Caradoc and the Romans


Panoramic Tour of Caer Caradoc

Panorama from Caersws


Dear Reader,

This article is based upon my own assumptions of where the last stand of the British chieftain; Caradoc (Caratacus) took place against the Romans in A.D.51. Its dependence on well-established histories, particularly the writings of the Roman historian Tacitus are fully acknowledged in the text, together with the works of other historians both ancient and modern. I have also included details of other speculative sites in the area, which will allow for comparison and conjecture.

It is my hope that the visitor to the beautiful Welsh Boarder region will not only reflect on the building of King Offa’s Dyke, and the deeds of Owain Glyndwr, but will also take time to search out the sites of a sadly overlooked British leader who deserves a special place in the folklore of this historical landscape-Caradoc.

The Roman Invasions of Britain

After Julius Caesar’s two abortive expeditions to Britain in 55-54 B.C., the first Roman Emperor and adopted son of Julius Caesar, Augustus (63B.C. -A.D.14) issued an edict banning further expansion of the Roman frontiers. In A.D.40 the Emperor Gaius (Caligula) [1]Caligula,” Little Boots”, a name given him as a child. received the surrender of a number of Britons while he was campaigning in Germany. Among those who paid homage to the mad Emperor was Adiminus, brother of Caradoc and son of the British chieftain Cunobelin. After being banished by his father, Adiminus sought help from Gaius in order to reassert his position in Britain. Thereafter Gaius took his army to the Channel coast of France (Gaul) and ordered his troops to gather seashells, considering that by doing so he had conquered the sea and therefore had obtained a victory over Neptune! [2]Rodney Legg,’Romans in Britain’, page 10

With the assassination of Gaius in A.D.41, his stuttering, lame footed uncle, Claudius became Emperor. Claudius now turned his attention to incorporating Britain into the Empire. The Claudian invasion of Britain in A.D.43 was based on little more than it being, ‘ the most obvious place for the acquisition of real military glory’, and the fact that, ‘ the impact of conquest on the public throughout the Empire was considerable, partly in response to Claudius’ precised wishes’. [3]Barbara Levick,’Claudius’, page 139

When we consider the massive logistic problems that have been entailed in such operations as the 1944 Normandy Invasion, The Falklands War and Operation Desert Storm, then the organization of men, horses, ships, food and equipment for the Roman invasion of A.D.43 must have been a tremendous undertaking for its time. In all some 45-50,000 men, in four legions-II Augusta; IX Hispana; XIV Gemina and XX Valeria, together with vast amounts of supplies, as well as ancillary workers were transported on over 1,000 ships, this fleet alone necessitating the felling of thousands of trees. [4]No real information on just how many ships were used in the Claudian Invasion, but see General J.F.C.Fullers book,’ Julius Caesar, Man, Soldier and Tyrant’, page 124. Here there is a very … Continue reading   The command of the invasion was given to Aulus Plautius, who was to become the first Roman governor of Britain.

In the late summer of A.D.43 the main force of the invasion fleet embarked from the port of what is now modern day Boulogne, landing at Richbrough, Kent; while smaller flotillas came ashore at Dover and Lympne. No serious opposition was met during these landings, but a decisive battle was fought at the Medway, which caused many of the British tribes in this area to make their peace with Rome. Caradoc and another one of his brothers, Togodumnus managed to pull a large force back to their stronghold at Camulodunum (Colchester); however Plautius did not attack the British capitol at once, but awaited the arrival of the Emperor himself, who had just landed with reinforcements, including elephants, from Rome. Claudius now took command and the assault began. We have little to go on as to what exactly took place during the battle, but it would appear that Caradoc offered battle outside the fortifications of Camulodunum in order to allow himself more room for maneuver. We do know that the elephants played a crucial role in causing panic in the British ranks, which in turn allowed the Romans to obtain a victory; nevertheless, Caradoc was able to escape to the west taking a large portion of his army with him. [5]       Peter Berresford Ellis,’ The Celtic Empire’, page 199

After receiving the surrender of Camulodunum, Claudius celebrated his triumph, and after only sixteen days on British soil he took himself off back to Rome, confident we may imagine that all had been done that necessitated his Imperial person remaining to checkout a British winter. What became of the elephants we can only speculate? It would be quite an achievement if all of these poor beasts retuned safely to a warmer climb, and after being engaged in a full blown battle one can say with some certainty that many must have suffered horrendous wounds when we consider the type of weapons used at this time.

After the battle at Camulodunum the only real problems the Romans encounter were in the southwest, which was dealt with in short order by the Roman general, and future Emperor, Vespaisian with the II legion Augusta. [6]Graham Webster,’ Rome against Caratacus’, page 13  Thus the Romans secured for themselves a large foothold in Britain, which stretched from the Humber along the Trent Valley as far as the Bristol Channel;  [7]See enclosed map however the tribes beyond this frontier were quite another proposition.

As was their wont, the Romans at once began to seek security and protection of the new frontier line by the creation of client states; this was a system of Roman patronage whereby each kingdom within the borders of the Roman frontier was allowed to be ruled by it as own chieftain. These friendly states paid tribute to Rome; while at the same time they took an active part in the prevention of outside forces invading the province, a practice that was dubious in the extreme.

Plautius did not have much success in recruiting many of the tribes into client statehood beyond the boundaries of his initial frontier line; however he did manage to pull-off one great coup, and that was to secure Queen Catimandura on the throne of the Brigantia tribe, which in turn brought together other smaller tribes from as far away as Yorkshire and Lancashire. Catimandura will appear again in the saga of Caradoc.

Plautius remained in Britain continuing the ponderous job of conquest until A.D.47 when he returned to Rome to receive his ovation, which was on the grand scale, with Plautius entering the city on horseback rather than in a chariot, a custom normally only reserved for members of the Imperial family. [8]Barbara Levick, ‘Claudius’, page 144

Once again, as they had always done the Romans, without reference to military precedence took away a commander who understood his job, and replaced him with a man of lesser talent. Publius Ostorius Scapula now became the second governor of Britain.

The Governorship of Scapula

Not much is known of Scapulas military career, but it is certain that his choice as commander was very much a case of who you know rather than what you know. His actions against the tribes were to prove,’ hasty and ill considered even by the standards of military expediency. His growing frustration and anger which led to his death, prompts the view that he was a sick man on entering office.’ [9]Graham Webster, ‘ Rome against Caratacus’, page 15 This shows how vulnerable the Roman structure of command could be. To keep on changing commanders, and not necessarily for the better while not even bothering to consider the consequences of these acts undermines the whole Roman military system.

Caradoc now had solid support in the west, and if left to his own devices he would once more move towards the Midlands and the Northwest; therefore it became necessary to take into account any potential hostility which this may cause on the other tribes who came within the Roman sphere of influence. It is not clear how this was done but by the time of Scapulas arrival in Britain the Romans had an even distribution of forts holding all the main means of communication. [10]See enclosed map. Nevertheless Scapula faced a, ‘serious disruption, hostile tribes had irrupted violently into the province’. [11]Graham Webester,’Rome against Caratacus’, page 20

In the Latin text the Roman historian, Tacitus makes use of the word “turbidae”, which has been taken to mean disturbance; however it is used elsewhere in the Latin to mean mutiny, and if this is the case then it does indeed look as if some kind of internal disorder had been planned to coincide with Caradoc’s raids, this in turn showing how precarious the Roman system of client tribes could be [12]Anthony Barrett, ‘Historia’, No 17, page 227.

With only four legions, Scapula had to use his mounted troops to try and break up the various bands of rebels and raiders, while also being obliged to carry out search and locate operations, which would bring Caradoc to battle. Moreover, since the province was now in such an unsettled state, he could not afford to split his forces into penny packets since this would allow for each part to be defeated in detail. Also Scapula could not move against Caradoc and leave revolt simmering in his rear, and to be effective against the British chieftain he would need at least two legions.

Tacitus, our only real source for this campaign, says that Scapula, ‘Prepared to disarm all suspects and reduce the whole territory as far as the rivers Trent and Severn’. [13]Tacitus, ‘The Annals of Imperial Rome’, XII, 30-31  A interesting point here is that we are given a real demarcation line for the Roman frontier during this period, which, as one would have expected is based upon river lines, although whether the Romans considered that the Severn and Trent could keep out raiders is another matter; far greater barriers such as the Rhine and the Danube never seemed to have created a problem for the Germans and Gauls. The other thing here is the matter of disarming the suspect tribes. Under Roman law no one was allowed to carry arms with the exception of soldiers and officers. Here we must ask the question of just how this law was implemented? The problem was that since there would have been a great reluctance to hand over these weapons how did the Romans know that the spears, swords, bows and arrows etc that were handed in were not just old scrap items set aside for just such an occasion, while the real tools of war remained well hidden? The reason for implementing this law allowed Scapula to terrorize the Britons. He also forced the Icini tribe into a conflict in which the Romans defeated them with only their auxiliary troops. This later incident however would leave a lasting hatred, which would erupt into a full scale rising against Roman rule in A.D.60.

After subduing a rising by a faction of the Brigantes tribe, Scapula turned his attention to the Silures in South Wales where Caradoc had been operating with great success against Roman incursions. Now comes a passage in Tacitus that is hard to understand in military terms. He tells how Caradoc cleverly transferred his operations into the territory of the Ordovices tribe around what is today Merioneth and Caernarfon. This is quite puzzling because it is normally the attacker, in this case the Romans who would have the initiative, but here it seems to have been in the hands of Caradoc as he decided where the campaign should be conducted. We may consider that because of the tribal problems that Scapula had encountered Caradoc used this time to consolidate his forces and move into that part of the country where his greatest strength lay. [14]Tacitus, ‘The Annals of Imperial Rome’, XX, 33.

Scapula’s plan of campaign for the year A.D.51 was simple, with two strong military bases, one at Kingsholm in Gloucestershire, and the other at Wroxeter in Shropshire, the Roman commander would move his two legions, XIV Gemina and XX Valeria from south and north, squeezing Caradoc between them. For his part the British chieftain, possibly because he had to do something to keep his coalition of tribes together, and a decisive battle would do this, together with the promise of loot that this would engender, now offered battle on ground of his own choosing.

Various sites of the Battlefield

Over the years many sites have been put forward as the place where Caradoc made his last stand against the might of Rome. Many of these are pure guesswork, and many are based on the description given by Tacitus in his account of the battle. Once again he is the only real source we have to go by, and the fact that it was written many years later, and that Tacitus himself never saw the battlefield means that we must allow him a good deal of artistic license in his dramatic description of the site.

Without going into the details of other possible areas where the battle may or may not have been fought, let us restrict ourselves to the region around the hill fort of Caer Caradoc, overlooking the three main rivers that have been speculated upon more than others-The River Teme, The River Redlake, and the River Clun. [15]        See enclosed map of the area around Bucknell, Brampton Bryan and Purslow Wood, O.S.Map 137;Ludlow and Church Stretton.

In his book,’ Battlefields of England’, volume II, Colonel A.H.Burne puts forward his own theory of where the battle took place. Burne suggests that the area around Purslow Wood, Black Hill and Clunbury Hill was the place where Caradoc ranged his army to meet the Roman legions as they crossed the River Clun. At first sight the ground here looks well suited to the defense of Caer Caradoc hill fort; however it is far too great an area to be covered in strength. Burne says that,’ It was desirable to hold Clunbury Hill as well as Purslow Wood. Moreover it would be unnecessary to hold the whole extent of the line everywhere in strength’. [16]   Colonel A.H.Burne,’ Battlefields of England’, Vol II, page 6  Since the extent of Caradoc’s position here would be close to three miles wide, he would have needed at least 30-40,000 troops to cover the terrain. If we give an average of 10,000 men to the mile, and allow for a reserve of 10,000, this still is far too thin a screen to have prevented a concerted effort by two legions and their auxiliaries from being overwhelmingly strong at one point. How long it would have taken Caradoc to pull-in troops from two miles away Burne does not recount. Burne’s own figures for Caradoc’s army he estimates at 15,000. This number, spread over three miles would have meant an average of only 5,000 per mile, without reserves, or approximately three, or at most four men per meter, hardly sufficient to hold back a Roman attack in strength. One last point that Burne does not consider is the fact that his chosen site is susceptible to being outflanked by cavalry, as the ground to the British left flank is excellent manoeuvring country.

The name Tacitus gives to the river is “amnis”, which could mean almost anything; however we do know that it was not the River Severn, as Tacitus calls this river, “fluvio”, a term still used today in relation to Britain’s longest river. The Clun is far too small, even though an article in the, ‘Archaeologia Cambrensis’ states that it formally carried a larger body of water. [17]            Ibid, page 8  If this is the case, then the River Teme must have carried an even greater amount, and is I think the river we are looking for. The River Redlake, which flows between the Clun and the Teme, has long had a reputation among the locals in this area as being a place where a battle took place, and if my own speculation is correct then it did indeed play a very important role in Caradoc’s position.

It is of the utmost importance when viewing the site of a battlefield that is almost two thousand years old, to consider the changes that have taken place in terms of natural erosion of the landscape, and the impact of agriculture and deforestation. No large-scale drainage was undertaken until the sixteenth century, and consequently river valleys would have been marshland and flood plains. The crossing points of rivers would have been well pronounced, with tracks leading to fords and causeways, but many of these would not have been adequate for the passage of large armies.

There is the possibility however that just such a place did in fact exist. Midway between the town of Knighton and the village of Lentwardine, on the A4113, at what is now Lingen Bridge over the river Teme, there are two weirs. These I speculate are what Tacitus calls, “vardo incerto” (shifting fords?). The crossing point is indeed very ancient, and the existence of a bridge here shows that it must have been a focal crossing point on the river. The weirs were probably fords, converted at a later date, and if this is the case then they could have been made even more accessible by the Roman legions constructing pontoon bridges or causeways. Why I choose this site is because in 1983 aerial photography discovered two large Roman marching camps; one at Brampton Bryan, the other at Walford. [18]S.S.Frere and J.K.S.St.Joseph,’Roman Britain from the Air’, page 98  These camps lie directly in front of the two weirs at Lingen Bridge. The Brampton Bryan camp is enormous, covering some 64 acres (25.9 ha), while the one at Walford is also of a substantial size at 25 acres (10 ha). The Romans never built a camp larger than they could defend, therefore the Brampton site shows the presents of at least two legions, together with their auxiliary troops; in all around 15,000-20,000 men, while the one at Walford may have been used for the cavalry, say 2,000-3,000 men and horses.

The reason for Scapula concentrating his forces here becomes obvious when we consider that it was Caradoc who offered battle. After years of disappointment at not being able to get to grips with the slippery British chieftain, Scapula now had a chance to destroy the last real threat to his conquest of Wales. He would not have been interested in any wide outflanking manoeuvres which would allow the enemy once more to retreat into the mountain fastness, nor would he be inclined to pursue any complicated military tactics that might jeopardize his chances of success. A set -piece battle, in which the Romans could use their discipline and drill to full effect, would do the job in short order, with no frills attached. Also it is possible that Scapula was aware of the morale of the British army, which may or may not have been eroded due to tribal disputes and needed a victory to reduce any signs of dissention.

My Own Speculative Site.

Let us now examine the ground on which Caradoc formed his forces for battle. The Rivers Teme and Redlake become separated by less than a mile of ground between the village of Bucknell and Lingen Bridge, an old Iron Age fort on Coxall Knoll, which sits between the two rivers at this point would not have been occupied because of its isolation, and therefore easily surrounded and cut off. Caradoc’s right flank is well guarded, not only by the Teme, but also by the steep gradient, which stretches back to Holloway Rocks. His left flank is similarly safeguarded by the marshy ground of the Redlake valley, which also becomes steeper on both sides as it cuts its way past Bucknell Wood and Bucknell Hill.Caradoc would most probably have placed a screen of skirmishers to contest the river crossing, as here the Romans would have been constricted in their deployment until they could gain enough ground to extend their battle line. Here the defending Britons would have poured-in a hail of missile fire as the legions attempted to shake themselves out and form their customary dispositions; we may be sure that they also returned fire using their deadly accurate artillery pieces, the “ballista”. These bolt-firing machines, resembling a huge crossbow had enough force to pass through steel, wood, flesh and bone and come out the other side.

The crossing would only have been contested for a short time as Caradoc would then have pulled back all his forces to the rising ground at the foot of Bucknell Wood (the wood is modern, and the hillside here would probably have been cleared to allow a clear view from the Caer Caradoc Hill fort). Here the British army would be able to contest the whole front of their position with maximum strength. The fallback line to the crest of the hill becomes progressively steeper and at the summit near Stow Hill a fighting withdrawal could be made to Caer Caradoc. All in all one may say that the site chosen fulfils all the criteria of a sound military position. Not only did Caradoc rely on the natural defenses of the terrain, he also strengthened them with the addition of dry stonewalls, from behind which the defenders could seek protection from Roman missile fire. [19]   Tacitus,’The Annals of Imperial Rome’, XII.33

The Opposing Forces

Before we look at the battle itself, we must first take into account the fighting qualities of both armies. Centuries of conquest and civil wars had made the Roman Army an awesome fighting machine. Its discipline was second to none, and its morale and confidence wren of the highest order. Each legion contained about 5,000 men, arranged into ten cohorts of around 500 troops each. The drill and battle movements were ‘a series of time-saving and orderly movements of line into column, and of echelons and squares, and in battle of the replacements of exhausted units and men by fresh ones’. [20]General J.F.C. Fuller, ‘Julius Caesar, Man, Soldier and Tyrant’, page 81  The tools of war consisted of one or two throwing javelins,’pilums’, a short thrusting sword,’gladius’, and a rectangular semi-cylindrical shield, which was made of two layers of wood glued together and covered with canvas and leather; their upper and lower edges were rimmed with iron to resist the enemy sword blows, and they had a metal boss at the center. The shield measured two and a half feet wide by four feet in length. [21]Ibid, page78

The auxiliary troops attached to each legion numbered about 4,000-5,000 men, but they were not just there to bolster the legion numbers. These, ‘auxilia’, as they were known played an important role on the battlefield, and helped avoid a dilution of the citizen manpower of Rome. [22]Edward Luttwark,’The Grand Strategy of the Roman Army’, page 40 Their main job was to act as light infantry and cavalry, engaging the enemy in skirmishing, information gathering and general hit and run tactics on the battlefield; they could also however be used in a full scale assault alongside their heavy brothers in the legions when the need arose. They were armed in a similar fashion to the legionaries, but the sword was larger and the shield could be oval. They also specialized in the use of various light infantry weapons such as the bow and arrow and the slingshot.

Much criticism has been leveled at the numbers and use of the Roman cavalry. [23]General J.F.C.Fuller,’Julius Caesar, Man, Soldier and Tyrant’, page 74  Suffice it to say that, allowing for the fact that the Roman’s never did master the art of cavalry tactics as they did for their infantry, their mounted troops were nevertheless a formidable force when dealing with the pursuit of a defeated enemy, or for sudden strikes against a disorganized flank.

Roman command structure, at least on the battlefield, was of a high order. Writing around A.D.60 Josephus tells us that,’ At Dawn the private soldiers reported to their tribunes to salute them, and the tribunes accompany all their superior officers to headquarters, where the commander-in -chief, in accordance with routine, gives them the password and other orders to communicate to their subordinates. They act in the same orderly way on the battlefield changing direction promptly as required, and whether attacking or retreating move as one man.’ [24] Josephus,’ The Jewish War’, III.87

The Romans also had one great advantage over the Britons, and that was the fact that their soldiers knew what to expect if they were wounded or became ill while on campaign. They could rely on a very advanced medical system, which was so sophisticated, that it could extract wounded men from the battlefield, and had the ability to apply tourniquets and restorative medicines. These factors enabled the Romans to maintain a high level of morale.

The Britons, or Celts still used war chariots, but we may say with some certainty that Caradoc would not have burdened himself with many of these owing to his need to effect a quick getaway and further concentration once any initial misfortune had overtaken part, or all of his army. The main strength of Caradoc’s force was its foot soldiers, and although some Celtic tribes used light cavalry, these would not amount to more than a few hundred in Caradoc’s army; not because they were not valuable for recognisance work and light raids, but that they required large amounts of forage which would necessitate scattering troops over a wide area in order to requisition (or steal) enough for their needs in all seasons, this in turn would slow down any chance of a rapid concentration.

When forced to give battle on the open plain the Celtic tribes were no match for the Roman legions, but if they could catch the legions strung-out on the march, or without the protection of their marching camps the Romans could come-off far worse against a rapid and well-planned surprise attack. Caradoc’s plan would be to lure the legions onto ground that would cause them to concentrate their attack on a constricted front, while having to assault against a hill position from where the Celtic missile fire could be used to greatest effect.

Caradoc was also aware of the uselessness of the Celtic hill fort. Time and again the Roman’s had demonstrated their ability to surround these massive constructions and either smother them with such a concentration of fire as to render movement inside them all but impossible, or just sit outside and starve the defenders into surrender. The fact that Caradoc took-up a position near his base of operations (Caer Caradoc), and not inside it shows how much he had learned from the disasters that had befallen other such attempts at checking the Roman advance by withdrawing into one of these death traps. [25]Phil Baker, ‘The Armies and Enemies of Imperial Rome’, page 4

The warrior army of Caradoc would have carried a wide selection of weapons; these included the large Celtic sword and shield, both of excellent workmanship. The quality of their metalwork was also seen on their horse harness and other accoutrements. It has long been thought that the Celts ran around semi-naked with their bodies covered in wode; however, depending on the time of year and the conditions under which they were fighting, most of Caradoc’s army would have worn some form of woolen cloak and baggy trousers. They were not hairy like the Germans, but wore their hair long and shaved their bodies. The moustache was worn but not the beard and helmets of very ornate design would sometimes protect their heads. Like other European tribes, the British Celts were expert in the use of the bow and arrow and the sling-shot; however they do not seem to have developed a throwing spear that was compatible to the Roman “pilum”, and in the main their heavier and longer weapons were used to thrust at the enemy. One of the most terrifying sounds on the battlefield would come from the Celtic horns (carnyx), which were sounded en-mass together with the ferocious yells of the warriors, many of whom were probably well fortified with strong beer and wine. [26]Derek Williams,’ Romans and Barbarians’, page 12-13

The Battle

We have no idea at what time of the year the battle was fought, but it would be nice to think that it was a sunny day, and both sides were able to display their full military panoply; however for such a bloody occasion perhaps a typical dull and damp British morning would be more fitting.

The Roman camp would be astir well before dawn, at which time the heralds would ask the legionaries if they were ready for war, to which they would reply,”ready, ready, ready”; thereafter, with their wolf-skin head dressed standard bearers at the head of the columns the Roman army would begin to march out of camp. Having already ascertained that the river (Teme) was fordable, the Roman cavalry would be sent upstream and downstream of the crossing point to break the force of the water, upstream, and to catch those who were carried away by the current, downstream; meanwhile the auxiliary troops would be first to cross over and engage the enemy while also endeavoring to form a bridgehead which would allow for the deployment of the legions.

As I have said, it is possible that Caradoc disputed the crossing for a short time, but we may be sure that he would have formed his army to show a solid front rather than have left any isolated units to be surrounded and cut off by the Roman cavalry. On their side the Romans would have found their mounted troops of little value owing to the restricted nature of the battlefield, and the chance of any outflanking manoeuvre by these troops would have been negated because of the security of Caradoc’s left and right flanks.

Spreading out from the bridgehead the legions would form into their customary battle formations while the auxiliaries continued to harass the British position with missile fire and faint attacks, gaining time while the legions formed-up. Tacitus tells us that the Romans (auxiliaries?) came off worse in this exchange of fire. [27]Tacitus,’The Annals of Imperial Rome’, XII.33  This would seem to confirm Caradoc’s dispositions, as the Roman’s would be attacking uphill into a hail of missile fire being poured down upon them from the Britons stationed on a higher gradient; this fire was hard to return with any effect owing to the Roman’s being unable to use their throwing javelins while climbing the hill, and also the stone walls that had been built in front of the British line which deflected the arrows, ballista bolts and slingshots that were being returned.

The British army under Caradoc would seem to have been better disciplined than the Anglo Saxons under King Harold at the battle of Hastings in 1066, at least their leader kept a firm hand and did not allow them to break formation in order to attack the Roman’s as they scaled the hill. Instead the Roman auxiliaries had to pull back and reform, while the main body of the legions grouped into their famous “Testudo” formation of locked shields. [28]    Tacitus,’The Annals of Imperial Rome’, XII, 33 There would probably have been several of these formations, consisting of anything from 50 to 100 men in each. Why the Romans had to use this type of grouping can only be speculated upon, as they normally only formed the Testudo when attacking a enemy encampment or against the walls of a besieged town. It is possible that Caradoc’s roughly constructed stone walls were higher-up the hillside, therefore the Roman’s had to seek protection under cover of their locked shields to avoid the constant hail of missiles that were being poured upon them; a fire which had already caused the auxiliaries to fall back. The legionaries may have been followed by the auxiliary troops, rather in the way that infantry followed close behind tanks in the Second World War. Ten or twelve of such groupings would move forward in a chequerboard formation so that if one or two were held back by a blocking force of extended spears or staves, others could come in at intervals along the line and exert the pressure still further. This tactic did the trick, and the Romans managed to pull down the stonewalls and come to grips with their opponents. Once the legionaries got to work with their thrusting swords the result was pure butchery. Using the shield to push into the body and face of their enemies the Roman infantry would cut-down all before them, being described by one historian as,’ like the action of a buzz-saw’. [29]Dr Richard Gabriel, quoted in the B.B.C2 television program, ‘Timewatch, The Roman Way of War’.With little or no protection in the form of body armour, the Britons were forced back fighting off the Romans on three sides, for as the legionaries attacked their front; the auxiliaries now fanned out and came in on both flanks. [30]Tacitus,’The Annals of Imperial Rome’, XII.33

Tacitus tells us that Caradoc’s wife and family were taken prisoner, and that his brother surrendered. They were in all probability captured in or around Caer Caradoc Hill Fort; however Caradoc himself made his escape and sort refuge with Queen Catimandua. Her loyalty to Rome appears to have been solid for she turned Caradoc over to them, thereby ending the military career of one of Britains first great warrior leaders. [31]Tacitus,’The Annals of Imperial Rome, XII.33

Caradoc and his family were taken to Rome where, in spite of all the problems he had caused, and the vast amount of time and money involved in the military operations against him, he was allowed to live out the rest of his life in comparative peace and quiet. The Romans were well aware of Caradoc’s fighting qualities, and it must have bolstered their inflated egos to have this man walk among them free, while at the same time knowing he could no longer cause them any problems.

As for Ostorius Scapula, he became so exhausted by his responsibilities, that shortly after Caradoc was taken to Rome he died. [32]Ibid. 37 Thus, without receiving his triumph, Scapula lay dead in Britain while Caradoc walked the streets of Rome.

Other Sites to visit

As well as the position of the battlefield given above, and the site put forward by Colonel Burne, let us look at other possible locations that have been mooted over the last one hundred and fifty years or so. The reader will also find sites given in Colonel Burne’s own account of the battle. [33]Colonel A.H.Burne,’Battlefields of England’, Vol II, page 12

Caradoc’s forces, because of being easily surrounded, as I have said, would not have occupied Coxall Knoll. The main reason why this has been seen as the site of the battle is because of the Iron Age Hill-Fort situated on Coxall Knoll itself. There can be no doubt that the fort would have been very effective in guarding the crossing points of the river Teme in this area, but would have proved ineffective against the skills of the Roman military machine; however do go and visit this site as it is close to the proposed crossing point of Scapulas army, and well worth a visit in its own right.

Holloway Rocks, two miles north -east of Knighton is another site put forward as the battlefield. Anyone walking the road from The Stud Farm, just off the A488 near Coed detton, towards Stowe will soon realize that no commander in his right mind would have expected troops to ascend the steep climb up to Holloway Rocks in the face of determined opposition, to say nothing of having to make the climb in full armour and to be expected to fight a pitched battle! The main reason for considering Holloway Rocks is that they served to secure Caradoc’s right flank against any Roman turning movement.

The site chosen by Colonel Burne has been dealt with in the text, however not only should we take into account the fact that his battlefield is far too large for the forces engaged, (It is almost twice the size of the position taken-up by the Allies at Waterloo) but also the fact that no Roman Marching Camps have been found in this area; these are almost always the precursor of a legion or an army securing itself before giving battle.

Finally it must be remembered that my own ideas concerning the site of the battle are, like others, only speculative. The two Roman marching camps at Brampton Bryan and Walford have not been dated and may indeed be of a later date than Caradoc’s last battle; however, even if this is proven to be the case we should still consider why the Roman’s would have concentrated such a large force in this area, it maybe that there was another encounter at or around these sites that has as yet not been discovered.

December 2000.


Cefn Carnedd Hill Fort

After a recent visit to Cefn Carnedd Hill Fort in Wales I decided to include this brief summery regarding the possibility of it being an alternative site for Caradoc’s last stand against the Romans.

Many historians believe that this area was indeed the place where the battle was fought, and it does meet almost all of the criteria described in the sources that are available to us concerning the battle, namely the description given by Tacitus in his work, “The Annals of Imperial Rome.”

The Caersws Basin. (Clwyd-Powys Archaeological Trust)

The Caersws Basin.
(Clwyd-Powys Archaeological Trust)

The River Seven meandering its way through the Caersws Basin flows at the foot of Cefn Carnedd and does have all the appearance of presenting a substantial barrier to an invading army. The steep slopes leading up to the hill fort would also greatly hinder the progress of any direct assault. There are also a series of mounds and ridges rising from the river valley and becoming steeper as you approach the hill fort which, if we are to believe Tacitus, could have been strengthened by the addition of rough built stonewalls. The countryside abounds with Iron Age enclosures and, ‘…A large oval enclosure surrounded by an interrupted bank and ditch has recently been discovered just to the north of Caersws, and excavations have provided an Iron Age date from the ditch silts.’ [34]Historic Landscapes Characterisation, Clwyd-Powys Archaeological Trust.

Anyone visiting the site cannot fail to be impressed not only by the beautiful landscape, but also, for all those who have an interest in Caradoc, by the fact that the place seems to fit-in exactly with the ground described by Tacitus. However, there is one problem, which, for the moment, cannot be resolved. No Roman marching camps have been discovered in the vicinity, and although there is a Roman fort just east of Caersws village, and that it dates from around AD 50, with a new fort replacing it in AD 75, these structures are far too small to accommodate the size of the Roman army that engaged Caradoc in battle. The initial marching camp, like the one at Brampton Bryan described on this site, would have covered over 60 acres (or over 20 hectares). Since no such camp has thus far been located then, sadly, no firm claim can be made that Cefn Carnedd Hill Fort was the place where a battle occurred.

I am very grateful to Robert Lee and Dr Bob who accompanied me on the visit to Caersws and Cefn Carnedd which, despite a brief panic attack when it was thought that the car keys had been lost, turned out to be a most pleasurable experience – thank you chaps!

Drawings and Photographs

Panoramic view from Cefn Carnedd Hill Fort

Panoramic view from Caer Caradoc.

Tribes in Wales at time of Roman Invasion (Wikipedia)
Cefn Carnedd Hill Fort (Royal Commission on the Ancient and Historical Monuments of Wales)
Cefn Carnedd Hill Fort (Royal Commission on the Ancient and Historical Monuments of Wales)
Roman fort near Caersws (Royal Commission on the Ancient and Historical Monuments of Wales)
Roman fort near Caersws (Royal Commission on the Ancient and Historical Monuments of Wales)
View of the Severn Valley from The summit of Cefn Carnedd Hill Fort. Photograph, Robert Lee 2005
View of the Severn Valley from The summit of Cefn Carnedd Hill Fort. Photograph, Robert Lee 2005



View from Caer Caradoc Hill Fort. The ground to the left of the photograph falls way to the River Redlake and to Bucknell Hill at the right.
View from Caer Caradoc Hill Fort. The ground to the left of the photograph falls way to the River Redlake and to Bucknell Hill at the right.
Proposed plan of the battle.
Proposed plan of the battle.







Baker, P. ‘The Armies and Enemies of Imperial Rome,’ W.R.G. 1972

Bowman, Alan. K.Life and Letters on the Roman Frontier,’ British Museum Press, 1998

Burn, Alfred H.The Battlefields of England,’ Vol I and II, Greenhill Books, 1996

Ellis, Peter Berresford.The Celtic Empire,’ Constable, London. 1993

Fuller, General J.F.C.Julius Caesar, Man, Soldier, and Tyrant,’ Da Capo Press, New Jersey, 1965

Joseph, St, J.K.S and Frere,S.S.. ‘Roman Britain from the Air,’ Cambridge University Press, 1983

Josephus, ‘The Jewish War,’ Penguin Classics, 1981

Legg, Rodney. ‘Romans in Britain,’ London, 1993

Levick, Barbara.Claudius,’ Batsford Books, London, 1993

Luttwark, Edward.The Grand Strategy of the Roman Empire,’ John Hopkins University Press, 1979

  1. The Annals of Imperial Rome,’ Penguin Classics, 1996

Williams, Derek.Romans and Barbarians,’ Constable, London, 1998


Aerial Photographs

Map of area shown in aerial photographs.  A & B show some woods marked on the photographs, two trees are at C and a bridge at D. Approximate location of camps shown by green hatching. Click photographs below to enlarge.


The Author would like to thank the following for permission to reproduce photographs and plans appearing in this article.

Mrs. Alice Tsui and Mr. Philip Sticker, together with the Committee of Arial Photography, University of Cambridge.

Shirley and Gary Brueggeman for Roman Camps. Vist their splendid Website at:

The National Museum of Wales.

And especial thanks to Lou for putting up with me and Dr Bob for his skill in web design and typography.



1 Caligula,” Little Boots”, a name given him as a child.
2 Rodney Legg,’Romans in Britain’, page 10
3 Barbara Levick,’Claudius’, page 139
4 No real information on just how many ships were used in the Claudian Invasion, but see General J.F.C.Fullers book,’ Julius Caesar, Man, Soldier and Tyrant’, page 124. Here there is a very interesting account of Caesar’s shipping figures for his invasion of Britain in 54 B.C.
5        Peter Berresford Ellis,’ The Celtic Empire’, page 199
6 Graham Webster,’ Rome against Caratacus’, page 13
7 See enclosed map
8 Barbara Levick, ‘Claudius’, page 144
9 Graham Webster, ‘ Rome against Caratacus’, page 15
10 See enclosed map.
11 Graham Webester,’Rome against Caratacus’, page 20
12 Anthony Barrett, ‘Historia’, No 17, page 227
13 Tacitus, ‘The Annals of Imperial Rome’, XII, 30-31
14 Tacitus, ‘The Annals of Imperial Rome’, XX, 33.
15         See enclosed map of the area around Bucknell, Brampton Bryan and Purslow Wood, O.S.Map 137;Ludlow and Church Stretton.
16    Colonel A.H.Burne,’ Battlefields of England’, Vol II, page 6
17             Ibid, page 8
18 S.S.Frere and J.K.S.St.Joseph,’Roman Britain from the Air’, page 98
19    Tacitus,’The Annals of Imperial Rome’, XII.33
20 General J.F.C. Fuller, ‘Julius Caesar, Man, Soldier and Tyrant’, page 81
21 Ibid, page78
22 Edward Luttwark,’The Grand Strategy of the Roman Army’, page 40
23 General J.F.C.Fuller,’Julius Caesar, Man, Soldier and Tyrant’, page 74
24 Josephus,’ The Jewish War’, III.87
25 Phil Baker, ‘The Armies and Enemies of Imperial Rome’, page 4
26 Derek Williams,’ Romans and Barbarians’, page 12-13
27, 30 Tacitus,’The Annals of Imperial Rome’, XII.33
28     Tacitus,’The Annals of Imperial Rome’, XII, 33
29 Dr Richard Gabriel, quoted in the B.B.C2 television program, ‘Timewatch, The Roman Way of War’.
31 Tacitus,’The Annals of Imperial Rome, XII.33
32 Ibid. 37
33 Colonel A.H.Burne,’Battlefields of England’, Vol II, page 12
34 Historic Landscapes Characterisation, Clwyd-Powys Archaeological Trust.

Leave a Reply

Your email address will not be published. Required fields are marked *