¹ The photographs (Pls. XIII and XV) were provided by Dr. Dimitrios Pandermalis of the Archaeological Institute of the University of Thessaloniki. It was through him and his good offices that I obtained permission to publish from Mrs. Ch. Koukouli, of the Archaeological Museum of Kavalla, as well as information on the circumstances and find-spot from Mrs. E. Juri, of the Archaeological Museum of Thessaloniki. I gratefully acknowledge my indebtedness to these Greek scholars. Above all, I would like to express my gratitude to Dr. H. Nesselhauf who first directed my attention to the monument and who made valuable suggestions for the commentary.

² Actually Decebalus committed suicide to avoid being captured alive (Dio 68, 14). Yet since he fell into the hands of the Romans as he was dying, official imperial propaganda, just like our inscription, held he was captured, cf. SEG IX, 101 : cp. Reynolds, J., ‘Four Inscriptions from Roman Cyrene’, JRS XLIX (1959) 96, n. 11.Google Scholar

³ The exact measurements are as follows: height, 2·64 m; width at top, 0·83 m; width at bottom, 0·90 m; depth, 0·26–0·28 m; upper relief panel, 0·57 × 0·37 m; lower relief panel, 0·42 × 0·31 m; height of letters in line 1, 10 cm; line 2, 9·5 cm; lines 3–10, 6·3 cm; lines 11–23, 4·5 cm.

⁴ At the end of line 11 the word FACTVS was repeated erroneously, then erased and replaced by the letters DVPLI. Because of the pitted surface the stops are not always easily distinguishable.

⁵ When awarded citizenship during their period of service, soldiers tended to adopt the name of the ruling emperor. Cf. Wagner, W., ‘Zur ala Pansiana, eine epigraphische Nachlese’, Germania 41 (1963), 317–27Google Scholar, esp. p. 318 f. Also Kraft, K., Zur Rekrutierung der Alen und Kohorten an Rhein und Donau (Dissertationes Bernenses 1, 3 (1951)), 101 f.Google Scholar; Speidel, M., Die Equites Singulares Augusti (Antiquitas I, II (1965)), 66.Google Scholar

⁶ Grammeni, where the tombstone was found, apparently belonged to the ager of Philippi, as did the neighbouring Prossotsani, cf. Collart, P., Philippes, Ville de Macédoine (École française d'Athènes, travaux et mémoires, fasc. 5, 1937), pl. XXXIV and pp. 276 ff.Google Scholar For veterans of the auxilia see Nesselhauf, H., CIL XVI, 195.Google Scholar For the legions: Forni, G., Il reclutamento delle legioni da Augusto a Diocleziano (Milan, 1952), 41 and 145–51.Google Scholar

⁷ Schmidt, J., RE XIX, 2233 f.Google Scholar, s.v. Philippoi. As shown by a wealth of inscriptions, even the rural communities in the region of Philippi spoke Latin, cf. Collart l.c., n. 6.

⁸ Legionaries from Philippi: CIL III, 2717; 14933; 14507, I, a, (VII Cl.). CIL III, 2031 (XI Cl.). CIL III, 5636; IX, 4684 (VIII AUG.). AE 1934, 62 (VI Ferr.). All except CIL III, 14057 are from the first century, according to Forni, Reclutamento 165, 175. Decorated with dona militaria from Philippi: Dessau, 2127 and AE 1933. 87.

⁹ For evidence, see Patsch, C., Der Kampf um den Donauraum unter Domitian und Traian (Sb. Akad. Wien, phil.-hist. Kl. 217, 1 (1937)), 27 ff.Google Scholar; Syme, R., CAH XI, 172.Google Scholar

¹⁰ It seems that only the dona received in the Parthian War are not mentioned in chronological sequence, doubtless to avoid repeating et ab eodem ob virtutem donis donatus in bello for the third time.

¹¹ Vegetius 2, 21—a general statement to this effect. See also the excellent discussion of this rule by Gilliam, J. F., ‘Dura Rosters and the Constitutio Antoniniana’, Historia 14 (1965), 74–92Google Scholar and the same, ‘An Egyptian Cohort in A.D. 117’, Bonner Historia-Augusta-Colloquium 1964/65 (Antiquitas IV, 3 (1966)), 91–7.

¹² Other detailed careers mention the positions of miles and discens equitum, e.g. Dessau, 9090; CIL VI, 3409; CIL XI, 5646 = Dessau, 2081; CIL XIII, 2602 = Dessau, 2118.

¹³ SHA, Max. 2, 2Google Scholar; Herodian 6, 8, 1.

¹⁴ This theory was first put forward by Marquardt, J., Römische Staatsverwaltung (3rd ed., 1881) vol. 2, 456Google Scholar, and was widely accepted; cf. e.g. recently Neumann, A. in Der Kleine Pauly III (1968), 541 f.Google Scholar, s.v. legio.

¹⁵ cf. Wilkes, J., Dalmatia (1969), 96, n. 1.Google Scholar

¹⁶ Forni, Reclutamento 76 ff. assigns Macedonia to the eastern recruiting district, yet the dividing line was by no means rigid; Macedonians served in the legio VII Claudia even while it was still stationed in Dalmatia, cf. Wilkes 106 f., 464 ff.

¹⁷ BGU 600 lists an eques as well as an optio who is not called eques, both in the same century. Other indications are CIL VIII, 2568 (as explained by Breeze, D., ‘The Organization of the Legion’, JRS 59, 1969, 50–5)Google Scholar, but it may not refer to centuries; and P. Gen. Lat. 1, verso 5 (= Daris, , Documenti per la storia dell'esercito romano in Egitto (Milan, 1964) no. 10)Google Scholar, but it may refer to an auxiliary, not a legionary cohort, cf. Forni, Reclutamento 32 ff. (contra: Marichal, R., ‘La solde des armées romaines d'Auguste à Septime-Sévère’, Ann. de I'Institut de Phil. et d'Hist. Orient, et Slaves, 13, 1953, 399–421Google Scholar).

¹⁸ Vegetius 2, 20; PSI 1003 = Daris no. 33, cf. J. F. Gilliam, ‘An Egyptian Cohort in A.D. 117’ (see note II), 91–7. Also Davies, R. W., ‘A Note on Lorictitis’, BJb. 168 (1968), 161–5.Google Scholar

¹⁹ Alae, auxiliary cohorts and numeri had treasuries called quaesturae, administered by a quaestor: P. Berl. 6, 866 = Daris, Documenti 35, 49; P.Dura 60; AE 1912, 5; AE 1950, 16; CIL III, 798 = Dessau, 2429; CIL III, 1369 = Dessau, 2630. In one case, it seems, horses were bought with money from the quaestura: P.Dura 97. cf. Davies, R. W., ‘The Supply of Animals to the Roman Army and the Remount System’, Latomus 28 (1969), 429–59Google Scholar, esp. p. 448 f. Perhaps quaestor is but a different name for the post otherwise known as curator, for which see Birley, E., Libyca 2 (1954), 123 f.Google Scholar

²⁰ Tabularium, cf. AE 1957, 85. The actarius in CIL II, 2663, it seems now, belonged to the office of the equites legionis, not to the office of the legatus legionis. Contra: von Domaszewski, A., Die Rangordnung des römischen Heeres (second ed. by Dobson, B., 1967 Beihefte der Bonner Jahrbücher 14), 39.Google Scholar

²¹ CIL VIII, 2532, B = Dessau, 2487; Arrian, Ektaxis 5; Josephus, Bell. Jud. 3, 120; Tacitus, Ann. 4, 73.

²² D. Breeze, 1.c., assumes they could have done without, which is certainly impossible.

²³ Their commander cannot have ranked lower than a centurion since their exercitator apparently was a centurion, see AE 1965, 223 with the commentary by Fitz, J., ‘La Carrière de P. Aelius Proculinus’, Latomus 24 (1965), 565–75.Google Scholar Domaszewski (Rangordnung 47) thought of the tribunus sexmestris as commander of the equites legionis, but the fragmentary inscription CIL II, 5682 he quotes proves nothing, and had it been such a high officer, Arrian (Ektaxis 5) would have mentioned him. As for Statius, Silvae 5, 1, 94, adduced by Domaszewski in this context, it probably refers to a legionary centurion, not to a tribune, cf. Birley, E., ‘Promotions and Transfers in the Roman Army II: The Centurionate’, Carnuntum Jahrbuch 8 (1963/1964), 21–33Google Scholar, esp. p. 22. Perhaps, CIL XIII, 8648Google Scholar = Dessau, 2244 should be read [(centurioni)] eq(uitum) leg(ionis); for a recent discussion of this stone see John, W., RE XXIV, 924 f.Google Scholar, s.v. P. Quinctilius Varus.

²⁴ Domaszewski, A. V., ‘Die Religion des römischen Heeres’ (Westd. Zeitschr. 14 (1895), 89Google Scholar) assumed a legionary commander needed equites legionis, precisely because he had no singulares. In his Rangordnung p. 40, he ascribed the singulares to the tribunus laticlavius and made him mounted for that reason, but this is now proved to be wrong.

²⁵ Equites legionis as guards: Josephus, Bell. Jud. 3, 120. Arrian, Ektaxis 5; for singulares, cf. M. Speidel, Die Equites Singulares Augusti 87 ff. Both naturally may have served as messengers (for singulares, see P.Oxy. VII, 1022 = Daris, Documenti 4; Lydus, De mag. 3. 7). I cannot see, however, how Livy 37, 7 would show that the main purpose of the imperial equites legionis was to serve as mounted messengers, as D. Breeze, l.c., assumes.

²⁶ Josephus, , Bell. Jud. 3, 120.Google Scholar

²⁷ Horovitz, Ph., ‘Essai sur les pouvoirs des procurateurs-gouverneurs’, Revue Belge 17 (1938), 53–62Google Scholar, esp. p. 58 and Pflaum, H. G., Les procurateurs équestres sous le haut-empire Remain (1950), 126.Google Scholar

²⁸ Pflaum, H. G., ‘A propos de la date de création de la province de Numidie’, Libyca 5 (1957), 61–75.Google Scholar

²⁹ AE 1957, 122 on which Pflaum based his conclusions, mentions equites singulares, not simply singulares, which does not, however, seem to make any difference for this question.

³⁰ CIL III, 4061 = Dessau, 2330; XIII, 6948; VIII, 16549 = Dessau, 2329; VIII, 2562 (?); AE 1957, 341;BGU 600.

³¹ Domaszewski, o.c. (see n. 24) 88, commenting on CIL VIII, 2562.

³² Thus Premerstein, , ‘Die Buchführung einer ägyptischen Legionsabteilung’, Klio 3 (1903), 1 ff.CrossRefGoogle Scholar, esp. p. 27; Zwikker, W., ‘Bemerkungen zu den römischen Heeresfahnen in der älteren Kaiserzeit’, in 27. Bericht der römisch-germanischen Kommission (1937), 7–22Google Scholar, esp. p. II. There is, however, no evidence for this, not even among the equites praetoriani, as Domaszewski (Rangordnung 23, n. 9) and Durry, M. (Les cohortes prétoriennes (1968), 99)Google Scholar assume.

³³ Like the four vex(illarii) belonging to one turma of the cohors XX Palmyrenorum, P.Dura 100, col. 38 and 39.

³⁴ For turmae and decurions of the legionary cavalry, see Passerini, A., Dizionario Epigraphico 4, 610Google Scholar, styling CIL XI, 1526 ‘di dubbia autorità’. The vexillum mentioned in CIL XIII, 8276 = Dessau, 2324 seems to be a unit of veterans rather than legionary cavalry, cf. Premerstein, l.c. (see n. 32) 27, n. 7.

³⁵ CIL VI, 32709a = Dessau, 9190.

³⁶ Rangordnung 47. The fragmentary inscription CIL II, 5682 quoted there proves nothing. The optio equitum may have done duty at the tabularium (AE 1957. 85), compare CIL VIII, 18072 = Dessau, 2446; XIII, 5970 = Dessau, 2444.

³⁷ cf. Ritterling, E., RE XII, 1621 f.Google Scholar, s.v. legio.

³⁸ CIL III, 1193 = Dessau, 2746; III, 6395 = Dessau, 2665 (‘ob triumphos belli Dacici’); III, 14416 = Dessau, 7178; VI, 31856 = Dessau, 1327; VIII 217 = Dessau, 2658; XI, 395 = Dessau, 2648. One may speculate whether a good many higher officers did not receive dona more or less honoris causa as in CIL III, 14187, 4 = Dessau, 4081.

³⁹ e.g. CIL V, 7495 = Dessau, 2337. For the form ob virtute see CIL III, 1193 = Dessau, 2746.

⁴⁰ In A.D. 20, however, the proconsul of Africa still had the right to award dona, cf. Tac, , Ann. 3, 21.Google Scholar

⁴¹ cf. Steiner, P., ‘Die dona militaria’, Bonner Jahrbücher 114–15 (1906), 1–98Google Scholar, esp. p. 89. For dona earned in other campaigns: CIL XI, 395 = Dessau, 2648 (Claudius/Nero); CIL V, 7425 = Dessau, 2720 (Nerva).

⁴² cf. Suetonius, de Gramm. 9, ‘corniculo, moxequo meruit’; Caesar, , BG I, 23.Google Scholar Domaszewski, Rangordnung 54.

⁴³ CIL III, 8438;V, 522; 8185; VIII, 2354; AE 1917/18, 74/75; P. Mich, III, 164, etc. See also above, n. 42 and below, 44. cf. Gilliam, J. F., ‘The Appointment of Auxiliary Centurions’, TAPA 88 (1958), 155–68.Google Scholar

⁴⁴ Equites legionis: CIL III, 7449;647 (?); P.Mich. III, 164(2). Stratores: CIL VIII, 9002 (?), 9370; X, 7580; AE 1958, 156. Beneficiarii: CIL VIII, 17619; II, 2554, with AE 1910, 1. Cornicularius: AE 1917, 74. It is ironical that Domaszewski's contention (Rangordnung 54), that they all must have been former ‘tactical ranks’ of the legions, is borne out by none of the inscriptions he cites, but by ours that he could not know. For analogies in the legions cf. Birley, E., ‘Promotions and Transfers in the Roman Army II: the Centurionate’, Carnuntum Jahrbuch 8 (1963/1964), 21–33Google Scholar, esp. p. 22.

⁴⁵ For the training of the equites legionis, see Speidel, o.c. (n. 5), 59.

⁴⁶ Suggested by Domaszewski, Rangordnung 72. But equites singulares Augusti and equites singulares consularis, generally of non-Roman origin, are trained and promoted in the same way, as I hope to show in a study on the decurions. See also Cheesman, G. L., The Auxilia of the Roman Imperial Army (1914), 38.Google Scholar

⁴⁷ cf. Domaszewski, Rangordnung 69 and CIL VIII, 8438 = Dessau, 2597.

⁴⁸ AE 1917–18, 74; BGU II, 696 ( = Daris, Documenti 9), 16–18. P. Mich. III, 164 ( = Daris, Documenti 27), 5–7.

⁴⁹ Ritterling, E., RE XII, 1558Google Scholar, s.v. legio. E. Birley, o.c. (n. 23), 26.

⁵⁰ For Troianus, CIL XIV, 3626 = Dessau, 2742;RIB 2263 = CIL VII, 1163; JRS LVI (1966), 219, no. 9. For the omissions, see Kneissl, P., Die Siegestitulatur der römischen Kaiser (1970), 89 f.Google Scholar

⁵¹ Domaszewski, Rangordnung 71, assumes there were only four pay-grades in the legions below the centurion, the vexillarius being the second lowest. The papyri of the auxilia hint at there being only three pay-grades below the decurion, i.e. the duplicarius, the sesquiplicarius and basic, cf. Gilliam, J. F., ‘The Moesian Pridianum’, in Hommages à A. Grenier (Collection Latomus 58, Brussels, 1962), 747–56.Google Scholar

⁵² Without question assumed by Domaszewski, Rangordnung 72; Cheesman, Auxilia 35; Webster, G., The Roman Imperial Army (1969), 260Google Scholar; Watson, G., ‘The Pay of the Roman Army, The Auxiliary Forces’, Historia VIII (1959), 372–8Google Scholar, esp. P. 377; Brunt, P. A., ‘Pay and Superannuation in the Roman Army’, PBSR 28 (1950), 50–71Google Scholar, esp. p. 64.

⁵³ One-third was suggested by Domaszewski, , ‘Der Truppensold der Kaiserzeit’, Neue Heidelberger Jahrbücher 10 (1900), 225Google Scholar; R. Marichal, l.c. (n. 17); Watson, l.c. For five-sixths: Johnson, M. A. C., Roman Egypt (Baltimore, 1936), 670, 673Google Scholar; Passerini, A., Le coorti pretorie (Rome, 1939), 101, n. 2Google Scholar; Forni, o.c. (n. 6), 32 ff.

⁵⁴ Watson, l.c. We assume a duplicarius alae really got double pay, as is probable from Arrian, Tact. 41.

⁵⁵ cf. Domaszewski, Rangordnung 49 and 71. For the rank of vexillarius, see above. In the cohors XX Palmyrenorum at Dura-Europos the vexillarius, though not a permanent post, is reckoned among the paygrades of the duplicarii, cf. R. O. Fink in: The Excavations at Dura-Europos, ed. by Perkins, A., Final Report V. 1: The Parchments and Papyri (New Haven: Yale Univ. Press, 1959), 32 f.Google Scholar

⁵⁶ Vegetius 2, 21. Equites had to pay, at least in part, for their horses, see R. W. Davies, ‘The Supply of Animals’ (above n. 19), 448.

⁵⁷ cf. the works cited in note 53. Brunt, o.c., assumes three-fifths: that, too, could be reconciled with Maximus' career, but hardly with AE 1967, 410 (a legionary promoted to signifer numeri).

⁵⁸ Nesselhauf, , CIL XVI, 35, p. 33Google Scholar, n. 1.

⁵⁹ Wagner, , Die Dislokation der römischen Auxilarformationen (1938), 60 ff.Google Scholar

⁶⁰ CIL XVI, 163 with the surname Veterana added, cf. Nesselhauf's remarks CIL XVI, p. 45.

⁶¹ IGRR 1, 824 mentions a former prefect of the ala II Pannoniorum who had earned dona in Dacia and, it seems, Parthia; if he did so while commanding this regiment (contra: Domaszewski, Rangordnung 138), then the entire ala, not just a vexillation, must have been in the eastern war.

⁶² For the list of auxilia known so far as participating in that war, Paribeni, R., Optimus Princeps (1927) 11, 287 f.Google Scholar

⁶³ CIL VI, 32933 = Dessau, 2723. See also IGRR III, 173 for a possible stay in Ancara.

⁶⁴ Fronto, Princ. hist. 9 ( = C. R. Haines' edition (1920) II, 204).

⁶⁵ It is possible, though, that Maximus got his missio from the commander there, while still belonging to the exercitus Daciae, cf. Nesselhauf, , CIL XVI, 99.Google Scholar

⁶⁶ It is attested there from A.D. 143 onward, cf. AE 1906, 112; CIL XVI, no (see Wagner, Dislokation 38 f.); XVI, 185. For the identity with the ala II Gallorum et Pannoniorum, see Daicoviciu, C., ‘Un nouveau diplome militaire de Dacia Porolissensis’, JRS 51 (1961), 63–70Google Scholar, esp. p. 66, n. 29.

⁶⁷ SHA, Hadr. 6, 6.Google Scholar

⁶⁸ A decrease was assumed by Stein, A., Die Reichsbeamten von Dazien (Diss. Pann. Ser. I, 12 (1944)), 18.Google Scholar No decrease: Syme, R., JRS 36 (1946), 159–68Google Scholar (review of Stein, Dazien), and Forni, G., ‘Contributo alia storia della Dacia Romana’, Athenaeum, N.S. 36 (1958), 183–218Google Scholar, esp. p. 197.

⁶⁹ P.Dura 100 (A.D. 219) of the cohors XX Palmyrenorum lists more than 15 exploratores, among them at least five foot-soldiers. See also P.Brit.Mus. 2851 (Hunt's Pridianum) ed. Fink, R. O., JRS 48 (1958), 102–16Google Scholar (a.d. 105/6) of the cohors I Hispanorum, detaching several horsemen under a centurion exploratum For an explorator in an ala see CIL VIII, 21516.

⁷⁰ Arrian, Ektaxis I, 1. Their commander seems to have been a praepositus, since Arrian mentions ordinary centurions by their proper title. The centurions in CIL III , 3254 and 3648 = 10422 may have led the exploratores of a smaller force. Centurions, of course, could command mounted units; contra: Fiebiger, , RE VI, 1690–3Google Scholar, s.v. exploratores. See also Caesar, , BG I, 21 and 22.Google Scholar

⁷¹ Arrian, , Ektaxis I, 1.Google Scholar ‘Hyginus’, De mun. castr. 24. cf. Vegetius 3, 6.

⁷² ‘Hyginus’, 30. This work belongs to the second century, cf. Speidel, o.c. (n. 5) 12 f. Birley, E., ‘The Epigraphy of the Roman Army’, Actes du deuxième congrès international d'épig. grecque et latine, Paris, 1952, 227–39Google Scholar, esp. p. 234.

⁷³ Vegetius 3, 6. See also Dig. 49, 16, 3 and 5, providing especially severe punishment for their failing in service; cf. Fiebiger, l.c.

⁷⁴ The first known instance is from around A.D. 172 (AE 1956, 124). See Stein, E., Die kaiserlichen Beamten und Truppenkörper im römischen Deutschland unter dem Prinzipat (1932), 233 ff.Google Scholar The remarks of Callies, H., ‘Die fremden Truppen im römischen Heer des Prinzipats’, 45. Bericht der Römisch-Germanischen Kommission (1964), 130–227Google Scholar, are here lacking in clarity.

⁷⁵ cf. P.Br.Mus. 2851 (see above, n. 69); P.Dura 100, 101. Also inscriptions such as CIL XIII, 8185 = Dessau, 4743: ‘Ulpius Acutus, du[p(liciarius)] al(ae) Sulp(iciae), sing(ularis) co(n)s(ularis).’

⁷⁶ Ritterling, E., ‘Zur Erklärung von Arrians Ἔκταξις κατ' Ἀλανῶν,” Wiener Studien 24 (1902), 359–72Google Scholar, esp. p. 370. Stein, Beamten 235. Rowell, H. T., RE XVIII, 1327 ff.Google Scholar, s.v. numerus, was right.

⁷⁷ cf. ‘Hyginus’, 30; contra: Stein, Beamten 236.

⁷⁸ The word bis in this line does not mean twice in the Dacian War as well as twice in the Parthian War: in such a case one would expect the word quater. cf. CIL II, 4461 = Dessau, 2661: ‘ter donis donato ab imp. Traiano torquibus, armillis, phaleris, corona vallari bis in Dacia, semel in Parthia.’

⁷⁹ cf. Büttner, A., ‘Untersuchungen über Ursprung und Entwicklung von Auszeichnungen im römischen Heer’, Bonner Jahrbücher 157 (1957), 127–80Google Scholar, pl. 13, fig. 2.

⁸⁰ P. Steiner, o.c. (n. 41), esp. p. 22; 28.

⁸¹ CIL XI, 5646 = Dessau, 2081, an eques praetorianus.

⁸² Domaszewski, Rangordnung 68, assumed they were discontinued by Hadrian. New finds, however, may show that they were not discontinued, especially since centurions continued to be awarded phalerae (e.g. CIL III, 7334 = Dessau, 2080; CIL VIII, 217 = Dessau, 2658) but only that almost no dona are recorded for men of the ranks after Trajan's reign. Other instances where torques and armillae were awarded without phalerae: CIL V, 4365 = Dessau, 2272; CIL XIII, 2230 = Dessau, 2313; Dessau, 9492.

⁸³ cf. CIL VIII, 2354 = Dessau, 305; transfer seems to be involved in P.Mich. III, 164, 11, edited by J. F. Gilliam, o.c. (n. 45).

⁸⁴ cf. CIL III, 14416 = Dessau, 7178.

⁸⁵ CIL III, 7449; P.Mich. III, 164 ( = Daris, Documenti 27), 2 and 14.

⁸⁶ CIL VIII, 2354 = Dessau, 305, a close parallel to Maximus in career and time. For other decurions reaching the centurionate, cf. CIL II, 1681; III, 11213; VIII, 21567.

⁸⁷ He must have been literate, though, as is clear from his having been quaestor.

⁸⁸ In scenes LXXV; XCIII; CXXXV; CXLV as numbered by Cichorius, C., Die Reliefs der Trajanssäule (Berlin, 2nd. ed., 1927).Google Scholar cf. Panaitescu, E., ‘Il ritratto di Decebalo’, Ephemeris Dacoromanae I (1923), 387–413Google Scholar, esp. p. 407.

⁸⁹ For a recent discussion see Daicoviciu, H., ‘Osservazioni intorno alla colonna Traiana’, Dacia 3 (1959), 311–28Google Scholar; Turcan-Deleani, M., ‘Les monuments répresentés sur la colonne Trajane’, Mélanges d'arch. et d'hist. de l'école française de Rome 70 (1958), 147–76.Google Scholar

⁹⁰ That the artist was with Trajan in Transylvania has been suggested by Richmond, I. A., ‘Trajan's Army on Trajan's Column’, PBSR 13 (1935), 1–40Google Scholar, esp. p. 3. The other differences between the two pictures seem to be less important: on the tombstone Maximus wears the sagum, on the column not; here the King's shield is pointed on the long sides, there not. But why is the King's shield shown in both scenes? Only to mark him as a defeated warrior?

⁹¹ o.c. vol. 2, p. 361 (a remarkable inference, though he was referring only to the men to the right of the tree), cf. Vegetius 3, 6: ‘exploratores—cum equis probatissimis’.

⁹² As thought by Cichorius, o.c, p. 368 ff. If we retain his suggestion that it was one of the later legionary fortresses, Apulum is the likeliest: it changed its name, Mommsen, Th., CIL III, p. 183Google Scholar, and ‘Apulum’ may reflect Apollo Raniskelēnos. Yet this is no more than a guess. C. Patsch, Der Kampf um den Donauraum, 119 thinks of Praetoria Augusta as Trajan's last camp, and tentatively equates it with modern Inlăceni. Dr. C. Daicoviciu, in a letter, suggests Hoghiz (Heviz) and thinks of Ranisstorum as a civilian settlement near the camp of Praetoria Augusta.

⁹³ Detschew, D., Die thrakischen Sprachreste (Oest. Akad. Wiss., Schriften der Balkankommission 14, (1957)), 389Google Scholar ( = IGBR, IV, 1923). Comparison with other such names, like (IGBR 11, 540) shows that the -s- belongs to Ranis-, thus the double -s- in our inscription is fully justified, and the explanation of Tomaschek, W. (Die alten Thraker, II. Sprachreste (Sb. Akad. Wien, vol. 130, 1893), 68)Google Scholar that it is an ethnicon from Rani-skelos (thus quoted by Detschew) cannot be entirely correct.

⁹⁴ Detschew, Sprachreste 480.

⁹⁵ Cichorius, Reliefs 368 ff.; Petersen, E., Trajans dakische Kriege vol. 2 (1903), 119Google Scholar; Lehmann Hartleben, K., Die Trajanssäule (1926), 11.CrossRefGoogle Scholar

⁹⁶ The folds are recognizable above the heads of the soldiers who stand in front of the two men presenting the King's head. On the reliefs of the column, only the emperor and his comites wear the tunic together with the sagum. This interpretation can be mentioned here only in passing; detailed evidence will be set forth elsewhere.

⁹⁷ cf. CIL VI, 1444 = Dessau, 1022, ‘Imperator Caesar Nerva Traianus [Augustus Germanicus] Dacicus gentem Dacor(um) et regem Decebalum bello superavit’, an inscription sponsored by Trajan himself and therefore not just the historian's preoccupation with great leaders as it is found, e.g., in Pliny, , Ep. 8, 4Google Scholar or Eutropius 8, 6, 2 (‘bellum Decebali’).

⁹⁸ For the wide publicity given to the ‘capture ’ of Decebalus see SEG IX, 101, cf. above, n. 2.

⁹⁹ Dio 68, 14, 3; Fasti Ostienses A.D. 106: ‘Decebali [caput … in sca]lis Gemoni[is iacuit].’ The form Decebalus is found more often than the form Decibalus on Latin inscriptions, cf. Stein, A., PIR ² III, 4 f.Google Scholar Contra: Degrassi, A., Inscr. It. XIII, p. 226.Google Scholar

¹⁰⁰ cf. AE 1956, 124: ‘ab imp(eratore) Antonino Aug(usto) coram laudato et equo et phaleris et armis donato, quod manu sua ducem Naristarum Valaonem interemisset.’

¹⁰¹ cf. Livy 3, 57, 9: ‘emeritis etiam stipendiis pars magna voluntariorum ad nomina danda praesto fuere’. For a full discussion of the various meanings of the term voluntarius, see Kraft, Rekrutierung 87–95.

¹⁰² Alföldy, G., ‘Zur Beurteilung der Militärdiplome der Auxiliarsoldaten’, Historia 17 (1968), 215–27Google Scholar, makes the point that the actual duration of service in the auxilia was fixed at 25 years by Trajan in his later years. His evidence needs to be supplemented by the data known for the equites singulares Augusti (29 and 27 years as late as A.D. 138, cf. Speidel, o.c. (above, n. 5), 7) and for the cohors XX Palmyrenorum (27 and 26 years in A.D. 219, cf. R. O. Fink, o.c. (n. 55), 33) as well as by individual tombstones like CIL III, 14217, 34 (34 years) or Dessau, 9492 (eques legionis, 35 years).

¹⁰³ cf. Kraft, Rekrutierung 82–100. Kraft objected to Mommsen's theory that the cohortes voluntariorum were recruited by Augustus from former slaves (Macrobius, Sat. 1, 11, 32). One would wonder in that case why Augustus left them the same legacy as the legionaries received (Tac, Ann. 1, 8). Mommsen's view is upheld by Schillinger-Häfele, U., ‘Eine neue Inschrift aus Stettfeld, Ldkrs. Bruchsal’, Badische Fundberichte 22 (1962), 85–8Google Scholar, on the grounds that otherwise the cohortes ingenuorum could not be explained. See also Neumann, A., RE Suppl. IX, 886–90Google Scholar, s.v. voluntarii.

¹⁰⁴ See the list of cohortes voluntariorum in Cichorius, C., RE VI, 351–6Google Scholar, s.v. cohors.

¹⁰⁵ Mommsen-Nesselhauf, , CIL XVI, p. 161.Google Scholar

¹⁰⁶ CIL XVI, 57 = Dessau, 2004; XVI, 160; XVI, 163; CIL III, 1443; III, 1081 = Dessau, 3594. The date of his consulate is lacking. For a sketch of what is known about him see Groag, E., RE VA, 669–72.Google Scholar Also R. Syme, I.c. (n. 68). For the earlier career of Scaurianus, Syme adduces CIL XII, 3169, and Groag CIL III, 14387; see below, note 118.

¹⁰⁷ CIL XVI, 160, 163. That he was the first governor is probable, but for a possible doubt see Syme, I.e.

¹⁰⁸ Baur, P. V. C. and Rostovtzeff, M. I., The Excavations at Dura-Europos, Third Season, Preliminary Report (1932), 60Google Scholar = CRAI 1930, 272. cf. E. Groag, I.c.

¹⁰⁹ The only other Terentius of consular rank during Trajan's reign was Scaurianus' son, D. Terentius Gentianus, for whom see Groag, E., RE V A, 656–62Google Scholar, s.v. Terentius.

¹¹⁰ The reading No[ricum is impossible, since the remnant of the last letter on the stone cannot belong to an R. Nor was Noricum ever a consular province.

¹¹¹ For evidence see Hanslik, R., RE Suppl. x, 1035–1102Google Scholar, s.v. M. Ulpius Traianus.

¹¹² For the conquest and subsequent loss of these provinces see F. A. Lepper, Trajan's Parthian War (1948), with bibliography, and the review of that work by Henderson, M. I., JRS 39 (1949), 121–32.Google Scholar ‘Assyria’ apparently was Babylonia, annexed as a province, as is argued persuasively by Maricq, A., ‘La Province d'Assyrie, créée par Trajan’, Syria 36 (1959). 254–63CrossRefGoogle Scholar; cf. also Tacitus, , Ann. 2, 62.Google Scholar

¹¹³ CIL X, 8921 = Dessau, 1041; Dio 75, 9, 6 = Exc. Urb. 16. PIR ² II, p. 127 f., n. 558 (non vidi).

¹¹⁴ Thus Lepper, o.c, 208. For a doubt see Syme, R., Tacitus I (1958), 243.Google Scholar

¹¹⁵ AE 1927, 161. Gould, S., Dura-Reports IV, 56 ff.Google Scholar

¹¹⁶ See, e.g., A. Stein, Die Reichsbeamten von Dazien cit., p. 54. Dobo, A., Die Verwaltung der römischen Provinz Pannonien (1968), pp. 41 and 129Google Scholar, quoting the inscription Acta. Univ. Debr. VI (1959/60), 201 f. (Savaria) of L. Neratius Priscus, cos. (A.D. 103–6)—which shows, together with our inscription, that the title was already in use in Trajanic time.

¹¹⁷ Thus, Ritterling, E., ‘Die Statthalter der pannonischen Provinzen’ (Arch. -Ep. Mitth. 20 (1897)), 2–40Google Scholar, conceived of the generals under Agrippa or Tiberius as legati pro praetore exercitus.

¹¹⁸ Syria? See Groag, E., RE V A, V A, 669–72.Google ScholarCIL III, 14387 d = AE 1934, 177 could indeed refer to Scaurianus. The best edition of this inscription remains that by Premerstein, A. v., ‘C. Iulius Quadratus Bassus, Klient des jüngeren Plinius und General Trajans’ (SB Bayer. Akad. Wiss., Phil. -hist. Abt. 1934), 51 f.Google Scholar The edition by Rey-Coquais, J. P., IGLS VI, 2775Google Scholar, is seriously defective.

¹¹⁹ As L. Vitrasius Flamininus may have done, CIL X, 3870: ‘leg. pr. pr. Italiae Transpadanae et provinciae Moesiae superioris et exercitus provinciae Dalmatiae’; cf. Fitz, J., ‘Contribution à la carrière d'un proconsul d'Afrique’, Latomus 27 (1968), 45–74.Google Scholar

¹²⁰ Guey, J., Essai sur la guerre parthique de Trajan (1937) 78 f.Google Scholar, ingeniously interprets the coin legend Armenia et Mesopotamia in potestatem p. R. redactae, as referring to this status (more than capta, less than a provincia) with a reference to Strack, P., Untersuchungen zur römischen Reichsprägung I (Stuttgart, 1931), 223Google Scholar (non vidi). A less likely possibility would be that Mesopotamia was already lost, and its army on the retreat in Syria in A.D. 117. The two Germanies, before they were formally constituted as provinces by Domitian (see E. Stein, o.c. (n. 74), 10 f.), were likewise governed by legati consulares exercitus; the difference is, however, that Scaurianus is here (incorrectly?) called consularis exercitus provinciae.

¹²¹ Syme, R., ‘Governors of Pannonia Inferior’, Historia XVI (1965), 359.Google Scholar

¹²² Dio 68, 30. Fronto, ed. Haines 2, 22. Fronto, Princ. Hist. 16. Perhaps the name Scaurianus underlies the ‘mindestens sechs ziemlich schattenhafte Buchstaben’ read in Fronto, Princ. Hist., 16 as SANTRA by Hauler, E., ‘Zu Frontos Principia historiae’, Wiener Studien 38 (1916), 166–75.Google Scholar But cf. R. Syme, o.c., 347 for other available candidates. Scaurianus could even have been identical with the general Maximus of the First Dacian War who captured Decebalus' sister (Dio 68, 9, 4). If he lived on, Hadrian may have reckoned him among the hostile marshals, no longer entrusting him with a great command, just as his son, D. Terentius Gentianus, went without an armed province: CIL III, 1463 = Dessau, 1046. cf. Syme, R., Tacitus I, 245Google Scholar. n. 7.