Guy Leven-Torres 1996
The Roman Army on Campaign
Agricola and the Invasion of the North: Tacitus and Historical Interpretation
It appears rather fashionable at present to downgrade the achievements of men from the past, especially if they were prominent and that eminence strikes a chord of insensitivity among modern day elites. For example, Napoleon would have been quite pleased with some of his later press many years after his death but recently several revisionist authors have questioned his greatness as a military man. These same revisionist historians have also questioned the accepted version of events of not only Napoleon's era but also even the Second World War wherein many survivors are still in the full bloom of a very healthy old age. The most famous of course is Mr Irving whose appearance at a libel trial caused a sensation in the press.
We are not here to debate the rights and wrongs of Mr Irving's case. However it does raise a number of very important questions as to how we should review history, especially when it comes to interpreting Tacitus. Mr Irving had the courage to question certain perceptions of the Holocaust held by modern society. One does not have to agree with Mr Irving but there is no doubt he is an excellent researcher who is thorough in his investigations and has had unusually close contact with many of the families whose fathers, husbands, or grandfathers fought on the German side at General staff and party level. He came to his own conclusions no matter how far fetched they seem to the rest of Academia and the world at large in the post war era. The point being made here is that an academic in order to be a scholar must follow the truth even if that is sometimes unpalatable.
However, any conclusions so drawn must be justified by hard evidence. We cannot grow as human beings if everytime somebody prints a controversial piece, the media and mainstream academia of the day attack such a person as Irving in paroxysms of outrage and hysteria. Free speech and debate is essential to a healthy democracy. People's opinions must be respected otherwise, truth and debate die and fear takes over. We are seeing this happen to a large degree already. This is one main reason why people today have such a distrust of politics and fail to vote. We should all take a leaf out of the great left-wing historian AJP Taylor who stated in his Foreword to Churchill's History of the English Speaking Peoples that if one 'ignored rubbish it generally went away'(1968). Those who pounced on Mr Irving might like to bear this in mind before they made him a martyr.
That does not mean one has to ignore Tacitus and his assessment of Agricola. But Tacitus belonged to the Roman Senate and therefore represented a very upper class view of his father in law. If we as historians are having difficulty concerning the veracity of events sixty or so years ago, how are we to cope with questions of history and campaigns from eighty years or so after Christ? Especially so when the only version of events is a testimony we possess in copies which are far more than third or fourth hand. The Agricola itself must have been copied many hundreds of time down the ages, and it is well known that scribes make errors in their transliterate work, which is why debate among modern classicists is usually about philological problems and often based upon the ending of one word or other which can literally alter an academic's view of his period.
Tacitus, like Winston Churchill in our time was obviously anxious to have his version of events held to be the true version. In both politics played a huge role. Churchill claimed in his version of the Battle of Britain that Air Chief Marshal Dowding informed him that twenty-five squadrons was all that he needed to defend Britain against the Luftwaffe. Dowding denied this (C.Wilmot, Struggle for Europe 1951. Churchill History of Second World War Vol. II, p38 1948).
Tacitus in a similar vein says of his hero after Agricola had reached a point no Roman Army had before and thrashing the Caledonii, 'Fired with self-confidence and the glory of his victory, the army protested that no obstacle could bar its brave advance... Even the conservative strategists of yesterday were forward and boastful enough after the victory (Agricola 27, Tacitus). Tacitus was like all senators a staunch Republican and only just tolerated 'good emperors' like Trajan. The truth is that Tacitus owed much to the Flavians and it was only after the death of Domitian that he dared write his eulogy to Agricola.
Agricola is above all a political pamphlet. The author goes at some length to compare the present good times with the bad old days of the former regime. Trajan had come to the throne under rather forced circumstances in that his predecessor did not have much choice but to adopt him as his successor, otherwise there would have been civil war. An ailing and senile old man is an easy target for rebellion. So anything, which enhances the new Establishment, is welcome particularly from one who had ingratiated himself with the old regime and had a little covering up to do and favour to win of the fellow at the top. However, we should not be blind to the informative material that is contained in the Agricola. It is to that we now turn.
The problems of trying to analyse Agricola's campaigns have been compounded by interpretations of modern historians and archaeologists. Many of them despite highly creditable work and investigations have consistently failed to agree the ultimate success or failure of Agricola in his military exploits. Ironically, it is writers such as Leonard Cottrell in the 1950s who give the most positive and interesting interpretations of Agricola's skills. Cottrell has been criticised as too unscholarly for modern academic tastes but from a modern soldier's analysis of campaign it is first rate, simply because it makes good military sense.
Choosing from today's scholars, the best seem to be Sheppard Frere and William Hanson. There are of course brief accounts by Peter Salway and Hugh Scullard but none approach the analytical qualities inherent in Hanson or the very fine logistical-historical interpretation of Frere. The best of Frere is to be found in the twelfth edition of the Scottish Archaeological Forum published in 1981. William Hanson took part in this study of Agricola's campaigns as well.
Vindolanda
From a purely military nterpretation, the best version of events is that written by Frere in the publication cited above, entitled, 'The Flavian Frontier in Scotland.' Frere has made the basic assumptions that the 'glen blocking forts' are indeed Agricolan and supports this by reference to The Agricola, Chap 25, 3. This is the point in Tacitus' narrative where the Caledonii attacked a fort or forts and this led some Roman officers to counsel withdrawal behind the Forth-Clyde Isthmus. Frere states that it hardly made sense for the Romans to retreat behind the Isthmus if the fort was not further north.
If we examine Agricola 23, we find that the Isthmus had been fortified and territory to the south secured. This event probably took place in 81AD but it does not have to mean that outposts did not exist further north. They certainly existed in Hadrian's time and that of the Antonine Wall. From a logistical and strategic point of view it made good military sense to have forward observation posts manned in some strength. It would be necessary to provide early warning of enemy movement and therefore provide suitable accommodation for forward cavalry and infantry patrolling and reconnaissance. Besides these forward posts would secure the area against surprise attack but more importantly for our purposes it is good military practice that still goes on today. Only a complacent commander would fail to secure not only his rear but frontal areas as well.
A good commander will always 'recce' an area himself rather than leave it to minions. Obviously intelligence is gathered from many sources, clandestine and otherwise. Modern operations are a lot more mobile and faster than Roman times with enemy troops advancing well into the rear of the defender. One role of the Special Services is to maintain ground observation posts in the captured areas in order to keep reporting vital information to their own side in preparation for a counter attack. But more than this these posts which have been bypassed by the enemy provide not only observation points but also centres of resistance against invading forces tying down valuable enemy troops, sometimes in large numbers. When the counter attack comes it means of course that one has men and equipment already to hand thereby maintaining a grip on territory and further confusing the enemy. These forward bases often serve as embarkation points for offensive operations in the enemy's rear. This was the lesson Orde Wingate taught the modern British Army in Burma. He took whole command and control headquarters with him, including field-hospitals, which gave him a series of huge independent military commands able to fight the Japanese in their rear but with all the materiel and men necessary to wage war supplied from the air. He tied down thousands of enemy troops.
At Arnhem, dropping paratroops behind German lines made a forward thrust. By doing this Montgomery had hoped to secure a bridgehead well in the rear of the enemy. He dithered by failing to send in heavy armoured and infantry support and subsequently the British at Arnhem had to fight against heavy German attacks. But they were very well dug in. They took a lot of enemy with them. Finally an ordered withdrawal was made. But the lesson is clear- forward defence is vital supported by a good heavily defended rear. If Frere is right then we are beginning to see something of the type of commander Agricola was.
Military science demands forward planning and observation/ intelligence gathering in great detail. Only a very stupid commander would fail to do so. It is more than likely that the Roman line of forts on the Forth-Clyde line was a bridgehead to be used as a 'jump-off point' for future campaigns further up the Island towards John O'Groats and hence Tacitus' carping about the willingness of the Roman Army to go on (Agricola 27). But to the rear of this line also existed a line of glen-blocking forts that performed another vital function in that it secured conquered territory. If Frere is in fact right in his assessment that they closed off the valleys or glens, then this is another move by Agricola to protect his left flank. This is nothing less than classical military containment of a perceived threat.
At Agricola 25, Tacitus tells us that 'He (Agricola) feared a general uprising of the northern nations and threatening movements by the enemy on land, he used his fleet too reconnoitre the harbours. It was first brought in by Agricola to bring up his forces to the requisite strength.' How else could he do this without a bridgehead into enemy lands? The Caledonii probably took to the hills as the Romans advanced. It sounds as though Agricola knew his Frontinus. His actions if they have been correctly interpreted are quite in keeping with modern military thinking. By this is meant the heavy build up of overwhelming force, intelligence gathering, the establishment of a bridgehead in secured territory, secure territory in the rear and on the flanks, forward observation posts. A general at the Ministry of Defence in London or the Pentagon would recognise all these activities. In fact a map of Frere's interpretation of Agricola's forts was shown to a former high ranking officer in Britain's armed services. He commented that Agricola acted and behaved like a modern strategist (comment by former RAF officer in conversation with G. Leven-Torres 1996).
Most archaeologists though are guarded about placing a precise date on these forts. However, at least one scholar points to the truth of Frere's assumptions. Hanson states, 'While not disputing that the mountains themselves are unlikely to have proffered any major attractions to settlement, the straths and glens which divide them have much to offer. In particular recent detailed investigations of Loch Awe and Loch Tay suggest that our estimate of population density in the Highlands in prehistory needs to be drastically revised to take into account the number of crannogs or man-made islands (Dixon 1982;Morrison 1985). Moreover, contrary to Hind's assertion, the place name evidence that he quotes for the location of the Caledonii...would place them within the Highland massif not east of it (Agricola and the Conquest of the Northern States 1986).
Frere dates these forts along the Isthmus and those he sees as 'glen blocking forts' further south as Flavian. He names these as Drumquhassle, Menteith, Bochastle, Dalginross, Fendoch, and most importantly for our purposes, the main legionary fortress in the area Inchtuthil. This is because its position at the top of the map is important militarily and gives a strong clue to the type of commander Agricola was. One can only trust that Frere's hypothesis is right and that the forts are Flavian. One has been dated through planking removed from the side of a Roman well. The dendrochronological dating is the summer of 80AD.
There is nothing particularly special or brilliant in the way Agricola placed his forts. He had learnt and honed his military talents under Frontinus who wrote a treatise on military strategy and Petilius Cerealis, the former commander of the ill-fated Ninth Hispana. Actually more revealing is the political relationship between these latter Roman commanders and Rome itself. Cerealis had something of a chequered career, especially on the Flavian's side during the civil war of 69AD when he disguised himself as a peasant to escape from Vitellius' Rome. But Cerealis was a careful judge in the political sphere (Histories iv, 86) and despite the odd military fiasco such as the rout of the Ninth Hispana at the hands of Boudicca was a competent and inspirational soldier. His time in Britain saw further consolidation of the North of England and the establishment of the Legionary fortress at York. This was common Roman practice to place a heavy force near at hand to hold a possible enemy in thrall by the threat of armed force. Brigantia had been a client kingdom under Cartimandua. The resultant problems led to the Romans annexing the rest of the territory and placing auxiliary forts around the region in order to patrol and observe. The placement of the fortress at York formed the same basis of military strategy that Agricola used in Scotland. Edward Luttwak who wrote the Grand Strategy of the Roman Empire (1976) called this type of deployment 'an economy of force.'
Leonard Cottrell gives a vivid account of this annexation. The Romans according to his version advanced on the two flanks of the Pennines. Agricola commanded the western advance and Cerealis tested Agricola's mettle as a commander. He had been sent to Britain to take over the notorious XX Legion Valeria Victrix that had elected to fight for Vitellius and the governor at that time, one Vettius Bolanus was a Vitellian appointee. The young thirty year old Agricola apparently had been sent to bring the mutinous and surly Twentieth back to its sense of duty.
In relatively quick succession, Agricola 'cut his military commanders teeth' under two very able Governors. The last of them was dear old Frontinus who set about the destruction of the Welsh tribes in 74AD. He used similar methods of contaiment to those of Agricola in Scotland. There are glen blocking forts at Tomen y Mur and Caer Gai for example.This fact above all else would seem to prove the hypothesis of Frere that the forts in Scotland are Flavian, in that Agricola seeing this strategy in Wales, copied his former commander's activities in Scotland. Further evidence seems to support this as well.
There is epigraphical evidence of Agricola at Chester from around the time of Frontinus' period of rule in Britain. The fact that pottery from slightly earlier has been found at Carlisle and Bowes strongly suggests the presence of Cerealis around the early seventies would seem to futher support Frere's idea that both governors had shown a keen interest in the Brigantian region. It seems that the historical key 'fits the lock.' The chronology of events wherein three Flavian men and 'experts' on warfare, especially in a region like Northern England would seem to indicate a consistent imperial policy. Agricola then spent time as governor of Aquitaine for four years. Vespasian thought highly enough of him to recall him and send him to Britain. Agricola was, like Petilius a Flavian man. He had raised troops in the region of his birth in order to support the usurper Vespasian. 
The actual dating is something of a problem. Agricola apparently arrived as governor in 77AD (Hanson 1986). Therefore taking the events as laid down by Tacitus, 77AD would be the first when he smashed the Ordovices late in the season (Agricola 18). He then established forts in Wales of the glen blocking type at Denbigh, Flint and Caernarvon, and used the Batavians to storm Anglesey. A legionary base already existed at Chester dated prior to Agricola's time as governor. AD 78 saw him fortifying and annexing more territory in Brigantia. He again advanced on both sides of the Pennines and eventually reached the Tyne. Corstopitum near corbridge appears to have been his jumping off point. In AD 79 he launched his main offensive into Scotland. By 80AD he had reached and consolidated his position on the Forth-Clyde Isthmus and built a line of forts (Agricola 23). Domitian succeeded Titus in 81AD so it is probable that Agricola ' kicked his heels ' for a year awaiting new orders. The penalty for exceeding gubernatorial imperium was harsh. Domitian was known to be fickle. The new emperor ordered Agricola to supply vexillations to his campaigns in Germany and along the Danube, It appears these came from poor old Nine Hispana. But he did sail along the Scottish coast and campaign, probably to keep the Caledonii occupied and test their strength. In 82AD he opened his big offensive (Agricola 25).
We find evidence of marching camps and forts along his line of advance at Dunblane, Ardoch. There was also a line of signal stations along Gask Ridge, although they have been archaeologically dated as 'before 90AD.' For our purposes we shall assume them to be Agricolan. Dating is not an exact science. There is also a road from Sterling to Falkirk. He followed this through Strath Allen and across Strathearn, with the Ochill Hills at his back, the Firth of Tay gleaming on his right and away to the northeast the Grampians. It was now that Calgacus attacked Agricola in several columns (Agricola 25). In response Agricola divided his forces. The poor depleted IX Hispana was Calgacus' primary target no doubt hoping to exploit the weakest spot in the Agricolan plan. Calgacus fell at the Mons Graupius in the following year 83AD (Agricola 29).
Inchtuthill was left unfinished. It was probably started around this time but its non-completion rather makes one feel that Domitian was less interested in Scotland and a wayward commander bent on his own glory since he was recalled to Rome in this same year (Agricola 39, 40).
Dating the forts above to Agricolan times has proved contentious; the best that can be said is that they are 'Flavian' (Hanson 1986). Hanson the archaeologist has taken the forts Barochan Hill, Cudder, Mollins, Castlecary, Camelon, Mumrills and Elgihaugh, measured the distance between them and deduced a 6-8 mile mean distance and feels this is strong archaeological evidence of the line of forts established by Agricola. Flavian dated artefacts have been found at all of them. Furthermore, Inchtuthill is not placed well strategically but tactically.
It would appear that all this evidence even allowing for the enthusiasm of archaeologists and historians, the whole system if it is 'Agricolan' seems to make good military sense in that it is consistent with what we know of Roman practice. Hadrian's Wall is good evidence and there are similar systems elsewhere in Europe dated to the same period. Looking at it from a soldier's point of view, it does make good sense. But what does it tell us about Tacitus' hero? He appears to be competent Roman general typical of the type. There does not seem to be anything particularly outstanding about him.
His actions concerning the attack upon the auxiliary cohort by the Ordovices (Agricola 18) at the start of governorship, appears something of a planned reaction. Obviously the failure of previous governors such as Cerealis to 'finish the job' was well known to the new governor upon his arrival in 77AD. There is also evidence that Agricola made certain new dispositions as regards Brigantia. There certainly appears to have been a general policy decision about this time to consolidate the Roman hold on Brigantia. This would make sense in order to ensure no trouble broke out in the rear. Vespasian would have been well aware of Agricola's knowledge of Britain. The fact that he had served under no less than four previous governors made him something of a specialist on the subject. He must have been fully aware of previous failings and inadequacies of his previous commanders.
His whole governorship has the heavy feeling of trying to consolidate and finish off the conquest of Britain. It tied down three legions and strategically speaking it was not that important. Nero thought seriously about abandoning the province only to be dissuaded by Seneca and Burrus on the grounds it would reflect badly on the new regime's reputation with the Army. There is in all this a hint of Julius Caesar and Augustus who established the Julio-Claudian dynasty. Octavian as he was then known set about apolicy of consolidation upon the frontiers.
Vespasian was to found a new dynasty. The army still remembered fondly the Julio-Claudians. He had an established military reputation though but being the far-sighted and practical man he was foresaw that Northern Britain would continue to be a drain on military manpower. This was especially so in the light of his experiences in Judea during the late 60s.
The Brigantian kingdom was no longer relevant to a secure Britain. Roman policy as regards Cartimandua had failed. But there is more than a hint of hesitation even though Cerealis placed the Ninth Hispana at York, he seems not to have consolidated this annexation fully. Hence Agricola's later work. Whatever the case obviously he felt that hostile parties could have caused a problem in his rear.
So it is fairly safe to assume that Flavian policy dictated the use of 'experts' like Agricola. Returning to Archaeology. Hanson and Frere (1986) both argue for the dating of the 'glen-blocking forts' as Agricolan. Furthermore bronze asses dated no later than 86AD have been found at Inchtuthill which was unfinished. These have also been found at Dalginross and Strathcaro. All provide a terminus post quem well in keeping with the departure of Agricola. From a military point of view they do seem remarkably too well placed to be the work of more than one commander. Even if one were to raise the argument that they are the result of different commanders at consecutive periods, thess fall down by virtue of the fact that most commanders in history upon taking command in a new theatre have their own ideas. One only has to examine recent history to establish this fact and this is also the reason why commanders are changed in the hope they will bring fresh impetus and outlook to a military problem.
From the above it appears that Agricola may have been a good general but it must be clearly understood that the Emperor was the Commander in Chief. Coins with the head of Domitian as 'Imperator' appear around this time as well. Some archaeologists and numismatists have believed the issue of these was to commemorate Mons Graupius.
Military commanders if they are good also tend to become arrogant. One only has to examine the likes of Alexander, Caesar, Patton, McArthur and even to slight degree Montgomery and Churchill. There is evidence in The Agricola that he developed this sense of his own pre-eminence. This was not good for realations with Emperors like Domitian (Agricola 39, 40).
Summing up the man, it could be stated that Agricola was typical of his type. Better than average, popular with his troops, concerned for the welfare of his men, a good tactical commander but not so brilliant at strategy and a tendency to fall into complacency in respect of his own talents. Domitian was in many ways right to recall the man. Such a one could pose a threat to the throne if allowed to grow to overconfidence. Besides war threatened from a far more dangerous people, namely Decebalus and the proto-state of Dacia, which left un-checked threatened the whole Northern Frontier.
Reading list:-
Tacitus, The Agricola, Penguin Classics edition.
Tacitus, The Histories, Penguin Classic edition,
Frere, Britannia. 1968 and revised copies 1993 etc.
Hanson, Agricola and the Conquest of the Northern States etc 1986.
Churchill, History of Second World War 1948.
Wilmot, The Struggle for Europe 1951.
Luttwak, Grand Strategy of the Roman Empire 1976 and revised 1994.
