Wednesday, January 8, 2014

Who Placed the Earliest Roman Footprint in Scotland?

by Nancy Jardine

Research on northern Britain for the post-Claudian period between AD 69 and AD 84 has been in a state of flux for some time now, due in part to recent archaeoscience techniques. These new procedures provide theories which change earlier perceptions of what was happening to the northern Celtic tribes of Britannia, and puts into dispute which Roman Governor of Britannia was responsible for the earliest major infiltration of the north of England and of Scotland. Keeping up to date with recent advances in archaeological research is not something I find easy to do as a non-professional in the field. In fact, by the time this article is posted, there may well be new additions to the most recent findings which may well change perceptions yet again.

What is known of the indigenous Celtic people who inhabited the areas tends towards the conjectural, written evidence left by the Celtic tribes of northern Britain ostensibly non-existent till someone proves otherwise. This leaves a researcher, like me, to depend on (translations of) Latin and Greek sources, men who wrote of Britannia from their own somewhat biased viewpoint. To a limited extent, one could also argue that some knowledge of those Celtic tribes is gleaned from their oral tradition and from comparisons with the barbarian ‘Keltoi’ tribes of continental Europe, the name coined by Latin and Greek writers who provide a little documentation on them. Science and traditional archaeology provide the researcher with other, non philological interpretations but it can take some time before information filters down to the layman.

In some Brigantian areas of northern England, stone foundations of Celtic roundhouses have been investigated giving a glimpse of the neighbouring Celtic lifestyle, though these are localised and not widespread throughout northern Britannia. Wooden roundhouse construction was much more common throughout the north, the scant remains of decayed wood leaving only traces to interpret. (An exception to that statement would be the stone built brochs of present-day Scotland - though those buildings and the lifestyle of the people inhabiting them are worthy of a completely separate study.) However, when other botanical and biological studies of plant and animal remains and findings of glass, pottery, metalwork and coins are added, the picture of Celtic life in northern Britannia builds up. A peaceful agrarian image emerges, the Celtic farmer perhaps only taking up arms to defend his plot of land, or warring on a small scale with rival Celtic tribes over localised disputes. In northern Britannia the landscape varies so widely; much of it is mountainous or hilly, inhospitable and barren; therefore only a small percentage of the whole was able to be farmed successfully by the Celts of the post-Claudian period.

At what point was that peaceful farming lifestyle interrupted by Roman Army advances?

By AD 69, Romanisation was well in place in the southern reaches of Britannia but the tentacles of Roman civilisation had yet to grip tightly in the north. Historical references about Queen Cartimandua of the Brigantes indicated that so long as she kept her federation of northern Brigante Celtic tribes from warring with Rome northern Britannia was largely left free from oppression by the Roman Army. Even after the fall of Cartimandua in approximately AD 69, the queen having been rejected by her own people and also by Rome, it was thought that Roman presence was still light in northern Brigantia for some further years– at least that was the supposition until approximately the 1990s when theories were re-evaluated. Before the 1990s, the land covered by present-day Scotland was thought to be relatively free from Roman invasion till sometime during the later AD 70s. Thinking changed – the goalposts of the Roman invasion of Scotland moved to the beginning of the AD 70s.

Dendrochronologist findings on wooden post holes and other wooden traces, point to Roman forts having been established in Brigante areas in the early AD 70s. Scientific studies on pollen, grain and other botanical remains like grasses point to an increase in farming productivity in parts of central Scotland during the late AD 70s, with more successful animal grazing occurring. The changes are thought to indicate a new market for the farmers’ labours. That market is now thought to have been for Roman consumption, and being seasonal such a demand would only be met with some forward planning indicating a longer settled Roman presence in those areas of Scotland – a surprise to those conducting the studies since regions where military upheaval occurs do not generally show successful farming.

The lack of plentiful hard evidence makes the study of this era a very fascinating one; though as a novelist of the period I find it is frustrating to not know the fine details. It is even more exasperating when new revelations and discoveries make rewrites to manuscripts appear inevitable when interpretations, which had been accepted for decades, become disputed and even discredited.

Recent scientific techniques providing this different information have laid major doubt on the chronological timescale as was set out by historians till the 1990s. The governor credited, by many scholars prior to the 1990s, with establishing initial Roman Army presence in Scotland was Gnaeus Julius Agricola, Roman Governor of Britannia from approximately AD 77/78 to AD 84. That theory was based on a chronological reliance on the military activities of Agricola, as written in the work named ‘Agricola’ by the Roman historian Tacitus, son-in-law to Agricola. For some decades now that ‘historical’ theory has been disputed. Current archaeological data just does not ‘fit’ a literal acceptance of Tacitus’ recounts of Agricola’s military movements in northern Britannia during his approximate seven year tenure of governmental office– though it has long been accepted by scholars that Tacitus never intended his account to be more than a posthumous accolade for his father-in-law. Recent use of dendrochronology and other archaeosciences, like the use of C14 dating, now indicate that forts, fortlets and signal towers in northern Britannia, which were formerly attributed to Agricola, were in place some years before he became governor.

If not Agricola, who can those military advances in northern England and Scotland be attributed to?

Tree ring dating for the ramparts of the permanent fort at Carlyle (Luguvalium) indicate that the posts were in place in late AD 72, some five or six years prior to the Agricolan surge northwards as documented by Tacitus. The date of AD 72 signifies that the fort was erected during the governorship of Quintus Petilius Cerialis (AD 71- AD 74). Though it can’t be discounted that if the posts were replacements for an earlier fort’s main structure, the original fort erection may perhaps even have been during the tenure of Vettius Bolanus (Governor AD 69 – AD 71), or earlier. Wooden structures have a short shelf life - weather conditions and type of wood used playing a determining role as to how long they can last for safe use without repair. A few years might be the best duration in some wet or inclement areas before replacement is necessary and Luguvalium, on the border area between England and Scotland, does not always enjoy fine weather. Dendrochronology, in particular, has been used to support theories that many sites of Roman occupation in northern Britannia had more than one construction phase - meaning that the earliest occupation of the site may have been pre-Agricolan and even as early as the governmental tenure of Bolanus. That Bolanus could potentially have been responsible for Roman incursions and the settlement of Roman troops in Scotland is an exciting theory. Cerialis was a proven and experienced Roman soldier. The notion that he marched his troops northwards in Scotland, possibly even to the very north-east of Scotland, is definitely feasible to me – and would possibly have been some six years, or more, before Agricola.

Roman sources exist which may back up the theory that Roman incursions in northern Britannia were made during the time Bolanus and Cerialis were in office and that one or other of them may even have ventured all the way up into northern Scotland before AD 74. Pliny the Elder’s “Natural History” makes references to Roman engagement with the ‘Caledonians’ by (or before) AD 73, and the poet Statius referred to Bolanus having set up watch towers and strongholds in ‘Caledonia’.

In addition to the findings at Luguvalium, evidence in pottery and glass, and dendrochronology verification of many small fortlets in Brigantia strongly suggests this lengthier, pre-Agricolan, occupation of northern Britannia. The strength of the argument put forward during earlier decades on coin evidence pointing to occupation by Agricolan forces in northern areas, during the period AD 78 to AD 84, is now less secure when taken alongside newer archaeo-scientific findings.

When Agricola became Governor of Britannia in either late AD 77 or early AD 78 his campaigns of the north seem much more credible if the way had mostly been paved by his predecessors and what he was actually doing was more a case of proclaiming his presence as Governor with his full legionary back-up, rather than actually subduing the natives along the complete route as he marched northwards. However, perhaps not the whole way, which might explain the two legions of troops who are claimed to have accompanied him at the time of the only large battle mentioned by Tacitus - but I will come to that later.

The Roman tendency to first establish their lines of advance, in order to be victorious during any engagement with resisting local tribes, would have been long done if Cerialis had already moved his troops northwards into Scotland in the early AD 70s.

The map which was created, according to the account written in Tacitus’ ‘Agricola’, indicates that Agricola dotted around a fair bit in southern Scotland during the years AD 79 to AD 82 and only advanced to north-east Scotland in AD 83 and AD 84. Pollen and cattle grazing studies in Angus indicates that there was no fall in production during this time as might have happened if the farmers had been killed in battle and the land and animals left untended. A more peaceable earlier occupation has been theorised, pre-dating Agricola. More evidence may yet be uncovered that the troops of Bolanus, or Cerialis or even Frontinus (AD 74-AD 78) had settled in the central belt of Scotland and up towards Tayside before Agricola.

North of the central belt of Scotland more than 70 temporary marching camps, 1 legionary fortress and at least 14 auxiliary forts were constructed and were thought to have been of Flavian origin. Earlier attributed to Agricola, dendrochronology and non-intrusive techniques used to survey some of these forts now indicate they had more than one construction phase, possibly even three structural periods. Three rebuilds would mean a longer time of occupation of that area and would put them at a time that was likely before Agricola set foot near them. Ditch rebuilding on permanent sites of Roman occupation lengthens the duration even further – turf walls and ditches having tended to last much longer than wooden posts. The habitation of the Inchtuthil legionary fortress now appears longer than the few years accredited to Agricola’s forces. If the Roman Army was pulled south from Inchtuthil around AD 86 or early AD 87, according to ‘surge year findings’ of coins, then the occupation time at Inchtuthil could put the earliest phase of wooden construction at a date that was pre-Agricolan, perhaps built by Cerialis or Frontinus.

Work during the last decade on the Gask Ridge series of forts, fortlets and watchtowers in Perthshire, Tayside and Angus - some of the northerly ones in the line earlier referred to as ‘Glen Blocker’ forts - continues to provide exciting new evidence. Discovering missing forts along the Gask Ridge, which is now believed to be the earliest lime in the whole of the Roman Empire, may mean even better answers. I look forward to the research surveys which will be done in 2014 to find ‘the missing’ fort/s because I find it perplexing that Roman construction stopped at Stracathro in Angus, the most northerly permanent fort. If Cerialis had anything to do with Gask Ridge construction why did he stop at Stracathro? He was a seasoned and well experienced soldier, having distinguished himself in mainland Europe and seemed to have been more than capable of carrying on northwards. Of course, other governmental duties and circumstances across the rest of Britannia may have played a significant part in any withdrawal by Cerialis from Scotland, leading to Agricola being credited with making the most northerly advances.

Beyond Stracathro in Angus, the Romans appear to only have laid down the defences for marching camps as they made advances north eastwards in Aberdeenshire, all the way to almost the Moray coastline. Till more evidence is uncovered north of Stracathro, I like the idea that this was the geographical cut-off point, a place where Agricola began to tread new ground when he tramped further northwards. The archaeological excavations at the Roman Marching Camp of Deer’s Den, Kintore, Aberdeenshire, between 2002 and 2006, revealed a much greater occupation than had previously been envisaged. Instead of 4,000 troops using the site as was imagined before the 1970s, the excavation findings of 2002-2006 uncovered sufficient Roman ovens to postulate that the site had been occupied by some 10,000 troops during the Agricolan period.

Just nine miles or so north-eastwards from Kintore the marching camp at Durno, the largest found in north-east Scotland, harboured perhaps as many as 30,000 troops, though at which time period may be debatable since both Kintore and Durno were inhabited on three different occasions spanning from the Agricolan period to the Severan campaigns of AD 210.

The battle, mentioned by Tacitus in his ‘Agricola’, involving the Caledonian leader Calgacus, a battle later named ‘Mons Graupius’ by Victorian historians, makes more sense to me if Agricola did not have to engage much with local tribes in central Scotland prior to his advancement in Aberdeenshire. His legions would have largely still remained intact on the northwards journey if he had not met with Celtic resistance till north of Stracathro. Those 10,000 troops harboured at Kintore, and then potentially 30, 000 troops at Durno would have been reasonably fresh for battle and victory as stated in Tacitus’ ‘Agricola’. From Tacitus, we read of Agricola’s ambitious plan to place the Roman footprint on the whole of Britannia, having established that it is definitely an island and navigable. That part of the recount of Agricola’s endeavours seems very plausible though the navigation around the northern coast of Scotland was perhaps well before any ‘battle at Mons Graupius’, since it appears that the Classis was available and ready to be used to swell troop numbers and had been patrolling the waters off north-east Scotland. Since some of those ships could carry 400 soldiers, it would not take too many ships to augment the Durno troops. Even if 30,000 does seem such an immensely inflated number of soldiers, the distance from shore to the battle site at Durno across from the hills named Bennachie is not so implausible. Tacitus being loquacious, enlarging those numbers at the site of battle in his writing, to glorify his father-in-law Agricola, is also very probable, and yet it does not account for the amount of hectares covered by the camp ramparts if they are indeed Agricolan. The hectares covered indicate some 30,000 troops sleeping inside the boundaries.

Tacitus’ account of the exploits of Agricola can be taken with a pinch of salt but that does not explain other Roman intentions in north-east Scotland. The size of Inchtuthil, a Legionary storage depot fortress and garrison, was home to 5,500 men. How temporary the stay at Inchtuthil for the soldiers of the Roman Empire is debatable. Such legionary storage fortresses in southern Britannia were places where the onward progress of cohorts of soldiers was organised. How long it operated for sending troops northwards is an interesting question. Even more interesting - the massive amount of iron nails, over three quarters of a million of them in different sizes, which were buried when the legionary fortress at Inchtuthil was abandoned (approximately AD 87), is astounding. Nails were used for hob-nailing boots…for repairs to equipment and weapons…and were for building new forts and watch towers. They were perhaps also for making repairs to existing structures.

The Inchtuthil fortress construction and use signifies the supplying of forward troops, but was it built by Cerialis, Frontinus or Agricola?

Why those nails were buried, and not used in construction of new fortifications north of Stracathro, is an itch I’m perpetually scratching right now. I look forward to reading the latest information on the Roman occupation of Scotland and to the responses anyone might have to my possibly misguided theories in this post.

Sources used:


Books 1 & 2 in Nancy Jardine’s Celtic Fervour, historical romantic adventure series - The Beltane Choice and After Whorl: Bran Reborn (Dec 2013) - are available from:
Amazon UK
Amazon US

(Book 3 of the Celtic Fervour series After Whorl: Donning Double Cloaks, which takes the action to north-east Scotland and battle at Bennachie, is due for publication in spring 2014. )


  1. Thank you Nancy! I'm fascinated by Agricola.

    1. Hello, Kathy. I find him fascinating, too. I don't have time just now for a more in depth research of him but would love to do that someday. Cerialis sounds like quite a plan as well!


Note: Only a member of this blog may post a comment.