by Hew Strachan
He was extraordinarily lucky. He had assumed that I corps would close up, having passed the forest of Mormal, and so secure his right: in reality, the effect of Landrecies had been to widen the gap between the two corps. However, Kluck still imagined that the British line of retreat lay to the south-west, not the south, and the Germans had therefore failed to appreciate the British alignment, and—as at Mons—had thus found their attack deprived of coordination. Once again poor German intelligence allowed inferior British forces to fight a successful delaying battle against Kluck’s army. During the battle von der Marwitz’s cavalry corps fought dismounted, and therefore did not push towards II corps’ flanks; after it, it failed to pursue. On the afternoon of the 26th Smith-Dorrien’s corps left the battlefield. By 28 August it was 56 kilometres to the south, and the gap between Kluck and itself had been widened as the former persevered with the south-westerly direction of his advance.
From Alsace, through Lorraine and the Ardennes to Mons and Charleroi, the allied armies had been forced to retreat. By the evening of 23 August the battle of the frontiers was over. At 9.35 the following morning Joffre reported to Messimy that his armies would be passing from the offensive to the defensive: ‘Our object must be to last out as long as possible, trying to wear the enemy out, and to resume the offensive when the time comes.’77
Joffre’s critics suggest that these initial defeats were self-inflicted, and harp on the slowness with which he realized the full scale of the German envelopment through Belgium. This is only partly justified. The German advance did not begin until 18 August. Thus, little could be clear before then: between 6 and 15 August Sordet’s cavalry corps conducted a long-range reconnaissance into Belgium but found nothing, as the Germans had not yet begun to move. However, after 18 August preconception played powerfully on Joffre: as late as 23 August he was underestimating the German strength in Belgium by 300,000 men. Not until 25 August was the veil truly pulled to one side.
PROBLEMS, AND SOME SOLUTIONS
What followed looked far worse. Companies, battalions passed in indescribable disorder. Mixed in with the soldiers were women carrying children on their arms or pushing little carts in front of them, girls in their Sunday best, old people, carrying or dragging a bizarre mixture of objects. Entire regiments were falling back in disorder. One had the impression that discipline had completely collapsed. Nonetheless all these units had regular cadres and belonged to the active army.78
This was how Le Matin on 24 August 1914 reported the retreat of Castelnau’s XV corps, shattered by the German 6th army at Morhange. It was not an isolated account. As the wounded were conveyed to the rear they propagated rumours and bore evidence on their bodies of realities which both shattered popular illusions and jeopardized the army’s morale. With them came others whose departure from the front was less honourable. In Caen, at 8 p.m. on 27 August, 120 men of the 27th Territorial Infantry alighted. Under the command of two officers, and bearing the regimental colours, they had panicked at Valenciennes, boarded a train for the capital without a travel order, and then been directed westwards.79 As the French and British armies fell back from the frontiers retreat threatened to become rout.
Joffre was alive to the danger. On 10 August the French military authorities had been accorded the right to use the death sentence whenever immediate repression was essential. On 1 September the ministry of war instructed officers to carry out death sentences within twenty-four hours unless there were good grounds for clemency. This became the norm. Soldiers were executed without trial.80 By October rumours had reached Britain that those who had abandoned Namur were punished by being drilled under heavy shellfire, and that a unit of Niçois who had run away in Lorraine had been stood with their backs to the enemy in front of their own trenches—again under shellfire.81
Retreat did not turn into rout. The resilience of the French army’s morale, despite the heavy casualties and tactical defeats of August, argues for the positive functions of punitive and repressive discipline. But this was not the only reason why the army did not collapse. The retreat generated its own determination. After Le Cateau a British unit, the 2nd battalion, the Cameronians, marched 91 kilometres in thirty-six hours, and the BEF as a whole covered 320 kilometres in thirteen days.82 On its right, the French 5th division, in Lanrezac’s 5th army, averaged 22 kilometres a day in the week before 6 September. Exhausted by continuous marching in late summer, the soldiers longed to halt. When the order to attack finally came it seemed to Joffre’s men that he had done no more than reflect their own mood: ‘the idea of the Marne’, wrote one junior officer, ‘surged up from the army itself.’83
Joffre’s own resolution never wavered. Outwardly he appeared imperturbable. He ate regularly and well; he insisted on undisturbed sleep. His calm conveyed itself to his staff, and from it to his armies. His claim to be a great commander rests on his conduct on 25 August and the days immediately succeeding it. During that time the competing interpretations of the unfolding situation could so easily have resulted in the implosion of French general headquarters, Grand Quartier Général (GQG). That it did not do so was a reflection of Joffre’s moral authority.
The dominant personality at GQG, after Joffre himself, was the energetic but corpulent deputy chief of staff, Berthelot. So modest and self-effacing was Berthelot’s direct superior, Belin, that this inversion of the hierarchy was not itself productive of tension. The trouble was that Berthelot embraced operational concepts that did not always accord with reality or with realism. Thus, the pre-war division between the 3eme bureau (operations) and the 2eme bureau (intelligence) persisted, and it was compounded by disagreements within the 3eme bureau itself. Joffre’s chef de cabinet, Maurice Gamelin, although as persuaded of the merits of the offensive as Berthelot, could not but recall Foch’s pre-war conviction that the German right wing would extend across the Meuse and up to the Channel coast. From 16 August others at GQG woke up to the reach of the German envelopment, but Berthelot was not one of them. Even as late as 29 August, according to Huguet, the French liaison officer with the BEF, ‘he did not appreciate fully the importance of the German threat to our left flank’.84 Joffre’s reluctance to listen to the Belgians needs little further explanation.
No generals in the history of warfare had commanded such large armies on such a broad front as did Joffre and Moltke. The problems of communication and control were enormous. However, by placing GQG close behind the front at Vitry-le-François, Joffre positioned himself so as to be able to maintain regular contact with his armies. He strove for ubiquity. On 26 August he drove 340 kilometres, on the 28th 390, and on the 29th 240.85 When he could not see the situation for himself, he listened to the reports of his liaison officers. These men, relatively junior but frequently brilliant, were Joffre’s eyes and ears. They picked up the mood of the generals, and compared their own impressions of the situation with the reports GQG was receiving. In so far as was possible, Joffre’s views became grounded in reality.
The opening defeats were the sanction which Joffre needed in order to reconstitute the French army. The apprehension, and even distrust, felt by the generals towards GQG’s liaison officers were well-founded. The average age of French generals on mobilization was 64.2. The mobilization of reserve divisions had resulted in the recall of retired officers, and their commanders were on the whole almost three years older than those of active divisions. General Brugère, who had been vice-president of the Conseil Supérieur de Guerre in 1900–6, assumed the command of a territorial division aged 73. Pau and Maunoury, both summoned to command armies, were aged 67. Some of these older men performed distinguished service; others lacked the fitness to cope with the physical strains. However, age was not the only problem of the French high command; German generals were no younger. In 1900, in the aftermath of the Dreyfus affair, promotions had become subject to ministerial approval. Between 1900 and 1904 General André had promoted officers of proven republicanism. During this period 290 brigadiers had been appointed and 125 divis
ional generals: many of them were among those recalled in 1914. Professional qualities were at a discount. Even more important than those who were promoted were those who were not. The average period spent as a lieutenant-colonel in the pre-1914 French army was 3.5 years. Sarrail, whose republican credentials were impeccable, held the rank for 1.5 years. By contrast, Foch and Fayolle, both to become marshals of France but both educated by Jesuits, were lieutenant-colonels for five and eight years respectively. In 1911 Joffre, with Messimy’s support, had begun the process of restoring professional qualifications as the prime claim to rank. More importantly, his cabinet had set about informing itself of the qualities of France’s officer corps. On mobilization the bulk of this cabinet became the staff of GQG. Messimy, on 10 August 1914, with his fondness for the memory of 1793, authorized Joffre to conduct a purge of the higher command. The biggest single factor uniting those who were limogés was age, and what united those who were not was professional competence. Many of those who suffered did so because they had been given in wartime the corps or divisional commands that had eluded them in peace. Of the ten commanders of territorial divisions eight were removed, and their average age was 62. By 6 September 1914 Joffre had dismissed three army commanders, seven corps commanders, thirty-four divisional commanders, and fourteen brigadiers. When casualties are included, almost one-third of the French high command had changed. By the end of the year 162 commanders of higher formations had been sacked; only three army commanders, six corps commanders, and twenty-one divisional commanders had remained in post since mobilization. Furthermore, the changes which Joffre effected determined the senior command of the French army for the rest of the war. Foch, commanding XX corps on mobilization, became an army commander on 30 August, and took over an army group, as Joffre’s deputy, on 5 October. Philippe Pétain, about to retire as a colonel when mobilization came, was a temporary general by the end of August, and an army commander by June 1915.86
Pétain was exceptional. He was one of only three officers aged 56 or over in 1914 who were not already generals but who rose to become army commanders in the course of the war (the others were Nivelle and de Mitry). Of the remaining thirty-six who claimed this accolade, the vast majority were therefore pursuing an upward trajectory already set in peacetime and determined by professional competence rather than by republican loyalties. Pétain had been held back by his views on tactics, not on politics. Thus, Joffre’s purge was dependent on his knowledge of, and belief in, the qualities of the officer corps as a whole. It reflected his conviction that the battle of the frontiers was a series of tactical defeats.
Lackadaisical leadership had characterized the battles in Alsace, in the Ardennes, and on the Sambre. But the defeats were not entirely the fault of the generals. The tactical instructions issued both by GQG on 16 and 18 August and also by individual army commanders reflected pre-war precepts about the difficulties of the offensive. Stress was put on artillery support, on the need to develop fire superiority, on the combination of artillery with infantry, and on immediately entrenching once ground was gained. In the event the infantry did not wait, did not prepare their attacks with method. This was less a consequence of the ‘spirit of the offensive’ and more of a lack of training. The three-year law had not only come too late, it had also been implemented by taking in two classes in 1913 rather than retain the 1911 class for a further year. Therefore, in July 1914 two-thirds of the conscripts then serving had been in uniform for less than a year, and none of them had been on manoeuvres; the 1912 class was still under arms but had only attended divisional or corps manoeuvres, as army manoeuvres were held only once every four years. Nor was this inexperience confined to the most junior ranks. A French infantry company had on average less than half the number of non-commissioned officers enjoyed by a German company. In 1913 the French army was short of 6,000 sergeants and 23,000 corporals. Thus, only one problem addressed by the three-year law, that of manpower, had been resolved; the other, that of tactical competence, would take longer to put right than the July crisis had allowed. In August 1914 the excitement of action overtook professional wisdom. The influence of rhetoric proved greater than that of common sense.87
Matters were not helped by the manner in which the French used their 75 mm field guns when on the attack. In 1897 France had led the world in the introduction of quick-firing field artillery, and in 1914 the 75 mm was still arguably the best of its type. Although the Germans had upgraded their 77 mm gun, first introduced in 1896, in 1905, they had concluded from the Russo-Japanese War that long-range fire with field guns was useless: its fall could not be adequately observed and its shrapnel shells in particular were ineffective against field fortifications and enemy gun-pits. These tasks were transferred to heavier artillery, and in particular the 105 mm light howitzer and the 150 mm heavy howitzer. The key attribute of the 77 mm became its mobility, and the facility with which it could accompany infantry over broken ground.88 By contrast, French artillery doctrine was built around the light field gun rather than a blend of such guns with heavier artillery. The French had no howitzers at any level up to that of the corps; they argued that the German 105 mm and 210 mm howitzers were slow to come into action, tended to fire high, and did more moral than material damage. Speed of deployment and rate of fire—it was claimed a 75 mm could fire twenty rounds a minute—were rated over weight of metal. Neutralizing the enemy was more important than destroying him. French field guns were therefore organized in four-gun batteries, not the six-gun batteries favoured by the Germans, and so a French corps could field thirty batteries, although armed with only 120 field guns, to a German corps’ twenty-four batteries made up of 144 field guns. In all, France could equip 999 batteries to Germany’s 812.
To support this approach to the use of artillery, its command was delegated down to subordinate formations. Percin, the inspector-general of artillery between 1907 and 1911, wanted to establish direct liaison between the artillery battery and the infantry company, his idea being that the infantry would use the support of the guns as it encountered resistance to its advance. Therefore, the French artillery did not prepare the attack, did not exploit its maximum range, and did not counter the enemy’s artillery. No range tables for the 75 mm firing high-explosive shell had been drawn up before the war, and its normal maximum range with shrapnel had become set at 4,000 metres. The guns’ task began when the infantry entered the danger zone from enemy small-arms fire, at say 800 metres, and lasted until it was within 350 metres. They could not perform this task from a distance or indirectly, as the army had insufficient telephone line to link the forward observers to the batteries: the guns had either to use direct fire or to be controlled by an observer on an adjacent hill. Although the 75 mm was technically capable of high-angle, indirect fire, using the ‘plaquette Malandrin’ to curve the shell’s trajectory, such tactics required the gun to be dug in and therefore militated against the mobility and speed which had become the artillery’s watchwords. Without such preparation the 75 mm’s maximum elevation was 18 degrees.89
The fighting on the frontiers had been in broken and wooded ground, demanding indirect fire and high-angle elevations. The French infantry had suffered from the attentions of the German heavy artillery, often dug in, in prepared and covered positions, and capable of inflicting damage at long ranges. But trials even at this early stage of the war showed the maximum range of the 75 mm with high-explosive shell was 6,800 metres to the 5,500 of the 77 mm, and at Mortagne, with a delayed-action fuse and its trail dug in to increase its elevation, it had ranged 9,000 metres and provided a match for the German howitzers.90 Ballistically the 75 mm was far superior to the 77 mm. Its cone of fire was long, deep, and dense compared with the more perpendicular cone of fire characteristic of the German gun, and its long carriage enabled it to sweep fire to left and right. Its shrapnel shell was slightly heavier, and its high-explosive charge was 650 grams compared to the 77 mm’s 190 grams. Using delayed-action fuses for ricochet fire, a 75 mm battery was able to sweep
an area of 4 hectares, 400 metres deep, in forty to fifty seconds, and three batteries could cover 12 hectares in a minute in the so-called rafale. Although the fire was not accurate, it was precise in the height at which it burst: between one and three metres. A 75 mm battery firing shrapnel, even if it only fired ten rounds per gun, discharged 10,000 balls a minute. Thus it was far more effective against advancing infantry than any small-arms or machine-gun fire.91
Therefore, the disadvantages in French artillery doctrine when on the offensive were neutralized on the defensive. Able to choose their ground, the French presented their 75 mms with good fields of fire. The comparative immobility of the heavy artillery worked against the Germans if they were the attackers; its shell supply could not keep up with the advance, and it could not choose prepared or concealed positions.
The change from the offensive to the defensive is the best explanation for the transformation in tactical performance. Joffre’s tactical instructions of 24 August, so often cited as the cause of the transformation in the French army’s performance, were little different from those of 16 and 18 August. He reiterated the need not to launch the infantry prematurely or from too great a distance. Above all, he emphasized that the role of the artillery was not only to support but also to prepare the attack.