The Augustan Empire 43 BC

Vitellius not only failed to reconcile the troops who had opposed him, but also failed to win popularity in other quarters. Coins advertising the imperial corn supply show that he was aware of the need for support from the plebs; and on the day after his arrival at Rome, he accepted the title 'Augustus' in response to popular demand (early coins describe him as 'Germanicus', with the praenomen 'imperator'). He accepted the dtle 'perpetual consul' which had been rejected by Nero. But he failed, or refused, to recognize the importance of being head of the domus Caesaris, and did not call himself 'Caesar' until the very last days of his reign. Instead, he emphasized the security which his new dynasty provided by the existence of a son and a daughter. His children appear on gold coins from Rome; the daughter was betrothed to Decimus Valerius Asiaticus, who as legate of Gallia Belgica had been an early supporter of Vitellius. Although only of praetorian rank, he was the son of the Asiaticus who had been mentioned as an imperial candidate in a.d. 41, and forced to commit suicide by Messallina in 47. The alliance with Asiaticus was perhaps both an attempt to reconcile those communities in Gaul that had backed Vindex, such as Asiaticus' origo Vienne, and to win the support of those families that had been associated with Vitellius' father during the early years of Claudius. Other aurei struck by Vitellius at Rome represent the censor. Another of his father's associates, Vespasian's elder brother Flavius Sabinus (cos. suff. c. 45), was confirmed in the position of praefectus urbi restored to him by Otho. His son was assigned a consulship. But neither Asiaticus nor Sabinus were able to give Vitellius much support when the army commanders appointed during Nero's reign came up with an alternative. Even Vitellius' own mother was sceptical of Vitellius' chances of establishing a new dynasty: she is reported to have said that the son she gave birth to was called Aulus, not Germanicus.⁸

» Vitellius' coin issues: IMP. GERMANICVS, Sutherland 1987 (в 3 5 8) chs. 47-9; CONSENSVS EXERCITVVM S.C.; FIDES EXERCITWM/ PRAETORIANORVM; ANNONA AVG. S.C., MW 36-9; LIBERI IMP. GERM. AVG., MW 80; L. VITELLIVS COS III CENSOR, MW 82 (we may note that Joscphus pretends to know nothing of these children: BJ iv.io.3(596)). Consul perpetuus-, ILS 242 = MW 81.

Foremost amongst the provincial commanders was C. Licinius Mucianus, legate of Syria. Mucianus preferred literature to soldiering, and did not propose to put forward his own candidature. Although we do not have enough information about his family to know how 'aristocratic' he was, his own career — governor of Lycia and Pamphylia in 5 7; consul towards the end of Nero's reign — gave him the authority to recommend a name to the Roman establishment. And those officers whose careers had been advanced by Corbulo before his execution in a.d. 66 now looked to him to protect their interests against their fellow- officers in the Rhine legions, who were being given swift promotion by Vitellius. Vespasian, like Mucianus, had loyally served Nero in his last years; he had had much military experience, and distinction; and he also had two adult sons, ensuring that there would be someone to succeed him. Between them the two legates commanded six legions, enough at least to challenge the rebellious Rhine armies. Vespasian was prepared to take the initiative. Despite initial disagreements of the kind only to be expected when Syria and its army had been divided between the two of them by Nero in the spring of 67, Mucianus was prepared to back him.

Vespasian's son Titus was instrumental in arranging Mucianus' support for his father. At the end of 68, he had left Palestine for Rome to submit himself as a candidate for the quaestorship; Galba had been his father's superior at Strasbourg in a.d. 41-3, when Titus had been a child, and Titus was certain that he would favour him. But at Corinth he heard both of Galba's assassination, and of Vitellius' proclamation, and decided to return to Palestine. Oracles and omens along the way confirmed him in the view that Vitellius should be resisted. When news of Otho's defeat reached them, Mucianus and Vespasian were in no doubt about their responsibility to restore legitimate government. They informed governors, imperial procurators and legionary legates throughout the empire of their intentions, and won the support of the network of client kings in the eastern Mediterranean. Minor military operations against the Jewish rebels in Palestine in June had left Vespasian in control of most of the province except Jerusalem and three other strongholds; most of the Judaean army was free for operations elsewhere.

By the time Vespasian was publicly proclaimed emperor, the Danube armies were already throwing off their allegiance to Vitellius. The process by which they were persuaded to support Vespasian rather than a more legitimate Galban successor is unclear. Personal animosities between officers played their part; in Moesia, discipline collapsed when the governor, Marcus Aponius Saturninus, tried to kill the legate of the Seventh Claudia, alleging treason. Another Moesian legion, the Third Gallica, had recently been transferred from Syria. As soon as it heard that the other Syrian legions were supporting Vespasian, it too expressed its support for his candidature. Its legate, Titus Aurelius Fulvus from Nimes, rose high in Vespasian's service. He was to be honoured with a first consulship c. 70; he was to be consul again as the emperor Domitian's colleague in 85, and his grandson Antoninus Pius was himself to hold the imperial office. By contrast, no rewards accrued to Antonius Primus, sdll in command of Galba's Seventh Legion, and loyal to his and Otho's memory. His legion, and the other Pannonian legion, the Thirteenth, that had been forced to assist Caecina and Valens' victory games in Cremona and Bologna, declared for the Flavians, but only because that allowed them to re-open hostilides against Vitellius. They accused the governor of Pannonia, M. Tampius Flavianus, of loyalty to Vitellius — he was a distant relative, but as we have seen had been honoured by Otho. Flavianus initially abandoned his office, but returned at the request of the procurator Cornelius Fuscus. Fuscus had been appointed to the post by Galba; he was to be another important supporter of the new dynasty.[428]

The first official formally to proclaim Vespasian was in fact the prefect of Egypt, Tiberius Iulius Alexander. Son of Alexander 'the Alabarch', a procurator of the younger Antonia, he was the nephew of the Jewish philosopher Philo, and his deceased brother had been the son-in-law of King Iulius (commonly but incorrectly 'Herod') Agrippa I. Not surpris­ingly, he had done well during the early years of Claudius; from a.d. 46 to 48 he was procuratorial governor of Judaea. After some quiet years probably to be explained by the primacy of Agrippina (when Vespasian, too, had been in disgrace) his experience and connexions throughout the eastern Mediterranean made him a suitable choice as an officer on Corbulo's staff in 63 (probably praefectus castrorum, in charge of the commissariat). In 66 he was appointed prefect of Egypt. The prefect of Egypt was the only Roman provincial governor who was not shadowed by an imperial procurator; this made it easier for him to declare his support for a new emperor. The acclamation of Vespasian at Alexandria on i July (two days before Vespasian's own army followed suit at Caesarea) was enthusiastically received; Alexandria had only restricted corporate rights as a city, and will have hoped that support for a successful pretender would be rewarded by the privileges appropriate for the second greatest city in the Mediterranean world. Vespasian was to disappoint any such expectations; he had no wish to represent himself as beholden to the Greek East.¹⁰

In mid-July, Vespasian and Mucianus held a conference at Berytus (Beirut) to plan their campaign. Agrippa, and representatives of other eastern clients of the Julio-Claudians such as Antiochus of Commagene, also attended. The Judaean countryside had been pacified; the glory of conquering the centre of the rebellion, Jerusalem, could safely be left to Titus, supported by Alexander as hispraefectus castrorum. Titus needed an experienced counsellor, and it was perhaps politic to prevent Alexander from becoming too popular in his home town. Mucianus would proceed to the Balkans with an army consisting of one full legion and 13,000 men in vexillationes from seven others. Ships were organized for the crossing of the Bosphorus, and later events suggest that the Flavians approached the prefects of both Italian fleets with a view to winning their support for ferrying Mucianus' army from Dyrrhachium to Brundisium. Whether the intention was to invade Italy from the south or the north (or both), Vespasian cannot have expected any major military operations in Italy before the spring of the next year. In the mean time, it would be made clear to Vitellius and his supporters that no one could be emperor with the support of the Rhine legions alone. The supplies of corn from Egypt, upon which Rome depended, were cut off. Vespasian himself was to await the outcome of events at Alexandria.

What the plan decided upon at Berytus actually was cannot be known because by the time Mucianus reached the Balkans two months later, he found that the Danubian legions had already begun their own war against Vitellius under the command of Antonius Primus. To do that they had left the Danube frontier almost unprotected against the continuing Sarmatian pressure, and Mucianus was forced to turn north to repulse a serious invasion. Leaving his army behind, he hurried on after the Danube legions, to reach Italy in December. His haste suggests that Primus' advance into Italy at the beginning of September was by no means in accordance with the plans drawn up at Berytus. Tacitus says that Primus ignored Vespasian's written instructions to hold back at Aquileia. Vespasian and Mucianus were not pleased at Primus' victories, and he had to spend the rest of his days in peaceful retirement in his home town of Tolosa. On reaching northern Italy, Primus had shown where his loyalty lay: at Padua, he called for the busts of Galba to be restored. It is also remarkable that no coins bearing Vespasianic legends can be assigned to his army; serious objections have been raised to the view that a series of coins bearing Galba's portrait and referring to him as P[a/er] P[atriae] were issued posthumously, perhaps at Lyons, but if such coins were indeed struck after Galba's death, it would be tempting to associate them with the pay given to Primus' soldiers.¹¹

From Vitellius' point of view, the immediate effect of Primus' action

¹¹ On the so-called 'posthumous' coins of Galba, see Kraay 1956 (в 532).

was to cut Italy off from any further reinforcements from the British and Rhine armies. It also made it clear to some of Vitellius' own supporters that there was now no chance that he would be accepted as a legitimate emperor. In September, Caecina and Valens entered upon the suffect consulship which was their reward for their victory. In response to Primus' invasion, Caecina took the entire army north (the praetorians, including the best of the soldiers who had come from the Rhine, were left behind; so was Valens, who was ill). He left it at Bologna, and then went to Ravenna to discuss with the prefect of the Adriatic fleet, Sextus Caecilius Bassus, the best way of solving the crisis without bloodshed. (Bassus was disappointed with Vitellius because he had hoped for promotion to praetorian prefect.) Bassus had already been in touch with an imperial freedman, Hormus, acting for Vespasian.

While military and civilian officers were trying to avoid bloodshed, that was exactly what the armies were looking for. Primus' legions, stationed at Verona, rioted against the two provincial governors, and both Tampius Flavianus and Aponius Saturninus were expelled from the camp. Meanwhile at Bologna, Caecina tried to remove Vitellius' por­traits, but could make no headway in persuading his legions to accept Vespasian as emperor. He fled to Bassus, who had brought his sailors over to the Flavians without difficulty. Two legionary tribunes, Fabius Fabullus and Cassius Longus, took over command of the Vitellian army until Valens was to arrive.

Primus saw that he should force a battle now, before Valens could restore the Vitellians' morale. He advanced on the strongly pro-Vitellian city of Cremona, forcing the Vitellians to try to reach it first. Battle was joined to the east of the city in the late afternoon of 24 October, and lasted through the night; ancient sources give the expected vivid accounts of the horrors of this 'Second Battle of Cremona'. There were said to be 50,000 dead, and worse than the actual battle was the destruction of Cremona by the victorious Flavians which followed; the fire, started by the Thirteenth Legion in revenge for the insulting way it had been forced to help build an amphitheatre for Valens' games after the first battle, was said to have lasted for four days.

Caecina's defection showed that Vitellius could no longer trust some of his own officers. The praetorian prefect, Publius Sabinus, had to be replaced before the major part of what remained of the Vitellian army, fourteen praetorian cohorts, moved north along the Via Flaminia in support of Valens, who no longer had an army. Meanwhile Cornelius Fuscus had occupied Rimini on behalf of the Flavians, and the Vitellian army fell back, ultimately taking up a position at Narnia, about 100 km north of Rome. Valens himself travelled through northern Italy to Gallia Narbonensis, hoping to raise another army on the Rhine; but Valerius

Paulinus, the procurator - who, like many of the imperial procurators of whom we know, gave immediate support to Vespasian's bid for the empire - arrested him and sent him to Primus, who had him executed.

Late in November, a centurion of the fleet at Misenum was instrumen­tal in effecting the fleet's transfer of loyalty to Vespasian. But here Vitellius had some success: he sent his brother Lucius to Campania with a few praetorian cohorts, and on 18 December - during the Saturnalia - Lucius' troops managed to recapture Terracina from the marines. But Lucius' actions came too late to save his brother: the cohorts at Narnia had surrendered one or two days before.

Vitellius' failure to bring about a swift end to the civil war after the second battle of Cremona has been unfavourably compared to Otho's suicide after the first. His indecision may be explained as due to uncertainty as to the extent to which Primus and his army were actually acting in support of Vespasian. Throughout the autumn, Vespasian's brother, the urban prefect Flavius Sabinus, had been available as a mediator; Vitellius seems to have been guaranteed his life, and the opportunity to retire to Campania. But Sabinus, too, was unclear about whether Primus would accept his authority (Vespasian's twenty-year- old younger son Domitian was not prepared to leave Rome in the company of Primus' messengers). Only when Mucianus himself had reached northern Italy could Sabinus and Vitellius act publicly. But the soldiers from the Rhine legions whom Vitellius had promoted to his praetorian guard had too much to lose to accept his abdication. When a formal contio was held in the Forum on 18 December to announce the surrender of Vitellius' imperium, those present shouted their opposition. An attempt to hand his dagger over to the consul, Caecilius Simplex, as a sign that he was resigning the imperial office, was rejected. Vitellius had to return to the palace while Sabinus (who, as praefectus urbi, was commander of the urban cohorts) and Domitian retired to the Capitol, assuming that they would be safe there until Vitellius regained control of his supporters.

While these negodations had been in progress, Primus was in no hurry to rescue Vespasian's relatives in Rome. Independent action was taken by a cavalry unit commanded by Petillius Cerialis, described as a close relative of Vespasian; he was almost certainly the husband of Vespasian's daughter Domitilla (now deceased). Cerialis' attempt to break through the Vitellian defences on the northern outskirts of Rome was repulsed, and in revenge Vitellius' soldiers turned against Sabinus. It appears that some of the Flavian supporters set fire to buildings on the slope of the Capitol in order to protect themselves: the fire spread and engulfed the principal temple of Rome. Several Flavian supporters were killed. Sabinus himself was captured and brought before Vitellius, whose attempts to save his life failed. Domiuan escaped in the garb of a priest of Isis, together with his cousin, Sabinus' son T. Flavius Clemens (consul in a.d. 95 as Domitian's colleague).[429]

When he heard of Cerialis' failure to enter Rome and of the destruction of the Capitol, Antonius Primus could no longer hold back his army. He may also have calculated that the death of Sabinus would enable him to present a candidate of his own choice to the Senate. Tacitus suggests that in the first days or weeks after the occupation of Rome, he tried and failed to persuade Licinius Crassus Scribonianus to become his own puppet emperor. As brother of the Piso adopted by Galba, and the Cn. Pompeius Magnus married to Claudius' daughter Antonia, Crassus had a stronger connexion with the household of Caesar than Vespasian. Primus entered the city on 20 December (possibly 21), and encountered considerable resistance. Vitellius attempted to flee the city - he may have heard of his brother's successes in Campania — but his praetorians would not let him go. He was discovered hiding in the deserted palace, dragged through the Forum by a mob of soldiers and civilians, and put to death.

Mucianus succeeded in reaching Rome within a few days of Primus, and acted swiftly to isolate him. Even before his arrival, Mucianus had sent written instructions to the Senate to ensure that it was Vespasian who was duly recognized as Caesar and Augustus, and that the people passed a law voting him all the legal powers that earlier emperors had had (one of the two bronze tablets bearing the text of this lex de Imperio Vespasiani still exists, and grammatical peculiarities suggest - the haste with which Mucianus drafted it).[430] It is not surprising that individual senators started to ask questions about just who it was who represented the new emperor. In January 70, with Domitian's consent, the Senate passed a decree honouring Galba and Piso; only later was it realized that this did not accord at all with the wishes of Mucianus and Vespasian. Domitian's re-appearance as Caesar provided a point of reference. He was duly elected as urban praetor, but with the unprecedented imperium of a consul. Mucianus arranged for rewards for those who had been a party to Vespasian's own plans. He himself was nominated to a second suffect consulship in a.d. 70, together with Petillius Cerialis; the client- kings who had supported Vespasian were honoured; and the freedman Hormus, instrumental in negotiations regarding the fleet, was granted equestrian status. But Antonius Primus, the man who had actually defeated Vitellius, was eased out of power and never again played a political role. For Vespasian's security, Mucianus arranged for the execution of those who might attract the support of Galban partisans as candidates for the imperial office: C. Piso Galerianus (son of the conspirator of a.d. 65), and his father-in-law Lucius Piso (cos. 57), current proconsul of Africa. Lucius was a brother-in-law of Galba's heir and of Licinius Crassus Scribonianus; both were therefore related to Augustus' first wife Scribonia. Piso was killed by Valerius Festus, legate of the Third Legion and a relative of Vitellius who needed to prove his loyalty to the new regime.

Mucianus' next problem was to defuse the rivalry between Vespa­sian's two sons. Titus had won military glory as his father's legate in Judaea (he was to remain there for the prestige of destroying Jerusalem that summer). Domitian suspected that he would have little chance of surviving for long if his brother ever came to the throne. One option open to him was to win military glory himself by leading the Flavian legions north to deal with the remaining Vitellian units in Gaul, Britain and the Rhineland. Mucianus had already sent Petillius Cerialis to the Rhineland, and allowed Domitian to follow (thus removing him from Rome); later tradition had it that Domitian personally received the surrender of the Lingones, but he seems to have been prevented from seeing any fighting. Instead of seeking to rival the military glories of his brother Titus and brother-in-law Cerialis, Domitian dedicated himself to writing poetry, including epics recording the fighting on the Capitol and his achievements in Gaul.

The failure of the Rhine legions to accept Vespasian after Vitellius' death proved a major embarrassment to the Flavians, and to pro-Flavian historians. The events of a.d. 69/70 in the Rhineland had to be re-written in such a way as to avoid giving the impression that Vespasian had been supported by Batavians and (some) Gauls, while the citizen legions and (other) Gauls continued to constitute a 'Vitellian' force. In consequence, Tacitus' Histories describe the rebellion against Vitellius led by the Batavian leader Civilis as though it was an uprising by provincials against Roman rule. But Tacitus also has to admit that when the rebellion began, it was welcome to the Flavians: he says that at first it was only in secret that the rebel leaders expressed anti-Roman views. If Civilis was a traitor, he was a traitor to Vitellius. In the autumn of 69, at the behest of Antonius Primus, he took the oath to Vespasian and besieged a Vitellian legion at Vetera (Xanten). Tacitus misleadingly suggests that by the beginning of 70, the legions too had taken the oath of loyalty to Vespasian. In fact, the legate of Upper Germany, Hordeo- nius Flaccus (who had supported Civilis' action) was killed by his troops when he tried to administer the oath, and Vitellius' portrait restored. A pro-Vitellian legate, Dillius Vocula, came to the help of the soldiers at Vetera; when the legionaries tried to evacuate the camp there and march south in March 70, they were massacred by Civilis' Batavians (Tacitus emphasizes the presence of Germanic warriors from across the Rhine among Civilis' soldiers).

The leaders of a number of Gallic tribes also remained loyal to the Vitellian cause. With the Flavians recognized at Rome and the arrival of Cerialis and Domitian in Gaul in the spring of 70, their resistance could be re-interpreted as a tribal uprising. But these men were as little Gallic nationalists as Vindex had been. Iulius Classicus had led the Vitellian advance as far as the Maritime Alps in early 69; the other leaders, Iulius Tutor and Iulius Sabinus, were 'Romans' to such an extent that Dillius Vocula's legions accepted their command after the disastrous retreat from Vetera. In the absence of any senator who might be put up as the Vitellians' candidate for the imperial office, Iulius Sabinus made a bid by claiming that his grandfather had been an illegitimate son of none other than Iulius Caesar himself. The 'Gallic Empire' (Imperium Galliarum) which they called for was not an empire controlled by the Gauls, but a Roman empire in Gaul, a compromise which could be supported both by legionaries who wished to remain loyal to Vitellius and by Civilis' Batavians and other Gallic tribes who had fought them.

It was the absence of a plausible leader that gave the legionaries no alternative but to accept Vespasian. Their last hope was to persuade Cerialis himself to take up their cause; he passed their offer to make him emperor back to Domitian. The Flavians took what measures they could to win the loyalty of these supporters of Vitellius. Four of the Rhine legions had to be disbanded (I, IV Macedonica, XV Primigenia, XVI), and replaced by new ones, whose titles proclaimed their association with the new dynasty (IV and XVI Flavia). The loyalties of the British legions during this period are even more difficult to reconstruct (see ch. 13?). Cerialis took over command of the British army, perhaps to balance his brother-in-law Titus' command in the East. The military activities of the next three years, involving the subjugation of Brigantia and the founding of a new legionary base at York, gave the legions stationed in Britain an opportunity to prove their loyalty to the Flavians. In Britain, as in the Rhineland, the legionaries' conditions were improved by the construction of more permanent, stone camps, such as the one at Caerleon. The story of their war against Civilis was re-written to make it seem that they had always been loyal to Rome, fighting German barbarians and Celtic and Batavian traitors. Unlike Galba, Otho or Vitellius, the Flavians managed to win the support even of those who had fought against them. The coinage broadcast not just military victory over Judaea and the security represented by the new emperor's two sons, but the 'Revival of Rome', Peace, Liberty, and concord between emperor and Senate. As censors (a.d. 72-4), Vespasian and Titus freed the Roman people from the moral stain, and from some of the memories, of civil strife. The account of Vespasian's reign as the recognized successor of the Julio-Claudians is to be found in another volume. Mucianus enjoyed a third consulship in a.d. 72, and then spent his retirement writing books.¹⁴

¹⁴ Civilis: Urban 1985 (c 406). Classicus' coins include the legends ADSERTOR LIBERTAT1S, LEGION XV PRIM and CONCORDIA: FIDES may be an appeal for continued loyalty to the Vitellian cause. Cf. Zehnacker 1987 (в 564). Tacitus' admission that the rebels were only 'separatists' in secret: Hist. 111.14. Brigantia: Birley 197; (e 529), Hanson and Campbell 1986 (e 544). Vespasian's coin issues: ROMA RESVRGENS, PAX P. ROMANI, L1BERTAS RESTITVTA, AETERNI- TAS P.R., CONCORDIA SENATV1 - MW 42-6; 90; 254.

CHAPTER 7 THE IMPERIAL COURT

ANDREW WALLACE-HADRILL

I. INTRODUCTION

If the powers of Augustus and his successors were monarchical, the most important arena where those powers were exercised was the court. Both as an institution and as a word, the court was alien to the Republic. Aula, a direct derivative of the Greek aule, the standard term in the hellenistic world for the courts of oriental and Greek kings, is almost unknown to republican literature (including Livy); but rapidly establishes itself under the early Empire (notably in the writings of Seneca under Nero) to refer both to the physical location of imperial power and to the type of power, the personnel, and the perilous way of life that were associated with it.¹ New though the phenomenon was to the Romans, they were well aware that what they now experienced was an old feature of monarchical societies. 'Reflect,' observed the emperor Marcus Aurelius in his Meditations, 'how all the life today is a repetition of the past... the whole court circle of Hadrian for example, or the court of Antoninus, or the courts of Philip, Alexander and Croesus. The performance is always the same; it is only the actors who change.'²

The historical and biographical sources recognize the role of the Julio-Claudian court. Stories told about Vespasian's early career encap­sulate assumptions about how court life worked. His success under Claudius was ascribed to the influence of the freedman Narcissus; he also had a mistress, Caenis, among the imperial freedwomen. His son, Titus, was brought up at court (in aula) with Britannicus. The fall of Narcissus and the rise of Agrippina meant his political eclipse. Nevertheless, he remained in the court circle, and was taken by Nero to Greece among the comites. But his unconcealed lack of enthusiasm for singing brought him into bad odour, and he was banned not only from the inner circle (contubernium) but even from the general audience {publico salutatio). He

See TLL 1.1457-8, s.v. aula II.j.c. Cic. Fam. xv.4.6 (of the court of Ariobarzanes) is apparently the only republican occurrence. Similarly used of foreign courts by Augustan and later writers, e.g. Virg. 11.504, Val. Max. vn.1.2; of court life in contexts applicable to Rome ĥrst in Seneca Ira 11. j}.», Tranq. vi.2; of Nero's court, [Seneca] Octavia 285 etc.; then regularly of the imperial court in Martial, Statius, Tacitus, Suetonius and later.

Med. 10.27. On the views of Marcus, Brunt 1974 (в 19).

283

learnt of his disgrace from one of the freedmen who controlled admissions (ex officio admissionis), whose treatment of him was so acrimonious that he was scarcely rescued by the intervention of other courtiers.[431] We meet here a string of assumptions that run through the historical accounts of the Julio-Claudian and later periods: the fragility of political success and its dependence on imperial favour; the role of freedmen and members of the imperial family as mediators of favour; the emergence of subordinate personnel who help to define access to and exclusion from the court; and the intertwining of political and social life at court, and the consequent importance of imperial tastes.

The work of the last generation of historians has represented a large step towards a better understanding of the early imperial court. Several major studies have extended our detailed knowledge of the freedmen personnel,[432] the equestrian amici principis,[433] and of links among the senatorial elite.[434] Above all, study of contacts between emperors and their subjects, the decision-making process and the distribution of resources and patronage, show us the network of imperial personnel in operation and reveal something of the structures within which they operate.⁷

But in spite of these advances, the court remains partly veiled from our sight. Historiographically it leads a sort of twilight existence. This is true both of the ancient sources and modern scholarship. The difficulties that obstruct the historian were articulated by Cassius Dio: monarchical rule involved a retreat of political life and the decision-making process from open places (the Senate and Forum) into privacy. Dubious official announcements and hearsay represent the only access to what was going on.[435] Tacitus reacts to this problem by the tactic of irony.⁹ Rather than focus on the court on the basis of suspect information, he directs his attention to public places in the style of his republican predecessors: he thereby underlines not merely the political impotence of the Senate, but the impotence of the historian, who can only approach the true locus of power indirectly. The majority of our direct information about the workings of the Julio-Claudian court is anecdotal: this is true not only of the biographies of Suetonius, but of the numerous reminiscences of contemporaries, Seneca in his philosophical dialogues, the elder Pliny in his Natural Histories, or the Discourses of the ex-slave philosopher Epictetus preserved by Arrian. The tendency to anecdote is not a personal weakness of our sources, but a structural consequence of the retreat of politics behind closed doors.[436]

Modern historians have reacted to the problem differently. Suspicious of anecdote, and disinclined to see history as made by feminine schemes and palace plots,[437] we have moved away from study of the Principate as a political system to study of administrative systems and hierarchies. The temptation has not always been resisted to substitute modern bureaucra­tic structures for the unfamiliar structures of a court society.¹² The world of kings and courts is one of which the present age has lost sight, and it requires an effort of historical imagination to take its structures ser­iously.[438] In consequence, this chapter represents a sketch not only of what we have learnt, but of what we stand in danger of forgetting.[439] In discussing the nascent court of the Julio-Claudian period, it will be necessary to generalize more broadly about the function of the court in the structure of imperial power.

ii. access and ritual: court society

The court and its membership had no 'official' definition, for this was a social not a legal institution, private in its composition though public in its importance. The contrast with the Senate is significant: membership of that body was a legal status, only open to certain social categories, age groups, and one sex, and Augustus at an early stage took measures further to define eligibility and to formalize procedures and conduct of business.¹⁵ The court remained in its nature undefined: membership was constituted by proximity to the emperor, and only social ritual could distinguish degrees of proximity. At the negative extreme, the renounce­ment of amicitia was a formal token of imperial displeasure and expulsion from court; but the amicitia enjoyed by those who had not fallen from grace was fluid and imprecise (a point obscured by attempts to catalogue the amici principis, as if they were officials with a rank).¹⁶ Many had access to the aula; far fewer were admitted to the private chamber, the cubiculum principis,¹⁷ Nor did the court have any official or public function. Events of public importance took place on the Palatine from Augustus on, such as the reception of embassies, councils of state and trials, but they did so not as 'court' events, but in virtue of the personal obligations of the emperor. By tradition, any public figure at Rome was liable to use his house for occasions of a quasi-public nature.¹⁸ This lack of definition only added to the power of the court: one of the secrets of power, the arcana imperii, was to be untramelled by rules.

Nor was its location fixed: aula represents an abstraction, not a description of a particular place. Under the late Empire the court was to be peripatetic, like the courts of many medieval monarchs; at all periods the court (but not necessarily all courtiers) moved with the emperor.¹⁹ This does not mean that the imperial presence transformed all contexts into the court, as when the emperor attended the Senate or the games: these were public venues, in contrast to the private and domestic venues of the court, even the praetorian tent on campaign.²⁰ But despite the string of properties across Italy already developed by Augustus, and the fondness of the Julio-Claudians for the Bay of Naples, and specifically Capri, it is notable that in practice the court was from the start firmly centred on the city of Rome, and particularly the Palatine Hill.²¹ This too has its echo in language. Palatium acquires the sense of'palace' by the end of the first century a.d. (the metaphorical usage goes back to Ovid), and as Cassius Dio later pointed out, it was the facts of life rather than any decree that turned palatium into the name for any imperial residence, no matter where its location.²² The rapid absorption of the show houses of the republican nobility on the Palatine, already far advanced by the end of Augustus' reign, neatly symbolizes the absorption of their social power.²³ Augustus and his successors manipulated this symbolism with care: the rich ritual and 'historical' associations of the hill of Romulus were exploited, and the potential of the site to overlook and dominate the public activity of the Forum and the mass meetings of the Circus Maximus was underlined by the choice of where to build.

Suetonius' emphasis on the modesty of Augustus' residence may create a false impression, engendered by the desire of a later age to idealize the simplicity of the past.²⁴ Contemporary reactions in the poets, explicit in Propertius and Ovid, veiled in Virgil, register the overwhelm­ing impression made by the novel complex of private house and public temple (Actian Apollo), portico (adorned with Danaids) and libraries.²⁵ The tantalizing fragments that have emerged from recent archaeological

'8 Vitr. De Arcb. vi.j.z; cf. Millar 1977 (a 59) i8ff. ¹⁹ Millar 1977 (a 59) 28-57.

Veyne 1976 (f 71) 682-5 perversely identifies the whole city of Rome as court.

Millar 1977 (a 59) 15-28.

Ov. Met. 175, Dio lin.16.4-6; cf. RE xvin 3 (1949) 10-15 ^{s v}- Palatium.

Wiseman 1987 (f 81).

Suet. Aug. 72. Sources are collected in Lugli 1962 (e 82) 154-61.

Esp. Prop. 11.31; Ov. Fast. ^.951-4; Tr. 111.1.31-48; Pont. 11.8.17; Virg. Aen. vii.^ofT, cf. Wiseman 1987 (f 81).

exploration give concrete documentation of the interweaving of public and private in the area of the temple of Apollo, approached from within Augustus' house by a series of ramps, which is more reminiscent of a hellenistic royal palace than a traditional Roman house.[440] This feature, dating back to 28 B.C., was extended in the course of the reign: in 12 в.с. the public cult of Vesta, symbolic hearthplace of the city, was incorpor­ated within the private house of Augustus as pontifex maximus, and in a.d. 3 after a major fire and rebuilding of the palace on public subscripdon, the whole residence was declared public property. Thus the architectural ambivalence of public and private embodies from the first the essendal ambiguity of the court as an institution, a private household with a central role in public life, the domus of a citizen and simultaneously the praetorium, the headquarters of a commander pro­tected by the praetorian guard.[441]

The Augustan development lacked unity; it was rather a string of separate households absorbed piecemeal, and this was still true of the palace as Josephus describes it at the time of Gaius' murder.[442] Nero's vast building activities, both before and after the great fire, imposed coherence for the first time, and eliminated the final traces of indepen­dent houses of the aristocracy on the Palatine, such as the house of the orator Crassus with its famous lotus trees, finally owned by Claudius' courtier Caecina Largus.[443] Even without taking into account Nero's extension of his Golden House onto the Esquiline, we may be struck, as were contemporaries, by the staggering extent of the palatial complex.[444]Covering some 10 hectares, it exceeded the palace of Attalus at Pergamum by a factor of 30, though indeed if the palaces of Alexandria or Antioch were preserved, they might have approached somewhat closer to the Roman scale. This vast development implies human activity on a corresponding scale. The so-called Aula Regia of Domitian's palace was preceded by an earlier and not much less impressive auditorium. A small indicator is provided by the lavatories which constitute one of the few fragments of Nero's rebuilding on the Palatine: with a capacity of over forty, they exceed the public lavatories attached to the fora of towns like Ostia or Corinth, and approach the level of a major modern railway station. The palace should be seen as a major concourse of human activity.³¹

Rome was where the early emperors held court for serious business: Italian villas and the Bay of Naples, even in the case of Tiberius' last years, represented an escape from the pressure of people into relative otium,[445] The choice of location had implications for the development of Rome as an imperial city and as the monumental showpiece of the empire. Many factors, not least tradition, may have dictated this choice; but one factor of paramount importance was the question of accessibi­lity. The emperor needed to be readily accessible to a very considerable number of individuals. The prime function of a court is to provide and control physical access to the ruler; the courtiers are those who simultaneously have achieved some degree of regular access for them­selves and are capable of mediating it to others. It is therefore the structures and rituals through which access to the ruler is mediated which give a court its distinctive character. Who could get at the emperor, and on what conditions?

The composition and rituals of the imperial court were evolved from patterns current among the Roman upper classes at large.[446] Three groups can be broadly distinguished: family, servile household, and friends. The first two represent the 'insiders', the domus or Jamilia Caesaris. Wives and children play a central role in court life. Other relatives were more loosely attached: Roman social custom did not favour the extended family, and many members of the imperial family kept separate households. The exceptionally diffuse family network built up by Augustus explains the physical structure of the palace in his day as a nexus of partially separate houses: even Tiberius in the last decade of his adoptive father's reign kept separate household in the Domus Tiberiana, while Gaius' father Germanicus had his own house in the reign of Tiberius.

Freedmen too, following Roman social custom, might be more or less loosely attached to their imperial patron's house: they might reside within the palace to perform daily services, but they might keep separate households of their own. Augustus used the houses of freedmen on the Palatine or elsewhere to escape from visitors or to watch the games, while the independent houses of Claudius' great freedmen like Posides and Callistus were among the wonders of the city.[447] What distinguishes both family and freedmen as 'insiders' is their relationship to the emperor, not their residential location. Fortune, whether through birth, marriage or the slave market, had placed them in a permanent proximity to the ruler to which no outsider had access. The imperial household, unlike that of the medieval or early modern king, opened no avenues to the talent and ambition of the subject: the element of sheer chance behind the making of a potent freedman was epitomized by Epictetus in the figure of Felicio, the cobbler slave who by an exchange of hands emerged as an imperial functionary, to the confusion of his old master.[448] To start with, the domus Caesaris was many households as well as many houses: different members of the imperial family kept their own establishments, and Antonius Pallas, the most famous of Claudius' freedmen, began his career as a slave in the confidential service of Claudius' mother Antonia.[449]

The court is not simply the ruler's household, but the household operating as an interface with the society over which he rules. The distribution of power in monarchical society is likely to correspond to the distribution of access to the ruler. In the hellenistic kingdoms there was marked conflict between the status systems of the court and of the cities. The royal philoi drew their status from proximity to the king; and the grades of court hierarchy depended not on functional differentiation but on closeness to the royal person — so in the Ptolemaic court the descent is from relatives (syngeneis), to those honoured as if relatives, to the bodyguard (in the sense of royal pages), to first friends, to friends. The kings paid no attention to the ascriptive status systems of the cities; consequently out of the court circle the royal friends were derided as unworthy climbers, 'flatterers' or 'parasites'.[450] Correspondingly the hellenistic courts developed rituals and ceremonials which opened a sharp gulf between the king and the norms of Greek or Macedonian society: pomposity of dress and setting (elaborately canopied thrones); rituals like proskynesis which, whatever its significance and appropriate­ness in Persian society, had in the context of Greco-Macedonian society a profoundly distancing effect; and ceremonial language drawing on that of cult.

The similarity has often been remarked between these hellenistic philoi and the amici Caesaris, particularly in view of the apparent (but ill- attested) distinctions introduced of a cohors primae admissionis (group of the first admission), secundae admissionis and so on.[451] Doubtless there was hellenistic influence on Roman social ritual, of which the Romans themselves were aware, just as the differentiation of the freedman secretariat is probably developed on a hellenistic model. But this obscures the fundamental gulf between the imperial court and any hellenistic analogue. For by and large the early Caesars paid elaborate attention to the status hierarchy of Roman society, dovetailed the privileges of their amicitia with the demands of ascriptive status, avoided rituals that set them apart from the aristocracy, and controlled the tendency of the court to generate a gulf between itself and society.[452]

The social rituals which channelled access, notably the morning salutatio and the afternoon cena, were those normal among the nobility of the late Republic and early Empire. Repeated descriptions of the bustle of the early morning salutatio at the great houses of Rome by Seneca and the satirists only underline its similarity to the imperial routine: the emperor was distinguishable in the scale but not the style of his admissions.[453] If he graded his friends into admissiones, so too did others; Seneca, our only informant on this, attributes the introduction of the custom to Gaius Gracchus and Livius Drusus.[454] Assuming that Vespa­sian followed the pattern of his predecessors, secretaries and officials were interviewed and their breviaria read before the admission of friends to the bedroom, followed by a general salutation. Vespasian may have started earlier in the day than some, but the daybooks of officials in Egypt show similar patterns of business.[455] Nor is there much trace at this stage of the evolution of distinctive imperial dress or pomp. The emperor wore the toga at his levee; if Caligula wore floral tunics, it was regarded as an aberration, and failed to establish a new ceremonial.[456]

Other institutions taken directly from the republican nobility include the appointment of comites (companions), duly rewarded with a salarium, to form a cohors amicorum, and to join the contubernium (mess) of the emperor on tour or campaign, and the summoning of amici to form a consilium to advise on specific issues.[457] Naturally, the 'friends' and 'advisers' of the emperor played a role in public affairs and wielded an influence which far outran any republican precedent, and the amici principis were busy men, and regarded by others with awe and even fear.[458] But it is an error to represent the imperial consilium as an established organ of government with a defined membership. Its informality was essential.[459] In building on republican precedent in all these varieties of amicitia, the Caesars not only established themselves as respectors of the mores maiorum, but integrated the behaviour of the court into the patterns of behaviour current in the aristocradc society around them.

Perhaps the most striking feature of the anecdotal descriptions of imperial admissions and receptions is the predominance of senators and members of the upper stratum of the equestrian order. There was evidently widespread attendance at salutations by members of the senatorial order (including their wives and children); not until a.d. i 2 in the infirmity of old age did Augustus ask the Senate to be excused his normal practice of greedng them all at his home.⁴⁷ As a rule they enjoyed precedence. Senators were greeted with a kiss - a hellenistic custom indeed, but one already current among the elite in Cicero's day.⁴⁸ Nero is said to have denied the kiss to all senators on his return from Greece: this was a powerful mark of imperial displeasure, not an attempt to reverse the assumption that senators were entitled to this mark of intimacy.⁴⁹ A vivid reflection of the social ties which interconnected the upper orders and linked them to the emperor is the elder Pliny's report of the outbreak of a facial disease in Tiberius' reign.⁵⁰ Pliny remarks on the way this epidemic was restricted in its incidence both geographically to Rome and socially to the upper orders {proceres): the disease was spread by kissing, and its extent and restriction reflected the exchange of kisses at the salutation. Tiberius, who appears to have been affected himself, put a temporary ban on the custom. The kiss was not reduced to a symbol of obeisance. Seneca vigorously protests at Gaius' gesture in proffering his foot to a consular to kiss: with its overtones of oriental court ritual, this was precisely the kind of gesture that did not establish itself as the Roman norm.⁵¹

Accounts of imperial dinners repeatedly feature senators and equites.⁵² Even if Gaius was tickled by the macabre thought that he could execute both consuls at will, they were reclining next to him in the positions of honour when the thought arose.⁵³ Conversely there is a dearth of anecdotes illustrating the entertainment of the socially humble, or complaining of their access to the imperial table. Augustus is said only once to have admitted a freedman (not his own) to his table.⁵⁴ His successors were not necessarily so strict; but there is no sign of imperial freedmen jostling for places with the proceres. The prime access of freedmen to the emperors was not on formal occasions, but informal and backstairs. Helico owed his influence with Gaius to his access to him at

⁴⁷ Dio lvi.26.2—j* ⁴⁸ Cic. Alt. xvi.5.2; Kroll 1933 (a 54) li-5gfF.

⁴⁹ Suet. Ner. 37. » Pliny, HN xxvi.3; cf. Val. Max. xi.6.17; Suet. Tib. 34.4; 68.2.

⁵¹ Alfoldi 1934 (d i), 4off; Sen. Вея. н.12.1; cf. Epictetus, Diss, iv.1.17.

^и Friedlander 1922 (a 30) 1. 98-103: Turcan 1987 (d 20) 2}7ff; cf. D'Arms 1984 (f 23).

⁵¹ Suet. Calig. 32.

⁵⁴ Suet. Aug. 74; but cf. Macrob. Sat. 11.4.28 for the entertainment of a slave dealer.

intimate moments, 'when he was playing ball, taking exercise, at his bath and at his breakfast, and retiring at night'.⁵⁵ But as far as social life was concerned, the early emperors behaved as members of their own social class, greeting, entertaining, and on occasion reciprocating offices by accepting hospitality and attending functions.[460]

Senators and equites were by no means the sole members of the court circle. One notable group which regularly met in the court of Augustus and his successors was that of Greek intellectuals and men of learning - the philosopher Areius at Augustus' court, the grammarian Seleucus or the astrologer Thrasyllus at Tiberius', the doctor Xenophon at Clau­dius', the musician Terpnus at Nero's. The majority of these are attested as living at court, sharing the contubernium principis.⁵¹ Here again, emperors were not setting themselves apart from, but assimiladng themselves to, the habits of the republican and early imperial nobility. When the historian Timagenes forfeited the amicitia of Augustus, he went to live with Asinius Pollio.⁵⁸ In supporting such intellectuals, emperors were not promoting a group otherwise neglected by society, but providing themselves and their friends with cultural stimulus of the type the Roman upper class had come to expect. On the other hand, because the resources and importance of the imperial house so far outran those of any aristocratic house, the effect was to introduce a new pattern of effectively 'public' patronage of the arts in place of the strictly private patronage of the Republic.⁵⁹

Because integrated into the social and cultural life of the Roman upper class, the court not only served to reflect existing norms but dictated the tone of society.[461] The emperor was seen as a model eagerly imitated by others. The hothouse atmosphere of the court helped to disseminate tastes and fashions as well as facial disorders. Fashions in hairstyles or the decoration of houses throughout the empire closely and rapidly respond to models set by the court in Rome, and art history points to the deep penetration of the lives of Romans by the stylistic and moral values of the imperial circle.[462]

The role of the court in shaping fashion was aided by its use as a place for the upbringing of the children of favoured courtiers (as well as the children of foreign and barbarian kings). In hellenistic courts, the pages or basilikoipaides were a formal institution, enjoying especial prestige, and kings took into their innermost circle the syntrophoi with whom they themselves had been brought up. At Rome there is no trace of royal pages as a formal rank, but the children of the distinguished certainly frequented the court, received schooling there (under Augustus at the hands of the grammarian Verrius Flaccus), attended dinners (explicitly attested under Claudius), and enjoyed the attentions of emperors and their wives.[463]

Looking back from the complacent respectability of the Flavian and Antonine eras, our historical sources regard the mores of the Julio- Claudian court with a mixture of shock and astonishment. Profligacy of sexual morals, grossness and wanton pursuit of the exotic in eating, above all lavish waste in the construction and decoration of houses combine with sophistication of taste in literature and an (unRoman) delight in music. In all this, the imperial court continues in a direct line the 'hellenizing' tendencies of the aristocratic houses of the late Republic. Such social and cultural trends could not be manipulated by the emperors at will: the attempts of Augustus and even Tiberius to impose restraint, whether by legislation or by example, proved futile. In fact they (probably unwittingly) promoted the trends they professed to oppose. For by suppressing the traditional channels by which prestige was generated and made visible under the Republic, through glory in war and demonstrations of popular favour,⁶³ they redirected the compe­titive energies of the elite into the social displays upon which success in a court society depended.

This display contained the seeds of its own destruction. Their very magnificence, as Tacitus observes, was the ruin of the great houses, and Nero, who outstripped all competition with the sumptuousness of his Golden House and the wasteful dinners when guests were drenched in perfume from the ceiling, was surely aware of the political advantages of ruining his rivals financially with the aid of his unique access to the wealth of empire.⁶⁴ But Nero in turn was ruined by employment of this technique, both financially and, more damagingly, morally. The accele­ration in extravagance of his reign produced a revulsion of taste within the court circle itself, among men from municipal and provincial backgrounds who perceived the implications of the way of life into which they found themselves sucked.[464] The tone of the Flavian court, for which the elder Pliny acts as spokesman, was palpably different.

Just as the court had a decisive impact on the culture and morality of Roman society at large, it is likely to have played a central role in the formation of opinion. It is frequently stated that the outlook of our sources is 'senatorial'. In some ways this is undeniable. Republican historiography had been dominated by senators, and imperial historians were conscious inheritors of the republican tradition. Respect for the upper classes in general and for the Senate in particular is one of the criteria on which emperors are most consistently praised or condemned. Social contacts within the relatively small group of senators could have been close, and doubtless many of them saw eye to eye on many issues. But what cannot be demonstrated is that such a 'senatorial' viewpoint is at variance with an alternative viewpoint, and that things looked rather differently from the perspective of the Palace.

It is notable that two of our major sources for the Julio-Claudian period, the elder Pliny and Suetonius, were men of equestrian rank who held posts in the service of the emperor. Their judgments of individual emperors and their underlying ideals do not appear to differ significantly from those of the senatorial Tacitus; on the other hand, both can be taken to reflect the views of the courts at which they served, Pliny in his loyalty to the Flavians and their puritanical morality, Suetonius in his implicit acceptance of the ideals of the 'golden age' of Trajan and Hadrian.⁶⁶ Other non-senatorial sources follow the same pattern. Josephus' black­ening of Gaius, though in line with senatorial opinion, was determined by his own Jewish sensibilities, and was evidently quite acceptable to his Flavian patrons. Epictetus' reminiscences of court life are based on his experience as slave of Epaphroditus; though his master was close to Nero, he fully shares the 'senatorial' view of Nero as a tyrant.⁶⁷

Without suggesting that the court always had a homogeneous point of view (there could be deep internal conflicts, as under Nero), it is not hard to imagine that it may have acted as a focus for discussion, gossip, and eventual opinion formation. Gossip it generated in abundance, and courtiers at all levels might be the source of anecdotes, from Augustus' attendant Julius Marathus who could describe his physique, and the interiores aulici who had theories about Gaius' Baiae bridge, to reminis­cences by consulars about what had been said at the imperial table.⁶⁸ Imperial freedmen were a source of valuable information to contempor­aries: leaking of inside information, or to use their own expression, the 'sale of smoke', became a familiar abuse in the Antonine court, but already we are told that Augustus broke the legs of a secretary for selling the contents of a letter.⁶⁹

Behind trivial gossip lies concealed the serious purpose of the

⁶⁴ Wallace-Hadrill 1983 (в 190) 998; Gascou 1984 (в 59) 71 iff; Lambrecht 1984 (в 103).

Rajak 1983 (в 147) i8jf on Josephus; Millar 1965 (d 14) on Epictetus.

Suet. Aug. 79 and 94.3; Calig. 19.3; Tib. 61.6.

Suet. Aug. 67; Friedlander 1922 (a 30) 1. 47 on the sale of smoke; cf. Mart, iv.5.7.

exchange of observations and impressions by those in the imperial entourage. Court life, as Saint Simon appreciated, is a watching game. It could be vital to second-guess the imperial mind, to see who was rising in favour and who falling, and what changes were in the wind, for on such observations, as Sejanus' facdon discovered to their cost, fortune and even life depended. Tacitus' descripdon of the dinner at which Britanni­cus was poisoned suggests something of the sense of urgency of the game, and of the simultaneous need to see into the minds of others while concealing one's own: 'those sitting nearby were thrown into confusion; the imprudent fled, but those with deeper understanding remained rooted to the spot and watched Nero'.⁷⁰

Assessments of individual emperors and their characters are surpris­ingly constant in the different sources, and it was once the fashion of source-criticism to posit a single source from whose initial assessment of an emperor all successive accounts derived. This perhaps underestimates the potential of the social circles around the court, the convivia et circuit of whose part in shaping public opinion Tiberius was aware,[465] to evolve a stereotype of the character of the ruler. In his lifetime assessments will have been fluid; but after his death, the court of the succeeding ruler could impose a definitive stamp. The image of Claudius as a fool was one Nero deliberately encouraged, both by his own chance remarks, and by the publication of the Apocolocyntosis by his closest adviser; Nero was surely drawing on and encouraging court gossip here, and there is no need to lay the blame for the image of Claudius solely on the malice of senators outraged by the power of the secretariat.[466]

In social terms, then, the Julio-Claudian emperors, whatever the political strains they may have experienced with the Senate, and however much power they may have allowed to their freedmen, drew their friends and companions from the upper class, afforded them easy access, failed to elaborate rituals that set themselves apart, and were bonded to them by the integrating force of common culture. Rather than regarding the court as an institution apart, we might think of it as the centre of a sort of solar system. Numerous houses of the rich and powerful in the city of Rome acted as lesser courts, centres of influence round which social activity clustered, to which visitors and clients thronged in the morning, and where sophisticated entertainment was provided later in the day. The palace was both similar to them and yet outshone them, the centre round which they themselves revolved, and from which ultimately they derived their own radiance.

iii patronage, power and government

The social rituals of a court may act as a fafade to screen the realities of power. The endlessly elaborate etiquette and ceremonial of the French court of the seventeenth and eighteenth centuries partly served to mask the diversion of power from the old nobility by substituting the fafade of social precedence for the realities of control.[467] The 'civility' for which 'good' emperors are praised by the sources has also been seen as a charade designed to screen the unpalatable truth of imperial power. The disjunction between appearance and reality has been greatly exagger­ated. For while emperors undoubtedly used the court to control and limit the power of the upper classes, they also used it to strengthen their own power by embedding it within the existing social structure. The relationship of emperor and upper classes is thus complex and ambivalent.[468]

What drew men to court was more than social life. The court was the font of power and favour - and so the scene of anxieties and humilia­tions. Men love or hate Caesar, according to Epictetus, only because of his power to confer and take away advantages, wealth, military rank, praetorships or consulships.⁷⁵ The court inspires fear, not just of bodyguards and chamberlains and the like, but because of anxiety to secure the benefits Caesar distributes, governorships, procuratorships, praetorships, consulships, money; the courtiers behave like children fighting in their scramble to gather the scattered figs and nuts.⁷⁶ The lure of court is irresistible: the returning exile who swore to live in peace could not resist the invitation to court, and found himself praetorian prefect.⁷⁷ Yet was success worth the humiliations involved? The rising early, the running around, the kissing hands, rotting at others' doors, speaking and acting like a slave, sending gifts?⁷⁸

From the first, emperors derived power from their ability to distribute resources. Claudius had shown, according to Seneca, how much more effectively imperial power was secured by favours (beneficia) than by arms.⁷⁹ The range of beneficia was enormous: status and legal privileges (citizenship, equestrian and senatorial rank, privileges like the ius trium liberorum etc.), magistracies, posts in the army and administration, financial benefits (fiscal concessions and immunities, subventions after disaster, grants to enhance status, and numberless liberalities to favour­ites and courtiers) and judgment (from resolution of disputes to cases of life and death). Documents and anecdotes evoke a vivid picture of the pressure of petitions and requests from individuals and communities across the empire on the person of the emperor, and the personal nature of his involvement.⁸⁰ Yet though he and not any subordinate bureau­cracy was the source of the benefits, inevitably the requests were mediated through others. Hence the patronage of the emperor is the centre of a complex web, in which the courtiers act as brokers as well as beneficiaries.⁸¹

The network emerged rapidly. One aspect is the swift evolution of a ramifying secretariat of slaves and freedmen. Over 4,000 inscriptions, mostly sepulchral, attest the sheer scale of the imperial secretariat over the course of the Empire.⁸² The shape of imperial business dictated the division and organization of labour, and it is significant that the lines along which it divided were not areas of government but the channels of communication between subject and ruler. The letters, petitions, embas­sies and legal hearings which brought contact with the emperor generated the Palatine 'offices' of ab epistulis, a libellis, a legationibus and a cognitionibus, and alongside these record-keeping (a memoria) and above all supervision of the vast imperial wealth, ambivalent in its status between the public and the private (a rationibus), account for the main activities of the secretariat.⁸³ Such divisions may go back at an informal level to Augustus,⁸⁴ but it is notoriously under Claudius that the formal titulature that became standard is first seen in the literary sources in the naming of Polybius, Narcissus and Pallas as a studiis, ab epistulis and a rationibus respectively, and on the testimony of one who himself held two of these posts.⁸⁵ At once, such titles acquired an imperial ring: the charge against the two Torquati Silani under Nero of nursing imperial ambi­tions in calling their secretaries ab epistulis, a libellis and a rationibus shows how for all its origins in the bloated servile households of the aristocracy, the imperial household had grown into something of quite another order.⁸⁶

In some respects, the familia Caesaris betrays characteristic features of bureaucratic government. We can detect the emergence of bureaux with their own hierarchy of subordinate posts, from slave tabellarii, through junior freedmen adiutores, tabularii and a commentariis, to the senior grade of proximus immediately below the head, himself known simply by the name of his officium (e.g. ab epistulis). The grades seem clearly distinguish­able in terms of age-range (senior officials were normally old men), even if a set salary structure must be regarded as hypothetical.⁸⁷ The personnel could be regarded as 'officials' embarked on a quasi-public career

⁸⁰ Millar 1977 (a 59)passim and 1967 (d i 5). ⁸¹ Sailer 1981 (f 59) 4iff.

⁸² Weaver 1972 (d 22) 8. 83 Millar 1977 (a 59) 20з(Г. м Boulvert 1970 (d 6) 5jff.

»5 Suet. Claud. 28; cf. Wallace-Hadrill 198} (в 190) ₇₃ff.

Tac. Ann. xv.jj and xvi.8.

Weaver 1972 (d 22) 227ff; Boulvert 1974(07) 127ff on grades is too schematic, cf. Burton 1977 (D 8).

partially analogous to the cursus bonorum: this much is implicit in Statius' panegyrical account of the career of the father of Claudius Etruscus, whose promotions through a series of posts brought him progressive honour,⁸⁸ and also in those epitaphs which imitate senators and eques­trians in listing posts in ascending or descending sequence.

But in analysing the functions and powers of the familia Caesaris, it is misleading to assimilate it to a modern bureaucracy. Much more fruitful analogies lie in the royal households of medieval and Renaissance Europe. One essential feature of the household is that it serves the person of the ruler in all his activities, private or public, small or large. Private functions of the ruler (the bedchamber, the table, the stables etc.) are hard to separate from the public and administrative. Just as the medieval English court generated numerous - and to us faintly ludicrous — subdivisions in the private sphere, of spicery, napery, ewery, and apothecary, of gargons of the sumpterhorse or valets of the garbage,[469] or as the court of Francis I of France gloried in its sixty categories of household officials, down to furriers, spit-turners, tapestry-makers and laundresses,⁹⁰ so the imperial court displays a dizzy proliferation of minutely defined functions, such as the many divisions of the wardrobe (a veste privata,forensi, castrensi, munda, alba triumphali, matutina venatoria, regia et Graecula etc.) or of the buttery (a crystallinis, a cyatho, a lagona, a potione etc.).⁹¹ The fact that a freedman might advance like Ti. Claudius Aug. lib. Bucolas from taster (praegustator) and butler (tricliniarchus) to procurator aquarum, with care for the aqueducts of Rome, and procurator castrensis, steward of the Palace,⁹² certainly affected contemporary per­ceptions of imperial freedmen, and should at least make us pause before categorizing them as 'civil servants'. Separation of domus and respublica was an empty promise.[470]

The range of posts within and without the Palace reflected the diversity of its activities, from distribution of resources and judgment to feasting and entertainment. Certainly the appointment of equestrians to the major secretarial posts which Vitellius initiated shows their develop­ment under the Julio-Claudians to a conspicuous role in public life; yet equestrians had been employed before this in the imperial household in less 'political' functions, like Pompeius Macer as a bibliothecis under Augustus, let alone Tiberius' shocking appointment of an equestrian to charge of his 'pleasures' (a voluptatibus), a post regarded by a later freedman as a splendid promotion.[471] It does not help to draw a hard and fast line between private and public funcdons. The role of the important post a studiis is notably obscure, but may have ranged from advice on imperial speeches to grammatical commentary on private reading of literature.⁹⁵

In trying to understand the power of the imperial freedman, then, it is not enough to say that the early emperors turned their household into a new arm of government (though this is clearly the case). The power of the freedman derived from his proximity to the emperor and his consequent ability to influence specific aspects of resource-distribution. The word even of a court-jester might cost a man his life.[472] Claudius' prepotent freedmen, who included Posides the eunuch and Harpocras as well as the 'heads of bureaux', owed their power to their master's combination of an insatiable appetite to bestow favours and judgment with an inability to control the detail of so many transactions. The mistresses of emperors, as of many later kings, were in an ideal position to extract favours, as Vespasian's Caenis, with her long experience of the court, well understood.[473] That the elite resented the wealth and influence which flowed from such brokerage is not surprising, not because the use of political position to amass gratia was new to Roman society, but precisely because the exercise of patronage was how the elite tradition­ally defined its own standing. Imperial freedmen established no mono­poly in this respect, and the fact that the court became the focus of elite patronage too underlay the tension.

The reign of Augustus was one of transition from the pluralist patronage system of the Republic, whereby the nobility competed with each other to maximize their following and thus their influence with the populus Komanus, to the imperial pattern under which the emperor monopolized the support of the populus, and the elite looked to him for favours, which they in their turn distributed to others.⁹⁸ The number of benefits within the imperial gift multiplied throughout the Julio- Claudian period: the number of posts in the imperial service rose, and rights and privileges like the ius trium liberorum or even leave of absence from the Senate were quietly absorbed by successive emperors by steps we can mostly no longer trace. But the core of imperial patronage, round which all else accrued, was there from the start: the wealth that flowed from victory in civil war, and the control over appointments in the army and 'imperial' provinces.

From the first, then, the elite looked to the emperor for favours, and their attendance at court was motivated by pursuit of favours. The court thus played a vital role in consolidating imperial power within the context of imperial society." First, it enabled the ruler to control the elite. In order to pursue power it was necessary to come to Rome and enter the intrigue of the court. That firmly established Rome as the arena of political conflict and discouraged the emergence of alternative regional power bases. The 'big men' of the empire were under the immediate eye of the emperor. He could manipulate their ambition by playing them off against each other, using his control of the distribution of resources to keep them on tenterhooks, withholding favours and elevating new favourites if the influence of old favourites threatened to become entrenched. Secondly, he could through the elite exercise a progressively wider control throughout the empire. The elite, senatorial and equestrian, was drawn from the municipalities of Italy and, in this period, increasingly the western provinces. Those at court acted as brokers for their contacts at home, securing benefits for them and drawing further compatriots into the circle of power at Rome — a marked example of this process is the rise of Spaniards in various posts in the administration during the Corduban Seneca's period of influence with Nero.¹⁰⁰

Within the broad circle of the hopeful and ambitious who attended the court, there was an inner circle of amici upon whom emperors called for advice in a variety of circumstances: to assist in giving judgment, whether in public imperial cognitiones, or in the more sinister trials intra cubiculum, and to handle a whole range of questions from the trivial and routine to matters of high state. Perhaps there were times when not even the amici could predict the gravity of the questions to be considered: Juvenal's picture of an imperial council debating the preparation of a fish may be satire, but Nero is said to have called theprimores to his house in the Vindex crisis only to spend the day, after brief political consultation, discussing types of musical organ.¹⁰¹ Augustus' innovation of a standing committee of senators with regular meetings and a defined and rotating membership which prepared business for the Senate was not continued by his successors; thereafter such business was dealt with on the same informal and ad hoc basis as other matters. There was no such thing, as the classic study of the subject has emphasized, as the consilium principis.¹⁰² Lack of definition, in membership and function, only increased the discretionary powers of the ruler: this too was among the arcana imperii. Even so, some were called for consultation more regularly, and on more

" Cf. Elias, Court Sociity, i46ff. ¹⁰⁰ Griffin 1976 (в 71), 81-96.

Juv. Sat. 4; Suet. Ner. 41.2, better than Dio lxin.26.4.

Crook 1955 (d 10) 8-20; i04ff.

sensitive issues, than others, and these could be seen as the friends of Caesar.

The accessibility of the emperor to the elite thus worked to their mutual advantage. Individual members of the elite had access to power and influence; the emperor was able to reduce the elite to dependence on himself. That does not mean that the court operated smoothly and without tension. On the contrary, it was a battleground — much more so than the Senate, where the only real battles were trials. In the Julio- Claudian period the battle was particularly bloody, for while the system was still emergent, major tensions were unresolved. The sharpness of the conflict is reflected in the bitterness of the accounts given by the sources, for instance the power of the praetorian prefect Sejanus under Tiberius, or that of the freedmen Pallas and Narcissus under Claudius. Two areas of tension are apparent: that within the senatorial-equestrian elite, and that between the elite and members of the inner imperial household, especially the freedmen officials.

Because of the obvious contrast between the monarchical nature of the court and the republican nature of the Senate, it is tempting to envisage a permanent tension between senators as a group and non- senators, whether equites or imperial freedmen, as an opposed group, a temptadon strengthened by the old theory of a legal separadon of powers between emperor and Senate. This is to understate the com­plexity of the conflict.[474] It is true that Augustus' creation of the great equestrian prefectures, and the power attained by the chief freedmen secretaries under Gaius, Claudius and Nero, created a new disjunction between power and status, which resulted in strange inversions of social precedence, as when the equestrian prefects followed the consuls, but preceded the other magistrates, in swearing the oath of allegiance to Tiberius, or Claudius' freedman Polybius walked in public between the two consuls.[475] A divorce between status and power meant that the emperor was less trammelled by social constraints in distribudng power, and could neutralize those by whom he felt threatened by palming them off with marks of high status that carried little power.[476] It is not unlikely that even Augustus saw the advantages of such a strategy and played it deliberately.

But it is wrong to represent the senators as a coherent group, either socially or politically. They were as much creatures of the court as the imperial freedmen. Patronage cut across status barriers: senators enlisted the support of equites and freedmen, but conversely equestrian and freedmen posts might be owed to the brokerage of senators. Alliances like that between Vespasian and Narcissus worked to the advantage of both parties. Within the Senate distinctions may be drawn: not perhaps between men in the 'imperial service' and others, since imperial patronage also affected posts that were not direct imperial appointments, but between a 'grand set' of those swiftly promoted in status who enjoyed little power, and a 'power set' of those who rose more slowly but were entrusted with greater responsibility.¹⁰⁶ But even this distinction may understate the influence wielded at court by members of the grand set, who having risen rapidly thanks to good connexions may well have continued to exercise their connexions to the benefit of others. The lines of division of the elite at court were not between the social ranks of senator, eques and freedman, which were united by multiple ties of family, friendship and interest, but between groups of mixed status: the fissures were vertical not horizontal.

The heyday of the power of freedmen coincides with a period of intrigue and influence among the female members of the imperial household. Wives and freedmen have it in common that they are 'insiders' and therefore stand apart from the 'outsider' elite. In no sense were freedmen in compedtion with members of the elite: they were not eligible for army rank nor senatorial positions (even if they could be awarded military and senatorial decorations); they did not function as amici, and there is no sign that they were invited to attend the consilium -it is with high irony that Tacitus depicts Claudius in consilio when consulting his freedmen.[477] Nor, as we have seen, do they appear to have shared in the social life of the court. Unlike elite brokers of patronage, they were not themselves competitors. Their competition was with each other (Pallas' award of the insignia of the praetorship reflects compe­tition not with senators but with his fellow-freedman Narcissus, pre­viously decorated with the quaestorship); in exactly the same way the imperial women competed for influence with each other, excluded by their sex from the men's world of offices. The influence of freedmen should therefore be seen in the context of the pattern of court intrigue in which the women were simultaneously involved. Their power came from the conflict of competing groups.

The women of the Julio-Claudian household were openly involved in the opieradon of patronage. We hear casually of Livia's role in promodng Galba and the grandfather of the emperor Otho.¹⁰⁸ An inscription shows her openly acknowledged by Augustus for her role in securing privileges for the island of Samos.¹⁰⁹ Networks of friendship extended from the palace among the women of the Roman elite. Seneca (who owed the start of his career to his aunt Helvia, and its furtherance to Agrippina) takes for granted that Marcia, as an indmate of Livia, used her influence to secure a priesthood for her own son.¹¹⁰ Messallina abused her position not by exercising but by selling patronage: together with Claudius' freedmen, she sold the cidzenship so liberally that it was said to be had in exchange for glass beads, and not only the citizenship, nor even military commands and provincial governorships, but everything in general.¹¹¹ Her presence at the trial intra cubiculum of Valerius Asiaticus was something altogether more sinister.¹¹²

Female involvement in patronage was not simply a side product of the system. From Augustus to Nero the imperial court is characterized by sharp intrigue that periodically surfaces in the eruption of major conflicts between competing groups; in almost all these conflicts, the women play a central role. The court of Louis XIV was analysed by pardcipants as split between cabals that clustered round various members of the royal family; any distinctions of political or religious principle that could be detected between the cabals were of secondary significance.¹¹³ A similar analysis seems to .apply to the Julio-Claudian court. The power groupings are heterogeneous in composition: female members of the domus Caesaris and their children, leading freedmen, senators and equites. Lucius Vitellius, that epitome of a courtier, thrice consul and censor, was said to have carried around Messallina's slipper and kissed it from time to time, and to have kept the images of Narcissus and Pallas among his lares.ⁿ⁴

The aim of a cabal is to maximize its own influence in the distribution of resources. Naturally groupings tend to form around potential candidates for the succession: there are already hints of rival groups round Octavia and her son Marcellus on one side, Agrippa, Livia and her sons on the other early in Augustus' reign,¹¹⁵ clear signs of rival groups round Julia, Livia and their respective sons later,¹¹⁶ and under Tiberius explicit feuding between the supporters of Agrippina and those of Sejanus, adulterously linked to Livilla.¹¹⁷ It should not be assumed that such cabals formed with explicit designs on the throne: the mere existence of a potential successor is enough to constitute a catalyst for intrigue, and much of the policy of intermarriage and interadoption,

¹⁰⁹ Reynolds 1982 (в 270) no. 13 line 5, cf. Suet. Aug. 40. ¹¹⁰ Sen. Cons, ad Marciam 24.3.

¹¹¹ Dio lx.17.3-8. ¹¹² Tac. Ann. xi.2.

^1,5 See E. Le Roy Ladurie 'Versailles observed: the court of Louis XIV in 1709' (in The Mind and Method of the Historian. Trans. S. and B. Reynolds. Brighton, 1981) for analysis of cabals.

¹¹⁴ Suet. Vit. 2. ¹¹⁵ Syme 1939 (a 93) 340-2.

Syme 1984 (a 94) hi. 912-36. ¹¹⁷ Levick 1976 (c 366) i48fF.

particularly as practised by Augustus, must have been designed (how­ever ineffectively) to frustrate the formation of rival cabals. The marriage of Tiberius to lulia, for instance, though it did little to clarify the line of succession to power, must have aimed to obviate precisely the sort of tensions and rivalries that erupted with such unfortunate consequences.

A characteristic of conflict between rival groupings is that they come to a head in accusations of adultery - against the two Iulias, Livilla and Sejanus, the sisters of Gaius, Messallina, and Nero's betrothed Octavia. The charge of adultery is often regarded as a sham to disguise political realities; indeed the strings of 'accomplices' of the adultery of the Iulias indicate that no ordinary adultery is involved.¹¹⁸ But we should not underestimate the threat posed to stability within the court by adulterous liaisons (nor overestimate the innocence of the accused). Since marriage was used as an official instrument of dynastic policy, to mark succession and to unify potentially divergent groups, adultery represented the inverse, the dark underside of intrigue and group formation out of the emperor's control. Sejanus' adultery was seen as a vital step in his rise to influence and his establishment of a stranglehold over the network of patronage. Of course, some accusations of adultery were false, and could be cooked up by rival interests to discredit the accused (Livia must be suspect on this count). But, as with accusations of magic, which was the inverse of the divine protection behind imperial power, the charge reflected a threat to imperial power which the participants felt to be real.

Finally, we should not exaggerate the rigidity of such cabals. Their membership was unstable and fluid. Loyalties and friendships could evaporate in a moment (it was the misfortune of Sejanus' supporters that they had no warning of his fall). Courtiers watched carefully to see whose stock was rising with the emperor, whose falling. 'Nothing in human affairs is so unstable and fluid as the reputation of power': Agrippina's crowded threshold was deserted in an instant when the whisper circulated of her son Nero's displeasure.¹¹⁹ Epictetus compares court life to the lot of a traveller who attaches himself to the convoy of a passing official for protection from bandits; the friendship of Caesar is an equally undependable method of progress, hard to pick up, easy to be lost, and limited by the life chances of the Caesar himself.¹²⁰ The point applies similarly to friendship with Caesar's friends. Moreover, the groupings were fissile, potentially divided into further groupings. Messallina was overthrown by a combination of her old supporters, Narcissus and Vitellius; during the crisis, Narcissus did not feel sure even of Vitellius and had him excluded from the imperial Utter.¹²¹ Though supported in the overthrow by Pallas, Narcissus was ruined by the combination of

¹¹⁸ Tac. Ann. ш.24; cf. Syme 1984 (a 94) 111.924^ »' Tac. Ann. xin.19.

¹²⁰ Epictetus, Diss, rv.1.91-8. ¹²¹ Tac. Ann. xi.33.

Pallas and Agrippina, having unwisely shown too much interest in Britannicus. Agrippina was abandoned by her proteges Seneca and Burrus. Such cases serve as warning against any attempt to detect long- term political groupings and alliances.

With new patterns of politics, the court generated new styles of life peculiar to itself. Even survival, let alone success, was fraught with dangers. Seneca reports the reply of the old courtier asked with amazement how he had reached old age at court: 'by accepting insults and expressing gratitude for them'.[478] Flattery and the concealment of true feelings were a structural necessity. Seneca goes on to tell the tale of the distinguished eques Pastor who, on the very day that his son was executed by Gaius, was bidden to make merry at the imperial table. There was a reason for the courtier's bizarre compliance with the invitation — he had a second son. A degree of self-abasement and hypocrisy seemed necessary even under the best-intentioned emperors: Tiberius complained of the servility of his senators, but failed to stop it. In this respect, the Senate acted as an extension of court life; the adulatio of which Tacitus complains, the incessant manufacture of honorific decrees and inflated language, came from men with an eye to promotion or merely survival at court.

Hypocrisy and flattery stood in direct antithesis to the libertas of frank expression and independent opinion on which the republican nobility prided itself.[479] It was not however mere traditionalist sentiment which made men under the Principate hanker for the old libertas. The new court life was highly unstable, and placed gross psychological strains on the courtier, who hardly knew whom to trust and whom to back from one moment to another. The agony felt by the friends of the disgraced Sejanus, eloquently voiced by M. Terentius, struck a chord with every anxious courtier: 'It is not ours to reason whom you choose to elevate above others and on what grounds; the gods have given you the final say; it is left to us to take pride in loyalty.'¹²⁴ But such obsequium was no defence for those who backed a loser.

In this context of instability and psychological strain, philosophy had an important role to play. Stoicism, with its stress on the value of single- minded pursuit of public duty and virtues irrespective of the dangers, offered a vital antidote to the hypocrisy of court life.¹²⁵ It is no coincidence that Stoicism flourished, in martyrs like Thrasea Paetus, when the excesses of Nero's court were at their peak. The philosophy of both Seneca and Epictetus emerges from men with a court background and offers explicit reaction against court morality. In the long run the Stoics carried their point, and the tone did change. Yet a century later, the Stoic emperor Marcus still needs his philosophy as antidote to court life, its vain pomp and superficiality, its transitory quarrels and ambi­tions, and the sheer irritation of working with the pettiness of his courtiers.¹²⁶

IV. CONCLUSION

The court, as social and political insdtution, lies at the heart of the new regime established by Augustus and his heirs. It also encapsulates the paradoxes of that regime, and the way it transformed the structures of the old city-state to create those of the new monarchy. The household of a private citizen, based on the forms and practices of the households of the republican nobility, became the centre of the state; the focus of political activity shifted irrevocably from a plurality of households to a single one, sprawling monstrously over the symbolical heart of Rome. In drawing to itself the threads of patronage, the court brought the transactions of political dealing under imperial surveillance.

The similarities to the royal courts of the East were only too apparent to participants. Court life brought servility in the place of the freedom of a society of citizen equals. The tone of public discourse changed, from bold self-advertisement and uninhibited attack on rivals, to self-conceal­ment and lip-service to the source of power. And yet the transition from city-state to monarchy was a hesitant and gradual one, and the reuse of old forms was essential. The Julio-Claudian court preserved the social hierarchy of the Republic, while yet seeming to undermine it and subject senators to slaves. The early emperors needed to exercise power with, not against, the traditional ruling class. They used republican forms to establish their own dominance while appearing to respect their fellow- citizens. The rituals of court allowed them at one level to use the republican status hierarchy to legitimate their own position, while at another playing off the aristocracy against new men promoted from the provinces and against liberti, ignoble but potent. The accessibility of the emperor to the upper classes and his 'civil' treatment of them as 'equals' was an essendal part of the strategy of power, and it makes the imperial court fundamentally different from the court of any hellenistic ruler.

Between Augustus and Nero the patterns of court life were develop­ing, and still far from fixed. But there is an unmistakable movement towards formalization and institutionalization. The differentiation of the secretariat and the evolution of its internal hierarchy is one tangible

¹²⁴ Cf. Brunt 1974 (в 19).

example of this. It is also right to emphasize the element of continuity.¹²⁷ When we ask what made possible the stability of the government evolved by Augustus, which despite its extraordinary lack of legal definition and its reliance on Augustus' own charismatic personality, nevertheless managed to survive the eccentricities of four members of his own house and a return to civil war, to become the system without which peace was unthinkable, the answer must He partly in the imperial court. Despite notable instances of the fall of political favourites, like Sejanus or Seneca, there was an underlying condnuity of personnel. The Flavians were served by many with long experience of power in the Julio-Claudian court. The anonymous father of Claudius Etruscus, who served as freedman of every Caesar from Tiberius to Domitian to die in his ninetieth year excited Statius' admiration by surviving so many changes of yoke and so many stormy seas.¹²⁸ But though few could rival him in longevity, imperial slaves and freedmen, originally personal to Augustus, came to transfer automatically to the new regime, giving rise to a stability of staff.

The same continuity can be observed at higher social levels. It is striking what long and intimate links each of Nero's successors display with the Julio-Claudian court. Galba started as a favourite of Livia, and served successive emperors, being especially favoured by Claudius who admitted him to his cohors amicorum.¹²⁹ Otho was grandson of another of Livia's proteges and son of one so admired by Claudius as to be honoured with a statue on the Palatine; his own intimacy with Nero was notorious.¹³⁰ Vitellius, grandson of an Augustan procurator, and son of that most adept of Claudian courtiers, also had an uncle whose links with Sejanus cost him his life; while he himself followed the tastes of each Caesar with remarkable pliability, a sexual favourite under Tiberius, a charioteer under Gaius, a dicer under Claudius, a musician under Nero.¹³¹ Vespasian, as we have seen, met both favour and disgrace at court, while his son Titus was intimate enough with Britannicus to have risked sharing his fate. Even in Nerva, at the end of the century, we find a sexagenarian, whose loyalty to Nero had earned him a statue on the Palatine, and a member of a family whose three generations of loyalty to the dynasty stretched back to the treaty of Brundisium in 39 B.C.¹³² If others were as well served biographically as were emperors, such family histories of continuous service would be multiplied.

Good friends, Trajan is supposed to have said, compensated for

Crook 1955 (d 10) 29, 11 j fF etc.

Stat. Silv. in.j.8jf, 'tu totiens mutata ducum iuga rite tulisti|integer, inque omni felix tua cumba profundo'; Weaver 1972 (d 22) 284^

Suet. Galba j and 7. ¹³⁰ Suet. Oth. 1. Suet. Vit. 2-4.

¹³² Crook 19j5 (d 10) i)9f; for the consulate of Nerva's father, AE 1979, 100.

Domitian's bad rule.¹³³ But emperors inevitably took over their pre­decessors' friends and servants, good or bad, since these made them­selves indispensable. Vested interests were at stake. Augustus and his successors needed a court in order to rule; but if imperial rule came under question, the court needed its emperor. Thus, despite its conflicts and distasteful features, the court was a system of power which tended to its own perpetuation.

¹³³ SHA Alex. Sen. 65.5; cf. Tac. Hist, iv.7.3, 'nullum maius boni imperii instrumentum quam bonos amicos esse.'

CHAPTER 8 THE IMPERIAL FINANCES

D. W. RATHBONE

The economic resources at the disposal of the emperors from Augustus to Vitellius and the uses which they made of them are most clearly explained against the background of the state expenditure of the Roman empire.¹

The empire required an army, and under Aifgustus a standing army was developed, of which the size and terms of service of the legionary component remained broadly stable throughout this period, although the nature of the auxiliary component took much longer to crystallize.² Annual pay for a legionary was 900 sesterces, while cavalrymen, higher ranks and the praetorian guard received considerably more. There were stoppages against this pay for replacement equipment and clothing and almost certainly for food. On discharge a surviving legionary in theory received a bounty of 12,000 sesterces - equivalent to over twelve years' basic pay, and so a third of a surviving veteran's total remuneration - but he may often have been given a plot of land in a frontier zone instead or in part payment. The conversion of auxiliary forces, traditionally supplied ad hoc by allied states, into regular units of the Roman army and the standardization of their terms of service and remuneration were slow processes which lasted into the Flavian era. The rate of pay for auxiliary troops remains frustratingly uncertain (footsoldiers may have received a half or five-sixths or some intermediate fraction of the basic legionary rate), as does the date of its standardization (perhaps under Claudius, but perhaps not until the Flavians). There is no evidence that auxiliaries in this period regularly received either cash or land on discharge. Instead, from Claudius on, Roman citizenship was used as a cheap reward, along with the limited tax immunities which were probably granted to all veterans. Pay for all soldiers was sometimes supplemented by bonuses given by emperors on political occasions (booty was another possible extra, though hardly state expenditure). Other military expenditure included materials for defences, camps, all kinds of equipment, transport and riding animals, and supplies. There were also the fleets to maintain.

¹ General treatments: Frank 1940 (d i 28) v. chs. I—II; Neesen 1980 (d 151); Lo Cascio 1986 (d 145); Noe 1987 (d 152). ^г See below, ch. 11.

309

The total annual cost of the imperial armed forces cannot be computed with accuracy because of the mass of variable and unknown factors. Most modern estimates of the average annual wage bill before Domi­tian's pay-rise would put it, if we include discharge bounties, at 400 million sesterces, at least.[480] Even if not fallacious, such estimates are misleading. Because of the system of deductions at source from pay, much of the theoretical wage bill was probably never paid in cash. On the other hand, the total bill will have increased steadily as the number of auxiliary units grew and their remuneration was regularized. Actual cash expenditure also swelled when campaigns were mounted, probably mainly to mobilize extra supplies — the slave dispensator for Nero's Armenian manoeuvres allegedly managed to siphon off 13 million sesterces with which to buy his freedom.[481] In general terms, however, military expenditure was kept artificially low insofar as conscription, rather than the payment of attractive salaries, was used regularly to fill auxiliary units and sometimes to fill legions.

The empire required administration, mostly in the spheres of finance and law and order. Salaried officials were few - the senatorial governors and legates and the slowly growing number of equestrian procurators - but their salaries were substantial, perhaps totalling over 50 million sesterces per annum, and presumably were paid in cash; revenues were also skimmed off by the increasingly numerous and permanent clerical staff in their offices.[482] However, many of the costs of administration were hidden. The emperor, senators and town councillors throughout the empire were meant to perform public functions at their own private expense, an obligation which helped to justify and to reinforce their economic dominance.[483] As subordinates they would also use their own dependants - which was initially the position of the familia Caesaris, the imperial slaves and freedmen, although it came to live at least partly off state revenues. The central government and its representatives also employed seconded soldiers in civil police and administradve roles. When transport, labour or supplies of any kind were required in the public interest both central and local governments and their individual representatives could commandeer virtually at will from the subject population. The prime examples of this are the cursus publicus and the uniquely well-documented local corvee obligations in Egypt.⁷

The empire had no economic or social programmes, but it still incurred massive expenditure on public buildings and roads, on the rituals of civic life such as sacrifices, games and banquets, on rewards to artists, athletes and educators, on minting coinage, and on ensuring a reasonably regular supply of staple foodstuffs to its urban populations, in short on producing and maintaining what we recognize as Roman civilization. In the provinces and Italy this expenditure normally fell on the local aristocracy, who were mostly, in this period, not unwilling to bear it in return for the prestige and power which it conferred. In Rome itself, though senatorial commissions to supervise public buildings and facilides had been instituted by Augustus, who had also revived the priestly colleges, a de facto ban on aristocratic initiatives had been imposed to reduce the risk of challenges to imperial munificence.⁸ Senators could still, on defined occasions, give games, but all main public buildings and facilities, the major festivals and the grain supply became the responsibility of the emperor. From Augustus on, emperors haphazardly extended their operations in this line to the towns of Italy and the provinces, using tactics which included, for example, paying for buildings through their relatives, and diverting or remitting imperial taxes to local councils to aid municipal projects.⁹

Beyond this state munificence which was arguably necessary there was the ad hoc liberality expected of all rich and prominent men in the empire, and most expected of the richest and most prominent of all, the emperor.¹⁰ Friendship with the emperor and his trust were demonstrated in a courtier's receipt of estates and other gifts in cash and kind. Individual deeds had to be rewarded appropriately, whether a huge sum to an important freedman or a few coins to a street poet. An emperor could remit some of the taxes due from a city purely as a mark of his favour; a Nero could remit those of a whole province. In an ego- boosting display of superiority as well as of generosity the emperor could throw to the Roman crowd tokens for mystery prizes including cash and all kinds of objects. The range of imperial giving cannot be described exhaustively, nor was it meant to be: 'there is nothing that might not be hoped for from my magnanimity', said Nero.¹¹ Since such 'spontaneous' giving was an integral part of the role of emperor as, on a smaller scale, it was of local magistrates, it must be counted as an area of state expenditure.

The cost of all this munificence, both necessary and spontaneous, is impossible to compute. More important is its size in relation to military expenditure. Under Claudius, for example, the draining of the Fucine lake over eleven years is said to have employed 30,000 men (though perhaps 30,000 was the aggregate total of man-days), and the estimated costs of the new port at Ostia were expected to kill off his enthusiasm for the project. There were other imperial building projects in Rome, lavish

⁸ Eck 1984(0 39). ' Bourne 1946(0 11 j); Corbier 1985 (o 124); Mitchell 1987(0 150).

"> Kloft 1970 (d 138). ¹¹ GCN 64 (lines 10-11).

shows and several handouts. The freedmen Pallas and Narcissus between them allegedly accumulated a sum equal to one and a half years' military budget.[484] It is likely that Claudius spent in and around Rome - and necessarily in actual coin - as much each year as the army in theory cost him, and in practice much military expenditure was notional since it was covered by supplies in kind. If we allow also for civil expenditure outside Rome and its environs, it is likely that the army, even if it was the single largest regular item in the imperial budget, in this period accounted on average for less than half of all imperial spending. The claims in later Roman writers that the reason for taxation was the need to pay for the armies which guaranteed peace have a propagandist whiff about them.

To meet this varied expenditure the state had a correspondingly varied range of assets and incomes. As heir to the ideology of the Greek city-state, the Roman government did not subject its own citizens, wherever they resided, to regular direct taxation on the person, and did not tax its own 'citizen land' (i.e. that held iureQuiritium), which meant mainland Italy and also the territories of Roman overseas colonies and of provincial cities which enjoyed the г us Italicum.ⁿ As an imperial power Rome levied direct taxes or rents on the rest of its subject lands and populations. It is dubious whether any coherent legal justification for this fiscal exploitation was elaborated under the Principate; instead pragmatism ruled.[485] Where sophisticated pre-Roman fiscal systems existed, mainly in the old hellenistic kingdoms, they tended to be adapted and maintained, and more generally there flourished a defensive ideology of fiscal minimalism (no new taxes, no increases to old ones). But, starting in Egypt, Augustus introduced an annual poll-tax in cash, Roman-style census arrangements gradually spread through the eastern provinces, and Roman fiscality — and, with it, monetization - was brusquely introduced to the northern and central European provinces.¹⁵ Although the new regular provincial poll-tax allowed Augustus and his successors to dispense with the irregular hellenistic capitation taxes which republican governors had continued to levy on occasion and to discontinue the revived triumviral levying of tributum in Italy, all Rome's subjects and even her own citizens remained liable to random summary exploitation such as confiscation of land for colonies or veteran settle­ment (not always to punish disloyalty), requisition of housing, animals and supplies for the use of the military and the administration, and conscription into the army.¹⁶

In the early Principate different direct taxes, assessed on different bases and according to different rates, continued to be levied from province to province. Republican modes of thinking and terms persisted: the fiscal value of a province was estimated as an annual cash sum, the word vectigalia could still be used of all fiscal revenues, direct and indirect, from a province, and stipendium of the totality of direct taxes from a senatorial province. But a new categorization was developing: vectigalia often now denoted only indirect taxes, and tributum was used of regular direct taxes (not, as in the Republic, of emergency cash levies), conceptually subdivided into those assessed on land {tributum soli) and those assessed on persons {tributum capitis). This was not a programmatic scheme for standardizing direct taxation - indeed some scholars deny that capitation taxes were levied in all provinces — but these terminological changes reflect some attempt to simplify and improve the overall administration of taxation and the loss by provincial governors of independence in fiscal matters in favour of the central imperial government.¹⁷

The collection of direct taxes was now mostly devolved to the theoretically autonomous cities and tribes of the empire, each of which was meant to produce a fixed annual sum of direct tax assessed in cash terms. The elimination of tithes and of their collection by Roman publicani in the Greek-speaking provinces seems to have been mainly the work of A. Gabinius and Iulius Caesar. Both the tithe and publicani persisted in Sicily, but neither Augustus nor his successors introduced publicani to collect direct taxes in newly created provinces.¹⁸ The total of direct taxes due from each community was computed by multiplying the taxable base — quantity of land and (probably) number of people — by the relevant rates. In some cases this will have followed on a Roman census; in others, presumably, it was simply what the city claimed was the traditional figure, while for many tribes it must have been an arbitrary guess. The city council (or tribal leaders) were obliged to make up any shortfall in the aggregate sum due, but probably more often made a nice profit, whether through extortion or because the actual taxable base had grown since the original assessment. The job of the local Roman financial official, the quaestor in a public province or the procurator in an imperial province, must have been mainly to ensure that the total due was paid on time, in full and (to introduce a further complication) in acceptable proportions of cash and kind.

Although the total tax dues of provinces and communities were usually expressed in terms of a lump cash sum, direct taxes on land were often assessed and collected in kind, mainly wheat, rather than cash. (Peasants presumably often paid local collectors in kind, and the collectors sold the produce and made the payments to the government in

Neesen 1980(0 151)25-9, 117-20.

Brum 1990 (d 119); Jones 1974(0 157) 164-8, 180-3; Cimma '9⁸¹ (^{D I21})'

cash, but this is a different matter.) The early evidence from Egypt and Britain for adaeratio, the commutation of wheat-dues for a cash payment at a fixed official exchange rate, and the more widely attested government purchases of grain (implying that, relative to needs, too much tax had in practice been paid as cash rather than in kind), suggest that from the start the Roman government could be flexible about the medium of pay­ment.¹⁹ The existence of an official exchange rate permitted the calcula­tion and recording of taxes in cash terms whatever the proportion actually paid each year in kind. No figure can be put on the average annual empire-wide ratio between direct taxes collected in cash and in kind, but probably more came in kind under the early Principate than is conventionally assumed.

Many indirect taxes, called vectigalia, were also levied in the Roman empire.²⁰ The main category of these were customs-dues (portoria) which were usually exacted at ports, on the imperial frontiers, at the boundaries between provinces or groups of provinces, and sometimes at internal boundaries within provinces. The rate on the eastern frontier was apparently 25 per cent of the value of all goods; known inter-provincial rates range from 2 per cent to 5 per cent. In Italy the imperial government drew revenue from a 1 per cent auction tax (centesima rerutn venalium), a 4 per cent (originally 2 per cent) tax on sales of slaves, and the tolls at the gates of Rome; it was also the recipient of the 5 per cent inheritance tax (vicesima bereditatum) which applied throughout the empire to Roman citizens of a certain wealth and without closely related heirs, and of a 5 per cent tax on the value of slaves manumitted by Roman citizens. In the cities of the empire other indirect taxes were imposed by and benefited the local authorities.

The collection of imperial indirect taxes continued in the early Principate as in the Republic to be farmed out to publicani. The old censorial task of fixing the contracts and supervising their execution must have passed to new imperial financial officials — in Italy this was certainly one function of the prefects of the state treasury.²¹ In theory the state conceded some profit margin to the contractors, but in pracuce the system avoided extra bureaucracy and stabilized receipts. The relative value to the imperial government of indirect as against direct taxes is impossible to assess, but they were probably crucial to the imperial finances. Being indirect they were politically easier to increase or invent than direct taxes, and in fact all the new taxes imposed in the early

" Tac Agr. xix.4; Neesen 1980(0 151) 104-16; Brunt 1981 (d 188) 161-2; Rathbone i989(e96o) >75-4-

General: de Laet 1949 (d 140); Neesen 1980 (0151)156-41; see n. 18 above. Cases of Asia and Egypt: Engelmann and Knibbe 1989 (в 229); Sijpesteijn 1987 (e 965); Wallace 1938 (e 979).

Dio Lx.10.3; Corbier 1974 (d 122); Millar 1964 (d 149); see nn. 18 and 20 above.

Principate were indirect. Other advantages were that they produced a fairly immediate cash revenue, which in several cases was actually paid over in Rome, and that Roman citizens, perhaps with the excepdon of veterans, were not exempt. Indeed, if we except the landholdings of Roman citizens in territories not exempt from tributum soli, indirect taxes were almost the only regular means of exploiting private Roman wealth open to the imperial government.

The state also had fixed assets consisting principally of land, urban properties and mines. In theory all ager publicus which had not been granted away into private ownership still belonged to the Roman state and bore a rent. It is unfortunately unclear how much remained, and whether and how rents from it were collected, but it is known that the government still farmed out to publicani the collection of fees, called scripturae, for the use of public grazing lands in Italy, Cyrene and perhaps elsewhere too. Many cities in their own right also owned and leased out estates, not just in their own territories, and this category of public ownership was constantly being increased by bequests from private individuals. As regards other fixed assets of the state, public buildings should perhaps be counted rather as financial liabilities. Temples, however, contained treasures which could be 'borrowed' in times of emergency, and warehouses, the shop areas in porticoes and other functional buildings could be leased out by the civic authorides.

The possessions of the emperor himself, his patrimonium, must also be counted as state assets.²² The emperor was not just another member of the empire-wide wealthy elite who discharged public functions and funded public projects out of their own private resources. Much imperial property may have been acquired through private transactions such as inheritance, personal gifting or purchase, and emperors made wills as if they were private persons. However the imperial patrimonium passed from emperor to emperor as part of the office rather than through normal inheritance, as is patent in the cases of the emperors from Otho to Vespasian but was perhaps first recognized on Gaius' accession, whereas no consul, for example, inherited his predecessor's personal fortune.²³ Furthermore, the patrimonium gradually established its claim to a number of 'public' sources of income, and although it was in theory managed separately from the state finances, its personnel, both equestrian procur­ators and imperial freedmen and slaves, soon became an integral part of the state bureaucracy.

The basis of the patrimonium was the family estates, urban properties, slaves and other possessions of the Iulii, Octavii and Claudii. Under Tiberius the patrimonium in Italy was still modest, according to Tacitus -

²² Millar 1977 (a 59) ch. IV and Apps. 1-3; Rogers 1947 (d 154); Crawford 1976 (d 125); Parassoglou 1978 (e 9)6); Rathbone 1993 (e 962). и Bellen 1974 (d 112).

that is by senatorial standards; the comment implies significant growth by the end of the century. Emperors were also from the beginning massive landowners in the provinces. Augustus' acquisition of substan­tial estates in Egypt (known locally as ousiat) is a prime example; another is Nero's confiscation of 'half' of Africa.[486] The patrimonium grew in ways unparalleled by any private estate because the emperor's position opened unique avenues for increasing his possessions. Like any Roman noble, he expected and received legacies from relatives and friends, but under an acquisitive emperor the category of 'friends' could embrace almost all the Italian nobility and some prominent provincials, especially client kings. In the first century the patrimonium gradually usurped from the aerarium the right to bona vacantia and caduca and bona damnatorum (that is property with no known owner, usually because the former owner had died intestate and without kin, property whose testamentary disposition was legally invalid and property of condemned criminals). Since in Egypt these had all fallen to the fiscus since annexation, this was clearly a royal prerogative adopted from hellenistic practice. The patrimonium was also the beneficiary of booty (manubiae) from imperial campaigns, and of the gold crowns sometimes spontaneously offered by communities to mark victories. The emperor's landed properties, like those of any noble, contained sub-enterprises such as transhumant flocks, clay pits and potteries, tanneries and textile processing facilities, urban craftshops and so on. Under Augustus and Tiberius almost all mines not already run by the state came into the hands of the patrimonium, and often if not normally were put under military supervision, and new mines, like those in Britain, followed suit. Some quarries too became imperial proper­ties.[487] In Rome itself the emperors had warehouses where they stored everything from produce of their own estates to exotic gifts from foreign embassies. There was also the palace, enlarged successively by each Julio-Claudian emperor, together with the imperial gardens; though the site and buildings were hardly saleable, the rich furniture and furnishings represented a significant reserve of wealth. The contribution of the patrimonium to the imperial finances cannot be quantified, but its political importance is clear: it enabled emperors to claim that they subsidized rather than exploited the state revenues.

These, in outline, were the resources available to the imperial government to meet its expenditure. The last topic which must be added before the management of the imperial finances can be discussed is the imperial coinage and its production.[488] The coinage of Rome as stabilized by Augustus in or by 19 B.C. was trimetallic, consisting of almost pure gold and silver coins and a range of what is for convenience termed 'bronze' (or aes) coinage, though some pieces were almost pure copper while others were orichalcum, an alloy of copper and zinc. In the system established by Augustus the main coins in circulation and their official relationships of value were as follows: the gold aureus, the silver denarius of which there were 2 5 to the aureus, the copper as of which there were sixteen to the denarius, and various fractions of the as\ the normal unit of account, however, remained the sestertius, equivalent to four asses, though the actual (orichalcum) coin was rare. As regards weight, forty or forty-two aurei were struck from one Roman pound of gold, and eighty- four denarii from one pound of silver. These standards held until Nero's reform of a.d. 64. He retained the relative face values of the Augustan system but struck forty-five aurei and ninety-six denarii respectively to the pound. The silver content of the denarius was also reduced to an average of 93.5 per cent. Although Nero's attempt to introduce a wholly orichalcum 'bronze' coinage was a rapid failure, his system in its essentials lasted until Commodus.

The various denominations in the Augustan-Neronian system were minted in varying quantities, often discontinuously, from two main and some minor mints. The mint at Lyons (Lugdunum) produced almost all the imperial gold and silver coinage from 15 b.c. onwards until Nero (or possibly Gaius) transferred production to the mint at Rome. From 23 or 19 b.c. the Roman mint produced most of the imperial 'bronze' coinage, but in most reigns there were sporadic and sometimes heavy regional issues of imperial type from provincial mints. Output of mainstream imperial coin was supplemented by the issue of silver tetradrachms, didrachms and drachmas by the mints of a number of Greek cities, notably Ephesus, Pergamum, Caesarea (in Cappadocia) and (Syrian) Antioch. These and other city-mints also produced sporadic issues, occasionally quite large, of bronze fractions. Egypt had its own internal coinage based on the Alexandrian tetradrachm. In the west local mints had always been rare. Most were in Spain, they produced only bronze coin, and those which survived Tiberius were shut down by Gaius. The broad pattern of supply of coinage in the period as a whole is thus that the mints at Rome and Lyons produced gold coins for the whole empire and silver and bronze for all the western provinces; western silver coins also reached the East but were outnumbered by the regional producuons there, and the eastern provinces were almost wholly dependent on very locally produced bronze coinage.

Minting was essentially controlled by the emperor. Most of the bullion used must have come from sources under imperial control — an early example is the exaction of bullion in Gaul by Augustus' freedman

3i8

8. the imperial finances

procurator Licinus, presumably to prime the new mint at Lyons.[489]Supervision of the state mints, at Rome at least, was again entrusted to young senatorial tresviri monetales, whose full title (aere argento auro flando feriundo) implies oversight of the production of all coins. Briefly under Augustus they were allowed to choose the types for some issues, but that was the extent of their independence. The letters 'S.C.' (senatus consulto) which appeared on Augustus' new bronze coinage, and on some provincial and some Neronian imperial issues, do not, it is now generally agreed, indicate any continuing senatorial control of minting, but advertise that this was the official Roman coinage, perhaps originally with reference to a senatorial vote of approval for the new weight standards of the Augustan system.[490] In the provinces many 'local' coinages, such as the cistophoric tetradrachms of the province of Asia (which bore Latin legends), were in effect 'imperial' coinages. The mint at Alexandria was under direct imperial control, and under Tiberius the silver-weight of its tetradrachms was adjusted to match that of the denarius; around the same time Palmyra and the Jewish rulers were made to bring their silver coinages into line. The closing of all the local mints in Spain must indicate imperial intervention, and it is noticeable that many sporadic eastern issues coincided with military operations in the area.[491] The emperor could control minting when and wherever he wanted; that he sometimes allowed local initiative is not evidence for a real division of authority. The emperor thus was in theory able to regulate in broad terms the quantity and type of coinage in circulation; the quesdons of whether and why he did or did not lead into the wider issue of the management of the imperial finances in general.

Detailed quantification of coin production in the early Empire must await systematic study of the number of dies used for each issue, although even this will leave considerable uncertainty about the scale of issues.³⁰ Compared to earlier and later eras the surviving gold and silver coinage of this period is relatively rare; significant quantities of the heavier republican denarii condnued to find their way into hoards through to the end of the first century a.d. Augustus had to mint extensively to establish his new system of bronze coinage, but there was a drastic fall in producdon later: Tiberius and Gaius, for example, closed the western provincial mints, and no imperial bronze was struck in the first ten years of Nero's reign. There is no evidence for regular recall and re-minting of old coins (which would have been very expensive). Old coins collected by the state were simply re-issued. The main sources of metal for minting new coins were bullion acquired through taxadon orconfiscation and above all the mines which had rapidly fallen under imperial control. It is therefore very likely that the overall stock of coinage in the early Empire was constantly if gradually increasing.

The rationale underlying this pattern of minting is a controversial topic.³¹ It is likely that the imperial government recognized some political responsibility, incurred through its near monopoly of minting, to maintain in circulation an adequate supply of the full range of denominations. The rare but heavy issues of small denominations, however, must be taken as one-off responses to particularly noticeable shortages and thus as indicators of a lack of any forward planning. The famous 'crisis of liquidity' at Rome in a.d. 33 tells the same story for the higher denominations.³² Clearly there can have been no government statistics for the volume of coinage in circulation, for any lump of gold or silver, including coins of the Roman Republic and of the hellenistic kings, could be used for exchange, while imperial gold and silver coins could be hoarded or melted down as bullion. These considerations undercut modern theories that changes in the rate of output and in the weight and purity of the imperial coinage represent attempts to keep it in tune with the changing market values of the uncoined metals; it is more plausible that the 'bronze' was a largely token coinage from the start, and that the denarius was deliberately overvalued in relation to the aureus so that it had a token premium against gold which discouraged private melting down of silver coins. Indeed it is very difficult to construct any satisfactory economic explanation for Nero's 'devaluation' of the silver and gold coinage, the only major monetary adjustment in this period. The common view that it was a device to stretch imperial funds is unsatisfactory, partly because earlier heavier coins were not all driven out of circulation, and mainly because it ignores the simultaneous attempt to introduce an all-orichalcum aes coinage.³³ Nero was probably trying to reform the whole monetary system for a mixture of administra­tive and aesthetic reasons. Normally, however, emperors seem to have thought little about minting, which was ordered primarily in response to specific immediate needs. As long as the mines, supplemented by booty and confiscations of bullion from individuals, continued to produce sufficient new metal for minting, there will have been no obvious need to worry about questions of policy.

State income and expenditure in cash in the Roman empire is best visualized not as a massive annual ebb and flow of coin between the provinces and Rome, but as a series of provincial whirlpools, some of them spilling over into others and all being sporadically topped up from the imperial mints at Rome and elsewhere. The whole system functioned

For example Crawford 1970 (d i 26); Lo Cascio 1981 (d 144); Howgego 1992 (d i }s).

Rodewald 1976 (в 548) ch. 1. ³³ Bolin 19)8 (o 11 3) ch. 4; Lo Cascio 1980(0 143).

}20

8. the imperial finances

largely under its own momentum with little direct intervention from the central government. It seems that, following republican practice, each province had a 'fiscus' (literally 'basket', sc. for holding coins), a sort of branch office of the main state treasury {aerarium). The chief task of each fiscus was to receive and record the lump sums of direct and indirect taxes due from the local communities and tax-farmers. It also had to pay out for expenses in that province: the salaries of the governor and his subordinates, any imperially funded building projects, and the cash costs of the garrison if there was one.

In republican Rome the central state treasury, to which all state revenues were in theory due and from which expenditure was made — though in practice many transactions were handled entirely by the provincial fisci — was the aerarium, located in the temple of Saturn. This treasury continued to exist in the Principate, now called the aerarium Saturni to distinguish it from the aerarium militare, the separate 'military treasury' established by Augustus in a.d. 6 with the new and limited function of paying the discharge bounties due to veterans out of the revenues earmarked for them.[492] In addition to these public treasuries formally constituted under senatorial supervisors, there existed the originally private administrative organization of the emperor's patrimo- nium or fiscus (as it was sometimes known), staffed by imperial slaves and freedmen, which swiftly came to assume the leading role in the administration of the state finances as a whole; hence the trend for fiscus to supplant aerarium as the general term for the fiscal and financial centre of the Roman state.

Admittedly the nature and origins of this imperial fiscus have been keenly disputed.[493] A common view is that a new imperial treasury called the fiscus, separate from the patrimonium, was set up parallel to the aerarium Saturni, probably by Claudius and perhaps together with the creation of an 'accounts department' {a rationibus) of the imperial familia headed by Pallas. Another suggestion is that this fiscus was a sub-unit of the aerarium which, on the analogy of provincial fisci, handled the finances of the emperor's composite provincia. The evidence, however, tells against any neat division between 'imperial' and 'senatorial' finances and their control. Under Augustus the aerarium Saturni was credited with the revenues of the new imperial province of Egypt, as was the aerarium militare in a.d. 17 with those of Cappadocia; the aerarium Saturni administered the financing of the vigiles, the new imperial fire-brigade, and continued to do so into the third century, and the aerarium militare functioned independently into the same period.[494] In the summaryaccount of the finances of the empire which Augustus left on his death, along with his private will, he listed the cash in the aerarium, the cash in the provincial fisci and the sums due from the tax-farmers.³⁷ Clearly no new imperial treasury was officially recognized under Augustus, and there is no good evidence for one under his Julio-Claudian successors. The emperors were able to control state finances without diverting revenues into a new separate treasury.

The aerarium Saturni had no real financial independence. Although it was supervised by senatorial officials, the changes from praetors selected by lot to quaestors and then to ex-praetors chosen by the emperor are one sign of subordination to imperial control.³⁸ The duties of these officials, as under the Republic, and of the senatorial prefects of the aerarium militare, were restricted to technical functions such as administering the tax-farming contracts, investigating accusations of tax avoidance and prosecuting defaulters; because this often meant dealing with upper- class Italians, it was politic to employ officials of senatorial status.³⁹ It is, furthermore, unclear what revenues and expenditure continued in practice to be accounted for — let alone actually received or disbursed — by the aerarium Saturni. When Augustus, for example, drew up his summary of the state finances, a large percentage of the sums involved will have been in the fisci of imperial provinces under the control of imperial freedmen or equestrian officials, and any cash he held in Rome was presumably accounted for as being 'in' these fisci or as 'due' to the aerarium. These sums, as well as not passing through the aerarium, had apparently not been reported to its officials, for Augustus referred the Senate for details to the members of his familia who kept the accounts. These imperial clerks, technically the financial administrators (a rationi- bus) of the patrimonium, were thus not invented by Claudius, even if he was responsible for giving them a more formal 'departmental' organiza­tion. This may have encouraged people to think in terms of an imperial treasury based on the administration of the patrimonium, and hence called the fiscus, and in practice the role of the aerarium Saturni may increasingly have been confined to receiving the fiscal surpluses from public provinces and revenues raised in Italy and to administering public expenditure in Rome and Italy which was nominally under senatorial control such as that on aqueducts and temple maintenance and rituals.

In some respects Augustus had behaved in the tradition of late republican commanders, notably Pompey. There had not, therefore, been any formal division of responsibility, and in theory the aerarium Saturni remained the state treasury. In practice, however, the emperors controlled all financial policy. After Augustus only Gaius ever again

³¹ Suet. Aug. 101.4. ³⁸ Millar 1964(0 149) 34.

³⁹ For example, the case of Claudius: Suet. Claud. 9.2.

offered any account of the imperial revenues and expenditure to the Senate. Instead of the emperor's agents reporting to the aerarium, we must suppose that its prefects had to make their records available to the imperial accountants who drew up overall statements of the state finances for the restricted benefit of the emperor and his advisers. The question of administration is really a red herring: Augustus and his successors controlled the state finances by monopolizing the decision­making on financial matters. More precisely state finances were depoliti- cized by the death of republican politics - it was no longer open to ambitious individuals to propose controversial expenditure (wars, buildings, doles) or fiscal changes. Now a standing army received automatic payment in cash and kind, the Roman populace had a permanent grain supply laid on by the emperor, the provinces had a system of regular taxation which for over two hundred years underwent only minor adjustments.

The stability of Roman taxation at a level which, if it hurt individual peasants, was low for each community as a whole is often used to help explain the acceptance and support of Roman rule by the upper classes of the provinces.⁴⁰ But the proposition should perhaps be reversed: the Romans were so dependent on this local co-operation that to avoid the risk of disaffection they rarely dared to increase provincial taxation, and its level constrained rather than was determined by imperial expenditure. In the Julio-Claudian period expenditure on the army must have increased gradually as auxiliary forces were turned into regular units. Total state revenues, however, will also have increased as new areas were converted into provinces subject to direct Roman taxadon. The evi­dence suggests that, outside Egypt, censuses were not regular and neutral operations but occasional deliberate attempts to increase the tribute assessments of individual provinces; if so, it would appear that as Gaul developed economically, its tribute was increased.⁴¹ Similar increases probably occurred in other relatively new and underdeveloped provinces as, for instance, in Moesia under Nero through the settlement of Transdanubians.⁴² In the Principate, however, only Vespasian is credited — and dubiously so — with widespread increases of tribute, examples of imperial caution about the general level of provincial taxation are numerous, and individual communities could petidon for reductions in their tribute assessment and doubtless frequently did so, somedmes with success.

It is difficult to estimate the size and nature of the public profit made from the provinces by imperial Rome. The situadon can be pictured as an outer ring of coin-hungry fisci of frontier provinces with large

Jones 1974 (d 137); MacMullen 1987 (d 147).

Cf. Brunt 1981 (d i 18), modified in 1990 (a 12) 533. «2 GCN 228.

garrisons which kept solvent by drawing on the cash surplus of the fisci of interior civilian provinces.⁴³ How much or little cash surplus this left to be shipped to Rome is unknowable; against it must also be set all the newly minted coinage injected into the provincial system. But the profits of imperialism did not come only in cash. Direct taxes, although assessed and accounted for in cash terms, were partly collected in kind. Thus, for instance, insofar as soldiers received supplies in place of cash remune­ration, the fisci of frontier provinces need not always have been seriously short of coin; on the other hand civilian provinces may have produced surpluses in kind rather than cash. More importantly, the one provincial revenue which is certainly known to have been shipped to Rome is the annonal wheat.

While the revenues which could be drawn in cash from the provinces were limited, emperors were under constant pressure to spend munifi­cently, especially in Rome. Tiberius was exceptional in his accumulation of a large cash reserve, and Gaius' immediate spending of it was almost inevitable. Such savings undermined the justification for taxation, a mentality which was in part the legacy of the republican system of ad hoc financial arrangements, but in part derived from the emperor's monopo­lization of the control of the state finances. While emperors were happy to take the credit for beneficial expenditure, they also had to face personal criticism for the level of taxation, and preferred to spend rather than save. There could normally be no centralized reserves of wealth at all comparable to those, for instance, of the Achaemenid kings. It is also clear why for emperors who wished or were obliged to fund major new projects such as wars or building schemes and whose needs were normally for ready cash, the income from indirect taxes, particularly those raised in Italy, and that from the patrimonium had a special importance. In effect needy emperors turned to the Senate (and other rich nobles), whether it was Augustus instituting the 5 per cent inheritance tax or the villain of later senatorial rhetoric, the emperor who killed and confiscated to raise cash. The imperial wealth was enormous but, through a combination of political weakness, difficulties of commu­nications and transport and incomplete monetization, much of it could not be mobilized effectively by the central government. Although the period from Augustus to Nero saw an overall rise in expenditure which was at least matched by an overall increase in revenues, the lack of central reserves was a weakness embedded in the system from its inception and one which was to cause problems for the rest of the Principate.

⁴³ Hopkins 1980(0 153).

THE SENATE AND SENATORIAL AND EQUESTRIAN POSTS

RICHARD J. A. TALBERT

I. THE SENATE¹

There can be no question that the 20s B.C. and the half century which followed were a time of unparalleled change for the Senate and its members. Augustus was its principal instigator. Once peace had been secured after the long civil wars, the 'restoration of the Republic' was one of his foremost aims. By definition that touched closely the central institutions of the Republic, the Senate among them. The size and quality of senatorial membership engaged his attention first. In size it had expanded to 1,000 or more, partly because of numerous adlections by Iulius Caesar as dictator, partly because following his death others successfully used influence and bribery to gain admission by the same means. Moreover, by raising the total of quaestorships from twenty to forty, Caesar had doubled the number of new members each year, since tenure of this junior magistracy in practice offered life membership of the Senate. As early as 29 B.C. Octavian (as he then was) used a review of the senatorial roll to exclude 190 members on one ground or another. It was

¹ Since contemporary testimony is largely lost along with the Lex Iulia of 9 b.c. which governed procedure, the main sources of knowledge for the Senate during the Julio-Claudian period are the later historical writers Tacitus, Suetonius and Cassius Dio — in particular Tacitus, who certainly drew upon the detailed record of senatorial proceedings {acta senates) for his Annals, although to what extent and by what means remain matters of considerable dispute (Talbert 1984 (d 77) ch. 9; Brunt 1984 (л io)). Inscriptions and papyri make a growing contribution. An impression of the nature and scope of senatorial legislative activity can be formed by drawing together material from legal writers and elsewhere (Talbert 1984 (d 77) ch. 15 sect. 5). Seneca's vivid sketch of the heavenly senate in session on Olympus, presided over by Jupiter (Apocol. 8-11), parodies its Roman counterpart of which he was himself a member, and offers a rare piece of contemporary insight. If it is accepted that Diocletian's Curia in Rome (built near the end of the third century and still standing today in a restored state) is in effect a reconstruction of the Curia Iulia, then it is possible to observe closely the meeting-place where most of the Senate's sessions were held: see further A. Bartoli, Curia Senatus, lo scam e H restauro (Rome, 1963).

Inscriptions are the main source for knowledge of senatorial and equestrian administrators and their work. Significant in this connexion from Augustus' reign onwards is the growing frequency with which records listing all the offices a man had held were no longer inscribed just posthumously, but during his lifetime too (Millar and Segal 1984 (c 176) ch. ;).

Modern discussion: Talbert 1984 (d 77) offers a starting-point on most aspects; for senators and their careers, see also Hopkins 1983 (a 46) ch. 3. Much relevant documentary material is assembled in FIRA i.

324

probably also during the 20s that he reduced the number of quaestor- ships to the old figure of twenty. Either then, or during the 'teens b.c., he took the consequential step of reducing the lower office holders (mostly aspirants to the Senate, not yet members) from vigintisexviri (twenty-six) to vigintwiri (twenty).

A Senate of about 800 still seemed too large. When Augustus returned to the task of reducing it further by another review of the roll in 18 b.c.,² his preference is said to have been for a body of just 300: the simultaneous removal of as many as 500 members would thus be required. Unless he was displaying an astonishing lack of foresight, a more profound reappraisal of the role of the Senate would have been called for next, since all the existing functions assigned to the corporate body and its members could barely have been carried out by such a reduced group. In the event, however, Augustus abandoned any drastic aims of this type, and enrolled about 600 members by a peculiar method which combined co-option and the drawing of lots. Thereby the Senate returned to the approximate size which the dictator Sulla had made it. Up to the end of the Julio-Claudian period there are known to have been at least two more revisions of the roll during Augustus' reign (around 13— 11 b.c. and in a.d. 4), and a third carried out by the emperor Claudius and L. Vitellius as censors in a.d. 47/8. But in none of these instances does there appear to have been further significant alteration to the size of the membership. Rather, the regular number remained about 600, though it should be understood that this figure was always just a notional optimum, never a fixed maximum or fixed total. The normal method of entry continued to be through the twenty annual vacancies in the quaestorship. On present evidence at least, the alternative of'adlection', or direct elevation of a non-member to a grade of membership within the Senate (at the emperor's instigation), was only used very sparingly indeed during the Julio-Claudian period.³

The quality of senatorial membership concerned Augustus, as well as its size. As his conduct of the reviews in 29 and 18 B.C. demonstrated, he was determined to rid the Senate of members who were immoral, irresponsible, or lacking means. His purpose was to create a body which should be an outstanding elite of princes - high-minded, statesmanlike, wealthy. He waited until 18 B.C. to translate this ideal into reality. From that time all members had to be worth at least one million sesterces rather than just showing the modest equestrian census of 400,000, which was all that had previously been required.⁴ He appreciated the strain which would result, and over the years did help both worthy existing members who could not show the increased amount, and many prospective

² Dio Liv.13-15. ³ Demougin 1982 (d 36) 81-2.

⁴ Nicolet 1976 (d j 3); Millar and Segal 1984 (c 176) ch. 4.

3*6

9- the senate

entrants. Among Augustus' Julio-Claudian successors similar assistance is known to have been given by Tiberius (albeit sometimes in rather grudging fashion) and by Nero.

Also from 18 в.с. in all likelihood, the old custom was abandoned whereby every prospective entrant wore the distinctive badge of the senator — the broad stripe (latus clavus) on the tunic — even before he had ever gained the lowest senatorial magistracy and actually joined the corporate body as quaestor.[495] In future this was to be the exclusive privilege of those senators' sons who chose to follow in their fathers' footsteps. Other young men seeking to become the first members of their family to enter the Senate could certainly pursue this quest, as ever, but they could not wear the coveted latus clavus until they became quaestors.

This particular way of marking out senators' sons and encouraging them to emulate their fathers was one of Augustus' many experiments which did not endure. The restriction had evidently come to be disregarded by the 30s a.d. at the very latest. Instead the practice developed whereby all equestrian aspirants to a senatorial career were obliged to gain the emperor's permission to wear the latus clavus. How selective successive emperors were in their consideration of such applications is completely unknown. None the less it is clear that Augustus' experiment formed part of a wider effort to exalt not just senators themselves, but also members of their families, whom he actually defined for the first time ever as a separate, superior 'senatorial class'.

The class first appears formally in Augustus' marriage legislation of 18 b.C., and of course it did endure. Membership belonged to senators and their descendants to the third generation, plus wives. Once a distinct class had been formed on this pattern, it was natural for a haphazard growth of privileges and restrictions to become attached to it. Among privileges, special front seats at shows and a certain precedence at elections were introduced early; limited exemption from particular local obligations may also have been granted.[496] Among restrictions, a series of bans on marriage with the lowest classes, prostitution, and appearances in shows or on stage, were all intended to maintain the dignity of the highest class in society.[497]

Regardless of how they gained the latus clavus, all those intending to pursue a senatorial career had to undertake the cursus honorum as reformed by Augustus.⁸ Tenure of one of the twenty minor offices in the vigintivirate bestowed annually by the emperor was now made acompulsory prerequisite. Either before or afterwards a limited period of service in a legion as tribunus mtlitum was recommended, though it was never compulsory and was often omitted by those of aristocratic background. Entry to the Senate itself was gained by election to one of the twenty annual quaestorships, for which a candidate had now to have reached his twenty-fifth year (previously the qualifying age had been thirty).Thereafter, notionally with minimum intervals of just over one year between each magistracy, plebeians had first to hold one of the six aedileships or ten tribunates (patricians were excused this stage); next all competed for the praetorship, which could not be held before a candidate's thirtieth year (previously thirty-nine or forty). The degree of rivalry sharpened at this vital stage, depending upon the number of praetorships, which it was the significant prerogative of the emperor to fix from year to year. Augustus at first permitted as few as ten praetors each year, and even by the end of his reign seldom more than twelve. As a result, at this date an average first-generation senator could take pride in having climbed even this high. Augustus' Julio-Claudian successors became somewhat more generous (not least because the range of duties assigned to senators of this rank was extended), so that by the end of the period the total of praetorships seems to have been fluctuating between fourteen and eighteen. None the less the risk of rejection was still a real one.

Beyond the praetorship a minority of favoured senators could sooner or later proceed on to the highest magistracy, the consulship. Both the number of consulships each year, and the choice of holders, in effect quickly came to be a choice for the emperor alone to make. Initially there was no more than one pair of holders for the entire year on the traditional republican pattern. But from 5 в.с. these two 'ordinary' consuls, who retained the prestige of opening the year, were regularly replaced by one or two further pairs of 'suffect' consuls at variable intervals, with the result that up to six men were permitted to attain this distinction within a single year. Thereby competition for it became less intense, and there were more members eligible to occupy posts reserved for senators of this standing. Certain highly distinguished men might be privileged to enjoy the supreme honour of a second, and even a third, consulship.

In time Augustus formed the opinion that it was not just the membership of the Senate which required his attention, but also the workings of the corporate body. His revival of fines for non-attendance in 17 в.с. is an early sign of his impatience with members who failed to match up to his ideals. Though in theory a presiding magistrate had always had authority to fine absentees, not since the second century B.C. perhaps had it been normal practice to do so, with the result that this clumsy measure by Augustus merely served to give offence. Only in 11 в.с. did he act further, when he formally abolished the quorum of 400 which was still required for any measure passed to be valid. In all likelihood it dated back to Caesar's dictatorship, but must have been a dead letter ever since the reduction of the membership to 600 in 18 в.с.

The abolition at least cleared the way for positive reform in the shape of the comprehensive lex Iulia de senatu habendo (9 B.C.), which was intended to regulate every aspect of the Senate's workings. The principal purpose of the law was seemingly to improve levels of attendance, which had for some time been giving Augustus cause for concern. To this end fines were increased, but they proved as ineffective as ever, and were quietly dropped, never to be revived. Quorums (a modest 200 is the only one known)⁹ were introduced for every kind of business: in themselves they were no novelty, but never before had they been laid down so comprehensively. Even more important was the innovation of fixed days for meetings, the Kalends and the Ides of each month, so that members would know to set these aside for attendance. As some alleviation, for the four stated meetings of the holiday months, September and October, the law did permit no more than a quorum chosen by lot to be present, while the likelihood is that perhaps two stated meetings were normally cancelled around our Eastertime, when traditionally there had been a recess (res prolatae or discessus senatus). However at all seasons special meetings in addition to the stated ones could be called, if necessary at very short notice. It was equally in connexion with regulating attendance that the law made two further provisions. First, it required a list of all senators' names to be displayed publicly and updated each year. Second, it introduced a 'retirement age' for senators. Previously the formal position had been that every member was obliged to keep up his attendance for life. Augustus appreciated that it would be neither practical nor sensible to insist upon this, and thus had the law stipulate that members were no longer required to come beyond the age of sixty or sixty-five (it is not known which). All the same, they were still welcome to come voluntarily, and many did.

Beyond all this the Lex Iulia codified senatorial procedure. That really did represent a new departure, since previously the proceedings seem to have been governed almost exclusively by custom, rather than by written statute. So it was probably now for the first time that features like the order in which opinions were to be asked for, or the manner in which a vote was to be taken, were actually written down. Such codification no doubt appealed to Augustus' sense of order. Even so it is striking that he does not appear to have exploited the opportunity to change procedure much. In practice meetings seem to have beeA generally conducted in just the same way after 9 в.с. as before. There is no foundation to the ⁹ FIR A i 68 col. V lines 106-7.

modern claim¹⁰ that the law in some way curtailed the ancient right of a member, when called upon for his opinion (sententia), to speak first without time limit on whatever subject he chose (egredi relationem). This right was retained and was still exploited.

Of course what neither the Lex lulia nor any other law ever codified was the position of the emperor in the Senate. His presence was a major new feature to which the corporate body had to adjust from the 20s B.C. All emperors were patrician senators and must have headed the list of members during their reigns, though Augustus alone of the Julio- Claudians took the title princeps senatus (from 28 B.C.). In his case, too, formal difficulties were few before 23 B.C., since he was always consul and frequently out of Rome. Thereafter, however, the need was felt to offer him the guaranteed opportunity of bringing forward one item at any stage of any meeting — what has been dubbed somewhat inaccurately the ius primae relationis — as well as authority to summon the Senate as often as he pleased (in theory he could already do this by virtue of his tribunicia potestas). In 19 b.c. he was granted the right to sit on the president's tribunal at meetings, in between the two consuls. At some stage, too, as early as Augustus' reign, there was recognition (not necessarily formal perhaps) of a unique right of the emperor to have business put forward by letter rather than in person. All these privileges must have been conferred upon subsequent emperors on their accession.

At least up until a.d. 8, when old age compelled him to reduce his activities, Augustus showed the Senate respect by attending not just as president, but also as a private member. The one meeting which we know him to have missed deliberately was the occasion in 2 B.C. when the discovery of his daughter Iulia's scandalous behaviour had to be made public: in his shame he could not face the Senate in person, but sent a letter instead. Unfortunately the source-material is lacking which would allow us to build up a picture of his participation and performance at meetings in the way that can be done for Tiberius through Tacitus' Annals. In general, however, it is clear that he did take an active enough part in debate, although two major difficulties in this connexion quickly made themselves felt.

The first was the nature of members' reaction to the superior position of the emperor, which might take the form of respect, or fear, or resentment, according to different individuals' viewpoints. These feel­ings sprang from a variety of causes: the knowledge that in practice nothing which the emperor requested or openly supported could be refused; the recognition that every senator's advancement depended in large measure upon his approval; and the realization that control of many key spheres of government had effectively become his alone. Even

¹⁰ Mommsen 1888 (a 65) 111.2. 940.

many of the Senate's meeting-places were now powerful symbols of the imperial regime — the Curia Iulia, begun by Iulius Caesar, dedicated by Octavian in 29 B.C., and thereafter adorned with a growing number of monuments and dedications in honour of the emperor and his family; the temple of Apollo on the Palatine, close by the emperor's residence; and from 2 B.C. the temple of Mars Ultor in front of which was sited a great statue of Augustus victorious in a chariot. Under such circumstances, and in such surroundings, members came to feel, more or less willingly, that it was pointless any longer to take an active, critical, independent part in sessions, when the result seemed a foregone conclusion, and no more than officially selected extracts from the detailed record of proceedings — acta senatus, instituted by Iulius Caesar in 59 в.с. - were now permitted to be made public. In addition certain matters of the highest importance were never even referred to the Senate at all. It is hardly surprising that the only two known instances of open senatorial disagreement with Augustus were cases where he perhaps expected opposition to be voiced anyway — a request to have not one colleague, but two, whenever he held the consulship, and an offer after his illness in 2} в.с. to read out his will. Perhaps more characteristic were the meetings under Augustus' presidency where frustration at members' reluctance to formulate independent opinions led him to call names at random rather than in the customary order of seniority.

Despite Augustus' efforts to counter the trend, this understandable reluctance was to persist indefinitely. Tiberius' impatience with it as emperor prompted his allegedly regular exclamation on leaving sessions 'O homines ad servitutem paratos', 'O men ready to be slaves!'.[498] It must be reflected again by the otherwise unknown Titius Rufus whose claim that 'the Senate thought one way and voted another'¹² led to his indictment in a.d. 39; and there is no doubt that it was a principal target of the consular Thrasea Paetus, who consciously risked Nero's disap­proval by his outspoken encouragement of greater independence on the part of fellow members in the late 50s and early 60s. The most vehement attack on such senatorial reluctance, however, is made in the speech of an unidentified senator (in all likelihood the emperor Claudius) preserved on a papyrus fragment:

If these proposals meet with your approval, Conscript Fathers, say so plainly at once, in your own considered words. But if you disapprove, find another solution, yet do so in this temple, or, if you perhaps want a more generous interval in which to think, take it, provided you remember that, whatever the place you should be summoned to, you must give us your own opinion. For

Conscript Fathers, it is most unbecoming to the dignity of this order here that just one consul designate should deliver a sententia, and even this drawn word for word from the relatio of the consuls, while others utter the single word adsentior, and then when they depart say 'Well, we spoke'.¹³

Even where the emperor took care not to express a view, his relatives (who generally pursued senatorial careers) might still be regarded as speaking for him. Thus in a.d. 13, when alternadves to the 5 per cent inheritance tax were under discussion, Augustus specifically forbade Germanicus and Drusus to make any suggestion, for fear that it would be regarded as his, and adopted without more ado.

The second main difficulty which acted as a curb on the freedom and vigour of senatorial proceedings in Augustus' reign was his introduction some time between 27 and 18 в.с. of a consilium to consider items of business in advance of their being laid before the full corporate body (distinct from the consilium principis, for which see p. 290). It must be acknowledged that this committee was intended to have no more than such a preparatory function. Yet for all Augustus' efforts to uphold that aim, members in the full Senate would hardly have been human if they still did not suspect that they could exercise only the most limited influence after the 'real' debate had already occurred in the committee, and the 'real' decisions had been taken there. Under such conditions few members were going to have the appetite for a wide-ranging, frank discussion in the full Senate. Their worst fears can only have been confirmed in a.d. i 3 when Augustus (now in extreme old age) had the membership of the consilium reformed and its decisions granted authority equal to that of the full Senate.

In a pithy summary Tacitus later wrote of Augustus 'drawing to himself the functions of the Senate, the magistrates and the laws'.¹⁴ There is a large measure of truth in the allegation: even though not only Augustus but also all his successors studiously derived their formal authority from the Senate, it did still have to adjust itself to a curtailed prerogative. Of course many traditional functions remained. The Senate legislated actively, and its resolutions came to be recognized as law without the need for confirmation by a popular assembly. Honours were bestowed in greater quantity and variety than ever. The Senate's authority in matters of religion was still accepted as supreme, and it continued to be approached by embassies, albeit in reduced numbers. On the other hand the emperor in large measure reserved to himself matters relating to the army and foreign affairs; public finance; and the adminis­trative oversight of a large group of existing provinces, together with that of all new ones. In consequence the Senate lost for ever the major

¹³ FIRA I 44 col. ill lines 10-22. ¹⁴ Am. i.i.

prerogative (already challenged formidably in the late Republic) of determining the disposition of the state's military forces year by year and the extent of the territory to which it laid claim. The creation of supervisors for roads, aqueducts, the distribution and supply of corn, and for other concerns (treated further below), in practice represented further encroachment upon its formerly exclusive authority.

However, despite the fact that the republican Senate had seldom shown more than the most desultory concern for such matters, Augustus was still scrupulous in arranging not just for the new officials to be appointed by the Senate, but also for their activities to be authorized by it. He likewise constantly informed and consulted the Senate about military, provincial, diplomatic and financial affairs, in addition to inviting its approval of significant changes or unusual expedients in these spheres.¹⁵ In many instances it may be that this was not merely tact or caution, but rather that he was genuinely seeking to hear a range of proposals, to test opinion and to mould his reaction to it, as well as ensuring reasonable acquiescence in whatever might finally be decided. More than anyone Augustus knew how vital it was that he should not lose touch with upper-class opinion or seriously alienate it. Yet however open to advice he might appear, it always remained awkward for members to be confident of his purpose, or to judge the point at which they might be considered to have overstepped the mark in risking a frank statement of views. In this dilemma the majority preferred to take no risk at all, and the Senate as a deliberative body suffered.

Altogether Augustus' impact upon the Senate proved a mixed one. He showed it the greatest respect. While reducing the size of the member­ship, he raised its moral and social standing, he promoted regular attendance by a variety of means, and codified (though hardly altered) procedure. But for all his assiduous consultation of the Senate, and his avowed encouragement of frankly expressed opinions, it was impossible for members to ignore his overriding supremacy in the state and his effective usurpation of certain major senatorial prerogatives. The sena­torial consilium, especially after the strengthening of its authority in a.d. i j, acted as a further discouragement to the corporate body.

Tiberius' impact was equally mixed. Up to a point in the case of the Senate, as elsewhere, he merely continued Augustus' approach. While this is by no means an unfair assessment, it perhaps fails to give due weight to our sources' emphatic claim that the widest possible range of issues, public and private, great and small, was brought before the Senate by Tiberius, at least in the earlier part of the reign. Discreet warnings against such openness from Augustus' confidant, the eques C. Sallustius ¹⁵ Brunt 1984 (d 27); FIRA i 99 lines 1-7.

Crispus, were ignored. Moreover the Senate could feel that it enjoyed greater freedom to handle all this business, following the radical step taken by Tiberius on his accession: he was not content merely to reduce Augustus' senatorial consilium to its status prior to a.d. ij, but actually abolished it altogether. As a result the primacy of the full Senate was quite unexpectedly reasserted.

The Senate received a further boost during the early weeks of Tiberius' reign when elections to magistracies were transferred to it from the popular assemblies (though the latter continued to meet for the purpose of ratifying the choice of candidates). To what extent this development was an idea of Augustus rather than of Tiberius is obscure: but on present evidence there is no sign that the former ever wanted to do more than give the upper classes a prominent role in assembly elections, while at the least there can be no question that the dming of the change must have been decided by Tiberius.¹⁶ The Senate, of course, gained no formal power from it. Neither was there any relaxation of the existing constraints imposed upon both candidates and voters by the emperor's interest. For the consulship he condnued to support as many candidates as there were vacancies. For all other magistracies, however, his candidates would usually comprise no more than a proportion of the vacancies, so that there was genuine, fierce competition for the remain­ing places. Thus the transfer still gratified members, and did offer the corporate body a regular, active function to which much significance was attached.¹⁷ The details of how far in advance magistrates were elected thus in the Julio-Claudian period, and at what times of year, remain almost a blank: in all probability no set pattern emerged until a later date. An attempt by Gaius to return the elections to popular assemblies was frustrated by senators and soon abandoned.

Even more welcome to members was the trend which Tiberius more or less consciously encouraged whereby the Senate should exercise a regular jurisdiction as a high court.¹⁸ It had never done this during the Republic nor during the reign of Augustus. Rather, in his scheme of things this function was to be fulfilled by the jury-courts (quaestiones), which he overhauled and added to, and in which he gave senators an established place; in addition, from 4 b.c. certain charges of extordon (repetundae) might be heard by small panels of senators. Only for needs and cases beyond the normal routine did Augustus occasionally turn to the full Senate — in particular cases where his own prestige and interest were closely involved, or where the complexity or novelty of the issues were beyond the competence of a quaestio. In the earlier part of Tiberius' reign such formerly occasional referral became so frequent as to

¹⁶ Brunt 1961 (c 47); 1984 (d 27) 429. ¹⁷ Talbert 1984 (d 77) 202-4 ^and 34>—5•

¹⁸ Bleicken 1962 (d 248); Garnsey 1970 (f j;); Talbert 1984 (d 77) ch. 16.

9- the senate

constitute regular jurisdiction, while many more repetundae cases were considered to require a hearing before the full Senate rather than mere reference to a small panel. The trial of Cn. Calpurnius Piso in a.d. 20 for the murder of Germanicus may have been a turning-point. According to Tacitus,¹⁹ Tiberius himself openly acknowledged that it was exceptional to bring the case before the Senate rather than a quaestio. Yet from the 20s there remains no doubt that the senatorial court was well established, and the likelihood is that the quaestiones for treason (maiestas) and extortion (repetundae) became practically defunct in consequence.

Established senatorial procedure required little adaptation to accommodate judicial hearings, especially as the Senate had long been accustomed to entertaining pleas and applications, and adjudicating disputes. It is unlikely that its regular jurisdiction was ever sanctioned formally by law: none was necessary if the development enjoyed the emperor's support. While in theory the Senate as a supreme legislative body claimed the right (unlike a quaestio) to hear any charge and to fix any penalty, certain conventions quickly developed. The Senate became the principal court chosen to take cases of maiestas and repetundae in the Julio- Claudian period. Otherwise it normally confined itself to cases where individuals of high rank were involved; where the issue was particularly serious or scandalous; or where an affair attracted a special degree of public attention. Thus, for example, the Senate was a natural choice of court to hear adultery cases where persons of high rank were implicated, and where there might be associated charges, not to mention delicate political overtones. It was equally well fitted to investigate the collapse of an unsafe amphitheatre at Fidenae in 27 which caused catastrophic loss of life among the spectators: this resulted in the banishment of the builder, a freedman, and the drafting of regulations to prevent the recurrence of such a disaster.

The further convention seems to have developed that the emperor remained aloof from repetundae trials, according the Senate complete freedom to decide these as it pleased — a detachment which represented no special sacrifice on his part. It could only be otherwise with cases of maiestas, however. These were often brought to the emperor in the first instance and only referred to the Senate on his initiative. Since by definition they did touch his own safety and interest, he considered it important to make his views known and to have them adopted by whatever means might prove necessary. As a result the Senate was seldom left free to decide such cases, and bitterly resented the inevitable imperial interference, especially when the defendants were from the senatorial class. It became a major tragedy of Tiberius' reign that he did

334

¹⁹ Am. hi. 12.

less and less to control the bringing of maiestas charges. Moreover in any politically sensitive case he was above all concerned to see his own wishes met, rather than to encourage senatorial independence.

No less harmful was his withdrawal to Capri in 26, which turned out to be permanent. Up till that time his attendance - at debates and trials, as president and private member, even on election days — had been outstandingly conscientious. He had participated actively in proceed­ings too — suffering insults, being drawn into embarrassing exchanges, and even on occasion finding himself outvoted. Taken together with his other measures this behaviour understandably increased the Senate's confidence in the nature and value of its role, so that the effect of the emperor's isolation from the corporate body after 26 was all the more damaging.

Gaius' declaration²⁰ at his accession that he would never write to the Senate (and thus by implication would always attend in person) did indicate a fleeting initial reaction against Tiberius' behaviour during the previous eleven years. But it was left to Claudius to make a serious effort in this regard. While perhaps never as assiduous as Tiberius had been, he did none the less regularly attend meetings and trials, both as president and private member, and was an eager participant, bringing much business before the corporate body. He seems also to have been exceptionally severe in insisting upon good attendance by others. The ban on unauthorized private travel beyond Italy (and after 49 Sicily and Narbonese Gaul) by senators was stringently enforced. Nero's personal­ity and lack of experience led him to attend the Senate much less than Claudius, in particular towards the end of the reign when he became more and more estranged from it. But strikingly Vitellius' background and training led him to revert to the example of Augustus, Tiberius and Claudius. Tacitus²¹ notes that during his brief reign in 69 he made a point of attending the Senate even when the items on the agenda were only trivial.

Of all the emperors between 37 and 69 it was Claudius who made the most lasting impact upon the Senate by widening its membership. It is true that he stressed to the Senate itself the desire of both Augustus and Tiberius 'that there should be in this curia all the flower of the colonies and municipalities everywhere, namely good men and rich'.²² Yet in making such a claim he appears to be over-generous. Even though Iulius Caesar had introduced a few provincials, both Augustus and Tiberius - whatever may have been their ideal - in practice seem to have continued this trend no more than cautiously. Despite the favour regularly shown ²⁰ Dio ux.3.1. ²¹ Hiit. 11.91. ²² FIRA 1 43 col. II lines 2-4.

by emperors to respectful senators of distinguished ancestry, many old families soon ceased to be represented for a variety of reasons.²³ As a result there was room for a steady influx of novi homines, or first generation senators, who at this date were still mainly Italian. There was evidently no shortage of aspirants except for a limited period during the 'teens B.C. New patricians were created by both Augustus and Claudius. But it was only because of further initiative by the latter that provincials became in any way a notable element in the membership of the Senate. Even then, the great majority of these newcomers originated from the West of the empire: by contrast, not until after the Julio-Claudian period did more than a handful of easterners have qualifications and contacts which encouraged them to put themselves forward.²⁴

Hostile emperors like Gaius and Nero inflicted no more than short-term damage upon the Senate as a corporate body. For by the latter part of Tiberius' reign reform of its membership and workings was complete, while its functions had been satisfyingly enough redefined within the new constraints which the Principate imposed. In the spheres of legislation and jurisdiction the Senate remained notably busy. Meetings might last the entire day from sunrise to sunset; even so, many were required beyond the minimum of two each month prescribed by the Lex Iulia. Such miscellaneous attendance and voting figures as survive range from respectable to high²⁵ and are all the more remarkable in view of the considerable proportion of members who would always have to be out of Rome on official business or had reached the 'retirement age'. Debate was often sharp, and participation in it by no means confined just to the two highest grades, consulares and praetorii, who were consulted first. Great pride was taken in senatorial membership, and there was evidently never difficulty in attracting fresh aspirants, or in inspiring loyalty to the institution on the part of those who were elected. Moreover, even though the Senate may no longer have exercised much formal power, its members individually and collectively still exerted a decisive influence upon all the empire's affairs. While in one sense the well-being of the Senate, like everything else, remained painfully dependent upon the emperor's pleasure, in another the attitude of Augustus and Tiberius during their long reigns set a standard which senatorial opinion could ever afterwards demand that each of their successors maintain. These values were strongly advocated by senators and to a significant extent observed by responsible emperors, very much to the benefit of the corporate body and its prestige. Thus, as Tacitus²⁶ has Otho emphasize in the most high-flown surviving statement of the Senate's significance

²³ Hopkins 1985 (a 46), ch. 5. ²⁴ Halfmann 1979 (d 44).

²⁵ Talbert 1984 (d 77) ch. 4 sect. 2; Gonzalez 1984 (в 234) 76. ²⁶ Hist. 1. 84.

for Romans, it was the institution which continued to be seen as the permanent embodiment of the ancient respublica.

II. SENATORIAL AND EQUESTRIAN POSTS

No princeps, however active, could run the empire single-handed. Moreover the administrative functions fulfilled by the annual magis­trates elected at Rome were deliberately curtailed in scope. It is true that they continued to preside over a variety of courts there, and that quaestors acted as financial officers in the ten or so senatorial provinces. In addition three magistrates in office acted as mint supervisors, while between 23 b.c. and a.d. 56 there were others who administered the state treasury in Rome. But that was about all. For everything else the emperor had to seek assistance, principally from the upper classes. Here the role of senators was an outstanding one. The individuals invited to advise the emperor in his private consilium would be drawn largely from their ranks. Their formerly exclusive privilege to govern provinces and command legions was barely infringed either during the Julio-Claudian period or long afterwards. These were two functions of vital importance which alone by a.d. 68 called for the services of over fifty members at any one time, nearly all of them consulares or praetorii (men who had been consul or praetor respectively); further senators would accompany governors as legates.

The proconsuls of the senatorial provinces were still chosen according to the traditional method of the lot to serve for just a one year term, which would normally be expected to begin between our Easter and mid-summer. The arrangements for drawing lots, and the timing, are mostly obscure. Appointment as proconsul of Africa or Asia came to be offered to the senior consulares who had not held either post already. In this instance, therefore, once the two men eligible and willing to accept appointment had been identified, the drawing of lots was confined to deciding which province each would take. It may be that a broadly similar procedure was followed in the case of other proconsulships too, all reserved for praetorii (although tenure of more than one such post was permitted). Since there were as many as eight posts to be assigned thus, the lot could operate very much at random, and it does seem to have been left to do so. Such instances of individual manipulation as have been suspected appear exceptional; the same applies to extended terms of office.²⁷

Apart from these ten or so proconsulships, all governors and all legionary commanders were appointed by the emperor to serve for as long as he required. The same in effect applied to most of the new

^я Talbert 1984 (d 77) ch. 10 sect. 3 and App. 8.

Table i New senatorial posts within Rome and Italy

Title

Function

Date

Remarks

Number and rank

PRAEFECTUS

FRUMENTI DANDI²⁹ CURATOR VIARUM»

PRAEFECTUS AERARII Management of state treasury SATURNI²⁸

Distribution of corn dole at Rome

Management of roads in Italy (though precise scope of functions remains obscure)

CURATOR AQUARUM³¹ Management of aqueducts of Rome

I praetorii

4 praetorii, notionally chosen

by lot Board of uncertain composition

3 notionally chosen by lot, comprising i consular is, i praetorius, i senator of lesser rank

29 to 23 в.с. and from a.d. 56

2 from 22 b.c., 2 more

added in 18 B.C. From 20 B.C.

From 11 в.с.

Function carried out between 2 3 в.с. and a.d. 56 by praetors and quaestors in office

Regular assignment 6f one or more named main roads to an individual senatorial curator almost certainly postdates the Julio-Claudian period Board was granted legal authority to maintain the responsibility exercised informally by Agrippa for just over twenty years prior to his death in 12 B.C.

PRAEFECTUS AERARII Management of military treasury 3 praetorii MILITAR1S³²

From a.d. 6

Archaic office permanently re- instituted in new form from a.d. 13

PRAEFECTUS URBI³³ Oversight of law and order in 1 senior consularis Rome, and command of the three urban cohorts (formerly under the direct control of Augustus)

senatorial posts within Rome and Italy established on a permanent basis by Augustus and Tiberius, albeit with the Senate's approval. Although these posts (set out in Table i) without doubt represent a haphazard growth, rather than a planned series, none the less all were equally intended to improve public services and thereby strengthen the em­peror's own position. At the same time the creation of one or more posts with a particular responsibility did not deter him from still taking personal initiatives in the same sphere from time to time.

Not only did adjustment and experiment continue, as Table i shows. Senators might also be called upon at any time to assist in tackling some short-term crisis or difficulty. But all the same it can be seen that the substantial group of new senatorial administrative posts within Rome and Italy was largely organized by early in Tiberius' reign. To some extent the same may be true of the new posts throughout the empire to which equites were appointed, although the ancient sources' lack of interest in tracing the development of the equestrian service usually makes it impossible to claim with confidence when a particular post was instituted.³⁹

Already during the late Republic certain officerships in the army were normally held by equites (a small number of whom would advance to pursue senatorial careers). Augustus increased the opportunities for military service of this type, so that in time there developed the pattern whereby most legionary tribunates and all auxiliary prefectures were reserved for equites-, some prefects of fleets were also equestrian (the others being freedmen). A limited proportion of all these officers were ex-centurions who had gained equestrian status through working their way up to the primipilate; but the majority were equites by birth, newly recruited into the army and likely to serve there for some years. It seems to have been understood that such military service would be required of any eques who aspired to a civil appointment in the emperor's service.

Like any republican magnate Augustus needed procurators to manage estates which he could not see to himself and to represent him in the courts. He generally asked equites to fulfil this function, and from the beginning of his reign he must have had such representatives in most, if not all, provinces. In senatorial provinces (where a quaestor was stationed) the procurator's function was technically confined to the administration of the emperor's private property. Even during the reign of Augustus, however, procurators in imperial provinces took on a wider role, handling public money and commanding troops; some were actually put in charge of a region or even an endre province, answerable either to the nearest army commander, or to the emperor direct. Most

³⁹ Hirschfeld 1912 (d 13); Stein 1927 (d66); Pflaum 1950(0 56); 1960-1; 1982 (d 59); 1974 (d 58). Many of Pflaum's dates for the creation of new posts should be viewed with caution.

notable among the latter was the prefect of Egypt, who was regarded as the senior equestrian official during the Julio-Claudian period, and whose immediate subordinates (even the commanders of the two legions stationed outside Alexandria) were all equites.[499] Some enlargement of the procurator's role inevitably developed in senatorial provinces too, and it must have been as a reflection of this general expansion that Claudius gave all his procurators jurisdiction in fiscal cases.[500] Indeed by his day there was even one eques who believed that in occupying posts normally given to members of his class he could achieve the same degree of wealth and influence as a consularis."'[501]

In Rome Augustus handed direct command of the praetorian cohorts to a pair of equestrian prefects from 2 в.с. A few years later crises in two spheres prompted him to tackle their persistent problems much more decisively than hitherto. First, after a serious fire in a.d. 6 he took the step of appointing an equestrian praefectus vigilum who commanded a force of 7,000 freedmen to combat fires. Though ostensibly experimen­tal,[502] this innovation soon became a permanent feature. Second, a severe shortage in the same year led him to appoint a pair of consulares to supervise the corn supply in two successive years; then at some date between a.d. 7 and his death in 14 he put the task in the hands of an equestrian praefectus annonae, whose office was permanent.⁴⁴ There is reason to believe that an equestrian prefecture of vehicles in Italy may also date from Augustus' reign,⁴⁵ while it was certainly from early in the Julio-Claudian period that equestrian assistants (adiutores) of various grades came to be attached to many of the senatorial and equestrian administrative officers mentioned above.

It should be stressed that the growth of all these equestrian posts was as much an unco-ordinated response to immediate problems as in the case of the senatorial appointments already outlined. There was no equestrian 'civil service' whose members were guaranteed permanent employment within a planned career structure which encouraged them to develop a particular expertise.⁴⁶ Augustus' general reasons for turning to the equestrian order for the assistance which he sought seem easy enough to conjecture. On the negative side, it might not have been diplomatic to appoint senators to some of the posts concerned, even had there been sufficient members of their class; there may have also been instances where senators' competence was doubted. On the positive side, while equites ranked below senators (and could thus accept orders more readily), they had always been inextricably linked with them; a favoured few were even numbered among the emperor's closest advisers. In addition as a group equites, like senators, were wealthy, educated, and conservative in outlook. Many had experience of public life as jurors, contractors and municipal magistrates, as well as through army service. In general they were an obvious recourse for the emperor in his search for administrative assistance.

All the same it is less easy to be sure why he specifically chose members of the equestrian class to occupy particular posts. In Rome for example, the prefectures of the fire brigade and of the corn supply (both spheres formerly of general concern to senatorial magistrates) could seemingly just as well have been senatorial appointments. Among provinces it is impossible to find convincing general characteristics which distinguish the diverse areas entrusted to equites from those continuing to be governed by senators. Even in the case of Egypt the claims of later ancient writers,⁴⁷ that the country was too turbulent and altogether represented too great a security risk to be safely assigned to a senator, hardly ring true, all the more so in view of the alarm created by the first prefect, the eques Cornelius Gallus. As to the choice of equites to fill procuratorships, the modern contention that the background of the class enabled its members to draw upon unique expertise in the areas of finance, trade and manufacture may seem an unsatisfactory oversimplifi­cation, which overlooks the fact that most equites were no more than owners of large estates, and that the type of expertise attributed to the class is not hereditary. Any assumptions that equestrian officials would generally prove more honest in their conduct than senators, as well as displaying greater loyalty to the emperor, are equally misplaced. It is worth recalling in this connexion the point made above that equestrian adiutores came to be attached to both senatorial and equestrian admini­strative officers. While their appointment may have been intended in part to provide a check on malpractices, it is equally likely that the burden of work carried by their superiors did genuinely call for some assistance.

It may be more satisfactory to admit that Augustus' motives for choosing to employ equites in the way that he did can no longer be identified with any certainty for the most part. At the least, however, his concern must have been to ensure that each individual responsibility was tackled in the most effective manner at the dme, rather than that assignment of posts to members of different classes should conform to some general system or theory. Later, Augustus' successors in all likelihood just continued to appoint to most posts men of the same class as the redring holders, partly out of respect for established practice, and partly because no pressing cause to overturn existing arrangements was apparent. Exceptionally, towards the end of the Julio-Claudian period

⁴⁷ Tac. Ann. ii. 5 9; Hist. 1.11; Dio Li.7.1.

pressure did develop on a number of grounds for equites to be appointed to senior posidons in the emperor's secretariat, which hitherto had normally been given to freedmen. Although the shift itself only occurred later, it does at least serve to highlight in conclusion the extent to which the ambition of equites had grown within the relatively short span since their first employment by Augustus. It confirms, too, their willingness to serve the emperor and their full appreciation by this date of his boundless prerogative as patron and ruler.

PROVINCIAL ADMINISTRATION AND TAXATION

ALAN K. BOWMAN

I. ROME, THE EMPEROR AND THE PROVINCES

The reorganization of provincial government which began with Au­gustus' so-called first settlement in January 27 B.C. gave to the imperial administration in the provinces a fundamental structure which it was to retain for more than three centuries. Its basis can only be fully appreciated in the light of the developments of the late republican period.¹ In the East the Roman organization of Greece and Asia had taken advantage of the urban legacy of hellenization and set the pattern of which the far-reaching arrangements of Pompey's eastern settlement were a logical extension. Here, the ubiquitous phenomenon of organiza­tion through the hellenized poleis, based on specific and definable relationships between the city and the ruling power, was to find its clearest expression, whilst the military and fiscal interests of Rome knitted diverse communities into a loose provincial structure. In the West, Spain, Africa and Narbonensis required a longer period of development and acclimatization to Roman rule, accelerating noticeably only in the last three or four decades of the first century в.с. and drawing in their wake the newly acquired regions of Gallia Comata. If East and West differed in pace of'Romanization' and in many a significant detail, the broad objectives did not: the need to encourage or create civilized and self-sufficient communities (whether based on polis or civitas) governed by their indigenous aristocracies; the need to ensure Rome's military security and the protection of her imperial interests in the broad sense, the cost of which would be met (at the least) by the revenue which Rome could draw from the province enjoying her protection; finally, as a natural corollary, the need to support and promote the interests of Romans in the provinces, senators and equites at the top of the social and