When he was killed in 1870, Ardant du Picq had already distilled many insights into combat psychology from his own experiences and from the existing rich fund of French military writing. After the débâcle, however, the duty of carrying forward his important work fell to a new generation. At first there was understandable turmoil and confusion, as a new army was gradually pieced together, but by the 1890s the newly-established General Staff had become an effective organiser of combat analysis. A Higher War School, later joined by a Centre of Higher Studies, had been set up alongside the historical section of the Staff.
The work of these organisations, although brusquely terminated in 1914, included an unsurpassed scholarly contribution to our understanding of Napoleonic and Franco-Prussian War history. This scholarship was never intended to be impartial, however, since the aim was always to use the lessons of the past as a guide to action in the future. Its authors looked forward not to retirement in a medieval university, as did an Oman or a Delbrück, but to glorious service on the field of honour. Many of the best of them were killed there, including Colin and de Grandmaison – but others found promotion to the very highest levels of command. De Maud’huy went on to command an army, Foch became generalissimo of the entire Western front … and Pétain eventually became the head of state.
We have already examined Colin’s perception that infantry combat was based essentially on firepower. He had his disciples, including Captain ‘Danrit’, Bressonet, Pétain himself, and several students of the Boer War.86 In 1904 the official tactical regulations also distanced themselves a little from the emphasis on the offensive that had been evident in the 1884–7 manuals, and still more so in those of 1894–5.87 The 1905 two-year service law and Japanese successes in Manchuria, however, soon swung the fashion back again. Many French officers thereafter agitated for the regulations to be rewritten in a more ‘offensive’ manner, and their efforts were crowned with success by the end of 1913, especially in the Réglement de Manoeuvre d’Infanterie of 20 April 1914.88 Ironically, this document starts by stressing the need to avoid constant changes of doctrine, since they would create a malaise, yet it represents precisely such a change itself.89 If we wish to find a reason for French failures in August, the fact that their tactics were only four months old may not be the least relevant.
The 1914 Réglement echoed many contemporary opinions when it claimed that
The bayonet is the supreme weapon of the infantryman. She plays the decisive role in the assault, towards which all attacks should resolutely aim, and which alone allows an adversary to be put definitively out of action.90
Nevertheless, the whole system envisaged by these regulations is based on an assumption that the young conscripts will be simultaneously enthusiastic for combat, yet calmly obedient to orders. The officers will have been carefully trained both to ‘read the battle’ and to keep a strong control of their troops. They will not attack until the enemy has first been reconnoitred and suppressed by fire, a process that will be helped by the battalion’s two machine guns advancing close behind the skirmish line.91 As far as it goes, this is scarcely a stupid recipe for victory, although its starting assumptions about the quality and training of the participants were doubtless wildly over-optimistic.
The French problem was not so much how to design theoretically effective assault tactics, but how to train the new two-year conscripts to carry them out reliably. One of the most interesting views on how they should be motivated appeared in de Maud’huy’s book Infanterie, which was published in 1911. De Maud’huy was a keen advocate of the work of Ardant du Picq, but he had also made a considerable study of the infant science of psychology. He corresponded with Gustave le Bon, among others, whose Psychologie des Foules (1895) was a runaway best seller, not least in proto-fascist and military circles.92 Le Bon was interested in the way large groups were motivated by irrational subconscious impulses, whereas individuals or small groups might behave perfectly sensibly. An individual leader might therefore be able to decide a correct policy, but he would have to use well-chosen symbolism and rhetoric, rather than logical reasoning, if he were to move the masses. By the same token, democratic or parliamentary decision-making was intrinsically flawed, since it would rapidly descend to the lowest, least intellectual, common denominator.
In terms of military tactics, this analysis led to an astonishingly ‘modern’ understanding of the fears and stresses which the soldier encounters in combat. Like S. L. A. Marshall a half-century later, de Maud’huy had a vision of the individual sensing himself alone in battle, prey to surges of irrational terror because of what he did not see, more than because of what he did. He would fight best if he knew his comrades were watching him, whereas fatigue and enforced immobility under fire would be vital contributors to increased stress.93 In a passage unconsciously echoing Jackson’s view of his 1815 militia at New Orleans, furthermore, de Maud’huy said that normal crowds were incapable of manoeuvres in open terrain, but could fight only from behind barricades or other fortifications.94 It was therefore the twofold job of the officer to manipulate the military crowd so as to draw the best from it, at the same time as converting it into a higher, more rational and more regular type of organisation. In combat the unit had to be prevented from degenerating into a formless mob for as long as possible, while the enemy had to be encouraged to take on crowd characteristics and abandon his own regularity. Racial factors would be very important in the way these military crowds behaved, since German crowds were less mercurial or audacious than French ones, so officers should have a deep understanding of this element.95
The French 1914 doctrines of the offensive drew heavily upon pseudo-scientific social, racial and psychological concepts, but they were mightily reinforced by analysis of the operations in 1870. This was not so much a matter of the most minor levels of tactics, however, where some new assault techniques had been glimpsed during the war, but at a higher level of army manoeuvres. Despite some effective local counter-attacks, the general French posture in 1870 had been passive and defensive, with both Bazaine in Metz and Napoleon III in Sedan contriving to get themselves encircled. Around 1900, therefore, the perception was that defeat in 1870 had been caused by allowing the Germans to seize the initiative and complete their encirclements undisturbed. In a future war this should not be allowed to happen again: it should be the French who seized the initiative, disrupted the enemy’s plans, and threw him into paralysis.
This view was expressed most pithily in the work of Colonel (later General) Louis de Grandmaison, a staff officer who had made a close study of recent military history. In 1906 he seemed to lead the widespread reaction to the ‘cautious’ regulations of 1904, by publishing Dressage de l’Infanterie en Vue du Combat Offensif.96 This was certainly concerned with minor tactics but, despite rhetoric condemning a passive approach, it actually deepened the 1904 call for increased supporting fire in the attacks. As de Maud’huy was to do more thoroughly five years later, he also recognised the growing importance of stress and fear on the modern ‘empty’ battlefield, and cited some of Le Bon’s ideas about crowd psychology.97 He then proposed that stress should be overcome by solid training, clear doctrine, small group cohesion, and plenty of artillery support. This is far from a pernicious approach to the problem, although in the event it was acclaimed more for its continued acceptance of the frontal attack than for its more useful mechanisms to make such an attack actually work.
More influential within the army – and correspondingly more damaging to de Grandmaison’s posthumous reputation – were to be two staff lectures advocating the offensive which he published in 1911.98 To these have been attributed the predominance of ‘Young Turks’ on the General Staff in 1914, and the idea that a ‘cult’ of the offensive was the touchstone of any commander’s efficiency. Two sets of regulations for large unit operations were inspired by de Grandmaison in 1913; although contrary to the popular misconception, his mature doctrines were concerned with minor tactics much less than with ‘the engagement
of large formations’. They were intended as a contribution to a complex technical debate that had been raging for many years, concerning manoeuvres at the ‘operational’ level of action.
Everyone agreed that Napoleon had been right to advance into contact with the enemy by using a dispersed fan of army corps, closing up for decisive combat once the key point had been identified – by 1911 no French commander worth his salt would contemplate setting about things in any other way – but the problem was to decide just precisely how Napoleon had intended it should happen, and how he would have applied it himself if he had been born a century later. Since 1895 the accepted orthodoxy had been that each large formation should be preceded by a vanguard, to find the enemy and guarantee ‘security-with-confidence’ (sûreté) for the commander while he stopped to design an appropriate plan. The need for security applied still more to the flanks, where there was an enormous nervousness that the Germans would apply their customary doctrine of encirclement. Yet in their exercises the French found that reconnoitring vanguards tended to fight only withdrawal actions when they met the enemy, while fear for the flanks tended to suck the follow-up troops outwards rather than forwards, thereby dissipating their offensive power.99 The overall effect was one of hesitation, and the initiative was left with the enemy.
In de Grandmaison’s two lectures, therefore, he argued that the French fan of dispersed formations should not be allowed to open out when the Germans were found, but should redouble its forward march. The attacking troops should also be well concentrated towards the front, not echeloned rearwards in an over-complex succession of detachments and reserves. Apart from anything else, he pointed out that such an arrangement would possess a simplicity that might, at long last, allow every participant to understand what was expected from him. De Grandmaison’s message was that the Germans’ strict march procedures were vulnerable to dislocation if their preconceived timetable could be upset. They themselves were aware of this vulnerability, however, and hoped to keep the French at bay through the use of strong and aggressive forces in their vanguards. By adopting a more flexible and tentative approach, with weaker vanguards, the French were therefore voluntarily abandoning the initiative to their opponents. They would do better to take a leaf out of the Prussian manual and batter down the initial opposition while it was still weak. By ‘biting fast and deep’100 the French could upset the rigid enemy plan and win an important advantage in both morale and operational posture. They could win security not by a physical defensive screen, but by a dynamic offensive movement.
In the event, in 1914, the Germans won some important advantages precisely because their vanguard screens in Belgium were heavier and more aggressive than those of the French. Both sides also suffered from just the type of worry about flanks that de Grandmaison had warned against,101 and it was often by threatening encirclement that the Germans were able to maintain their mobility forward into northern France. They seized the initiative in very much the manner that had been expected, and de Grandmaison cannot be faulted for these aspects of his analysis.102 Where his recipe came unstuck, however, was in the way his brusque attacks were actually executed.
In Joffre’s conception of Plan XVII, as it had evolved by the middle of August, there were to be three separate offensives – an early right hook from Nancy passing Metz to the south, a left hook attacking frontally through Charleroi to repulse the enemy spearhead in Belgium, and a slightly delayed main assault through the Ardennes to cut through the Germans’ rear.103 Each attack consisted of two armies in line abreast, with those in Lorraine and Charleroi pressing forward for a frontal clash in very much the concentrated formation that de Grandmaison would have wished. The attack into the Ardennes, however, was more tentative, more widely dispersed and fragmented. It cannot be attributed to de Grandmaison’s vision, but to Joffre’s more cautious and traditional idea of waiting to collect intelligence, and then launching a central breakthrough ‘in the style of Austerlitz’ where the enemy was believed to be weak. When the intelligence turned out to be wrong, and the enemy was found to be advancing through the Ardennes with numerical superiority, both the strategic and the operational conceptions collapsed. Equally, Lanrezac’s Fifth Army at Charleroi lost its offensive impetus when it found an attractive position along the south bank of the Sambre. Far from making an attaque à outrance, Lanrezac stuck fast and fought defensively for three days before following the BEF in the general rearward movement.104
The best test of de Grandmaison’s theory of the offensive should perhaps be sought in the Lorraine attack conducted by Dubail’s First and de Castelnau’s Second armies. Between 19 and 20 August these came into action at Sarrebourg and Morhange respectively, on an overall frontage of around thirty miles. They attacked, but often without following de Grandmaison’s principles at all closely. For a number of reasons they were held fast by an enemy standing in position, and then retreated due to counter-attacks and threats to their flanks: it all made rather an ignominious end to three years of doctrinal theorising and agitation.
At Sarrebourg the French VIII Corps attacked against approximately its own numbers on a seven-mile front to the north-east of the town. An initial frontal attack by three regiments from the town itself was stopped at the foot of a long open slope by artillery and machine gun fire. The assault made better progress when it was renewed on the 20th with increased artillery support, but an enemy counter-attack finally drove it back. In the afternoon the French had to fight a defensive house-to-house combat in Sarrebourg, to cover their retreat.
The battles for Lorraine, 19–20 August 1914.
Further to the left the terrain was more broken, wooded and undulating. Despite a failure to assault at dawn, two regiments of the 15th Division each took their initial objectives – one by skirmisher infiltrations, the other by cold steel alone. Heavy artillery fire and a solidly fortified infantry position soon halted both attacks, however, and a third regiment failed to take even its first objectives. Pressure on the flanks forced a retreat by lunchtime, although the French observed that their own artillery could be just as effective in the defence as the Germans’. The ‘decisive battle of Sarrebourg’ was thus a very short-lived and small-scale episode indeed, with little more than 24,000 French troops engaged for little more than a day, and around 8,000 casualties. This was the highest loss of any corps in First Army, but by nineteenth-century standards it was scarcely a shocking figure for such a crucial engagement.105
A very similar story, albeit on a rather larger scale, can be told of the battle of Morhange. Second Army attacked with three corps in line on a twenty-mile front, against an essentially equal enemy force. On 19 August Foch’s XX Corps on the left established itself on the ridge running east towards Morhange village, but there was some heavy fighting. De Grandmaison himself was wounded twice, and his regiment was to lose 1,100 casualties in 24 hours. Meanwhile, in the natural amphitheatre below, an eyewitness reported the 37th regiment’s ‘magnifique’ advance under shelling, in bounds by half platoons (sections), to capture Conthil. Alas, they were evicted at 7 o’clock next morning by a counter-attack, although pursuit was halted by two companies and two machine guns from the 26th regiment, supported by artillery. The 79th regiment subsequently fought a heavy battle around Lindrezing, but neither side seemed to have arranged proper artillery co-ordination, and there was an echeloned disengagement in the early afternoon, Foch’s intended grand offensive to capture Morhange on the 20th had been both pre-empted by the enemy’s own offensive, and countermanded by the army commander’s orders. It was scarcely what de Grandmaison can have had in mind in 1911.106
On the centre and right of de Castelnau’s army the advance was no more spectacular. XV and XVI Corps made a few gains, notably the village of Bidestroff, but were unable to debouch either from there or from the edge of a long wood facing the fortified villages of Cutting and Loudrefing. When the advance bogged down on 19 August there was a good deal of digging in, with some units ‘attacking’ only by fire.107 Th
en next day the German counter-attacks came in as heavily as at Conthil and Lindrezing, making gains particularly along the junction between the XX and XV Corps areas. Bidestroff fell after severe shelling and two assaults, and by mid-morning de Castelnau was facing the possibility of a central breakthrough of his position without reserves to fill the gap. He became convinced that he had to pull back his entire army out of contact.108
One important feature of the French system was that above the level of the three-battalion regiment, each unit had two sub-units. There were two regiments to a brigade, two brigades to a division and two divisions to a corps. This meant it was awkward for a commander to keep a proper balance between the troops committed to action and those kept in reserve, since an attack ‘one up’ would leave a massive 50% in reserve, while ‘two up’ would leave nothing at all. At Sarrebourg-Morhange the two armies certainly attacked in a long line of divisions, usually without a significant operational reserve; and this surely deprived them of a measure of resilience. De Grandmaison had happily accepted this effect in principle, but only on condition that the line was not too long and all its parts attacked simultaneously and vigorously. In practice, however, his preconditions were not properly met. The attack frontage of First and Second armies in 1914 was about twice as long as he would have envisaged, and the advance of each element was not properly synchronised. This in turn opened the door to all the anxieties about encirclement that he had originally feared and tried to avoid.
Forward into Battle Page 14