1. See Syme, R., Tacitus (Oxford 1958), 481–85Google Scholar, 517–19; against Syme see Goodyear, F.R.D., The Annals of Tacitus Vol. 1 (Cambridge 1972), 127fGoogle Scholar., I83f.; also Bowersock, G., ‘Tacitus and the Province of Asia’, in Luce, T.J. and Woodman, A.J. (eds.), Tacitus and the Tacitean Tradition (Princeton 1992), 1–10Google Scholar, at 4 and 10. The focus of Syme’s study was author-centred, looking to authorial intent; Syme’s reading categorised Tacitus as either ‘pro’ or ‘anti’ the princeps under whom he wrote. Such terms, I believe, are invalid, based on an intentionalism which cannot be definitively retrieved; see Kennedy, D.F., ‘“Augustan” and “Anti-Augustan”: Reflections on Terms of Reference’, in Powell, Anton (ed.), Roman Poetry and Propaganda in the Age of Augustus (Bristol 1992), 26–58, at 34f.Google Scholar

2. See Shotter, D.C.A., ‘Tacitus’ View of Emperors and the Principate’, ANRW 2.33.5 (1991), 3263–328Google Scholar. History was always, in a sense, politically charged by the very nature of the genre and its didactic intent to prepare men for political life. See Luce, T.J., ‘Tacitus on ”History’s Highest Function”: praecipuum munus annalium (Ann. 3.65)’, ANRW 2.33.4 (1991), 2904–27Google Scholar. The Agricola is one of the best examples of this utilitarian view of history (42.4, 46.2f.).

3. Consequently my reading is very much in agreement with Henderson, J., ‘Tacitus/The World in Pieces’, Ramus 18 (1989) 167–210CrossRefGoogle Scholar; see esp. 194: ‘As against attempts to read the Annals as the wilful imposition of a verbal definition on a period of history, even an epoch, something we challenge, outdo with an approximation of our own.’ The ‘meaning’ of the Annals is fluid, comprehended, in part, through the understanding of one’s own epoch and its interpretation is defiantly multivalent.

4. Potter, D.S., ‘The Inscriptions on the Bronze Herakles from Mesene: Vologeses IV’s War with Rome and the Date of Tacitus’ Annales’, ZPE 88 (1991), 277–90Google Scholar, esp. 287 and 287 nn.35–39 citing the bibliography on the current state of the question (of which see Beaujeu, J., ‘Le mare rubrum de Tacite et le problème de la chronologie des Annates’, REL 38 [1960], 200–35Google Scholar; S. Borzák, ‘P. Cornelius Tacitus’, RE suppl. 9.467; Bowersock, G., ‘The Greek-Nabataean Bilingual Inscription at Ruwwafa, Saudi Arabia’, Bingen, J., Cambier, G. and Nachtergael, G. [eds.], Le monde grec: Hommages a Claire Préaux [Brussels 1975], 518–20Google Scholar; Syme [n.1 above], 768; id., ‘How Tacitus Wrote Annals I-III’, Roman Papers III [Oxford 1984; orig. publ. 1977], 1014Google Scholar–42; id., ‘Tacitus: Some Sources of His Information’, JRS 72 [1982], 69–71Google Scholar; Goodyear, F.R.D., The Annals of Tacitus Vol. 2 [Cambridge 1981], 387–93Google Scholar; Martin, R.H., Tacitus [London 1981], 31)Google Scholar. For the most recent discussion see Bowersock (n.l above), 3–10.

5. Syme (n.l above), 768–70; see Ten Studies in Tacitus (Oxford 1970), 144f.Google Scholar, and ‘How Tacitus…’ (n.4 above), 1037–40.

6. Potter argues that the epigraphic record indicates that Parthia was not able to reassert control over the kingdom of Mesene in Lower Mesopotamia after Roman withdrawal under Hadrian and that it remained allied to Rome. Mesene had the status of a Roman client kingdom, and such a status helps to solve the question concerning to which sea mare rubrum refers; it is Potter’s argument that Tacitus included this client kingdom in the sphere of the Roman Empire since Rome still exerted considerable influence there, and hence the reference is probably to the Persian Gulf. That does not necessarily put Arabia Felix, reduced to a province in 106, out of the running (a key difficulty in accepting Potter’s argument based on the inscription); it merely makes it likely that Tacitus could use the client kingdom of Mesene as a boundary of empire. This, contends Potter, could imply a late date for the Annals. See Potter (n.4 above), 277 and 289.

7. See Potter (n.4 above), 289f.

8. For the date of Pliny Letters Book 8, see Syme (n.1 above), 661; also Sherwin-White, A.N., The Letters of Pliny: A Historical and Social Commentary (Oxford 1966), 38fGoogle Scholar. (see 39–41 for a similar date for Book 9 which could refer to the reading of Tacitus’ Historiae [Ep. 9.24]; also see comments by Syme [n.1 above], 120).

9. See e.g. Plin. Ep. 7.20.5; 7.33.1–2; 9.23.3.

10. Brink, C.O., ‘Can Tacitus’ Dialogus be Dated? Evidence and Historical Conclusions’, HSCP 96 (1994), 251–80Google Scholar. The difficulty with Brink’s dating is that it relies too much on the Panegyricus. Pliny’s letters on that work could date to as late as 103, when he was still in the process of revising the speech. See also Barnes, T.D., ‘The Significance of Tacitus’ Dialogus de Oratoribus’, HSCP 90 (1986), 225–44, at 229–32 and 243f.Google Scholar

11. The possible discrepancy between his rate of composition between c.97-c. 100–103 and c.100–109 further indicates that he did not work at a consistent rate of composition. Syme’s conjecture for the length of the Histories (n.l above, 686) is, I believe, correct.

12. See Plin. Ep. 5.8.12f.: tu tamen iam nunc cogita quae potissimum tempora adgrediar. uetera el scripta aliis? parata inquisitio, sed onerosa collatio. intacta et noua? graues offensae leuis gratia, nam praeter id, quod in tantis uitiis hominum plura culpanda sunt quam laudanda, tum si laudaueris parous, si culpaueris nimius fuisse dicaris, quamuis illud plenissime, hoc restrictissime feceris (‘Consider, however, which times I should above all treat. Ancient ones written about by others? The evidence is at hand, but bringing it together is burdensome. For, aside from the fact that there are more things worthy of blame than praise amidst the great vices of men, if you do praise them, although you do so most fully, you are called sparing, if you blame them, however sparingly, you are said to have done so excessively’). This is not to be confused with the deliberate introduction of a subtext into a work, concerning which see Cic. De Or. 3.53; Quint. Inst. Orat. 9.2.65. See Ahl, F., ‘The Art of Safe Criticism in Greece and Rome’, AJP 105 (1984), 187–203Google Scholar for discussion; for a related study see Bartsch, S., Actors in the Audience: Theatricality and Doublespeak from Nero to Hadrian (Cambridge MA 1994)CrossRefGoogle Scholar, passim. For the role of genre in encouraging the reading of a subtext see Kennedy (n.l above), 47: ‘The identification of the genre of a particular work is one of the chief ways of determining its reception, and of reproducing a reading practice.’ I focus here specifically on historical texts; naturally this could extend to any number of genres, as Tacitus himself was aware; hence the Dialogus is set the day after Curiatius Maternus gave a reading of his drama, the Cato, in which influential members of the court had detected offensive material. On the safety of poetry and oratory see the first pair of speeches, Dial. 5.3–13.6.

13. See Kennedy (n.l above), 31, who notes that the stability of a text is illusory, open to disruption, contestation and change.

14. See, e.g., the case with Phaedrus, the writer of fables under Tiberius. For an excellent general discussion see Bartsch (n.l2 above), 67–82. See, too, Tacitus’ treatment of Mamercus Aemilius Scaurus (Ann. 6.29.4f.) and Cremutius Cordus (Ann. 4.34–35).

15. Similar to what Henderson (n.3 above, 167) called a ‘Politics of interpretation’: ‘No way can the interpretation of politics exorcise from its workings the “Politics of interpretation”.’ Meaning is negotiated between the juxtaposition of past politics and the recollection of more current events. Further indication that Tacitus’ audience would make such connections and draw parallels between the past and present is found in Ann. 1.9. At the funeral of Augustus, many of those in attendance made comparison and connections between Augustus and other Romans of the past: multus hinc ipso de Augusto sermo, plerisque uana mirantibus, quod idem dies accepti quondam imperii princeps et uitae supremus, quod Nolae in domo et cubiculo in quo pater eius Octauius uitam finiuisset. numerus etiam consulatuum celebratur, quo Valerium Coruum et C. Marium simul aequauerat, continuata per septem et triginta annos tribunicia potestas, nomen imperatoris semel atque uiciens partum aliaque honorum multiplicata aut noua (‘Then there was much conversation concerning Augustus himself, with many marvelling at meaningless things—the fact that on the same date he had assumed supreme power and died, that he had died in the same room of the same house where his father, Octavius, had died. The number of his consulships was even discussed, in which he equalled the number Valerius Corvus and Gaius Marius had held, that his tribunician power continued for thirty-seven years, that he obtained the name imperator twenty-one times, and there were other honours multiplied or new’). The case was the same with Germanicus at his triumph in 17, where people were reminded of the fate of his uncle and father; while still later at his funeral, more striking parallels were drawn between Germanicus’ life and that of Alexander the Great (Ann. 2.41.5; 2.73.2–3).

16. One could further argue that Tacitus’ text, like his audience, collapses the boundaries between past and present. The two essential chronological terms by which Tacitus attempts to establish his authority as an author are recens and procul (Ann. 1.1.4–5): Tiberii Gaique et Claudii ac Neronis res florentibus ipsis ob metum falsae, postquam occiderant recentibus odiis compositae sunt, inde consilium mihi pauca de Augusto et extrema tradere, mox Tiberii principatum et cetera, sine ira et studio, quorum causas procul habeo (‘The reigns of Tiberius, Gaius, Claudius and Nero, when they themselves flourished, were falsified due to fear; after they had died, the histories of their reigns were written with hatreds still fresh. My plan, therefore, is to relate a few things concerning Augustus and the last part of his reign, then the principate of Tiberius and the rest, without anger or partisanship, the reasons of which are far from me’). While these terms ostensibly establish a chronological boundary, they simultaneously break down this boundary for Tacitus’ readers. That Tacitus will not touch a subject which might call into question his accuracy due to recens odium establishes his authority as an author, implicitly impugns those who do not follow suit, and dictates his choice of subject matter set well in the past; Tacitus’ own recens odium, in the present, forces him to look procul, far into the past, for his subject. In this sense, the present can be said to construct a foundation for the past Tacitus depicts; the boundaries set up by procul and recens, then, are rendered dynamic and permeable rather than static. This framework of contraposition was one Tacitus employed with relative frequency, as Williams, B.’ study, ‘Reading Tacitus’ Tiberian Annals’, Ramus 18 (1989), 140–66CrossRefGoogle Scholar (esp. 140–46), has shown. Courage/fear, liberty/servitude, arrogance/civility, etc., are contrasted, and thus create a complex ‘politico-moral’ nexus, which also serves to collapse past and present (see esp. 143).

17. And which it was up to those in the senate in Tiberius’ day to scrutinise as well. For one of the most famous (and unsuccessful) instances of this see Asinius Gallus’ misreading of Tiberius, Ann. 1.11.1–4.

18. For Tiberius as a constructed text see Dunkle, J.R., ‘The Rhetorical Tyrant in Roman Historiography: Sallust, Livy, and Tacitus’, CW 65 (1971), 12–20Google Scholar. It is worth noting that Tacitus’ reading of Tiberius uses the very methods of scrutiny he condemns in those such as delatores who were adepts at detecting subtexts and subversion in literary works (see e.g. Tac. Ann. 4.34.1f.). For this aspect of Tacitus’ discourse see Kennedy (n.l above), 40: ‘Establishment discourse is shaped by and contains traces of its opposition (and vice versa), even if the conflicting voice is not heard in its own right.’

19. There are other aspects of Tacitus’ history which would invite the reader’s scrutiny. Not the least of these is Tacitus’ style, which looks back to Sallust, which in turn draws on Cato the Elder’s; one could argue an ideological program behind Tacitus’ style which is embedded in a discourse which is anti-imperial (in the form of a Sallustian style that looks back to the republic).

20. See Waters, K.H., ‘Traianus Domitiani Conlinuator’, AJP 90 (1969), 385–405Google Scholar; Ahl (n.12 above), 174–208 (esp. 187–203).

21. Ahl (n.12 above), 207, argues we should beware of Tacitus’ assertion of libertas in Hist. 1.1.4, basing his arguments on the fact that ‘The verdict of extant literature is very different. Latin poetry flourished under Domitian. The closest thing to epic Trajan’s reign has left is his column, and Suetonius, in his Lives of the Caesars, never even mentions Trajan’s name. The Caesars are twelve and end with Domitian.’

22. See Bennett, J., Trajan, Optimus Princeps (Indianapolis 1997), 210.Google Scholar

23. For a good discussion of Trajan in the historiographic tradition see Waters, K.H., ‘Trajan’s Character in the Literary Tradition’, in Evans, J.A.S. (ed.), Polis and Imperium: Studies in Honor of Edward Togo Salmon (Toronto 1974), 233–52Google Scholar; see also Cizek, E., L’Époque de Trajan (Paris 1983), 26–52Google Scholar. Pliny’s more positive judgment on Trajan’s character is of dubious value; Pliny’s assessment took place while Trajan was still alive and is little different from the flattering posture we find in literati under Domitian such as Martial and Statius.

24. Tac. Dial. 28.1–41.5; Hist. 1.1.4; Agr. 2–3. For general discussions of literature under Trajan, see Bennett (n.22 above), 131–37; Cizek (n.23 above), 483–91.

25. People and events mentioned in Augustus’ obituary will surely have been connected in the audience’s mind with contemporary personalities and events. The mention of the luxus (‘luxury’) of Tedius and Pollio will have both had clear and current parallels in Trajan’s close friendship with the likes of the luxury loving L. Sura. On Sura see Mart. Epig. 6.64.13; 1.49.40; 7.47; Arrian Diss. Epicteti 3.17.4. Tacitus had also criticised Augustus for starting the practice of divine worship of his person (Ann. 1.10.5), a practice which saw further development under Trajan. See Cizek (n.23 above), 130f.; cf. Jones, B., The Emperor Domitian (London 1992), 99f.CrossRefGoogle Scholar; for discussion of Ann. 1.10 see Luce (n.2 above), esp. 2925f.

26. Vespasian the sole exception according to Tac. Hist. 1.50.4.

27. Cf. Syme (n.1 above), 11, 35, 130; for the most recent discussion see Bennett (n.22 above), 46–48.

28. uideretur…senili adoptione inrepsisse, ‘he seemed to have crept in through senile adoption’, Ann. 1.7.10. Galba’s adoption of Piso would no doubt appear to present a much more obvious (and current, given the date of the Histories) parallel than that in the Annals, complete with the set speeches which elucidate Nerva’s own approach to the problem of succession. The episode, on the whole, would reflect favourably on Nerva and Trajan; the adoption of Tiberius by Augustus, however, invites an alternative (negative) response.

29. Thus the death of the four consulares at the opening of Hadrian’s reign thought by Syme (n.1 above, 484f.) to be parallel to the murder of Agrippa Postumus (primum facinus noui principatus, ‘the first crime of the new principate’, Ann. 1.6.1); Nero’s reign also begins in murder (prima now principatu mors, ‘the first death in the new principate’, Ann. 13.1.1), as do Galba’s and Otho’s. Indeed, the opening of Galba’s reign more nearly offers a parallel to Hadrian’s, for it opens with the death of two ex-consuls, Cingonius Varro and Petronius Turpilianus (Hist. 1.6) and the deaths of two governors, Clodius Macer, governor of Africa, and Fonteius Capito, governor of Lower Germany and consul in 67. Otho’s reign commences with the deaths not only of Galba and Piso, but of Titus Vinius, the consul (Hist. 1.42), Cornelius Laco, the Praetorian praefect, and Icelus (Hist. 1.46), a freedman of Galba’s—it appears that old scores were inevitably settled upon succession. Hence Vitellius’ succession is marred by the death of Cornelius Dola-bella (Hist. 2.63.1), due to his association with Galba and, presumably, his illustrious name (Hist. 1.88.1); his death, notes Tacitus, was one of the first acts to bring the regime into disrepute (Hist. 2. 64.1). Vespasian’s tenure as princeps opened with the senate’s mind collectively bent on settling old scores before Vespasian’s arrival and—in an episode similar to that at the opening of Hadrian’s reign—the senate was preparing to take its revenge autonomously on particularly offensive parties before the princeps arrived (Hist. 4.10–11; cf. HA Hadr. 7.2), and there was the usual rash of executions and suicides by prominent individuals, including the death of Calpumius Galerianus, son of Gaius Piso (Hist. 4.11.2), the suicide of Iulius Priscus, the Praetorian prefect (Hist. 4.11.3), and of Asiaticus, Vitellius’ freedman.

30. See also Bartsch (n.12 above), 164, for discussion of these two passages. The refusal motif was an old and specious one. See Aug. Res Gest. 5, 10 and 21.

31. See BMC, Imp. III, p.38, no. 55. There is a denarius dated to 98–99 with the head of Trajan on the obverse, while on the reverse is either Nerva or the senate with the legend TR. P. COS. II P.P. PROVID. Trajan’s second consulship will have been in 98, while Nerva’s third was with Verginius Rufus in 97. On his acceptance of the title see Plin. Pan. 21.If.

32. Waters, K.H., ‘The Character of Domitian’, Phoenix 18 (1964), 49–77CrossRefGoogle Scholar, esp. 67; see also Henderson (n.3 above, 177) for discussion of Tac. Ann. 1.72.2. Henderson is right when he notes that Tacitus’ (and Trajan’s) audience will have seen these types of recusationes for what they were—‘Imperial jabber’.

33. Syme (n.l above), 484.

34. Contra Shotter (n.2 above), who does not appear to view Tiberius as an archetype; see Dunkle (n.18 above), 12–20 (also contra Shotter).

35. See Syme (n.l above), 470f. and 496.

36. See the discussion on date in part I above.

37. Arrian Parthica fr. 39 Roos; Fronto Princ. Hist. 14–17 (= ii.212–16 Haines); Dio 68.20.3f.

38. See Magie, D., Roman Rule in Asia Minor, Vol. 1 (Princeton 1950), 476, 485, 496Google Scholar; Sherwin-White, A. N., Roman Foreign Policy in the East (London 1984), 326Google Scholar; Chaumont, M., ‘LéArménie entre Rome et l’Iran’, ANRW 2.9.1 (1976), 73–84.Google Scholar

39. See Fronto Princ. Hist. 14 = Haines ii.211.

40. Fronto Princ. Hist. 7 = Haines ii.206. Indeed, to judge from the sources, the reputation Trajan acquired from his eastern campaigns was less than flattering: sed etiam fortisssimi imperatoris Traiani ductu legatus cum exercitu caesus et principis ad triumphum decedentis haudquaquam secura nee incruenta regressio (‘But even a legate under the leadership of the strongest of emperors, Trajan, was killed with his army, and the retreat of the Emperor departing for his triumph was by no means secure and without bloodshed’, Fronto Princ. Hist 3 = Haines ii.202). The tradition of Trajan’s exploits in the East is not an unblemished record. See Amm. Marc. 25.8.5: prope Hatram uenimus, uetus oppidum in media solitudine positum, olimque desertum, quod eruendum adorti temporibus uariis Traianus et Seuerus principes bellicosi, cum exercitibus paene deleti sunt (‘We approached Hatra, an ancient city long deserted in the middle of a desert which the warlike emperors Trajan and Severus had attacked at different times for the purpose of destroying, although both nearly perished with their armies’). It should be noted that Trajan’s ingratitude towards his legates probably did not go unnoticed by the author of the Agri-cola, who had deplored Domitian’s ungracious behaviour towards his father-in-law.

41. It should be noted that lack of expansion does not mean Tacitus will de facto criticise an emperor. Thus it was Vespasian, not Claudius (the one emperor who did add a province in the early empire), who comes in for praise (relatively speaking: see Hist. 1.50.4).

42. Cf. Syme (n.1 above), 489 (citing Orosius 7.3.7 and Floras 1.47.4): Tacitus approves expansion, but prudent expansion.

43. Trajan’s sexual proclivities are well attested to in our sources (which are prejudiced against them), such as Dio 68.7.4, where he remarks upon Trajan’s passion both for small boys and wine; Dio in fact lists these at the end of a string of praises for the Emperor, after which he succinctly reports the peculiar set of less flattering details (for Dio as a historical source see Millar, F., A Study of Cassius Dio [Oxford 1964], 34–38, 62–72Google Scholar). In Julian’s Caesares Trajan’s sexual tastes are noted (and parodied) more than once; upon his entrance in the work Zeus is bidden to hide Ganymede (Iul. Caes. 311c). Elsewhere in the same work Silenus rails at Trajan for his love of ‘a pleasure most shameful and worthy of reproach’ (333a). Still less reliable sources report a similar tradition (specifically HA Hadr. 2.7). Trajan kept company with gladiators and dancers of pantomime (Dio 68.10.2). His ministers were little better—one, Licinius Sura, had had a liaison with a dancer, Philostorgos (Arrian Diss. Epicteti 3.17.4). We can presume that Trajan’s excessive love of games would not have been to Tacitus’ taste (nor, one suspects, to that of his audience), as his criticism of Drusus (Ann. 1.76.5–7) indicates. The games of 107 celebrating Rome’s victory over the Dacians were the largest ever seen, and were well remembered by later ages. See Anonymi Valesiani Pars Posterior 60: exhibens [Theodericus] ludos circensium et amphitheatrum, ut etiam a Romanis Traianus uel Valentinianus, quorum tempora sectatus est, appelaretur (‘[Theodoric] exhibiting games in the circus and amphitheatre, is even called Trajan or Valentinian by the Romans, whose times he followed’). For Tacitus’ negative sentiments on games and entertainment in general see Dial. 26.2f.; 29.3.

44. Though this did not prevent emperors, such as Nero, from taking to the stage. On the political expediency of the theatre in ancient Rome see Manning, C.E., ‘Acting and Nero’s Conception of the Principate’, G&R 22 (1975), 164–75Google Scholar. For further discussion see Yavetz, Z., Plebs and Princeps (Oxford 1969), 62–67, 103–29Google Scholar. For actors and their legal status see Dig. 3.2.1; Dio 67.3.1.

45. The passage is corrupt and has been restored to read illud eti<am commune ut>ri<que est uitio> datum, histriones ex urbe in Suriam accisse (Fronto Princ. Hist. 18 = Haines ii.214); Van Den Hout (Cornelii Frontonis, M.Epistulae [Leiden 1954], 199)Google Scholar reads illud eti<am> opprobrio ductum bello <incipiente> histriones ex urbe in Suriam accisse (‘It was considered reproachful that, at the start of the war, he summoned actors from the city to Syria’).

46. Tiberius expelled actors from Rome in 23; see Tac. Ann. 4.14.4.

47. Plin. Pan. 46.4.

48. See Bartsch (n.12 above), 84–86, for Cremutius Cordus; see also Luce, T.J., ‘The Causes of Bias in Historical Writing’, CP 84 (1989), 31.Google Scholar

49. By choosing to write about the early principate, moreover, Tacitus looks to his own time as a teleological conclusion to the oppressive nature of the early principate which has now, through the auspices of Nerva and Trajan, reconciled libertas et principatus. Tacitus thereby confers a type of legitimacy on Trajan. This view is challenged, however, by the open nature of Tacitus’ text.

50. Against Shorter (n.2 above), 3271: ‘Tacitus’ political creed persuaded him to identify evils not so much with the system as with individual emperors.’ Tacitus’ mistrust is institutional and based on caricature. For similar discussion and argumentation see Schmidt, E.A., ‘Die Angst der Mächtigen in den Annalen des Tacitus’, WS 16 (1982), 274–87.Google Scholar

51. See McCulloch, Harold Y. Jr., ‘The Historical Process and Theories of History in the Annals and Histories of Tacitus’, ANRW 2.33.4 (1991), 2928–48Google Scholar, esp. 2929: ‘Surely Tacitus tempered his view toward the reign of Trajan once he delved into the research and began the composition of his major historical works; the initial felicitas temporum of Trajan’s principate (Agr. 3.1 and cf. Hist. 1.1.4) cannot have withstood the inevitable onslaught of mala, despite the shortlived remedia against destructive impulses.’

52. So Bartsch (n.12 above), 84–86. For a recent study on the emergence of personality in Tacitus see Woodman, A.J., ‘Tacitus’ Obituary of Tiberius’, CQ 39 (1989)CrossRefGoogle Scholar, passim.

53. For the instability of social status during the principate, see e.g. Plin. Pan. 34.1, nullius status certus (‘no one’s status certain’). For the instability of a text’s interpretation see, e.g., Mam. Aemilius Scaurus’ tragedy (possibly of Atreus) and the attendant results (Tac. Ann. 6.29.4f.); see also Cremutius Cordus’ prosecution (Tac. Ann. 4.34–35); cf. Dio 67.12.5 for the execution of the sophist Maternus in 91 CE for writing a declamation concerning tyranny. For a good general discussion on the suppression of literature under the early principate see Cramer, F.H., ‘Bookburning and Censorship in Ancient Rome: A Chapter from the History of Freedom of Speech’, JHI 6 (1945), 157–96.CrossRefGoogle Scholar

54. Felicitas (‘happiness’) itself could also be a dubious concept. See Zieske, L., Felicitas: Eine Wortuntersuchung (Hamburg 1972), 303–06Google Scholar; he argues that the word is used for propaganda purposes, refers primarily to military success, but also has overtones of hybris. The word is similar to a campaign slogan. Such a meaning may not be too far wrong; the phrase felicitas temporum occurs in an edict of Nerva’s in 96 (Plin. Ep. 10.58.7).

55. Tacitus’ expressed hope in the Histories (1.1.4), to write about the reigns of Nerva and Trajan, subjects with materia securior et uberior, was dubious at best; uberior materia (‘richer material’) is an ambiguous phrase. And with the historic record—of Nerva’s role as poet and favoured courtier under Nero, of his humiliation as an aged emperor by the praetorians, of the secretive rise to power of Trajan with the armies of the Rhine at his back, and taste for less refined pleasures—what chance did either stand against Tacitus’ penetrating, if occasionally ambiguous scrutiny? For Nerva as a poet under Nero see Mart. Epig. 8.70, and cf. 9.26 and Pliny Ep. 5.3.5; for Nerva’s humiliation see Dio 68.3.3f.; for movements on the Rhine and their implications in 97, see Syme (n.l above), 632f. Nerva was also reported to have seduced Domitian (Suet. Dom. 1.1). For discussion of Nerva’s poetry see Dilke, O.A.W., ‘The Literary Output of the Roman Emperors’, G&R 4 (1957), 78–97, esp. 94.Google Scholar

56. Ann. 1.1.4–6. As Williams (n.l6 above, 162) has noted, Tacitus’ assertion of non-involvement and want of personal bias is instantly refuted by his question at 1.3.7, quotus quisque qui rem publicam uidisset, ‘How many were left, and who, who had seen the republic?’ For a discussion of the possible Sallustian influences in this passage see Woodman, A.J., ‘The Preface to Tacitus’ Annals: More Sallust?’, CQ 42 (1992), 567f.CrossRefGoogle Scholar

57. These considerations, it is true, did not prohibit his composing a history of more recent times; however, it must be noted that it was not a history of the current regime, but one in which events were set in the recent past under Nerva’s and Trajan’s discredited predecessor Domitian.

58. Cf. Bartsch (n.12 above), 122–25, who discusses a similar ambiguity in the Dialogus where she remarks Maternus’ two-edged encomium of Vespasian and its implications for the rest of Tacitus’ writings: ‘Given that Tacitus’ historical works are in an important sense about [her italics] the way political language masks its opposite and can be read as a glossary of the false official language of virtue imposed by imperial ideology upon its subjects, his own use of terms like freedom, happiness, and safety are always already undermined by their unveiling as hollow in the very works in which they appear.’

59. I would like to thank Ellen O’Gorman and Victoria Pagan for their useful comments and suggestions which helped me to improve this work, and the Graduate Research Board at the University of Maryland, whose generous support allowed me the time to research and complete it. I would also like to thank the readers and editors at Ramus for their kind advice from which this essay has benefited. Any errors are the author’s fault alone.