by Robin Prior
The 6 Division, which had failed before the Quadrilateral on the 15th, was assigned the task of assailing it again. At least the position of the German defensive work had now been established with some accuracy by the British artillery, and a bombardment of it began on the 17th.8
According to the 6 Division's War Diary, there is very little to say about this operation. The Quadrilateral fell after a brief fight, conducted like clockwork.9 One is always wary of such descriptions of operations on the Western Front and indeed according to one participant in the attack nothing remotely approaching precision occurred on that day. Lt-Col. Dillon was an officer who had been left in reserve on the 15th. On the morning of the 18th he was ordered forward to report to his brigadier about the forthcoming operation. His diary then takes up the story:
He [the Brigadier] then told me that the Colonel [of the attack battalion] had broken down, that the whole situation had altered, that I was in Command of the battn. and that we would attack at dawn, and that he would let me have orders as soon as possible. I had ... a conference of C.O's, all of which was absolute Greek to me not knowing where the Bn. was or what we were going to attack etc. So I sat quiet and said yes and no and tried to look dignified. At about 5 p.m. we were all asked if we quite understood and as we all said yes the show finished. I then got hold of an intelligent young staff officer and in ten minutes, with my map, had chalked the whole thing and made him promise to send me orders at the first possible opportunity and then we went to my headquarters .... On arrival I found the adjutant in a state almost of collapse, the doctor wounded, and apparently no one knew where the companies were. Fortunately, the pressure was relieved with the arrival of orders .... [But] by this time ... it was 8 p.m. and quite dark and the attack was to start at 5.50 am. Nobody seemed to know the way and everybody we met was dead or wounded .... It was getting very late now and the shelling was vy heavy, and I was really rather in trouble, as had I got straffed I had all the orders on me and nobody would have known what was happening.... At 5.30 I woke my Adjt..... The barrage was just starting.... Our own guns had it all their own way for 15 minutes by which time the lads were over....
What happened was two of our Companies went over the top in front and got hit badly, a third (and this is where I disobeyed orders) came in from [a flank]. Rushed the Hun strong pts bombed and bayoneted down the trench. The two unfortunate Companies got up and went for the boche and it seems incredible but in 10 minutes they had the whole boche line. In the meantime my 4th Coy. had gone straight on and made a strong pt 600 yards beyond and they caught the remaining Germans across the open. We captured 500 yards of front, advanced 1,000 yds. One mortar, one machine gun, 100 German prisoners and countless bombs. German dead God knows.10
By such imprecise and un-clockwork methods did the dreaded Quadrilateral fall to the British.
Far to the left of 6 Division in the area of III Corps two new divisions (50 and 23) which had replaced 15 and 47 were also making ground. One of the battalions involved in this forward movement has described exactly how it was done:
Orders were received from Bde for the Batt ... to dig at 7.30 pm a jumping off trench, midway between the front line ... and the Starfish Line, with a view to an attack on the latter, but the orders also stated that if, on completion of the jumping off trench, the covering party found that the Starfish line was lightly or not held at all, then the covering party was to enter it, and a garrison be put in.11
And this is exactly what happened. The Starfish Line was occupied and garrisoned without opposition. Later in the day Prue Trench was captured in a similar manner.12 So by these means 50 Division had gained its start line for the big attack.
Meanwhile, similar operations were taking place on the front of 1 Division where other sections of the Starfish Line were captured,13 and on the front of 23 Division where the remainder of Prue Trench was secured.14 Though little heralded at the time, later in the war such operations, especially those carried out by the Australian Corps before the Battle of Amiens in August 1918, would become known as ‘peaceful penetration’ operations. The expression as it happened was coined by Rawlinson in his notes on these events.15
II
The attack was to commence at 12.35 p.m. on 25 September. It was to be carried out by 10 divisions, five of which (23, 1, 55, 21, 5) had not been in action on 15 September. All corps would attack, but the main undertaking, the capture of Gueudecourt, Lesboeufs, and Morval – and consequently the German third line – was to be the task of XIV and XV Corps. III Corps would merely keep pace on the left to provide flank protection for XV Corps. There was on this occasion very little mention of the cavalry. One division of horsed soldiers was to be available to Fourth Army for unspecified purposes. There were to be no distant objectives. And because of this the artillery concentration against the defenders of the German front system was formidable, although Fourth Army headquarters seemed not to have calculated exactly what it would be.
In the two days before the battle, the British threw approximately 400,000 shells weighing about 7.5 million pounds into not more than 18,000 yards of enemy trench.16 The bombardment therefore achieved a concentration of about 400 pounds of shell for every yard of trench attacked. This was 40 per cent higher than the bombardment of 15 September. As for the defences which received this weight of shell, they were more rudimentary than either the first or second German trench systems on 15 September. In addition the incessant British attacks since the 15th and the rainy weather between the 16th and the 22nd had prevented the Germans from strengthening their line. On the right, the XIV Corps was to capture Morval and Lesboeufs. The 56 Division was merely to provide flank protection, the main operation being carried out by (from right to left) the 5, 6, and Guards Divisions. The advance was to be made behind a creeping barrage fired across the whole front in three short jumps. Between each jump there was to be a pause of an hour for consolidation and reorganisation.17 The total distance to be traversed was no more than 1,500 yards, giving the operation all the characteristics of a bite and hold attack.
The 56 Division accomplished its task with relative ease. Advancing under ‘a most efficient enfilade artillery barrage’ the battalions cleared Bouleux Wood and advanced some way beyond.18 Combles was now outflanked to the right. Casualties, at 164, were modest by Somme standards.19
The main attack by XIV Corps was carried out by 5, 6, and the Guards Divisions. The 5 and 6 Divisions had a relatively easy time, gaining all their objectives including the ruins of Morval according to timetable. The pauses in the advance were found most useful for regrouping under the creeping barrage. There is no doubt that success on this part of the front was due to the combination of an accurate preliminary bombardment, accurate wire-cutting, and the relentless destructive capacity of the creeping barrage. The War Diary of a Norfolk battalion which led the 5 Division's assault makes this clear:
Our artillery barrage was excellent, & we advanced with it, practically in it, & got to the objective at the very second it lifted...
The Germans had got out of their trenches & were largely in shell-holes, a few in front (who were killed as we went up), the majority were in rear of the trench. Germans were killed in the shelters & in the trench, & at first, some as they attempted to come forward from the shell-holes – the remainder surrendered (about 150).20
Thus were the lessons of ‘bite and hold’ revealed once more to the lower levels of com mand. They would be lessons that the higher echelons of the British army would learn, discard, and learn again until the end of the war.
On the left of XIV Corps' attack, the experience of the much-tried Guards Division would demonstrate that many things could go wrong even in a successful attack and that the experience of battalions in the same attack could be quite different. On the right of their front the Guards faced a series of sunken roads and wire concealed in fields of crops.21 In this sector the 2 Grenadiers found that the artillery had failed ‘to cut a single strand of wire’.22 They lost most of their of
ficers in the process of cutting their way through it and then charging the German trenches.23 Their War Diary has nothing but derision for the gunners:
the co-operation of the Artillery was remarkable for its absence and a great deal of ammunition was uselessly expended on ground where no Germans were, and places where Germans could be seen were left untouched.24
On the other hand, the battalion immediately to the left of the Grenadiers found all the wire cut and resistance weak. They took their objective with just 14 casualties.25 In the end the Guards Division secured all their objectives, including the ruins of Lesboeufs, but at a cost of 1,900 casualties – twice those of any other division of XIV Corps.26
On the left of XIV Corps were the three assault divisions of XV Corps, from right to left 21, 55, and the New Zealand Division. The 21 Division met with varying fortunes. Their 64 Brigade (on the left of the Guards) suffered severely from a German barrage which came down just two minutes after zero.27 They then found their objective (Gird Trench) strongly manned with machine-guns and riflemen who took a heavy toll on the advancing troops.28 Finally, when the survivors closed on Gird Trench they found that the artillery had left strong belts of wire in front of it untouched.29 The few men now remaining had no option but to take shelter in shell holes and await the protection of the night.30
The 110 Brigade of 21 Division fared little better. The wire had not been cut on some sections of its front, the trenches were strongly held, and German fire soon reduced the brigade to such small numbers that they were forced to withdraw. A night attack was ordered but after a discussion between the divisional commander (Jeudwine) and the XV corps commander (Horne) more cautious counsels prevailed and it was cancelled.31
This failure left the British in a considerable quandary. To the north-east of XIV Corps the Gird Trench remained in enemy hands and it provided an excellent starting point for a counter attack southwards into the flank of Cavan's divisions. Should such an attack succeed, the whole British position around Lesboeufs and Morval would be threatened. Luckily for the British help was at hand. During the night a tank had been hidden in the ruins of Flers. It now moved off to a section of Gird Trench where the Guards were still hanging on. At 7.30 a.m. on the 26th, accompanied by bombing detachments from the 7 Leicesters, the tank
started moving South Eastwards along the Gird Trench, first on one side of the trench and then on the other, firing its machine guns and being well supported by the bombers.... As the ‘tank’ moved down the trench the enemy surrendered freely.... By 8.30 a.m. the whole length of trench had been cleared.... In the capture of Gird trench, 8 officers and 362 other ranks of the enemy were made prisoners besides many killed. Our casualties were ridiculously small, about 5 in all. Over 1,000 yards of trench were captured in about 1 hour. What would have proved a difficult operation involving probably considerable losses was made easy entirely owing to the assistance rendered by the ‘tank’.32
With the capture of Gird Trench, the road to Gueudecourt lay open. Patrols from the 6 Leicesters found the southern end of the village unoccupied. They then ran into some resistance from snipers and machine-guns but by dusk most opposition had ended and the village fell into their hands.33 So the last of the three villages protecting the German third line (Gueudecourt, Lesboeufs, and Morval) was now secured.
The remainder of the battle need not detain us. The 55 and New Zealand Divisions, advancing behind an accurate and ‘extremely steady’ creeping barrage, captured the trench lines to the north-west of Gueudecourt with relative ease.34
As for III Corps, we have noted, most of its objectives were captured by ‘peaceful penetration’ methods before zero hour. On the 25th, the 1 and 50 Divisions gained ground to the north-west of Martinpuich. The 23 Division in the face of strong opposition on the Albert–Bapaume road broke off the attack.35
So ended the action of 25 September. In Somme terms it had been an outstanding success. The limited objective of the German third line had been captured from Martinpuich to Combles. As by default on 14 July and by design on this occasion, the efficacy of the bite and hold operation in conjunction with a formidable artillery accompaniment had been demonstrated. One tank had played a useful role around Gueudecourt, the cavalry no role at all. It was of course the guns that had effected the victory. The artillery had delivered sufficient of the infantry to their objectives to enable the small gains to be made. In terms of casualties as well, 5,000 in total for the 10 divisions, the cost was low by Western Front standards.
Did the fall of the third German line mean then that Rawlinson's forces now confronted open country where the cavalry might roam? It did not. The inexorable logic of Western Front fighting meant that while the British were making their modest gains in September, the Germans had not been idle. While Rawlinson's men crept towards Bapaume the German command ordered that a new defensive line be dug between Ligny-Thilloy and Le Transloy, a fifth line in front of Bapaume, and a sixth line further back.36 Now these lines too would have to be breached before open country was reached. And by late September summer had departed, autumn had commenced and winter with its promise of rain and mud was very close. Whether Haig would respond to these matters was another question.
Thiepval Ridge, 26–30 September
23 ‘The Tragic Hill of Thiepval’,1 26–30 September
I
When we left the Reserve Army in early September the II Corps and the Australians were struggling towards Mouquet Farm. There were few further large operations on Gough's front until 26 September. One, however, is worthy of note, not for the results achieved, but for the reactions of the command to its complete failure. On 3 September, two divisions of II Corps (39 and 49) attacked to the north of Thiepval. The 49 Division was given the task, after a short bombardment, of assaulting the Schwaben Redoubt frontally. There was to be no attempt at surprise. Not unexpectedly, as this strongpoint had withstood the five-day artillery assault before 1 July, it failed. There were 1,800 casualties. No ground was gained. Gough, Jacob (the Commander of II Corps), and the divisional commander were in no doubt as to the cause of failure. It was not due to the total lack of originality or subtlety in the plan – it was due to the lack of ‘martial qualities’ in the troops.
The chorus of invective by the command against the assault battalions is remarkable. As well as the lack of martial spirit, Gough contributed want of ‘discipline and motivation’, ‘ignorance on the part of the C.O.s’, and ‘poor spirit in the men’. Jacob added ‘want of direction’, ‘stage-fright’ (!!), and cowardice on the part of the brigadier. He pointed out that there were ‘no casualties amongst C.O.'s’ (a damning indictment indeed) and wanted entire battalions disbanded if they were unable to pull themselves together. The Divisional Commander was by comparison restrained, merely pointing to a lack of training and discipline (as though these matters were not part of his responsibility). These judgements, it may be noted, were passed on battalions that had lost between 30 and 50 per cent of their strength in failing to capture one of the strongest positions on the Western Front. Perhaps it was as well that most of the Reserve Army front was quiescent for the remainder of the month.2
II
The origins and timing of the major attack by Gough's force on 26 September are shrouded in mystery. Earlier he had stated that the whole purpose of the crawl towards Mouquet Farm was to position his force to attack Thiepval from the rear. At some point in September this plan was scrapped (but operations to capture the farm were not) and a frontal assault on Thiepval substituted. We may surmise that the difficulty in capturing Mouquet Farm was the reason for the change, although none of the Reserve Army Papers, or Haig's diary, or any other source is specific about this. The date of the operation, as we have noted, was changed from the 25th (when it would have coincided with efforts by the Fourth Army) to the 26th in order to conceal the tanks in the early morning mists of the Thiepval Valley. But as was also noted, for reasons that are opaque and on which the sources are also silent, the start time was mov
ed from morning to afternoon – 12.35 p.m., exactly the same time as Fourth Army operations, but a day later.
III
For the attack on Thiepval and its ridge, Gough had assembled four assault divisions on a 6,000-yard front. On the right, the 2 and 1 Canadian Divisions would advance to the north and east of Courcelette and attempt to capture Regina Trench which lay just beyond the ridge line. Further left the 11 Division, in its first action since the Suvla Bay landing in August 1915, would capture Mouquet Farm and Zollern and Stuff Redoubts, and then advance beyond to Stuff Trench (as the continuation of Regina Trench was known in this area). The most difficult task would fall to the 18 Division, which had played so conspicuous a part in early operations of the Somme campaign. Since then it had been ‘resting’ in Flanders. Now it had been brought south and given the intimidating tasks of capturing Thiepval and the Schwaben Redoubt, strongpoints which had taken a fearful toll on the 36 and 32 Divisions on 1 July and which – apart from the ill-fated attempt of the 49 Division on 3 September – had not been attacked since.
The whole operation presented enormous difficulties to the Reserve Army. The final objective (Regina and Stuff Trenches) lay on a reverse slope, out of direct observation from the British line. But before that point was reached, in the II Corps sector, lay the five formidable objectives just mentioned: Mouquet Farm, Zollern Redoubt, Stuff Redoubt, Thiepval, and Schwaben Redoubt. These fortified areas represented slightly different problems for the attack. On the surface, Mouquet Farm was now little more than rubble. However, underneath the debris lay three groups of cellars connected by a tunnel which ran in part behind the British front line. The entrances to the cellars were concealed in the ruins and not at all obvious to the attacking infantrymen. So even if an accurate creeping barrage could carry the assault battalions to the farm, it must lift while the soldiers sought out the entrances. This was the moment of danger. If the enemy machine-gunners and riflemen were to emerge before these entrances were located the attackers could find themselves assaulted from all sides.