Succession and security remained the principal factors governing imperial policy under Diocletian and his successors. Once Diocletian determined that he could only survive by establishing regimes loyal to himself in different parts of the empire, the empire for most intents and purposes ceased to be a single entity. There would be a sole emperor for only 13 years between 285, when Diocletian appointed Maximian, a man related to him by neither blood nor marriage, as his deputy emperor, or Caesar, and the death of Constantine in 337: this was the sole reign of Constantine after 324. After Constantine, there were brief periods when there were not multiple holders of the title Augustus: the sole reigns of Constantius II from 353 to 360 (ending when Julian, his nephew and Caesar, rebelled), the reigns of Julian and Jovian from 361 to 364, and the last months of Theodosius in 395. Even this is deceptive, for Constantius appointed Julian as his Caesar in 354 precisely because he felt that he needed a family member to carry the banner of his regime in Gaul while he was busy elsewhere; and Constantine had elevated his sons to the rank of Caesar well before 324, giving them ever more responsibility as time passed. Indeed in the years immediately prior to his death, Constantine appears to have envisioned a sort of return to the Diocletianic system, with no fewer than five Augusti ruling concurrently once he died. In 334 he elevated his nephew Dalmatius to the rank of Caesar with responsibility for the Balkans, while of his sons, Constantine II administered Western Europe, Constans took Italy, some portion of the Balkans and Africa, and Constantius II held the east. Another nephew, Hannibalianus, was designated King of Armenia with the view that he would rule whatever portion of the Persian Empire Constantine could conquer in the war that he planned in the last years of his life (T. D. Barnes 1985: 132).
What is perhaps most interesting about Constantine’s plan is that it was biologically unnecessary. With three sons, he could simply have divided the empire three ways. He seems, however, to have felt that it would be better not to do this, and the best explanation is that he felt that a tripartite division of power would not be successful (in this view he was proved correct).
What differentiated Constantine’s plan from that of Diocletian is that Constantine decided that blood relationship was a prerequisite for membership in the imperial college; Diocletian, at least initially, had decided that proven ability should govern membership in the college, though it would appear that he changed his mind on this point on more than one occasion. His system of government was very much the work of trial and error (for details, see Elton, this volume). There is no reason to think that when he appointed Maximian Caesar, he envisioned that he would soon make him Augustus. But the situation in the west was complicated - there would be a revolt in Gaul soon after he left Maximian in Italy, and it is likely that he felt that Maximian needed the authority conferred by the title Augustus to deal with his subordinates. When he raised Maximian to the rank of Augustus he also took the title Jovius for himself, giving Maximian the title Herculius (Nixon and Rodgers 1994: 44-5). The divine names were selected to make clear the ranking of the two men: Diocletian was senior, Maximian was junior.
In the early 290s it may have become clear that two emperors were insufficient. In 286 a general on the Rhine, Carausius, had rebelled against Maximian, and Maximian was having grave difficulties bringing the revolt to an end (Casey 1994: 42-3). At the same time Diocletian appears to have been devoting himself to the Danubian frontier. He journeyed only once to the east, and he may have sensed that here, too, the imperial authority was under-represented. In late 290 or early 291, after what appears to have been a catastrophic failure on the part of Maximian in the war with Carausius, Diocletian met with his colleague at Milan (T. D. Barnes 1982: 52). It may have been then that they decided that they would appoint two Caesars, when circumstances made it possible. In 293, Maximian scored a major triumph over Carausius, driving him from the mainland. The man responsible for this victory was his son-in-law, Constantius. On March 1, 293 Constantius was made Caesar; either on the same day, or in May, Diocletian proclaimed Maximianus Galerius, who married his daughter, Valeria, Caesar as well (T. D. Barnes 1982: 62 n. 73; Kolb 1987: 73).
There were many idioms through which the new college of emperors could be presented to their subjects. On the remodeled coinage that was introduced as they were in 293, they tend to look rather alike, men with beards and close-cropped hair. On the famous porphyry statue group built into the Cathedral of St. Mark at Venice, the two Caesars are distinguishable from the two Augusti because they have less full beards, suggestive of their ‘‘youth’’ (neither was a young man). Most spectacularly, we now have a group of mosaics from the palace that Galerius constructed as his retirement home at Romuliana in what is now Serbia. Here is perhaps the most nuanced view, as befits one who had a stake in the system. Each of the tetrarchs has a distinctive appearance (Srejovic 1994: 143-52).
The wisdom of Diocletian’s action was revealed a few years later. A poet who most likely wrote in Egypt at the end of the decade offers a very interesting picture of the tetrarchic regime. The event that inspired his poem was Galerius’ victory over the Persian king, Narses, in 297. In describing the outbreak of the war he says:
Other kings would have rushed to his aid from Italy, if Iberian Ares had not restrained one, and the din of battle on the island of Britain had not flared up around another. Just as one god comes from Crete, another from sea-girt Delos - they are Zeus over Mount Othrys and Apollo to Pangaion - the throng of Giants trembles as they put on their armor: so did the elder king, bringing an army of Ausonians come east together with the younger king. (Sel. Pap. 135)
Aside from the war with Persia the events mentioned here are Constantins’ victorious campaign in Britain, which ended the revolt of Carausius (who was murdered by his own men just before the end and replaced by a man named Allectus), and a revolt in Spain, for which this text is our sole evidence. Diocletian in fact played no direct role in the Persian war, leaving the command to Galerius, who suffered a serious defeat near Carrhae in 296 before advancing through Armenia in the following year to win a decisive victory. Even as Galerius advanced, Diocletian was called away to Egypt to suppress a revolt. The series of victories won by the four members of the college finally solidified the control of the empire, and, for a time at least, of the eastern frontier. According to the terms of the treaty imposed upon the Persians, five provinces were added to the empire in the east that enabled Rome to control potential invasion routes through Northern Mesopotamia and Armenia (Winter 1988: 171-82). In Egypt, Diocletian followed up his victory with a campaign into the Sudan against the Blemmyes. Constantius returned from Britain to Trier, which, in effect, became his capital. Maximian moved on from suppressing the revolt in Spain to campaign against nomadic tribes in North Africa from 297 to 298. After a brief stop at Rome in 299, it appears that he intended to make Milan his capital. Diocletian’s principal palace would be at Nicomedia in Bithynia, and Galerius would establish himself at Sirmium, close to the Danube frontier.
In the wake of the great successes won by himself and his colleagues, Diocletian would increasingly turn his attention to the reorganization of the empire. His policies included re-tariffing the new coinage that he had introduced in 293, an effort to regulate prices throughout the empire, and a persecution of the Christians. These followed a series of earlier reforms: in addition to the aforementioned reform of the coinage, there was a regularization of the tax system of the empire, probably in 296 (and quite possibly the cause of the revolts that broke out at that time), and an effort to codify Roman law (Corcoran 2000: 25-42). Diocletian also appears to have supported a scheme for the continuation of the collegial system that would involve his retirement, along, ultimately, with that of Maximian, the promotion of the Caesars to the rank of Augustus, and their replacement by two new Caesars: Con-stantius’ eldest son, Constantine, who had served on Diocletian’s staff in the 290s, and Maximian’s son, Maxentius, who married Galerius’ daughter.
Shortly after 300 something seems to have changed. The image of the tetrarchy that Galerius portrayed in his palace had perhaps some unfortunate consequences for himself. If the Augusti retired in the order in which they had entered power, he would be left as Caesar, while Constantius would become the colleague of Maximian. It does not appear to have been until 303 that Diocletian decided that this could not happen, for in that year he went to Rome for perhaps only the second time in his life, to celebrate the twentieth anniversary of his joint accession with Maximian: the count of Maximian’s regnal years was suddenly changed to make it equal with that of Diocletian’s (Chastagnol 1967). Maximian, willingly or no, swore an oath to Diocletian that he would step down at the exact same time as his colleague. Diocletian’s decision that it was time to retire may have been inspired by a serious illness that began while he was in Rome. It may also have been at this same time that he decided that Constantine and Maxentius would not succeed to the position of Caesar. Sources hostile to Galerius (none have survived that would tell his side of the story) assert that he forced Diocletian to make these decisions. We will never know the truth; the one thing that does seem true is that, although Diocletian had long since decided upon abdication, and had had a retirement palace built for himself at Split (in modern Croatia), he put off a final decision as to whether to endorse the notion of biological succession to the very end. His decision against it would ruin the system of government that he tried to introduce. The two new Caesars, Severus and Maximinus Daia, were seen as too closely aligned with Galerius.
The retirement of Diocletian and Maximian on May 1, 305 precipitated a crisis. Constantius demanded that Galerius send him Constantine, who was serving under Galerius in the Balkans, while Maxentius took up residence outside of Rome (we have no way of knowing when he did this). Constantius may have sensed that his own health was failing, and he was determined that his own son would succeed him. In the summer of 306, if not before, Constantine joined his father for a campaign in Britain. When Constantius died on July 25 at York, the stage was set for the proclamation of Constantine as Caesar. Galerius, who may well have expected the move, acquiesced (Potter 2004: 346). What he could not tolerate was the revolt at Rome in October, by which Maxentius, with the aid of his father Maximian, proclaimed Augustus once again, claimed a place in the imperial college for himself as a princeps.
Maximian remained a powerful figure, and his own discontent with the settlement imposed by Diocletian is manifest in his subsequent conduct. He appears to have been popular with the army that he had led for many years, so that when Severus attempted to lead it against Rome, it deserted. Severus was taken captive and placed under house arrest. Maximian then traveled to Gaul in the company of his very young daughter, Fausta, whom he would marry to Constantine, whose wife, Minervina, had died after bearing him a son (T. D. Barnes 1982: 42-3). Maximian stood as a champion of biological succession against the collegial model promoted by Galerius and Diocletian. What is perhaps most interesting about this is that, after realizing that he lacked the military force to overthrow Maximian and Maxentius in 307, Galerius was willing to negotiate. Indeed, throughout the rest of his reign, Galerius appears to have tried his best to keep the peace. Although he would not recognize Maxentius, and could not after Maxentius murdered Severus and styled himself Augustus, he was willing to negotiate with both Constantine and Maximian. In 308, after Maxentius quarreled with Maximian and forced him to flee Italy, Galerius hosted a meeting, which included Diocletian, who had assumed the consulship, at Carnuntum. Max-imian agreed to retire once more, going to live with Constantine in Gaul, while Galerius elevated Licinius, said to be a close friend, to the rank of Augustus to replace Severus. Constantine and Maximin Daia remained Caesars. Maxentius was not accorded recognition in the imperial college.
The settlement of Carnuntum assured the peace of the empire until Galerius’ death in 311. As Maximian had either been killed or committed suicide after a failed revolt against Constantine in 310, the old generation had now passed away. The new generation was not interested in peace; each of the emperors now wished to rule at least half of the empire himself. The man with the most obvious difficulty in reaching this goal was Licinius, who found himself between two hostile powers, Maximin and Maxentius. He sought an alliance with Constantine, who could scarcely refuse, since the failure of Licinius would result in his having to face an overwhelming force in the aftermath. In the end it would be Constantine who struck first.
Constantine’s campaign in 312 is a moment of world historical significance. Constantine would have been only too aware that both Severus and Galerius had failed dismally in their attempts on Rome. In each case failure ensued when the campaign bogged down in siege operations or their prospect - Severus’ army had deserted on the way to Rome, and Galerius had not even attempted to besiege the city. Constantine, who seems to have had a strong sense that a god was helping him, even if he was not sure which one, appears in the period leading up to the campaign to have decided that the Christian God would be the one who could help him. No rational political calculation could have influenced Constantine. He had previously claimed a special relationship with Apollo in his guise as a sun god, and he would continue to advertise that connection long after he was identifying himself as Christian to Christians. Indeed, it seems to have taken him some time to move from a henotheistic understanding of the Christian God as the ‘‘highest God,’’ to the monotheistic understanding that is enshrined in the best-known religious document of the period, the Nicene Creed of 325. We will probably never know why he decided that the Christian God, whom he plainly did not understand very well at this point in his life, would be his protector. The fact, however, that he did take this position would result in the transformation of the Christian Church from a fringe group within the empire to a religion that would shape human understanding, for good or ill, in the centuries to come.
Once he had decided upon his protector, Constantine moved his army with exemplary speed, drawing a series of armies that had been sent by Maxentius to northern Italy into open battle. He was able to destroy them, and captured cities that resisted by storm. He reached Rome by October, and it is quite possible that his success had so demoralized the opposition that Maxentius felt that he had to risk an open battle, even under deeply unfavorable conditions (with the river Tiber at his back). The destruction of Maxentius’ last army, and the death of Maxentius himself in battle, made Constantine ruler of half the empire. In February, 313, he met Licinius at Milan; the alliance between the two was there strengthened by Licinius’ marriage to Constantine’s half-sister, Constantia.
In the spring of 313, Maximinus invaded the territory of Licinius. The result was spectacular military failure: Maximinus committed suicide at Tarsus in July. Peace between Constantine and Licinius did not, however, last long. They were at war in 316-17, with Constantine winning a limited strategic victory to back up further spectacular successes on the battlefield by the spring. Although Licinius retained most of his portion of the empire, Constantine gained a solid foothold in the Balkans as a result of the peace treaty. In 324, he exploited his advanced positions by launching another invasion of Licinius' territory. Heavily defeated, Licinius retreated to Nicomedia, where he surrendered to Constantine in the autumn. Licinius was allowed to retire to Thessalonica, where Constantine had him murdered on what may have been bogus charges of conspiracy less than a year later.