Hostname: page-component-7479d7b7d-t6hkb Total loading time: 0 Render date: 2024-07-11T00:17:54.408Z Has data issue: false hasContentIssue false

Octavian in the Senate, January 27 B.C.

Published online by Cambridge University Press:  24 September 2012

W. K. Lacey
Affiliation:
University of Auckland, New Zealand

Extract

Reviewing Peter Sattler's Augustus und der Senat (1960), J. P. V. D. Balsdon remarked that the background to the events which marked the establishment of the principate is a field which has been ploughed, even deep ploughed times without number. This must be agreed, and the sceptical need go no further than Lothar Wickert's article s.v. princeps in Pauly's RE xxii, 2 (1953), esp. 2002–2004, to confirm the truth of the remark. Much of the ploughing however has been concerned with the question, ‘By what legal right did Octavian/Augustus govern before, during, and after the period in which he claimed to have “transferred the res publica into the discretionary power of the Senate and Roman people”?’ The interest of this particular question has somewhat declined recently, perhaps rightly so, in an age in which there is a score of Octavians in the world, governing by right of victory in a civil war, and the governed populations tolerate these rulers without constantly examining their constitutional credentials, because they have one all-important virtue—they have put a stop to civil war.

Type
Research Article
Copyright
Copyright © W. K. Lacey 1974. Exclusive Licence to Publish: The Society for the Promotion of Roman Studies

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

1 Gnomon 1961, 393 ff. For a more favourable view of Sattler, Brunt, P. A., JRS 1961, 234-5Google Scholar. Sattler's views on the opposition to Augustus have been taken up and amplified by Schmitthenner, W., Historia xi (1962), 31 ffGoogle Scholar. Their basis is a belief in Dio's basic veracity, at least in the narrative. Less confident is Millar, F., A Study of Cassias Dio (1964)Google Scholar. But all historians should heed what Tacitus tells us about his predecessors in imperial history (Annals i, 1); they were Dio's predecessors too, and Dio was less perceptive, and less Roman, than Tacitus.

2 Cf. the valuable summaries by Chilver, G. E. F. (Historia i (1950), 409 ff.)Google Scholar; more recent but more discursive, Salmon, E. T., Historia v (1956), 456 ffGoogle Scholar. Both reveal how much tilling has been in the same furrow—imperium, potestas, auctoritas. A new line of thought was suggested by Sir Frank Adcock (CQ 1951, 132-5), but it did not convince Salmon (op. cit. 457, n. 7), and another by Jones, A. H. M. (Augustus (1970), 46Google Scholar, and A History of Rome Through the Fifth Century (1968-70), 25, 26, 41) which does not convince me. Both involve a formal vote to give Octavian overriding powers. Grenade, P., Essai sur les origines du Principat (1961)Google Scholar, adopted a similar but more sweeping view on the totality of Octavian/Augustus' legally bestowed powers. For a favourable review, Béranger, , Gnomon 1961, 387-93Google Scholar, for a hostile one, Brunt, , JRS 1961, 236-8Google Scholar.

3 See F. Millar (op. cit. 118) for Dio. For this reason, I share Syme's view of Augustus' interpretation of his imperium: JRS 36 (1946), 155Google Scholar, despite the theoretically valid but arm-chair objections of Salmon, op. cit. 465; cf. Syme, Roman Revolution (1939), 307, ‘had the question (of the name of Octavian's powers) been of concern to men at the time’. It wasn't. Cf. id. 324.

4 Mommsen, , Staatsrecht iii, 918-9, 956 and n. 3 (on p. 957Google Scholar). Examples of formal announcement ex s.c.: Caelius in Cicero, , ad fam. viii, 8, 5, and 6Google Scholar, six months' notice of a debate on the consular provinces (51 B.C.); Cicero, , ad fam. i, 9, 8Google Scholar, five weeks' notice of a debate on the ager Campanus (56 B.C.). Ex s.c. embassies from provinces and socii had priority in February (ad fam. i, 4, 1), unless they were explicitly put off (ad Att. 1, 14, 5). Less formal announcements were made by consuls from time to time: Antony let it be known that there would be an important debate on 1st June, 44 B.C., and Cicero was advised to stay away (ad Att. xv, 5, 2 with Phil., ii, 108); Hirtius kept out of Rome in order to avoid having to debate. On Antony's proposed s.c. about Brutus and Cassius ad Att. ibid.; it was to be debated on 5th June (ad Att. xv, 9, 1). On 31st August in the same year, it was known that the following day Antony would propose not the res publica infinite but honours to Caesar, and Cicero was explicitly told that this was so (Philippic i, 11-12 and 8). In December 63, Cicero told those who had arrived at his house how he proposed to deal with the letters he had seized from the Allobroges (in Cat. iii, 7), and on the ‘immortal Nones’, two days later, some Senators stayed away because they knew they were going to be asked to discuss the fate of the accomplices of Catiline, whom the Senate had decreed to have acted contra rem publicam (in Cat. iv, 10).

5 Of Cicero's seventeen surviving senatorial speeches, two, de lege agraria i and in Catilinam i, are of this type; so was Antony's attack on Cicero to which Philippic i is a reply. Cf. Hirtius' and Pansa's speeches on 1st January, 43 B.C.: Cicero, , Phil, v, 1Google Scholar.

6 In Catilinam iv re-opens a debate, and makes clear what the Senate must vote on ‘before nightfall’ (in Cat. iv, 6), though it proposed no motion.

7 Post reditum in senatu and Philippic i clearly belong to this class. So does the lost speech in which Cicero proposed Pompey's corn commission in 56 B.C. (ad Att. iv, 1-6).

8 Philippic iii introduces a motion (38), so do viii (33) ix (15), x (25) and xiv (36); a series of motions in Philippic v, 31, 38, 53; whatever may have been the case in the later speeches, Cicero's motion in Phil, iii was certainly not known or advertised in advance. Phil, xiii (50) and vii (27) introduce proposals in the course of debates already under way; cf. de prov. cons. (1), a debate whose subject was known in advance, and in which Cicero's speech was to support an already-proposed motion.

9 Protracted debates include the restoration of Ptolemy Auletes in 56 B.C. (which was never concluded; for the variety and complexity of the proposals, Cicero, , ad fam. i, 1, ff.Google Scholar); the establishment of a court to try those charged with sacrilege at the Bona Dea festival of 62 B.C. lasted about 2½ months (Jan-mid-March, 61 B.C.) and was settled only by one party capitulating; the revision of the Asian tax-contract (ad Att. i, 18, 7) lasted nearly a year. More briefly, ‘eo die res confecta non est, eo die nihil perfectum est,’ Cicero, , ad Q.f. ii, 3, 1 and 3Google Scholar: many other examples.

10 Cicero's in Pisonem, the fragmentary in toga Candida and in Clodium (et Curionem) are written-up altercationes; de haruspicum responso and Philippic xii combine self-defence with attacks on opponents; pro Marcello is a written-up version of what was a spontaneous and complimentary contribution to a discussion following a consular announcement; compare Cicero's account of Crassus' and his own contributions to the discussion which opened with the consul asking Pompey's views on the court to investigate the Bona Dea affair (ad Att. i, 14, 2-4).

11 Cicero, , Philippic iii, 13Google Scholar: ‘quamquam vos nihil aliud nisi de praesidio, ut senatum tuto consules Kalendis Ianuariis habere possint, rettulistis, tamen mihi videmini magno consilio atque optima mente potestatem nobis de tota re publica fecisse dicendi.’ The implication that Cicero's extension of the field of debate was with the leave of the presiding tribunes is very clear.

12 Augustus 84; for this occasion, Dio liii, 2, 7, and cf. 11, 1.

13 The speech in Dio is generally thought an invention—certainly it cannot be put into Latin in the simple style Suetonius describes as Augustus' (Augustus 86). The atmosphere of hysterical amazement in the Senate may, however, be authentic. Schmitthenner (op.cit. 36, following Sattler) emphasizes the opposition, describes the settlement as a compromise, and, accepting Dip's statements about doubling the praetorians' pay (liii, 11, 5), asks if the meeting was in fact intimidated by troops. If the answer is ‘yes’, the Romans played farces with straight faces superbly—better than I think credible.

14 Corona quern[a uti super ianuam domus imp. Caesaris] Augusti poner[etur senatus decrevit quod rent publicam] P(opulo) R(omano) rest[t]tuit. CIL i 2, 231; most accessible in Ehrenberg, V. and Jones, A. H. M., Documents illustrating the reigns of Augustus and Tiberius (1955), 44 ffGoogle Scholar. (E/J Docs hereafter). This, the official document, says p. R. restituit: re publica restituta exists only on a private document, the famous elogium of ‘Turia’(E/J Docs 357). Millar, F., ‘Two Augustan notes’, CR 1968, 263-6Google Scholar, questions the validity of this document. I agree. For the history and credentials of the fasti, see Gagé, J., 2nd ed. of Res Gestae (1950), 155 ff.Google Scholar, with bibliography of post-Mommsen discoveries, p. 161.

15 Dio liii, 16, 7 says explicitly that the name Augustus was the last honour to be granted, but see below, p. 182. For two stages in awarding honours, Mommsen, Staatsrecht iii, 2, 745–6Google Scholar.

16 OB CIVES SERVATOS is the legend of the first oricalchum SC coins from the Roman mint, dated not before 23 B.C., E/J Docs 19, q.v. for references. Cf. Val. Max. ii, 8, 7. For these coins and others, Bay, A., ‘The lettering SC on the Augustan aes coinage’, JRS 62 (1972), 114Google Scholar, for their date as 18-15 B.C- M. H. Crawford (private communication to the writer) observes ‘there was a hiatus in the coinage between 27 and 19 B.C.’ CIVIBUS SERVATEIS appears on an aureus dated CAESAR COS VII, Mattingly, H., British Museum Catalogue of Coins of the Roman Empire (BMCRE) i, CXXIV and 106–7Google Scholar, nos. 656 and following. The reverse is AUGUSTUS SC; on this issue SC records the gift of the name Augustus by the Senate : Mattingly and Sydenham, Roman Imperial Coinage 62, n. 1.

17 Val. Max. l.c.; cf. Gellius v, 6, 11; Pliny, NH xvi, 7 and 13 for the associated honours. Full references to ancient sources in RE iv, 1639 ff. (Fiebiger, 1901); Mommsen, Res gestae 2 149 ff.

18 Livy ii, 8, 1: rejected as a Livian error, Ogilvie, R. M., Commentary on Livy i–v (1965), 252Google Scholar. Could Augustus be one cause of the ‘error’?

19 IMP CAESAR DIVI F COS VI LIBERTATIS P. R. VINDEX, tetradrachm (cistophorus) BMCRE i, 112, no. 691, attributed to Asia (the Ephesus mint) by Mattingly, to Bithynia (the Nicomedeia mint) by Woodward, A. M. in Roman Coinage, Essays presented to Harold Mattingly (1956), 152Google Scholar, with references; cf. Grant, M., Roman Imperial Money (1954), 24Google Scholar; Syme, Rom. Rev. 306. The reverse says PAX: Grenade (op. cit. (n. 2) 62-3) argues that this was the more important legend. I doubt it, but peace was the natural (and asserted) corollary of the victories proclaimed on the dated quinarii ASIA RECEPTA (IMP VII), BMCRE i, 105, nos. 647-9; on denarii AEGYPTO CAPTA (COS VI), ib. 106, nos. 650 ff., and on an aureus (COS VII), ib. no. 655. Cf. also Livy i, 19, 3, written before 25 B.C., Syme, R., Harvard Studies lxiv (1959), 42Google Scholar. See also Ch. Wirszubski, , Libertas (1960), 4–5Google Scholar for Libertas populi Romani as signifying Republican government, and 100 ff. for libertatis p. R. vindex.

Millar, Fergus (JRS lxiii, 1973, 59 ff.Google Scholar, which was not available to me until after this article was written) has righly emphasized the lack of evidence for a routine personal jurisdiction by the triumvirs in Rome and Italy; we must agree, but there is also no evidence of the normal functioning of the Republican courts of justice. Moreover, the coniuratio totius Italiae must at least have seemed to put those who took the oath in verba C. Caesaris into the same relationship to Octavian as every soldier put himself in relation to the commander under whom he enlisted—a relationship which certainly gave the commander summary powers. This relationship could also only be dissolved by the acto of the commander releasing his troops at the end of his campaign, that is, in Octavian's case, at the end of the civil wars, which were what tota Italia had enlisted for. Note also that Dio liii, 2 suggests that there was no praetor urbanus in 28 B.C. until Octavian appointed one.

20 The date is uncertain, and disputed. The s.c. quoted by Macrobius, (Sat. i, 12, 25Google Scholar) gives the month as August. The battles at Naulochus and Actium were both fought in September (Dio li, 1, 1; for the calendars of the Arval Brethren etc., E/J Docs p. 51; Gagé, op. cit. (n. 14) 180); the s.c. lists Octavian's triumph in 29 and the capture of Egypt in 30 separately; 28 looks the most promising year (so Grenade, op. cit. (n. 2)), except that Janus had been closed (hence peace proclaimed) in January 29 (so Brunt, op. cit. (n. 2)). But it is possible that the s.c. was written well after 27, and called the abolition of the illegal measures of the triumvirs ‘the end of the civil wars’. This involves believing that the s.c. to call Sextilis ‘August’ belongs to the date when it was passed, i.e. to 8 B.C. It is tempting to assign also to 28 the edict and oath of Suetonius, Augustus 27, 2 (so Grenade, 68), and to associate it with the opening of the gardens round his mausoleum to the public, dated to 28 by Suetonius, Augustus 100, 4, who omits the month.

21 Octavian's measures had been confirmed in 39 (Dio xlviii, 34, 1) and again in 29 (Dio li, 20, 1); this step distinguished him from the other two triumvirs (Sattler, op. cit., 34). But propaganda exhibiting deference to A. Cascellius was also perhaps important. Cascellius had refused to accept the validity of the acts of the triumvirs at any time (Val. Max. vi, 2, 12): he was still alive, and probably active, and widely, respected, RE iii, 1635 (Jörs, 1899)Google Scholar.

22 Res gestae 3, 1 and Velleius ii, 86 both exaggerate as far as 31 was concerned; however, the propaganda of 28 must have laid the ground for an award clementiae causa.

23 BMCRE i, nos. 351-6, and Plate 7, nos. 5-8 (a Spanish mint).

24 BMCRE, i nos, 134, 139, 148, 157, 165, 171, 175, etc.; Plate 18 for illustrations of these and others, cf. Plate 2i, 8 (no. 737). The laurel wreath with this legend is a civil war (A.D. 68-9) coin, BMCRE i, no. 289 and n.

25 E/J Docs. 35, from the Capitoline fasti triumphales; cf. the coins cited in n. 19, above.

26 Velleius ii, 89.

27 Dio li, 25, 2; E/J Docs. 17 ( = ILS 81) records Octavian as IMP VII this year, and the dedication as RE PUBLICA CONSERVATA; Mommsen, res gestae 2, 12. ‘Some people say‘ (ὡς γέ τινές φασιν), says Dio, that Crassus also was saluted as imperator; ILS 8801 (from Athens) supports them. See Syme, art. cit. (above, n. 19) 46 for both this point and Crassus‘return (below).

28 Crassus on 4 July, not over the Bastarnae, but ex Thraecia et Getis; Messalla on 25 September (E/J Docs. 35, from the Capitoline fasti triumphales).

29 Syme (n. 19 above) against Grenade, op. cit. (n. 2 above) 171.

30 Livy iv, 20, 5-11, for Octavian's results; Ogilvie, op. cit. (above, n. 18), 563 f., with references to earlier literature. Syme (n. 27 above) disbelieves Groag's suggestion (RE xiii, 283 ff. (1926Google Scholar)) that Crassus‘claims motivated the return of the res publica to the SPQR; so do I. For the relevance of Romulus’honour, Grenade, op. cit. 171; on the other hand (a) Crassus was the only triumphator of the time who had won his victory in ‘Antony's’ provinces; (b) Crassus was the one whose salutation as Imperator Octavian had seen fit to share, though Carrinas' victories had been celebrated in Octavian's Dalmatian triumph (Dio li, 21, 6; cf. above, and n. 27). Perhaps, too, the temple of Jupiter Feretrius was not yet completed; although the restoration had begun before Atticus' death in 32 B.C. (Nepos, , Attictts 20, 3Google Scholar), the temple had been extremely dilapidated, so much so that Augustus counted it as his own work (res gestae 19; Livy iv, 20, 7). Completion might have been delayed, too, if Crassus had made his inconvenient claim as early as 29. Augustus did not become IMP VIII till 25 B.C.

31 E/J Docs. 21 for Gallus' claims; dated 15 April, 29 B.C. Dio liii, 23, 5-7 (dated 26 B.C.) for his disgrace and suicide; his exact offence is not clearly stated. Syme, Rom. Rev. 309 dates it 27 B.C.

32 This might be what Livy, Epit. 134 is trying to say: ‘rebus conpositis et omnibus provinciis in certain formam redactis Augustus quoque cognominatus est’. Moreover, cum per totum imperium populi Romani terra marique parta esset pax, the formula for the closing of Janus'temple, could be held to exclude foreign wars, and was evidently so held, since the temple was not re-opened for Augustus' widely-anticipated campaign in Britain (Dio liii, 22, 5; Momigliano, , JRS 40 (1950), 39Google Scholar), but only for his Spanish campaign (or the war against the Salassi): Plutarch, Mor. 322 B–C (= de fortuna Romanorum 9); Dio liii, 26, 5; cf. Orosius vi, 21, 1 (though he has the date wrong).

33 Fasti i, 589-90: ‘redditaque est omnis populo provincia nostro/et tuus Augusto nomine dictus avus’. est omnis has the support of the best MSS, Lenz (Teubner editor), J. G. Frazer and others; immunis was read by Merkel (1891), and he has convinced some historians (e.g. Gagé, 164). Merkel also proposed res publica for provincia, which Mommsen, accepted (CIL i, p. 384Google Scholar), but later regretted (‘hodie paenitet’, Res gestae 2 147). vestro has MS support, but is not read.

34 E/J Docs., 45: cf. Mommsen, Staatsrecht, l.c.

35 Whether Agrippa, the other consul, got a province is discussed below (p. 183).

36 Dio liii, 4, 3 and 5, 4, heavily underlined by Tiberius' speech as given in lvi, 39 (esp. 4 for the aerarium); but for the credentials of the latter, F. Millar, op. cit. (n. 1), 101. I do not believe in Dio's φροντίδα τήν τε προστασίαν τῶν κοινῶν πᾶσαν (=cura and principatus of the whole res publica), even in Schmitthenner's ‘staatsrechtlich unverbindliche Formel’(op. cit. in n. 1, above, 36, q.v. n. 44 for references), but agree with Syme, Rom. Rev. 313.

37 Strabo xvii, 840; he says the right to make war and peace was granted for life. He may be right, but where did Dio get the information (liii, 13, 1) that Octavian playfully (or boastfully, like a νεανίας) added that he would give back his three provinces to the Senate if they were pacified within the period of his (10-year) proconsulate? This does not look like a historian's—or even a rhetorician's—invention. The ‘frontiers and neighbours’of Syria were, of course, stretched to include the whole Eastern frontier and Egypt.

38 Strabo (l.c.) says ‘to the People’, a clear mistake. The people did not allocate provinces, except whe nunder the Republic they overruled the Senate or preempted provinces for powerful populates.

39 Cicero, , ad. Att. iv, 1, 67Google Scholar ‘factum est s.c. … ut cum Pompeio ageretur ut eam rem susciperet, lexque ferretur … Postridie … nihil Pompeio postulanti negarunt, ille legatos cum XV postularet … legem consules conscripserunt qua Pompeio … potestas … daretur.’

40 Dio liii, 2, 1; one of the occasions referred to in res gestae 17, 1.

41 Most recently, Bay, JRS 1972, 119 ff.

42 The treasury (aerarium Saturni) had always financed provincial government; to make the reorganization coincide was only natural. Cf. Bay, art. cit. 120, who points out that a reorganization of the aerarium preceded both Augustus' constitutional reorganizations. Nobody (then or now) would be foolish enough to suppose that this would exempt the Senate from having to finance the provinces (ornare provincias); that is why the aerarium never had any money. See Polybius vi, 13; Mommsen, , Staatsrecht iii, 1097Google Scholar; RE, Supp. vi, 741 (O'Brien Moore, 1935), for the norm.

43 Velleius ii, 89, 3; ‘restituta vis legibus, iudiciis auctoritas, senatui maiestas, imperium magistratuum ad pristinum redactum modum….; prisca ilia et antiqua rei p. forma revocata. Rediit …’ prisca ilia et antiqua rei p. forma revocata. Rediit …’ This punctuation (Krause, Gagé) should be preferred This punctuation (Krause, Gagé) should be preferred to … modum … Prisca. … revocata, rediit. … which appears to reduce prisca … revocata from being the triumphant capstone to a subsidiary ablative absolute clause; but it is the reading of Halm (Teubner text) and others; some texts print the conjecture renovata;—haud scio an recte (Halm). Sattler, op. cit. 41, n. 95, cites parallels from Cicero for res publica as meaning the traditional functioning of the traditional organs of state: cf. Millar, F., CR N.S. 18 (1968), 263 ffGoogle Scholar. In JRS lxiii (1973)Google Scholar Millar's discussion ignores the aerarium; yet control of this by the Senate was the surest proof of the maiestas senatus in the res publica. The Senate cannot even have seemed to control the aerarium until (at least nominally) it controlled the ornatio provinciarum and the sending out of governors, which, as Millar agrees, was a power the Senate did not recover before 27 B.C. Dio's account (liii, 2) represents Octavian as treating the aerarium very much as within his own prerogative in 28 B.C.

44 One may speculate on these: de re publica SPQR restituenda? Surely not. It was a good ex post facto claim, but not a preparatory motion: de potestate extraordinaria imp. Caesaris divi f. deponenda? or imperio triumvirali? or potestatibus? each gets more blatant, and improbable.

45 Nicolaus of Damascus, a contemporary writer, gave Dio grounds for his interpretation of Octavian's monarchical wishes—e.g. in Caesar's reasons for choosing Octavian as heir, Jacoby, FGH ii, 416Google Scholar, esp. c. xxx (120); Salmon (art. cit., above, n. 2), 458. The study cited there is not available to me.

46 Dio reports that Octavian immediately procured a grant of double pay to those who were going to be his cohors praetoria (liii, 11, 5). Dio's future tense suggests that on 13 January Octavian did not have such a cohors: this is natural, since one is appropriate to a proconsul, not to a consul; as triumvir, Octavian had had one, see Millar, , JRS lxiii (1973), 59 and n. 55Google Scholar.

47 Dio li, 19, 3 for 30 B.C.; E/J Docs. p. 45; the fasti Verulani explicitly state the reason.

48 Pace Syme, Rom. Rev. 313. The 15th, being the Carmentatia, was nefastus parte; this must warn us against supposing too much done.

49 liii, 16, 4 for the former: ὅτε … τὰ περὶ τῆς τῶν ἔθνων διανομῆς … διελέχθη; 16, 6 for the latter: ἐπεὶ δὲ καὶ τῷ ἔργῳ αὐτά διετέλεσεν.

50 34, 2: ‘[et clujpeus [aureu]s in [c]uria Iulia positus quem mihi senatum pop[ulumq]ue Rom[anu]m dare virtutis clement[iaeque e]t iustitiae et pieta[tis caus]sa testatu[m] est pe[r e]ius clupei [inscription]em.’

51 SPQR IMP. CAES. DIVI F. AUGUSTO COS. VIII DEDIT CLUPEUM VIRTUTIS CLEMENTIAE IUSTITIAE PIETATIS ERGA DEOS PATRIAMQUE : full text in An. Ép. 1952, 165. It was found at Aries in 1951: Benoit, , Rév. Arch. 6, 39 (1952), 48Google Scholar, for an illustration. Cf. S. Weinstock, Divus Iulius, Plate 18. For coins showing CL(UPEUS) V(IRTUTIS), or SPQR. CL. V, BMCRE i, nos. 321-3, 333-43. 352, 381, etc.

52 As argued by W. Seston, CRAI 1954, 286 ff.

53 E.g. die rigged voting of 61 B.C., Cicero, , ad Att. i, 14, 5Google Scholar; the riot at Milo's prosecution in 56 B.C., Cicero, , ad Q.f. ii, 3, 2Google Scholar.

54 For the description, Syme (art. cit. n. 19), 55, with the evidence.

55 In E/J Docs. p. 35, the omission of Octavian's second triumph on the fasti triumphales Barberini is a warning against over-confidence about this.

56 M. P. Charlesworth, ‘The Virtues of a Roman Emperor’ (Raleigh lecture, Proc. Br. Academy, 23) 10 ff.

57 S. Weinstock, op. cit. (n. 51), ch. xi.

58 Dating perhaps from the pamphlets published in 36 B.C. (Appian, B.C. v, 130) or to a response to Antony's propaganda before Actium: Suetonius, Augustus 28; Dio 1, 7: cf. Grenade, op. cit. 77, but pushed too far.

59 The opening three chapters of res gestae also seem to have the imperial virtues in mind; Octavian's virtus liberated the res publica from a factio (1, 1), to spare veniam petentibus was a n act of clementia (3, 1), avenging Caesar was an act of pietas, and pursuing the vengeance legitimis iudiciis one of iustitia (2, 1). In res gestae, however, the order of virtues does not correspond with those on the shield; they are therefore probably in the background only.

60 Suetonius, Augustus 7.

61 Dio liii, 16, 7; Balsdon, op. cit. (n. 1) 394, and Syme, op. cit. (n. 19) 55, give reasons.

62 Texts in Gagé, op. cit. (n. 14), 145. Velleius ii, 91 says Senate and People, but res gestae must be preferred. The date is in the fasti of Cumae and Praeneste, E/J Docs, p. 45. Censorinus even gets the day wrong (21, 8.)

63 For its progress to his regular nomen, Syme, , Historia vii (1958), 176 ffGoogle Scholar.

64 Syme, Rom. Rev. 313.

65 Most scholars use his background with confidence, but his speeches with great caution. Even in the background, though, there are blunders: Balsdon, art. cit. (n. 1).

66 lii, 42.

67 Octavian himself, Dio liii, 1, 3; cf. res gestae 7, 2 (not wholly accurate).

68 Dio, loc. cit.; res gestae 8, 2.

69 Suetonius, , Augustus 26, 3Google Scholar: for shared fasces, Dio liii, 1, 1: Staveley, E. S., Historia xii (1963), 458484 (esp. 478Google Scholar) for th e importance of this.

70 P. 180 above; note especially: (1) the control by praefecti of praetorian rank, Bay, op. cit. 120; (2) the cancellation of the pre-Actium contracts, except for those concerned with public buildings; (3) the general cancellation of treasury-debts.

71 Innovations, but not startling ones; indeed, Dio half suggests that the aediles had usurped from the praetors; his comment on Augustus' ‘frequent appointment of the praetor urbanus’ is a comment on later practice, if it is true.

72 Dio (liii, 1, 1) adds that Octavian's other acts were κατὰ τὰ νομιζόμενα; this hardly justifies Sattler's claim (op. cit. 34) that the eiuratio must mean that in the whole of 28 Octavian had made no use of extra-constitutional powers.

73 Dio liii, 2, 4: this sentence needs careful punctuation and translation: τὰ μὲν ἱερὰ τὰ Αἰγύπτια (Egyptian rites) are opposed to τῶν δὲ δὴ ναῶν (temples of (Roman) gods), a class divided in turn into μέν some repaired by the descendants of the original dedicators, δέ others by Octavian himself. Cf. res gestae 20, 4.

74 Dio liii, 1, 4–6: it was an elaborate show with a special (temporary) wooden stadium and contests of gladiators (prisoners).

75 Dio liii, 1,3; explicit evidence in the Antium fasti, E/J Docs. p. 53.

76 J. A. Crook, Consilium Principis 11. Since Augustus left for Spain quite soon, it would have little chance to operate in 27 B.C.

77 Dio liii, 21,6; Dio's exact use of language for once — ὅ τε δῆμος … καὶ τὸ πλῆθος (populus and plebs) adds some confidence in a good source at this point.

78 Dio liii, 22, 1-2.

79 Suetonius, Augustus 26, 3 for the date of his departure. Magie, D., Class. Phil., xv (1920), 327Google Scholar suggests it was not till after Messalla Corvinus' triumph on 25 September, but this seems rather late, unless he had no real intention of invading Britain. Schmittenner (op, cit. 48) suggests a personal handover in Gaul in the summer of 27.

80 Dio liii, 26, 1; the poets had great expectations, Momigliano, , JRS 40 (1950), 39Google Scholar. Orosius (vi, 21, 1, perhaps derived from Livy) suggests that planned aggression in Spain was his object, and that Janus' temple was opened before he departed. For this Spanish campaign, most recently, Schmitthenner, op. cit., 48 ff.

81 Dio liii, 23, 1-2 and 27, 1-2.

82 RE iii, 2635-6 (Fiebiger, 1899).

83 Reinhold, M., Marcus Agrippa (1933), 42-3 and 60–1Google Scholar, with notes.

84 Suetonius, Augustus 40 confirms this.

85 I wish to thank my friends Michael Crawford and Elizabeth Rawson for their advice and help: they should not be held responsible for the views expressed.