^* It is a pleasure to acknowledge the help and advice I have received from Professor G. B. A. Fletcher, Mr. J. J. Paterson, and Mr. R. J. Seager, none of whom necessarily agrees with my conclusions. An abbreviated version of this essay was delivered as part of a Departmental Lecture in the Department of Latin, University of Liverpool, in May 1974. For an explanation of my system of references and abbreviations see the end of this essay.

^{page 272 note 1} F. E. Rockwood produced a school edition with commentary on chapters 41–131 in 1893; F. Portalupi's commentary of 1967 amounts to little more than a translation. J. C. Silverberg wrote a dissertation entitled ‘A Commentary to the Roman History of V.P. (Book II, 1–28)’ (Harvard, 1967), which remains unpublished, as does the ‘Historischantiquarischer Kommentar zur AugustusPartie des V.P.’ (Vienna, 1968) of B. Massauer, who deals with chapters 90–123. Neither of these dissertations incorporates literary material. I am myself preparing a critical edition with commentary, of which the first volume will (I hope) appear in the not too distant future, and would be glad to receive any criticisms or suggestions relevant to the points discussed in this essay.

^{page 272 note 2} I am thinking particularly of those by Haase (1884 edn., including Mommsen's emendations), Halm (1876), Ellis (1898), and Shipley (Loeb, 1924). There was also Bolaffi's edition (1930).

^{page 272 note 3} Schanz-Hosius, 580–7.

^{page 272 note 4} Stegmann's Teubner text remains standard largely because it is the only critical edition still in print (repr. 1965, 1968): Shipley's Loeb, which is also in print and the text of which is in most places superior to Stegmann's, would hardly claim to be a critical edition. The bibliography in SchanzHosius of course needs to be brought up to date: see the subsequent bibliographies by Dihle, 655–9, and by H.-D. Blume in the reprint of Stegmann. J. Hellegouarc'h is at present working on a Budé edition of V., and has also written a judicious bibliographical essay entitled ‘État présent des travaux sur l'Histoire romaine de V.P.’ which is to appear in a volume of Aufstieg and Niedergang der rön. Welt (ed. H. Temporini). I am most grateful to Professor Hellegouarc'h for sending me a copy of his essay in advance of publication.

^{page 273 note 1} The issues seem to me to be sufficiently important to justify the publication of this essay here and now; but in some places, particularly in part II, I have reserved for my commentary details which might inappropriately distract the reader of this essay. For illustration of V.'s neglect by literary scholars see my paper in Empire and Aftermath: Silver Latin II (ed. T. A. Dorey, 5975), 1 with nn. 2 and 4.

^{page 273 note 2} Addresses or apostrophes to Vinicius at 1.13. 5, 1 o 1. 3, 113. 1, I30. 4; cf. also 96. 2, 103. I, 104. 2. Dating from Vinicius' consulship usually takes the form ante annos quam tu, M. Vinici, consulatum inires followed by a numeral: see 1. 8. 1, 1. 8. 4, 1. 12. 6, 2. 7. 5, 49. 1, 65. 2. For the relevance of these allusions in determining the dates of V.'s composition see the next section.

^{page 273 note 3} At 104.3 V. describes himself as ‘successor officii patris mei’, from which Sumner (264–5) deduces that his father had been serving in Germany under the elder M. Vinicius (cos. 19 B.c.), the younger Marcus' grandfather. That is a possible but not inevitable interpretation of the text. It is reasonable to assume that V. elaborated on his relationship with M. Vinicius junior or his family in the lost preface (so, e.g., M. Broiek, ‘De V.P. opusculo mutilato’, Eos lii [1962], 125).

^{page 273 note 4} For the ages of the two Vinicii see R.E. ix A. I. 119–20 and 116–17 respectively.

^{page 273 note 5} Sumner, 288. For Vinicius' admiration of Ovid cf. Sen. Contr. to. 4. 25; for V.'s cf. 36. 3. Previous scholars had naturally attempted to see political significances behind V.'s literary predilections, cf. F. della Corte, R.F.C. XV (1937), 154–9; Lana, 280 ff.

^{page 273 note 6} I hope to deal with this subject in a future essay to appear elsewhere.

^{page 274 note 1} So, e.g., Dodwell, cxiii; Kritz, xxi; Syme (1933), 147 = DP 32, and most other places in his writings. At RR 384, however, Syriac curiously describes the elder M. Vinicius as V.'s patron. If this is not a slip, it is certainly incapable of proof. We know that V. himself went to Germany in A.D. 4 with Tiberius (104. 3), but we do not know the exact dates of the elder Vinicius' command there. Syme himself is most uncertain: at DP 31 he is tentative, ‘A.D. 1–4?’, but at DP 55 more convinced, ‘A.D. I–4’; then at DP 71 he suggests that the command did not begin till A.D. 2, while in Hist. xi (1962), 148, he is properly dubious, ‘Vinicius’s war in Germany c. A.D. 2'. Even if we assume the longest period possible, A.D. 1–4, that would leave little time for V. to acquaint himself with Vinicius who was at once replaced by Sentius Saturninus (105.). Sentius may in fact have taken over in the previous year, as Syme hesitatingly suggests in Hist. xiii (1964), 165; but this suggestion is probably based on a common misunderstanding of V.'s text (see the end of this note), and Syme is probably correct at DP 55 when he sees no reason for Vinicius' replacement until Tiberius' return to public life. But to return to the elder M. Vinicius' patronage of V.: Syme anyway contradicts himself since elsewhere in the same work he says that this Vinicius is treated coolly by V. (RR 431, repeated DP 33). It must be said that Sumner rejects this last point and accuses Syme of ‘malice’ in his reading of the passage in question, 104. 2 (268 n. 75). There is, paradoxically, some truth in each interpretation. I conclude by explaining V.'s text at 105. I since it has misled not only Syme (see above) but also Ruhnken, Groag (R.E. ii A. 2, 1521–2) and Lana (157 n. 685). The text reads: ‘Sentium Saturninum qui turn legatus patris eius in Germania fuerat’, and means: ‘who at that time was a legate of his father [i.e. of Augustus] in Germany’—the reason being that Latin idiom permits the use of the pluperf. instead of the imperf., especially with fueram and habueram (see Kühner-Stegmann, i. 140. 4).

^{page 274 note 2} Excellent remarks on patronage in Williams, 44–5.

^{page 274 note 3} On this whole tradition see esp. Dodwell, cdv; cf. also Syme, Tac. 672. On Claudian note in particular A. Cameron, Claudian (1970), 30–45. Cicero of course not only asked his friend Lucceius to commemorate his consulship (Fam. 5. 12), but also wrote a poem about it himself!

^{page 274 note 4} See G. Williams in Quality and Pleasure in Latin Poetry (ed. Tony Woodman and David West, 1974), 35–6.

^{page 275 note 1} The advantage would not, of course, be such as to jeopardize his political career, Tiberius remains the unmistakable hero col V.'s final chapters. Failure to appreciate this point leads Kritz into difficulties (xxiii).

^{page 275 note 2} So, e.g., Sauppe, 4 and g; Dihle, 640.

^{page 275 note 3} Cf. H. Peter, Der Brief in der röm. Literatur (1901), 242 ff.

^{page 275 note 4} Cf. R. Graefenhain, De more libros de. dicandi apud scriptores Graecos et Romanos obvic (1892), 33, 24.

^{page 275 note 5} See Janson, 106–12, 116. I have not seen J. Ruppert, Quaest. ad historiam dedicationil librorum pertinentes (1911). Val. Max. dedi. cated his contemporary collection of exempla to Tiberius, though not in epistolary form.

^{page 275 note 6} Sumner, 284. For the formulae with which V. dates events from Vinicius' consulship see above, p. 273 n. 2.

^{page 275 note 7} Some scholars (e.g. Sauppe, 9) have seen no reason why V. should not have continued writing until the end of June A.D. 30, since that was the moment at which Vinicius relinquished the consulship; thus Syme, presumably as a consequence, more than once states that V. ‘was writing in 29 or 30’ (e.g. Tar. 368), a view found repeated by others also. But the arguments which I give in this paragraph tell against such a view. For a refutation of Sumner's view see below, p. 276 n. 3.

^{page 276 note 1} Such words of appreciation could hardly have occurred in the lacuna at the very end of the work since the text breaks off in the middle of a prayer for the emperor. Clearly no reference to a consul could follow that.

^{page 276 note 2} Cf. Tac. 5. 5–5.

^{page 276 note 3} Sumner (284–8) follows Lana (299) and SO/Ile others (see above, p. 275 D. 7) in believng that V. continued writing until the late summer of A.D. 30. His evidence falls into two categories. (1) Dates given in V.'s text as ‘abhinc annos xxvii’ (503. 3) and 'horum xvi annorum' (126. 5). But no reliance can be placed upon this evidence. It is clear from the dedication to Vinicius that V. assumed in advance that his work would be published in A.D. 30. Therefore he would naturally date events from that year. This satisfactorily explains the numeral at 126. 1 (Tiberius' accession in A.D. 14 is 16 years from A.D. 30) but not that at 103. 3 (Tiberius' adoption in A.D. 4 is not 27 years from A.D. 30). The latter figure would require a notional publication date of A.D. 31, which is impossible. The numeral, like another in the same sentence, is clearly corrupt: Aldus‘xxvi’ should be read. (2) At 127–8 V. refers to Sejanus' elevation to high office, which Sumner interprets as the consulship of A.D. 31. This is a stronger argument since, if Sumner's interpretation is correct, V. could not have known about Sejanus' elevation before the election in the summer of A.D. 30. But since the vocabulary of chapters 527–8 strongly suggests that a totally different honour is being referred to (see below, pp. 305–2), Sumner's theory must be rejected.

^{page 276 note 4} I say ‘five or six months’ here because Kritz, one of the most eloquent exponents of this hypothesis, only allows ‘vix quinque menses’ (xxiii). Steffen is prepared to consider even four months (2).

^{page 276 note 5} Thus Dodwell, cxiii–cxv; Kritz, xxi–xxiii, lxv sqq.; Schanz-Hosius, 581–2, 585–6; Dihle, 640. So too Massauer, i.

^{page 276 note 6} Full credit should be given to the four scholars whose perceptive comments are quoted below (p. 279 n. p. 282 n. 2, p. 286 n. p. 287 n.; but the incidental nature of their remarks, which I shall quote in full, precludes them from constituting a serious challenge.

^{page 276 note 7} Since most of Book I is lost we can only guess at the total number of words written by V. It is reasonable to assume that Book I would be comparable in length to Book 2, which contains roughly 24,200 words. Book as it stands contains about 3,900 words.

^{page 276 note 8} Thus Schanz-Hosius, 299; cf. Syme (1959), 39 n. 51, 41. To be fair, I should point out that Cicero appears to have written the De Divinatione, a work of roughly 27,000 words, in two and a half months (see Pease's edn., pp. 13–15). But I think it is also fair to regard the comparison with Livy a more realistic reflection of a historian': work-rate. Cicero himself would possibly have agreed: ‘Neque enim occupata opera neque impedito animo res tanta suscipi potest… historia vero nec institui potesi nisi praeparato otio nec exiguo tempore absolvi’ (Leg. i. 8-g).

^{page 277 note 1} So Kritz xxi–xxiii; Schanz-Hosius, 585–6: Welleius Ofters sein beabsichtigte: grösseres Werk in Gegensatz zu dem Abris: stellt, wird man rich die Entstehung de: letzteren kaum anders denken können, al: dass die Designation des M. Vinicius zum Konsul den Autor veranlasste, auf Grund seines gesammelten Materials einen flüchtigen Abriss der römischen Geschichte zu geben. Damit steht im Einldang die häufige Woederholung derselben Wörter and Wortverbindungen; auch die Nachlässigkeit in der Periodenbildung wird nur zum Teil ihre Rechtfertigung durch den veränderten Geschmack der Zeit finden.' The ‘collected material’ for a major work has become an essential part of the hypothesis: cf. Kritz, xvi-xvii; F. Burmeister, De Fontibus V.P. (1894), 14; Sumner, 284 n. 143.

^{page 277 note 2} See E. A. de Stefani, ‘De V.P. periodic’, S.I.F.C.xviii (1910), 19–31; Woodman, A. J., Latomus xxv (1966), 564–6.Google Scholar

^{page 278 note 1} Milkau, 9–10. See now Adams, J. N., P.C.P.S. xvii (1971), 6–10.Google Scholar

^{page 278 note 2} A long list of examples in Kritz, lxviilxx.

^{page 278 note 3} e.g. Cicero, cf. E. Laughton, C.P. xlv (1950), 73 ff.; Livy, cf. K. Gries, C.P. xlvi (1951), 36 f.; Curtius, cf. H. Lindgren, Studia Curtiana (1935), 1–35; Fiona, cf. S. Lilliedahl, Florusstudien (1928), 57–61. Repetitions of various kinds have attracted an enormous amount of attention: to the refs. given by Kenney, E. J., C.Q. ix (1959), 248 n. I, add D. R. Shackleton Bailey, Propertiana (1956), 9; id., Towards a Text … (1959), 28; N. I. Herescu, La poésie latine (1960), 581–203; M. von Albrecht, Meister röm. Prosa (1971), 231 (index); Diggle on Eur. Phaeth. 56; Hofmann-Szantyr, 820–2.Google Scholar

^{page 278 note 4} So Kritz, xv-xvi; Schanz-Hosius, 581–2; Dihle, 641; Sumner, 282. Scholars make this deduction since the first occasion on which V. mentions his major work is at the outbreak of civil war in 49 B.C. (48. 5): ‘harurn praeteritarumque rerum ordo cum iustis aliorum voluminibus promatur, turn, uti Spero, nostris explicabitur.’ In fact, they do not do their case justice since (1) V. could have mentioned his major work in the large portion of his history now lost, (2) at 48. 5 he says ‘harum praeteritarumque rerum’, where praeteritarum would seem to imply at least some pre-49 B.c. history (but see next note). As it is, however, there is too much (fruitless) speculation about V.'s future work to justify basing any argument on it. Kritz, e.g., thinks it would conclude with the reign of Tiberius, Sumner that it would exclude the reign of Tiberius. Cludius (edn. xx), and M. Manitius (Rh. Mus. xlvii [1892], 467) think that V. intended to write four further works; Sauppe toyed with two (it); Schanz-Hosius and Sumner envisage only one; Sauppe (it), Teuffel (17), and Peter (WK 366) doubted whether V. intended seriously to write another work at all. In view of the evidence presented below (pp. 287–8), these last three scholars may well have been right, in which case the present argument is even less soundly based.

^{page 278 note 5} Why did V. not simply present Vinicius with a history ranging from the civil wars to Tiberius? It is hardly possible to argue that everything which precedes 48.5 is an inferior summary ‘stitched on’ to the ‘serious’ history which begins at that point and for which V. is assumed to have collected proper material. It is true that E. Gabba suggested that a similarly modest introduction was ‘stitched on’ to Polio's history of the civil wars (Appian [1956], 207 ff., esp. 232 ff.), but his suggestion has not been generally accepted. Besides, Gabba's hypothesis envisages only 80 years' introduction, whereas we should have to credit V. with more than 1,000 years!

^{page 278 note 6} So, e.g., Dihle, 640: ‘Die Schrift wäre dann in wenigen Monaten zusammengestellt worden. Dazu passt die fluchtige and summarische Erzahlung’.

^{page 279 note 1} Milkau, 10: ‘festinatio illa totiens como memorata magis de operis tenuitate el brevitate quam de tempore urgente accipienda sit.’ He presented no evidence for the: view.

^{page 279 note 2} Further examples of this are given below and in the following section. For brevit) coupled with speed cf. also Arist. Rhet. 3. 9 (1412^b), Cic. De Or. 17, 3. 202, Hor. Sat I. 10. 9. Interesting is Sen. Ep. 40, whicl deals with oratorical delivery rather thar style, but (as Summers in his corrunentar) observes) Seneca ‘does not keep the twc things quite distinct’: ‘oratio illa apuc Homerum concitata’ (2), ‘Istarn vim dicend rapidam’ (3), ‘quemadmodum per proclive currentium non … gradus sistitur. si ista dicendi celeritas’ (7), 'talem dicend velocitatem … ‘tantum festinet’ (8), etc.

^{page 279 note 3} This is made clear not only by Quintiliar (quoted) but also by Sen. Contr. 9. 5. 13 ‘curs sit praecipua in Thucydide virtus brevitas hac eum Sallustius vicit; nam in sententi: Graeca tam brevi habes quae salvo sense detrahas. at ex Sallusti sententia nihi demi sine detrimento sensus potent’.

^{page 279 note 4} For this cf. P. Freitag, Stilistische Beiträge zu V.P.: Pleonasms and Parenthese (1942).

^{page 279 note 5} Cicero's sentence describes the point most aptly, but Quint. also makes a similar distinction himself between the two types (cf. 4. 2. 41, with a reference to ‘speed’, celerius). The distinction is appreciated by such modern scholars as C. 0. Brink, Horace on Poetry (1963), i. 262, ‘As a stylistic feature it (brevity) could be discussed under the heading of style; as a selective principle of subject-matter under the heading of unity, order or content’, and H. Lausberg, Handbuch der literarischen Rhetorik (1960), i. 173, who refers to the brevitas of verba and the brevitas of res. Brevity is an extremely complicated subject and it is surprising that (to my knowledge) no full-scale treatment exists apart from J. Stroux, De Theophrasti virtutibus dicendi (1912), which I have not seen. However, Lausberg's collection of usages of brevitas and related terms is most helpful (pp. 169 ff.), and cf. Brink on Hor. A.P. 25–6.

^{page 279 note 6} Brink's excellent note on 148 points out that ‘This way of looking at poetry is Alexandrian and is perpetuated in the Homeric and Virgilian scholia’: that is, critics of an Alexandrian affiliation had to reconcile the poetry of Homer and Virgil (which by definition had to display all virtues) with brevity (which was an Alexandrian virtue). This paradoxical situation is perpetuated in any age where brevity is popular: see below, p. 286 and n. 2, and the reference to Pliny. Pliny, as it happens, is another author who distinguishes between brevity of style and brevity of content: in Ep. I. 20. 11–17 he discusses the latter (using the terms brevitas and ), then later in the letter (19–21) discusses the former. Between these two passages, however, at 18 he confusingly says ‘brevitate vel velocitate vel utraque (differunt enim)’. Yet we should not take his terminological differentiation too seriously: all Pliny means is that he prefers to use a separate term for each type. All the other evidence shows that both terms (i.e. brevity and speed) can be used to describe both sorts of brevity. Such looseness is of course typical of ancient literary critical terminology.

^{page 280 note 1} See Avenarius, 62–3 and I27–30 respectively on Lucian's distinction. In her commentary on Hust. Conscr. H. Homeyer is wrong at § 56 when she compares Quintilian's description of Sallust at 10. I. 102 (velocitas) since, as I have shown, Quint. there means the other type of brevity. Interestingly Cicero did not approve of brevitas in rebus (Lucian's ) in historiography: he demanded ‘exaedificatio in rebus et in verbis’ (De Or. 2. 63), scorning those old annalists who ‘unam dicendi laudem putant esse brevitatem’ (ib. 2. 53). Cicero is not inconsistent when he praises Caesar's works for their brevitas (Brut. 262) since brevitas was an indigenous quality of the commentarius genre within which Caesar was working (see P. T. Eden, Glotta xl [1962], 75–8). Nevertheless it is well known that Cicero characteristically eschewed brevity in his own commentarii (cf. AU. 2. I.) I It is also interesting that when the third-century rhetorician Aquila illustrates percursio (which is hardly different from ) with the sentence ‘Caesar in I taliam evolavit, Corfinium Domitio deiecto ceperat, Urbe potiebatur, Pompeium persequebatur’, this reads just like the commentarius style as written by Cicero in, e.g., Att. 5. 20. 3 (Aq. 6, p. 24 Halm).

^{page 281 note 1} Tac. I. 8s deals only with the actual electoral process which took place each summer and is thus no evidence for a private arrangement of the kind here envisaged.

^{page 281 note 2} Tac. 2. 36.

^{page 281 note 3} So Shotter, 326–7, and Seager (next n.).

^{page 281 note 4} Seager, 126.

^{page 281 note 5} This point, which is important for my argument, is confirmed by an analysis of the fasti for Tiberius' reign: see Seager, 124 ff., esp. 126–8.

^{page 281 note 6} P.B.A. xxxix (1953), 202. The italics are mine.

^{page 281 note 7} For the standard age see Syme, Tac. 653–6.

^{page 281 note 8} Tac. 6. 15. 1. A number of scholars (e.g. Z. Stewart, A.J.P. axis, [1953], 74 f.; Syme, Tac. 384; R. Sealey, Phoenix xv [1961], 102 ff.) have imagined that Vinicius was a supporter of Sejanus; but his elevation to a ‘royal’ marriage after Sejanus' downfall would seem to discredit this view: cf. Dihle, 640, and esp. Steffen, 194.

^{page 281 note 9} So Syme (1933), 143 DP 27–8.

^{page 281 note 10} Even regardless of specific hints or promises from Tiberius, one could be reasonably sure that Vinicius, in view of his background, would get to be consul sooner rather than later.

^{page 282 note 1} The promise need not of course have mentioned a precise date, merely some general date in the future. That would have been enough to prompt V. to start writing. He could always have changed or, if the worst happened, deleted the references to Vinicius altogether. Changing or deleting references is quite different from adding them. The hypothesis that the completion of V.'s history luckily happened to coincide with the announcement of Vinicius' election in 29 is at first sight attractive (and apparently hinted at by Steffen, 1–2): V. would then have gone through inserting references to Vinicius' consulship into his finished text. But Sumner has pointed out (284 n. 545) that the references at 1. 8. i and 2. 7. 5 are most unlikely to be late additions.

^{page 282 note 2} Jodry (271) stated that in his opinion V. began writing in A.D. 141/15 as soon as he had left the army (see below, p. 287 n. r). He is followed in this by Portalupi (edn. xvi), who quotes his article in a different connection three pages earlier.

^{page 282 note 3} I have chosen Ammianus because he happens to illustrate best the points which I wish to make, especially in terminology. But what I say about him could be said with equal truth about any full-blooded historian writing in the rhetorical or ‘Ciceronian’ tradition, almost all of whom (e.g. Livy, Tacitus) claim to be recounting only the most important material: see Avenarius, 128–9. Sallust is another particularly good example (with the proviso mentioned in p. 283 n. 2). He says that his work is ‘in primis … memorabile’ (Cat. 4. 4), and elsewhere he excuses discursive material with phraseology such as: ‘res postulare videtur … paucis exponere … cetera quam paucissumis absolvam’ (lug. 17. 1–2), ‘de Carthagine silere melius puto quam parum dicere, quoniam alio properare tempus monet’ (ib. 19. 2), ‘Si singillatim aut pro magnitudine parem disserere, tempus quam res maturius me deseret’ (42.5).

^{page 283 note 1} For similar formulae cf., e.g., Nep. praef. 8 ‘sed hic plura persequi cum magnitudo voluminis prohibet tum festinatio ut ea explicem quae exorsus sum’, Sen. N.Q. 3 praef. 4 ‘festinemus et opus nescio an insuperabile…tractemus’, Plin. N.H. 28. 87 ‘ut festinet oratio’, Min. Fel. Oct. t t. 5. These are by no means the only instances where festinatioxs is used thus.

^{page 283 note 2} Sallust (see p. 282 n. 3) would seem to differ from Amm. on this last point since at Hist. 1. 4 he appears to be claiming brevity for the whole of his work; but this is probably due either to the monographical nature of the work (cf. Cat. 4. 3) or to the fact that he means the brevitas in verbis for which he was pre-eminently famous (see above, pp. 279 f.).

^{page 283 note 3} See Janson, 154.

^{page 284 note 1} Some of his references are particularly interesting. E.g. in the discussion of provincialization at 38–9 (which is clearly marked as a digression, cf. 39. 3 ‘sed revertamur ad ordinem’) his introduction is: ‘haud absurdum videtur propositi operis regulae paucis percurrere …’, where percurrere seems to be an allusion to the technical percursio which is defined by Aquila as ‘distantia plura inter se percurrens velocitate ipsa circurnponit’ (6, p. 24 Halm: see above, p. 280 n. 1). For a similar allusion see Brink on Hor. A.P. 18. There may be a less obvious allusion in V. at 117. r ‘persona moram exigit’: mora is a natural word to use in a context of a ‘speedy’ narrative, but it may also be a quasi-technical term = ‘digression’: cf. Amm. 15. I (quoted above, p. 283), Gran. Licin. 36a-b (p. 59 Camozzi) ‘mora et non urgentia’ (in a ref. to Sallust's insertions into his narrative of ‘contiones loca montes flumina et hoc genus alia’), Pallad. 1. 1. r ‘sed nos recidamus praefationis moram’, and Geoffrey de Vinsauf in C. S. Lewis, The Discarded Image (1964), 192.

^{page 284 note 2} Justin's work of course covers events from the origins of Rome, but he is an epitomator, by which I mean an author who has digested one single work only: i.e. he is not ‘original’. By the same token V. is not an epitomator, though often described as such, e.g. recently by S. Usher, Historians of Greece and Rome (1969), 242. See n. 5 below.

^{page 284 note 3} Fest. 1 ‘brevem fieri dementia tua praecepit.… ac morem secutus calculonum, qui … brevioribus exprimunt, res gestas signabo, non eloquar. accipe ergo quod breviter dictis brevius conputetur.’

^{page 284 note 4} Most of the works mentioned in this and the following paragraphs are usually known as breviaria or chronica (there seems little difference between the two terms): these differ fundamentally from epitomes inasmuch as they draw their material from more than one source, and they generally make use of their authors' personal experience: i.e. they are ‘original’. There is something to be said, however, for avoiding these terms and employing the more appropriate ‘summary’ (for which cf. Sen. Ep. 39. 1): for breviarium is conventionally reserved for late (usually fourth-century) works, while chronicon implies a concern for chronology (cf. Gell. 17. 21. 1) which is not always justified. See also next n.

^{page 284 note 5} P. Jal has rightly stressed that Florus' work is in no way an epitome (R.E.L. xliii [1965], 358 ff., and in his edn. [1967], xxi sqq.), though often described as such—e.g. by J. W. Eadie in his edn. of Festus (1967), ii, an error which leads Eadie to state (13) that we have no evidence of a breviarium between the end of the Republic and Eutropius. Yet Florus' work, not to mention V.'s, is essentially no different from that of Orosius, Sulpicius, or any other breviarium-writer. That his work is never described as such is due to the restricting conventions of terminology mentioned in the preceding note.

^{page 285 note 1} For this motif, of the reader ‘hurrying’, cf. Liv. praef. 4 ‘legentium plerisque … festinantibus ad haec nova’.

^{page 285 note 2} Further parallels in E. Klebs, Philol. xlix (i 890), 289. One of the purposes of Kiebs's article was to attempt to show that Sulpicius' phraseology is modelled directly on that of V. This would not, however, necessarily disprove my point: it merely shows that Sulpicius realized the nature of such phraseology in V.'s text and thus was able to use it for a similar purpose in his own universal summary. Similarly with ‘carptim’ at Chron. 1.1.1, which echoes Sall. Cat. 4. 2.

^{page 285 note 3} For the currency of summaries in the fourth century cf. A. Momigliano, The Conflict between Paganism and Christianity in the Fourth Century (1963), 82 ff.; also R. Syme, Ammianus and the Historic Augusta (1968), 104–3. There did of course already exist a tradition of summary histories at Rome when V. came to write: see below, p. 286 n. 4.

^{page 286 note 1} J. Wight Duff, Lit. Hist. of Rome in the Silver Age (3rd edn., 1964), 71, ‘This brevity was a kind of literary fashion, comparable with that to which Phaedrus lays claim at the same period.’ (Vitruvius is not, of course, as exactly coeval with V. as is Val. Max.)

^{page 286 note 2} Plin. Ep. 5. 6. 42 ‘primum ego officium scriptoris existimo, titulum suum legat atque identidem interroget se quid coeperit scribere, sciatque si materiae immoratur non else longum, longissimum si aliquid accersit atque attrahit. vides quot versibus Homerus, quot Vergilius arma hic Aeneae Achillis ille describat; brevis tamen uterque est quia facit quod instituit’: cf. Hor. A.P. 148 ff., quoted above, p. 279 n. 6. It is true that none of this evidence comes from historians, but of course almost no first-century historian has survived. We do know that Sallustian brevitas in verbis became popular with the historian L. Arruntius (cf. Sen. Ep. 114. 17). Scholars have detected a rise in the popularity of brevitas, of one sort or another, from about the first century, reaching its climax in the Middle Ages: see Curtius, 487 ff.

^{page 286 note 3} On book totals see E. Badian, ‘The Early Historians’, in Latin Historians (ed.T. A. Dorey, 1966), 11 n. 51. His scepticism does not contradict what is a clear trend.

^{page 286 note 4} Cf. Catull. 1.5–6 ‘ausus es unus Italorum / omne aevum tribus explicare cards’, a description of Nepos' Chronica which indicates that he was doing something as new in historiography as Catullus himself was in lyric (cf. F. Cairns, Mnem. xxii [1969], 153–4): see HRR ii². mood sqq., 25–6. Nepos had been preceded by Cicero's friend Atticus, who had ‘omnem rerum memoriam breviter…. complexus’ (Brut. 14) in one book apparently entitled Annalis (HRR ii². xxiii sqq., 6–8). The polymath Varro had also written a short work in three books called Annales (HRR ii². xxxviii, 24), but we do not know when. We know that Ateius Philologus had provided Sallust with a ‘breviariurn rerum omnium Romanarum’ (Suet. Gramm. 10), but this may have been little more than a private notebook not unlike the epitomes which Brutus made, apparently for his own private use (cf. Cic. Att. 12. 5b). I cannot see that Mölfflin (335) has any evidence that Brutus epitomes were published; but he is right to stress (342) that Suet. Gramm. so is no evidence for the use of the term breviarium in Sallust's time.

^{page 286 note 5} The apparently recent popularity in epitomes is a useful parallel. Dionysius had epitomized his own Ant. Rom. (zo books) into possibly three books (RAC 5. 948); Martial tells us (14. 190) that Livy had been epitomized by the end of the first century, although it has been generally thought that the epitome is to be dated earlier—possibly (so Wölffiin) as early as V. himself (see C. M. Begbie, C.Q. xvii [1967], 332 ff. foi discussion); Vibius Maximus, a contem porary of Statius, produced what seems t have been a short work based on Sallust ant Livy (cf. Stat. Silv. 4. 7. 54–6); Fenestella': Annales (21 books) were epitomized (HR1 ii2. 87) but we do not know when. A valuably list and discussion of epitomes are provide by I. Opelt, RAC 5. 944–50; cf. also Wölff lin. I do not wish to imply that the produc tion of large-scale works dried up in the firs century A.D., which is quite clearly not tht MSC (see Syme 1959, 64 ff., J. Wilkes ‘Julio-Claudian Historians’, C. W. lxv [1972] 177–203). Yet V. has survived and they hay(not. Is this an indication that he sensed th literary mood of the time better than they? He was certainly in tune with the mood of later antiquity.

^{page 287 note 1} To his great credit Jodry had also come to this conclusion (274): ‘La 'festinatio’ qu'allégue V.P. n'est pas la hâte avec laquelle it rédige, pressé par le temps, son Histoire, mais la compression du récit commandée par les lois du genre de l'abrégé. But he unfortunately produced no argument whatsoever to substantiate his thesis.

^{page 287 note 2} The reason is often panegyrical: the events to be described are so great, magnificent, etc., that they cannot be done justice in the present work: see further below, p. 290 n. 3.

^{page 288 note 1} On the poets see W. Wimmel, Kallimachos in Rom (1960), Doblhofer, Nisbet-Hubbard on Hor. Odes i. 6, Janson, 75–6; on Tacitus see Syme, Tac. 219 fT.; on the later authors see Peter, WK 414 and n. 2. It should be noted that assertions of the modesty, or even of the incapacity, of one's present style or work imply no insult to the subject under discussion but rather the opposite: see, e.g., Janson, 124 ff.

^{page 288 note 2} The Aeneid is an awkward case: although it undeniably glorifies Augustus, it is hardly the poem foreshadowed in Georgics 3.

^{page 288 note 3} The ref. at 119. t does not at first sight appear panegyrical (the context is the Variana clades); but there is evidence that V. used his treatment of the disaster to bring out the military qualities of Tiberius.

^{page 288 note 4} prior servitus is an allusion to Domitian's reign, which occupied a part, presumably, of the Histories when published.

^{page 288 note 5} Sumner, 283.

^{page 288 note 6} Prof. D. A. West has pointed out to me a very interesting parallel. In the last chapter of Pantagruel Rabelais ‘promises another book after this one, even adding a summary of what it will contain. This is a normal fifteenth-century convention for ending a book’ (D. G. Coleman, Rabelais [1971], 55).

^{page 289 note 1} Die röm. Literatur (6th edn., 1961), 91.

^{page 289 note 2} Teuffel, 17.

^{page 289 note 3} Paladini, 469–78.

^{page 289 note 4} Steffen's work is mentioned only by Koestermann, whose pupil he was, of those scholars I have read.

^{page 289 note 5} ‘So darf man dem Tiberius-Bild des V.P. einen unverachtlichen historischen Wert beimessen, sobald man nur berücksichtigt, was für ein Mann es 1st, der es zeichnet’ (647).

^{page 289 note 6} Sumner, passim.

^{page 289 note 7} To be strictly accurate Marsh's book, which contains 335 pages, makes one further ref. in a footnote on p. 152. (The quotation is from p. 5 n. 2.)

^{page 289 note 8} E. Kornemann, Tiberius (ed. H. Bengtson, 1960), does refer to V. less infrequently than Marsh, but without conviction, and certainly far less than one would expect from a historian with a sympathetic attitude towards Tiberius.

^{page 289 note 9} Syme, Tac. 358, cf. 200, 271.

^{page 289 note 10} Respectively (5933), 547 n. 3; (1934); 121; RR 393 n. 5; (1956), 262; Tac.367; (1959), 69; TST 47.

^{page 289 note 11} Balsdon, J. P. V. D., Rome: the Story of an Empire (1970), 52–53, is one such example.Google Scholar

^{page 289 note 12} Seager, 266–9.

^{page 289 note 13} e.g. M. A. R. Colledge reviewing Balsdon's book: 'Why is Velleius Paterculus, of all people, given an outsize blurb?' (C.R. xxiii [5973]; 243.)

^{page 289 note 14} J.R.S. lvii (1967), 275.

^{page 290 note 1} Tac. 759 11. 2 and 08 respectively. My analysis of these chapters will be stylistic, hence the title of part II of this essay. But it should not be thought that this is a full account or evaluation of V.'s style as a whole: see, e.g., next n.

^{page 290 note 2} imagine was plausibly placed here by T. Sinko, Eos xx (1954–15), having already been suggested elsewhere in the sentence by Friebel in 1837. Casaubon ad loc. had noticed that Suet. Aug. 9 (‘proposita vitae eius velut summa, partes singillatim neque per tempora sed per species exsequar,!’) closely resembles V.'s sentence. See the excellent remarks of Leo, 241, who explains the difference between the two authors—although he is only one of many scholars to note the importance which biography plays in V.'s history. But this is unfortunately one of several subjects outside the scope of the present essay.

^{page 290 note 3} See, e.g., Curtius, 83–5,159 ff., 411 ff.; also above, p. 287 n. 2.

^{page 290 note 4} I have assigned V.'s rhetorical question at 126. I to this topos on account of the word partibus. It could perhaps be argued, by stressing the cum-clause, that V. has a different idea in mind; but panegyrical apologies often hide behind apologies of a quite different nature (thus the Augustan poets by means of Callimachean apologies: see Nisbet-Hubbard on Hor. Odes. 6 intro.). Besides, the notion ‘everyone knows’ lies behind such formulae of the praeteritio as ‘quid loquar?’ or ‘quid memorem?’ and these are also very common in panegyrical writing (for examples, cf. W. S. Maguinness, Hermath. xlviii [1933], 131–2).

^{page 291 note 1} Seager, 269, who at least is fair. Con. trast Syme, Tac. 367, ‘… no more meritiN credence than what [V.] has to relate about the phenomenal virtues of the Tiberiar regime, instantaneously apparent in tota contrast to all that went before’.

^{page 291 note 2} So Schäfer, 24.

^{page 291 note 3} Cf. Plin. N.H. 7. 149 f.; Syme, Tac. 427

^{page 291 note 4} So, rightly, Steffen, go. For Tiberius devotion to Augustus cf., out of man? examples, Tac. 1. 77. 3, 4. 37. 3; Grant Aspects, 39–40.

^{page 291 note 5} Plin. Pan. 53. ‘nihil non parum grass sine comparatione laudatur’ (cf. 53. 6) Menand. rhet. 376. 31–377. 2.

^{page 291 note 6} See Fraenkel, 450 f.

^{page 291 note 7} Another quasi-difficulty is the deification of Augustus at 126. r, which had already been mentioned at 124.3. Yet the deification of one's predecessor was to become one of the conventional introductory motifs of panegyric: cf. Plin, Pan. I 1, and Durry, 31.

^{page 291 note 8} Instructive for comparison are: J. Mesk, W.S. xxxii (1910), xxxiv (1912), Rh.M. lxvii (1912), 569 ff.; A. Klotz, Rh.M. lxvi (19, I), 513; L. K. Born, A.J.P. lv (1934), 20 ff.; Doblhofer, and review by R. G. M. Nisbet, C.R. xix (5969), 573–5; Gärtner, H., Einige Überlegungen zur kaiser zeitlichen Panegyrik … (1968); Cairns, 105 ff. One of the virtues of Nisbet-Hubbard's commentary on Hor. Odes s is the use they make of comparisons with V.Google Scholar

^{page 292 note 1} e.g. Solon fr. 5, Democr. B39, 79, Plato, Laws 711, Xen. Cyr. 6. 13, 8. 1. 12, 21 ff., 8. 5, Ages. 7. 2, Isocr. Areop. 21–2, Nicocl. 37; Cic. Leg. 3. 3I, Rep. 5. 47, 2. 69, Fam. 1.9. I2, Res Gest. 8, Ov. Met. 15. 834, Plin. Ep. 3. 18. 2–3, Sen. Clem. i. 1. 6, 1. 9. 1, Ep. 1i. 8.

^{page 292 note 2} The contrast between exemplum and imperium is found elsewhere (Tac. G. 7. s, Plin. N.H. 35. 86), but in view of the similarity of context and the close resemblance of Plin. Pan. 50. 4 to Vell. 126. 4 (quoted above), it would seem that Pliny here echoes V. See below, p. 295 and n. 6.

^{page 292 note 3} See Dodds on Eur. Bacch. 882–7, Nisbet-Hubbard on Hor. Odes I. 15. 19, Orelli on ib. 3. 2. 31 f., Ogilvie on Liv. 3. 56. 7, Smith on Tib. I. 9. 4, Mayor on Juv. 13. ID); Otto, Sprichwörter, III.

^{page 292 note 4} For which cf. Cairns, 218.

^{page 292 note 5} Cf., e.g., Seager, 144 ff.

^{page 292 note 6} Cf., e.g., Tib. 2. 5. 59–60 (and Smith's n.), Hor. Odes 4. 15. 14 ff., Anth. Pal., 6. 61, Curt. 7. 8. 12 (and Wilhelm 41), Pan. Lat. 2. 23. I; esp. Fraenkel 451 n. 4.

^{page 292 note 7} e.g. Hor. Odes 4. 5, Tac. D. 38, Inscr. Brit. Mus. 894.

^{page 292 note 8} The literary conceit has its roots in the attitude which persuaded Romans to see the pax augusta in terms of the power which made that pax possible. Cicero voiced the attitude (Off. 34–5) but it blossomed under the empire, cf. Virg. Aen. 286 ff., Val. Max. 2. 7 init.: see further H. Fuchs, M.H. xii (1955), 203–4; Fraenkel, 376; S. Weinstock, I.R.S. 1 (1960), 45 f., 49 f.; Williams, 435–8

^{page 292 note 9} Cf. Maguinness, W. S., Hermath. xlvii (1932), 42–61; xlviii (1933), 117–38.Google Scholar

^{page 293 note 1} The evidence for this practice is assem bled by Durry, 3–5.

^{page 293 note 2} ‘Zur Quellenanalyse des Plinianischer Panegyricus’, W.S. xxxiii (1911), 85–7 Earlier scholars who had described V.'. final pages as ‘panegyrical’ (e.g. Peter, GI i. 388, Leo 241 f.) were speaking no mon specifically than Syme.

^{page 293 note 3} Several scholars have naturally pleader that we should distinguish what V. says from the manner in which he says it: e.g. Laistner, M. L. W., Greater Roman Historians (2947), 110; Paladini, 477.Google Scholar

^{page 293 note 4} So too Paladini, 476, and others.

^{page 293 note 5} Cf. Vell. 224. 3. There is no reason to doubt V.'s statement, as does Frei-Stolba, 145–6 and n. 31.

^{page 294 note 1} So Canesi in his commentary (1955), ad loc.; Lacey, 171–2.

^{page 294 note 2} Cf. Shotter, 323 and n. 3.

^{page 294 note 3} A similar metonymy at Liv. 9. 46. I I; cf. Cic. De Or. 3. 167.

^{page 294 note 4} Dio 58. 20.3–4 [i.e. other than the consulship] [i.e. the com. centuriata and com. tributa respectively]

^{page 294 note 5} e.g. Lacey, 172; Frei-Stolba, 139. Tiberius' electoral reforms have caused a large amount of discussion, for which cf. Goodyear on Tac. 1. 14.4–15. 1, and Seager, 124 nn. 2–3. 6 Tac. 427.

^{page 294 note 7} e.g. Seager, 123 ff., with references.

^{page 294 note 8} Further confirmation in Rogers, 3 ff., 14 ff.; Lana, 183 ff.

^{page 294 note 9} See also RIC I. I05; Lana, 185 f.; Rogers, 15 ff.; Koestermann on Tac. 2.47.2.

^{page 295 note 1} For dementia cf. Rogers, 35 ff.; Grant, RAI 47–51; Béranger, 271, all with further refs.; for moderatio see Goodyear on Tac. x. 8. 5.

^{page 295 note 2} Tac. 420 n. 2. Tacitus places the dividing line in A.D. 23 (4. 6), Suetonius in 26 when Tiberius retired to Capri (Tib. 26 ff., 39–41). The real trouble, however, did not start until the conviction and execution of Sejanus in 31, by which time V. had stopped writing. This fact, always brought up by those who wish to defend V. (e.g. Paladini, 477), is always discounted out of hand by those who wish to condemn him (e.g. Teuffel, 17 'It will not do to excuse this unworthy attitude by the plea that Velleius wrote his work before (Tiberius') last … years'). But why not?

^{page 295 note 3} Camb. Anc. Hist. x. 369 ff., on the Pannonian Revolt (the quotation from p. 369 n. i), an attitude also found, e.g., in H. J. Rose, Handbook of Lat. Lit. (1936), 355, ‘[IV.] is at times of use … when the fuller and better historians fail us’. In an article on the Pannonian Revolt, however, J. J. Wilkes goes one better (Acta Ant. Hung. xiii [1965], 112–14). Modern students of ancient history are constantly lamenting the paucity of ancient historians who comply with Polybius' conditions for being ‘pragmatic’ historians, i.e. who actually took part in the events which they relate. V.'s account of the Revolt, however, is pragmatic, being based on his own experiences as a soldier on the western front. Wilkes complains that this ‘produces a serious lack of balance’ because V. gives no details of the eastern front!

^{page 295 note 4} Cf. DP 68.

^{page 295 note 5} J.R.S. xlv (1955), 25 = TST 47 (my talks). The context of Syme's statement (a liscussion of M. Lepidus, cos. A.D. 6) is interesting for those concerned with the tech3ique of some modern historians. Syme there states that in A.D. 14 Lepidus ‘held Hispania Tarraconensis with three legions’ (TST₃₅), a statement which he had already made in RR 138 and which he subsequently repeated also n Tac. 382 n. 9: the same statement occurs n P.L.R.² i. 61 no. 369 (x933); G. Alföldy, Fasti Hispanienses (1969), 12–13 (who quotes 3yme's statements); and L. Hayne, ‘The Last of the Aem. Lepidi’, A.C. xlii (1973), Soo. Now the only evidence which these scholars are able to quote in order to subrtantiate their assertions is Vell. 125. 5. Yet the paradosis of that passage reads as follows: ‘†ad Hispanias exercitumquet† virtutibus:eleberrimaque in Illyrico militia praedixinus, cum imperio obtineret, in summa pace <et> quiete continuit, cum …’—which is pure nonsense, hence the daggers. It is perfectly true that Madvig suggested a supplement which contained Lepidus' name, but that is no reason for his suggestion to be universally accepted as if it were a fact. A glance at Kritz's long note will show that the whole textual problem is far more complicated than these scholars have allowed. Only if Lepidus' position in A.D. 54 were to be confirmed by other evidence would we be able to contemplate inserting his name here. As it is, there are arguments in favour Df L. Aelius Lamia (Sauppe); and other arguments that the passage does not refer to A.D. 14 at all (as every ed. before Haase believed).

^{page 295 note 6} 'Velleius avait eu le mérite d'inaugurer Sur le mode des rhéteurs des thémes d'actualité propres au régime alors récent, des thémes romains et impériaux (Durry, 31, who proceeds to remark: ‘Pline n'a pas manqué de s'en inspirer’; ‘Velleius enrichit le genre d'éléments neufs que Pline s'est empressé de faire siens’). I shall provide two further examples of V.'s characteristic blending of rhetoric and fact. (1) At 129. 4 he says of the defeat of Sacrovir and Florus, ‘ante populus Romanus vicisse se quam bellare cognosceret nuntiosque periculi victoriae praecederet nuntius’. The sentiment owes much to panegyrical formulae found in many authors (e.g. Liv. 44. 32. 5, Curt. 3. 5. 6, 7. 4. 14, Sen. Ben. 6. 31. 9, Flor. 18. 6, 1. 29. 2, 2. 13. 63, Just. 11. 2. 10, Pan. Lat. 10. 5. 4; C. v. Morawski, Eos V [1898–9]i 5; Wilhelm, 7 ARM. 2; Korzeniewski, D., Die Zeit des Curtius Rufus [1959]Google Scholar, 61 f.). Yet, as Rillniken saw, Tiberius himself had in fact used a similar formula when announcing the defeat in the senate (cf. Tac. 3. 47. ‘Tiberius ortum patratumque bellum senatu scripsit’). (2) At 130.5 V. records the death of Livia, ‘cuius potentiam nemo sensit nisi aut levatione periculi aut accession dignitatis’. This is a topos of the consolatio (e.g. Sen. Cons. Pol. 3. 2 ‘nemo potentiam eius iniuria sensit’, 13. 1 ‘nee in isto potentiam tuam nisi in ea parte qua prodes ostenderis’, id. Anth. Lat. Ii. 405.3 ‘Crispe, potens numquam, nisi cum prodesse volebas’, Eleg. Maec. 15 f., Cons. Liv. 47 etc.; Esteve-Forriol, J., Die Trauer- and Trostgedichte in der röm. Lit. [1962], 136). Yet Livia had, e.g., saved C. Corn. Cinna Magnus from the wrath of Augustus (Sen. Clem. 1. g, D10 55. 14 ff.), and had used her influence to gain senatorial rank for M. Salvius Otho, grandfather of the emperor (Suet. Otho 1).Google Scholar

^{page 296 note 1} Att. 1. 19. so; cf. Kroll, R.E. Suppl. Vii. I 128–9.

^{page 296 note 2} Syme, DP 33, describing what is admittedly a characteristic of panegyrics. Yet two recent historians, Sumner (265 if.) and Seager (268–9), are undistressed by V.'s treatment of others.

^{page 296 note 3} Such a perspective might take into account writers like Cicero (Leg. 1. 5 'ut ea (patria) … per to eundem sit ornata'), Livy (cf. praef. 1), and Florus (1 praef. 3 ‘non nihil, ut spero, ad admirationem principis populi conlaturus’).

^{page 296 note 4} Tac. 368, 571.

^{page 296 note 5} Tac. 29.

^{page 296 note 6} The most recent account is Seager, 178–223, with further references.

^{page 297 note 1} Dio's is very clost to V.'s ‘adiutorem in omnia’. There art further correspondences. Tac. 4. 2. 3 ‘ut socium laborum non modo in sermonibus sed aped patres et populum celebraret’ is very clost to Vell. 128.4 ‘ad iuvanda vero opera principis … senatumque et populum Romanum ec perduxit ut …’; and Tac. 4. 40. 7 ‘nihil esse tam excelsum quod non virtutes istae (i.e. of Sejanus) tuusque in me animus mereantur, datoque tempore vel in senatu ye, in contione non reticebo’ is very close to Vell 128. 1 ‘in huius virtutum aestimatione … iudicia civitatis cum iudiciis principis certant’.

^{page 297 note 2} Thus Abraham, 8; Rogers, t 39; Steffen, 188; Koestermann on Tac. 4. 2. 3.

^{page 297 note 3} V. was himself a senator and would have heard such references on innumerable occasions.

^{page 297 note 4} A subject much discussed, cf. the excellent analysis of A. Ferrill, Hist. xx (1971), 718–31.

^{page 297 note 5} Senatorial opposition is rightly given great stress by Steffen, 186–7.

^{page 297 note 6} On the sources of the correspondence see Koestermann on Tac. 4. 39. I.

^{page 298 note 1} This aspect is well brought out in Tac. 6. 8. 3–4, and rightly stressed by Marsh, 189 f.

^{page 298 note 2} I should say that my analysis, which is similar in some respects, was written before I had been able to see Steffen's dissertation. His work, and that of Sumner, are by far the best guides to chapters 127–8.

^{page 298 note 3} e.g. 2. 9. 1 ‘eodem tractu temporum’, 90. 2 ‘in quas provincias …’.

^{page 298 note 4} e.g. 1. 24. I, I. 16. I, 36. 2, 66. 3.

^{page 298 note 5} Koestermann on Tac. 4. 39. 2: see esp. Steffen, 186 f., 194; Grenade, 466 ff. For example, V. stresses Sejanus' labor, cf. Tac. 3. 72. 3, 4. 39. 2 where also his vigilantia, for which cf. Vell. 127. 4; for his indifference to success cf. 227. 4, Tac. 4. 39. 2, 40. 5

^{page 298 note 6} So Rockwood, ad loc., and others.

^{page 299 note 1} The sense is: ‘Sejanus always reckonec himself below the aestimatio of others (127.4): BUT there was no doubt about the aestiomatio others made of his true worth (128. 1)’.

^{page 299 note 2} e.g. Sall. Iug. 85. 15 ‘fortissumum quern. que generosissumum’ (where generosus = nobilis, as often, cf. Gelzer, Roman Nobility 38), 85. 17 ‘ex virtute nobilitas coepit’, Cic Ep. Hirt. fr. 3P ‘nobilitas nihil aliud sit quan cognita virtus’, Sest. 136, Liv. I. 34. 6 ‘exvirtute nobilitas sit’, 4. 3. 13. There is a vas literature about this; cf. most recently T. P Wiseman, New Men in the Roman Senate (1971) 107 ff. on 'the ideology of novitas.

^{page 299 note 3} For tutela cf. Cic. Rep. 2. 51, Hor. Ode, 4. 24. 43–4, Epist. 2. I. 2, Strabo 17. 25 (p. 1172 Meineke), Vell. 105. 3, Tac. 1. 2. I, 12. I, Val. Max. g. a ext. 4 (!), Suet. Aug. 94. 8, Tit. 6. 1, Sen. Clem. I. 1. 5, Plin. Ep. 10. 52, 10. 102; A. von Premerstein, Vom Werden and Wesen des Prinzipats (1937), 117; Béranger, 204, 257–60, 266; Grenade, 446 ff.; Grant, M., From Imperium to Auctoritas (1946)Google Scholar, 452–3; RAI 89–90. For securitas cf. Instinsky, H., Sicherheit als politisches Problem des riim. Kaisertums (1952),Google Scholar ff.

^{page 299 note 4} Steffen (193) notes the inconsistency of 129. concluding (194–5) that chapters 127–8 are a later addition to the main narrative, inserted for the reasons given below, p. 301 n.1

^{page 300 note 1} See refs. above. The only Ciceronian echoes in these chapters are 127. 3 ‘compage corporis’, cf. Senec. 77 ‘compagibus corporis’; 128. 4 ‘exempli imitatio’, cf. Flacc. 24 ‘Imitationem exempli’. This is not the place to illustrate the many Ciceronianisms in 129–31. An abrupt change of style may be an indication of discomfort or stress. An interviewer of Stangl, the Nazi war criminal, has recorded that he ‘had a curious habit of changing from the semi-formal German he usually speaks, to the popular vernacular of his childhood whenever he had to deal with questions he found difficult to answer’ (Daily Telegraph Magazine, 8 October 1971).

^{page 300 note 2} For which cf. Bonner, S. F., Roman Declamation (1949), 51 ff.Google Scholar

^{page 300 note 3} e.g. Hor. Sat. 1. 6, Sen. Contr. s. 6. 3 fr., 7. 6. 8, Val. Max. 3. 4, 3. 5, 4. 4. II, Sen. Ben. 4. 30, 44; A. Oltramare, Les Origines de la diatribe romaine (1926), 266.

^{page 300 note 4} His prestigious position is discussed below; the idea of humble status is insisted on throughout 128. 1–3: ‘hominem novum’, ‘equestri loco natum’, ‘novum’, ‘ignotae originis’. Steffen gets it right in general terms (187: ‘Diesen Vorwurf, den man in manchen Kreisen gegen Sejan erhob, zu entwerten, ist das eigentliche Anliegen des Velleius’); but his interpretation of chapters 127–8 (for which see below, p. 301 and nn. 1–2) prevents him from realizing that V. has isolated these two areas of complaint in 127 and 128 respectively.

^{page 300 note 5} Syme (1956), 257–66, esp. 262–5, has ingeniously suggested that V. chose Carvilius as one of his exempla because the man had built a shrine to ‘the most patently plebeian of the Roman deities’ Fors Fortuna (Liv. ro. 46. 14), a goddess who had especial significance in Tiberius' reign (Tac. 2. 41. 1), particularly for Sejanus himself (Dio 58. 7. 2–3). V.'s practice in this section seems to be to mingle conventional exempla of the ‘humble birth topos’ (e.g. Cato, cf. H. W. Litchfield, H.S.C.P. xxv [f914], 33) with more subtle instances such as Carvilius and others (see below, p. 301 n. 4). The same is true of the exempla at 127. I: Val. Max. 4. 7. 7. compares the amicitia of C. Laelius and Scipio Africanus with that of Agrippa and Augustus in much the same way as V., attributing to both Laelius and Agrippa many of the standard virtues to be expected from those who devote their lives to the welfare of the state. Sure enough, in A.D. 28 the senate voted for the erection of an ara amicitiae, to be flanked by statues of Tiberius and Sejanus (Tac. 4. 74. 2; cf. Grant, Aspects 54). The astuteness of V.'s reference to Agrippa has long been recognized: e.g. L. Levy, Quomodo Tiberius se gesserit (1901), 78; Syme, Tac. 402 f.; Grenade, 168 f., 466 ff.

^{page 301 note 1} Steffen, 591 ff.

^{page 301 note 2} Sumner, 291 ff., refuting not Steffen (whom he had been unable to see) but Syme, Tac. 368, to whom a similar notion had occurred independently. My analysis in the next paragraph, if it is accepted, shows that Steffen's interpretation of 127–8 is also unlikely.

^{page 301 note 3} Sumner, 286.

^{page 301 note 4} His actual observation is (286 n. 157): ‘It could be held that Velleius manages to hint at more than the consulship for Sejanus, but it will not be denied that he alludes to the consulship specifically.’ Well, does he? V. alludes to the office of Pont. Max., consulships, censorships, triumphs, and the six consulships of Marius—hardly a specific reference to the consulship. He then mentions Cicero, of whom he says 'ut paene adsentatione sua quibus vellet principatus conciliaree; then Polio, to whom the Roman people ‘nihil negaverunt’; then the section concludes ‘plurimtun esse tribuendum’. These three statements, none of which makes any reference to the consulship, are particularly apt descriptions of the unusual power wielded by Sejanus throughout the twenties A.D. (cf. esp. Tac. 4. 2. 3 ‘neque senatorio ambitu abstinebat clientes suos honoribus aut provinciis ornandi’, 6. 8. 4 ‘ut quisque Seiano intimus, ita ad Caesaris amicitiam validus’). I thus deny the latter part of Sumner's observation, but readily accept the former part. Steffen (1g1) also sees chapter 128 as defending ‘unusual’ power, but he of course sees this as foreshadowing Sejanus' appointment as Tiberius' successor.

^{page 301 note 5} Denied by Steffen, 191.

^{page 302 note 1} Strabo 6. 4. 2 (p. 396 Meineke) .

^{page 302 note 2} For the date cf. Ehrenberg, V. and Jones, A. H. M., Documents illustrating the Reign of Augustus and Tiberius (ed. 2, 1955), 52.Google Scholar

^{page 302 note 3} We may accept the evidence put forward by Kornemann, E., Doppelprinzipat and Reichsteilung im Imperium Romanum (1930), 35 ff., without accepting the interpretation he places upon it.Google Scholar

^{page 302 note 4} I am not saying that chapters 127–8 were written in A.D. 23; but I do believe that they would have been appropriate at any moment in the twenties from then on.

^{page 302 note 5} The view is particularly common among those who, like Syme (Tac. 368), believe that V. perished in the aftermath of Sejanus' execution. Notable exceptions to this view are Abraham, 13 f.; Dihle, 640. Steffen and Sumner both argue, in my opinion rightly, that V. was actually antagonistic towards Sejanus: see below.

^{page 302 note 6} Steffen (194 f.) also makes this point.

^{page 302 note 7} There may have been a similar opportunity for V. at 101–2. We know from these chapters that V. was on the staff of Gains Caesar in the East at the turn of the century, and it is likely (though not provable, cf. Sumner 292 n. 197 on Tac. 4. 1. 2) that Sejanus was there too. It is usually assumed (e.g. by Syme 1956, 265) that the acquaintance of the two men must date from this period; yet V. makes no mention of it at any point during his elaborate account of Gains' oriental activities, even though this is the kind of personal detail he usually delights to include (cf. e.g. 101. 3). But the reason for this ‘omission’ may simply be that Sejanus was not in fact there at the time.

^{page 303 note 1} So too Steffen, 195, who avoids incon sistency in his argument by assuming that V pays to Sejanus merely ‘eine konventionelli Reverenz’. He rightly avoids using word like ‘eulogy’ or ‘panegyric’ throughout hi careful discussion, preferring instead tho term ‘Charakteristik’.

^{page 303 note 2} This is, after all, a natural conclusioi since Tiberius was held to be responsible fo Sejanus' promotion (above, p. 298), and it i Tiberius who as hero occupies the final page of V.'s work. There are further indications Sejanus is mentioned remarkably infre quently (only twice) in a discussion whicl has generally been regarded as his pane gyric; and when he is mentioned, it is officialese (see above), and as the recipien of honours either from Tiberius (127. 3) a from Tiberius acting with the senate and people (128. 4). Kritz mistakenly described these chapters as ‘Seiani laudes, vilissimae et abiectae’, but added ‘paene excusationem Tiberii suscepisse videatur’ (xvii sq.). He was followed by Sumner, who called the passage a ‘long eulogy’, but also said: ‘Vell. is not so much concerned to praise Sejanus as to justify Tiberius’ treatment of him…. The idea that the historian was enthusiastic about Sejanus' prospective ascendancy is based on an insufficiently attentive reading of his words. … An unprepossessed reading of Vell.'s work will reveal that he was devoted to Tiberius, not Sejanus' (292–6). Sumner comes to this conclusion on evidence which is different from both mine and Steffen's, some of it (I believe) dubious.

^{page 304 note 1} I here give two final examples of the way in which scholars assume that V. is falsifying evidence because they are predisposed to believe in his ‘adulation and mendacity’. It does not occur to them how ludicrous it would have been for V. to repeat lies which his contemporaries would have dismissed out of hand. It is more realistic, not to say fair, to work from the assumption that V. is telling the truth. (I) At 106. 2 V. writes: ‘denique —quod numquam antea spe conceptum, nedum opere temptatum erat—ad quadringentesimum miliarium a Rheno usque ad flumen Albim, qui Semnonum Hermundurorumque fines praeterfluit, Romanus cum signis perductus exercitus, et eodem, mira felicitate et cura ducis, temporum quoque observantia, classis, quae Oceani circumnavigaverat sinus, ab inaudito atque incognito ante maxi fiumine Albi subvecta, <cum< plurimarum gentium victoria, cum abundantissima rerum omnium copia exercitui Caesarique se iunxit.’ Now Syrne (1933, 147 = DP 33 n. 23) and C. M. Wells (German Policy of Augustus [1972], 1.59, 218–19), understanding V. to claim for Tiberius the honour of being the first to penetrate to the Elbe, have accused him of deliberate falsification since we know that Drusus and L. Domitius Ahenobarbus had reached the Elbe in 9 B.C. and A.D. 1 respectively. Yet the much criticized parenthesis (quod numquam antea ….) refers not simply to the clause which ends perductus exercitus but to the following clause as well, i.e. to the rest of the chapter. It is clear from the words which link the two clauses, et eodem, that V. thought of the passage as a single sentence describing a single operation. V.'s admiration is directed not at Tiberius' reaching the Elbe but at the joint military and naval operation which turned out so successfully. As nominal commander—in—chief of the expedition (Plin.

^{page 304 note 2} . 167 ‘auspiciis divi Aug.’) Augustus was equally delighted, but since (as Syme himself has remarked, Camb. Anc. Hist. x. 340) he did not have ‘the instincts of a soldier’, his delight is restricted to the more spectacular naval part of the campaign (cf. Res Gest. 26. 4). V., on the other hand, the professional soldier, sees the joint operation as a whole, and is impressed as much by the technical expertise of his immediate commander (temporum quoque observantia) as by his success (mira felicitate et cura duds). When we consider what such an operation must have involved in A.D. 5, we can appreciate that not only V.'s account but also his admiration are fully justified. (The alleged lie in al inaudito atque incognito mari was explained long ago by Ruhnken, q.v.) (2) At 112. 7 V. writes: ‘hoc fere tempore … crescentibtu in dies vitiis dignum furore suo habuit exitum’ of Agrippa Postumus. Now W. Allen (‘The Death of Ag. Post.’, T.A.P.A. burviii [1947] 139) and Syme (Tac. 367 and n. g) accuse V. of implying that Postumus ‘died some time before the end of Augustus' reign’. Yet they forget that the Latin word exitus is nc less ambiguous than the English word ‘fate’ it can of course mean ‘death’, but it can also mean ‘the final state of a person’ (O.L.D. s.v. 4)—as, e.g., in sentences very similar to V.'s, Cic. Verr. 2. 5. 189 ‘Verrem … dignus exitus eiusmodi vita atque factis vestro iudicio consequatur’, 15. II. 2 ‘tibi velim ei sint exitus quos mereris et quos fore confido’, Sen. Ira 3. i6. 4 ‘habuit itaque quem debuit exitum’. Since Postumus' ‘final state’ was exile, and since this took place in A.n. 7, the precise year which V. is discussing at 112. 7, it is surely the man's exile, not his death, to which reference is being made. (Syme may perhaps be excused a little because he mistakenly attributes V.'s account to A.D. 6, thus missing the significance of the year of Postumus' exile.) Seager (268) has recently described these two items as ‘perhaps (the) only two major instances of dishonesty’ in V.'s account of Tiberius. Now even these must go.