Ancient History & Civilisation

Chapter Ten

The Consequences of Gordian (1): Successful Emperors

One of the results of the events of 238 was to emphasize once again the possibility that a man might, by rebellion, make himself emperor. The events of 68–69 and 193 had already done so, and more recently in Syria Elagabalus had successfully seized power by rebellion.

A distinction must be made between the rebellions of Elagabalus and Gordian against the actions of other emperors in killing or acting against an emperor and taking his place, as Maximinus and Macrinus had done most recently by coup d’état. A rebellion was mounted away from the court and required an army to march on the capital. The most successful recent example was Septimius, but Gordian clearly intended to do the same; ironically, one result of his rebellion was that the Emperor Maximinus ended up in the posture of a rebel marching on the city.

The events of 238 were so widespread, involved so many groups and were so widely broadcast that the possibilities became much more appreciated. This was clearly one of the reasons why the next half-century saw a great number of pretenders and rebels to add to the confusion generated by the rapid turnover of 238, the ‘year of seven emperors’. Yet this was a time when the Empire went through an agonizing and extended period of foreign invasion and monetary inflation, and the rebels and pretenders were all part of that overall problem; at once cause, consequence and solution. The confusion was such that it is not too much to say that the Empire had collapsed.

The rebels are often pushed aside or out of consideration, so that the main story of the ‘legitimate’ emperors is usually that which is told, and it must be said that story is certainly complex enough even without the rebels. Here, however, rebellion is central to the concerns of this study, and so we shall consider the rebels as a separate group in the next chapter. Previously (in Chapters 6 and 8) they have been included in the main account, but they are so numerous in the mid-third century that they deserve a more detailed discussion. This chapter, therefore, will look at the ‘legitimate’ emperors – those generally supposed to be so, that is – and their modes of succession.

From 238 to 284 a superficial consideration of Roman politics might suggest that it was the Roman army that determined who should occupy the imperial throne. The emperors were certainly usually proclaimed by their soldiers, who had often murdered their imperial predecessors, hence the term ‘military emperors’ or ‘military monarchy’ which is sometimes used for this period, but such a description is not necessarily accurate. After all, the basis of Augustus’ power, and hence of every later emperor, was control of the army.

Also, the revival of senatorial influence during 238 cannot be ignored. After all, influence was the only power ever wielded by the Senate once the imperial system had been installed. Only in the case of Nerva before the Gordians had the Senate’s choice of emperor been free of outside pressure, and Nerva and the Gordians had scarcely been the most effective emperors. In all other cases the Senate had, sometimes after debate, accepted a candidate put forward from another authority: from the former emperor, the current emperor, the army, even on one recent occasion from the Guard. That is, the role of the Senate was to ratify a candidate nominated by one or more of these authorities, and at times (as with the crisis following the death of Gaius) the senators had to choose between several possible candidates. The basic reason, of course, was that the Senate itself controlled no armed force; its influence came from its collective experience, and this was not enough in some cases. Yet it is clear that in 238 it was able to use that influence and experience with success.

The reassertion of senatorial authority in 238 remained effective for some decades. Gordian III died in the war in the East in 244, killed by some means now quite unascertainable but probably in battle or by accident during a battle. One of his Praetorian Prefects, M. Iulius Philippus, recently appointed, was, after some delay and confusion, made emperor by the army on the urging of his brother, Priscus, the other prefect. This army was a major expeditionary force, so Philip’s elevation had powerful authority. Philip rapidly made peace with the Persians and removed his army to safety and himself to Rome, just as Hadrian had done in 117, and as Macrinus had failed to do in 217 or Maximinus in 235. The Senate’s endorsement of the army’s nomination was clearly of the first importance to him, even more than achieving victory in the war in the East. He had also, presumably, learned the lesson of Maximinus’ fate: that the Senate’s clear and willing acceptance of the new emperor was necessary.1

Philip succeeded in his endeavour, and soon his son was made Caesar as well. It may be that it was the fact that he had a son, about 7 or 8 years of age, and so was able to found a dynasty, that was one of the reasons he was promoted to the post by his brother. He reappointed Priscus as Praetorian Prefect and then as Corrector of the Eastern Provinces, a sort of super-governor. His brother-in-law Severianus was appointed to a similar position in the Balkans for a time; these were the two really troublesome areas, militarily speaking, at the time. These measures may be seen as revivals of similar emergency measures going back to Augustus’ use of Agrippa in various areas, or Avidius Cassius by Marcus Aurelius in the East. (For the family, see Genealogical Table XIII.)


A revolt broke out in the Balkans and Philip at once consulted the Senate. The rebel was Pacatianus, who was the successor of Severianus in overall command in the Balkans. Philip sent a senator, P. Messius Decius, who came from that region, to deal with it. Decius had demurred at first, arguing that the rebellion there would soon end and, sure enough, he achieved its suppression with little difficulty. He was a man of some experience, a former governor of Moesia Inferior and of Tarraconensis, and he had been City Prefect at Rome when Philip appointed him to suppress the revolt. However, it was his behaviour in the crisis of 238 that seems to have been the key to his general policy and attitude: as governor in Tarraconensis he was loyal to Maximinus until that emperor’s death. He presumably then made his peace with those who succeeded him, namely the regime of Gordian III, though Spain had remained resistant to Gordian’s authority for some time.2

Having won a battle with invaders who sought to exploit Pacatianus’ revolt, Decius then organized his own proclamation as emperor as soon as he had finished his campaign. The story as it is usually told leads directly from Decius’ suppression of the rebellion to his proclamation by his soldiers, and so to the battle in which he defeated and killed the Emperor Philip, but the events were more spread out than that version implies. The original rebellion by Pacatianus began late in 248, and was suppressed, judging by the coin evidence, by April of the next year, 249. The Battle of Verona, in which Philip died, did not take place until September 249. In other words, Decius spent several months in his province before he was made emperor, and several more months passed before the decisive battle. It is said that Decius corresponded with Philip over the problem they faced. It seems that Philip had been depressed by the rebellions against him (there was another in the East at the same time) before sending Decius north. Perhaps Decius believed that Philip would now resign. He did not, and Decius had to invade Italy. They met in battle at Verona and Philip, fighting apparently in the front rank of the army, was killed. It may well be that he was still depressed, even suicidal.

Decius, however, though a senator, was of the persuasion of Maximinus in regard to governing the Empire; that is, he insisted on the primacy of the needs of the army over those of the taxpayers. He was supported by his own soldiers, and by those who had supported Pacatianus. This second group of men were afraid that they would be punished for their rebellion, and the two groups had joined together to make Decius emperor. He was clearly not unwilling, and the length of time that passed between his victory and his proclamation suggests that he was using the time in a search for other supporters, as well as, perhaps, negotiating with Philip. One source roundly says he plotted his way to the throne. One measure he appears to have taken was to contact the Guard in Rome, where Philip’s son, now Augustus, had been left when the emperor marched north. As soon as news arrived that Philip was dead, the boy was killed, having apparently been kept in the Guard’s camp. This was surely prearranged, with the murderers hedging their bets by agreeing with Decius in advance that they would act in his support only if the elder Philip died in the fighting.3

Decius returned to Rome, where he spent some time. He was invested by the Senate in the normal way, but his reign was less than successful. He claimed a military ability that he did not in the end display. He promoted his son Herennius Etruscus as Caesar, which was once again an attempt to found a ruling dynasty. It was perhaps more likely to succeed than Philip’s, whose son had been killed while he was still a child. Decius himself was in his 60s and Herennius was an adult in his 30s. There was also a second son, Hostilianus, who was much younger than his brother and who remained in Rome when the Augustus and the Caesar went off to war in the Balkans once more. (For the family, see Genealogical Table XIV.)


Despite his time in Rome, nothing Decius did indicated any affection for the Senate. He clearly despised Philip, and his speech as recorded on the occasion of the debate about Pacatianus is a thinly-veiled attack on the emperor’s indecision, which may actually be an attack on the process of consulting the Senate. He had supported Maximinus against the first two Gordians and against the senatorial emperors in 238. While he was on campaign in the north, and despite the presence in Rome of Hostilianus and the boy’s guards, a senator, D. Iulius Valens Licinianus, staged a coup with the support of the plebs in the city. He may also have had senatorial support, in a mild way, but he was quickly suppressed, presumably by the Guard. This event was a clear sign of Decius’ unpopularity, however, and it has all the appearance of an attempted coup on the pattern of those against Maximinus.4

The war in the Balkans resulted in the deaths of both Decius and Herennius Etruscus. They were succeeded by an army colleague, C. Vibius Trebonianus Gallus, governor in Moesia, who the army chose as emperor as soon as Decius was dead. Like Philip, he headed straight for Rome, having made a rapid and generous peace with the invading Goths first; the same sequence of events that occurred when Philip succeeded Gordian III. Here was another indication of the new potency of the senatorial investiture process, in that it was considered necessary for a new emperor to report to the Senate first. Gallus adopted Decius’ son Hostilianus, and made both him and his own son Volusianus Caesars. He even put Hostilianus first, before Volusianus, for he had been Caesar under his father already. (Gallus was from Etruria; Decius’ wife was also from Etruria; it may be that she and Gallus were related, hence the adoption, which is otherwise difficult to explain.)5

The whole dynastic scheme quickly collapsed. Hostilianus died, apparently of disease – it was a plague period in the Empire – and rebellions broke out, both in the East and in the Balkans. Gallus commissioned P. Licinius Valerianus, governor in Germany, to cope with the rebellion in the Balkans, which was led by a man called Aemilius Aemilianus.

In Italy the soldiers were now operating a system of improvisational election, with a fearsome penalty for the defeated candidates. Gallus and his son Volusian led an army to face Aemilian, while Valerian brought a third army south from Germany. Instead of fighting, the several armies – or rather their commanders – consulted. Out of it Valerian emerged in power, while each of the other armies killed its own emperor.6 There was, apparently, no actual fighting. The army was acting as a political constituency, nominating an emperor, and its choice was ratified by the Senate. Gallus had been Decius’ heir, and Decius had been a Maximinist; that is, Decius, Gallus and Volusian were relying wholly on the army as their basis of power with no regard for the Senate, disdain for which Decius had earlier made clear. The same was the case with Aemilian, who is said to have written to Gallus offering to share power with him; in other words, he also had no regard for the Senate’s preferences and prerogatives.

Valerian, on the other hand, was a senatorial candidate. He had been in Africa in 238 and had been one of Gordian’s messengers to Rome, probably of consular rank even then. He was about 60 in 253, with an adult son, Gallienus, and the year before he had been in Rome and associated with Hostilianus, perhaps as his guardian, while Decius was on campaign. He therefore straddled both parties: the militarist (Maximinus, Decius and Gallus) and the senatorials (the Gordians and Philip). His age commended him to the Senate, which always preferred elderly candidates, and his adult son reassured those seeking the stability of dynastic continuity, while his military experience and ability recommended him to the soldiers, who had demonstrated their preferences in the most drastic style. (Valerian’s wife was Egnatia Mariniana, and seems to have been a relative of the frightened ‘competitor’ of 193, Triarius Maternus.)

Valerian’s position in the army/Senate conflict permitted him to be invested by the Senate and still spend much of his reign away from the city, while his son Gallienus operated there. Yet it was in his time that the role of the Senate in public affairs was potentially reduced to a largely ceremonial and deliberative one by legislation. The repeated rebellions and invasions had emphasized the necessity for a reform of the army, and this was undertaken by the joint emperors Valerian and Gallienus. One result was to remove senators from military commands; only emperors who were well entrenched in the goodwill of the Senate could have done this, for an army man might advocate the reduction of senatorial involvement in affairs, but would inevitably be accused of partiality. The new system left to the Senate the consulship and the government of several of the provinces, but put into the hands of the professional soldiers the control of frontier provinces and the command of the legions and armies.7

This was done principally by the work of Gallienus, who remained in command in Italy and the West while his father campaigned. Senators were thus to be deprived of the experience their forebears had gained while commanding legions and governing provinces. It was not a change done suddenly, for in many ways it was only a recognition of the preferences the senators had shown by their earlier unwillingness to take up these more arduous posts, and senators are still found later at times in command of armies and military provinces, but their exclusive role had gone. The Senate did retain its great prestige for some time, of course. Senators were always wealthy – it was a condition of membership – and the Senate was a legitimate and legal body, but inevitably the promotion of non-senators to the most important military and administrative positions gradually degraded its importance; by refraining from those forfeited positions, senators ceased to be listened to. It was still possible for senators to be employed as commanders and governors, but this was now at the discretion and choice of the emperors, not by right or by election or appointment by the Senate. The result was a divorce of senatorial competence from the reality of the Empire. For another generation there were in the Senate men who had governing and command experience, but as they died off, the Senate as a whole ceased to have a direct role in governing the Empire, and this applied as much to its role in installing emperors as it did to its function as a legislative and consultative body.


Valerian went east to deal with another Persian attack, and Gallienus placed his own son Saloninus as Caesar in nominal command of the Rhine frontier at Cologne. Once more a dynasty was in the process of formation, and Gallienus had a second son, another Valerian, available as well. Dynastically, however, the family failed: Valerian the elder was captured by the Persians; Saloninus and the younger Valerian were murdered; and eventually Gallienus, apparently for failing in all his policies, was murdered by a cabal of his generals.8 (For the family, see Genealogical Table XV.)

In the meantime the capture of the emperor precipitated a collapse of the Empire. The westernmost provinces were detached from Gallienus’ authority by the governor of Germania Inferior, M. Cassianus Latinius Postumus, who killed the boy Emperor Saloninus and made himself emperor in his stead. His programme was the defence of Gaul, which Gallienus had done much to improve already, but which he had then left in the hands of his adolescent son and his adviser. Postumus gained control of both Germanies, and of their frontier armies, of the Gallic provinces including Narbonensis, and the Spanish and British provinces as well, which also held armies. He took the full panoply of imperial titles – imperator, Caesar, Augustus – but made no serious attempt to take over the rest of the Empire. From the point of view of this study, it is especially interesting that Postumus also organized a Senate, which sat presumably in one of the cities close to the German frontier, and held elections to the usual Roman magistracies. When Postumus was killed by his troops at the instigation of a usurper in 269 – this was indeed a true version of the Roman Empire – he was succeeded by a former consul who put down the murderer and another usurper, and in turn was succeeded by another senatorial governor.

The ‘Gallic Empire’ is conventionally regarded as a case of the secession of a section of the Roman Empire, which was therefore ruled by rebels, but this interpretation is difficult to maintain. Postumus seized power by murdering his predecessor, or having him murdered, certainly, but this was almost normal in the century of Commodus, Caracalla, Maximinus and many other such murderers. He was acclaimed as emperor by his army, was accepted without serious demur by the people of his part of the Empire, and he was able to assemble a group of wealthy men to form a Senate – many of them would have had senatorial rank already – and he himself took up the office of consul five times. It was a process very like that of Galba two centuries before, except that Postumus did not immediately head for Rome. The only element in the usual array of imperial powers that is absent is therefore possession of the city of Rome, but ‘legitimate’ emperors such as Macrinus and Maximinus did not visit that city either.

It is best to see Postumus and most of his successors as normal Roman emperors, though there are some of his successors to whom such recognition cannot be extended. Postumus was killed by some of his soldiers, who proclaimed one of their commanders, Marcus Aurelius Marius, as the new emperor. He must be considered ‘legitimate’ since he would seem to have ruled for some time, perhaps three months. On the other hand, Ulpius Cornelianus Laelianus may be regarded as an unsuccessful rebel, for he did not command much beyond his province, which was probably Germania Superior, and he had been suppressed by Postumus just before the emperor himself was killed. Marius was himself killed by Postuman loyalists, headed by M. Piavvonius Victorinus, a former high official of Postumus’, and his colleague in the consulship on one occasion. Victorinus was clearly Postumus’ successor, and was successful in holding his empire together, ruling for more than two years.

The justification for the secession led by Postumus had been the defence of Gaul against barbarian attacks, and the rulers had accomplished that; at least, there were no records of any attacks after Postumus won victories along the frontier in 263 and 264. However, the murder of Victorinus in 271 – supposedly at the hands of an outraged husband – brought collapse. No successor had been chosen in advance of the murder, which rather confirms its spontaneity and personal nature. Victorinus’ mother Victoria took control, persuading the army to wait for the arrival of the governor of Aquitania, C. Pius Esucius Tetricus. He was duly proclaimed on his arrival, but Spain seceded from the secession and rejoined the main Empire, which was now under the rule of the Emperor Aurelian. Barbarian invasions penetrated once more into Gaul, and Aurelian invaded in 273. Tetricus, in the usual style, had his son made Caesar, but either lost heart in the face of Aurelian’s invasion or lost the decisive battle.

Between them, Postumus, Marius, Victorinus and Tetricus ruled a quarter of the Roman Empire with some success for fourteen years. Postumus’ reign of nearly ten years was the longest, except for that of Gallienus, between Alexander Severus and Diocletian. Aurelian’s recovery of control of the West was marked by the reception of Tetricus and his son into the Italian governmental system, and no doubt into the Roman Senate, so their Gallic senatorial colleagues can have achieved no less. The Gallic emperors were in truth true Roman emperors and Aurelian’s treatment of the last of them confirms this. The ‘Gallic Empire’, in fact, in most respects looks like a version of the Roman Empire before the reforms of Gallienus, and maybe this was one of the sources of Postumus’ support; that is, it represented a conservative reaction against those reforms. This would in part explain his constitution of a regional Senate, and one in which the reforms were resented. Yet it was Gallienus’ reforms, in the army and in the government, that were the future for the Empire. Just to remain the same, in a conservative way, as was the aim of the Gallic emperors, was no longer enough.9

Gallienus had clearly alienated many men by his reforms, and perhaps by his apparent legitimacy as emperor. His generals in the end conspired to kill him; the Senate may well have been dismayed by his reforms, and his reputation has suffered by the accounts of the ancient historians, though moderns have been more indulgent. He had, above all, ultimately failed as emperor: the West had gone to Postumus and Gallienus had failed to regain it; in the East his father Valerian was defeated and captured by the Persian invaders in 260. The whole East had then gone first to the Persians, then to the Palmyrenes, who were claiming to restore the Empire, acting as local emperors in the same way as Postumus and his successors. Gallienus’ fall was triggered by the rebellion of one his generals, Aureolus, in northern Italy, perhaps in concert with Postumus. It was this rebellion that finally provoked Gallienus’ killing.

The rebel Aureolus was swiftly disposed of, fighting a final battle after he had despairingly proclaimed himself emperor.10 The army – that is to say, the generals – chose one of their own number to be the next emperor, M. Aurelius Claudius, called ‘Claudius II’ or ‘Claudius Gothicus’. The ordinary soldiers were unhappy about the change and the Senate was persuaded to deify Gallienus to mollify them. That is, Claudius turned to the Senate, as had his predecessors, for the validation of his election by the generals.11 Gallienus had carried through a reduction of the Senate’s powers, but they had not yet seriously affected its prestige, and the institution of the Senate in Gaul told the same story: a Senate, even one as speedily recruited and provincial as that in Gaul, was required if a regime was to be seen as legitimate. It was the same measure as that taken by Galba two centuries earlier.

Claudius had a brother, M. Aurelius Claudius Quintillus, who was given some sort of authority, but not, so far as can be seen, the title of either Caesar or Augustus. When Claudius died of disease in 270, after a reign of only two years, however, Quintillus assumed the imperial power as of right; the dynastic principle was once more in operation. He did not have much support in the army, not apparently being a military commander, nor did he have time to secure recognition in the Senate. After a reign of less than a month, just long enough to have coins minted in several mints and to give a good impression to the Senate, as the later senatorial historians show, but not being vigorous enough to be acceptable to the army, he ceased to reign. He had initiated some negotiations with Claudius’ commanders and probably with the Senate, but the proclamation of a more widely acceptable man disheartened him. He was either killed or committed suicide, probably the latter.12

The army’s new choice was L. Domitius Aurelianus. He was proclaimed somewhere in the Balkans where the army was on campaign, marched by way of Aquileia (where Quintillus had died) and on to Rome for the senatorial ratification. This was apparently willingly given. He was a well-known and energetic military man who had been one of Gallienus’ senior commanders (and one of his murderers, as was Claudius, probably) before becoming one of Claudius’ senior generals. He was in the line of emperors from Gordian III and Philip who had been keen to seek the support of senatorial opinion, perhaps in some cases too much so for the taste of the senior soldiers. The generals who killed Philip and those who killed Gallienus were clearly convinced that they could do a better job of defending the Empire than them, just as Aurelian and his soldiers were convinced that he could do it better than Quintillus. Those generals – it is a characteristic of the species – were also quite certain that they did not need support or interference from the senators. That is, within the army there was always a Maximinist persuasion, yet each time an emperor was struck down, his successor rapidly came to the conclusion that the support of the Senate really was necessary. It had happened with Claudius, it was happening with Quintillus when he died, and now with Aurelian.

When Gallienus was killed, the rank-and-file were very displeased and the participation of the generals in the killing was hushed up, both at the time and in the later histories. The rank-and-file were thus of the opinion that the Senate’s accolade was necessary. So Claudius sought Gallienus’ deification, and Quintillus and Aurelian headed for Rome as soon as proclaimed, though only the latter actually arrived. So for the time being the men called ‘Maximinists’ by the author continued to be out of power; like the Guard in earlier years they were able to kill or remove emperors, even to put one of their own in his place, but the successors had immediately become ‘senatorials’. This may well be the reason that the decision as to who should be the next emperor when Aurelian died was left to the Senate.

This, of course, was not the way it was normally done, for the Senate did not usually do more than ratify a candidate chosen elsewhere, either by the previous emperor or by a cabal of generals (or in the occasional case by the Roman plebs). The normal method was for the army to nominate one man, as had been the case with Aurelian. This made sense, since his main task at this time was to lead the army into battle and he had to be able to command it. The days of a mainly civilian administration were now, at least for the moment, over. The Senate was no longer qualified to select such an emperor, and as those senators with experience of commanding government died off, it would become decreasingly capable of correct military judgement. However, the army now insisted that the Senate select the new emperor – or at least the army commanders did so – and the rank-and-file acquiesced. This is not really surprising, given the rapid requirement of senatorial acceptance shown by the emperors since Gallienus’ death. Indeed, if the reaction of the troops after Gallienus’ death is anything to go by, it was their attitude that insisted on it. The army by now was seeing itself once more as a representative of the citizens, and ‘Senate and People of Rome’, inscribed on their standards, was clearly being taken seriously by them.

Aurelian was murdered in a plot involving court officials and disaffected officers, but the plotters did not have a replacement emperor ready to hand, suggesting spontaneity and panic. It was clearly the absence of a candidate, and perhaps paralysing rivalry among the generals, that compelled the commanders to involve the Senate. The sources imply that the subsequent negotiations between the generals and the Senate lasted some considerable time, and indeed, there was a period of several weeks with no emperor. The Senate finally nominated M. Claudius Tacitus. He was probably another old soldier; certainly he was old, he showed a certain command ability during his short reign, and he was acceptable to the army. The senators had thus shown good sense in choosing a man likely to be liked by the troops; or rather perhaps it was a case of each side blocking the other’s selections until Tacitus’ name became a compromise choice. His age was no doubt one of the main recommendations from the Senate’s point of view.13

Tacitus, as it turned out, did not last long, nor could he be expected to, being too old for the vigorous life required. He was probably in his 60s – if not actually 75, as one source says – and so was a fairly typical choice of the Senate. He died while moving the army eastwards in pursuit of some of the murderers of Aurelian. Whether his death was from disease, as a result of an accident or murder is not clear, though given his age a natural death is the most likely. He was succeeded by his half-brother M. Annius Florianus, who had been his Praetorian Prefect, apparently by acceptance by the army in the Balkans where he commanded, and then by the Senate. It was a succession by dynastic right, and it would appear that Florian was accepted throughout the Western parts of the Empire without difficulty.14

Another general, M. Aurelius Probus, was proclaimed by the army in Syria, where he commanded. The situation was reminiscent of the Quintillus/Aurelian dispute and came to the same result: the general triumphed over the heir, mainly due to Florian’s army being damaged by disease which broke out when he allowed himself to be blocked up in Tarsus in a siege. Probus’ forces did little actual fighting, and it was Florian’s army that finally ended the matter by killing their man.15

It is worth noting that frequent disputes between commanders nominated by different parts of the army only rarely resulted in serious fighting. The armies of Philip and Decius had fought a battle, though perhaps only doing so until Philip was killed; in later cases the result was often achieved when one army eliminated its own candidate, either because it faced unwelcome odds or when it was persuaded that the opposing candidate was a better man for the job. So Valerian did not have to fight to triumph over Aemilian and Gallus; when Gallienus did get involved in a siege against a rebel he was removed (and so was the rebel); Quintillus’ army peacefully accepted Aurelian; and Florian’s army accepted Probus. Actual and prolonged civil war between these forces was therefore very rare. The only serious exception was the secession of the Gallic Empire and of the East under Palmyra, both of which are rather different situations, and even in the case of the suppression of the Gallic secession, it was not clear that there was serious fighting; the final battle may not actually have happened. These Eastern and Western events were not disputes over the succession so much as disputes over the integrity of the Empire. The unity and defence of the Empire was the basic programme of the army, and this was the justification for all the political manoeuvring and murders of the armies and the Guard.

Probus went to Rome for his formal senatorial investiture and made a good impression, by all accounts.16 He faced rebellion and invasion during much of his reign. The Historia Augusta is eloquent on the Senate’s favour towards him, but the passage must be judged an invention. Nevertheless, Probus’ reputation is of an emperor in alliance with the Senate, and his employment of a senior senator, Virius Lupus, as governor in Syria would suggest that this reputation was in part earned by a decision to employ senators once more in government positions. He was killed by his officers, apparently because of an offhand remark he made that he had been so successful in pacifying the Empire that he might soon be able to do without soldiers. The implication was that he would be able to reduce costs to the civilians of the Empire by at least reducing the number of soldiers and therefore their wages. He seems also to have annoyed his soldiers by making them work on construction projects. This gave his Praetorian Prefect, M. Aurelius Carus, a platform. The troops under Carus’ command hailed him as emperor, and then the soldiers sent to suppress him joined in. Probus was then murdered.17

Carus took power smoothly when Probus died. He was one of those emperors who did not go to Rome for the investiture, and the fact that his son Carinus did so a year later was no substitute. Given the circumstances of his coup and the men who supported him, this must put Carus and his two sons into the party of the Maximinists, the believers in a purely military organization of the Empire; Carus promoted himself and his sons to the rank of Augusti and simply reported his elevation to the Senate with the implication that it did not really matter whether the Senate approved of him or not.18

Carus died on campaign in the East; his younger son Numerian followed him, perhaps killed by his Praetorian Prefect; the elder son Carinus faced rebellions by a local commander, M. Aurelius Iulianus, in Venetia in northern Italy, and by the generals in Numerian’s army, who proclaimed C. Aurelius Valerius Diocletianus as the new emperor. Carinus destroyed Iulianus in a battle near Verona, but was then killed by some of his own men during the battle with Diocletian’s army. Diocletian then publicly cut down the man accused of the murder of Numerian.

This confused set of coups, murders and battles is difficult to sort out. If Carus represented the militaristic (Maximinist) tendency, as his disdain for the Senate implies, then Iulianus should perhaps represent the senatorial party, and since he was operating as a governor, he was one of those senators who still played an active governmental role. His suppression by Carinus does indicate, besides his rebellion, that he was a political enemy of the Maximinists. Carinus in turn had referred to the Senate in dictatorial terms rather than with the more traditional request for confirmation of authority. So the deaths of Carus and his sons might be seen as coups on behalf of the revival of the Senate’s power.19

This may well have encouraged the senatorial party, but Diocletian proved to be just as unwilling to engage with the Senate as Carus had been. Probably Diocletian visited Rome, but only for a very brief visit; there is no sign of the Senate’s attitude to him. Probus had been conciliatory towards the Senate, but he was the last emperor to be so. After him, none of the emperors felt much need to gain the Senate’s approbation. The acceptance by the emperors of the refusal of most senators to engage in the government of the Empire had resulted inevitably in the emperors’ engrossing of powers to themselves, including the power of making a successor; the Senate had moved on to irrelevancy.

The accession to power of Diocletian proved to be the turning-point. The new system he constructed proved able to control the Empire and to ward off its external enemies but, of course, at the cost of increased autocracy. He was reacting against the political system that had existed before his seizure of power, and had existed during the preceding chaotic half-century. The balance of power that had been established by the upheaval of 238, between the Senate and the army, was perceived to have actually promoted disorder.

Such a balance had proved very difficult to maintain. There were always groups in the army who were liable to advocate a return to Maximinus’ disdain for the Senate and others who were much more willing to have the Senate taking a central role in the political and governmental process. The circumstances of the deaths of emperors, invariably with the army on campaign, whether by disease or assassination and invariably in the midst of a crisis, meant that it was always urgent for a new emperor to be chosen as quickly as possible. The command of an army by a single ruler was the only way of keeping the Empire united. The Senate was then invited to ratify the army’s choice, but not always. Mainly it did so, because the army and the new emperor made efforts to be considerate, but there were always those who despised such considerations and manoeuvres.

On the whole, the system had operated successfully. The mid-third century is regarded as confused and confusing, but this is largely a product of the few sources, which are late and difficult, together with the repeated barbarian invasions. Yet the succession of emperors was generally achieved by agreement until the dynasty of Carus (and dynastic succession was repeatedly attempted). We can see, therefore, that beyond the apparent confusion, the same factors in determining the succession were operating as in the previous centuries: the previous emperor’s preference, if he had indicated one, the army and then the Senate, in that order. The balance between these shifted for five decades, and it seems to have been very largely the Senate that predominated, even if the army tended to be the source of each new emperor. Finally the balance tipped and power came down on to the army and on dynastic succession, very largely cutting out the Senate. This was in part due to the governmental reforms of Gallienus, which reduced the role of the senators in public affairs. Yet the Senate survived, and its ambitions were undimmed.20

However, it was not simply a matter of the previous half-century, but of the previous two and a half centuries. For what had collapsed during the period since 238 was the system of selecting an emperor that had been developed by Augustus. This had, in a variety of ways, operated since his death. It had been altered in the course of that long period, as the basic military situation of the Roman Empire gradually obtruded itself. This was realized first by Vespasian, then by Trajan, and brought to its fully military refinement by Septimius. Changes such as this is what one would expect from any system, and the fact that it had operated for almost two centuries with only minimal changes was a tribute to its originator and to those who operated it. Nevertheless, it failed and its failure produced the chaos of the middle of the third century. It was Diocletian’s achievement to realize that tinkering with the system was no longer an option.

The Eastern Problem


Odainathos (250).


Zenobia (270–272).


Vaballathus (270–272).


Balista (c.260).

The Restorers Of Unity


Claudius Gothicus (268–270).


Quintillus (270).


Aurelian (270–275).

Continuing Consolidation


Tacitus (275–276).


Florian (276).


Probus (276–282).

The House of Carus


Carus (282–283). (Met Museum of Art/Public Domain)


Carinus (282–285).


Numerian (282–284).

If you find an error or have any questions, please email us at Thank you!