Hadrian's Wall

by Adrian Goldsworthy

  Antoninus Pius died in 161 and was succeeded by his adopted son, Marcus Aurelius, an earnest man and a philosopher, whose Meditations survives. It is a deeply personal book, made all the more fascinating because it was written by the ruler of most of the known world and deals with his struggle to live a good life, to rule well, and not to be corrupted by power struggles or too upset by criticism—‘it is the king’s part to do good and be ill spoke of.’ Death is a constant theme. Marcus Aurelius had a dozen children, most of whom died in infancy, and his sorrow preoccupied him, as did the prospect of his own end—‘in a short while you will be no one and nowhere, as are Hadrian and Augustus.’ Hadrian is mentioned elsewhere in the book, but this was not a work dealing with the details of government and events. Although some of the Meditations was written during the long and brutal campaigns Marcus Aurelius waged on the Danube, these conflicts are not discussed.2

  The same is true of the troubles that had broken out again in Britain early in his reign (161–180). These led to another change of policy, and around this time the Antonine Wall was abandoned and Hadrian’s Wall once again became the main feature in the frontier. The transition was carefully managed and prepared rather than the result of disaster or panicked retreat, for excavation of sites on the Antonine Wall reveal deliberate demolition rather than violent destruction. An inscription records building work at one of the forts on Hadrian’s Wall and also on the Wall itself in 158, while in the same year the abandoned outpost fort at Birrens was rebuilt. This was just the start of substantial restoration and remodelling of the Wall. Gates were replaced in milecastles and the remaining stretch of Turf Wall rebuilt in stone, which makes it clear that the original decision under Hadrian to build it in turf was not simply due to lack of suitable stone close to the line of the Wall. A smaller mound—known as the marginal mound—on the south side of the Vallum ditch may date to this period, blocking off the causeways so that the Vallum once again became an effective obstacle. Refinements continued for decades. One of the biggest changes was the construction of the Military Way, a road running between the Wall and the Vallum and connecting milecastles and turrets as well as forts. In the original design of the wall under Hadrian, a footpath had crossed the rivers. Now these were bridged to take the road across these obstacles.

  Northern Britain, showing the lines of Hadrian’s Wall, the Antonine Wall, and earlier frontier deployments such as the towers and other installations along the Gask Ridge. Tribal names are based on Greco-Roman sources, and the areas shown are approximate at best. The lighter shaded areas are high ground and mountains.

  In 180, Marcus Aurelius died and was succeeded by his eighteen-year-old son, Commodus, the only one of his sons to survive him. Unlike his father in almost every way, the new emperor soon rejected the dull routine of administration and the hardships of supervising the army on campaign. Instead he preferred to return to Rome and spent most of his twelve-year reign in and around the great city. The senator Cassius Dio, who knew both father and son, described Commodus as ‘not naturally evil, but simple minded.’ He and other senators were appalled by Commodus’s growing obsession with circus games and gladiators. The young emperor was fond of showing off his skill with a bow by shooting down from the imperial box in the Colosseum and slaughtering animals in the arena. On other occasions he fought display bouts as a gladiator, his opponents armed with blunt weapons to make sure that no accidental injury was inflicted on the emperor.

  Most of the troops stationed on Hadrian’s Wall were the noncitizen auxiliaries recruited from all over the empire. This picture shows a typical auxiliary infantryman in the second century AD. His bronze helmet is simpler than legionary patterns; he wears a mail shirt and carries a flat oval shield. He carries a spear that could be thrown or thrust and has a dagger on his left hip and a gladius sword on his right. Note the practical clothing. The popular stereotype of homesick Italians, shivering as they stood guard on Hadrian’s Wall wearing short tunics and open sandals, is wrong in every respect.

  Yet for all his antics, in the early years of Commodus’s reign problems in the provinces were dealt with competently by his subordinates. A Byzantine epitome of Dio’s text—the full version is lost—also tells us that ‘the tribes in that island [of Britain], crossing the wall that separated them from the Roman legions, proceeded to do much mischief and cut down a general together with his troops.’ The experienced Ulpius Marcellus was given charge of the province and ‘ruthlessly put down the barbarians of Britain.’ Coins commemorating a victory in Britain were issued in 184. These days, archaeologists are reluctant to interpret burnt layers on a site as the mark of violent destruction. This is a healthy reaction against earlier scholars who were inclined to assume fires were deliberate rather than accidental, and who were overfond of tying the remains as closely as possible to any event mentioned in our literary sources. Yet the pendulum may have swung too far the other way, and recently one scholar has argued that work at Corbridge and Halton Chesters (the closest fort on the Wall to Dere Street, the main north-south road) does suggest a layer of burning on each site dating to around the 180s that was probably deliberate, making it quite possible that they attest to the destruction caused by this war.3

  Dio does not say whether the commander killed by the invaders was a legate commanding one of the legions in Britain or the provincial governor, the legatus augusti himself in charge of the whole province. Ulpius Marcellus is attested as legate of Britain in 178 on a document recording the honourable discharge and grant of citizenship to auxiliary soldiers, which means that he arrived to govern Britain under Marcus Aurelius and so was not sent by Commodus. This has led to the suggestion that it was his successor as legatus augusti who died in the fighting or was recalled for his failure, and that Marcellus was sent back to the province and served a second spell as governor—a very rare occurrence, but not impossible. On the other hand, if the officer killed was a legionary legate, then Marcellus may already have been in the province rather than having been sent by Commodus to deal with the crisis. As usual, a brief description in one of our sources raises almost as many questions as it answers. We simply cannot say which peoples were involved in the war or why it broke out, nor can we trace its course, in particular the circumstances in which the attackers overran the Wall.

  MARCUS AURELIUS’S REIGN HAD SEEN the start of the Antonine plague, an unidentified epidemic that was carried by troops returning from a successful invasion of Parthia and later spread throughout the empire. Cities and army bases were especially prone to outbreaks, as both brought large numbers of people into close contact. At the same time, there were arduous campaigns on the upper Danube. Both factors may well have depleted the garrison in Britain, so that it was less able to dominate the northern frontier. Later in Commodus’s reign, discontent among the troops in Britain led to mutiny on more than one occasion.4

  Commodus was strangled in his bath on New Year’s Eve 192. His successor lasted only a few months before he was murdered by the praetorian guardsmen, who were angered when he failed to pay them all of a promised bounty. The civil war that followed was fought between the governors of the three biggest military provinces in the empire: Britain, Upper Pannonia on the Danube, and Syria. The legate of Britain was Decimus Clodius Albinus, and he drew off a substantial part of the garrison along with troops from other provinces to amass the huge army that he led to defeat at Lugdunum (modern Lyons) in 197. The victor was Lucius Septimius Severus, a native of Lepcis Magna in North Africa but a Roman senator just like the ‘Spanish’ Trajan and Hadrian.

  Severus was a hard-nosed, ruthless politician who realised that he might easily face a challenge from another senator in charge of an army. Soon he named his infant sons as co-rulers, in an effort to show that the dynasty would survive whatever happened to him. He also declared himself the son of Marcus Aurelius, even though there was no basis for the claim. Although feared by senators, Severus was never popular and relied openly on force to maintain his power. He reinforced
the already substantial total of guard units in Rome with a newly formed legion stationed a short distance away. During his reign he fought two major foreign wars—leading an expedition against the Parthians and then later against the Caledonians—and it was no coincidence that this gave him a chance to confirm the loyalty of the armies in Syria and Britain, both of which had fought for his enemies in the civil war.

  That is not to say that either war was unnecessary. At some stage during the early years of Severus’s reign, there was trouble in the north of Britain, centering on a previously unknown group called the Maeatae, who may have lived near the River Forth and the line of the abandoned Antonine Wall. Breaking their old alliances with Rome, many of the Caledonian tribes joined the Maeatae. Severus’s newly appointed governor judged them too strong to fight with the weak garrison of the province and, according to Dio, ‘was compelled to purchase peace from the Maeatae for a large sum,’ receiving some hostages as surety. Some of the hoards of silver denarii found in Scotland and dating to the turn of the second to third centuries AD may represent subsidies paid at this time to buy peace.5

  The peace did not prove permanent. There was fighting in 207, and in 208 Severus came to Britain and spent the next three years campaigning in the north. He brought substantial reinforcements, including detachments (or vexillations named after the flag they carried, or vexillum) from a number of legions. Marching camps in Scotland associated with these campaigns are some of the largest known in Britain, and for a while, bases were re-established beyond the Forth-Clyde line. The Caledonians proved elusive, avoiding pitched battle and instead fighting by ambush and raid, making the most of their knowledge of the country. Dio says that they used sheep and cattle as bait to lure the Roman soldiers into ambushes. Severus ordered brutal reprisals and eventually received the submission of the tribes, but a renewal of war prompted him to order even more savagery in the hope of terrifying the tribes into surrender. The emperor’s health was poor, so he had to be carried in a litter for much of the time, and the rigours of campaigning used up his last strength: he died in York in 211. In spite of the advance to the far north, Septimius Severus’s reign saw a good deal of work restoring and repairing Hadrian’s Wall, and there is no hint of any plan to abandon it in favour of a more northerly line.

  A painting of Vindolanda c. AD 200, the fort lying on the same ground as the stone remains visible today. Note the rows of barrack blocks and the central range of buildings, with granaries just to the right of the gateway, principia, and praetorium. Although much like other forts, the rows of Iron Age–style round houses inside the fort on the left are unusual and not yet understood. In the foreground we have part of the vicus settlement, part of which is surrounded by an earth rampart. In the bottom right is the mansio, the inn or waystation for those travelling on government business. To the left, the building with the apse is the bathhouse. Remains of both these buildings can be seen on the site today. At the top left in front of the fort is the Stangegate, the road running east-west, which predated the construction of Hadrian’s Wall.

  Our literary sources are silent for most of the rest of the third century, so it is impossible to say how long the tribes remained cowed by the emperor’s demonstration of Roman might. Late in the century, the name Picts is first attested for some of the peoples of Scotland. Most scholars assume that groups already living in the area came together under some loose confederation—a trend shared with other tribal societies living beyond the frontiers of the Roman Empire. The name means ‘the painted’—as early as Caesar’s day, Roman commentators talked of the Britons’ fondness for tattooing and painting themselves—so it may simply be a slang term imposed by outsiders and may long predate its first appearance in our meagre sources. We simply do not have enough evidence to say whether or not the tribes of the north changed much in their social and political organisation.

  Yet for the Romans, this was certainly an era of change. The dying Severus made his two sons joint successors, supposedly giving them the grim advice, ‘love one another, indulge the soldiers, and despise everyone else.’ Within a year, the older brother had murdered his younger sibling. He was in turn killed in 217, stabbed to death by a centurion in his own bodyguard, and this was the start of generations of civil war unlike anything in Rome’s past. Long-term stability never really returned, and the Western Roman Empire would fade into history with the deposition of its last emperor in 476. For much of the time, Britain escaped the worst of the internal chaos, although it must have seen frequent troop withdrawals as soldiers were drawn off to back leaders in the long succession of civil wars.

  Some of the power struggles and attendant wars lasted for many years, so that several rival emperors ruled simultaneously, each controlling only a part of the whole empire. Starting in 260, for more than a decade a succession of emperors ruled an empire consisting of the Rhineland, Gaul, much of Spain, and Britain. Scholars conventionally refer to this as the Gallic empire, but as far as the emperors were concerned, they were the legitimate emperors of Rome, and it was simply that they were unwilling or unable to bring the other provinces under their sway. Eventually, weakened by the same sort of internal power struggles that ravaged the whole empire, the western provinces were brought back under the control of central authority by the Emperor Aurelian. In 286, Carausius, a commander sent to deal with seaborne raiders in the English Channel, declared himself emperor, and once again Britain became part of a separate regime that still considered itself Roman in every important respect. At least one major attempt by the emperors controlling the rest of the empire to invade and reclaim the island was defeated, but in the end, Carausius was murdered by a rival, who was in turn defeated by one of the emperors from the ‘main’ Roman empire. In both these periods, Britain was Roman culturally and politically—it was just that for a while there was more than one Roman empire. Inevitably, this must have disrupted the army’s system of administration, recruitment, promotion, and supply, but our meagre evidence makes it impossible to trace this dislocation and gives a false impression of continuity and stability.

  By the early third century, most of the forts on Hadrian’s Wall were garrisoned by units that would remain in place for the remainder of Roman rule in Britain. This may mean no more than that their headquarters and records were based there and that maintaining this arrangement was seen as sensible. As always, it is harder to say where the majority of the soldiers in each unit actually were at any time—and indeed how strong the unit was in practice and not simply in theory. Probably around the 230s, the barrack blocks in several forts were reduced in size by roughly half, which does suggest that it was expected that fewer men would be present. This most likely reflects a reduction in the nominal strength of cohorts and alae. Other, less regular units appear at the same time at Housesteads and elsewhere, so perhaps the smaller regiments were augmented by less formally trained warriors from other frontiers, but the result may still have been to reduce the total number of troops available in the area of the Wall.

  Sometime late in the second or early in the third century, many turrets on Hadrian’s Wall were decommissioned and their entrances blocked. They were subsequently demolished and not replaced. This was especially true in the central section, where only a few remained in use. It looks as if most or all of the smaller installations on the Cumbrian coast fell out of use sometime before this, and limited excavation has shown no trace of reoccupation of towers and mile-fortlets after the return from the Antonine Wall. However, the forts along the coast continued in full use, which suggests that a significant military presence was still felt necessary in this area.

  Around the time that the turrets on the Wall fell out of use, other changes were made. In many milecastles, the northern gateway was walled up, sometimes leaving only a narrow doorway permitting access to the outside, sufficient for maintenance of the Wall. Wherever this occurred, the causeways across the ditch were removed or blocked. Yet in spite of these changes, the milecastles continued to be used by t
he army, and it is now time to look in more detail at the physical components of Hadrian’s Wall as we try to understand how it was intended to function and how this functioning developed over the years.

  Five

  THE ANATOMY OF HADRIAN’S WALL

  THE NORTHERNMOST ELEMENT OF THE barrier was the ditch, which ran along most of the frontage of the Wall, with a few exceptions, usually where natural features rendered it either impractical or superfluous. More often than not it was V-shaped, some nine feet (2.74 m) deep and twenty-eight feet (8.53 m) wide. The spoil from the ditch was piled on the northern side and then smoothed into a gently sloping mound, which effectively made the ditch deeper but offered no cover to any attacker. Although in older studies, diagrams of the ditch have a rectangular trench at the bottom, it is now clear that this was rare. In many respects the ditch was similar to those surrounding Roman forts. Causeways were provided in front of fort gateways and, in the original design, appear to have been standard in front of milecastles. As we have seen, most of the latter were removed or blocked in the late second or early third century.

  However, it should not surprise us that there was considerable variation over the course of an obstacle stretching some eighty Roman miles. Some parts of the ditch are more U-shaped than V-shaped. In sections where the ground sloped down away from the Wall, there are stretches where the mound was made steeper on the inside—creating a counterscarp in the language of early modern fortifications. Another device employed where the natural gradient was insufficient was to dig out a steeper inner side of the ditch, at times creating a one-sided ditch. Some of these alternative approaches provide a shallower trench, but all perform the same basic function. The ditch was intended to slow down anyone trying to get across the Wall from the north, while keeping them in plain sight of the defenders. Variations in its design appear most common in the craggy central sector of the Wall. Famously, at Limestone Corner near Milecastle 30, the party digging the ditch left a number of granite boulders in place. Marks from their tools as they tried to drive wedges into the stone and break the pieces up can still be seen today, for at no point in the following centuries did anyone bother to complete the work.