Rome and the North Britons

Reconstruction of the Antonine Wall at Callendar Park, reproduced courtesy of Falkirk Museum and M. J. Moore DA FSA Scot

Consolidation of the Frontier

If the imperial authorities hoped that the Antonine Wall would bring a period of stability to Roman Britain, their optimism was dashed when trouble broke out among the northern tribes in 154 or 155. Which tribes were involved is a matter of debate, as is the question of how much disruption was caused. It is possible, for instance, that the unrest was confined to communities living north of Hadrian’s Wall, or that these were joined by neighbours in Dumfriesshire, or even that the main troublemakers lay further north in Caledonia. Whatever the location of the uprising it was put down by Julius Verus, governor of Britain, and special coins were minted to celebrate the restoration of order. In the next few years, however, a decision was taken to abandon the Antonine frontier and withdraw to Hadrian’s Wall. The presence of troublesome natives in the region between the two walls may have influenced the decision, but other factors, such as the strain on military resources, could have played a bigger role. When the withdrawal commenced in 158, it evacuated the Antonine line but stopped short of abandoning the region between the two walls. Some of the intervallate forts were even refurbished at this time. Buildings destroyed by fire at Birrens, known as Blatobulgium (‘The Flour Sack’) because of the distinctive shape of the nearby Burnswark Hill, were once thought to have succumbed to the native uprising of AD 154/5, but were more likely to have been demolished by the fort garrison during a makeover.

Before 160 the Antonine Wall was recommissioned and its soldiers came back to the forts, if only for a brief time. Their return to the Forth–Clyde isthmus was temporary and did not outlast the end of the decade. Trouble flared again in the early 160s, soon after the accession of Marcus Aurelius as emperor. A Roman general with the portentous name Calpurnius Agricola was ordered to quell it. The contemporary sources do not identify the culprits, who were either rebellious Britons on the northern frontier or Caledonian raiders from the lands beyond. Whoever these troublemakers were they were defeated and a semblance of stability returned. Roman sources describe another outbreak of hostilities in 169 when unidentified Britons caused trouble somewhere in the North. A war was seemingly averted by nipping the unrest in the bud, but, by 170, the Antonine Wall was again evacuated when Marcus Aurelius needed reinforcements for a campaign on the Danube. This time the troop withdrawals were intended to be permanent and many forts sustained deliberate demolition of buildings and defences. The turf frontier was abandoned, the Stirlingshire forts were left empty and the imperial boundary shrank back to Hadrian’s Wall. Some forts in the intervallate region remained in use, but these were engulfed in 181, during the reign of the emperor Commodus, when the Caledonii swept down from their Highland fastnesses to plunder the wealth of the Roman province. A high-ranking general marched out to meet the marauders, but he and his troops were slain. The ensuing wave of destruction left several forts along Hadrian’s Wall in ruins and spelled disaster for vulnerable outposts such as Newstead. Commodus, son and successor of Marcus Aurelius, dismissed the hapless governor of Britain and appointed a more effective replacement. The new governor, Ulpius Marcellus, defeated the Caledonii and restored control before following up his victory by making changes to troop dispositions in the interval-late region. Some forts were rebuilt and regarrisoned, but others, including Newstead, were condemned to dereliction. By the end of the second century, only a handful of outposts north of Hadrian’s Wall remained in use, their soldiers providing a token military presence in a region now regarded as a buffer-zone between the imperial province of Britannia and the badlands of Caledonia. The outposts lay in the south of the intervallate region, in lands nominally given over to native rule but under the watchful eye of Rome. Beyond them, in a broad band of territory encompassing Clydesdale and Lothian, the North Britons retained a measure of independence under the authority of their own leaders. It is likely that this arrangement was monitored by the Roman army during ceremonial events and tribal assemblies at specific sites called loci. The Latin word locus simply means ‘place’, but in the context of barbarian tribes bound in clientship to Rome these ‘places’ may have held administrative and diplomatic significance. Each of the four major groupings of North Britons had one or more loci within its territory, some being centred on sacred stones of immense antiquity which had long been used for ceremonial purposes. A public gathering at a locus would have given Rome an opportunity to remind the natives of their obligations to her Empire. How much autonomy was actually delegated to the intervallate Britons is unclear, but the surviving outpost forts were doubtless a constant reminder of imperial authority. At Birrens the Roman garrison used a native hillfort at nearby Burnswark for target practice by bombarding its decaying ramparts with catapults, an exercise which may have served the dual purpose of providing in-house artillery training as well as discouraging dissent among the North Britons. The latter thus approached the third century sandwiched between two implacably hostile forces: the Empire to the south and Caledonia to the north. Treaties forged in the aftermath of troop withdrawals from the Antonine Wall created an uneasy peace between the protagonists, but neither side, still less the Britons caught in the middle, expected it to last. It was little more than a temporary respite, a breathing-space, before a new round of raiding and retribution began.

Native communities in the land between the two Roman walls dwelt in the shadow of a conquering power. Their fellow-Britons living south of Hadrian’s Wall in what is now northern England were subjects of the Empire and, by the third century, had grudgingly or willingly accepted the situation. Earlier revolts by the Brigantes had been brutally crushed and were never to be repeated. Acceptance of subjugation was an easier option, even if it meant a loss of pride and a tax obligation to the imperial treasury. North of the Hadrianic frontier the Britons of the intervallate region remained nominally independent while acknowledging some measure of Roman authority. Unlike their Brigantian neighbours they continued to be ruled by their own leaders but these had presumably forged long-term treaties with Rome.

South of Hadrian’s Wall, the Brigantes and other conquered Britons experienced the full impact of the Roman occupation. The native upper classes, comprising the major landowning families, had watched their privileged status slip away after the conquest. Their lifestyles had collapsed as soon as Rome dismantled the old economic networks. Tithes of agricultural produce formerly rendered to local headmen were now collected by imperial tax-gatherers, while a strict prohibition of civilian military activity brought an end to tit-for-tat raids by predatory bands of Britons upon their neighbours. The resulting net loss of plunder severed the native upper class from its traditional methods of amassing surplus wealth through the acquisition of cattle and slaves. In such circumstances the neutered elites of Brigantia had little choice but to accept new roles delegated to them by the imperial administration. Some were probably allowed to retain a measure of authority in local contexts, as leaders nominated by Rome to oversee districts where their ancestors had once held substantial power. Such folk would have become more or less Romanised, maintaining their elevated status by exploiting opportunities for social advancement in the northern military zone. Some, no doubt, were allowed to remain on their ancestral estates and would have continued to receive tithes from tenant farmers.

A wholly new type of civilian settlement, the vicus, appeared in the wake of conquest. The typical vicus was a small village established outside the main gate of a Roman fort and along the primary access road. It tended to attract entrepreneurs seeking business opportunities at places where large numbers of military folk had disposable incomes. By definition, the vicus owed its existence to the presence of the fort and was wholly dependent on the patronage of soldiers. Its inhabitants, known as vicani, were generally a mixed bag of individuals drawn from local native communities and from places further afield. Some manufactured clothes, shoes or craft goods in small workshops, while others established taverns and hotels. Female vicani included wives and girlfriends of the garrison, their residence outside the fort initially being a requirement of the Army’s prohibition on married soldiers until Septimius Severus changed the law. In reality, even before the Severan reform, the authorities routinely turned a blind eye to liaisons between soldiers and native women, many of whom bore sons who eventually succeeded their fathers in the garrison.

North of Hadrian’s Wall, the much briefer occupation of Roman forts made the vici a fleeting addition to the landscape. Even when the Antonine Wall provided a temporary screen against Caledonian incursions, the intervallate region was not a place where civilians could put down roots outside a fort. Thus, while some vici south of Hadrian’s Wall thrived for two hundred years or more, in the lands further north a long period of habitation for vicani was out of the question. No fort north of today’s Anglo-Scottish border was permanently garrisoned after the end of the second century, a statistic which helps to explain why archaeologists have identified so few vici in Scotland. One of the few examples unearthed by excavation is a large village clustering outside the east gate of the fort at Inveresk near Musselburgh in East Lothian. Another has been discovered at Carriden, known to the Romans as Veluniate, a fort perched on the eastern extremity of the Antonine Wall overlooking the Forth estuary. The vicani at Carriden were a community of sufficient stability and cohesion to be granted a measure of self-government by the military authorities. However, neither of these settlements endured for long. They were wholly dependent on their forts and disappeared when these were abandoned.

Caledonii and Maeatae

Beyond the Antonine Wall lay the enemies of Rome: the Caledonii and their neighbours. During the northern campaigns of the second century, the Empire’s relationship with these barbarians was characterised by raid and counter-raid across the borderlands around the Firth of Forth. This region became a volatile conflict zone while Roman troops still garrisoned the Antonine forts, and likewise in the years following its final abandonment in the 160s. Hostilities continued until Rome forged treaties with the main barbarian groups at the end of the century, probably by paying them to stop raiding. At that time the Caledonii were still the main threat, but another people, the Maeatae, were recognised as an equally belligerent foe. Roman writers located the Maeatae immediately north of the Antonine Wall in what are now Stirlingshire and Clackmannanshire. They may have been a fusion of smaller groups on the model of the Caledonian ‘confederacy’ further north. Some historians wonder if these political fusions may have occurred because an aggressive foreign power held sway south of the Forth–Clyde isthmus. They see Rome’s occupation of the southern part of Britain as a catalyst for political developments in the North. In this scenario the creation of large tribal confederacies is viewed as a logical progression arising from the proximity of large numbers of Roman troops. An alternative theory sees amalgamation as an outcome of conflict between neighbouring communities, rather than as a voluntary or co-operative response to the threat of Roman invasion. Indeed, Rome might even have been responsible for creating tensions by favouring some native groups while neglecting others. Thus, it is possible that the pro-Roman queen Cartimandua might not have been a ‘pan-Brigantian’ sovereign after all, but merely a local ruler who exploited imperial patronage to impose her will on other Pennine peoples. By applying this model further north, we might envisage the Caledonii not so much joining with their neighbours as subjugating them by force. Such a process may have placed the Caledonian leadership at the head of a large, powerful amalgamation of tribes in a region centred on the valley of the River Tay. If this is what happened, then the Maeatae may have similarly seized the initiative among their own weaker neighbours.

High on a shoulder of the Ochil Hills, commanding a wide vista across Stirlingshire and the Firth of Forth, stood the great oppidum or tribal centre of the ancestors of the Maeatae. This stronghold may already have been abandoned when the Maeatae themselves first came to Rome’s attention, but it remained an imposing feature in the landscape. Its ancient name is unrecorded, but the hill on which it stands is known today as Dumyat, a name deriving from Gaelic Dun Myat (‘The Fort of the Maeatae’). Five miles south-east, and a little to the south of the modern town of Clackmannan, stood an unshaped boulder venerated in pre-Christian times as a sacred stone. In the medieval period this monument became known as King Robert’s Stone after its role in a folktale about Robert the Bruce, but its original name was Clach Manonn (‘The Stone of Manau’). The stone’s proximity to the heartlands of the Maeatae suggests its adoption by their forefathers as a venue for sacred rites and public ceremonies. It now sits on top of a pillar beside the old tolbooth in Clackmannan and has given its name to the town.

The Maeatae make their first appearance in the historical record around the year 200. At that time, according to the Roman writer Cassius Dio, they overturned a treaty with Rome and mustered their forces for war. They chose the right moment, for substantial numbers of Roman troops had recently been withdrawn from Britain by Clodius Albinus, an ambitious governor who hoped to set himself up as emperor. Seeking to exploit the situation, the Maeatae crossed the abandoned Antonine Wall to rampage southward, wreaking havoc wherever they went. To make matters worse, the Caledonii were preparing to break their own treaty with the Empire by joining the assault. In a desperate bid to avert a major crisis the newly appointed governor of Roman Britain, Virius Lupus, tried to placate the Maeatae with a substantial payment. The offer was accepted: the raiders went home and released a small number of Roman prisoners. But peace did not last and a new spate of raiding began. This time, no bribe was forthcoming from the imperial treasury. What the barbarians received instead was a full-scale assault. In 208, the warlike emperor Septimius Severus arrived in Britain to deal personally with the situation on the northern frontier. With him came his sons, Geta and Caracalla, two young men rescued from the sleaze of Rome by a father who regarded the Forth borderlands as a somewhat more wholesome environment. Assembling a large army, Severus marched north to hammer the Maeatae into submission and to discourage the Caledonii from joining them. His strategy seemed to work: he received pledges of peace from the barbarians and returned to his base at York. In 210, however, the Maeatae again reverted to their old ways. They may have heard a rumour that Severus was sick and unable to leave his bed. He was indeed too ill to command a new campaign, but, despite his infirmity, he had no intention of letting the enemy run amok. Leadership of the counter-attack was delegated to Caracalla who unleashed upon the Maeatae a harsh retribution. He arrived in Stirlingshire with a clear instruction from his father to slaughter the natives and to leave none alive. Until this point, the Caledonii had merely observed from the sidelines, but new tales of Roman savagery towards their neighbours brought them swiftly into the fray. They had another incentive to confront the invader, for Severus intended to build a massive fortress at Carpow at the mouth of the River Earn on the southern edge of their heartlands. The new base was designed to accommodate an entire legion and, when completed, would have posed a major threat to native ambitions. A prolonged and bitter conflict seemed unavoidable until fate intervened to remove Severus from the equation. In February 211, at his military headquarters in York, he finally succumbed to illness. Caracalla became emperor, but no longer shared his father’s enthusiasm for the northern campaign. He saw little gain in resuming it: the fighting was hard, the short-term rewards were meagre and the prospect of a lasting solution looked increasingly remote. Moreover, the drain on military resources was becoming acute and difficult to justify at a time when other parts of the Empire demanded urgent attention. Foremost among Caracalla’s anxieties was a bitter rivalry with his younger brother, Geta, whose growing influence at the imperial court was an irritation. Caracalla therefore called a halt to the war, made peace with the Maeatae and Caledonii and relinquished any serious claim on their lands. He returned to Rome to assert his authority and, within a few months, masterminded his brother’s assassination. Meanwhile, in northern Britain, the forward bases occupied during the Severan campaign were evacuated. Construction at Carpow was halted and the soldiers withdrew. A token military presence lingered at Cramond on the Forth until it, too, was abandoned in the 220s. The imperial frontier again retreated to Hadrian’s Wall, leaving only four outpost forts in the lands beyond: Risingham and High Rochester in the east; Bewcastle and Netherby in the west. With this retreat the Roman adventure in the Highlands finally came to an end.

The Emperor, his sons and the military leadership wintered in York. Sadly for them however the terms which had so satisfied the Romans in AD 209 were not so agreeable to at least the Maeatae as in AD 210 they revolted again. The Caledonians predictably joined in, and Severus decided to go north again to settle matters once and for all. On this occasion he’d obviously had enough of the troublesome Britons, giving his famous order to kill all the natives his troops came across.

This second campaign re-enacted the AD 209 campaign exactly, though this time solely under Caracalla as Severus was too ill. It was even more brutal than the first as afterwards there was peace along the northern border for four generations afterwards, the longest in pre-modern times. Archaeological data is now emerging to show this was because of a major depopulation event, indicating something close to a genocide was committed by the Romans in the central and upper Midland Valley.


Some Roman writers poured scorn on Caracalla’s readiness to let the barbarians off the hook, but his treaties held firm and ultimately proved the doubters wrong. The third century passed in relative peace. No new outbreaks of trouble on the northern frontier are known from the surviving literary sources. Only in the century’s last decade did the situation once again grow volatile. In 297, the poet Eumenius referred to a people called Picti (‘Picts’), whom he named alongside the Irish as enemies of the Britons. He did not say where they came from, but they plainly lived outside the Empire. Their location was made clearer by an anonymous writer of the early fourth century who referred to ‘the woods and marshes of the Caledones and other Picts’. This clearly identifies the Caledonii of earlier times as a component of the Picti. It also shows that Perthshire, the ancient Caledonian heartland, must have lain within Pictish territory. Later in the fourth century, the historian and ex-soldier Ammianus Marcellinus regarded the Picts as a fusion of two distinct peoples, the Verturiones and Dicalydones. The latter name relates in some way to Caledonii and indicates that this ancient grouping still functioned as a political force three hundred years after the Agricolan invasion. The Verturiones are previously unknown, but their name connects them to Fortriu, an area of importance during the second half of the first millennium AD. In the nineteenth century, the Scottish antiquary William Forbes Skene equated Fortriu with the later earldom of Strathearn and Menteith. This identification remained largely unchallenged until 2006, when its weakness was highlighted in a groundbreaking paper. Fortriu is now regarded as a more northerly territory centred on Moray. In another recent development, some historians have adopted the adjective ‘Verturian’ when referring to the land and people of this region.

Picti means ‘Painted People’ or ‘People of the Designs’. When and why this name originated are questions to which several plausible answers can be offered. So far, no consensus has yet been achieved. The name may be derived from, or related to, a collective term used by the Picts of themselves, but it is equally possible that no such term existed until the Romans began to distinguish the peoples of northern Britain from one another. Sadly, the Pictish language vanished after c.900 and, as no Pictish writings have survived, there is now little hope of ascertaining whether or not a native precursor of Latin Picti ever existed. Historians are left instead to muse on the nature and purpose of the ‘designs’ that gave rise to the name. Did the Picts tattoo their skin, or did they merely daub their bodies with warpaint? Tattooing was regarded as archaic and primitive by the Romanised Britons living south of Hadrian’s Wall, but it possibly lingered as a custom further north. If so, its continuing use far beyond the frontier might explain why the poet Claudian, writing in the late fourth and early fifth centuries, referred to Roman soldiers observing the decorative body-art of slain Pictish warriors.

Whatever the origin of their name, the Picts posed a major threat to Roman Britain throughout the fourth century. They were a numerous people whose lands encompassed a broad swath of territory stretching from the Western Isles to Fife and from Shetland to the Ochil Hills. Within this large area many communities shared cultural traits we now regard as essentially ‘Pictish’. They shared a common language similar to, and no doubt once indistinguishable from, the language of the Britons. On a political level, however, the Picts were not a single entity but a patchwork of separate groups, each of which was probably ruled as a small kingdom. In early times, when they first came to Rome’s attention, their most frequent foes were likely to have been fellow-Picts rather than people living south of the Forth–Clyde isthmus. Much of the slave-raiding and cattle-reiving undertaken by Picts in Roman times was surely conducted within their own homelands. Ambitious leaders would have had little incentive to act in unison against an external power unless persuaded or coerced to do so. Thus, although the notion of pan-Pictish unity might have simplified matters for Roman chroniclers, we should not feel tempted to run too far with it. Temporary pooling of military forces in response to Roman aggression perhaps occurred from time to time, but the Picts were not a homogeneous group. Their default political framework was rooted in local allegiances rather than in abstract concepts of nationhood. Although this pattern began to change in the sixth century, with the emergence of one or more Pictish overkingships, the marauding bands of ‘Painted People’ who troubled Roman Britain in 297 were almost certainly not acting in unison.

What distinguished the Picts from other indigenous peoples of the British Isles? The simplest answer to this question is that Pictish culture must have been unique, distinctive and recognisable to outsiders. It was sufficiently distinct for Roman writers to differentiate the Picts from the Britons and the Irish. All three were part of a Celtic cultural zone, but, despite this shared heritage, they each exhibited certain traits that set them apart from one another. One important difference was language: the Picts and Britons spoke variants of a Brittonic language of the ‘P-Celtic’ group, while the Irish used Goidelic or Gaelic speech which modern linguists define as ‘Q-Celtic’. The Pictish and British varieties of Brittonic represented separate dialects which, although mutually intelligible, may have sounded quite distinct when spoken. The date at which the two diverged is unknown but their separation perhaps began in Roman times, when the influence of Latin south of Hadrian’s Wall might have made northern dialects seem barbarous and different. By 297, when the Picts emerged into recorded history, it is possible that their speech already sounded sufficiently different to set them apart.

In ethnic terms the Picts of the third and fourth centuries were simply the most northerly of the Britons. There is no doubt that they were a ‘Celtic’ people. Like their southern cousins they had been exposed to Celticisation during the first millennium BC when cultural influences from Continental Europe spread throughout the British Isles. Unsurprisingly, the Pictish landscape contains a number of ‘Celtic’ features, the most visible being hilltop fortresses defended by concentric walls of unmortared stone laced with timber. Certain other structures are not found elsewhere in the Celtic world, or are encountered only rarely, and seem to be indigenous to the Pictish zone. Of these, the best-known are the brochs, the enigmatic towers found all over the Pictish area, with a major concentration north and west of the Great Glen. Isolated examples in southward districts such as Lothian suggest that the design was not confined to what is usually regarded as the main Pictish zone. As previously noted, archaeological study has dated their main occupation phase to the period 500 BC to AD 100 which means that they had probably fallen out of use when Roman writers first mentioned the Picts. The northern concentration of brochs has led to their builders being seen as ‘proto-Pictish’ ancestors of the later raiding bands. A simpler explanation is that the brochs were built by ‘Britons’ whose descendants in the early centuries AD remained largely untouched by Romanisation.

The Picti were none other than the Caledonii, Verturiones and other indigenous peoples previously recorded as separate entities but now appearing under a new collective name. Aside from this ‘rebranding’ of Rome’s old enemies, the situation on the northern frontier remained largely unchanged for much of the fourth century, except perhaps for an increasing number of barbarian raids. Whether these incursions became as serious as those of the Severan era in the early 200s is unknown, but they caused sufficient anxiety to provoke a Roman response. In 305, the respected general Constantius Chlorus marched from his base at York to deal with the Picts. He presumably defeated them. Likewise, his son Constantine, whom the frontier army proclaimed emperor in 306, took a break from civil war in Europe to wage a Pictish war in 312. Hostilities with the Picts continued up to the middle years of the fourth century when the emperor Constans, son of Constantine, came to Britain to oversee the imperial response.