Hadrian's Wall

by Adrian Goldsworthy

  Housesteads is probably the most famous and most often visited of all the forts on Hadrian’s Wall. It provides direct evidence that the decision to add forts to the Wall was made not only after the initial design but after substantial work had already been completed, because it required the demolition of a turret on the Wall. Built on the slope of a hill, Housesteads is unusual for a Wall fort in that it is oriented to the east, looking along the Wall, rather than to the north. It was laid down for a double-sized milliary cohort but in other respects is fairly typical in its layout. The Vallum was some distance away, near the bottom of the valley. The civilian vicus covered a wide area, reaching not only to the Vallum but even beyond. A number of temples were located south of the Vallum on a low ridge known today as Chapel Hill.

  As ever, the precise sequence of decisions and construction can only be guessed. At some point later in Hadrian’s reign, the stretch of Turf Wall for several miles to the west of the River Irthing near Birdoswald was replaced in stone, as was the fort itself. There may have been a substantial pause in construction, and similar halts occurred elsewhere, if not necessarily at the same time. There is evidence for heavy fighting under Hadrian, apart from the war at the start of his reign, and various dates have been suggested for this, notably c. 123–124 or later in the decade. The need to draw troops away for active campaigning is an attractive explanation for such lulls in building work, as is the idea that the Romans’ new fortification and stronger military presence provoked hostility in the people of the area, but there could be other explanations as well, for the army was given many tasks and had limited manpower. The extension of the observation system to the Cumbrian coast does suggest a real threat, presumably of raiders coming by boat and bypassing the Wall altogether.

  We shall examine each element of the Wall in more detail, but it is worth noting that in its original and modified designs, Hadrian’s Wall was a formidable obstacle to unauthorised movement. It was not meant to hinder the Roman army, which controlled its numerous crossing points. Compared to other Roman frontiers, the system is on a far grander scale. The line Hadrian ordered built in Germany was a simple stockade fence with a ditch behind it—an obstacle to slow down anyone trying to cross it, but nothing more.

  Three

  BUILDING AND MANNING THE WALL: LEGIONS AND AUXILIA

  THE ROMAN ARMY WAS BASED around the legions recruited from Roman citizens. There were thirty of these when the Wall was built, so the garrison of Britain represented one-tenth of the entire legionary strength of the empire. Each legion had a theoretical strength of around 5,000 men, almost all of whom were heavily armed infantrymen. These were organised into ten cohorts, each consisting of six centuries of eighty men who were commanded by a centurion. (The First Cohort was different, having five double-strength centuries and the task of guarding the precious eagle standard of the legion.)

  The legions were supported by the auxilia, men recruited from the provinces who would gain citizenship only at the end of their military service. The auxilia included both infantry and cavalry as well as specialised troops such as archers. They were not divided into legion-sized formations but formed in independent units similar in size to a legionary cohort. There were infantry cohorts of 480 or 800 men, mixed cohorts (cohortes equitatae) of 480 infantry and 120 horsemen or 800 infantry and 240 horsemen, and cavalry alae (all-cavalry regiments) of 512 or 768, respectively. The theoretical organisation of the various types of army unit in the second and early third centuries is given in Table 1.

  From Augustus onwards, there were at least as many auxiliaries as legionaries, and by the second century AD, there were significantly more. Overall, the garrison of Britain probably had a theoretical strength of more than 35,000 men, making it larger than the army in almost every other province of the empire. It was commanded by the provincial governor (legatus augusti), a senator, usually at least in his forties and who had already held command in a less prestigious military province.

  Although legionaries were Roman citizens, only a tiny minority of them came from Italy, and most were from the substantial communities of citizens living in Gaul, Spain, North Africa, and other provinces. Auxiliaries were organised into units with ethnic names—for instance, Spaniards, Gauls, Dacians, Thracians, Hamians from Syria, and Batavians and Tungrians from the Rhineland. When first raised, auxiliary units were recruited from these peoples. Over time, especially when a unit was posted far from its homeland, it is clear that men from other races enlisted and served. The language of command and administration in all units was Latin, although Hadrian encouraged units to keep ethnic war-cries. Britons served in distinct units but were also present in many other cohorts and alae stationed on the Wall.

  Some units had simple titles, such as cohors II Delmatarum eq., or the second part mounted regiment of Delmatae, which was recruited from modern-day Croatia and served at Carvoran in the third century; or cohors I Hamiorum sagittariorum, or the ‘first regiment of Hamian archers’, which had been at the same fort under both Hadrian, then on the Antonine Wall in the 140s and 150s, and then back at Carvoran under Marcus Aurelius. A few units acquired additional titles as rewards for valour or loyalty. Thus, there was the ala Augusta ob virtutem appellata, the ‘cavalry regiment called Augustan on account of its valour’, and the even more elaborately named ala Augusta Gallorum Petriana Milliaria civium Romanorum, or ‘double-strength cavalry regiment of Gauls, originally raised by Petrianus, the emperor’s own, of Roman citizens’. Granting citizenship to an entire regiment of soldiers was a rare reward for conspicuous service. The grant only applied to soldiers currently serving with the unit, but the unit title ‘of Roman citizens’ was kept permanently, even when all the men who had been made citizens had served their time and been discharged from the army.

  A Roman legionary in Hadrian’s day. Men from the three legions stationed in Britain provided the bulk of the manpower and technical skill employed in building the Wall and its bases. He wears the famous banded armour (lorica segmentata) and has additional protection from an iron helmet and his large semicylindrical shield, here protected by a leather cover. He is armed with the pilum javelin and the short gladius sword, worn on the right hip. His short-sleeved tunic is of a traditional pattern, but note the enclosed hobnailed boots, usually worn with socks. Breeches were a common addition for added warmth.

  In the second century AD, most recruits were volunteers, with only occasional periods of conscription, often when auxiliary units were raised as part of a treaty with a tribe or when a major war occurred. All had to be freeborn, for the army was not supposed to be a refuge for runaway slaves. Army pay was similar to the wages of a farm labourer, but it was regular, and soldiers were fed, clothed, and provided with better medical care than the poor in civilian life. Literate recruits stood a good chance of promotion, which brought better pay and conditions. All this was set against extremely harsh discipline and a minimum enlistment of twenty-five years. Even the food and clothes issued to soldiers came at a price deducted from their pay.

  Officers were drawn from higher up the social scale than the men they commanded, and the most senior were invariably citizens of considerable status. In charge of each legion was a legate (legatus legionis) who was a senator, usually in his early thirties, supported by a senior tribune in his late teens who was just starting a senatorial career. Equestrians, or ‘knights’ (equites), the next social class below senators, provided the bulk of the army’s senior officers. Typically, such men first commanded an auxiliary cohort, then became one of five junior tribunes in each legion, before receiving the command of a cavalry ala. The officer commanding an auxiliary cohort or ala was called a prefect (praefectus), except in the case of some more prestigious units, where his title was tribune.

  Senators were obliged to own land in Italy, although many, like Hadrian, had grown up in the provinces. Equestrians came from all over the empire and were men of considerable property and education. A few chose to serve instead as legionary centurions, b
ut most centurions in the legions and the auxilia lacked equestrian status. Even so, they had to be highly literate, and many appear to have come from the minor gentry of Italy and the provinces, serving all or most of their time in the army as centurions. Centurion was not a specific rank but a grade of officer whose responsibilities, pay, and status varied considerably. Much less is known about the background and careers of auxiliary as opposed to legionary centurions.

  The Roman army was very much the army of the empire, recruited from the provinces and led by men from the imperial elite. Soldiers usually served their entire twenty-five years with the same unit, and increasingly often that unit remained in the same province for generation after generation. In contrast, senatorial and equestrian officers moved from posting to posting around the empire, and many legionary centurions also served in a succession of different units in different provinces. This meant that northern Britain, like any frontier zone, drew men from far afield to serve there, the more senior usually only for short tours. In the second century AD, a significant minority of the Senate’s six hundred members had been to Britain, and many had been to the north. Although on the edge of the empire, the sheer size of the provincial garrison meant that Britain was always important, its governorship a rare honour reserved only for the most distinguished and loyal senators.

  HADRIAN’S WALL WAS BUILT BY legionaries, the work being divided up and allocated to sub-units that then commemorated what they had done in simple inscriptions: ‘From the fifth cohort the century of Caecilius Proculus built this’ (coh[ortis] V c[enturia] Caecili Procul[i]) or ‘The Sixth Legion built this’ (leg[io] VI V[ictrix]). Every legion included a large number of specialists and craftsmen, with other soldiers providing more or less skilled labour. An inscription from the Vallum recording work done by an auxiliary cohort suggests that this huge but simpler task was allocated to these soldiers, whose units generally contained a much lower proportion of specialists. In later years, auxiliaries would undertake more complex building projects, especially in the forts.1

  The Wall itself was a simple structure, and the reliance on legionaries to build it has more to do with these bigger units being easier to organise on a large scale than auxiliary units one-tenth of their size. Stone was quarried from the nearest available source, usually less than two miles (c. three km) from the site. Squared rubble—the term used by stonemasons for stones more or less roughly squared off rather than precisely worked—was used for the two facing walls, with a core of smaller stones bonded with clay. The curtain wall was not particularly well made by Roman standards, so the priority was to complete this immense structure as quickly as possible. Some of the stone used was of very poor quality, so that in patches it was barely squared off at all.

  In the original design, the western section of Hadrian’s Wall was built in turf and timber. Its construction used materials and techniques routinely employed by the Roman army, and this no doubt allowed it to be completed more quickly. This reconstruction of a short section based on the northern wall of a milecastle was built at Vindolanda several decades ago, in part to test how long it could last. It has survived the elements remarkably well. As with all reconstructions of the Wall, much is guesswork. Note that the timber wall and tower are set in a much wider turf bank.

  More care was given to the important features, such as turrets, buildings, and especially the gateways in forts and milecastles. These required considerably more skill than building the wall itself, which suggests that all the best craftsmen and teams were assigned to these tasks, leaving the bulk of other work to be performed by the less skilled legionaries and supervising engineers. Interruptions in the process may well have meant that one team laid foundations before moving on, with another or even several other parties completing successive stages of the building. There is clear evidence of haste in the last phase of construction, so that standards of work became lower in the hurry to finish things off. The gateways at Housesteads and Chesters, and the stone fort that replaced the timber one at Birdoswald, all made use of cruder, less finished stone in the upper courses. At Milecastle 37 subsidence caused a crack next to the northern gate that was made good but was still visible in the twentieth century when archaeologists temporarily cleared away stonework from a later period. Hadrian’s Wall in its early phases presents a strange mixture of grandiose imperial project alongside haste and corner cutting by the men tasked with turning this concept into reality. Yet whether in design or implementation, their willingness to adapt shows a desire to make the whole system work.

  In the end, the Wall’s effectiveness would depend on the troops stationed on and around it. As far as we can tell, every fort on the Wall was designed to house an entire auxiliary cohort or ala, rather than a mixed force drawn from detachments of several units. For much of the time, the picture may well have been less neat, with parts of the garrison posted away and other troops passing through and living in a fort for short or long periods, but it does show an expectation that in normal circumstances, each sector would be controlled by a distinct auxiliary unit. Thus, even though legionaries had built the Wall, it would be manned primarily by auxiliaries.

  Building work continued throughout Hadrian’s reign until his death in 138 (it was claimed that he was such a skilled astrologer that he was able to predict the day and time with precision). His successor was his adopted son, Antoninus Pius, who insisted that the Senate deify his ‘father’ in spite of its reluctance to do so. Deification was the normal honour voted to a good emperor on his death, and most senators were keen to show their dislike of Hadrian, now that he was dead and unable to strike back. Realisation that his successor was eager for the honour to be awarded soon persuaded them of the advantages of pleasing the new emperor. Unlike Hadrian, Antoninus Pius enjoyed the security of knowing that his claim to the imperial throne was unimpeachable, but he was also a very different character than his predecessor, having a mild nature and lacking both energy and passion. During his reign, no senator was executed, and his popularity never waned. Unlike Hadrian, Antoninus Pius spent his entire reign in Italy. Yet although he never saw Britain or its northern frontier, he had very different plans for his army there.

  As far as we know, two turrets were always included in between each milecastle in the original design of Hadrian’s Wall. The only exception is at Peel Gap, between Milecastles 39 and 40, where a third turret was added. It stood in a valley running across the line of the Wall, which would have been largely invisible to anyone on the higher ground to either side. This appears to have been a practical response to the local topography. In the modern era, this section of the Wall was excavated and has been conserved slightly differently from much of the rest. If you look closely, you can see Broad Wall foundations, with Narrow Wall constructed on top. Also visible is the third-century work demolishing the turret and separating what was left from the Wall itself.

  Four

  FRESH MINDS: ANTONINUS PIUS TO SEPTIMIUS SEVERUS

  WAR APPEARS TO HAVE ERUPTED in Britain early in the new emperor’s reign. The Greek writer Pausanias says that ‘the Brigantes… had begun a war, invading Gerunia, which is subject to the Romans.’ The Brigantes were a large tribe occupying much of what would become northern England. Most lived south of the Wall, although it is quite possible that it had cut through tribal territory, so that some Brigantes or closely related peoples now lived north of the Roman military zone. On the other hand, the name was often applied loosely by the Greeks and Romans to Britons in general, and perhaps especially the northern tribes. The district Pausanias calls Gerunia is unknown, and the name may be garbled, though some scholars prefer to believe that the passage talks of troubles in Raetia, on the continent, rather than in Britain. Yet disturbances in Britain, most probably well to the north of Hadrian’s Wall, make most sense of what happened, and perhaps tribes allied to Rome were attacked by neighbours.1

  The result was a major change in Roman deployment. Hadrian’s Wall ceased to be the main part of the provincial fro
ntier, and instead the army advanced into Lowland Scotland and constructed the Antonine Wall on the line between the estuaries of the Rivers Forth and Clyde. The new wall was built of turf and timber. From the start, forts formed part of the new wall, but unlike on Hadrian’s Wall, not all were designed to house a complete ala or cohort. Instead they were bases for composite groups mixing infantry and cavalry and troops from two or more separate units. There were also fortlets similar to milecastles along the rampart, but a line of turrets was not part of the system and was presumably seen as unnecessary. As with Hadrian’s Wall, there were outposts beyond the line of the Antonine Wall itself. Antoninus Pius was acclaimed as imperator or ‘victorious general’ by the Senate for a victory in Britain, the only time during his reign that he accepted this title, so the operations that led to the construction of the new wall were paraded as a major achievement and success by the emperor.

  The move stripped Hadrian’s Wall of its main function, and it was decommissioned. Gates from milecastles were removed, and the pivot stones in which they had been mounted were smashed, so that anyone could pass through. At the same time, earth was taken from the mounds on either side of the Vallum to help make causeways across it, at intervals of forty-five yards (forty-one m) or so along much of the line, although in some areas this was never completed or even started. It is possible that a handful of the forts on Hadrian’s Wall continued to be manned, albeit probably by substantially reduced garrisons. The rest were abandoned or maintained by very small numbers of personnel as holding units.