9
FROM GORDIAN III TO VALERIAN
The bloodletting of summer 238 ended, in Italy at least, in a relatively stable peace. The 12-year-old emperor Gordian III, acclaimed caesar in February, and then augustus in May or June, was watched over by a cabal of mainly equestrian officials who determined imperial policy. These men were led by C. Furius Sabinius Timesitheus, the praetorian prefect, whose power was such that he married his daughter Tranquillina to the emperor himself in May 241, as soon as she was old enough. But whatever stability had accompanied the end of the civil war proved illusory. The proconsul of Africa, M. Asinius Sabinianus, who had replaced old Gordian, revolted soon after the new regime had settled. He was an old Severan, having been consul in 225, and it may be that he objected to the sidelining of the consular elites after the failure of Pupienus and Balbinus’s regime. Sabinianus’s putsch failed, put down by the procurator of Mauretania, Faltonius Restitutianus. He was replaced as proconsul by L. Caesonius Lucillus Macer Rufinianus, who had been one of the vigintiviri alongside Balbinus and Pupienus, which suggests that we must not read events in terms of senatorial and equestrian factions at court, but instead as rival factions within equestrian and senatorial ordines.
Nevertheless, with Timesitheus at the centre of things, the government continued to have the professional, systematising outlook of the bureaucratic classes – among the equestrians known to have prospered greatly during the reign we find M. Gnaius Licinius Rufinus, as a libellis; C. Attius Alcimus Felicianus, whose career had begun under Elagabalus and whose financial posts clearly made him something of an expert in the field; Gnaeus Domitius Philippus, who was prefect of the vigiles at the start of the reign; Faltonius Restitutianus, who put down the revolt of Sabinianus in Africa; and two brothers from Arabia, Julius Priscus and Julius Philippus, the latter becoming emperor within a few years: by the early 240s, the world in which the accession of Macrinus had been resisted because he was not a senator no longer existed.
In 238, the regime’s first order of business was Persia. As we saw in chapter 8, the Sasanian kings who replaced the Arsacids in Persia and Mesopotamia were much more expansionary than their predecessors had been, bringing to heel the semi-independent satraps of Mesopotamia and the Caucasus, in part provoked by the survival of the Arsacids in Armenia, where kings like Tiridates II (r. 217–52) attempted to rally other frontier dynasts as far away as India against Ardashir. By the time the regime of the young Gordian had settled, much of Roman Mesopotamia was exposed to Persian invasion, Nisibis, Carrhae and Hatra had all fallen, and other fortress cities like Singara could not be re-enforced. The prestige of the new regime at Rome would be much enhanced if it could secure the eastern frontiers in a way that Maximinus had failed entirely to do.
Leaving behind Alcimus Felicianus to run Rome, and having appointed Julius Priscus as his fellow praetorian prefect, Timesitheus and the young emperor marched east in 242, evidently having found it difficult to muster a campaign army. They signalled the gravity of their intentions by opening the doors of the temple of Janus in Rome itself, probably the last time in history that this archaic ritual declaration of war was performed. They then progressed overland through Moesia and Thrace, their crossing from Europe to Asia, probably early in the summer of 242, being commemorated with an issue of gold medallions with the legend traiectus, ‘crossed over’, to important courtiers. Antioch in Syria served, as it usually did, as the staging point for the campaign against Persia, but delays were endemic. The year 243 was spent in Syria, but not until the winter of 243–4 do we find the army fighting along the Euphrates. It may be that the death of Timesitheus sometime in 243 had contributed to the delay. His successor as prefect was Julius Philippus, brother of the other prefect Julius Priscus. The joint prefecture of two brothers was unheard of, but in the circumstances, the two men had become indispensable: they came from Shaba in Arabia, had risen through the equestrian grades (Priscus, at least, had once been a fiscal procurator), and their local connections made them a good conduit to the region’s elites whose cooperation was necessary if the campaign was to go smoothly and the army was to be properly supplied.
At first, the fighting went the Romans’ way and a victory over Persian forces is recorded in the sources, perhaps at Resaina in Osrhoene, and possibly in battle against the Persian shah himself. This was now Shapur I, ruling alone after the death of Ardashir in 242, though he had been effectively in charge since 240, when he was crowned co-ruler with his father. Shapur, even more than Ardashir, was the true architect of Sasanian power. Though he had not yet articulated an ambition to recreate the Persian empire of the Achaemenids, there is no question that he embraced an Achaemenid more than a Parthian model of display. What is more, he took to new levels his father’s militance, campaigning on every frontier of his empire and imposing Sasanian governors, often members of his own house.
Much of what we know about Shapur’s early reign comes from the monumental inscription he put up to commemorate his victories at Naqsh-e Rustam. The site is significant, for it lies a few miles outside the ancient Achaemenid city of Persepolis and houses the rock-cut tombs of several Achaemenid kings, including Darius the Great and probably Xerxes I. By appropriating for his own display the necropolis of the last conquering dynasty to come from Fars, Shapur was perhaps implying a continuity with them, and certainly displaying himself firmly in a Persian rather than a Mesopotamian or Parthian light.
Of the two earliest Sasanian reliefs at Naqsh-e Rustam, one shows Shapur on horseback, with a Roman emperor kneeling in supplication before him. The other shows Shapur’s father Ardashir being invested with his crown by the supreme Zoroastrian divinity Ohrmazd. A more elaborate version of the Shapur monument appears at Bishapur, a town that served as a staging post between the Sasanian dynastic centre at Istakhr and the old Parthian capital at Ctesiphon in Mesopotamia. Back at Naqsh-e Rustam, a square tower known as the Ka’ba-i Zardusht (or the Kaaba of Zoroaster) stood opposite the rock-cut tombs of the Achaemenids, and had served as a Zoroastrian fire sanctuary since the reign of the Achaemenid Darius. On it, Shapur’s son Ohrmazd I had inscribed a text in three languages – Parthian, Middle Persian and Greek – that his father had composed in the last years of his long reign, outlining an official version of his glorious deeds and his piety. The inscription gives Shapur’s title as His Mazdayasnian Majesty Shapur, King of Kings of Iran and not-Iran (or of the Aryans and the non-Aryans) Whose Seed is from the Gods. Shapur’s father Ardashir had already revived the Achaemenid title of shahanshah (‘king of kings’) and asserted a personal connection to Ohrmazd, but Shapur now added an explicit claim to universal rulership that matched that of the Roman emperors. The rest of the text gives us the Persian account of Shapur’s struggles with his neighbours and is often strikingly different from what we find in the Greek and Latin sources, sometimes completely contradicting them. It also, to some extent, confirms the strong focus of Shapur on his conflicts with the Romans, rather than with other parts of his realm, for the battles that he commemorates on his inscription are all those that he fought against the Romans, rather than on his eastern and north-eastern frontiers.
And yet we also know that Shapur inherited the same problems his Arsacid predecessors had faced on the eastern frontier. Early in his reign, he may have subjugated Khwarezm, the northernmost of the Central Asian oases, at the delta of the Amu Darya beside what was then still the Aral sea, though it never became a province of the empire. It was also Shapur, though when in his reign we do not know, who reduced the Kushanshahr to a client state of his dynasty, marked by a series of coinages that we know as Kushano-Sasanian. Here, as elsewhere, we continue to learn a great deal that is new in Sasanian history from the numismatic evidence – much of it, sadly, coming to light from the clandestine excavation and looting made possible by conflict in present-day Afghanistan and Pakistan. Most significantly, the heavy concentration of Sasanian minting in places like Merv and elsewhere in the east, some of it by die-workers clearly trained in the metropolitan mint at Ctesiphon, demonstrates how much military force was needed to control the east under Shapur and how many campaigns he must have fought there.
In his inscription, he lays claim to Sind and ‘the Kushanshahr up to Peshawar and up to Kashgar, Sogdiana and the mountains of Tashkent’, but he does not tell us about the fighting that was clearly necessary to control that region. Instead, he focuses on his Roman victories – reciprocating the kind of focus that Roman emperors had for their eastern neighbours. This Sasanian focus on Rome marks a change from the Arsacids’ more balanced division of attention between their eastern and western frontiers, although Shapur did continue the Arsacid policy of farming out control of the Syrian and northern Arabian deserts to clients, most particularly the Lakhmid king of Hira, Imru’ulqais, who had succeeded his father some years before and went on to serve not just Shapur but also his successors Ohrmazd I and Varahran I as satrap of Iraq and the Hijaz.
For historians of the Roman empire, this Sasanian interest in Rome offers a valuable counterpoint to the sparse imperial evidence. On the Naqsh-e Rustam monument, Shapur claims to have defeated and killed the emperor Gordian at Misiche (or, in Persian, Mishik) on the middle Euphrates. Of the three Roman emperors depicted on the victory monument, one lies dead on the ground, one is supplicant and the third has been taken captive – Gordian, Philip and Valerian, respectively. Shapur also renamed Misiche Peroz-Shapur, or ‘victorious is Shapur’. By contrast with Shapur’s explicit claims, the Roman sources are ambiguous. None straightforwardly attests Gordian’s death in battle. Indeed, the best Roman evidence suggests that Gordian died further north than Misiche, at Zaitha, sometime between mid January and mid March 244. He was certainly buried there, at least temporarily, in a massive tumulus that could still be seen more than a hundred years later, when another Roman army was invading the region.
What actually happened will never be known, but Gordian was a teenager and had little military experience. Few can have expected great things of him on the battlefield. It may be that a mid-winter defeat by Shapur on the edge of Persian territory, one in which the emperor himself was perhaps badly wounded, prompted a disgruntled soldiery to assassinate him at Zaitha. An opaque passage in the life of the philosopher Plotinus, who had been accompanying the imperial expedition on a sort of research trip for esoteric knowledge, suggests that there was rioting in the Roman camp when Gordian was killed. As one would expect, many sources – from the near-contemporary apocalyptic text known as the Thirteenth Sybilline Oracle to the fourth-century Latin tradition of abbreviated histories – blame the man who profited from Gordian’s death for causing it, Julius Philippus, the praetorian prefect who had succeeded Timesitheus. But Philippus (or Philip the Arab, as he is conventionally known) was not with the army at the time of Gordian’s death, though his brother Priscus may have been. That fact could explain why the older and presumably predominant brother did not himself take the throne – throughout late Roman history, councils of army officers sometimes chose an imperial candidate who could achieve consensus precisely because he was not present on the spot to take part in debate about the succession.
Regardless of how Gordian died, it took a lot of negotiation for the army to extricate itself from disputed territory, and the new emperor had to act personally as a supplicant in the peace talks. The settlement, which was all Philip could have hoped for in the circumstances, was the root of endless further conflict between Persian and Roman monarchs. According to the Roman sources, Philip ‘betrayed’ Armenia to the Persians, which must mean that he acknowledged Shapur’s right to determine that kingdom’s succession, as the Arsacids had done for centuries. Shapur, in his victory inscription from Naqsh-e Rustam, claims both that Philip became his tributary and that he was made to pay an indemnity of half a million gold aurei, a seemingly impossible sum, but at the very least an approximation of the scale: to get his army out of Persia safely, Philip mortgaged his throne. In all likelihood, along with acknowledging Sasanian hegemony in Armenian affairs, he also transferred the traditional supplementary payments for guarding the Caucasian passes against nomadic incursions from the Armenians to the Persians themselves, hence the shahanshah’s willingness to claim that Philip was offering him tribute – and hence, too, the reason the Roman sources are silent on such details.
Philip, for his part, put as happy a face on things as he plausibly could. Back in Antioch, he struck antoniniani with the legend pax fundata cum Persis (‘peace made with Persia’), took for himself the titles Parthicus and Persicus Maximus, and began to establish the dynastic image of his family. His wife, Marcia Otacilia Severa, was made augusta and named mater castrorum, a direct assertion of the regime’s concern for its soldiers. Then, on two key frontiers, Philip installed relatives as his representatives: his brother Julius Priscus in Syria and their brother-in-law Severianus (Otacilia Severa’s brother) on the Danube in Moesia. He himself returned to Rome as quickly as possible, taking the sea route up the coast of Asia Minor, reaching the imperial capital early in the summer of 244. Severianus’s command shows the growing importance, in these years, of the Danubian armies by contrast to those of the Rhine frontier; it may also be an early sign of the increasingly assertive Roman self-consciousness in a region that was one of the last Latin-speaking parts of the empire to gain widespread access to the Roman citizenship – certainly men from the Danube would come to dominate politics in the latter half of the century.
The Syrian command of Julius Priscus, meanwhile, demonstrates Philip’s determination to keep an eye on Persian developments and to maintain the family’s close connections to their native east. The reconstruction of the dynastic home town of Shahba under the new name of Philippopolis was a truly massive endeavour, one financed in part by stricter financial exactions under the supervision of Priscus. Priscus’s role is also significant: he was not merely the governor of Syria but also corrector, a nebulous word that signalled his precedence over other officials. He possessed, in other words, supra-regional jurisdiction over the other governors of the east, an important precedent for later third-century experiments in government. Priscus was, to all practical purposes, Philip’s co-ruler in the east. Severianus probably disposed of a similar authority in the Balkans, though his title is not explicitly attested in the way that Priscus’s is.
At the same time that Philip was securing his familial power in this way, he was also ensuring that he had no challenges to his own legitimacy – he had not forgotten how popular the young Gordian had been with the people of Rome, and so he put about word that Gordian had died of an illness and brought his body back to Rome, burying it with honours. He also asked the senate to approve the dead boy’s consecration as Divus Gordianus, which it did, and though Tranquillina disappears from the historical record, she is likely to have enjoyed an honourable retirement, since she and Gordian had lacked any worrisome heirs. Philip’s own son, born around 237/238 and thus five years old, was now made caesar. Because of the poor documentation for this period, we do not know how long this relative tranquillity lasted, or how popular Philip was in Rome itself. But trouble at the frontiers occupied the middle of his reign.
Just as the earlier 200s had seen upheavals among the barbarian polities of the upper Rhine and Danube, so the middle years of the century brought major cultural and political change beyond the lower Danube and the northern Black Sea coast. Scholars have traditionally associated these changes with the arrival in the region of Goths migrating from their former homes in what is now Poland. This narrative is shaped around what we find in the Getica of Jordanes, a tale of ethnic origins composed hundreds of years later, in sixth-century Constantinople, by a Latin-speaking Roman of Gothic descent. Archaeological evidence has been consistently distorted to fit a legendary saga of mass migration and Scandinavian origins. But this simplistic model is not well supported by the evidence.
It is true that a new and relatively homogeneous archaeological culture developed between the Carpathians and the Donets river in the second half of the third century, and that by the 320s this region was ruled by a number of tribal polities whose main language was Gothic and whose military elites are lumped together as ‘Goths’ in the Roman sources. This archaeological culture, which we call Sântana-de-Mures/Černjachov, is named for two cemeteries, one in modern Rumania, one in modern Ukraine, that share characteristic types of grave goods and burial ritual. New forms of settlement appear in this region during the second half of the third century as well, with farming villages clustered along the arable river valleys, large compounds for the elites dominating the landscape, a predilection for high-value Roman imports and a symbiotic relationship with nomadic pastoralists where agricultural lands bordered the grassy steppe. The decorative and dress styles revealed in the grave goods share some affinities with those found a century or so earlier in northern Poland, but they also show elements of central European origin, and quite a lot of influence from the nomadic art of the Eurasian steppe.
Rather than shoehorning this evidence into a narrative of mass migration from the north in order to fit our late literary evidence, responsible scholars recognise a local development, involving migration from both the steppes and parts of northern Europe, reshaping the more fragmented and less hierarchical agricultural societies that had preceded them. The formation of new, and often socially more complex, societies on the fringes of empires is well known from comparative evidence, both from antiquity – such as the Alamanni, discussed in chapter 8 – and from such modern examples as Tsarist Russia in its expansionist phase or the frontiers of the British Raj. The existence of the imperial power gave tribal military leaders a powerful structure against which to fight, but also something to learn from, and a supply of resources – either in loot, subsidy or trade – which they could distribute to increase their own power. Given time, and the consolidation of new tribal polities, a new and more or less uniform material culture developed out of numerous different antecedents, and a common language among the elite population in the form of Gothic.
Somewhat ironically, one of the first pieces of evidence we have for Goths and their relationship with the Roman empire is Shapur’s monumental inscription at Naqsh-e Rustam, which lists Goths among the nations he defeated in Rome’s armies and thus demonstrates that by the middle of the century the emperors were recruiting men from beyond the Danube into units that they could designate as Gothic. That said, the consolidation of the lower Danube and Black Sea regions under Gothic hegemony is really only apparent in the fourth century, when it was already complete. During the third century, all we see are its by-products – invasions from beyond the Danube and the Black Sea that were a constant problem for third-century emperors, whether led by assertive new rulers seeking plunder or their defeated opponents looking for greener pastures. In what follows, we will refer to the various third-century invaders from this region not as Goths but as ‘Scythians’, the generic word used in our Greek sources, rather than making presumptions about how they identified themselves or retrojecting Gothic hegemony into a period when it cannot be documented. These Danubian ‘Scythians’ joined the Persians and Alamanni as the most formidable of Rome’s neighbours from Philip’s reign onwards.
By mid 245, Philip had marched with an army to Dacia, and he is known to have been at Aquae in November of that year. Trajan’s Dacian provinces had always been an experiment and Roman civic life never put down roots in Dacia, as it had begun to do south of the Danube from the time of Marcus Aurelius onwards. Pannonia and Moesia, and the Balkans more generally, began to look more and more like the rest of the Latin provinces as the later second century turned into the third. Civilian population followed in the wake of the huge military investment of the Marcomannic wars and three generations later, at mid third century, civilian life flourished both in the great military cities that led from the Alps to Asia and in rich villas where the topography was suitable to agriculture. Dacia, by contrast, had been built up hastily after Trajan’s defeat of the last Dacian king Decebalus, with few monumental cities and a network of roads and way stations that primarily served a military presence needed to protect the mineral resources of the province.
Even though the Sarmatians and Carpi, whose territories neighboured the province to the west and east respectively, were relatively easy to control compared to the more powerful Alamanni, it does not seem that Roman-style living penetrated very deeply into the cultural fabric of the Dacian provinces, or that a civilian infrastructure developed save to cater for the direct needs of the military and the mines. Things might, of course, have changed – Pannonia and Moesia did not really start to develop into fully Romanised provinces until the time of Marcus, a hundred years or more after their conquest – but the upheaval to the east of Dacia, and in particular its challenge to the local hegemony of the Carpi, meant that the very real revenues from the Dacian mines can hardly have been worth the expense of keeping the region garrisoned.
Perhaps disturbed by events among the ‘Scythians’ north of the Black Sea, the Carpi began causing trouble on the Dacian frontier in 245, and Philip continued fighting there in 246, probably well beyond the imperial borders. Late in the summer of 247, he returned to Rome and celebrated a triumph, taking the official title Carpicus Maximus and perhaps also being hailed as Germanicus Maximus. The imperial family stayed in Rome for the winter of 247–8 and the following spring Philip celebrated the completion of the thousandth anniversary of Rome’s history, on 21 April of that year. As we saw in the discussion in chapter 5 of the previous secular games, there was some confusion among the Romans themselves about their secular games, and so two rather different sequences of them developed. One was the supposedly archaic and Etruscan celebrations revived by Augustus in 17 BC, which were conducted every 110 years, supervised by the quindecimviri of the official priesthood and celebrated over a triduum from the nights of 31 May to 3 June. The others were centenary celebrations to commemorate centuries (saecula) since the foundation of Rome, which took place on her dies natalis, 21 April – the feast of the Parilia (or Natalis Urbis, as it was uniformly referred to in our period). In their different ways both sorts of ‘secular’ games publicly performed a link between the past and the present of the Roman state, asserting an essential continuity in Rome’s identity over the years. Domitian and Septimius Severus had celebrated games in the Augustan sequence, Antoninus Pius in the second, and Philip’s would continue that of Pius.
There can be no doubt that Philip’s millennial games were even more symbolically significant for those purposes. We do not have an involved description of Philip’s games, but one (not very reliable) source reports that he used the gladiators and animals that had been intended for the Persian victory of Gordian, a victory that of course never came. We do, however, have epigraphic attestations of the importance of these millennial games, and they are noted in the various strands of Greek and Latin chronicling from late antiquity. Philip’s coinage is particularly rich in references to their celebration. Struck both in commemorative aurei and regular antoniniani and sestertii, in the name not just of Philip himself but also of the young Philip II and Otacilia Severa, they depict various animals killed in the Circus Maximus: lions, antelopes, hippopotami, ibexes, stags and gazelles, as well as pictures of a temple to Roma with her statue among the columns. We also see depictions of the chariot races – ludi circenses – that had been a regular part of the annual celebration of the Natalis Urbis since the time of Hadrian. But to the beast hunts and races, Philip added the rituals of the ‘authentic’ secular games in the Augustan sequence, with a triduum of theatrical spectacles on the Campus Martius, as well as singing competitions and other events whose victors are recorded in surviving inscriptions. This wide variety of attestation – particularly by contrast to the games of Antoninus Pius – is a marker of the celebration’s millennial importance.
Perhaps the most surprising evidence of this is also the strangest: common terra sigillata – the red-slip tableware that graced every Roman table – decorated with medallions honouring Philip’s games rather in the manner of today’s painted place settings commemorating royal weddings or presidential inaugurations (and perhaps just as kitschy then as now). The pomp and circumstance of Philip’s games was clearly felt out in distant provinces, but scholars remain divided about whether the sort of ‘millennial fever’ that was felt in Christian Europe around AD 1000 was there in Philip’s time as well. The Latin sources are ambiguous at best, but there is one eastern source that does seem to make the case: a collection of Greek texts that came to be known as the Oracula Sybillina. These had nothing whatsoever to do with the original Sybilline oracles of the republican era, but were both a response to the crises of third-century politics and genuinely apocalyptic in outlook – in the Thirteenth Oracle, much the most famous in the collection, the millenarian expectations of Rome’s thousandth anniversary meets with a fierce sense that the Roman empire needed to end.
Yet despite any such currents of unease, Philip’s lavish celebrations seem to have been a propaganda success. The completion of one millennium could also be seen as the start of another, just as glorious: certainly successors of Philip continued to strike secular coins a full half decade after his games had been celebrated. In general, however, Philip’s government was not faring well. In 248, in Cappadocia or Syria, a nobleman from Commagene named Iotapianus was proclaimed emperor, supposedly to protect the provinces from the heavy-handed exactions of Julius Priscus. There may also have been a revolt on the Rhine, led by one Marinus Silbanniccus, although the date of this revolt is uncertain and only two coins of the usurper are known to exist. More dangerously, the consular governor of Moesia, Claudius Marinus Pacatianus, was proclaimed emperor, possibly in April 248, probably at Viminacium where his coins were minted. But he was swiftly killed by his own soldiers when the neighbouring provincial governor led his army into Moesia.
It is possible that the senator Pacatianus represented the hostility of his order towards the junta of bureaucrats that surrounded Philip, one that resented the supra-regional commands awarded to men like Severianus and Priscus. It is also possible that Philip’s celebration of the Roman millennium, with its heavy freighting of tradition, contributed to his decision to rise up. Regardless, Pacatianus’s rebellion was put down. Unfortunately for Philip, the commander who suppressed him immediately allowed himself to be proclaimed emperor in turn. C. Messius Quintus Decius Valerinus was a senatorial commander in his mid-forties who came from the region, having been born near Sirmium. Under Severus Alexander he had governed both Moesia and Germania Inferior, while he held the prestigious command of Hispania Citerior under Maximinus and remained loyal to that emperor in the face of the senatorial revolt. In 249, he was serving as a legate of both Moesia and Pannonia, in one of the extraordinary commands that Philip so favoured. It was in Pannonia that his rebellion was declared, the mint at Viminacium striking issues in his name almost at once. He also changed his name to C. Messius Quintus Traianus Decius, reflecting the glory of the emperor Trajan who had conquered the Dacians, in what may have been a reference to Decius’s own Balkan origins.
A bigger puzzle surrounds the whereabouts and actions of Philip himself. He was either still in Italy, where he mustered an army to personally confront this latest usurper, or he was deep in Thrace, marching eastwards to deal with the rebellion of Iotapianus, having thought it safe to leave Decius to deal with the Danube frontier. Either way, having suppressed a brief mutiny at Viminacium, Decius either advanced from Pannonia at the head of the Balkan legions, crossed the Julian Alps and defeated Philip at Verona in September 249, or else he marched down the main Balkan highway in the opposite direction and defeated Philip at Beroea in Thrace. Philip was killed by his own soldiers, the young Philip was put to death in Rome by the praetorians when news of his father’s defeat reached them, and the emperor’s other relatives are never heard from again.
After defeating Philip, wherever he did so, Decius advanced to Rome, and was by September recognised as pontifex maximus and pater patriae. He immediately launched a programme of great ambition. His ostentatious traditionalism was obvious from the outset, starting with his decision to change his name to include that of a conquering emperor remembered for his goodness as much as his prowess. The names he gave his children by the Roman matron Herennia Cupressenia Etruscilla are equally redolent of the past: Q. Herennius Etruscus Messius Decius and C. Valens Hostilianus Messius Quintus. Just as extraordinary were his numismatic initiatives. Coins were, of course, among the most widely seen and handled items to emanate from the imperial court and the emperor’s circle. While we should not presume that emperors were personally responsible for every change and fillip in the numismatic iconography, when a major programmatic initiative in the coinage differs distinctly from what has gone before, we should take notice, as is the case with Decius.
After the new emperor had reached Italy, the mints at Milan and Rome began to issue not just the expected sorts of coins honouring Dacia, Pannonia and the genius exercitum illyriciani (‘the genius of the Illyrian armies’) but also an unprecedented series of antoniniani celebrating the deified emperors of Rome. These coins showed an altar on their reverse with the legend consecratio, while on their obverses there appeared a series of imperial portraits of the divi, beginning with Augustus and taking in Vespasian, Titus, Nerva, Trajan, Hadrian, Antoninus Pius and Marcus Aurelius. That imperial pantheon is standard and to be expected, but the selection of names that follows is more interesting: Commodus, Septimius Severus and Severus Alexander. No Pertinax and no divus Gordianus, the latter having been deified a mere five years previously. Now to be sure, false genealogies and the retrospective ascription of dynastic ties are something we have become used to already in this book – from Severus’s shifting decisions about which ‘ancestors’ to honour, to the sedulously cultivated rumours of a filial connection between Caracalla and the two young cousins who succeeded him. Equally, Greek and Roman viewers were used to the periodic erasing of emperors from official memory, the physical chiselling away of names from monuments both effacing memories and simultaneously forcing remembrance, the better to damage the reputation of the one thus effaced. But the ‘virtual’ erasure of memory as we see it enacted here is something different and new. Because the remembrances with which it so visibly tampers were so recent, it can only be read as a deliberate rejection of the previous decade. Decius was asserting his direct succession to Severus Alexander and retaining none of the intervening emperors, not even the youngest Gordian whose consecratio was a matter of living and public knowledge. The miserable decade of invasion and civil war, it was meant to be clear, had ended.
In the end, Decius did not reign long. Yet he is one of the most famous emperors of the third century, not on account of his extraordinary coinage but for another measure altogether: an order for universal sacrifice to the gods, for the good of the Roman state. Like the new coin types, it was meant to signal a new beginning, the return of prosperity to the empire, but for it to work, everyone would need to sacrifice. The plan for carrying out the new emperor’s order was modelled on the collection of taxes and registration for the census, and the evidence for it survives to a remarkable degree in the papyri of Egypt. These papyri are libelli that record a citizen’s act of sacrifice to the ‘ancestral gods’ before public officials, who are named in the documents as having presided over the act of sacrifice and the consumption of sacrificial food. As with census registration for tax purposes, the inhabitants of each province were assigned a day on which they were required to appear before their municipal magistrates to perform sacrifice by burning incense to the gods, and were then issued with a chit proving that they had done so – with failure to comply carrying serious consequences. This devolution of imperial directives on to local authorities is entirely characteristic of the way regional administration worked in the early empire, but we know much more about Decius’s edict than most others, for two reasons: one being the survival of the Egyptian libelli; the other being the impact it had on Christians.
Although the edict’s wording was vague, and did not specify which or whose ancestral gods were being referred to, Christians seem to have interpreted the order as requiring them to sacrifice to the gods of the Roman state. Christian monotheism rejected any distinction between belief and cultic acts. Christians were prohibited from worshipping any gods but their own one god, and they regarded the many other gods they saw all around them as demons. Most Christians, that is to say, believed full well in the gods of the Roman state, knew them to be quite real, but regarded their reality as demonic rather than divine. An order to sacrifice along the lines of the Decian edict could not help but challenge them. Choosing to comply meant not just sinning, but exposing their souls to eternal damnation. Failure to comply, by contrast, meant exposing each one individually to the more immediate bodily vengeance of the Roman state. Thus, whether or not Decius had intended it as such, it was impossible for Christians to interpret his edict as anything other than a deliberate attack on them and their beliefs. Decius is duly remembered by Christians as the second of the great Roman persecutors after Nero, the scale of whose persecution is much exaggerated.
Decius’s intentions have long provoked debate. Some elements of his plan are uncontroversial. He clearly believed that a single, unifying ritual act was necessary to please the gods and ensure the state’s safety. Perhaps he was mainly concerned with shoring up his own rather weak claims to the purple. Perhaps he was responding to millenarianism provoked by Philip’s celebration of Rome’s thousandth anniversary. He may also have been influenced by the outbreak of a new and frightening disease that was taking hold in the eastern parts of the empire: beginning in 249 in Alexandria, a severely contagious illness that caused high fever and conjunctival bleeding had begun to spread across the empire. Fading away in high summer, it returned in autumn, reaching Rome and Carthage by 251 at the latest and recurring periodically for at least a decade thereafter. The precise virus responsible cannot be determined, but scholars have recently begun to recognise just how serious it was, affecting town and country, rich and poor, and killing off as much as two-thirds of the population in some cities, if our better sources are to be believed. Haemorrhagic fever similar to the Ebola virus has been hypothesised: the combination of seasonality with high morbidity and mortality makes a filovirus of that sort a likely candidate. As the recent Ebola outbreak shows, the sudden impact of a new epidemic can prompt hysteria and the demand that our leaders take drastic measures to save us. We should not be surprised if ancient people, lacking modern epidemiological knowledge, thought it necessary to propitiate angered divinities, and worried that those who habitually refused to honour the gods were responsible for their anger.
Whatever the admixture of foresight and reaction, or of reason and hysteria, that lay behind the Decian measure, we also need to understand it as an early fruit of the equestrianisation of the empire – the increasing dominance of men of equestrian rank with bureaucratic careers, and the type of ‘governmentality’ that went with that. Decius undoubtedly remembered Caracalla’s citizenship edict, with its universalising rhetoric and still-more universalising impact in extending Roman law to the whole imperial population. Almost forty years after the Antonine constitution, a whole generation whose parents had been born as non-citizen peregrini was having to negotiate an adherence to Roman laws that were entirely outside the customary behaviours of their locales. Experienced administrators like Decius were aware of this before and after, and will have seen an ideological value in this universal conformity. We should thus think in terms of two governmental impulses, each functioning in parallel to the other: Caracalla’s grandiose edict had made it possible to imagine the whole of the empire as a single world in which all could and should do things the same way; at the same time, equestrianisation and the relentless expansion of governmental routine made the aspiration to enforcing such uniformity seem both possible and achievable.
The ad hoc nature of early Roman government had stemmed from a recognition that ruling the empire could be done most cheaply, peacefully and efficiently if local customs and local ways of keeping the peace and extracting tribute were maintained – and also from the incapacity of a polity that had developed out of the familia and clientela of Augustus to administer an empire of such scale according to uniform rules. Two centuries of expanding government and structural stability had altered that picture, so that what was once both inconceivable and unnecessary could now seem both possible and worth having. Whatever else we see in Decius’s edict to sacrifice – and it is clear that later emperors who ordered the deliberate persecution of Christians did see his edict as a model – we should also see it as an important stage in the development of a specifically late imperial approach to government, one that reached its fullest expression in the fourth century, and was thereafter sustained for a century or more in the west, and for a full three centuries in the east.
Decius’s assertiveness extended to more than matters of religious practice or the power of the state. He seems also to have been intent upon asserting himself as a more effective military leader than Philip had been. We have already seen how changes north of the Black Sea and east of the Carpathians had seriously altered the balance of power there, with the imposition on a settled agricultural society of a new military elite whose culture combined elements of central European dress and language with fighting styles and decorative schemata from the steppe. By 249, Roman military planners were conscious enough of these changes that Decius sent soldiers to the Bosporus to report on developments, the rump of the Hellenistic Bosporan kingdom still hanging on as the world changed around it. Along the lower Danube, meanwhile, Philip had exacerbated an already bad situation by stopping traditional subsidies to the Carpi and campaigning against them, alleging that they had broken the peace with Rome. But the withdrawal of imperial subsidy was rarely a successful technique of frontier management: Decius had personally to lead an army to Moesia Inferior against barbarians who are variously called Carpi, Borani, Ourogundoi and Goths by sources with some limited claims to authenticity, and are generically known as Scythians by the contemporary classicising authors. The confusion of names attested in the sources is a fact of prime historical importance: contemporaries on both sides of the frontier had little real idea of what was going on. Only with hindsight can we understand these events as the by-product of a developing Gothic hegemony in the whole region, in consequence of which the barbarian polities along the fourth-century Danube bore little traceable relationship to those of the third.
We need to resist the temptation to paste our disconnected pieces of evidence into a single, neat narrative, but the one thing we can say with certainty is that Decius faced a truly disastrous situation when he led his army into Moesia Inferior, where an invading army of Scythians had placed Marcianopolis under siege. Newly discovered fragments of the Athenian historian Dexippus have confirmed details known from much later sources whose authenticity has been doubted, and it is now clear that several groups of Scythians were led by commanders named Ostrogotha and Cniva, and that they inflicted heavy losses on the emperor himself at Beroea. News of this debacle probably triggered a short-lived coup in Rome by the senator Iulius Valens Licinianus, but that appears to have been suppressed almost instantly, perhaps by the praetorians, for no coins were struck in Valens’s name. Then, back in the Balkans at either the end of 250 or early in 251, Philippopolis did fall to Cniva’s invaders, after a short-lived usurpation by the governor of Thrace, T. Julius Priscus, had taken place there. Finally, in the first half of June, Decius confronted the invaders at Abrittus, north-west of Marcianopolis, in a pitched battle that went disastrously wrong. His opponents had dug themselves in across treacherous, marshy terrain ill suited to the mass infantry engagements at which Roman armies generally excelled. In a monumentally foolish move, Decius personally led his forces into battle across the marsh, where they became entangled and were massacred. The emperor himself died, as did his son Herennius Etruscus, and his body was never recovered. Later Christian authors could gleefully imagine this as the condign fate that awaited persecutors: ‘as an enemy of God deserved, he lay stripped and naked, food for the beasts and the carrion birds’.
News of the disaster reached Rome in mid June, as did word that the troops had raised C. Vibius Trebonianus Gallus, governor of Moesia Inferior, to the purple. He negotiated the withdrawal of the Scythians back across the Danube, though it appears that they took a large part of Decius’ imperial treasury with them: what had been a silver-based economy north of the imperial frontiers was rapidly transformed into one based on gold, with aurei of Decius seized at Abrittus its primary model. As soon as the Scythians departed, Gallus returned to Rome with great haste, arriving there in late summer. Gallus was an Italian of high rank – that we even know the name of his father, whose career was underway during the reign of Septimius Severus, is rare enough in this period to mark the family’s distinction. Despite ongoing trouble on the eastern front, Gallus’s hasty journey to Rome was a wise idea: the symbolic conciliation of senate and plebs remained a necessary duty if an imperial acclamation was to win any long-lasting acceptance, and Gallus issued adventus coins to commemorate his arrival. The senate swiftly consecrated the dead emperor as divus Decius, and Gallus at first accepted the latter’s younger son Hostilianus as a caesar alongside his own son Volusianus. It is unclear whether or for how long this situation lasted: there is some evidence that Decius suffered damnatio memoriae along with his sons, but it is not widespread enough for us to be sure and may instead reflect later Christian defacement of the hated persecutor’s name. Certainly, the surviving son Hostilianus was either put to death or died of natural causes before the year was out. At any rate, before the middle of 251, only two emperors were recognised by the senate, C. Vibius Trebonianus Gallus and his son, C. Vibius Volusianus, the caesar and princeps iuventutis. Senatorial approval or acquiescence having been vouchsafed, however, more pressing matters needed attention.
The revolt of Iotapianus had been suppressed either at the end of Philip’s reign or at the beginning of Decius’s, in circumstances that remain unknown to us, but a new revolt had broken out led by an Antiochene notable with the Syrian name of Mariades. By the time imperial troops loyal to the Italian regime got properly involved, Mariades determined to flee to Persia and seek refuge with Shapur. Shapur, in his victory inscription, claims that Rome had violated the peace that Philip made in 244 by sheltering the Arsacid heir Tiridates of Armenia, who had sought refuge with the Romans after his father Khusrau II had been assassinated, probably at Shapur’s instigation. In response, Shapur annexed Armenia, eliminating its Arsacid rulers and placing it under the rule of his son Ohrmazd.
Much more elaborate versions of these events survive in the Armenian tradition, with added folkloric elements of subversion, betrayal and massacre. One detail they preserve that may have some value is the extent to which the elder Tiridates and Khusrau had been irritants to Ardashir and then Shapur, promoting rebellions as far away as the Kushanshahr – which would help to explain the amount of time Shapur had to spend fighting in the east. Now, with the strategic mountain kingdom of Armenia in his possession and his eastern frontiers secure, Shapur again turned against the Romans, assisted by the rebel Mariades. In 252 or 253, the shahanshah attacked the Syrian provinces, not by the usual approach via Singara, Resaina and Carrhae, but by an unexpected route that caught by surprise the whole Roman garrison of Syria, which was being brought together in anticipation of a Persian campaign.
Shapur claimed that in this victory, at Barbalissos on the Euphrates, he defeated a full 60,000 Romans – a number that, even if wildly exaggerated, speaks to a major Persian victory. Antioch itself, along with other important Syrian cities like Hierapolis, were taken in this campaign or campaigns, Persian armies advanced as far as Cappadocia, and the Italian government of Trebonianus Gallus was powerless to do anything about it. Shapur deported a great many captives to Khuzistan, where he founded a new city that he called Veh Antiok Shapur (‘Better than Antioch Has Shapur Founded This’), a name that was later corrupted into Gundeshapur, an important town in the zone between the Zagros and the Tigris. Shapur’s motives went beyond military glory: Khuzistan was to become the economic engine of the Sasanian state, peppered with industrial centres, many populated by deported captives, specialising in production that enhanced the royal revenue and paid for further conquest. By suppressing older towns – perhaps most importantly Susa – with traditions of independent government and favouring instead royal centres governed directly by the imperial administration, Shapur inaugurated a long-standing policy that made the Sasanian dynasty the richest and most powerful of the ancient Near Eastern empires
In the third-century moment, however, the failure of the Roman emperors to act meant local easterners were compelled to take matters into their own hands. It may be at this time that Odaenathus, a nobleman from the flourishing caravan city of Palmyra, and an intermittently important figure in eastern affairs until his assassination in 267, came on the scene, but the chronology of events in this period is almost hopelessly tangled: some would attribute to this Odaenathus the defeat of a part of the Persian army that others attribute to one Uranius Antoninus of Emesa. That obscure figure’s full titulature – Iulius Aurelius Sulpicius Severus Uranius Antoninus – clearly proclaimed his kinship with the Severan house. He may have been a devotee and priest of Elagabal, the god of Emesa, and if so he was perhaps a relative of the family of Julia Domna. Given that he minted coins, which are the best evidence we have for his existence and his sphere of activity, it seems clear that Uranius was actively challenging Gallus’s claim to the imperial throne. In the later romanticised version of John Malalas, a sixth-century author who related a great deal of local Antiochene lore of varying reliability, Shapur himself is said to have been killed in this Emesene encounter, although the usurper Uranius has been blotted out and replaced by a noble priest named Sampsigeramus. While the story is clearly fiction, Emesa is conspicuously missing from the list of cities Shapur claims to have conquered in his inscription at Naqsh-e Rustam, and it seems very likely that he suffered a substantial defeat there. The fact that it took place at the hands of a local ruler, rather than an army even putatively loyal to the legitimate emperor, reminds us of the ongoing difficulty western emperors faced in retaining control of the east, in part because of their inability to protect the eastern provinces from Persian threats: it was not that they did not want to act, but that other threats constrained them. Thus Trebonianus Gallus could not, in 253, have acted upon the devastating news from the east even if he had wanted to, because he was facing the threat of another rebellion, closer to home and thus more immediately dangerous.
This new challenge was triggered by events that should now be numbingly familiar to the reader: a general on the frontier wins a battle against some barbarians, takes the purple, marches against the reigning emperor, and both would rather allow the provincials to suffer than let their rival gather momentum. As so often, the state of the sources leaves us guessing about details, but it is pretty clear that although Gallus had neutralised the group of Scythians that had taken Philippopolis and killed Decius, these were just one small part of a larger problem: because there were as yet no organised polities beyond the Danube with whom to treat, no single ethnic or tribal group responsible for the problems the empire was facing, the suppression of one challenge meant nothing for the status of many others. Thus the Scythians with whom Gallus had dealt might very well have respected their treaty with him, but that left dozens of opportunistic raiders happy to take advantage of imperial distraction and weakness. So it was that in 252 there were major seaborne raids into the Aegean from the Black Sea, presumably by barbarians who had joined with, or conquered and seized the resources of, the remaining Bosporan Greeks. The raiders were successful and elusive and, although it is hard to gauge the actual level of damage, some of what they did was quite shocking to contemporary sensibilities, for instance, the burning of the temple of Artemis at Ephesus in Asia Minor, one of the most famous shrines of antiquity.
Then, in 253, Moesia Inferior was invaded by a different group of Scythians, who this time met with effective resistance from a general named Aemilius Aemilianus, about whom precious little is known. He was probably born in the first decade of the century, starting a career under Severus Alexander and rising through the ranks, but that is about all we can say. Now he demonstrated a fact that was becoming obvious to everyone – the loyalty of provincial armies to a distant emperor was impossible to guarantee. Aemilian’s victorious troops declared him emperor and, though there was no way the Balkan frontier could yet be considered secure, he turned immediately to the march on Italy, necessary if he was to secure his hold on the purple. Crossing into Italy, he met and defeated Gallus at Interamna in Umbria. Gallus and the caesar Volusian were killed by their own troops and Aemilian won the recognition of the senate, who also acclaimed his wife Cornelia Supera the augusta. But Aemilian’s reign was to prove a short one. As soon as Gallus received news of the uprising, he sent word to Gaul to summon support from Publius Licinius Valerianus, more generally known as Valerian. The latter heard of Gallus’s death while still en route to Italy, and in Raetia he was proclaimed emperor in turn. In September 253, during the same campaigning season in which he had first been proclaimed, Aemilian was killed in battle with Valerian’s army at Spoletium in central Italy. The rapid turnover of emperors had become bewildering. The reign of Valerian would last somewhat longer, but would be no less traumatic for the high politics of empire.