The Somme

  The Guards had gained 2,000 yards on a 1,500-yard front, quite a feat in the circumstances. But this did not alter the fact that Haig's conception even now lay in ruins. The distant villages which were meant to be in British hands before the cavalry went forward were still 2,000 yards away. Along the whole of XIV Corps' front, only the Guards had captured even their first objective. Far from the tanks having ‘flabbergasted’ the Germans, many of the enemy in this sector would have been unaware of their existence. Overall, the optimistic predictions of collapsing German morale had proved unreliable as ever.

  II

  The failure of the right wing of the British attack ensured that there would be no large cavalry sweep on the Somme. The task of the left flanking divisions was subsidiary: to guard the right flank as it moved forward in conjunction with the horsed soldiers. Ironically, they were completely successful.

  Attacking were 2 Canadian Division from the Reserve Army and on its right 15 (Scottish) Division of Fourth Army.21 Their task looked anything but simple. Standing in the path of the Canadians were the ruins of the large village of Courcelette. Similarly the ruins of Martinpuich confronted the 15th. As it happened, neither proved much of an obstacle. The Canadians had two advantages. First, they had detailed their seven tanks for mopping-up purposes only. So they would follow the infantry and no tank lanes were left in the barrage.22 Second, the Germans conducted an ill-thought-out attack on some Canadian positions just before zero.23 Not only was this operation quickly dealt with by Canadian front line units, but it left the forward German trenches packed with infantry, most of whom were caught by the preliminary bombardment. So when the Canadians followed what they described as ‘our terrible Artillery barrage’, they found the remaining Germans unusually willing to surrender.24 Follow-up troops, pushing on to Courcelette, encountered troops described as of ‘very poor quality’,25 who in any case were new arrivals on the Somme battlefield.26 By 7.30 p.m. the village was in Canadian hands and patrols had even pushed beyond their designated final objectives.27

  On the right of the Canadians the 15 Division also captured all of its objectives, including the ruins of Martinpuich. There was no assistance from the tanks, all of which arrived too late to take part in the battle.28 In this area the enemy seemed to be taken by surprise, many of the defenders being caught in their dug-outs.29 In addition, the morale of German troops around Martinpuich actually seemed in line with GHQ predictions. The defenders of the village, although having ample time to emerge from the cellars and strongpoints and employ their weapons, surrendered.30 No fewer than 700 Germans including a battalion headquarters, were captured.31 The left flank of the great advance, although already aborted by events on the right, was securely in British hands.

  The 15 Division was part of III Corps which had two other divisions attacking that day, 50 to the left of High Wood and 47 from the wood itself. Their task was to capture the third German line to the left of Flers and later to form part of the northern advance of the infantry. Neither division got so far, although the 50 made reasonable progress. In this area the bombardment had proved particularly accurate and the attackers found few Germans remaining to defend the first line.32 Further progress was then possible on the left because of the success of the neighbouring 15 Division. At the end of the day this sector of the divisional front lay just short of the German third position.33

  On the right it was quite different. There, enfilading machine guns from High Wood halted the advance before it reached the first German line. Some further progress was made after the wood had fallen, but by this time casualties were high, troops had lost direction because of the dust kicked up by the barrage, and the two tanks which had accompanied them and performed some useful tasks in enfilading trenches were out of action. The men were eventually forced to dig in just short of the German second position, well short of their final objective.34

  It may be imagined that the 47 Division had more immediate concerns than the grandiose objectives laid out for them in their operations orders. Their first objective was High Wood. This had been one of the objectives on 15 July, and of a series of subsequent operations. None had succeeded. On 15 September the Germans were still in occupation of the Switch Line which ran through the northern sector of the wood. Now the 47, a Territorial unit from south London which had fought at Aubers Ridge, Festubert, and Loos in 1915, had been given the task. It had one advantage. For the first time since 14 July, the wood was to be attacked as part of a broad-front operation, so that the Germans would not be able to concentrate all their artillery fire against it. There were, however, some significant negative factors. The first was General Pulteney, the commander of III Corps, who had decided to use the four tanks allotted to 47 Division inside the tangled wilderness of the wood instead of in outflanking movements around it. Furthermore, none of the tank officers had visited 47 Divisional staff to discuss the employment of artillery fire or where they proposed to enter the wood.35 Then there was the fact that the British and German front line trenches in the wood were so close together that bombardment was difficult. For the period of the preliminary bombardment this was easily solved. The British troops were withdrawn and the wood subjected to heavy fire. But it was then decided that, just prior to zero, the men should reoccupy their original front line.36 This meant that the problem of proximity which had been over-come for the preliminary bombardment now applied to the creeping barrage. And here the solution was quite lamentable. The command decided to commence the barrage (which anyway would have lanes left in it for the tanks) 150 yards beyond the German front line to avoid hitting their own troops.37 This decision was neither sensible nor necessary. A new front line could have been dug in rear of the existing British line and the creeping barrage commenced in front of that. Finally, the map co-ordinates given to the artillery were wrong, so the creeping barrage actually commenced 200 to 250 yards behind the German front line.38

  The result was predictable. Of the four tanks assigned to the division, two were ditched in old trenches behind the British front line in the wood. The third found the going so rough that it turned right out of the wood and shot up a trench full of troops, who happened to be British.39 Nevertheless, the tank commander was awarded the Military Cross.40 In the restrained words of this unit's War Diary, ‘after a heated argument ... the tank ceased fire and took no further part in the day's proceedings’.41 A fourth tank by good luck and skilful driving actually arrived at the German front line and shot up a small group of enemy defenders, then proceeded through to the next rearward trench and shot that up as well. At this point its engine failed and the crew abandoned it. Making their way back to the British lines they accepted the surrender of a group of Germans on the way.42 Despite this small success, Pulteney's decision to use the tanks inside the wood certainly prevented them from fulfilling even the limited potential they possessed in 1916.

  As for the infantry, they were having to battle their way towards the distant creeping barrage which had 100-yard gaps in it. The men were met with such heavy fire from unbombarded German machine-gun positions that few survived to contribute to their respective war diaries. The best account of what happened comes from the unlikely source of a French interpreter assigned to the division.

  The 15th [Londons] were to rush a [German strongpoint] with the help of the Tanks. The 8th [Londons] were to debouch from the Wood and make for some redoubts on the downward slope beyond. The 6th [Londons] were to pass through the 15th and 8th, and make for a certain point in the German 3rd line .... The Artillery was put on the back positions principally and the Tanks and the 15th were left to deal with the first line. The Tanks are certainly wonderful things, and can go over all sorts of country except, perhaps, ground where tree trunks lie criss cross in all directions. Therefore when the 15th went over the top, they had a bad time with machine guns, though they took the [strongpoint] and lay down in front of the Old German Front Line. The 8th, coming up to pass through the 15th and carry on, instead of carrying on, had to support
the 15th themselves .... Old Goodes and his merry men were then called in and did a record shoot with their Trench Mortars. The Huns stuck it until 750 bombs had been hurled at them and then held up a white flag. This ended the business in the Wood.43

  It was a notable event. High Wood, which had resisted so many British attacks, had fallen. The combination of the broad-front attack, the heavy preliminary bombardment, and the final hurricane of trench mortar shells had overcome the ineptitude of the original plan. But it was only the first objective of the day. The German second and third lines were still to be assaulted. Not surprisingly, further substantial movement proved beyond the division. Most of its reserve battalions had been drawn into the fighting inside the wood. It was not until 6 p.m. that a motley composite force could be organised for a further advance. Some ground was gained but at heavy cost, one of the assaulting units being reduced to just 62 men of the 567 who attacked.44 The survivors could do nothing more than dig in just short of the second German line.

  III

  It fell to the divisions of XV Corps to make the most substantial advances of the day and it was in this sector that the tanks made an effective contribution. The corps had a series of formidable objectives. First they were to capture the Switch Line, then later the German second line and the fortified village of Flers, and then move on to the German third line around Gueudecourt.

  Least successful was 14 Division on the right flank of the corps. This unit was heavily enfiladed by the machine-gunners who had wrought such devastation on the early attack of the Guards Division.45 The unit also had trouble locating the Switch Line, which had in fact been obliterated by the artillery.46 Nor could the division establish contact with the fast-moving 41 Division on its left. So as the reserve battalions attempted to move forward they found both their flanks unsupported. They dug in and held their positions, but they were short of the German second line and had no chance of advancing beyond.47

  The two divisions of the corps which achieved the major advances of the day were the New Zealand and the 41, neither of which had seen action on the Somme. Along some sections of their area of attack the preliminary bombardment had been extremely effective. In the 41 Division area very few defenders in the front positions survived it and so despite the lanes left for the tanks the Switch Line was taken with few casualties. The attackers were also fortunate to have three tanks which were able to operate in advance of them and subdue or put to flight those who had survived the bombardment.48

  In the New Zealand sector (where none of their four tanks appeared before zero) numerous German machine-gunners were still in their dug-outs when the infantry, moving fast behind the barrage, arrived.49 So by 6.50 a.m., just 30 minutes after zero, the Switch Line had been captured. The only severe casualties had been suffered by the left of the New Zealand Division by flanking fire from High Wood.50

  The second objective consisted of a line of trenches just south of Flers. It fell without incident. The troops ran into desultory German artillery fire, but when they reached the new line they found it almost deserted. By 7.20 a.m. detachments from the two divisions were in sight of Flers.51

  Their third objective lay just beyond the ruins of Flers, in some trenches which made up part of the German second line. This was the moment when the tanks played their most significant part in the operations. On the front of 41 Division four tanks were still operational. These machines followed the creeping barrage (which it might be noted now had no gaps in it, demonstrating how unnecessary the tank lanes had been) into Flers village with the infantry close behind. Three machines fanned out to the right and left of the rubble and subdued any machine-guns that remained. The fourth lumbered up what remained of the main road and dealt with a machine-gun that was holding up the advance. All of this proved too much for the defenders. They fled back towards Gueudecourt and by 10 a.m. Flers and some of the German second line had been captured.52

  Tanks were also assisting the New Zealanders. On the left of their advance,

  The 3rd bn. had been hung up by intact wire and heavy machine gun fire. At 10.30 a.m.3 tanks arrived – one tank went to the extreme left of the sector ... while another dealt with the wire and machine guns .... This enabled the 3rd bn. to take the Flers Line.53

  On the right the panic caused by the tanks in Flers made the capture of the left of the village a relatively easy matter.

  A portion of the German second line now lay in British hands, and that part of it running northwards towards Le Sars had been outflanked. Gueudecourt lay not far distant. It seemed that the centre of the Fourth Army was on the brink of a major victory. But this was not the case. The 41 Division's casualties were by this time considerable, and the units much intermixed. Such was the confusion that at one point an entire brigade actually fell back from Flers, leaving it open. The situation was restored by some reserve battalions before the enemy could take advantage of it, but these units could make no further advance. All tanks were now out of action and the German defence had recovered to the extent that they delivered several heavy counter-attacks from Gueudecourt. Just after 4.30 p.m., the order was sent out to consolidate what had been gained.54 Nor could the New Zealanders progress. They attempted to advance north-west of Flers, but reports were coming in of German counter-attacks on High Wood and on Flers itself. It was thought prudent to dig in.55 The great attack was over.

  22 25 September

  I

  On the evening of 15 September Rawlinson received reports from his corps commanders and issued his orders for the new day. He demanded that nothing less than a full-scale attack be undertaken immediately to enable the ‘Cavalry Corps to push through to its objectives and complete the enemy defeat’.1 The unreality of this order hardly needs emphasising. The Cavalry Corps had precisely the same chance of effecting a victory on the 16th as it had had the day before – that is, no chance at all. As for the infantry, the 6, Guards, and 47 Divisions were in no state to resume full-scale attacks and the remainder were in various degrees of disorder. What followed then from Rawlinson's instruction was not a concerted attack across the whole front but the series of small-scale, sporadic actions that had become a familiar pattern after a major attack on the Somme. On this occasion they produced exactly the same results as those following 14 July – no advances for quite heavy casualties.

  By the morning of the 17th, the deteriorating weather and the exhaustion of British divisions holding the line made it obvious even to Rawlinson that there could be no large operations for some days. In any case, by this time Haig had a larger vision. After speaking to Foch and Gough he determined that a major offensive would be carried out from Thiepval to Maurepas. The Reserve Army and the French would be engaged as well as Rawlinson's forces.2 In all, 10 divisions would be involved, with the aim of capturing Thiepval and the German third line around Le Transloy.

  In the event, the same chronic inability to co-ordinate the action of the different armies foiled Haig's intent. The first destabilising factor was the weather. Rain persisted on the 18th, making it impossible to bring forward ammunition for the guns and supplies of all kinds for the troops. By the 20th the weather had become so bad that operations were postponed until the 23rd.3 Further bad weather imposed another delay until the 25th.4

  25 September

  Then there were the French. On the 18th Foch indicated that he would not be ready until the 21st because of ammunition shortages. He also insisted that any attack should take place in the afternoon so that artillery fire could be observed. (This was to have consequences for the operation as a whole which will be discussed later.) As the 21st approached Foch then made it clear that he could not meet that deadline. So heavy had been his expenditure in ammunition in beating off a German counter-attack on the 20th that he would need 48 hours to replace it.5 This pushed the operation forward to the 23rd and then the bad weather delayed it again until the 25th. Nevertheless, the Fourth Army and the French were at least to go forward on the same day.

  The same could not be
said of the Reserve Army. Gough had decided to use a number of tanks to assist him in capturing Thiepval. This presented a problem. Because of French insistence that the attack by their forces and the Fourth Army take place in the afternoon, Haig reasoned that if Gough attacked at the same time his tanks would have to be brought forward in daylight, exposing them to the German artillery. He therefore decided that they must be used in the concealing ‘mists of early morning’6 and this meant that Gough's attack would go in on the 26th, the day after Rawlinson and the French. This made no sense. The advantages of a really broad front attack were being sacrificed for a weapon which had hardly proved itself of startling utility on the 15th. Nevertheless, Haig had made his decision. Then for reasons that are totally obscure, Gough made a different one. When the day of his attack arrived the tanks were not used in the early morning mists but in the early afternoon. Gough in fact attacked on the 26th at exactly the same time that the French and Rawlinson attacked on the 25th.

  Rawlinson's plan therefore (Gough's will be discussed separately in the next chapter) was for 10 divisions to assault the German third line from Martin-puich on the left to Combles on the right. The French I and XXXII Corps would co-operate on the right. The decision to attack in the afternoon, forced on Rawlinson by the French, meant that no tanks could be used for the same reason expounded earlier by Haig regarding Gough's forces – tanks would be too vulnerable to artillery fire in broad daylight. So instead of leading the attack, several would follow the assault around dusk and deal with any German resistance lingering in the ruins of the villages in their third line.7 What this meant was that the creeping barrage could be reintroduced as a method of infantry protection all along the front, for there would be no lanes required for the tanks. This decision, which was to have a crucial impact on the outcome of the battle, therefore came about not because of a reflective investigation of operations on the 15th but by accident. While the plan for the 25th (and, as it happened, the 26th) was being decided the start line for that operation had yet to be captured. This meant that the Quadrilateral and the unattained objectives between High Wood and Martinpuich, including Prue Trench and the Starfish Line, had first to be taken.