Hostname: page-component-74656f9884-npj9w Total loading time: 0 Render date: 2026-11-07T17:30:56.067Z Has data issue: false hasContentIssue false
  • English
  • Français

FABIUS AND MINUCIUS IN TACITUS: INTERTEXTUALITY AND ALLUSION IN ANNALS BOOK 15

Published online by Cambridge University Press:  16 October 2017

Arthur J. Pomeroy
Arthur J. Pomeroy*
Affiliation:
Victoria University of Wellington
*
Arthur.Pomeroy@vuw.ac.nz
Rights & Permissions [Opens in a new window]

Extract

Roman conflict with Parthia in the mid first century for control of Armenia and Domitius Corbulo's exploits in the East, culminating in the Parthian candidate for the throne, Tiridates, receiving his diadem from the hands of the Emperor Nero in Rome, have frequently been studied for what they reveal about military and diplomatic manoeuvres under the later Julio-Claudians. The historiographical investigation of our main source, Tacitus, particularly through comparison with the fragments of Cassius Dio, is also important for the light this sheds on the Roman senator's methods. The intention of this paper is to draw attention to the complexity of Tacitus’ account by indicating his use of multiple and sometimes contradictory viewpoints in his narration of Caesennius Paetus’ unsuccessful Armenian campaign of a.d. 62–63 (Ann. 15.1–17) and to highlight an unrecognized echoing of his predecessor Livy by the historian. The examination of a text that is as elusive as it is allusive will require a careful study of what must have been a debated incident, but, as I hope to show, will give a broader insight into the historian's methods and, perhaps, wider intentions. I begin with an outline of historical events and the sources available to the historian. Next, I present a narratological examination of the text to show how differing viewpoints are highlighted within a contested storyline. I then show how this confusion may appear to be dispelled by a strong plot-line that borrows heavily from Livy, wherein the actions of an older, experienced general are contrasted with those of his subordinate. However, aspects of this account suggest that the tensions within the narrative are not resolved and that this has wider implications for our understanding of the historian and his work.

Information

Type
Research Article
Information
The Classical Quarterly , Volume 67 , Issue 2 , December 2017 , pp. 583 - 596
Check for updates
Copyright
Copyright © The Classical Association 2017 

Roman conflict with Parthia in the mid first century for control of Armenia and Domitius Corbulo's exploits in the East, culminating in the Parthian candidate for the throne, Tiridates, receiving his diadem from the hands of the Emperor Nero in Rome, have frequently been studied for what they reveal about military and diplomatic manoeuvres under the later Julio-Claudians.Footnote 1 The historiographical investigation of our main source, Tacitus, particularly through comparison with the fragments of Cassius Dio, is also important for the light this sheds on the Roman senator's methods.Footnote 2 The intention of this paper is to draw attention to the complexity of Tacitus’ account by indicating his use of multiple and sometimes contradictory viewpoints in his narration of Caesennius Paetus’ unsuccessful Armenian campaign of a.d. 62–63 (Ann. 15.1–17) and to highlight an unrecognized echoing of his predecessor Livy by the historian. The examination of a text that is as elusive as it is allusive will require a careful study of what must have been a debated incident, but, as I hope to show, will give a broader insight into the historian's methods and, perhaps, wider intentions. I begin with an outline of historical events and the sources available to the historian. Next, I present a narratological examination of the text to show how differing viewpoints are highlighted within a contested storyline. I then show how this confusion may appear to be dispelled by a strong plot-line that borrows heavily from Livy, wherein the actions of an older, experienced general are contrasted with those of his subordinate. However, aspects of this account suggest that the tensions within the narrative are not resolved and that this has wider implications for our understanding of the historian and his work.

EVENTS AND SOURCES

The basic outline of events in the East that led to conflict between Rome and Parthia over Armenia is scarcely in dispute. Domitius Corbulo, having been dispatched by Nero to deal with the Armenian problem, had expelled the Parthian king Vologaeses’ brother, Tiridates, from the country in the late 50s a.d. and installed on the throne a Roman nominee, Tigranes. The appointment in a.d. 62 of Caesennius Paetus as commander with the task of defending Armenia was the Roman response to renewed Parthian activity in the 60s a.d. Paetus was to act in consultation with Corbulo, now appointed governor of Syria. The two divided between them their recently increased forces, around three legions going to each. Corbulo guarded the Euphrates crossings into Syria, even occupying high ground across the river in enemy territory, while Paetus invaded Armenia on a punitive expedition in response to the Parthian king's restoration of Tiridates. However, after skirmishing with Vologaeses’ army, Paetus withdrew past the Taurus mountains and reluctantly requested assistance from Corbulo. Vologaeses, after breaking through the Roman defence lines, besieged Paetus’ forces at Rhandeia near the river Arsanias.Footnote 3 Eventually Paetus capitulated on terms that involved the return of all booty taken during his expedition and the abandonment of any claims on Armenia, before retreating south to meet Corbulo's relief force. A year later, Corbulo, now sole commander in the East after the dismissal of Paetus, met Tiridates at the site of Paetus’ last camp and negotiated terms that involved the Parthian claimant travelling to Rome to receive official approval from the emperor of his occupation of Armenia.

That Roman forces were forced to leave Armenia in a.d. 62 was clearly seen as a stain on national honour, even if attempts may have been made to portray the result as closer to a draw than a defeat, since the Parthians had had to withdraw from Armenia in exchange for the removal of Roman military installations across the Euphrates (Ann. 15.17.3). However, as Tacitus indicates in a telling postscript, an unwillingness to halt the construction of a planned victory monument on the Capitoline only underlined the lack of Roman success (Ann. 15.18.1). Modern scholars place the blame firmly on Caesennius Paetus for failing to respond appropriately to the military situation he was facing. So Devillers claims that Tacitus never misses an opportunity to underline Paetus’ incompetence in order to stress Corbulo's talents when saving the Roman forces from complete defeat.Footnote 4 One explanation for this is that given by Ronald Martin, who believes that the historian's ultimate source for this interpretation of events was Corbulo himself.Footnote 5 Yet, as will be seen, Tacitus’ narrative is much more ambiguous than such remarks suggest.

To begin, the view that Corbulo is directly or indirectly the major source of Tacitus’ account needs re-examination, especially given that even modern researchers sometimes still view the historian as a mere compiler of earlier written accounts.Footnote 6 In reality, there was a range of sources that Tacitus could draw on and it is likely that his audience, many of whom had connections with the actors in these events, would be quick to make their feelings known about material with which they disagreed.Footnote 7 Corbulo is certainly one source, since Tacitus declares that the general was able to make the claim (prodiderit), based on the agreed fact (constitit) that, since the Romans had plentiful grain in reserve, the Parthians would have abandoned their siege from lack of supplies if Paetus had not come to terms with Vologaeses. Anyway, he and his army were only three days’ march away on their relief mission (Ann. 15.16.1).Footnote 8 This statement is sometimes described as originating in Corbulo's memoirs, but Tacitus’ use of the potential subjunctive suggests that the actual statement could equally well have come from official dispatches or even have been recounted by one or more of his supporters.Footnote 9 Moreover, Paetus presumably put his own slant on affairs, as the suggestion in Tacitus that Corbulo was initially slow to assist (Ann. 15.10.4) indicates.Footnote 10 Other informants were available as either written or oral sources. Pliny the Elder, for instance, criticizes the use of the term ‘Caspian Gates’ by people who had served with Corbulo in Armenia, since the pass should actually have been called the Caucasian Gates (HN 6.23). Clearly he knew this usage from oral accounts or had read the works of members of this group. The hostile aspects of the portrayal of Ummidius Quadratus, governor of Syria, in Ann. 13.8–9 are likely to derive from a source associated with Corbulo. On the other hand, the accusation that in seizing the hostages, whom the Parthian king had surrendered under an agreement with Quadratus, Corbulo was stealing someone else's glory may stem from official reports from the Syrian legate or have been a grudge kept alive by the governor's daughter, the long-lived socialite Ummidia Quadratilla.Footnote 11 Whoever depicted the senior centurion Paccius Orfitus as being berated by Corbulo for incompetence and cowardice when faced by the Parthians, not once but twice (Ann. 13.36 and 15.12.2), surely favoured the general. As Verwaet has shown,Footnote 12 there were many supporters of Corbulo in high positions under the Flavian regime, most notably Licinius Mucianus and Corbulo's daughter, Domitia Longina, the wife of Domitian who survived well into Hadrian's reign. It can readily be inferred that Crispinus, the son of Vettius Bolanus, was keen to advertize his father's military service under Corbulo in Armenia given its prominent inclusion in Statius’ laudatory poem for the son (Silv. 5.2.31–50). The opinions of Funisulanus Vettonianus, commander of the Fourth Scythian Legion under Paetus (Ann. 15.7.1) and later decorated for services in the Dacian Wars of the late 80s a.d. (CIL 11.571), might not have been so favourable. Those of the sons of Paetus, the homonymous L. Iunius Caesennius Paetus (suffect consul in a.d. 79, I.Smyrna 2.826), and Caesennius Sospes, whose name suggests that he was the military tribune humiliatingly chosen by Corbulo to bury the dead left behind after his father's defeat (Ann. 15.28.2) and later praetorian legate in Galatia (CIL 3.6818)—perhaps also the suffect consul of a.d. 114 (RMD 1978 no. 14)—would definitely have been hostile. In brief, while after Corbulo's death a virtual hagiography was constructed of the Neronian general that is most evident in Dio (62.23.5–6, 63.6.4), there were also many who criticized him as arrogant and self-serving. Tacitus would have been aware of these conflicting descriptions from written sources and, equally importantly, from conversations in his social circles. It must not be forgotten that the major advantage of being a senatorial historian in Imperial times did not lie so much in access to privileged information about political affairs as in interaction with one's peers, most obviously in Tacitus’ case, but surely not limited to, his membership of the college of quindecimuiri sacris faciundis.Footnote 13 If, as Birley has argued, Tacitus had the cognomen, Caecina, and an additional cognomen, Paetus, he may have had a range of relatives to praise and offend.Footnote 14

TACITUS’ PRESENTATION

The topic of Roman-Parthian relations, particularly with regard to Armenia, was also highly relevant, given Trajan's campaigns in the region and Hadrian's consequent abandonment of his predecessor's conversion of the kingdom into a Roman province.Footnote 15 Of course, we do not know exactly when Tacitus was composing Annals 15, but a time after Trajan began preparing for eastern intervention is almost certain. If, as Birley argues,Footnote 16 the earliest date for the publication of the Histories should be set as a.d. 112, then the later books of the Annals would definitely be Hadrianic. However, it is difficult to read any specific comment on contemporary events from Tacitus’ description of the a.d. 62 campaign. On the basis of the omens reported at Ann. 15.7, Meulder thought that Paetus’ invasion of Armenia was being treated by Tacitus as an impious attack in the manner of those launched by Crassus or, later, Julian.Footnote 17 Meulder linked this with Dio's description of Trajan's campaigns to indicate how that emperor too could be seen as acting against the wishes of the gods. Yet, apart from the methodological problem of attributing Dio's opinions to his historical predecessor, it must also be noted that the divine signs in Tacitus are directed against Paetus and Paetus alone. Corbulo's actions (which had included an invasion of Armenia in a.d. 58–60) are nowhere subjected to the sanction of the gods. Indeed, despite the obvious topical connotations of Armenia, it is difficult to detect the historian's judgement on contemporary imperial policy in the East. Likewise the biographer, Suetonius, writing in the same period, refrains from any evaluation of international relations in the area.Footnote 18

Leaving aside the wider implications of these events, Tacitus appears, nevertheless, in line with the normal pattern of exemplarity in Roman history, to be portraying the actions of Paetus as doomed from the beginning. To quote Livy (Praef. 10), the enterprise was foedum inceptu foedum exitu. Certainly the use of counterfactuals offers a critical description of his conduct.Footnote 19 Paetus gave out (ferebat, Ann. 15.8.1) that he was invading Armenia to recover Tigranocerta and to plunder the areas that Corbulo had left intact. He did capture a number of strongholds and gain significant military renown and booty—but he needed to have kept his glory-hunting within limits and take care of his plunder (si aut gloriam cum modo aut praedam cum cura habuisset, Ann. 15.8.2). This is not the tension-raising alternative that Jonas Grethlein has labelled Beinahe Footnote 20 but rather the same sort of gap between ideology and performance that appears in Tacitus’ famous evaluation of Galba as capax imperii nisi imperasset (‘a man who could have been a good emperor except for the fact that he became one’, Hist. 1.49.4). Similarly, the need to call up the Twelfth Legion (Ann. 15.10.1), rather than strengthening Paetus’ position, actually exposes the reality of the limited forces under his command when faced by the Parthian army. This point is reiterated when Paetus later attempts to bluster his way out of the siege, suggesting that, while Vologaeses might well have attacked in person (ipsum) and with all the forces of his kingdom Paetus’ mere two legions, the Romans still had the ability to provide Paetus with reinforcements from the rest of the world (15.13.3). Even given this disparity in forces, the historian opines, the camp could have been defended and the Parthians frustrated by dragging out the war, ‘if Paetus had shown a clear intention to follow a single plan whether it was the one he had proposed or that of others’ (si Paeto aut in suis aut in alienis consiliis constantia fuisset, 15.10.1). Indeed, the whole campaign seems to offer alternative scenarios. If Paetus had not divided his army by sending some of his men to guard his wife and son at Arsamosata, the combined forces would have been better prepared to meet the enemy's waves of assaults (in uno habitus uagum hostem promptius sustentauisset, 15.10.3).

However, the voice of the narrator also views the acts of others as contributing to the result. Corbulo was slow to respond since the growing danger to Paetus’ forces also increased the glory to be gained through their rescue (quo gliscentibus periculis etiam subsidii laus augeretur, 15.10.4). After Roman capitulation, Corbulo complains that the war could have been concluded with the rout of the Parthians (potuisse bellum fuga Parthorum finiri), a chance now missed; Paetus counters that neither general has had his hands tied and they could jointly make an invasion of Armenia now that Vologaeses has withdrawn (ille integra utrique cuncta respondit: conuerterent aquilas et iuncti inuaderent Armeniam abscessu Vologaesis infirmatam, 15.17.1). In brief, the argument has broken down to ‘he said, I said’, and the focalizers for the narrative need to be carefully considered. When Corbulo refuses to accept the excuses of Roman stragglers, he sends them back to Paetus, declaring that he only showed mercy to the victorious (se nisi uictoribus immitem esse, 15.12.2). There is an indirect suggestion here that Paetus is already defeated. Common belief (rumor) exaggerated events to the extent of having the Roman legions sent under the yoke. This disgrace is denied, but the behaviour of the Armenians in reclaiming war booty from the Roman army gives the appearance of the aftermath of a military disaster (alia ex rebus infaustis, quorum simulacrum ab Armeniis usurpatum est, 15.15.2). The Roman granaries are described as undoubtedly (constitit) still filled with grain at a time when, so Corbulo claimed, the Parthians were about to abandon the siege from lack of supplies. In addition, Paetus is criticized for solemnly taking an oath before the standards and in the presence of Parthian representatives that no Roman would enter Armenia until Nero had indicated whether or not he approved of the peace terms (15.16.1–2). Tacitus characterizes this version of events as one circulated by Corbulo to discredit his co-commander further (augendae infamiae composita, 15.16.3). However, there is no doubt (focalizing as the historian) that Paetus force-marched his troops forty miles in one day, leaving the wounded behind along the way, and the general state of his forces when they met Corbulo was no different from soldiers retreating after being defeated in battle (15.16.3). This leads to group-focalization, as Corbulo's men cannot hold back their tears when they see their fellow legionaries (uicem commilitonum miserantes ne lacrimis quidem temperare). Competitive heroism and rivalry for fame (certamen uirtutis et ambitio gloriae) are emotions of the fortunate (felicium hominum adfectus), that is to say, the senatorial commanders, who are now laid aside. For lesser men, pity outweighed all other feelings (sola misericordia ualebat, et apud minores magis, 15.16.5).

The concentration on the mass might suggest that typical senatorial history with its accounts of the victories of notable commanders might actually be irrelevant. Yet, Tacitus is not creating a military history that equates with the modern mythologizing of the First World War to emphasize the suffering of ordinary soldiers as a result of the stupidity of their generals. Senators always had a duty of care for their clients and senatorial commanders for their armies. Rather, the whole narrative of the Armenian campaign with its ‘might-have-beens’ and shifting viewpoints maintains the tension between escape and disaster and reaches its apex with this final cathartic scene. I will now argue that, along with his use of the narratological devices as indicated above, the historian also constructs his account around a recognizable historical narrative that creates an apparent unity out of highly disparate elements.

INTERTEXTUALITY

The treatment of Paetus’ Armenian expedition is exemplary, in the sense of offering a warning of how not to fight a war. However, the use of exemplars in itself has a tendency to create a chain reaction of allusion in Roman historical writing. Thus, Tacitus’ portrait of Sejanus in Ann. 4.1 draws not only on Sallust's character-sketch of Catiline (Cat. 5) but also on Livy's description of Hannibal (21.4), itself based on Sallust's Catiline.Footnote 21 It is this aspect of Tacitus’ historical method that is worth exploring in this case. If we break the campaign down into its narrative tropes, this is the result:

  1. 1. An army is divided under two commanders who cannot cooperate (Ann. 15.6.4).

  2. 2. The rash commander boasts that he will outdo the achievements of his predecessor (15.6.4).

  3. 3. The campaign begins with bad omens (15.7).

  4. 4. The rash commander ignores the omens and without sufficient preparations hastens to begin his campaign (15.8.1).

  5. 5. Initial successes create a false confidence verging on arrogance (15.8.2; cf. si … gloriam cum modo … habuisset; composuit … litteras quasi confecto bello, uerbis magnificis, rerum uacuas ‘if only he had set a limit on his glory-hunting; he concocted letters in brilliant prose that made it appear that the war was over but which ignored the facts’).

  6. 6. The wiser commander maintains his vigilance against the enemy (15.9).

  7. 7. The enemy chooses to attack the rash commander and takes him by surprise (15.9.2).

  8. 8. The rash commander shows a lack of resolution (Paetus cannot settle on a plan of attack or defence and divides his forces, weakening them, 15.10).

  9. 9. The rash commander is forced to appeal to his wiser colleague (15.10.4).

  10. 10. The wise commander rushes to the rescue (subverted here, as Corbulo takes his time, 15.10.4).

  11. 11. The enemy's initial attack is successful (despite the individual aristeia of Tarquitius Crescens, 15.11.1).

  12. 12. The troops of the rash commander are routed (15.11.2).

  13. 13. The appeal to the wiser colleague is repeated with some desperation (15.11.3).

  14. 14. The wise commander rushes to the rescue (15.12.1).

  15. 15. The wise commander encourages his troops to rescue their fellow citizens (15.12.3).

  16. 16. The wise commander meets retreating troops from the other army and encourages them to return to the fight (15.12.2). The enemy besieges the troops of the rash commander who refuse to fight (15.13.1–2, a variant on #12).

    1. 17. The rash commander appeals for terms from the enemy (here, Paetus is forced to act by his troops but his missive shows continuing arrogance, 15.13.3).

    2. 18. The enemy commander responds coldly (15.14.1–2); eventually terms are agreed (15.14.3).

    3. 19. The rash commander undergoes humiliation (subverted: having to swear an oath before the standards to withdraw from Armenia is Corbulo's version, reported later, 15.16.2; denied: Paetus claims that building a bridge over the Arsanias was to expedite his withdrawal, 15.15.1).

    4. 20. The rash commander's army is humiliated (the plundering of the camp by the Armenians, rumour of sending under the yoke, 15.15.2).

    5. 21. The enemy commander refuses to be personally involved in the humiliation of the defeated foe (partially subverted here: Vologaeses does not watch the legions’ retreat but crosses the Arsanias on an elephant to display his power).

    6. 22. The army of the rash commander flees in disorder (15.16.3).

  17. 23. The army of the wise commander arrives to rescue their comrades (subverted: the defeat has already occurred, 15.16.4; Corbulo stresses that surrender was unnecessary and especially disgraceful, 15.16.1–2).

  18. 24. The soldiers of the two armies embrace one another in joy, the rescuing army refusing to criticize their fellows (15.16.4).

  19. 25. The two generals embrace and the rash commander admits his mistake (subverted: Paetus refuses to admit his mistakes; Corbulo insists on returning to defend Syria, 16.17.1–2).

Excluding numbers 17–22 (an embedded story of a siege and surrender), the account is a traditional one of a rash warrior and a wiser leader, a theme that dates back to Homer and is famously exploited by Thucydides in his portraits of Alcibiades and Nicias.Footnote 22 This typical historiographical construction has, for instance, been recognized by Kristine Gilmartin and Rhiannon Ash.Footnote 23 Furthermore, this also would appear to be the way in which Dio narrated events (62.21–2  =  Exc. U r 12 page 392). Yet, the later historian's brief account actually seems to be closer to Corbulo's alternative version in Tac. Ann. 15.16.1–2 than to Tacitus’ nuanced reconstruction. In Dio, Paetus is forced to retreat from Tigranocerta by Vologaeses’ superior forces and is besieged. Nevertheless, Vologaeses is on the point of withdrawing owing to lack of supplies, and even after Paetus’ capitulation he has to let the Roman army go free since Corbulo is quickly approaching. The latter arrives at the Euphrates and waits for Paetus to join him. The scene of the forces reuniting emphasizes the pride of one army in its speed and the despondency of the other in its humiliation owing to the fault of its commander. The fault is Paetus’ and Paetus’ alone.

Tacitus, by contrast, is clearly drawing parallels with earlier Roman history in his account, by recalling scenes from Livy, some indirectly but two specifically.Footnote 24 Under siege, Paetus’ soldiers restrict themselves to guarding their positions, some under orders but others from cowardice and hoping for Corbulo's arrival. If pushed (uis si ingrueret), they have in mind the outcomes at the Caudine Forks and Numantia (prouisis exemplis Caudinae Numantinaeque  <pacis> Footnote 25 ). The tone is satiric (a Caudina pax is hardly a pax Romana, particularly when one remembers that both the Caudine and the Numantine treaties were disowned by the Senate), a point underlined by the belief that ‘it was a reliable and respectable plan of action in the past whenever luck went against you to consider how to save yourself’ (ualidam quoque et laudatam antiquitatem, quotiens fortuna contra daret, saluti consuluisse, 15.13.2). The Samnite Pontius’ victory at Caudium (Livy, Book 9) and Mancinus’ disaster at Numantia (from the now-lost Livy, Book 55), two events with strong intertextual connections within Livy,Footnote 26 are clearly being referenced here.

However, the Livian reminiscences in this episode do not stop here. MeulderFootnote 27 had already seen that the omens that greet Paetus when he invades Armenia are similar to those described when Flaminius set off on his ill-fated campaign against Hannibal in 217 b.c. Paetus’ horse carrying the consular insignia turns tail when crossing the Euphrates bridge; during the construction of winter camp the sacrificial victim breaks out of the camp; Paetus’ soldiers’ javelins seemed to be on fire (arsere), a particularly disconcerting prodigy, given the Parthian preference for fighting with missiles (Ann. 15.7.2). In Livy, the ill omens are briefly separated in the narrative. First, the sacrificial victim escapes when Flaminius enters office at Ariminum on 1 March 217 b.c. (21.63.13–14). There then follows a lengthy description of Cn. Servilius’ traditional entry into the consulship at Rome and the attempt to placate the gods for only a single consul seeking divine auspices, before the narrative returns to Flaminius setting out for Etruria in spring. Here the omens are clearly unfavourable: Flaminius falls from his horse and the standards cannot be pulled out of the ground (Livy 22.3.11–14).Footnote 28

However, in terms of exemplars, an even stronger narrative parallel exists between Tacitus’ account of Paetus’ Armenian campaign and the well-known tale of the division of command between Fabius Maximus and Minucius in the same year as Flaminius’ disaster at Trasimene (Livy 22.24–30). Minucius, the magister equitum to Fabius, achieves success at Geronium, although in Livy's version it is a Pyrrhic victory that he advertises as a glorious success (cf. Paetus’ missives to Nero, ‘written as though the war were over in splendid style but empty of real accomplishments’).Footnote 29 The tribune M. Metilius seems to suggest that Minucius and his troops are at another Caudium thanks to Fabius’ unwillingness to fight (clausos prope intra uallum retentos; tamquam hostibus captiuis arma adempta, Livy 22.25.8), a pattern which is actually realized in Tacitus when Paetus is besieged at Rhandeia. As a result of domestic pressure, the Roman forces facing Hannibal are divided between Minucius and Fabius, who strongly opposes his subordinate's actions. Hannibal sets a trap for Minucius, but Fabius, who has been watching events, intervenes with his troops, encouraging them with the thought of defeating the enemy and forcing Minucius and his men to admit the error of their ways (cf. Corbulo's exhortation to his troops in Ann. 15.12.3). In turn, Minucius’ men take heart, even rallying from flight,Footnote 30 and form a single unit in combination with Fabius’ army. Faced with such resistance, Hannibal gives the signal for his troops to withdraw. When Minucius returns to camp, he assembles his men and declares that he has learnt to obey his betters and intends to reunite the camps under Fabius’ command. All march to the dictator's camp where Minucius addresses Fabius as parens in recognition of his superior understanding and status, while his soldiers salute their peers as clients would their patrons (Livy 22.30.7):

tum dextrae interiunctae militesque contione dimissa ab notis ignotisque benigne atque hospitaliter inuitati laetusque dies ex admodum tristi paulo ante ac prope exsecrabili factus.

Then they shook hands and after the assembly was dismissed (Minucius’) soldiers were invited in as guests in a spirit of generosity both by those who knew them and by strangers; the day turned to one of rejoicing after just recently having been ominous and verging on the unmentionable.

The day ends in celebration among the soldiers. Furthermore, in contrast with Minucius’ earlier claims, Fabius has won true glory that is now acknowledged in Rome. Even Hannibal has to accept that he is fighting with a worthy opponent (22.30.8–10). Tacitus seems to be deliberately frustrating the reader's desire for a happy ending: when Paetus’ army is reunited with that of Corbulo, the predominant emotion is pity (maesti manipuli ac uicem commilitonum miserantes ne lacrimis quidem temperare, Ann. 15.16.4).

Apart from the general narrative form, there are other indications that Tacitus had this episode in mind when framing his account of Paetus’ campaign. Perhaps surprisingly, it is Vologaeses who is described as cunctator ingenio (15.1.1);Footnote 31 yet, with his cunning and unwillingness to take risks the Parthian king is more like Fabius than his rash Roman opponent. In Livy, Hannibal provokes the division of command between Fabius and Minucius by acting like Fabius at Gereonium (artibus Fabi sedendo et cunctando bellum gerebat, 22.23.10). The dissonance between reports reaching Rome and the strategic reality (cf. Livy's description of Minucius’ victory in 22.24.14) is enhanced by Tacitus’ account of the construction of a victory monument and triumphal arch on the Capitoline, continued even after the truth was known (15.18.1). This is even more notable when Parthian envoys arrive at Rome in a.d. 62 with their version of events that refutes Paetus’ earlier dispatches. Corbulo is given full control of the East, ‘because of the expertise he had gained over the years with regard to his soldiers and the enemy and because, after the loss of confidence in Paetus, there was no wish to have someone else out of ignorance making the same mistakes again’ (ne cuius alterius inscitia rursum peccaretur, 15.25.2).Footnote 32

By the late Republic the exemplary nature of the story of Fabius and Minucius was sufficiently well established that Livy could expect his readers to note the parallels when he described the conflict between Camillus and his magister equitum, L. Furius in the war against the Volsci of 381 b.c. (6.22–5).Footnote 33 Given the increasing recognition of Tacitus’ indebtedness to Livy, his use of the well-known tale of Fabius and Minucius should hardly come as a surprise.Footnote 34 Livy's tale is perfectly exemplary in teaching respect for one's superiors and obedience to the traditional chains of command, linking senatorial values with the more general Roman systems of patronage and familial order. Popular politics are shown as being, at heart, un-Roman.

It is common for allusion to Republican history to be seen as a negative judgement on the process of Imperial government.Footnote 35 Even if traces of a version of events that portrayed Corbulo as a Republican-style hero can be detected,Footnote 36 Tacitus by using Republican exemplars need not be showing nostalgia for that past. In fact, in his description of Corbulo's character, he is clearly indicating that the competitive values of the Republic are not something to emulate.Footnote 37 While Paetus should have been content to play second fiddle instead of denigrating Corbulo's achievements and asserting that he would impose real Roman rule on Armenia, Corbulo himself at the same time could not brook a rival (neque Corbulo aemuli patiens, 15.6.4). If this is a ‘Republican’ image, it is one that draws on Lucan's depiction of Caesar and Pompey, one brooking no superior, the other no equal (Luc. 1.125–6). Corbulo's tardiness in setting out to assist in order to allow the danger to grow and so increase his own glory as rescuer (neque a Corbulone properatum, quo gliscentibus periculis etiam subsidii laus augeretur, 15.10.4) is hardly the unselfishness of Fabius in Livy 22.29.1–2. Furthermore, Republican exemplars are even mocked when used as models for self-preservation by Paetus’ troops (common soldiers, it should be noted, who appear in Tacitus’ account to be very well trained in declamatory skills). The narration of the exploits of centurions as a reflection on their commander (Paccius Orfitus shows himself as cowardly as his commander in leading a charge to the rear at 15.12.2, while Tarquitius Crescens alone of Paetus’ soldiers defends his post even at the cost of his life at 15.1.1) appears to follow a late Republican historiographical innovation, as commanders wished to display their close bonds with their troops rather than general loyalty to the Republic.Footnote 38 Finally, there is no sudden peripeteia in this story: Paetus and his men suffer a humiliating defeat. Yet, Paetus remains in a state of denial about what has occurred, even hiding the outcome of his campaign in his reports to Rome, while Corbulo himself shows no empathy for his colleague's plight.Footnote 39

A SENATORIAL HISTORIAN

While it would be easy to dismiss the allusions and complex intertextual relationship between Tacitus and the work of Livy merely as a sign of the later writer displaying his influences, that would be to downplay the significance of the whole episode.Footnote 40 Paetus’ Armenian campaign ended in the greatest military humiliation for the Roman state since Augustan times. As with defeats in modern times (for instance, the disastrous Dardanelles campaign of the First World War that has since been elevated to become the origin of Turkish, Australian and New Zealand nationalism), the outcome could be either downplayed as insignificant or blamed on the incompetence of a single commander (here Paetus) to highlight the qualities of another general (Corbulo). Some indication of the first approach can be seen in Tacitus’ account. Corbulo, like Fabius, is confirmed in his command after the failure of his rival, while Paetus’ defeat simply becomes the subject of an imperial bon mot. Nero offered an immediate pardon, ‘in case, given his tendency to panic, Paetus’ health might suffer from prolonged anxiety’ (ignoscere se statim, ne tam promptus in pauorem longiore sollicitudine aegresceret, 15.25.4). This is a private matter; for public consumption, the victory monuments remain.

The second approach is also visible and, though restrained in Tacitus, with some reason. In Silius Italicus’ quite recent recounting of the Fabius and Minucius story, Fabius had gained a semi-divine status at its conclusion, just as Scipio does at the end of the Punica.Footnote 41 There the general who had triumphed over Hannibal is compared to the victorious Dionysus or Hercules, a true son of Jupiter, in the line of Romulus and Camillus. The conclusion is obvious. These are the Republican avatars for the Flavian emperors.Footnote 42 It would have been possible to follow this pattern and portray Corbulo as capax imperii—it is likely that some Flavian accounts dimly visible in Dio treated him so. Still, whether as a deliberate answer after the death of Domitian to previous Flavian propaganda or as an acknowledgement of the difficulty in sorting out conflicting claims, Corbulo is no saviour in Annals 15.Footnote 43

Any reader of Tacitus’ account would have been aware of the outcome of events in Armenia. However, by structuring his narrative as a rescue mission—indeed, Corbulo uses the analogy of a ciuica corona in 15.12.3—the historian maintains the fantasy that events might turn out otherwise as long as possible. This is fully consonant with Tacitus’ mimetic method as noted by many scholars.Footnote 44 But, along with this overarching counterfactual, the historian also indicates the choices, real or imagined, that were given to the participants in events. This should not be viewed as a failure in methodology.Footnote 45 While Sailor has suggested that Tacitus by his writing constitutes himself as a virtual princeps,Footnote 46 I would suggest that the depiction of Paetus’ campaign most clearly shows him as a senatorial historian. Relations between the emperor and his provincial commanders are tangential to events in Ann. 15.1–17, and criticism or approval of imperial strategy is impossible to detect.Footnote 47 The claims and denials that men of senatorial status presented are much more obvious. Embedded within Roman society, Tacitus could not afford to be too partial in his version of highly controversial events. From the Agricola onward, his historical writing is concerned as much about the behaviour of the senatorial class under the system of the Principate as with the deeds of the emperors. Sometimes senators behaved egregiously badly, other times with remarkable courage.Footnote 48 But much of what they did showed the possibility of both virtues and faults. Relatives and friends of those who participated in the Armenian campaign might not wholly agree with the historian's version of events, but they could feel that the blame was spread more widely than previously. That would be enough to assuage any resentment at damage to family honour and bring them back to the next public recitation of the Annals.

References

1 For Corbulo's exploits, see most recently Vervaet, F.A., ‘CIL IX 3426: a new light on Corbulo's career, with special reference to his official mandate in the East from ad 55 to ad 63’, Latomus 58 (1999), 574–99Google Scholar; id., ‘Domitius Corbulo and the senatorial opposition to the reign of Nero’, AncSoc 32 (2002), 135–93Google Scholar; id., ‘Domitius Corbulo and the rise of the Flavian dynasty’, Historia 52 (2004), 436–64Google Scholar. A detailed bibliography is conveniently compiled in Cornell, T.J., The Fragments of the Roman Historians (Oxford, 2013)Google Scholar [hereafter FRH], 1.538. For Tacitus’ depiction of Corbulo, see Ash, R., ‘Following in the footsteps of Lucullus? Tacitus’ characterization of Corbulo’, Arethusa 39 (2006), 355–75CrossRefGoogle Scholar.

2 Sources for events: Tac. Ann. 13.8–9 (a.d. 54), 34–41 (a.d. 58; cf. Cass. Dio 62.19); 14.23–6 (a.d. 60); 15.1–17 (a.d. 62; cf. Cass. Dio 62.19–22), 24–9 (a.d. 63; cf. Cass. Dio 62.23). Tiridates’ coronation in a.d. 66 is adumbrated in Ann. 16.23, but the text of the Annals breaks off before the actual event, leaving us to rely on the description in Cass. Dio 63.1–7 (= Xiphilinus 172–5).

3 Cass. Dio 62.21.1; Tacitus’ only specific geographical details are that Paetus placed his wife and son in the safety of the fortress of Arsamosata (Ann. 15.10.3) and that the river Arsanias flowed near the main Roman camp. It was close enough for the Romans to build a bridge across the river during negotiations with the Parthians, although they actually took another route in leaving the area (Ann. 15.15.1). All indications are that the scene of conflict should be on the borders of Cappadocia-Sophene and Armenia, perhaps on the road from Amida to Melitene where Paetus could have attempted to block access to the Arsanias river valley. The general area is indicated by Vologaeses’ ability both to besiege the Roman camp and threaten the castellum of Arsamosata (15.13.4). Dio specifies the site of the main Roman camp as the otherwise unknown Rhandeia, ‘near the river Arsanias’ (62.21.1).

4 Devillers, O., L'art de la persuasion dans les Annales de Tacite (Collection Latomus 223) (Brussels, 1994), 91 Google Scholar.

5 For instance, Martin, R., Tacitus (London, 1988), 180 Google Scholar: ‘In 15.16.1 Tacitus expressly names Corbulo as a source for one of his details. It is hard to resist the conclusion that for much of the eulogistic matter and tone the same source was responsible.’

6 So Power, T., ‘Suetonius’ Tacitus’, JRS 104 (2014), 205–25, at 218Google Scholar: ‘the twice removed version [of history] of Tacitus was superfluous, derivative and, given the rhetorical nature of Roman historiography, doubly tainted. It would have been deemed a poor source in contrast to the earlier, more direct evidence by any responsible biographer or historian, and rightly shunned.’

7 Cf. Plin. Ep. 9.27, where an unnamed historian allows himself to be persuaded not to expose the misdeeds of certain members of the audience by a public recitatio of the relevant part of his work. This is not regarded as bringing his reliability (fides) into question, since his published text, read in private, continued to provide a permanent record of these crimes. The historian is often identified with Pliny's friend, Tacitus (e.g. Birley, A.R., Onomasticon to the Younger Pliny [Munich, 2000], 53)CrossRefGoogle Scholar. Cf. Ann. 3.65.1 on the duty of the historian to keep alive the memory both of good qualities (uirtutes) and evil words and actions (praua facta et dicta).

8 This passage is listed as Corbulo, fr. 2 in FRH (n. 1), 2.1032 (commentary, 3.612), but there is no reason to consider this as more than a paraphrase of Corbulo's opinion of events. FRH prints Briscoe's suggestion prodidit (turning Corbulo's words into a direct quotation) instead of MS M's prodiderit (probably a potential subjunctive, suggesting a vivid depiction of events that allows for fulfilment i.e. ‘was able to claim/had the opportunity of claiming’—and clearly did claim, perhaps on more than one occasion; cf. Woodcock, E.C., A New Latin Syntax [London, 1959], 91)Google Scholar.

9 For Corbulo's writings, see the account of B. Levick in FRH (n. 1), 1.541–5. Questions of whether Corbulo wrote an autobiography (possibly a Res Gestae) or only commentarii and even the nature of the latter remain insoluble. See also Devillers, O., Tacite et les sources des Annales (Louvain, 2003), 37–9Google Scholar.

10 Vervaet (n. 1 [1999]), 596 n. 72.

11 Famously portrayed by Pliny the Younger in Ep. 7.24.

12 Vervaet (n. 1 [2002]).

13 On the quindecimuiri, see Syme, R., Tacitus (Oxford, 1958), 1.65–6 and Appendix 22Google Scholar.

14 Birley, A.R., ‘The life and death of Cornelius Tacitus’, Historia 49 (2000), 230–47, at 230–4Google Scholar.

15 For Hadrian's withdrawal from Armenia, see SHA Hadr. 23.10; on the Armenian revolts, Cass. Dio 74.9.6. In general on this topic: Chaumont, M.L., ‘Armenia and Iran ii. The pre-Islamic period’, Encyclopaedia Iranica 2 (1986), 418–38Google Scholar. For an overview of Tacitus’ handling of Eastern matters and events in Armenia prior to the reign of Nero, see Malloch, S.J.V., The Annals of Tacitus Book 11 (Cambridge, 2013), 116–26Google Scholar.

16 Birley (n. 14), 241–7.

17 Meulder, M., ‘L. Caesennius Paetus, un avatar du guerrier inpie chez Tacite (Ann. XV, 7–8)’, Latomus 52 (1993), 98–104 Google Scholar.

18 So, for example, Wardle, D., Suetonius Life of Augustus (Oxford, 2014), 9, 174Google Scholar. However, the provisos necessary to detect contemporary references within the Augustus are so substantial that it is difficult to imagine Suetonius’ words conflicting with the aims of either Trajan or Hadrian.

19 In this section, I use a number of the themes that are important to narratologists not to indicate Tacitus’ vivid mimesis (that hardly needs proving) but to show the complex manner in which he constructs a dual narrative (for instance, Paetus’ and Corbulo's activities are synchronic but can only be recounted consecutively). For an introduction to narratology, see De Jong, I.J.F., Narratology and Classics: A Practical Guide (Oxford, 2014)CrossRefGoogle Scholar, in particular ch. 7 (167–96), ‘Narratology and historiography’.

20 Grethlein, J., Experience and Teleology in Ancient Historiography (Cambridge, 2013), 15–16 and 156–9CrossRefGoogle Scholar, on how the possibility of alternative actions increases the vividness of narration.

21 Cf. Shannon, K., ‘Livy's Cossus and Augustus and Tacitus’ Germanicus and Tiberius: a historiographic allusion’, Histos 5 (2011), 266–82Google Scholar, who uses the intertext between Inguiomerus in Ann. 2.21.1 and Tolumnius in Livy 4.19.2 to create a parallel between Germanicus and Cossus and to raise the question of Augustus and Crassus’ eligibility for the spolia opima. This is additionally seen as suggestive of Tiberius’ attitude to Germanicus’ desire for triumphal glory.

22 Livy then uses the Nicias–Alcibiades contrast to frame his account of the debate between Fabius and Scipio over the invasion of Africa (28.40–5), a depiction that has been analysed in detail by Rodgers, B.S., ‘Great expectations: Livy on Thucydides’, TAPhA 116 (1986), 335–52Google Scholar. It is worth noting that Livy's use of the theme in this case turns it on its head, since the ‘rash’ commander is later notably successful, despite the objections of the cautious Fabius.

23 See Gilmartin, K., ‘Corbulo's campaigns in the East: an analysis of Tacitus’ account’, Historia 22 (1973), 583–626, at 608–19Google Scholar, concentrating on the psychological depiction of the two generals; and Ash (n. 1), 371–3, who also argues that Corbulo is regularly contrasted with Lucullus to the former's disadvantage.

24 One might easily find parallels for the actions of the rash commander throughout Livy, as, for instance, in Hannibal's deliberate choice to lure Flaminius who was spoiling for a battle into an ambush at Trasimene (22.4.1). In contrast with Paetus, Flaminius, however, shows courage and makes the most of his military training when surrounded by the enemy (22.5.1).

25 The supplement is Madvig's on the basis of Livy 9.5.3: non … foedere pax Caudina sed per sponsionem facta sit. <pacis; neque> will have been omitted in Tacitus by a haplography of neque after numantin(a)eque.

26 Oakley, S.P., Commentary on Livy, Books VI–X (Oxford, 1997–2005), 3.27–31Google Scholar and Appendix 2.

27 Meulder (n. 17), 100 n. 22.

28 Comparison with Sil. Pun. 5.59–76 shows the poet placing the omens directly before the battle, not at the inauguration of the consulship. Before Trasimene the sacred chickens refuse to eat (not paralleled in Tacitus but perhaps similar to omens reported in Livy 22.1.8–20), the sacrificial bull escapes, the standards cannot be lifted out of the ground and fiery lightning appears over the lake.

29 Minucius’ dispatches: in tam prope clade famam <uanam> egregiae uictoriae cum uanioribus litteris magistri equitum Romam perlatam, Livy 22.24.14; Paetus’ report: litteras quasi confecto bello, uerbis magnificis, rerum uacuas, Ann. 15.8.2.

30 qui solutis ordinibus uagi dissipati errant, Livy 22.29.5; cf. the stragglers who meet Corbulo in Ann. 15.12.2 and are sent back to the front.

31 As E. Koestermann observes (Cornelius Tacitus Annalen [Heidelberg, 1963–1968], ad loc.), the phrase is a direct quotation from Livy's famous obituary of Fabius Maximus (30.26.9). The description is almost a foreshadowing of Vologaeses’ way of dealing with Paetus, avoiding any real negotiation on the pretext that he was waiting for his brothers to arrive, thus dragging out the siege while indicating that the fate of the Roman legions was in the hands of the Parthian royalty (Ann. 15.14.1). For a generally different, less complimentary reading of the Livian allusion (but one which acknowledges that Vologaeses at times might be more Roman than the Romans), see Clark, A.J., ‘Vologaeses as mirror’, Histos 5 (2011), 208–31Google Scholar.

32 Cf. Fabius’ reminder to the Senate of clades per temeritatem atque inscientiam ducum acceptas, 22.25.12.

33 Oakley (n. 26), 1.580–2. For another version of the Minucius story, see Polyb. 3.103–5. Polybius’ deduction in 3.105.9 that the episode made clear to the Romans that there is a difference between a (real) general's planning, logical resolution and good sense and the indiscipline and foolishness of a (mere) soldier is typically practical, in contrast to Livy's emphasis on moral factors ( Walbank, F.W., Commentary on Polybius [Oxford, 1957–1979Google Scholar], ad loc.). App. Hann. 12–13, Val. Max. 5.2.4, Cass. Dio fr. 57.19 and Plut. Fab. 8–13 may preserve traces of an earlier tradition. Silius Italicus’ Punica had already recounted the same episode in 8.377–408, 8.494–750, raising Fabius to the semi-divine status of a Hercules (8.746–50; cf. Verg. Aen. 8.273–5; see Littlewood, R.J., A Commentary on Silius Italicus’ Punica 7 [Oxford, 2011]Google Scholar, on 7.732, 7.735–6, 7.746–7, 7.748, 7.750). Since Tacitus’ friend, Pliny, was well acquainted with Silius and his literary interests (Ep. 3.7.4–5), it is likely that the historian was fully aware of this most recent version of the story also.

34 Livian influences in Tacitus were detected by Syme (n. 13), 200–1, who noted the parallels between the Roman surrender in the Batavian revolt in Hist. 4.62, 4.72 and Livy's account of the Caudine Forks, and also traced stylistic parallels in Appendix 54 ([n. 13], 733–4, drawing partly on the work of G. Andresen and G.B.A. Fletcher). Kraus, C.S., ‘“No second Troy”: topoi and refoundation in Livy, Book V’, TAPhA 124 (1994), 267–89, at 287–8Google Scholar highlighted echoes of Livy, Book 5 in Ann. 15.43.1–5, while articles by R. Ash (‘Waving the white flag: surrender scenes in Livy 9.5–6 and Tacitus, Histories 3.31 and 4.62’, G&R 45 [1998], 27–44 Google Scholar) and Woodman, A.J. (‘Mutiny and madness: Tacitus Annals 1.16–49’, Arethusa 39 [2006], 303–29CrossRefGoogle Scholar) have demonstrated further substantial allusions to Livy in the later historian. Ash's edition of Tacitus Histories Book II (Cambridge, 2007)Google Scholar is particularly good in listing Livian parallels. Further Livian parallels are discussed at length by Shannon (n. 21) and Clark (n. 31).

35 Shannon (n. 21), 277: ‘reference to a work written about, or under the Republic, has definite political implications when it appears in a work written about, and under, the Principate’. More circumspectly, Clark (n. 31), 212: ‘the reader is reminded of how times and circumstances have changed’.

36 Most obviously in Cassius Dio: Corbulo, although widely admired and with a large army under his command, never considered rebellion against the emperor, 62.23.5; Tiridates found one fault alone in Corbulo, namely that he put up with a master such as Nero, 63.6.4; Corbulo realized, when ordered to commit suicide, his mistake in sparing ‘the lyre-player’, 63.17.4.

37 Vervaet (n. 1 [2002]), 187–93; Ash (n. 1), especially 363–4, 373–4. For Tacitean awareness that all was not necessarily fine in ‘the good old days’, see Maternus’ closing remarks in the Dialogus (e.g. 41.1; Republican speeches by their very existence show the problems of their age) or the portrait of the aristocratic Piso the Elder in Annals 2 and 3.

38 The most obvious examples are the exploits of Caesar's centurions T. Pullo and L. Vorenus in BGall. 5.44 and the heroic exploits and deaths of L. Fabius and M. Petronius in the attack on Gergovia (BGall. 7.47, 7.50). Scaeva, in defending his outpost against all odds at Dyrrachium (BCiv. 3.53, notoriously expanded in Luc. 6.140–262), provides a notable precursor to Tarquitius Crescens in Tacitus.

39 Levene, D.S., Livy on the Hannibalic War (Oxford, 2010)CrossRefGoogle Scholar, 171 n. 18 notes that in Livy there is redemption for Flaminius (by his heroic end), Varro (as not despairing for Rome after Cannae) and Minucius (by his admission of error). In Tacitus, there is no such outcome for Paetus.

40 The nature of intertexuality and allusion in history is highly topical: see Pelling, C., ‘Intertextuality, plausibility, and interpretation’, Histos 7 (2013), 1–20 Google Scholar and Elliott, J., ‘The epic vantage-point: Roman historical allusion’, Histos 9 (2015), 277–311 Google Scholar for bibliography. The debate is centred on the level of recognition, not on whether intertexts and allusions exist.

41 A thousand altars raised and libations to Fabius before feasting, Sil. Pun. 7.746–50; cf. the glorification of Scipio at 17.645–54, esp. salue, inuicte parens, non concessure Quirino | laudibus ac meritis, non concessure Camillo, 17.651–2. If Fabius becomes the surrogate father for Minucius and his troops, Scipio becomes pater patriae.

42 See Marks, R., From Republic to Empire (Frankfurt am Main, 2005), 81, 219, 224–5Google Scholar.

43 By contrast, Silius appears to have been the first writer to combine elements of Livy's Caudine Forks narrative with the rescue of Minucius, portraying the unexpected salvation in terms of a mini-katabasis in Pun. 7.723–9 (Stygiae … fugere tenebrae … nox atra recessit, | coniuent solemque pauent agnoscere uisu). Cf. Livy's Roman soldiers leaving the defile after passing under the yoke who blink at the light: uelut ab inferis extractum tum primum lucem aspicere uisi sunt, 9.6.5. In Livy 22.29.3, the simile is not of Minucius’ troops rising from the dead but of Fabius’ forces descending as if from heaven (Fabiana se acies repente uelut caelo demissa ad auxilium ostendit).

44 See most recently Grethlein (n. 20), 178–9. Numerous studies of enargeia throughout Tacitus’ works illustrate this point (e.g. Morgan, M. Gwyn, ‘The smell of victory’, CPh 87 [1992], 14–29 Google Scholar, on Vitellius’ visit to the battlefield at Bedriacum, or Levene, D.S., ‘Pity, fear, and the historical audience’, in Braund, S.M., Gill, C. [edd.], The Passions in Roman Thought and Literature [Cambridge, 1997], 128–49CrossRefGoogle Scholar, on his capture and death in the Histories; O'Gorman, E., Irony and Misreading in the Annals of Tacitus [Cambridge, 2006], 49–53 Google Scholar, on Germanicus’ inspection of the remains of Varus’ army, or Quinn, K., Latin Explorations [London, 1963], 110–29Google Scholar, on the death of Agrippina).

45 Cf. Martin (n. 5), 181, who opines that Tacitus’ narrative includes criticisms of Corbulo, ‘without effectively trying to assess where the balance of truth lay’.

46 Sailor, D., Writing and Empire in Tacitus (Cambridge, 2008), 178–82CrossRefGoogle Scholar.

47 Keitel, E., ‘The role of Parthia and Armenia in Tacitus’ Annals 11 and 12’, AJPh 99 (1978), 462–73, at 472–3Google Scholar argues that only in these books is there a close analogy between events on the periphery of the empire and its centre. Contra, Ash (n. 1) and Clark (n. 31).

48 See n. 7 above.