The Somme

  There was a clear alternative to this proceeding, that of tactical withdrawal which husbanded reserves and shortened the line. Some German commanders recognised this wisdom. On 2 July, a subordinate commander south of the Somme sanctioned a local withdrawal just for the purpose of straightening the line. This brought down on him the wrath of Falkenhayn. The local commander was sacked and a stern warning issued by the commander-in-chief:

  The first principle in positional warfare must be not to give up an inch of territory and when an inch is lost to launch a counter-attack until the last man.43

  It is important to take note of this directive. We have witnessed so many examples of ineptitude on the part of the British command that it is important to remember that they could be equalled or exceeded by the Germans whenever the opportunity presented.

  VII

  It seems certain that the British command derived great comfort from the reasonably rapid gain of ground made in the period from 2 to 13 July. But in doing this they were ignoring two factors. The first was the fact that in proceeding with unrelenting, poorly planned and uncoordinated attacks, they were incurring very heavy casualties. The second was that such a method was only having success because of the shaken German defence. The same methods employed against resolute Germans manning intact defences around Ovillers and La Boisselle had brought disaster. In other words, the circumstances which made progress possible in this period would have to be replicated if such illprepared and costly methods were to lead to progress in the future.

  The Plans, 14 July

  13 ‘Cavalry Sharpening Their Swords’, 14 July

  I

  Preparations for a major new attack by Fourth Army had been under way from the second day of the campaign. On that day Haig had pointed out that if a modest advance in the south could be accomplished, it would open the way for an attack on the German second line from Bazentin-le-Grand to Longueval. If this attack coincided with an assault further south by the French, ‘very considerable results could be achieved’.1 The prerequisites for the operation were that Mametz Wood and Trones Wood should be in British hands to secure the flanks of the new endeavour.2

  As noted earlier, Rawlinson had briefly considered renewing the attack in the north. But by 4 July he had accepted that this was impractical and was prepared to go along with Haig's scheme. On the 8th he had issued his first operation order for an assault by four divisions on ‘the enemy's second line between Longueval and Bazentin-le-Petit Wood’.3

  On a first reading, Haig seemed quite satisfied with this plan. Nevertheless in one important respect the two commanders were working on different assumptions. Haig thought that before attacking the German second line, Rawlinson would undertake preliminary action which would advance his front to within a few hundred yards of the German second position.4 Rawlinson considered this impractical along the whole line of his front. The most he could do was to get forward on the flanks. But this would mean that in the centre there would remain a no man's land 1,500 yards wide and entirely devoid of cover – just the kind of scenario which had led to the destruction of III Corps on 1 July. To resolve this problem, Rawlinson was proposing to undertake an operation commencing at night in which the two divisions from XV Corps (3 and 9) to be employed in this area would form up in no man's land in the dark and close on the German front line. As soon as dawn broke these divisions, along with the two from XIII Corps on the right, would attack.5

  Haig first heard of Rawlinson's plan on the 10th.6 Initially he made no comment on it.7 But by the next day he had formed a strongly negative opinion. He now refused to sanction the night attack, informing Rawlinson:

  Our troops are not highly trained and disciplined, nor are many of the staff experienced in such work, and to move two divisions in the dark over such a distance, form them up, and deliver an attack in good order and in the right direction at dawn, as proposed, would hardly be considered possible even in a peace manoeuvre.8

  Instead, Haig proposed an attack by XV Corps on the left of Rawlinson's proposed operational front, from around Mametz Wood, which would then turn eastwards to roll up the German line around Longueval. If this attack was successful, XIII Corps would then attack Longueval next day. Haig's proposal had little to recommend it. XV Corps' attack was to take place on a narrow front and involved a highly complicated turning manoeuvre. The eastward attack towards Longueval also risked being taken in flank by a German counter-attack from High Wood. Finally, as Congreve (XIII Corps commander) pointed out, Haig was embarking on a two-stage attack which invited defeat in detail.9

  Presumably most of these arguments were levelled against Haig's plan at a meeting between Rawlinson and his corps commanders (Horne and Congreve) at 2 p.m. on the 11th. Certainly Horne noted that ‘some modification of plans [was] suggested’.10 In any case, as a result of that meeting Rawlinson told Kiggell that the corps commanders were both strongly in favour of the original (Rawlinson) plan. He added that Horne was ‘much averse’ to attacking without the support of XIII Corps. Kiggell relayed these views to Haig and then spoke to Horne who reiterated his objections to Haig's plan. Haig refused to budge. Rawlinson was instructed to prepare the attack along the lines stipulated by the commander-in-chief.11

  Rawlinson passed on this message to his corps commanders. But after a further talk with General Montgomery, his Chief of Staff and Horne, he decided to press Haig once more to countenance a night attack.12

  Haig called a meeting of his own staff on the 12th to discuss what he called Rawlinson's ‘amended plan’.13 (It was in fact the same plan.) Birch, Haig's artillery adviser, who had been in touch with Congreve, assured the Commander-in-Chief that the British could dominate the German guns, and after some discussion Haig agreed that Rawlinson's plan should be tried.14 He insisted, however, that Rawlinson re-examine his counter-battery programme, that a contingency plan in case of failure be worked out and that, if available, the use of motorised machine-guns be considered. But on the main issue there would be no further discussion: the night attack would take place.

  As long as it remained undetected, the attack by night would certainly solve one of the problems that had doomed the attempted advance on 1 July – the crossing of no man's land in full view of the enemy machine-gunners and artillerymen. Preventing detection was therefore a crucial matter. Although many historians appear unaware of the fact, night operations had been attempted already and had not proved successful. They had been detected by alert machine-gunners and had also run into uncut wire. We have already noted the failure of such a night attack by 12 Division on 3 July. The key factors for a successful night attack were stealth and the ability of the artillery to eliminate or suppress sufficient of the German machine-gunners and hostile guns to give the troops a reasonable chance.

  With the decision to undertake a night attack in place, other aspects of planning for the new operation could commence. In one respect the process seemed to indicate that the command had absorbed an important lesson from 1 July. If we compare the bombardment on 1 July with that of 14 July we find that – for the length of trench attacked – the new operation would deliver an intensity of shell five times that of the disastrous first day. Yet there was a singular aspect to the new artillery plan: on this occasion all the shell would be concentrated against the German front system, instead of, as previously in any Rawlinson–Haig operation, spreading the fire across all the German defensive systems within range.15

  One other aspect of the planning for the 14 July operation depressingly paralleled that for 1 July. As the day of battle drew closer the objectives to be aimed for by the Fourth Army were further and further extended. But there was a bizarre difference between the two occasions. In the second instance, it was Rawlinson who kept extending the objectives while Haig tried (unsuccessfully) to rein them in.

  Rawlinson's plan began modestly enough. Initially, there was little mention of any objective beyond the line running through the Bazentins to Delville Wood. Some elements of the 2 Indian Cavalry
Division were to be on hand to seize High Wood, but that was all.16 Three days later, however, at a conference with the cavalry commanders, we find Rawlinson – while entering the usual caveats about the cavalry being held back until infantry operations had succeeded – considerably enlarging the cavalry aspect of the plan. The 2 Indian Cavalry Division was not only to seize High Wood but push out patrols to Flers and Le Sars, some two and three miles beyond. Then the 1 Cavalry Division would be sent through to hold Morval and Les Boeufs until relieved by the infantry of XV Corps. Meanwhile, the 3 Cavalry Division would operate towards Martinpuich and Le Sars.17

  What Rawlinson was trying to accomplish was not merely the capture of the German second line. He was aiming to seize a third line between Le Sars and Morval, which aerial reconnaissance had only just revealed to him.18 This third line was continuous but the aerial photographs had revealed some gaps in the wire, and anyway Rawlinson did not consider that the Germans had enough troops to hold it in strength.19 Whether there might be the odd machine gunner or reserve infantry division in the German rear areas over which the cavalry were to roam was not discussed. Nevertheless, the intent of Rawlinson's orders was clear – the objectives of the operation had been extended from the capture of the German second position (a depth of advance of a few hundred yards on a 6,000-yard front) to an advance of roughly four miles on a front of 10,000 yards.

  Surprisingly, given his hankering after distant objectives before 1 July, this plan did not find favour with Sir Douglas Haig. At a conference with Rawlinson on the 12th he stated that he

  [did] not consider that large forces of cavalry should be pushed throuh at once as there would not be sufficient room for their action at first and they might be thrown back and cause confusion. A few squadrons could probably, however, be used in the circumstances without the danger alluded to and probably with good effect.20

  And just in case Rawlinson had not got the point, later in the day Haig had Kiggell write to Archie Montgomery to repeat his doubts about the cavalry and to make the additional point that ‘there would be too much danger of their [the cavalry] coming unexpectedly under heavy fire from the enemy's rallying points’.21 Kiggell added for good measure that Haig wanted the 3 Cavalry Division withdrawn into reserve.22

  Nothing followed from this. Far from changing his plan, Rawlinson on the following day (the 12th) issued a new set of orders for the cavalry which gave them precisely the same objectives as he had stipulated before Haig's intervention.23

  Perhaps as a result of these orders Rawlinson received a visit from Haig on the 13th. Haig noted in his diary:

  I visited Sir H Rawlinson after lunch. I spoke about the use of cavalry. The divisions were not to go forward until we had got through the Enemy's fortifications, except a few squadrons to take “High Wood” ... I also stated his objectives as follows:–

  (1) Occupy position Longueval–Bazentin-le-Petit, and consolidate it.

  (2) Take High Wood, and establish right flank on Ginchy and Guillemont.

  (3) At the same time (if possible, as there are ample troops) extend left and take Pozieres ridge and village of Martinpuich.24

  Only in the last aspect did Haig's instructions induce Rawlinson to modify his plan. He now incorporated III Corps into his operation and instructed them to take Pozières, Courcelette, Martinpuich, and Le Sars as soon as XV Corps had secured the Bazentins. As for the rest, he not only ignored Haig's instructions a second time but incorporated distant objectives into his infantry plan and gave the cavalry yet wider horizons beyond even the third German line. So, if operations succeeded, the infantry would occupy a line through Mouquet Farm to Guillemont (as indicated on the map) while the cavalry would proceed about seven miles further east to the dotted black line.25 This would entail an infantry advance of about 2–5 miles on a 15,000-yard front, and a cavalry advance of 7=8 miles on a 20,000-yard front. The German second line would have fallen into British hands from Mouquet Farm to Guillemont and sections of it north of Thiepval would have been outflanked; and the third German line would have been captured from Pys to Leuze Wood.

  Quite apart from Rawlinson's apparent disregard for his commander's strictures, this was an extraordinary exchange between the two men. In substance it was almost a mirror image of what had passed between them in the planning stage before 1 July. Then Rawlinson's preference had been for a modest operation, while Haig had opted for more distant objectives. Now the reverse was largely the case.

  How do we explain this turnabout? Apparently Haig was no longer as optimistic about imminent German collapse as he had been before 1 July and in the aftermath of that day's endeavours. Whereas GHQ had written on 6 July about the enemy's waning morale leading to ‘serious demoralisation [which] may set in at any time, possibly tomorrow’, Haig only two days later was taking a more sober view. So he wrote to Robertson, the CIGS, that ‘much remains to be done’ before the Germans were defeated.26 And three days later he warned Rawlinson that the enemy would rally at their strongpoints behind the front and offer much resistance.

  Rawlinson to all appearances was oblivious to these changes in Haig's appreciation. His plans for 14 July seem to be based on the view that once the German front had been breached by the devastating bombardment, no serious resistance was to be anticipated, only a thoroughly demoralised enemy. Hence, he could plan to devote all of his artillery bombardment to the German forward position and yet still anticipate a sustained advance over several miles of unbombarded ground.

  This failure of Rawlinson to take note of Haig's increasingly realistic assessments set the scene for a paradoxical operation. In it a strikingly successful first phase resulted from a considerable underestimation of enemy strength. But it led not to further triumphs, only to a restoration of futile operations and deadlock.

  This point needs to be stressed, on account of the many occasions when commentators have hailed the plan for 14 July as being a great advance in tactical sophistication over that for 1 July. It must be emphasised that a night attack in itself did not guarantee success (as the failure of 3 July had made clear). And the aspects of the 14 July scheme envisaging wide-ranging exploitation by both cavalry and infantry were in themselves no more soundly based than similar plans before 1 July. Only one facet of Rawlinson's 14 July plan shows any improvement on the first day, namely the execution of a bombardment on the front German system of far greater intensity than anything delivered on 1 July. Here Rawlinson may have revealed a measure of insight, a learning curve perhaps. But two things tell firmly against any such notion. First is the fact that fundamental to the proceedings of 14 July was the delusion that the bombardment would not only overwhelm the German front defences but lead to a decisive collapse in German morale, opening the way for a wide-ranging cavalry sweep. The second is that Rawlinson, in his subsequent operations on the Somme, never again employed a bombardment of the intensity of 14 July against the enemy's foremost defences. In short, Rawlinson's delivery of a devastating bombardment on 14 July was based on delusion, not judgement – the operation was not planned as an example of bite and hold, but as the prelude to a Haig-like cavalry advance.

  II

  The German second line had been under attack by the Fourth Army artillery since 11 July.27 This line was in general much weaker than those sections of the first line captured on 1 July – the dug-outs were not so frequent or formidable and the trenches were not nearly as extensive. The line was well-wired however, with two belts of wire each five yards wide.28

  In the early hours of 14 July the troops from the four assaulting divisions (from left to right 21 and 7 Divisions from XV Corps and 3 and 9 of XIII Corps) moved into position. For the men of XV Corps this presented no difficulty. The distance between the front lines in their area was small and the remains of Mametz Wood still offered some cover for the assembling troops. The situation facing XIII Corps was quite different and provided Rawlinson's justification for a night attack. Here the British line was 1,000 to 1,500 yards from the German front posi
tion and in between the bare ground sloped gently up to the Bazentin Ridge. So in this area very careful preparations had to be made to get the troops within close proximity to the German line by zero hour (3.25 a.m.).

  In the 9 Division this was done by sending forward small parties equipped with Lewis guns to form a screen at the crest of the ridge. Then even smaller parties crept forward to place marker tapes within 500 yards of the German front line. When these were in place the bulk of the assault battalions moved up by companies in single file until they reached the tapes. From there they slowly crawled forward until the enemy line was just 50 yards away.29 This last phase could be a slow business. The 3 Division, which had moved forward in similar manner to the 9, started creeping forward at 2 a.m., gaining just 20 yards every quarter of an hour, so that by 3.15 a.m. they were 120 yards from the German trenches.30 In what was a remarkable piece of staff work, these divisions were in place at zero without the Germans being aware that a major attack was imminent.

  Despite the crushing intensity of the bombardment – by far the heaviest weight of shell fired by the British so far for the length of trench attacked – its effect varied along the front. On the left, 21 Division found the enemy front line obliterated and the wire well cut. Still, some units suffered heavy casualties from machine-guns in the rear of the German front system which had been left untouched by the artillery.31