1. Thus (synthetized) Marsh, F.B., The Reign of Tiberius (Oxford 1931)Google Scholar; Syme, R., Tacitus (Oxford 1958)Google Scholar; Walker, B., The Annals of Tacitus (Manchester 1952).Google Scholar

2. See, for example, Quinn, K., ‘Tacitus’ Narrative Technique’ in Latin Explorations (London 1963)Google Scholar; Segal, C.P., ‘Tacitus and Poetic History: The End of Annals XIII’, Ramus 2 (1973) 107–26CrossRefGoogle Scholar; more recently, Woodman, A.J., Rhetoric in Classical Historiography (Portland, Oregon 1988)Google Scholar, has a chapter on Tacitus.

3. Daitz, S.G., ‘Tacitus’ Technique of Character Portrayal’, AJP 81 (1960) 30–52Google Scholar, compares Tacitus’ Annals with a modern historical novel (52 n.97); the need to distinguish between narrative voice and authorial voice which is an obvious aspect of that genre is no less important here. Tacitus never speaks in what Luce, T.J., ‘Tacitus’ Conception of Historical Change: The Problem of Discovering the Historian’s Opinions’ in I.S. Moxon, J.D. Smart, A.J. Woodman (edd.), Past Perspectives (Cambridge 1986) 143–57Google Scholar, describes as sua persona (146f.).

4. The percentage devoted to senatorial debate increases through Annals 1–3, making its sudden diminishment in Annals 4 the more noticeable. Martin, R., Tacitus (London 1981) 107fGoogle Scholar. sees Tacitus exploiting a necessary break between the themes of Germanicus and Sejanus. However, the period at issue is really 2½ years rather than 4 years as Martin suggests (from 3.20–4.1); and the treatment of A.D. 27 (4.63–67), for example, indicates how quickly Tacitus might dispense with a year.

5. Contrary to Shotter, D.C.A., ‘Tacitus, Tiberius and Germanicus’, Historia 17 (1968) 195–214Google Scholar, who takes 1.33 at face value — and regards the rivalry between Tiberius and Germanicus as a simple personality clash (196f.). Nevertheless Shotter’s essay instigated a much-needed revision of critical attitudes to Tacitus’ Germanicus.

6. Criticism of Tacitus’ Germanicus has done an about-face: with Syme (n.1 above) 418, ‘From first to last, Germanicus Caesar is adorned and enhanced’ compare Jr.Ross, D.O., ‘The Tacitean Germanicus’, YCS 23 (1973) 209–27Google Scholar, ‘Germanicus then becomes … a figure of almost comic failure’ (215). Ross follows Shotter (n.5 above, 194) in contradicting the opinion that evidence against Germanicus is only found by ‘reading between the lines’ (cf. Syme [n.1 above] 418 and Walker [n. 1 above] 8). Goodyear, F.R.D., The Annals of Tacitus Vol. I (Cambridge 1972) 240fGoogle Scholar, suggests very tentatively a middle line — ambivalence, ‘or plain inconsistency’ — in which he is followed by Luce (n.3 above) 151, who sees in ‘anomaly, inconcinnity and complexity’ a more realistic approach to human affairs. Criticism of Tacitus’ Tiberius will benefit from analogous revisions.

7. Note the loading of the latter part of the sentence with ‘deception’ words.

8. Compare the parallelism of the relationship between the favour in which Germanicus is held and the hope of freedom which he represents: credebaturque, si rerum potitus foret, libertatem redditurus; unde in Germanicum favor et spes eadem (‘and it was believed that, if he had gained control of things, he would have restored the republic; hence for Germanicus favour and the same hope’, 1.33). It follows that any undermining of Arminius’ claims as the defender of German freedom raises questions in the mind of the reader with respect to his Roman counterpart.

9. On Tacitus‘ condemnatory treatment of Germanicus’ generalship see Shotter (n.5 above) and Ross (n.6 above).

10. On Germanicus’ accountability here see Ross (n.6 above) 218ff.

11. Pragmatism, too, seems implicit in Tacitus’ succinct treatment of Germanicus’ response to the carnage at 1.49 — a perfunctory exclamation followed immediately by the order for burial: mox ingressus castra Germanicus, non medicinam illud plurimis cum lacrimis sed cladem appellans, cremari corpora iubet (‘next Germanicus, having entered the camp, amid many tears labelling this not a remedy but a disaster, ordered the bodies to be burned’, 1.49).

12. Ross (n.6 above) 218: ‘It is in the people’s mind alone that Germanicus … is Tiberius’ foil: for the historian Germanicus is merely a lower representative of the supreme power.’

13. Woodman, A.J., ‘Tacitus’ Obituary of Tiberius’, CQ 39 (1989) 197–205CrossRefGoogle Scholar, notes dialectic between 1.4 and 6.51: ‘It will not be forgotten that in the obituary he also describes Tiberius’ life before A.D. 14 as egregium, which is not at all the impression given of the same period at 1.4.3–5: “multaque indicia saevitiae, quamquam premantur, erumpere” ’ (205). Elsewhere Woodman (n.2 above, 196 n.83) suggests that the regularity of the discrepancy between political image and reality ‘almost constitutes a warning to view the narrative itself in that light’.

14. Individual episodes also. The German campaign of A.D. 15, for example, is framed by Arminius and Segestes, political advancement for Germanicus, and the Varus incident. Chapter 55 decrees Germanicus a triumph, introduces the German leaders and recounts the fall of Varus. Chapter 71 witnesses the surrender of Segestes’ brother, whose son is pardoned only after some hesitation, because of a general belief that he insulted the corpse of Varus. Germanicus’ popularity and the loyalty of his soldiers to himself is strengthened by his ‘kindness’ (comitas).

15. Arminius’ death is brought forward for the purposes of design effectiveness.

16. Perverted ritual is a feature of Tacitean historiography as it is of Senecan drama and the Petronian novel. Degenerate or infamous banquet scenes are common to all three.

17. Syme (n.1 above) 345 notes also the ‘strange emphasis’ acquired by ‘normal words’ which are associated with Tiberius (taciturnitas, haesitatio).

18. Stylistically a Sallustian inheritance — cf. Woodman (n.2 above) 167f.

19. On Tacitean innuendo and the evocation of prejudicial impressions see esp. Ryberg, I.S., ‘Tacitus’ Art of Innuendo’, TAPA 73 (1942) 383–404Google Scholar; and Walker (n.1 above).

20. Cf. Ryberg (n.19 above) 390.

21. For a more extended discussion see Ryberg (n.19 above) 391ff.

22. Note also Sallust Cat. 10: saevire fortuna ac miscere omnia coepit, ‘fortune grew savage and began to disturb everything’.

23. Woodman (n.13 above) 203 translates aut. ‘or rather’ — to make a match between 4.1 and the remainder of Annals 4. The ambiguity is intentional, however.

24. Woodman (n.2 above) 196 n.83: ‘Certainly Tacitus … likes to manipulate his readers in contrary directions.’

25. ignotae … virtutes, nova vitia; et quia ipsorum moribus aliena perinde odium pravis et honestis (‘unfamiliar … virtues, new vices, and being different from their own ways, inspired hatred alike for bad and good characteristics’, 2.2).

26. Which distinction does not contradict, but may prove more useful ultimately than, Daitz’ (n.3 above) ‘devices’ — direct description, innuendo, etc. — for handling, say, the speeches of Tiberius (‘many of them … actually quite good in content,… nevertheless generally derided’, Daitz, 45). The tension between propositional and dramatic, and also within modes, challenges the reader to confront the complexity of the problem of analyzing Tiberius.

27. We can take on board the dialectic between egregium vita famaque (‘excellence in life and reputation’, 6.51) and the rumours concerning Tiberius expounded at 4.1 — cf. Woodman (n.13 above) 205 n.42.

28. Annals 6.51 continues to excite debate — and in relation to 6.48. Critics are no longer hamstrung necessarily by the view that Tacitus ‘conceived of character as a wholly static and immutable thing’ (Marsh [n.1 above] 14; cf. Alexander, W.H., ‘The Tacitean “Non Liquet” on Seneca’, California Publications in Classical Philology 14 [1952] 356Google Scholar; Martin [n.4 above] 105). Woodman (n.13 above) follows Gill, C., ‘The Question of Character-Development in Plutarch and Tacitus’, CQ 33 (1983) 469–87CrossRefGoogle Scholar, in reaffirming the familiarity of the ancients with the concept of character change, but rejects (with Hands, A.R., ‘Postremo Suo Tantum Ingenio Utebatur’, CQ 24 [1974] 312–17CrossRefGoogle Scholar) the suggestion of Koestermann, E., Cornelius Tacitus Band I (Heidelberg 1963) 38Google Scholar, that Arruntius’ opinion is posed deliberately as an alternative to the historian’s opinion. (Hands’ cause for doubt is ‘the brevity of this episode as against the whole structure of Books 1–6 and the emphasis inevitably assumed by the final chapter’, 315). The latter point is an important one; but as to the former, 6.48 may be more consistent with Annals 1–6 on the whole). It is assumed generally that Tacitus’ opinion was that Tiberius, at least, did not change — which view seems rather more puzzling if not explained by a general theory of immutability. At this point it is especially useful to be able to distinguish between narrative and authorial voice. Two attempts to sidestep the issue prove unconvincing. Woodman (n. 13 above), citing Histories 1.90 as precedent, retranslates suo tantum ingenio utebatur as ‘had only himself to rely on’ in an attempt to defuse the moral volatility of the final paragraph. The moral force not only of the expression itself but of the surrounding text (which verbally recalls 3.26 — pudore/pudore, metu/metum, scelera/scelere, libidmibus/libidme, suo … ingenio, suopte ingenio — as 1.4 anticipated it) is against him. Luce (n.3 above) attempts to reconcile 1.48 with 1.51 by arguing that both are concerned with behavioural change; but here again the strong language employed by Arruntius with Tiberius named explicitly as subject (cum Tiberius … convulsus et mutatus sit, ‘when Tiberius … was distorted and changed’) militates against his theory.

29. Syme (n.1 above) 399: ‘The attitude of Tiberius was neutral and exemplary.’ Compare Walker’s account (n.1 above, 91ff.) of the historian’s manipulation of the material in order to cast Tiberius in the ‘ruthless tyrant’ role; or the account of Quinn, K., Texts and Contexts (London 1979) 237ffGoogle Scholar. Actually it is both/and — somewhere in between.

30. Tiberius’ impartiality is a source of dismay to Piso: nullo magis exterritus est quam quod Tiberium sine miseratione, sine ira, obstinatum clausumque vidit, ne quo adfectu perrumperetur (‘nothing terrified him so much as to see Tiberius, without pity, without anger, resolute and closed off, lest he be assailed by emotion’, 3.15). On adfectu here and affectus (Hist. 1.15) see Shorter (n.5 above) 212f.

31. Amoliri (‘to eliminate’) is deliberately ambiguous — see Ryberg (n.19 above) 392. Contrary to Shotter (n.5 above, 205) ‘eliminate’ implies no more than Tacitus intended his reader should infer.

32. For more detail on the recitation of hearsay, opinion and belief to build the case against Tiberius, see Ryberg (n.19 above) 393ff.

33. Victories rightly described by H. Furneaux, The Annals of Tacitus Vol. I (2nd ed. London 1896) 248 as ‘insignificant’. See also Shotter (n.5 above) 200.

34. Shotter (n.5 above, 202) reads against the text in describing the alternatives as ‘valid’ and ‘in no way malicious’. Whereas the irrational motive plainly derogates Tiberius, the events of 1.63–66 force a double-take.

35. Shotter (n.5 above, 203) draws attention to the fact that Tiberius’ arguments at 2.26 are self-contradictory — presenting Tiberius as an awkward diplomat, working hard not to embarrass Germanicus. There is something in this characterization — although the alleged inconsistencies are not irreconcilable. (One is a qualification of the other).

36. Compare 3.24, where Tiberius follows Augustus in personal resentment against Decimus Silanus.

37. Compare exuta aequalitate (‘with equality discarded’, 1.4).

38. Ryberg (n.19 above, 386f.) cites 4.10f. as sample of Tacitean innuendo against Tiberius.