by Hew Strachan
The French army’s conduct of the campaign was not necessarily more distinguished. But it was not burdened with the continuous pressure for a decisive victory in short order. After the Aisne, Joffre could view protracted operations—and therefore more cautious tactics—with equanimity. The preservation of the army in the field was quite sufficient as an immediate objective. Indeed, in the light of this the strategy which Joffre adopted seems almost rash. From 15 September Joffre had planned two levels of operation against the exposed German flank. In the first place, he wanted Maunoury’s 6th army to advance up the right bank of the Oise. Secondly, he envisaged a much larger turning movement between the Oise and the Somme. The caution of Maunoury’s advance, his decision to go up the left bank of the Oise, and therefore his failure fully to threaten Kluck’s right flank meant that he was slow to execute Joffre’s wishes. When, on 17 September, he threw XIII corps onto the right bank of the Oise it hit Kluck’s right wing proper, and the whole French 6th army was checked. But simultaneously the blocking of Bülow’s attack increased Joffre’s confidence in once again stripping his left to feed his right. The 2nd army at Nancy was dissolved, and a new 2nd army, still commanded by Castelnau, was formed south of Amiens. Four territorial divisions, covering the French line of communications in the Amiens area, extended Castelnau’s left. Thus, shorn of cavalry, France’s marching wing was entrusted to over-age and under-trained troops, commanded by a recalled 74-year-old. The hope that they could envelop the Germans by way of Arras and Bapaume was doomed to disappointment.158
The opportunity presented in the east by this further French move to the west was not neglected by the Germans: the planned attack on the Meuse forts was met in somewhat dilatory fashion by Sarrail’s 3rd army, and by 25 September the Germans had formed a salient south of Verdun round St Mihiel. Furthermore, with the failure of Bülow’s attack, Falkenhayn felt able to resume his own conception, and on 18 September ordered the movement of the 6th army from Lorraine to the area around St Quentin. The moves of each army mirrored those of the other.
The task which Crown Prince Rupprecht was set was poorly defined. His army, Falkenhayn told him, was to seek a decision on the northern wing and to secure the right flank. Falkenhayn had therefore failed to resolve his ambivalence between recognizing the defensive needs of his army and the pressure for quick victory in the west; furthermore, he gave Rupprecht no indications as to how he should execute his open-ended mission. Falkenhayn’s inclinations were to stress speed, to forestall the French advance between Roye and Montdidier, and so emphasized the defensive. Rupprecht overruled him, recognizing that the railway problem would delay the concentration of the 6th army, but anxious to postpone operations until he had sufficient strength mustered for a decisive effort.159
Not until the evening of 23 September did the 6th army begin its advance, moving up the Oise to the left of Compiègne. Bad weather had impeded aerial reconnaissance and the French cavalry was exhausted. Lack of intelligence buttressed Joffre’s belief that a decisive victory was in Castelnau’s grasp if only he would push northwards with enough energy. But Castelnau had acquired enough information during the course of 22 and 23 September to realize that the Germans were pre-empting him and were in strength. He proceeded with caution. By the 24th the limited encounters of the previous few days had given way to fierce fighting along the front from Noyon to Albert. The Germans struck Castelnau’s right flank as he pushed across the Somme. Their efforts were bent on breaking IV corps at Roye, so as to isolate the French to the north.
The French 2nd army hoped to neutralize this danger by continuing to outflank the Germans, but the latter were already at Bapaume on the 26th. The attack at Roye, although held, forced Castelnau onto the defensive. Falkenhayn, encouraged by the limited success at St Mihiel to believe that the French must be weakening their line elsewhere, responded to their lateral moves by resuming the attack in the centre. Launched on 26 September in heavy columns against prepared positions, this advance was repulsed with heavy losses. By the 27th it was clear that Falkenhayn had succeeded neither in destroying the French 2nd army on Rupprecht’s flank, nor in breaking through in the centre.
While Falkenhayn’s attention was being pulled away from his right to his centre and left, Joffre’s purpose remained fixed. On the 25th itself he shifted XI corps from the 9th army to Amiens: by 1 October, using road as well as rail, two more corps, plus three infantry and two cavalry divisions, had set off for Amiens, Arras, Lens, and Lille.160 Castelnau’s army now embraced eight corps and extended along a 100-kilometre front. Its task was no longer to outflank but to hold, while a detachment under Maud’huy of two of its corps and a cavalry corps manoeuvred on Arras. Maud’huy was drawn to the untenanted north-east, to Vimy and the Scarpe valley. The line south of Arras, between Maud’huy’s force and the 2nd army proper, was held by the territorials.
Falkenhayn too continued to manoeuvre on his wing. But although he pushed all his disposable forces towards Rupprecht, and on 28 September directed the 6th army to attack Arras, he did not forsake the possibility of a breakthrough in the centre. Thus, unlike Joffre, either by inclination or as a consequence of the logistical constraints he was not prepared to weaken the centre to reinforce the wing. Given the balance of forces, Rupprecht’s plans seem ambitious. He hoped to hold Maud’huy frontally at Arras and, wheeling north round the city, to envelop the French left wing. To do this, on 3 October he reinforced the reserve corps operating north of Arras, and sent IV cavalry corps from Valenciennes, north of Lens, towards Lille. However, chances of success also beckoned south of Arras, on Maud’huy’s right, where the resistance of the territorials was weakening.
By the evening of the 4th Maud’huy’s position at Arras was threatened with encirclement. The Germans were north-west of Vimy, at Givenchy, and Maud’huy’s left-hand division had lost contact with the cavalry corps to its north. To the south a breach was opening between the territorials and X corps. Maud’huy said he must retreat to avoid ‘a new Sedan’, and Castelnau asked whether Maud’huy should go in the direction of St Pol or Doullens. Joffre’s belief that Castelnau was given to pessimism, formed in the light of the battles round Nancy, found corroboration in his conduct of the battles of Artois and Picardy.161 He feared that the 2nd army would withdraw south of the Somme, so abandoning not only Maud’huy but all northern France. His response was threefold. First, he confirmed Maud’huy’s independence of Castelnau by making his command the 10th army. Secondly, he reiterated to Castelnau that the pressure on the 2nd army would be relieved as the allied strength accumulated in the north: he was on no account to fall back. Thirdly, he acknowledged that the northern armies now constituted a separate area of operations in need of more direct control than he personally could give it. He therefore appointed Foch as his deputy, making him responsible for coordinating the operations of the 2nd and 10th armies and the territorials. Foch’s own army, the 9th, was suppressed.
Castelnau now had the indignity of being under the command of his erstwhile subordinate—the man whom he could with reason see as responsible for the disaster at Morhange. To personal pique was added a difference of temperament. Both were devout Catholics, but Castelnau, André Tardieu recorded, had ‘a taste for nuances’, while Foch had ‘a passion for unities’. Castelnau believed the conception of operations entertained by Foch and by GQG—that the decisive battle of the war had been won and that they were embarked on the exploitation of France’s victory—to be fundamentally misconceived. He was concerned about his right, under heavy attack at Roye, and saw plans for retreat as a sensible precaution.162 Foch continued to look to the left and told Castelnau that withdrawal was out of the question. Maud’huy too was instructed to hold Arras. Energy and determination, not operational or strategic insight, constituted Foch’s contribution in 1914. While this spirit could not easily percolate through to the soldiers of the front line, it was most certainly conveyed to their commanders. Furthermore, he succeeded in projecting his doggedness onto the Br
itish and the Belgians. Within fifty-seven hours of his appointment on 4 October Foch had motored 850 kilometres.163 By 6 October the allied line from the Oise to north of Arras was secure. That evening Falkenhayn decided to cease all further attacks in Picardy; his attention—and that of Foch—was now moving north of Lens, to Flanders and the area between the River Lys and the sea.
The engagement of, first, Heeringen’s 7th army and then Rupprecht’s 6th in the battle on the German right wing had underpinned the original purpose in their withdrawal from Alsace-Lorraine. In the midst of the Marne battle Moltke and Hentsch had plotted their redeployment to quell their fears for the slender line of communications through Belgium. Both in the retreat from Mons and again during the battle of the Marne the Belgian army had sallied forth from its Antwerp base and mounted raids in the direction of Brussels. Later in August a brief appearance by some British marines at Ostend had been seen as a portent of the operational flexibility which seapower vouchsafed the Entente. The fear of a major landing, combined with a Belgian operation from Antwerp, was fed by some of the wilder intelligence received at OHL. The capture of Antwerp and the destruction of the Belgian army therefore promised to remove any invitation to the British, to safeguard German communications, and to release the German troops masking the city for operations on the main front.
Falkenhayn was later to claim in his memoirs that he was anxious to secure the Channel ports as a base for operations against Britain. The very title so often given to the operations consequent on the Marne and continuing until November—’the race to the sea’—supports the idea that the armies were battling for comparative coastal advantage. If the German army secured Calais and Boulogne, then Britain’s Channel fleet would be outflanked and Germany would have opened the southern exit from the North Sea. Britain’s sensitivities to exactly this threat have made its historians ready to impute to OHL the strategy which they themselves would have followed. But there is no evidence that Falkenhayn was thinking on these lines in September and October 1914. The withdrawal of the Grand Fleet to Scapa Flow meant that the German High Seas Fleet was closer to the Belgian coast than were the principal formations of the Royal Navy. Moreover, as the fighting developed along the coast, and specifically on the Yser, the guns of Germany’s battleships could have made a significant tactical contribution. But the fleet—partly out of ignorance as to the Grand Fleet’s position as well as of its own army’s situation—did not push its cause, and Falkenhayn did not ask for its aid. He decided to finish with Antwerp because of the needs and security of the right wing, in the pursuit of a quick victory in France, not because of a belated recognition of the tactical or strategic desirability of securing the Channel ports for the conduct of a long war.164
On 28 September von Beseler’s III reserve corps, made up of five inferior divisions but reinforced with 173 heavy guns, began the siege of the great port. The Belgian perimeter was 96 kilometres in length, but its main defences were concentrated in an outer ring of forts. These provided concentrated targets for the German artillery attack, and because they had not been modernized since 1900 were not proof against the plunging fire of calibres in excess of 210 mm or 220 mm. A dispersed defence, less reliant on armour and concrete, would have stretched the Germans’ heavy artillery and ammunition supply to far greater effect. The city’s physical vulnerabilities were compounded by the divided counsels of those charged with its defence. The general staff, whose chief was responsible for the army when it was in the field, favoured operations that were independent of the city. This line of thought tied in with the hopes of the French mission, who wished to see the Belgian army co-operating in Joffre’s great allied envelopment developing from the south. But while the field army was within the fortifications its command was in the hands of Antwerp’s governor. His problem was that he had insufficient men to hold the perimeter and to keep open a line of retreat to the west. His need was to pull the British and French to the Belgians, rather than to have the Belgians pulled to the British and French. In this respect Belgium’s priorities reflected those of Britain, anxious for the security of the Channel coast. The problem for both the Belgians and the British was that Joffre’s envelopment, even if successful, could not get to Antwerp in time. On 30 September the Belgian government appealed to France and Britain for direct support. On the following day the German infantry began to advance.165
Joffre’s response to Belgium’s appeal was half-hearted. He refused to release anything more than a scratch force of territorials, Zouaves, and marines. Furthermore, their task as he saw it was to cover the Belgian army’s retreat into France, so that it could join the main operations and extend his left wing. While the British were urging the Belgians to prolong Antwerp’s defence, Joffre was simultaneously using his emissaries to persuade them to abandon it.
The British reaction to Belgium’s request was altogether more urgent, if somewhat more muddled. Anglo-French co-operation in the field had produced victory on the Marne, but once more lost its appeal on the exposed slopes of the Aisne valley. On 24 September Henry Wilson suggested that the BEF move back to the allies’ left, so bringing it nearer to its bases and simplifying its lines of communication. Wilson’s proposal was of a piece with Joffre’s grand envelopment, but French was hesitant for that very reason: having nearly lost the BEF in August, he did not want to find himself isolated once more. What swung the commander-in-chief in favour of the idea was the promise of the First Sea Lord, Winston Churchill, that the navy would provide support from the coast. This scheme smacked less of Anglo-French co-operation and more of British independence, a return to the options adumbrated at the beginning and end of August. By early October French was, on the one hand, suggesting to Kitchener that the entire BEF would come to Antwerp’s rescue, while on the other, continuing to give priority to Joffre’s operations. The irony in French’s enthusiasm for Churchill’s plan was that, if effected, it would have exposed the BEF far more than had the French at Mons. Churchill hoped that Antwerp would lock the allied left wing, and that the line would then run through Ghent to Lille. The BEF would have been the occupants of an exposed salient, vulnerable to artillery attack from the south, and its line of communications continuously exposed to flanking fire. Antwerp itself could only have been directly supplied by sea if the allies had been prepared to breach Dutch neutrality in the Scheldt estuary. Joffre agreed to the BEF’s redeployment, but managed the situation by arguing that it could only be effected in stages because the troops were in such close contact with the enemy. The withdrawal from the Aisne began on the night of 1–2 October.
Britain’s real problem was that it had not the men to pursue two strategic options simultaneously. Its decision to send the newly formed 7th division to Belgium did not stop the Belgians concluding on 2 October that the government should leave Antwerp the next day. This was widely interpreted as a resolve to abandon the city entirely, a confusion compounded by Churchill, who thought that what was intended was the evacuation of the field army. On 3 October the Belgians made clear that they would hold Antwerp for a further ten days provided the British and the French could dispatch a relief force in three days. On 4 October Kitchener promised 53,000 men—a force whose main components would be Joffre’s territorials and marines, which the latter was meanwhile directing on Poperinghe and Lille, and the British 7th division, which Joffre wanted landed at Boulogne so that it could extend his left flank.166
In the event, Britain’s direct contribution to Antwerp’s defence was limited. Winston Churchill had himself appointed the British plenipotentiary, and together with the Royal Naval Division undertook to revitalize the city’s defence. The First Lord’s actions generated severe doubts as to his judgement. It was not appropriate for a cabinet minister to leave his post in the middle of a war in order to undertake a de facto field command. Nor was the division which he brought of much use: alongside a brigade of marines it contained two barely trained and inadequately organized naval brigades. Churchill justified his unconventiona
l behaviour by arguing that he had prolonged the defence of Antwerp until 10 October, and that in doing so he gave sufficient time for the allies to secure the Channel ports.167 In reality, Belgium’s decision-making was determined not by Churchill but by the continuing power struggle between its senior commanders. By the 10th only one section of the outer forts had been destroyed, and both the second line and the city’s enceinte were intact; furthermore, von Beseler’s guns were running short of ammunition.168 Arguably Churchill failed to get the Belgians to hold out as long as they might have done. Britain’s main role was not to sustain Antwerp’s powers of defence, but, through the arrival at Zeebrugge and Ostend of the 7th division—reinforced by a cavalry division, and somewhat grandly called IV corps (commanded by Henry Rawlinson)—to expedite the Belgian withdrawal.
As IV corps fell back it was brought into line with the BEF proper, being directed to the area north of Ypres. It took up its positions on 14 October, and was joined by I corps on 19 October. Cavalry covered the gap south of Ypres, while by 12 October II and III corps were advancing to the front La Bassée-Armentières. French had suggested to Joffre that the Belgian plain would give freedom of action for his cavalry,169 while Joffre had hoped that the BEF would be in time to secure the industrial centre of Lille. But the strength of German cavalry now concentrated in the area (eight divisions) forced the BEF to detrain so far to the west—at Abbeville, St Omer, and Hazebrouck—that it was too slow coming up. No serious attempt was made to defend Lille, and on 12 October it was occupied by the Germans.
Joffre failed to grasp how disorganized the retreating Belgian army had become. He assumed that it could return to the fray within forty-eight hours, intending it to swing inward towards Ypres, and so extend once more his left wing. King Albert had different ideas: his priority was to preserve his army and even, if possible, a bit of Belgium. He wanted to move south-west, not south-east. Moreover, by 14 October Beseler’s III reserve corps had entered Bruges and was reaching towards Ostend and Roulers, thus threatening the Belgian wings. Rawlinson’s IV corps was pushed out from Ypres to Roulers as part of the more general move to the east. It was checked almost immediately, but its effect was to draw the Germans away from the Belgians. The latter wanted a defensive position behind which they could consolidate, absorbing Antwerp’s fortress troops into a reorganized field army and restoring their shattered morale. The line of the Yser, its flanks resting on Dixmude and Nieuport, had been identified as just such a position in 1913, and it was thither that the Belgian army moved on 15 and 16 October. The French marines were told by Foch to hold Dixmude at all costs.
