Underground Warfare 1914-1918

by Simon Jones

  The Ronville caves were by far the larger, some of them immense caverns hundreds of feet in diameter and twenty to forty feet high. Pillars of chalk had been left at regular intervals to support the roof, but in the centuries that had elapsed since their abandonment, falls had gradually formed them into inverted cones springing away to the domed roof between, so high that it was barely discernable in the candle light.

  To make these caverns safe and habitable was no easy matter, for as soon as they were opened to the cold wet winter air the chalk commenced to swell and crack, and slabs weighing many tons would come crashing down without an instant’s warning. To timber up to the heights of these roofs was out of the question, so instead the floor was raised by dumping therein the chalk cut from the galleries and dugouts until the roof was close enough for clear observation and support. (Lieutenant J.C. Neill, New Zealand Tunnelling Company)26

  These caves were to accommodate 11,500 troops and the largest, Christchurch, held 4,000. The caves were provided with water and electricity and protected by gas-proof doors. They carried telephone cables, exchanges and line testing-points. A tramline ran from the sewer to the St Saveur caves. Just over half a mile from the front line a cave was equipped as a main dressing station, as it was anticipated by VI Corps that the congestion in Arras during the attack would prevent the evacuation of wounded for treatment, so advanced dressing stations were equipped and manned to perform the functions of a main dressing station. These were established in caves, cellars and the basements of buildings. The chief was close to the 3rd Division trenches, two sloped entrances allowed stretchers to be carried direct from the communication trenches, while an approach at the back allowed motor ambulances to collect casualties. Named Thompson’s Cave after the Corps Director of Medical Services, it was able to accommodate 700 casualties on stretchers fitted in two tiers. It also contained dressing and operating rooms, kitchens and latrines. The accommodation close behind the front line enabled it to act as a corps dressing station for severely wounded requiring urgent attention. The cave was used for the first two days of the offensive until a large shell exploding on the top of the cave burst a water main, which caused the roof to collapse in two places.27

  The New Zealanders also constructed two long tunnels towards the front line, one linking the St Saveur cave system and the other the Ronville caves so that men could if necessary get all the way to the front line through the caves and tunnels and avoid especially the heavily shelled area of the railway station. The St Saveur tunnel ended in five exits from Russian saps in no man’s land and the Ronville with one, although others were probably under preparation. About seven miles of tunnels and subways were constructed in just over four months.28

  Plan of the main dressing station established in Thompson’s Cave for the Battle of Arras in April 1917. Thompson’s cave is shown in the illustration on just below Guernsey. From Medical Services General History, Vol. 3.

  Along the brightly lighted gallery a constant stream of traffic went safely from the heart of Arras to the front line – ammunition, rations and engineers’ supplies were no longer carried painfully by fatigue parties along the wet and much strafed communication trenches, but went easily forward on trucks dry and safe.29

  By January 1917 the Russian saps were being driven across no man’s land but, as at the Somme, progress was hampered by the saps being blown in. However, the most frequent cause was British shell fire, and on occasion trench mortar ‘toffee apple’ bombs when cutting the British wire fell short, trapping parties of tunnellers at the ends of collapsed tunnels. On one occasion a 50ft tunnel that had to be driven to rescue them was dug in a record time of nine hours. The saps were cut quietly through clay using push-picks and to avoid alerting the Germans the men wrapped their boots in sacking and speaking above a whisper was forbidden. When one of the saps holed into a German gallery it was carefully closed and charged with 2,200lbs of ammonal (this appears to have been the only mine blown on the Arras front on 9 April).30

  As this work was progressing, however, the Germans made a dramatic move which seriously dislocated the tunnellers’ preparations for the Arras offensive. In February they withdrew from the Somme battlefield up to 30 miles to the prepared positions of the Hindenburg line. These positions were described by a British staff officer as:

  …two great lines, from five hundred to two thousand yards apart, each consisting of front and support trenches. The system constituted in all probability the most formidable fortification constructed in the course of the war. The Germans had sited and dug it at their ease, to a great extent with gangs of Russian prisoners and forced civilian labour. The trenches were wide, deep, and well revetted. The mined dug-outs were all designed to a pattern; the stairways, supports, and all timber used in them having been turned out by the sawmills in replica by the thousand pieces. They represented the first successful application of mass production to the construction of dug-outs.31

  These new positions were to provide formidable obstacles for the Allies to assault in 1917 and 1918. Between 17 and 28 March the Germans withdrew additionally from the line south-east of Arras up to a point just south of the Cambrai road. The immediate consequence at Arras was that, after the British had moved forward to locate the new German line, the end of the Ronville tunnel was left 1,000yds behind the new British front line. Of nine Russian saps extended beyond the British front line only three, from the St Saveur tunnel either side of the Cambrai Road, were still actually in no man’s land. There was no time to lengthen the remainder and so on the day of the attack only some units of the 12th Division were able to pass underground all the way to their own front line or beyond. By the time of the attack the description in a War Diary was applicable only to that Division: ‘Now it is possible to get from the crypt of the Cathedral to under the German wire without braving one shell in the open.’32

  Equally, it was only the 12th Division that was to derive potential benefit from the three remaining Russian saps. Wombat bores were driven from two (I.54 and I.56) to form communication trenches linking them with the front line. From the other machine-gun emplacements were prepared which were to be broken out immediately before zero on 9 April. The bores were blown at zero, 5.30am, and soon afterwards were connected with the old German front line. Neill was stationed in the northernmost of the three saps (I.57), which was about 25yds from the German front line trench and from which machine guns were supposed to fire from forward emplacements to sweep the German parapet. Before zero, however, the plan went badly wrong:

  Our orders were to remove this cover just before Zero and to open the end of the gallery itself into No-man’s land, continuing it as an open trench into the German front line, immediately the battle opened.

  At 10 p.m. a concentrated bombardment with gas shells fired from trench mortars was opened on the enemy positions, continuing for more than an hour. Many of these shells fell short and landed about our gallery, through the much broken roof of which the gas percolated, filling the gallery.

  A gas curtain was hastily rigged just behind the first opening – an incline up to the front trench – and efforts were made by a fire to draw the gas from the gallery, but in vain.

  There was nothing for it but to go forward and open up the end so that a natural current of air would clear the gas out. These forward galleries were not electrically lit so it meant groping by dim candle light, through a hundred and fifty yards of narrow winding gallery with eyes streaming and smarting from the tear gas that got in under the box respirators. Arrived at the end desperate efforts were made to remove the cover to the open air, but to work effectively in a box respirator, almost blinded and after coming through the ordeal of the gallery, was a physical impossibility and constant relays of fresh men were necessary before at last the work was done and it was possible to tear off the suffocating gas masks and breathe the clear night air.

  A machine-gun section was waiting back behind the gas curtain to man the forward positions but not being miners they could n
ot be induced to face the still gas filled gallery and so arrived at their battle positions too late, though so successful was the attack that their guns were not required. A wonderful battle picture was the reward of the solitary watcher who remained on guard at one of these openings.

  As the watch hands crept towards the fateful half-past five the first faint streaks of the wintery dawn began to colour the sky so that the dim outlines of the German parapet close at hand and the shattered stakes and wires of his entanglements all round could be faintly seen.

  An ominous stillness had fallen along the whole battle line, almost terrifying after the infernal racket of the last few days.

  Suddenly three tremendous reports in quick succession – the company’s mines – a pause of a fraction of a second and then the enemy parapet disappeared in a spouting wall of flame, the opening barrage.

  The watcher turned back to look towards our own lines and saw wave after wave of dim figures advancing steadily at a leisurely walk – Jocks and Tommies – as steady and unconcerned as at a field day at the base.

  They waved and shouted cheerily to the Tunnellers as they passed and some of the wounded were glad to be helped to the shelter of the gallery.

  All was hustle therein now and it was not long before our part was finished and the complete communication made to the German trench. (J. C. Neil, New Zealand Tunnelling Company)33

  The failure of the advanced machine-gun position did not result in the failure of the attack along the Cambrai road. The Russian saps were quickly connected to the captured German front line, but because of the success of the attack were not needed. As at Vimy, the attackers at Arras made dramatic advances during the first few days of the battle and the whole of the German front line was quickly taken, despite the loss of the benefits of the Russian saps. The initial success, however, at Vimy and Arras did not last and the bitter frustration felt by the tunnellers was expressed by Neil:

  …to those on the spot the high hopes based on the tremendous success of the first day were quickly dashed to earth when it was realized that no fresh troops were available to relieve and carry on the work so splendidly commenced by the few divisions which made the attack.34

  Graham of 185 Company recalled that the offensive marked the end of mining on the Arras–Vimy front, but was not a decisive victory and brought instead a change of the role of the tunnellers to dugout construction:

  We were bitterly disappointed that all the huge preparations had come to this, as we were dead sick of trenches and the smell of dug-outs. True, we had got all the high ground and a vastly superior position for observation, but we had had high hopes that the offensive would bring in its wake some open fighting and perhaps the beginning of the end… We did not anticipate another phase of mining as the lines settled down at an almost prohibitive distance apart and I think we were all content to help the infantry and gunners by making them comfortable dug-out accommodation. As a unit we grew less important from a front line point of view, although our efficiency was always felt. At mining we were omnipotent and our own masters; at dugouts we were ‘nobody’s children’. Still, mining was a dirty game in more senses than one.35

  During the Vimy-Arras offensive German defensive tactics were undergoing a radical change, which had begun on the Somme following the appointment of von Hindenburg. The German policy of strong defence of the front line, and immediate counterattack to regain it if lost, had caused losses during the Somme almost as high as the British attackers, even though they should have held an advantage. General Ludendorff revised German defensive policy to replace the rigid defence of the front line at all costs with a more intelligent and flexible system. The increased Allied use of heavy artillery, and in particular as the attackers learned to coordinate artillery with the infantry, caused the deep dugouts to become a liability. During the Somme fighting from July to November 1916, the Germans took to abandoning their prepared defences, targeted by a devastating weight of artillery, and crawled into shell holes in no man’s land. Instructions issued in November and December emphasized weaker garrisons on the front line, which could change position to avoid bombardment. This was developed into a system of elastic, in-depth defence, in which lightly-held outposts would screen main defences and force the attackers to advance beyond the support of their field artillery and fall victim to counterattack troops.36

  The German command was becoming conscious of the danger that deep dugouts were beginning to present, but it was not until after the Allied offensives of April and May 1917 that the Germans prohibited them in the front line. For their attacks in April, May and June 1917 the French and British used very heavy bombardments to attempt to destroy underground defensive positions, and also coordinated the forward movement of a creeping artillery barrage with their infantry advance. Communication and accommodation tunnels repeatedly proved disastrous for the Germans when used defensively in 1917. The new German tactics were not thoroughly applied by army commanders and usually the Allies were able to achieve success on the first day of operations before inertia gripped the advance. On Vimy Ridge a programme of replacing deep dugouts with concrete bunkers had been delayed by winter weather and shortage of labour. The Canadian attackers were aware of the extensive German dugouts, especially on the rear slope protecting the reserve companies, which had to be reached before the barrage moved off to prevent the Germans counterattacking. Machine-gun barrage fire in particular was used to keep the Germans trapped in these dugouts and two communication tunnels until the infantry reached them. The Schwaben tunnel on Vimy Ridge ran about 1,350m from the German second trench almost to the second line. The Volker tunnel ran about 520m, linking the German first, second and third line trenches. The Canadians learnt of these several months before the attack and specially targeted them on 9 April 1917 by assaulting infantry as well as parties of tunnellers with demolition charges. The officer leading the tunnellers was shot rushing the Schwaben tunnel, but the forward part was quickly taken and the 2nd Canadian Mounted Rifles captured over 150 Germans, many of whom were still half-dressed. German communication along this tunnel broke down on the morning of the attack. The Germans were captured at the forward end at 6am, but no news reached the battalion headquarters near Bois de Bonval until an hour later when a wounded man who happened to pass the entrance informed them that the English (i.e. the Canadians) were in the battalion position and already held the third trench.37

  The best-known of the German communication tunnels were on the west bank of the Meuse, where they lacked cover or concrete shelters on the exposed slopes of Hill 265 and Hill 295 (Mort Homme). Towards the end of 1916 they began the excavation of the Gallwitz, Kronprinz and Runkel (or Bismarck) tunnels, which were opened at the beginning of May 1917. The 3rd Company of the 7th Westphalian Pioneers was assigned to work on the Runkel tunnel at three points on 4 April using large timber frames 120 x 180cm and in the hard rock only three or four were required for support each day. The tunnel ran just under 1km from Hill 265 south-east to Hill 295, allowing access and also accommodating reserves.38 No sooner were these tunnels in use than there occurred an event in the Champagne sector which demonstrated the danger of German over-reliance on tunnels and the failure to adopt the system of defence in depth. Mont Cornillet, a 207m high German-held hill, was a powerful artillery observation post which dominated the surrounding area and formed one of the bastions of the R1 Stellung. French attempts to capture Mont Cornillet during the Third Battle of Champagne failed on 17 April and 4 May 1917 and the French learned from a prisoner that the Germans were using a system of tunnels with three exits on the German side of the hill to protect a large number of reserves, who were able to carry out strong counterattacks.

  Bavarian miners of the 4th Field Pioneer Company at work on a communication tunnel, sketched by Pioneer Götz. From Lehmann, Bayerische Pioniere im Weltkriege.

  The 740m German Gallwitz communication tunnel on the Mort Homme. Based on Topographische Strukturfragmente auf der Höhe Mort Homme.

  T
he tunnels, apparently originally chalk quarries, consisted of three timbered galleries, about 3m wide and 2.3m high, between 120m and 143m in length and joined by a single transverse corridor of 154m. The tunnels ran into the side of the hill with up to 18m of cover, but a series of shafts provided light and ventilation. The maximum capacity was three battalions, one in each gallery, which included battalion headquarters, field kitchens, an aid post and ten days supply of water, food and ammunition. Resupply took place at night by carrying parties. The German front line was on the forward slope of the hill and on the rear slope between the summit and the tunnel entrances there was a line of shelters. The Germans were resolved to hold the hill and placed almost all their manpower within 500m of the front line. Thus two battalions of the 173rd Regiment (223 Division) were deployed in the tunnel or the shelters with the third battalion in reserve.39

  The scene of the disaster of Mont Cornillet showing the situation on the morning of 20 May 1917, with collapses caused by French direct hits and the spread of carbon monoxide. In the three galleries driven beneath the hill the remains of 321 German soldiers were found sixty years later. From Knies, Das Württembergische Pionier=Bataillon Nr. 13.

  Diagram of the Cornillet tunnel, showing the entrances on the German side and the ventilation shaft opening just behind the German front line, which was on the forward slope of the south side of the hill. The French front line was on the lower slopes. From Seesselberg, Der Stellungskrieg.

  When the French attacked the position on 17 April they managed to destroy the German surface positions facing them on the southern slope of the Cornillet, but those hidden on the northern slope were still intact. The French did not manage to take the summit and only got into the southern part of the circular trench around the hill. The tunnel entrances were damaged, but were repaired, and the Germans were able to launch effective counterattacks from the tunnel and the smaller Voss and Bremen shelters nearby. When the French attempted another attack on 4 May the two German battalions could again leave the tunnel and immediately counterattack. The French struggled to reach the summit of the Cornillet and failed to destroy the tunnel or its entrances. On 17 May the 173rd Regiment was relieved by the 476th (242 Division). By this time there was no surface trench system left, just connected shell holes which were occupied during the day by small groups while the rest remained in the tunnel. The French positions were only 25-100m away. The tunnel seemed to hold a mythical strength to the French, expressed in a report: