Rome's Greatest Defeat

by Adrian Murdoch

  The Roman troops could make no serious defence. Under normal circumstances, the cavalry would have been deployed as a protective fence with detachments on all sides. But this was impossible in a forest. And it was no easier for the infantry. They were also struggling because of the difficulties in the terrain and hampered by civilians and their baggage wagons. In the confusion and the rain, the number of casualties from friendly fire must have been high. It is curious that Varus appears to have organised his train so badly. Certainly it was one of the more serious criticisms levelled against him by Roman authors. ‘The Romans were not proceeding in any regular order, but were mixed in helter-skelter with the wagons and the unarmed, and so, unable to form readily anywhere in a body, and being fewer at every point than their assailants, they suffered greatly and could offer no resistance at all,’ writes Cassius Dio.14

  Yet again, the answer lies in the level of Varus’ faith in Arminius. We shall never know exactly when it was that the general realised how misplaced this had been. But certainly it was not quite yet. There was no need for Varus to insist on war-readiness, because he was marching through not just subdued, but actively loyal, territory.

  Despite the assault and the shock that it caused, Varus maintained admirable control and presence of mind. He knew what he needed to do. He built a camp, of which sadly no trace has been found. Under normal circumstances, as with everything that the Romans did, there was a procedure to this. Ten men detached from each century carried coloured flags and marched in the front of the army. They were charged with identifying and plotting out a camp every afternoon, the coloured flags signifying the different areas of the camp; the gates, the commander’s tent and so on. When the rest of the army arrived, they set to work digging the protective ditches and ramparts, hammering in the wooden palisade, all before meals or sleep could be contemplated.

  Here, there would have been no such luxury. The enemy were all around. The cavalry and half of the infantry would have had to stand, battle-ready, protecting their colleagues digging the ditches, building the ramparts and erecting the palisades. Under normal conditions it took three to five hours to complete digging a ditch. Under these circumstances, with the enemy and rain, did they struggle through or did they cut their labours short? Tents should have been put up before work on the ditches and ramparts started, but one wonders whether, in the confusion of battle, this was done.

  A sign of how serious the attack was is that Varus insisted that the baggage train be burned or abandoned along with everything that was not strictly necessary. But there is no reason to think that Varus believed that his army would not survive at this stage. He had bought himself a brief respite. There was little chance that the Germans would attack a Roman camp directly. As the senior commanders met that evening to form a plan of action, it is likely that they agreed to aim for a river, where the army could pick up transport to take it to Haltern and then down to Xanten.

  The next morning, one day after the attack, the Romans left their camp and began the march towards the oak-covered slopes of the Kalkriese Berg. Arminius will have known with moderate certainty that this was the general direction his opponent would take. It is wrong to suggest that the Romans were being herded in this direction. The point is that there was no need to lure the Romans to their doom or force them into the corridor: they were always going to come this way. It was pretty much the only option open to Varus, where the main west–east routes from the mid-Weser to the lower Rhine converged, avoiding more difficult terrain to the north and the south. Indeed, until 1845 this was one of the main routes through the area and it is preserved on maps up to that time as the Alte Herrestraße – the Old Military Road.

  Generally speaking, however, although there was woodland, it was farmed and cultivated rather than wild and ancient forest. This now was not the landscape of thick, dark oaks that the more hysterical ancient commentators mention. The obsession that modern historians have with trying to reconcile the Teutoburg Forest that Tacitus mentions with the landscape could plausibly be the result of a misinterpretation. As has been explained above, while the phrase that the Roman historian uses, Teutoburgiensi saltu, can indeed be translated as ‘Teutoburg Forest’, it may also be rendered ‘Teutoburg Pass’.15 At times, Varus would have looked out to see a damp, agrarian landscape not unlike the Fens of England. It had been farmed for millennia, while wetter, marshier areas were a source of pasture or used for timber. When looking at landscape like this, Tacitus’ comment that the Cherusci were ‘experienced at fen-fighting’16 begins to make sense.

  As they came round the Kalkriese Berg, the Romans were approaching a narrow pass called the Kalkrieser-Niederweder-Senke, the mountain rising up 110m above the pass to the south, the Great Moor to the north. It would be difficult to think of a more perfect spot for an ambush, now accented by the archaeological traces which let you see how the deceit was accomplished. The analogy has sometimes been used that it is like a lobster pot that allowed the Romans in, but not out. The Kalkrieser-Niewedder-Senke is a narrow corridor, some 6km long and only 1km wide. But because of the high water table at the time, it was only passable at the edges, on the ridges of sand that had accumulated, 200m or so wide. At the same time, with the nearest Roman relief forces some 100km to the south, there was no chance of the alarm being raised.

  Varus and his men had been harried all the way. The Germans saw no moral virtue in standing their ground against heavily armoured Roman legionaries. Instead, the constant attack then retreat of the Cherusci began to take its toll. But the losses the Roman forces suffered there were minimal compared to the hell that was now unleashed as they marched into the pass. This is also where archaeology begins to add another dimension to our knowledge. This is the site that has been found. Arminius attacked again, this time with the full weight and anger of the forces with him.

  The Cheruscan commander had had the opportunity to line the pass with arc-shaped turf walls and sand ramparts that curved with the shape of the hill. Three have been found to date, with a total length of 400m. Speed was of the essence for him and his troops. From the construction and variety of materials used it is clear that they were built quickly with anything that came to hand (turf predominates where there were meadows, sand at the eastern end, and a mixture of turf, sand and limestone at the western end), all of which suggest that the work had taken at most a few weeks to complete. The style of build of the walls is peculiarly Roman; indeed when they were first discovered it was thought that they were rapidly constructed Roman defensive positions. Arminius had learned the lessons of Rome well.17

  Despite the speed and crudeness with which they were constructed, these walls were massively effective. Some 4 to 5m wide at the base – something you can work out easily by measuring the space between the drainage ditches that had been dug on the German side and the start of the Roman finds – they were not much more than 1.5m high, though in all likelihood this was raised by a palisade.

  It goes without saying that the ramparts did have a defensive role. And as just mentioned, on the inner side, facing the Germans, there were drainage pits. Given the weather, the pouring rain, these must have been significant in the hours before the ambush was sprung, as the natural clay in the soil will have stopped natural drainage. But the walls’ primary purpose was offensive. Arminius knew that if his men could attack a fragmented Roman tactical formation, they could beat it. For the Cherusci this was less about matching the Roman gladius than ‘out-psyching’ the legionaries. The forest might help the tribesmen achieve surprise, but it was the basic organisation of the column that needed to be shaped and then ruthlessly exploited. Not only did the very construction of the walls preserve the element of surprise for the Germans, it also managed to narrow the path, to guide the Romans into the dampest, most difficult part of the pass, where the legionaries could be massacred when the Germans leapt out through the gaps they had left in the walls.

  It was now that the worst of the fighting took place and the greatest Roman casua
lties occurred. If the human remains that have been found at Kalkriese provide only a very narrow snapshot of the thousands who died, then it is a terrifyingly high-definition one. Virtually all of the human remains that have been found are of men of military age, between 20 and 40 years old. All of their deaths come from offensive weapons, many of them from sword cuts.

  The Roman soldiers were cut down all too easily. Floundering through the pass it was impossible for them to form the impregnable legionary lines that had seen off Hannibal and numerous other enemies. Just as had happened on the previous day, the cavalry units hindered rather than helped. In trying to mount any kind of defence, let alone attack, both types of unit got entangled, crashing into each other. It is possible that Numonius Vala’s troops did not even fight on horseback. It was not unknown in forests for cavalry to dismount and fight on foot, certainly in an extreme situation like this. Their disorganisation made it even easier for the German forces.

  Arminius may have convinced his warriors of the benefits of Roman ambush techniques, even of using swords, but in terms of tactics it was a different matter. Here the native traditions had the advantage. Speed and agility were what counted and the chief hope of victory was a rapid, overwhelming attack. Could he have convinced them even if he had wanted to change their approach? That is doubtful. The indiscipline of the Germans might have been as much of a recognised literary theme as the softness of the Syrians, but it was their lack of restraint that made it so successful.

  A Roman historian who had himself fought the Germans, describes what Varus and his men would have been up against in those first moments: ‘The Germans rushed forward with more haste than caution, brandishing their weapons and throwing themselves on our squadrons of horses with horrible grinding of teeth and more than their usual fury. Their hair streamed behind them and a kind of madness flashed from their eyes. Our men faced them stubbornly, protecting their heads with their shields and trying to strike fear into the foe with drawn swords or the deadly javelins they brandished.’18

  In the face of this assault, there was no time to draw on any of the known and trusted techniques to avoid ambush. Army training recommended that soldiers feign fear and pretend to run away before regrouping. Here, though, their terror was real. They had nowhere to run. They could not re-form. Roman remains behind the wall in the drainage ditches suggest that some legionaries did go on the offensive, scrambling over the top to get at their attackers. Certainly the earth rampart, built without wooden supports, began to collapse even during the battle. From an archaeological point of view this was fortuitous as it protected finds, but in the heat of combat, the disintegration of the walls added to the confusion.

  The natural focus of the battle has been on the human element. But another dimension to our knowledge about the battle is added by the animal remains that have been discovered – the few mules that survived Varus’ burning of the baggage train. The remains of a mule were found in 1992 – or rather its skull, vertebrae and shoulder blade – along with the pendants and decorative glass pearls that had fallen off their mountings, highlighting the close bond between owner and animal. The iron clapper on the bronze bell round its neck had been muffled with a handful of oats its owner had grabbed from a field while passing, to keep them moving as silently as possible. A tangible sign of the state of extreme tenseness of the Romans who had survived the first day’s attack, it is also the analysis of the oats which has allowed archaeologists to date the battle to September. In the heat of battle the mule had broken free from its wagon, still wearing its metal harness, iron ring snaffles and iron rein chains. It fell in front of the wall, which collapsed on top of it, preserving the remains.

  Another, almost complete, skeleton of a mule was found in 2000 at the western end of the wall. Frozen in the moment of death, its head facing west and feet south, jaws still clamped on the snaffle, the animal had tried to escape over the wall and broken its neck in the fall. The wall, there built of sand reinforced with sandstone, then collapsed on it, protecting its body both from plunder and from scavenging animals. As Susanna Wilbers-Rost writes, ‘In Kalkriese, the collapsed wall has permitted the preservation of snapshots, which are otherwise discovered only rarely in archaeology.’19

  But the series of finds is not limited to the foot of the Kalkriese mountain, as might be expected of an army trying to push its way through the pass to safety. Traces can be found branching off, away from the main army, in a 2km-wide strip that leads north-eastwards to the edge of the Great Moor. It is clear that one part of the army split off and tried to make a break for it along the sandbars through the boggy forest.

  It is tempting to ascribe this attempt to Numonius Vala’s flight. In Velleius Paterculus’ account, Varus’ legate, his deputy, did try to escape, taking the cavalry with him. It is always dangerous to argue from a lack of evidence but there are remarkably few traces of cavalry and horses at the foot of Kalkriese mountain.

  Were Vala’s actions cowardice and a loss of nerve or a calculated saving of skin? We shall never know. Paterculus, as a cavalry man himself, is damning, suggesting that Vala was deserting the legionaries, trying to reach the Rhine and the safety of the Roman zone. Certainly when Varus’ deputy left, he was effectively deserting the legionaries. But it is also possible that he was acting on orders. The cavalry was proving a liability in the narrows of the Kalkrieser-Niewedder-Senke. For Varus to send Vala away might have seemed appropriate, though we are firmly in the realms of speculation here. Whatever the circumstances, he did not make it and his troops were slaughtered to a man. ‘Vala did not survive those whom he had abandoned, but died in the act of deserting them,’ writes Velleius Paterculus coldly.20

  The fourth day since they had left Minden brought no let-up from the weather; if anything it got worse. The violent winds made throwing javelins impossible and rendered the surviving Roman archers useless. The legionaries could not even defend themselves properly. The rain had soaked through the leather of the heavy Roman shields into the wood, making exhausted and demoralised soldiers carry even more weight.

  It was perfect weather, however, for the Germans. Given their light equipment, the rain and the wind if anything made their guerrilla attacks easier. They could pick off the enemy and quickly retire before the sodden Romans were able to react. Varus and his forces might have been exhausted – they had after all marched through the night – but Arminius was able to draw on fresh men keen for a fight. Overnight, the German ranks had been swelled by other tribes which initially had kept their distance, cautious about throwing their lot in with Arminius. Now, however, they had no such qualms. A much-weakened Roman army and the thought of plunder were enough of an inducement to overcome any misgivings.

  Ranks thinned, surrounded by Germans, with even the weather against them, Varus and his senior officers took their final, unimaginably difficult decision. With ever-strengthening German forces and two eagles already captured, it must have been apparent now that the Roman forces had no chance of escape. They would never make it back to safety. The senior commanders were already wounded and there was only the question of how they would die.

  Tales of Germanic blood rites were well known and only increased their desperation. Varus and his colleagues would all have heard how priestesses, dressed in white, met those unfortunate enough to be taken prisoner. These POWs would be crowned with wreaths and then led, one by one, to platforms suspended over great bronze cauldrons into which their throats would be cut. It was irrelevant that this dated back centuries. The story was still current, preserved at the time of the disaster in Rome by the Augustan geographer Strabo.21

  What was worse, suicide or capture? Ceionius, one of the camp commanders, decided to surrender. His colleague, Lucius Eggius, the other legionary commander, had already died in battle. A wounded Varus decided to die by his own hands. Even this, the final gesture of the commander-in-chief, did not pass by Velleius Paterculus’ jaded eye. ‘The general had more courage to die than to fight, for, following the ex
ample of his father and grandfather, he ran himself through with his sword,’ he wrote.22

  In more modern times, Paterculus’ comment has usually been reproduced without comment because it chimes with contemporary thoughts on suicide. To Christians, Jews and Muslims, it is a sin. Yet other Roman commentators were more understanding and sympathetic. For Cassius Dio, the action was ‘terrible yet unavoidable’; for Florus it was ‘noble’.23 The reactions of all three to Varus’ final action encapsulate the classical dichotomy with suicide. While impulsive suicide or some methods, like hanging or drowning, were condemned by Roman society, it was admired for reasons of self-sacrifice or to sponge away shame, just like Japanese seppuku. After all, Brutus, Cassius and Mark Antony had all committed suicide, as had, of course, Varus’ father.

  After their general’s suicide, the surviving Roman forces lost all hope. There was no chance of escape. Some joined Varus in suicide, others took their cue from Ceionius and surrendered. The manuscript of Cassius Dio breaks off in mid-sentence, his account mirroring history. ‘To flee was impossible, however much one might desire to do so. And so, every man and every horse was cut down without fear of resistance and the . . .’24

  Arminius had won. The pass was littered with dead men. The few who groaned had passed away or been dispatched in the night. Roman legionaries lay individually and in groups in unnatural positions: their necks were bent back or their heads face forward in the mud. Even the ones who lay on their back were clearly not at rest. The surprise of death on their faces, they stared up at the bleak German sky.

  As the battle was dying down, it was time for the mop-up operations to begin. Germanic troops looked to their wounds, tried to discover if their friends were alive and began the honours for their own dead. Their weapons were carefully collected. It was important for burial rites that a soldier’s arms were buried with him, something that accounts for the lack of native weaponry found by archaeologists at Kalkriese itself. But for Arminius’ opponent, not even the peace of death. Although the body of Varus himself had been partially burned and then buried by his adjutants, it was disinterred and mutilated by the vengeful Cherusci. The corpse was decapitated and Varus’ head was sent as a trophy to Maroboduus, along with an invitation to join Arminius in his war against Rome.