,” but the tone of the verses suggests much rather that they are the spirits of the other dead, deified as in the Dura text.

²⁷ Cf. in general the writings of Rohde, Otto, Jacobsen, Cumont, and Boyancé already cited and also, in addition to the handbooks, Santoro, B., Riv. di Fil. XVII (1889), 1–62Google Scholar; Norden, E., Maro, P. Vergilius, Aeneis Buch VI (1916), 16–48Google Scholar; Capelle, P., Luna, De, Stellis, Lacteo Orbe Animarum Sedibus (Diss. Halle, 1917)Google Scholar; Foucart, P., Le Culte des Héros chez les Grecs (Mém. Ac. d. Inscr. XLII, 1918), pp. 130–167Google Scholar; Farnell, L. R., Greek Hero Cults and Ideas of Immortality (1921), pp. 361–402Google Scholar; Nock, A. D., Sallustius, Concerning the Gods and the Universe (1926), pp. xxxii, lxvi, xcivGoogle Scholar; Moore, C. H., Immortality (Our Debt to Greece and Rome, 1931)Google Scholar; Festugière, A. J., L'Idéal Religieux des Grecs et l'Évangile (1932)Google Scholar; Nock, Hellenistic Religion, Ch. 19 (forthcoming). Much useful discussion occurs in the studies of the Somnium Scipionis by R. Harder (Schriften d. Königsberger Gelehrten Gesellschaft, Geisteswiss. Kl. VI, 3, 1929) and Boyancé (Thèse Paris, 1936). For studies of the epigraphic and archaeological material, in addition to the works of Kaufmann, Stettner, Schwarzlose and Galletier, cf. Schröder, B., Studien zu den Grabdenkmälern der römischen Kaiserzeit (Diss. Bonn, 1902)Google Scholar; Lier, B., Philologus LXII (1903), pp. 563–603Google Scholar; Tolman, J. A., A Study of the Sepulchral Inscriptions in Buecheler's “Carmina Epigraphica Latina” (Diss. Chicago, 1910)Google Scholar; MrsStrong, Arthur, Apotheosis and After Life (1915)Google Scholar; Cumont, F., Syria X (1929), pp. 216–237Google Scholar.

²⁸ The well known phrase, mortua heic ego sum, et sum cinis, is cinis terra est, sein est terra dea, ego sum dea, mortua non sum (Buecheler, 1532; Cholodniak, 1133h; similar also Buecheler 974 = Cholodniak, 1123; both from Rome). Cf. Lier, op. cit., pp. 588 f. Related is the common notion of the soul or spirit returning to and being dissipated in the winds, or the ether; cf. the texts assembled by Pease, A. S., Publi Vergili Maronis Aeneidos Liber Quartus (1935), p. 537Google Scholar.

²⁹ This includes a number of changes introduced by Buecheler. Cholodniak, 1158m, reads without emendation: nunc modo ad infernas sedes Acheruntis ad undas, taetraque Tartarei sidera possideo, but admits that the text is corrupt. Note especially the adjective Tartarei without a noun. For the meaning cf. F. Cumont, C.R.A.I. 1920, p. 282.

³⁰ A more remarkable confusion exists in Kaibel 288 (Leucosia in Cyprus), where the tomb is completely overlooked. Earth covered the body, then ἡ γάρ μοι ψυχὴ μὲν ἐς αἰθέρα καὶ Διòς αὐλάς, ὀστέα δ᾽ εἰς Ἀίδην ἄτροπος εἶλε νόμος. This anomalous descent of the bones to Hades may be connected with the subsequent reference to the deceased as a bridegroom only there: γαμικὸς μοῦνος ἐνì ϕθιμένοις (he was a bachelor while living).

³¹ Cf. Rohde, Psyche, pp. 162–174, 524–527; Jacobsen, vol. I, pp. 19–25, 61–71, 94–111; vol. II, ch. XVIII, for the tomb cults in general. The sentiment is common in the epigrams, e.g. Pal. Anth. VII, 339, 7/8: λοιπόν μοι τὸ κύπελλον ἀποστίλβωσον, ἑταῖρε, κωλυτὴν δ᾽ ὀδύνης τὸν βρόμιον πάρεχε. Thus the dead may be regarded formally as the owner of the tomb, its “house”; cf. Wiesner, J., Arch. f. Rel., XXXV (1938), pp. 314–328Google Scholar.

³² Festugière, op. cit., pp. 150 f. (Aristoph., Peace, 832–837, and later instances;) Cumont, After Life, p. 105 (IG XII, 7, 123 [Amorgus]): ἀστὴρ γὰρ γενόμην θεῖος ἀκρεσπέριος.

³³ The question of the θεῖος ἄνθρωπος (L. Bieler, Θεῖος ἀνήρ, Das Bild des “Göttlichen Menschen” in Spätantike und Frühchristentum, 1935/6) is quite different (Festugière, op. cit., pp. 40 ff.). His divinization was due to special qualities or special grace, from the latter point of view being more or less accepted by Judaism and Christianity; cf. Goodenough, E. R., By Light, Light (1935), pp. 126 f., 199, 202, 221–229Google Scholar; Festugière, op. cit., p. 130, n. 4, with special reference to Hippolytus, Elench., X, 34, pp. 292 f. (Wendland).

³⁴ Cf. Kaufmann, op. cit., p. 32; F. Pfister, Die Reliquienkult im Altertum (Dieterich-Wünsch, Religionsgesch. Versuche und Vorarbeiten, vol. V, 1909/1912); Rohde, op. cit., ch. IV, and pp. 527–533; Jacobsen, op. cit., vol. II, ch. I–VIII.

³⁵ For this text cf. Eitrem, S., Arch. f. Rel. XXXIV (1937), pp. 316–321Google Scholar; Nock, A. D., Gnomon XV (1939), p. 367Google Scholar.

³⁶ The editors consider, but properly reject, the hypothesis that this is a dedication to a Πατὴρ θεός, a local form of the god Papas.

³⁷ For the confusion of ideas cf. Nock, H.T.R. XXV (1932), p. 332.Google Scholar Cf. also the tomb oracle from Thyatira, L. Robert, Études Anatoliennes (1937), p. 130.

³⁸ It is essentially different from the question of the divinity of rulers and other great men, living or dead, and I forbear to cite the extensive discussion concerned with the ruler cults. Cf. most recently Wilcken, U., “Zur Entstehung des hellenistischen Königskultes,” SB Ak. Berlin (1938)Google Scholar, XXVIII; Kern, O., Die Religion der Griechen, III (1938), pp. 111–125Google Scholar; Otto, W., SB Ak. München, Phil.-hist. Abt., 1939, 3, pp. 1–16.Google Scholar

³⁹ E.g. IG XIV, 641 and 642 (Thurii, 4th cent. B.C.) cf. Kaufmann, op. cit., p. 22; Cumont, op. cit., pp. 148 f.; W. K. C. Guthrie, Orpheus and Greek Religion (1935), pp. 171–182; Bikerman, E., Journ. Warburg Inst. II (1939), 369Google Scholar, 371 (but cf. A. D. Nock, H.T.R. XXXIII (1940), 301 ff.); K. Kerényi, Pythagoras und Orpheus (1940), Part I.

⁴⁰ This usage leads easily to misunderstanding. So Altheim, A History of Roman Religion (tr. H. Mattingly, 1938), p. 361, remarks in connection with the problem of the divinity of the Emperors, that every man, when dead, became a god and joined the ranks of the Di Manes. That is true, of course, but in becoming a “god” in the older Roman conception, a dying Roman lost his individuality. In this respect, the Greek terminology is clearer and less open to confusion.

⁴¹ Kaufmann, op. cit., p. 22.

⁴² Immortality and escape from Fate were said to be given by various gods and mysteries (Jacobsen, op. cit., pp. 26 f.; Cumont, Syria X (1929), pp. 227–237; Festugière, op. cit., Pt. II, ch. II and IV, pp. 317–328). The power of the dead to harm, and the converse, their power to help, are alike primitive and general conceptions. Cf. Jacobsen, op. cit., p. 21; Rohde, op. cit., p. 533; Frazer, J. G., The Fear of the Dead in Primitive Religion, vols. I–II (1933–34).Google Scholar

I am disinclined to see in the θεαὶ ψνχαί of the Dura text the presence of Iranian or of Semitic religious beliefs, which offer certain tempting parallels. There is no indication of other influences than those of the Greco-Roman religious koine. The “Goddess Psyche” of Iranian and Hellenistic myth (e.g., R. Reitzenstein, SB Ak. Heidelberg VIII [1917], no. 10) is something else altogether.

⁴³ Rostovtzeff, Dura-Europos and its Art, pp. 50–54.

⁴⁴ It is probably unnecessary to devote much space to a discussion of the names of Julius Terentius and his wife, Aurelia Arria. Under the Empire we have often occasion to remember the remark of Plutarch (Vita Marii, 1) on this subject: εἰς μὲν οὖν ταῦτα πολλὰς δίδωσιν ἐπιχειρήσεις ἡ τῆς συνηθείας ἀνωμαλία. No Arrii nor Terentii have occurred hitherto at Dura (note however the Ter in Rep. VI, p. 488, no. 816, and Arrianus in Rep. IV, p. 87, no. 203, possibly also Rep. VI, p. 210, no. 719), and the tribune and his wife are outsiders, as is natural. Their second gentilicia had been adopted, according to the usual practice, in honor of the imperial family and certainly with its permission, Aurelia from Caracalla (or Heliogabalus, or Alexander), Julius from Maximinus (or Julia Domna).

⁴⁵ Published by C. Hopkins, Rep. VI, pp. 146–167; cf. Rostovtzeff, Dura-Europos, pp. 94 f.

⁴⁶ Cf. for example Meissner, B., Babylonien und Assyrien I (1920), p. 426.Google Scholar The practice may have existed also in primitive Greece (Rohde, op. cit., p. 166), but in classical times even city burial was reserved for a formally divinized hero of some sort (Rohde, op. cit., p. 128; Pfister, op. cit., pp. 445–465; Jacobsen, op. cit., vol. I, pp. 14, n. 1, 18–20, 75). Terentius was hardly this.

⁴⁷ Cf. Rostovtzeff, op. cit., pp. 55 f.

⁴⁸ Not unreasonably to be connected with the earthquake of A.D. 160 (Rep. II, p. 86, no. H 2 = SEG VII, no. 360). It is natural to suppose that the city was cleaned up by the Romans after their occupation. They would be little concerned with the sanctity of the necropolis, and were doubtless in haste to dump refuse as quickly as possible in the nearest spot outside the walls. [Cf. the legal maxim sepvlchra hostium religiosa nobis non sunt (Paulus ap. Dig. XLVII 12.4). Roman jurists went further: cf. Pomponius ap. Dig. XI 7.36 cum loca capta sunt ab hostibus, omnia desinunt religiosa vel sacra esse, sicut homines liberi in servitutem perveniunt: quod si ab hac calamitate fuerint liberata, quasi quodam postliminio reversa pristino statui restituuntur, Rose, H. J., Class. Quart. XXIV (1930), 134 f. — A. D. N.]Google Scholar

⁴⁹ Two names scratched over loculi, Mαθβαλάτ and Аὐρέήλιος KOϒ∑. These will be published in Report IX.

⁵⁰ It is not easy to think of a cenotaph, in view of the verb θάψε. This need not involve inhumation or even deposit of ashes in a tomb, but certainly means that regular funerary ceremonies took place. While the diwan may have constituted a heroon without a tomb, I cannot understand the epitaph apart from the grave. [ἐλαϕρὰ καλύψαι δὲ γαῖα, unless verbiage, suggests that the remains of the dead man were very near. Is it possible that the widow of Terentius decided to leave Dura, and deposited his ashes in what had been their house? — A. D. N.]

⁵¹ Rostovtzeff, , Gesellschaft und Wirtschaft im römischen Kaiserreich (1931), II, pp. 137–141.Google Scholar

⁵² For the units of the garrison cf. Rostovtzeff, Dura-Europos, pp. 24–26; Rep. VII/VIII, pp. 86 f.

⁵³ For their identification cf. most recently Seyrig, H., Syria XIII (1932), pp. 190–195Google Scholar; C. Hopkins, Rep. VII/VIII, pp. 365–367.

⁵⁴ Cumont, Fouilles, pp. 89–114, pls. XLIX–LI; Rostovtzeff, Dura-Europos, pp. 71–74.

⁵⁵ Cumont, Fouilles, p. 363, no. 8a.

⁵⁶ Cf. Rostovtzeff, Papyri und Altertumswissenschaft (Münchener Beitrage zur Papyrusforschung und antiken Rechtsgeschichte XIX, 1934, pp. 367–370); Rep. V, pp. 296 f., 299.

⁵⁷ Cumont, Fouilles, p. 363, no. 8b; cf. ibid. p. 113.

⁵⁸ Especially Moqimu, which is common. Taime occurs more properly spelled (Θαίμη) on a fragment of painted wall from the Temple of the Gaddé (Rep. VII/VIII, p. 283, no. 913b), and with the spelling Θαῖμος in the southwest tower of the Main Gate (Rep. V, p. 23, no. 394).

⁵⁹ Cumont, Fouilles, p. 376, no. 14; cf. p. 113. Cumont, knowing less of the Dura garrison than is now known, understood legionarius here in the loose general sense of “military,” and this may well be right, although the presence of legionary detachments in the city makes it possible to take the term in a more technical sense. For the priests of the 3rd Gallic and 1st Illyrian legions in Egypt in the fourth century, cf. J. G. Milne, Greek Inscriptions (Cat. Gen. d. Ant. Egypt, de Musée du Caire, vol. XXIII [1905], p. 45, no. 9238b) (A.D. 323); Rostovtzeff, Yale Class. Stud. V (1935), p. 247, n. 123. In Egypt, however, as Professor Nock reminds me, there was a priestly class, in Syria only priests of individual temples.

⁶⁰ There is a certain difficulty here which must not be minimized. Mr. A. S. Hoey points out that the fresco represents an act of worship in honor of Palmyrene gods who have no place in the official military calendar, the Feriale Duranum (DP 2, now in Yale Class. Stud. vol. VII). He feels that Themes cannot have been an official chaplain of the cohort, but merely have represented it in the worship of its native gods just as others properly qualified might offer worship in its behalf to Mithras or Jupiter Heliopolitanus or other popular deities not officially recognized. He calls attention to two priests of Jupiter Dolichenus in contemporary texts, cited by Domaszewski, Die Religion des römischen Heeres (1895), p. 64. They are as follows: W. Brambach, Corpus Inscr. Rhen. (1867), 645 = CIL XIII, 7786, A.D. 250 (Remagen): In h(onorem) d(omus) d(ivinae) Arcias Marinus sacerdos Dolicheni donum donavit equitibus C(o)hortis I F(laviae); W. Henzen, Ann. Inst. Corr. Arch., LVII (1885), p. 290, no. 38 = CIL VI, 31181 (Rome): Soli Invicto pro salute Imp(eratorum) et Genio n(umeri) eq(uitum) sing(ularium) eorum M. Ulp(ius) Chresimus sace[rd(os)] Iovis Dolich[eni] v(otum) s(olvit) l(ibens) l(aetus) m(erito). In his opinion, none of these priests were in any way official. The question is obscure, but in my opinion Themes' mention with the officiales of the cohort does attest his official position, and justifies the view expressed in the text. A full exposition of Mr. Hoey's views is presented in the T.A.P.A. LXX (1939), pp. 456–481.Google Scholar

⁶¹ “Die Rangordnung des römischen Heeres” (Bonner Jahrbücher CXVII (1908)), pp. 58 f.Google Scholar

⁶² Cf. most recently Ensslin in Cam. Anc. Hist. XII (1939), pp. 69, 86 f.

⁶³ Herodian, VI, 5/6.

⁶⁴ Rep. VI, pp. 433–435.

⁶⁵ It is desirable to take this occasion to correct a mis-reading of this text which was published in Rep. VI. The reading there given, ἐν καύνῃ πρὸ[ς] χειμασίᾳ σπείρης, κτλ., was based on the theory that καύνη could be a phonetic development from canabae, for which the late variant cabanae is found. This is now clearly impossible, and the true reading appears from a comparison with DP 101, of A.D. 227 (Arch. Hist. du Droit Orient., I [1937], pp. 261–288; Rep. VII/VIII, pp. 433–441): ἐν ∑αχάρη παραχειμασίᾳ as follows: ἐν Kάτνῃ παρ[α]χειμασίᾳ. No Qatna seems otherwise known in this region (R. Dussaud, Topographie Historique de la Syrie [1927], Index), but the papyrus cannot have been brought from far.

⁶⁶ Herodian, VI, 6, 5/6; cf. id., VII, 8, 4. The unreliability of Herodian was pointed out long ago by H. Schiller, Geschichte der römischen Kaiserzeit, I, 2 (1887), p. 781, n. 1.

⁶⁷ Ensslin, op. cit., pp. 86 f.

⁶⁸ Rep. VII/VIII, pp. 172 f., no. 876; cf. p. 179.

⁶⁹ Rep. IV, pp. 112–114, no. 233; cf. Rostovtzeff, Dura-Europos, p. 27. It is immaterial whether ὑμῶν be interpreted as “you,” that is, the Romans or the Roman garrison, or as miswritten for ὑμῶν, “us.” Even as a Roman citizen, Nebuchelus may not have been reconciled to the life of Dura as a Roman garrison town. The presence of so much soldiery in a small place must have disrupted completely the normal life of the citizens and placed all civilians in a more or less uncomfortable situation. There is no indication among Nebuchelus' accounts that he lived by selling to the troops of the garrison, and he need not have regarded the “descent” upon the arrogant westerners with entire dissatisfaction.

⁷⁰ Herodian, VI, 2, 5; cf. Rep. IV, p. 114.

⁷¹ The standard view is that dux was not used as a regular title before the time of Diocletian (Seeck in R.E. V [1905], 1869), that is, dux ripariensis or limitaneus. Earlier the term is applied to army leaders (from the reign of Marcus); cf. Schiller, Gesch. d. röm. Kaiserzeit, I, 2, pp. 769, 773, 819, 869; II, p. 93. The title at Dura was δοὺξ τῆς ῥείπης, a clear anticipation of the Diocletian institution, to the development of which Aurelian also contributed. Cf. Rostovtzeff, Dura-Europos, pp. 27 f. We have the names of four of these duces. One, Domitius Pompeianus, is known from undated texts from the palace in the northeast corner of the city (Rostovtzeff, op. cit., pp. 28, 52, and pl. X, 1). Another, Julius Julianus, is mentioned in two dedications from the Dolicheneum (Rostovtzeff, op. cit., p. 25), one of which is dated in the Seleucid year 562 (A.D. 249–50). He was previously known from a text from Bostra dating from Philip's reign (Dessau, 2771; cf. id., Prosop. Imp. Rom., II [1897], p. 197, no. 243. Stein in the R.E. X (1919), 653, no. 286, adds nothing) where he was praefectus Legionis I Parthicae, and is called dux devotissimus. This may have been his station prior to becoming dux ripae at Dura. The other two holders of this office known to us are named in the horseregister DP 3, verso (Rep. V, p. 297), Ulpius Tertius as of the year 251 and Aulus Licinnius Pacatianus in 245. The last named has the earliest date, and may very well have been the first of the series. It is now clear, as was noted by Mr. Brown, that he is the person honored in the inscription on the column found just west of the city arch, and published by Mr. A. M. G. Little, Rep. IV, pp. 72–74, no. 169. It was restored as follows by the editor: [Tὸν] λαμπρό | [τα]τον ὑπατι | [κὸν Kλαύδι | ον Σολλέμ] | νιον Πακατι | ανὸν Ἀντίγον | ος Mαρίωνος | ἀρχιερεùς τὸν | [εὐεργέτην]. (This was the legatus pro praetore provinciae Arabiae under Alexander Severus [E. Groag, A. Stein, Prosop. Imp. Rom., II², 1936, p. 250, no. 1030] who had no known connection with Dura.) It is now legitimate to restore the beginning as follows: [Tὸν] λαμπρό | [τα]τον ὑπατι | [κὸν καì δοῦ | κα A. Λιμίν |]νιον. Later, of course, the duces as other military commanders were chosen from the equites, but the same need not have been true at this time. The office of the dux developed out of the practice of combining several armies under a single extraordinary commander with this title. Such a post may have been held under Alexander by Maximinus, if it is to this that he refers in his speech to the army in Herodian VII, 8, 4: Πέρσαι … ἡσυχάζουσιν … πείρᾳ τῶν ἐμῶν πράξεων, ἃς ἔγνωσιν ὅτε τῶν ἐπὶ ταῖς ὄχθαις στρατοπέδων ἡγούμην. The “deeds” performed by Maximinus when commander of the legions (or “camps”) “on the banks” must have been accomplished during Alexander's campaign of A.D. 232. Later, according to the same Herodian, there was peace. I do not believe that Maximinus was ever an administrative dux ripae, even if, as is possible, στρατοπέδων here means castra and not legiones.