In this extremely valuable book E. lucidly sets out his interpretation of imperial policy in the years between Julian’s Persian disaster and Theodosius’ death, years which saw the formal division of the empire between East and West in 364, the founding of a relatively durable dynasty, the establishment of a large Gothic presence in Roman territory and the emergence of a western form of Christianity as the official religion of the empire. After an introduction successive chapters deal with the major divisions of the subject: emperors and dynasties; foreigners and frontiers; the government; Rome; Constantinople; religion and the state. E.’s fundamental contention throughout, for which he makes a largely convincing case, is that in this period, as earlier, emperors and their chief subordinates rarely approached the issues of the day with any preconceived policy. Policy consisted rather of a series of ad hoc reactions to specific events or other stimuli, often of a localised regional nature.
The introduction singles out the division of the empire in 364 as crucial. It created the conditions for a progressively widening divide between East and West (despite the myth that all emperors ruled everywhere), which nobody made any serious effort to prevent. Rome had long since ceased to be the administrative centre or even an imperial residence. Military realities and the foundation of Constantinople both contributed to a gradual eastward power shift, complicated by the retention of the three great praetorian prefectures from the time of Constantius II. The chief problem of external security was an eastern one: the Goths. Any internal challenge to an emperor’s position was bound to be regional in origin, but responses to such challenges did nothing to counter the basic trend towards separatism, as was demonstrated by Theodosius’ reaction to Magnus Maximus and Eugenius. For him partition was clearly acceptable provided both parts remained in the family.
In the selection of emperors the dynastic principle continued paramount. There might, however, be practical problems, as Constantius II found with his appointment of Gallus and Julian as Caesars. (E. is uncharacteristically incautious in accepting as fact Constantius’ supposed deathbed recognition of Julian as Augustus.) After Julian’s death Jovian’s embryonic efforts to create a dynasty were frustrated by fortune and his own speedy demise. Three major crises in as many years led to the demand that Valentinian I choose a colleague. So military necessity produced that division of the empire which, despite Themistius’ efforts to promote it as strengthening unity, in fact made separate development inevitable. E. is inclined to believe that Valentinian refused to help Valens against Procopius, though he admits there is no evidence that Valens asked for aid. The fact that Valens coped proved that the regional division was viable.
The dynasty developed with the advancement of Gratian and Galates, but the deaths of Galates and Valentinian were setbacks, and the elevation of Valentinian II was questionable: Valens and Gratian were not consulted and were slow to acknowledge him as a colleague till the activities of the Goths made recognition inevitable. After Valens’ death five months elapsed before the succession of Theodosius, but he and Gratian were then at pains to advertise their unity. When Maximus rebelled and murdered Gratian Theodosius was at first in no position to fight, though he did not recognise Maximus as consul for 384. His inaction may have encouraged Maximus to hope that Theodosius would again accept a fait accompli when he invaded Italy in 387. Valentinian II escaped to Thessalonica, but Theodosius was still slow to move. His eventual advertised motive was dynastic loyalty to Gratian, but his real reasons remain as obscure as they were at the time. After Maximus’ death Theodosius may have considered taking over the West while leaving Valentinian II in Gaul with Arbogast as his minder. Valentinian’s suicide changed things. Arbogast realised that Theodosius would not leave him in sole control or put him in charge of Honorius, so he set up Eugenius. Theodosius promptly made Honorius Augustus, but Honorius did not enter Italy until after the Frigidus. Theodosius’ early death renders futile speculation as to whether he intended a three-way division of the empire between himself and his sons. E. rightly insists that the whole story consistently displays the importance of the dynastic principle. One might add that it also reveals the extreme vulnerability of that principle to disloyalty and sudden death.
Foreign policy meant frontier policy. The tradition was one of Roman response to external challenge, with Julian’s Persian expedition a disastrous exception (but even that was bizarrely presented as pre-emptive defence). The point of a frontier was ‘to control and tax passage’. This is true, but one might add that the Romans’ conception of a frontier was not egalitarian; they saw frontiers as more binding on others than on themselves. Julian’s neglect of the West made it a priority after his death. E. is sound on Valentinian I’s objectives in Gaul and on the Rhine. The most important disturbance, perpetrated by the Lentienses, caused Gratian (or perhaps gave him an excuse?) to be too late to help Valens at Edirne. E. suggests that Ammianus may exaggerate the elder Theodosius’ achievements in Britain; in fact Ammianus’ panegyric is subtly ironic. The situation on the Danube was roughly similar to that on the Rhine. By 374 it demanded Valentinian’s presence, but after his death nothing of importance happened until the immigration of the Goths. There was nothing unusual about their request for land or Valens’ response; problems arose from the inaccurate estimate of their numbers and the corruption of Roman officials. Theodosius’ treaty of 382 had far-reaching consequences. It gave Rome peace, security and recruits, but it established a permanent bridgehead of unassimilated barbarians in a frontier province.
On the Eastern frontier Jovian’s treaty brought stability for 140 years. E. sets the reasonable judgement of Orosius against the hysteria of Ammianus and others. Problems lay in the control of Armenia and Iberia. E. is right to deny excessive importance to Mavia, whose activities did not demand the involvement of Valens. In Africa there was the rising of Firmus, in connection with which E. again rashly takes at face value Ammianus’ praise of the elder Theodosius.
The three praetorian prefectures had their fortuitous origin in the fact that Constantine had three sons. As long as there was only one emperor unity might be fostered, but after 364 the divisions contributed to the general tendency towards separatism. The Gothic crisis led to the bisection of Illyricum along linguistic grounds, and by Theodosius’ death this arrangement had become permanent, though the date cannot be precisely determined. Most of the chapter on government is devoted to the Codex Theodosianus, which was only the first stage in a process never completed and so of little practical use to lawyers. E.’s treatment of the nature of imperial laws, the queries that prompted their issue, the officials who might receive them, and the varying degrees of publicity they were given is lucid and largely convincing. Despite the fiction that all laws were issued by all emperors, all legislation was in the first place regional and co-ordination of measures between East and West was rare.
Rome was largely ignored by emperors. Since Constantius II there had been the possibility of friction between the praefectus urbi and the vicarius urbis Romae. Under Valentinian I a high level of interference can be detected, notably in the cases of the episcopal election of 366 and the trials for abuse of magic under Maximinus. Gratian was in general more liberal and less proactive, except in fostering Christianity; he notably refused to grant Symmachus an audience in the matter of the altar of Victory. When Symmachus was prefect under Valentinian II, it is clear that all important decisions were taken at court, not in Rome. Theodosius needed the support of the Roman aristocracy. So at first he showed no religious bias, was in 387 the first emperor to visit Rome since Constantius II in 357, and showed clemency towards the supporters of Maximus, including Symmachus. The speech of Pacatus (of which E. offers an eminently sane discussion) recommends Theodosius to the senate and offers conciliation. But the Christian backlash began in 391 and triumphed after the defeat of Eugenius and the suicide of Nicomachus Flavianus.
The emergence of Constantinople was of vital importance. Constantius II meant it to be just another major city, but he gave it a senate. Again 364 was a watershed. Valens largely stayed away, but developed the infrastructure. The Gothic crisis brought the city into the foreground. Theodosius resided there from 380 to 386 and turned it into a capital. When he left, Arcadius and the praetorian prefect remained. Its forum and column constructed for Constantinople an imperial history on the Roman model. But the creation of a senatorial class further weakened the ties between East and West. After Constantius II’s visit to Rome in 357 Themistius received orders to recruit, and numbers soon reached c.300. In 359 the city acquired its own praefectus urbi; new rules for recruitment came in 361. Under Theodosius a much increased tendency to legislate for Constantinople reflects its growing importance: bread rations and senatorial rights were constant problems, the public image of the ruling class a matter of concern. But there is no sign of any effort to promote Christianity. As E. dryly observes, the banning of circuses on Sundays, except on the emperor’s birthday, reveals Theodosius’ priorities.
The most valuable section of the book is that devoted to religion. Agreement on the nature of God contributed to unity and civic order; the church was influential because of its wealth and the tax privileges it enjoyed. It was therefore desirable to know what constituted ‘the church’. Constantius II might have solved the problem had he lived longer, but Julian left chaos behind him. The Codex Theodosianus records no law on paganism from Valentinian I or Valens, and very little on religion. Valens had the greater problems; he had to recreate a systematic approach, which was bound to revive that of Constantius II. But there is little evidence for the oppression of non-homoians, and the orthodox eager to blacken Valens had problems with the favour he showed to Athanasius and Basil. A few orthodox bishops were exiled, but probably not on dogmatic grounds. In discussing these matters and the complex situation at Antioch E. is commendably sceptical of orthodox propaganda. In the West Valentinian I preserved the Nicene status quo, but was tolerant except where magic (Manichaeans) and public order (Donatists and the battle for the bishopric of Rome) were concerned. Gratian issued no general laws on religion and took no initiative in the Priscillianist controversy. But once the court had been forced to move from Trier to Milan in 381 he came under constant pressure from Ambrose to be actively pro-Nicene. In 382 he removed the altar of Victory, stopped subsidies for pagan cults (probably in response to an approach by the praefectus urbi), and refused to grant an audience to Symmachus. Valentinian II and Justina were homoians, which made Maximus attractive to Ambrose as long as Theodosius showed no sign of intervening. So Ambrose went to Trier to negotiate with Maximus, but in 385-6 had a series of confrontations with him before Maximus invaded Italy in 387.
E. rightly maintains that Theodosius had no religious programme, being more concerned with security than Christian unity or exterminating paganism. Yet thanks to Ambrose he came to be hailed as a pillar of orthodoxy. He did create a symbiosis of church and state in the East, but his religious policy, such as it was, was conditioned by external secular factors. Constantinople constituted a problem, since it enjoyed no outstanding ecclesiastical status. At the time of Theodosius’ accession the city was homoian, but Theodosius wanted a unified church, which could only be Nicene. The idea of church unity was not new, but Theodosius was unprecedentedly energetic and authoritarian in its pursuit. E. picks his way lucidly through the complicated struggle for the bishopric of Constantinople, in which Milan, Rome and Antioch all had a lively interest. The city eventually emerged with a dramatic upgrade, now standing second only to Rome. More pro-Nicene legislation followed throughout the East, though the object may have been only to frighten into compliance, since the drastic penalties threatened seem not to have been imposed.
Theodosius spent five years sorting out the Christians while paying little attenton to paganism. But there was a fairly high level of tolerance of Christian vandalism, by monks and others, notably over the destruction of the temple of Sarapis in 391/2. The result was, as E. cogently puts it, ‘a general climate of opinion within which highly placed Christian extremists… could act virtually uncontrolled’. In Italy Theodosius needed the support of Ambrose; indeed he and Ambrose needed each other. The Callinicum incident seems paradigmatic: Theodosius was offended by Ambrose’s behaviour and at first made the right noises, but in the end Ambrose got his way. Ambrose also triumphed in the matter of the Thessalonica riots in 391 (well treated by E.), but his success was shortlived. E. argues convincingly that the later Christian interpretation of the last campaign against Eugenius in 394 as ‘Christians v. pagans’ is mostly fiction. Arbogast and Flavianus were not Christians, but Theodosius appointed them and trusted them until their treason. The unreliable Theodoret was the first to present the war as a crusade. Even Ambrose, praising Theodosius a few weeks after his death, consigned Maximus and Eugenius to hell not on doctrinal grounds but for rebellion against a legitimate ruler. Other religious ornaments to the story of the Frigidus are late; the effect of the bora is first found in Claudian (1/1/396), the pagan statues on Eugenius’ side are an even more belated addition.
The conclusion to this fascinating and important book recapitulates E.’s main points. There were practical reasons for the split in 364, but it made a gradual separation and the evolution of independent administrative traditions inevitable, while the retention of the three great prefectures promoted regional separatism as well as regional centralisation. Policy in all fields continued to be reactive and pragmatic, even in religious affairs, the overriding aim being political and economic stability. But E. admits that a regular pattern of similar reactions could be interpreted as policy, and indeed the pagans must have felt that all dodgy refereeing decisions went in the Christians’ favour. Increased dependence on the regional bureaucracies made the emperors less peripatetic, a development in which the Gothic war was a crucial catalyst. The most ominous feature of the time was Theodosius’ grant of territory to the Goths while allowing them to reject social and religious integration, which paved the way for the ultimate collapse of the empire.