FOUR

THE BATTLE OF THE FRONTIERS AND THE MARNE

STATESMEN WERE FILLED WITH foreboding by the coming of war but its declaration was greeted with enormous popular enthusiasm in the capitals of all combatant countries. Crowds thronged the streets, shouting, cheering and singing patriotic songs. In St Petersburg the French ambassador, Maurice Paléologue, found his way into the Winter Palace Square, ‘where an enormous crowd had congregated with flags, banners, icons and portraits of the Tsar. The Emperor appeared on the balcony. The entire crowd at once knelt and sang the Russian national anthem. To those thousands of men on their knees at that moment the Tsar was really the autocrat appointed of God, the military, political and religious leader of his people, the absolute master of their bodies and souls.’¹ The day was 2 August. On 1 August a similar crowd had gathered in the Odeonsplatz in Munich, capital of the German kingdom of Bavaria, to hear the proclamation of mobilisation. In it was Adolf Hitler who was ‘not ashamed to acknowledge that I was carried away by the enthusiasm of the moment and . . . sank down upon my knees and thanked Heaven out of the fullness of my heart for the favour of having been permitted to live in such times’.² In Berlin the Kaiser appeared on his palace balcony, dressed in field-grey uniform, to address a tumultuous crowd: ‘A fateful hour has fallen upon Germany. Envious people on all sides are compelling us to resort to a just defence. The sword is being forced into our hands . . . And now I command you all to go to church, kneel before God and pray to him to help our gallant army.’ In the Berlin cathedral, the Kaiser’s pastor led a huge congregation in the recitation of Psalm 130 and at the Oranienstrasse synagogue the rabbi conducted prayers for victory.³

There were to be similar scenes in London on 5 August. In Paris it was the departure of the city’s mobilised regiments to the Gare de l’Est and Gare du Nord which brought forth the crowds. ‘At six in the morning’, an infantry officer reported,

without any signal, the train slowly steamed out of the station. At that moment, quite spontaneously, like a smouldering fire suddenly erupting into roaring flames, an immense clamour arose as the Marseillaise burst from a thousand throats. All the men were standing at the train’s windows, waving their képis. From the track, quais and the neighbouring trains, the crowds waved back . . . Crowds gathered at every station, behind every barrier, and at every window along the road. Cries of ‘Vive la France! Vive l’armée’ could be heard everywhere, while people waved handkerchiefs and hats. The women were throwing kisses and heaped flowers on our convoy. The young men were shouting: ‘Au revoir! A bientôt!’⁴

All too soon, for most of the young men, the summons to follow would come. Reservists not yet called were already putting their affairs in order; in most armies the day before the stipulated date for reporting was a ‘free day’ for farewells to family and employer. ‘Complete strangers’, recorded Richard Cobb, the great historian of France, ‘could be heard addressing one another in bizarre fashion, as if Parisians had all at once become figures out of Alice [in Wonderland]: playing cards, days of the week, or dates in a new sort of calendar. “What day are you?” And, before the other could get in an answer, “I am on the first” (as if to suggest: “beat that”). “I am the ninth” (“Bad luck, you’ll miss all the fun, it’ll be over by then”). “I am the third, so won’t have to wait too long.” “I am the eleventh” (“You’ll never make Berlin at that rate”).’⁵ A German officer-candidate reservist gives a more prosaic account of how the procedure swept up the individual. He was on business in Antwerp. His military document told him he had to report

to the nearest regiment of field artillery on the second day of mobilisation . . . When I reached Bremen on 3 August, my family was frantic. They thought the Belgians had arrested and shot me . . . on 4 August, I presented myself to the army as a reservist and was told I now belonged to Reserve Field Artillery Regiment No. 18, which was forming in Behrenfeld near Hamburg, about seventy-five miles [away]. Relatives were not allowed near the building where we had to assemble. As soon as I could I gave a message to a little boy so my family knew . . . Relatives were not allowed on the railway platform either, only Red Cross people who gave us free cigars, cigarettes and candy. On the troop train I was glad to see friends I knew well from my rowing and tennis clubs . . . On 6 August I was issued my field-grey uniform which I had never worn before. The colour was grey-green with dull buttons, the helmet was covered with a grey cloth so that the ornaments would not glitter in the sun and the high riding boots were brown and very heavy . . . All soldiers and most of the officers were reservists but the commanding officer was a regular . . . Most of the NCOs were regulars. The horses were reservists, too. Owners of horses – sportsmen, businessmen and farmers – had to register them regularly and the army knew at all times where the horses were.⁶

Horses, like men, were mustering in hundreds of thousands all over Europe in the first week of August. Even Britain’s little army called up 165,000, mounts for the cavalry and draught animals for the artillery and regimental transport waggons. The Austrian army mobilised 600,000, the German 715,000, the Russian – with its twenty-four cavalry divisions – over a million.⁷ The armies of 1914 remained Napoleonic in their dependence on the horse; staff officers calculated the proportion between horses and men at 1:3. Walter Bloem, a reserve officer of the 12th Brandenburg Grenadiers, packed as much luggage for his two horses as himself when he mobilised at Stuttgart: ‘my trunk, my brown kitbag, and two boxes of saddlery . . . with the special red labels. “War luggage. Immediate”’ before sending them ahead by train to Metz on the French border.

Trains were to fill the memories of all who went to war in 1914. The railway section of the German Great General Staff timetabled the movement of 11,000 trains in the mobilisation period, and no less than 2,150 fifty-four-waggon trains crossed the Hohenzollern Bridge over the Rhine alone between 2 and 18 August.⁸ The chief French railway companies, Nord, Est, Ouest, PLM, POM, had since May 1912 had a plan to concentrate 7,000 trains for mobilisation. Many had moved near the entraining centres before war began.

Travellers coming in [to Paris] from Melun brought extraordinary accounts of empty, stationary trains, engineless, and often of mixed provenance, the carriages from different companies strung up together, passenger ones mixed up with guard trucks, many with chalk marks on their sides . . . waiting on side-lines the whole way from the chef-lieu of the Seine-et-Marne to the approaches of the Gare de Lyon. Equally bizarre were the reports brought in by travellers to the Gare du Nord of the presence along the immense sidings of Creil of several hundred stationary locomotives, smokeless and passive.⁹

They were not long stationary. Soon they would be moving, filled with hundreds of thousands of young men making their way, at ten or twenty miles an hour and often with lengthy, unexplained waits, to the detraining points just behind the frontiers. Long prepared, many of the frontier stations were sleepy village halts, where platforms three-quarters of a mile long had not justified the trickle of peacetime comings and goings. Images of those journeys are among the strongest to come down to us from the first two weeks of August 1914: the chalk scrawls on the waggon sides – ‘Ausflug nach Paris’, and ‘à Berlin’ – the eager young faces above the open collars of unworn uniforms, khaki, field-grey, pike-grey, olive-green, dark blue, crowding the windows. The faces glow in the bright sun of the harvest month and there are smiles, uplifted hands, the grimace of unheard shouts, the intangible mood of holiday, release from routine. Departure had everywhere been holidaylike, with wives and sweethearts, hobble-skirted, high-waisted, marching down the road to the terminus arm-in-arm with the men in the outside ranks. The Germans marched to war with flowers in the muzzles of their rifles or stuck between the top buttons of their tunics; the French marched in close-pressed ranks, bowed under the weight of enormous packs, forcing a passage between crowds over-spilling the pavements. One photograph of Paris that first week of August catches a sergeant marching backwards before his section as they lean towards him, he like a conductor orchestrating the rhythm of their footfalls on the cobbles, they urgent with the effort of departure and the call to arms.¹⁰ An unseen band seems to be playing ‘Sambre-et-Meuse’ or ‘le chant du départ’. Russian soldiers paraded before their regimental icons for a blessing by the chaplain, Austrians to shouts of loyalty to Franz Joseph, symbol of unity among the dozen nationalities of his creaking empire. In whichever country, mobilisation entailed enormous upheaval, the translation of civil society into the nation in arms. The British army, all-regular as it was, stood the readiest for war; once its reservists were recalled, it was prepared to deploy. ‘We found the barracks full of Reservists – many still in civilian dress – and more were flocking in by almost every train’, wrote Bandsman H.V. Sawyer of the 1st Rifle Brigade at Colchester on 5 August. ‘Fitting them out with uniform, boots and equipment was proceeding rapidly but in some cases was no easy job. I remember one man in particular who must have weighed eighteen stone . . . It was hard on the Reservists, leaving good jobs and comfortable homes to come back to coarse uniforms and heavy boots.’¹¹

Bandsman Shaw packed his peacetime kit and sent it home by rail. ‘As it turned out, I needn’t have bothered. But I wasn’t to know that I’d packed that lovely dark green review order tunic for the last time in my life.’¹² In Paris Lieutenant Edward Spears, 11th Hussars, on exchange from the British to the French army, changed into khaki. ‘“How funny you look, disguised as a dusty canary”, observed the female concierge who let me in at one of the more obscure entrances to the Ministère de la guerre. This was disappointing, but one became used to the fact that for a long time the French thought that to go to war in a collar and tie [British officers wore an open-necked tunic in service uniform] represented an attitude of levity quite out of keeping with the seriousness of the situation.’¹³ The British, as a result of the Boer War, had decided on a sartorial revolution the French had not been able to make. Despite much experiment and debate, it went garbed for war in 1914 much as it had done in 1870, almost as under Napoleon. The heavy cavalry wore brass helmets with a long horsehair plume, the light cavalry frogged jackets and scarlet trousers; some of the heavy cavalry were burdened with breastplates unchanged in pattern from Waterloo. The light cavalry of the Armée d’Afrique were dressed in sky-blue tunics, the Spahis in flowing red cloaks, the Zouaves in baggy red breeches and Turkish waistcoats. Most conspicuous of all, because of their numbers, were the infantry of the metropolitan army. Under long, turned-back blue greatcoats, their legs were encased in madder-red trousers tucked into calf-length boots.¹⁴ All was made of heavy wool; the stifling weight of antique uniforms was to prove one of the additional ordeals of combat in the sun-drenched autumn of 1914.

The Austrian cavalry rode to war in uniforms as antiquated as the French; only the infantry had been re-equipped with service grey. The Russians were unexpectedly modern. Their service dress was a loose olive-green overshirt, the gymnastirka modelled on an athlete’s tunic; but there were exotic exceptions, notably the Astrakhan caps of the light cavalry. Only the Germans had made as clean a sweep as the British. Their army was uniformly field-grey. With an antiquarian deference to tradition, however, each branch of the service was outfitted in a camouflage version of parade-ground finery. Uhlans wore double-breasted lancer tunics and hussars field-grey frogging, while cuirassiers, dragoons and infantrymen kept their spiked helmets, disguised with field-grey covers. Little patches of colour and braid and lace distinguished regiment from regiment in almost all armies; the Austrians meticulously differentiated between ten shades of red, including madder, cherry, rose, amaranth, carmine, lobster, scarlet and wine, for collar patches, six shades of green and three of yellow. The Hungarian regiments of Franz Joseph’s army wore braided knots on their trousers and the Bosnian-Herzogovinian infantry the red fez and baggy breeches of the Balkans. Even the British, whom Captain Walter Bloem would describe on first encounter as wearing ‘a grey-brown golfing suit’,¹⁵ excepted Lowlanders and Highlanders from the uniformity of khaki. They preserved their tartan trews or sporrans and kilts.

However clothed, the infantrymen of every army were afflicted by the enormous weight of their equipment: a rifle weighing ten pounds, bayonet, entrenching tool, ammunition pouches holding a hundred rounds or more, water bottle, large pack containing spare socks and shirt, haversack with iron rations and field dressing; that was a common outfit. The British, after the experience of long marches across the veldt in the Boer War, had adopted the ‘scientific’ Slade-Wallace equipment of canvas webbing, designed to distribute weight as evenly as possible over the body; even so, it dragged on the shoulders and waist. The Germans clung to leather, with the greatcoat hooped outside a stiff back-pack of undressed, and so water-repellant, hide. The French piled everything into a mountainous pyramid, ‘le chargement de campagne’, crowned with the individual’s metal cooking pot; gleams of sunlight from such pots would allow young Lieutenant Rommel to identify and kill French soldiers in high standing corn on the French frontier later that August.¹⁶ The Russians rolled their possessions, greatcoat and all, into a sausage slung over one shoulder and under the other arm. However arranged, no infantryman’s marching load weighed less than sixty pounds; and it had to be plodded forward, mile after mile for an expected twenty miles a day, in stiff, clumsy, nailed boots – ‘dice-boxes’, brodequins, Bluchers, to the Germans, French and British – which were agony until broken to the shape of the foot.

Feet were as important as trains in August 1914, horses’ feet as well as men’s feet for, after detrainment in the concentration area, cavalry and infantry deployed on to the line of march. That, for the Germans, presaged days of marching west and southwards, days in which human feet would bleed and horses throw shoes. The telltale clink of a loose nail warned a cavalryman that he must find the shoeing-smith if he were to keep up next day with the column; the same sound to the senior driver of a gun-team threatened the mobility of his six harnessed animals. There were 5,000 horses in an infantry division in 1914, more than 5,000 in a cavalry division. All had to be kept shod and healthy if the twenty miles of the day were to be covered to timetable, the infantry fed, reconnaissance reports returned, small-arms combat covered by artillery fire should the enemy be encountered. Fourteen miles of road was filled by an infantry division on the march and the endurance of horses – those pulling the wheeled field-kitchens, cooking on the march, quite as much as those drawing the ammunition waggons of the artillery brigades – counted with that of the infantry in the race to drive the advance forward.¹⁷

The race was tripartite. For the French it was north-eastward from their detraining points at Sedan, Montmédy, Toul, Nancy and Belfort behind the 1870 frontier. For the British Expeditionary Force, which began to disembark at Boulogne on 14 August, it was south-eastward towards Le Câteau, just before the Belgian border. These were short marches. For the Germans the marches planned were long, westward first and then southward towards Châlons, Eperney, Compiègne, Abbeville and Paris. General von Kluck’s First Army on the right faced a march of 200 miles from its detraining points at Aachen to the French capital.

Before Paris, however, there was Liège and Namur and the other fortresses of the Belgian rivers which impeded any easy crossing for a German army into France. Belgium, small but rich out of proportion to its size, its wealth the product of an early industrial revolution and the colonisation of the Congo, had invested heavily in fortification to protect its neutrality. The forts at Liège and Namur, guarding the crossings of the Meuse, were the most modern in Europe. Built between 1888 and 1892 to the design of General Henri Brialmont, they were constructed to resist attack by the heaviest gun then existing, the 210 mm (8.4 inch). Each consisted of a circle, twenty-five miles in circumference, of independent forts, arranged at sufficient distance to protect the city itself from attack and to lend each other the protection of their own guns. At Liège there were 400, of 6-inch calibre or less, disposed in the twelve forts of the complex, all protected by reinforced concrete and armour plate. The garrison of 40,000 provided the gun crews but also ‘interval troops’ who were supposed, at the threat of invasion, to dig trenches between the forts and hold at bay enemy infantry attempting to infiltrate through the gaps.

The strength of the Belgian forts had alarmed Schlieffen and his General Staff successors. They were, indeed, immensely strong, subterranean and self-contained, surrounded by a ditch thirty feet deep. Infantry assault upon them was certain to fail. Their thick skins would have to be broken by aimed artillery fire, and quickly, for a delay at the Meuse crossings would throw into jeopardy the smooth evolution of the Schlieffen Plan. No gun heavy enough for the work existed at the time of Schlieffen’s retirement in 1906. By 1909, however, Krupp had produced a prototype of a 420 mm (16.8 inch) howitzer powerful enough to penetrate the Belgian concrete. The Austrian Skoda company was meanwhile working on a 305 mm (12.2 inch) model which was ready the following year. It had the advantage of being road-transportable, when broken down into barrel, carriage and mount, on three motor-drawn waggons. The Krupp howitzer, in its original form, had to be transported by rail and embedded for action at the end of a specially built spur track in a concrete platform. Until a road-transportable model could be perfected, Austria lent Germany several of its 305s; only five of the Krupp rail and two of the new road-transportable guns had been finished by August 1914.¹⁸

Yet Liège had to be taken. Such was the necessity, and such the urgency, that the German war plan provided for the detachment of a special task force from Second Army to complete the mission. Commanded by General Otto von Emmich, its start line was drawn between Aachen and Eupen, at the north of the narrow corridor of Belgian territory lying between Holland and Luxembourg: Luxembourg, though independent and neutral, was to be overrun in the great German advance a few days after Emmich’s task force struck. The time allotted for the mission was forty-eight hours. It was expected by the Germans that Belgium would either not resist an invasion of its neutral territory or, should it do so, that its resistance would be swiftly overcome.

Both expectations were to be proved wrong. One of the clauses of the oath sworn by the Belgian sovereign on accession to the throne charged him with defence of the national territory, while Article 68 of the constitution appointed him commander-in-chief in time of war; he was also constitutionally president of his own council of ministers and therefore head of government, with executive powers unusual in a democracy. Albert I, King of the Belgians, was a man to take his responsibilities to heart. Intellectual, strong-willed, high-minded, he led an exemplary private life and set an example of fine public leadership. He was aware that his uncle, the aged Leopold II, had been bullied by the Kaiser in 1904: ‘You will be obliged to choose. You will be with us or against us.’ He had experienced the same treatment himself at Potsdam in 1913, when his military attaché had been warned that war would be ‘inevitable and soon’ and that it was ‘imperative for the weak to side with the strong’.¹⁹ Albert was determined not to take sides, correctly interpreting the treaty of 1839 to mean that Belgium’s right to neutrality was balanced by the requirement to avoid commitment to any foreign power.²⁰ It was for that reason his government had so peremptorily rejected a British offer of 1912 to lend assistance in the event of a German invasion; to have accepted it would have been to prejudice Belgium’s enjoyment of the international guarantees of its independence.

The British proposition, and the knowledge that only diplomatic delicacy deterred France from duplicating it, had the effect, however, of compelling the Belgian staff to confront the realities of national defence. Any intervention by the British or French, though necessitating resistance, would be benevolent. It would not threaten Belgian independence in either the long or even short term. A German intervention, by contrast, would have as its object not only the pre-emption of Belgian territory for a wider aggression but quite possibly the requisition of Belgian resources for the German war effort, and the subjection of Belgium to German military government for the duration of hostilities. From 1911 onwards, therefore,

Belgium’s political and military leadership had undertaken a major re-evaluation of Belgian policy. Three questions in particular worried Brussels: how to devise a military strategy that would limit the destruction of Belgium, how to ensure that a guarantor nation did not force Belgium into a war against its will, and how to ensure that a protesting power, once invited, would leave. Slowly, over a period of months and after much debate, the answers emerged. Militarily the Belgian General Staff planned to oppose any violation of Belgium; at the same time they hoped to confine all the fighting to a small area, possibly to the province of Belgian Luxembourg. Simply stated, Belgium would resist, yet seek to avoid losing either its integrity or its neutrality.²¹

Easier said than done. Belgium had adopted the principle of compulsory military service only in 1912, following the strategic review, and it had taken little effect by 1914. The army was one of the most old-fashioned in Europe. The cavalry still wore early nineteenth-century uniforms, crimson trousers, fur busbies, Polish lancer caps. The infantry were in dark blue with oilskin-covered shakos, feathered bonnets or grenadier bearskins. The few machine guns were drawn, like the Flemish milk carts much photographed by tourists, behind teams of dogs. Most of the artillery was allotted to the fortresses of Liège and Namur and the older defences of Antwerp. The army was actually outnumbered by the Garde Civique, the top-hatted town militias which descended from the days of the Thirty Years’ War. Belgium’s soldiers were patriotic and to prove themselves notably brave, but their capacity to confine any fighting for possession of their country to its eastern corner was delusory.

Yet, at the outset, they made a bold stab at enacting the General Staff’s strategy. The German ultimatum, fictively alleging a French intention to violate Belgian territory and asserting Germany’s right to do so in anticipation, was delivered, with a twelve-hour time limit, on the evening of Sunday 2 August. King Albert, acting as president of a council of state, considered it two hours later. The meeting lasted into the early hours of the morning. There were divided counsels. The Chief of Staff, General Antonin de Selliers, confessed the weakness of the army and advocated retreat to the River Velpe, outside Brussels. The Sub-Chief, General de Ryckel, demanded a spoiling attack into Germany: ‘Send them back where they belong.’ This fantasy was rejected. So, too, was Sellier’s defeatism. The King was most concerned that no appeal should be made to France or Britain, whose aid was assured, unless they reasserted their respect for the country’s independence. Eventually a middle way was decided. Belgium would not appeal for French or British assistance until her territory was physically violated, but the German ultimatum would meanwhile be rejected. The reply, described by Albertini as ‘the noblest document produced by the whole crisis’, ended with the resolution ‘to repel every infringement of [Belgium’s] rights by all the means at its power’.²²

It was delivered to the German Legation at seven o’clock on the morning of 3 August and received in Berlin shortly after noon. The Germans contrived to believe, nevertheless, that the Belgians would make no more than a show of force, sufficient to demonstrate their neutrality, before giving them passage. Later that evening the Kaiser sent a personal appeal to Albert – a member of the House of Hohenzollern-Sigmaringen and so a distant relative – restating his ‘friendliest intentions’ and claiming ‘the compulsion of the hour’ as justification for the invasion that was about to begin.²³On its receipt, the Belgian King gave way to his first outburst in two nerve-racking days: ‘What does he take me for?’ He immediately gave orders for the destruction of the bridges over the Meuse at Liège and the railway bridges and tunnels at the Luxembourg border.²⁴ He also charged the commander of the Liège fortress, General Gérard Leman, ‘to hold to the end with your division the position which you have been entrusted to defend’.

Leman, the King’s former military tutor, was a long-service professional soldier in the nineteenth-century tradition. Thirty years of his life had been spent at the Belgian War College. He was also a man of honour and, despite his advanced age, of courage and an unyielding sense of duty. The Meuse, which he was entrusted to hold, is a mighty river. ‘Sambre-et-Meuse’ is a traditional marching song of the French army, for the two rivers form a barrier which the revolutionary armies had defended against their enemies in 1792. At Liège the river runs in a narrow gorge 450 feet deep. It cannot be crossed in the face of a determined defence. So Emmich was to discover. His command entered Belgium early on the morning of 4 August, the outriders distributing leaflets disclaiming aggressive intent. Soon they came under fire from Belgian cavalrymen and cyclist troops who showed a quite unexpected resolution to oppose their advance. Pressing on to Liège, they found the bridges above and below the city already blown, despite the warning given that demolitions would be regarded as ‘hostile acts’. The Germans responded as threatened. Memories of ‘free firing’ by irregulars against the Prussian advance into France in 1870 were strong and had been re-enforced by official stricture. Despite the heroic place allotted to the Freischütze who had waged the War of Prussian Liberation against Napoleon in 1813–14, official Germany interpreted international law to mean that an effective occupying force had the right to treat civilian resistance as rebellion and punish resisters by summary execution and collective reprisal.²⁵ There were, later enquiries would reveal, few or no franc-tireurs in Belgium in 1914. It was an unmilitary nation, prepared for war neither in mind nor body; the loyal government, though determined on a legal defence with the inadequate means it possessed, showed itself anxious from the start to deter citizens from useless and dangerous opposition to the German invasion. It issued placards urging avoidance ‘of any pretext for measures of repression resulting in bloodshed or pillage or massacre of the innocent population’.²⁶ The government also advised civilians to lodge firearms with the authorities; in some places the Civic Guard took the warning so seriously that it deposited its government weapons at the local town hall.²⁷

Non-resistance did nothing to placate the invaders. Almost from the first hours, innocent civilians were shot and villages burnt, outrages all hotly denied by the Germans as soon as the news – subsequently well attested – reached neutral newspapers. Priests were shot, too, perhaps because German officers remembered that it was the priests who had led the resistance of Catholic Brittany against the armies of the French Revolution in 1793. The ‘rape of Belgium’ served no military purpose whatsoever and did Germany untold harm, particularly in the United States, where the reputations of the Kaiser and his government were blackened from the outset by reports of massacre and cultural despoliation. The reputation of the German army was dishonoured also. On 4 August, the first day of the Emmich incursion against the Meuse forts, six hostages were shot at Warsage and the village of Battice burnt to the ground. ‘Our advance in Belgium is certainly brutal’, Moltke wrote on 5 August, ‘but we are fighting for our lives and all who get in the way must take the consequences.’²⁸ The consequences were to get worse. Within the first three weeks, there would be large-scale massacres of civilians in small Belgian towns, at Andenne, Seilles, Tamines and Dinant. At Andenne there were 211 dead, at Tamines 384, at Dinant 612. The victims included children and women as well as men and the killing was systematic; at Tamines the hostages were massed in the square, shot down by execution squads and survivors bayoneted. The execution squads were not, as were the ‘action groups’ of Hitler’s holocaust, specially recruited killers but ordinary German soldiers. Indeed, those who murdered at Andenne were the reservists of the most distinguished regiments of the Prussian army, the Garde-Regimenter zu Fuss.²⁹

Worst of all the outrages began on 25 August at Louvain. This little university town, the ‘Oxford of Belgium’, was a treasure store of Flemish Gothic and Renaissance architecture, painting, manuscripts and books. Panicked allegedly by a misunderstood night-time movement of their own troops, the occupiers, 10,000 strong, began to shout ‘snipers’, and then to set fire to the streets and buildings where franc-tireurs suspectedly operated. At the end of three days of incendiarism and looting, the library of 230,000 books had been burnt out, 1,100 other buildings destroyed, 209 civilians killed and the population of 42,000 forcibly evacuated.³⁰ The worldwide condemnation of Germany’s war against ‘culture’ bit deep in the homeland. There academics and intellectuals were in the vanguard of the appeal to patriotism, representing the war as an attack by barbarians, philistines and decadents – Russians, British and French respectively – on high German civilisation. On 11 August, Professor von Harnack, director of the Royal Library in Berlin, had warned that ‘Mongolian Muscovite civilisation could not endure the light of the eighteenth century, still less of the nineteenth century, and now in the twentieth century, it breaks loose and threatens us.’³¹ ‘Light’ was a cherished idea to the Germans. The eighteenth-century Enlightenment of Lessing, Kant and Goethe – who had called for ‘more light’ on his deathbed – had been Germany’s passport into Europe’s life of the mind. Enlightenment had been the inspiration of Germany’s enormous contributions to philosophical, classical and historical scholarship during the nineteenth century. For Germans to be found out as book-burners cut educated Germany to the quick. Even harder to bear were the expressions of disgust from the world’s great centres of learning and research; American as well as European universities denounced the atrocity and committees were formed in twenty-five countries to collect money and books for the restoration of the Louvain library.³² Germany’s scholars and writers responded by a ‘Call to the World of Culture’, signed by such pre-eminent scientists as Max Planck and Wilhelm Röntgen, which ‘endorsed the franc-tireur hypothesis and the right to reprisal, and claimed that if it had not been for German soldiers, German culture would long have been swept away’.³³

The call fell on deaf ears. The damage had been done. It had been done, ironically, by latecomers to the invasion, the 17th and 18th Reserve Divisions, which had been retained for three weeks in their home district of Schleswig-Holstein to guard against the supposed danger of amphibious attacks by the British on the North Sea coast.³⁴ Far from the scene of action, the divisions imbibed to the full the newspaper propaganda about franc-tireurs, as well as the objective reports of the Belgian army’s wholly unexpected tenacity in defence of the Meuse forts. It is difficult to estimate with hindsight which more enraged the Germans. Perhaps the latter: the myth of franc-tireurs in rooftops and hedgerows had the force of alarming rumour; the fact of real Belgian resistance not only exploded the fictive belief in Belgian passivity but threatened the smooth unrolling of the German advance in the west at its most critical point.

Emmich’s task force, composed of the 11th, 14th, 24th, 28th, 38th and 43rd Brigades, specially detached from their parent divisions, together with the 2nd, 4th and 9th Cavalry Divisions and five élite Jäger (light infantry) battalions, all drawn from the peacetime army but reinforced for the operation, crossed the Belgian frontier on 4 August. It headed straight for Liège, twenty miles to the west, along the line of what today is the Aachen-Brussels international motorway. With them the units of the task force brought two batteries of 210 mm (8.4 inch) howitzers, the heaviest available until the Austrian and Krupp monsters could be got forward. On the morning of 5 August Captain Brinckman, recently the German military attaché in Brussels, appeared in Liège to demand Leman’s surrender.³⁵ He was sent packing. The German bombardment on the eastern forts opened shortly afterwards. When the infantry and cavalry attempted to advance, however, they found the way barred. Because of blown bridges, the 34th Brigade had to be ferried across the Meuse in pontoons. The garrisons of the forts returned fire steadily, while the ‘interval troops’ of the 3rd Division, manning the hastily dug entrenchments, fought manfully whenever the German advance guards tried to penetrate the line. Throughout the night of 5/6 August German casualties mounted steadily. They were particularly heavy at Fort Barchon, where the attackers ‘came on, line after line, almost shoulder to shoulder, until as we shot them down the fallen were heaped on top of each other in an awful barricade of dead and wounded’.³⁶ There was, in the confused and bitter fighting of the night, a ghostly foretaste of what would ensue at places not yet touched by the war, at Vimy, Verdun and Thiepval.

Yet there was also opportunity for success through leadership that the barbed wire and continuous trench lines of the Western Front would deny. Early in the morning of 6 August, General Erich Ludendorff, the liaison officer between Second Army and Emmich’s command, rode forward into the confusion to find that the commander of the 14th Brigade had been killed. Instantly assuming the vacancy, and ordering up a field howitzer to provide firepower at the point of assault, Ludendorff fought his new command through the straggling village of Queue-de-Bois to a high point from which he could look down, across the Meuse and the two unblown city bridges, into Liège itself. Unknown both to the Belgians and to the German high command, with which Ludendorff had lost touch, a force of 6,000 Germans had penetrated to the interior of Leman’s circle of defences. From his vantage point, Ludendorff ordered forward a party under a flag of truce to demand Leman’s surrender, which was again refused; a raiding force that followed was shot down at the door of Leman’s headquarters.³⁷ Ludendorff’s bold sally nevertheless prompted Leman to leave the city and take refuge in Fort Loncin on the west side of the outer ring. Leman also decided to send the infantry, the 3rd Division and its supporting 15th Brigade, back to join the field army on the River Gette outside Brussels, believing that they would be overwhelmed in a battle with what he calculated were five German corps. There he miscalculated. The German brigades merely represented the five different corps to which they belonged. In the long run, however, his decision was justified, for it spared one-sixth of the Belgian army to fight in the defence of Antwerp, which King Albert had already chosen to make his strongpoint in Belgium’s last stand.

A moment of equilibrium ensued. Ludendorff was inside the ring, but without sufficient force to compel a surrender. Most of Emmich’s command was outside the ring. Leman was determined to continue resistance as long as the forts remained intact, as all still did. The French government, to which Albert appealed for help, promised only to send Sordet’s cavalry corps and then just to reconnoitre. The British, who had been expected to deploy their Expeditionary Force of six divisions into Belgium, now decided to retain two at home. Joffre refused to extend the mass of his army northwards, since to do so would detract from his planned offensive towards the Rhine; he actually wanted Albert to bring the Belgian army down from Brussels, away from Antwerp, to join his left wing. The situation map showed a French army aligned towards Lorraine, a German army whose weight had not yet crossed either the Belgian or French frontier, a British army still mobilising to embark, a Belgian army concentrated in the centre of its homeland and, at Liège, a small German striking force immobilised by a handful of Belgian fortress troops guarding the crossings on the possession of which the future of military events in the west turned.

The equilibrium was upset by Ludendorff. Large in physique and personality, utterly devoid of moral or physical fear, indifferent to the good opinion even of superiors, dislikeable, insensitive – he was to suffer the death of two stepsons during the coming war without faltering in his exercise of high command – Ludendorff resolved on the morning of 7 August to launch the 14th Brigade into the centre of Liège and take the chance that he would be opposed. He was not. Driving up to the gates of the old citadel, he hammered on the door with the pommel of his sword and was admitted.³⁸ The surrender of the garrison gave him possession of the city. His bold sortie had put the bridges into his hands. He decided to return post-haste to Aachen and urge forward Second Army to complete his success.

While he was away Emmich’s task force broke the resistance of Forts Barchon and Evegnée, though more by luck than deliberate reduction. That would wait upon the appearance of the monster howitzers which General von Bülow, at Ludendorff’s insistence, despatched on 10 August.³⁹ The first road-transportable Krupp 420, diverted by demolitions, eventually arrived within range of Fort Pontisse on 12 August. After it was emplaced, the bombardment began. The crew, wearing head-padding, lay prone 300 yards away while the gun was fired electrically. ‘Sixty seconds ticked by – the time needed for the shell to traverse its 4,000 metre trajectory – and everyone listened in to the telephone report of our battery commander, who had his observation post 1,500 metres from the bombarded fort, and could watch at close range the column of smoke, earth and fire that climbed to the heavens.’⁴⁰ The first of the shells, delay-fused to explode only after penetration of the fort’s protective skin, fell short. Six minutes later, the next was fired and then five more, each ‘walked up’ towards the target as the elevation was corrected. The relentless approaching footfall of the detonations spoke to the paralysed defenders of the devastation to come. The eighth struck home. Then the gun fell silent for the night but next morning, joined by the other which had completed the journey from Essen, the bombardment reopened. The range had been found and soon the 2,000-pound shells were ‘stripping away armour plate and blocks of concrete, cracking arches and poisoning the air with heavy brown fumes’.⁴¹ By 12.30 Fort Pontisse was a wreck, its garrison physically incapacitated, and it surrendered. Fire then shifted to Fort Embourg, which surrendered at 17.30; Fort Chaudfontaine had been destroyed by the explosion of its magazine at nine o’clock. On 14 August it was the turn of Fort Liers, 09.40 hours, and Fléron, 09.45 hours. Finally, on 15 August, the howitzers, one of which was by now emplaced in the main square of Liège, reduced Forts Boncelle, 07.30 hours, and Lautin, 12.30 hours, before turning their fire on to Fort Loncin, to which General Leman had shifted his headquarters nine days earlier. After 140 minutes of bombardment the magazine was penetrated and the fortress destroyed in the resulting explosion.

The German pioneer troops who advanced to take possession found ‘a miniature Alpine landscape with débris strewn about like pebbles in a mountain stream . . . Heavy artillery and ammunition had been thrown everywhere; a cupola had been blown from its place . . . and had fallen on its dome; it now looked like a monstrous tortoise, lying on its shell.’ Amid the ruins General Leman was found lying insensible. To Emmich, whom he had met on manoeuvres some years previously, he said from the stretcher on which his captors placed him, ‘I ask you to bear witness that you found me unconscious.’⁴²

The last two forts, Hollogne and Flémelle, surrendered without further fight on 16 August and the Krupp and Skoda guns were then broken out of their emplacements and diverted towards the forts of Namur, where they would arrive on 21 August and repeat the victory of Liège after three days of bombardment on 24 August. These two ‘naval battles on land’, in which guns heavier than those mounted by any Dreadnought had cracked armoured targets incapable of manoeuvre, spelt the end of a three-hundred-year-old military trust in the power of fortress to oppose the advance of a hostile army without the active intervention of supporting mobile troops. That trust had never been more than conditional in any case. The Prince de Ligne, one of the leading generals of the eighteenth-century fortress age, had written, ‘The more I see and the more I read, the more I am convinced that the best fortress is an army, and the best rampart a rampart of men.’⁴³ Forts – at Maubeuge, at Przemysl, at Lemberg, at Verdun – would form the focus of intense fighting in 1914, 1915 and 1916 – but only as fixed points of encounter around which decisive battle would be waged by fluid masses and mobile weapons. Ramparts of men, not steel or concrete, would indeed form the fronts of the First World War.

Just such a rampart was in the making far to the south of the Meuse crossings even while Emmich’s task force was battering Liège and Namur into fragments. If the Emmich element in the German plan was bold, the French plan for the opening of the war was bolder in a different dimension, nothing less than a headlong offensive across the 1871 frontier into annexed Alsace-Lorraine. ‘Whatever the circumstances’, Plan XVII stated, ‘it is the Commander-in-Chief’s intention to advance with all forces united to attack the German armies.’⁴⁴ Those the French expected to find, as in 1870, deployed along the common frontier between Luxembourg and Switzerland. Joffre’s scheme of operations was to throw forward his five armies in two groups, Fifth and Third on the left, Second and First on the right, with Fourth echeloned slightly in rear to cover the gap between the two masses, into which topography and fortification, the French calculated, would funnel any successful German advance.

Had the Germans not long committed themselves to an entirely different plan that made the French dispositions both irrelevant and dangerous, Plan XVII was not ill-conceived. It was well adapted to the military geography, natural and man-made, of eastern France. Germany’s annexations of 1871 had robbed France of long lengths of her ‘natural’ frontier, including the Rhine between Strasbourg and Mulhouse. They nevertheless left strong positions in French hands, including the high ground of Côtes de Meuse between Verdun and Toul and, further south, the crests of the Vosges mountains above Nancy and Epinal.⁴⁵ The unfortified opening between, known as the Trouée de Charmes, was the trap into which the French hoped to tempt the Germans. The buttresses to left and right – Meuse heights, Vosges mountains – provided in any case firm points of departure, well furnished with road and rail heads and strongly fortified, for the two groups of armies to begin their descent into the Moselle and Rhine valleys. The two thrusts, by the Fifth and Third and the Second and First Armies respectively, were the essence of Plan XVII.

Before either could be set in motion, however, Joffre had unleashed a preliminary assault, designed, as was Emmich’s into Belgium, to open the way for the larger offensive to follow. On 7 August General Bonneau’s VII Corps, based at Besançon, moved forward to seize Mulhouse in Alsace and, it was hoped, raise the countryside against the Germans. Bonneau expressed reluctance and showed it in practice. He took two days to cover the fifteen miles to Mulhouse and allowed himself to be driven out within twenty-four hours when the Germans counter-attacked. Worse, he then beat a retreat to Belfort on the Swiss frontier, the only fortress to have sustained resistance to the Germans throughout the Franco-Prussian war. The humiliation, actual and symbolic, incensed Joffre. He dismissed both Bonneau and Aubier, commander of the accompanying 8th Cavalry Division, on the spot. It was a warning of a greater purge to come. Joffre was a sacker. He had removed two obviously incompetent generals after the 1913 manoeuvres and already seven divisional commanders who had shown themselves torpid or unfit in the period of mobilisation andcouverture.⁴⁶ By the end of August he would have dismissed an army commander, three out of twenty-one corps commanders and thirty-one out of 103 divisional commanders. In September he was to dismiss another thirty-eight divisional commanders, in October eleven and in November twelve.⁴⁷ Others were to be transferred, from active to territorial divisions, or demoted. In some divisions generals were given only a month to show their paces, sometimes less. The inappropriately named Generals Superbie and Bataille lasted respectively five weeks and ten days at the head of the 41st Division. Bolgert, who succeeded Bataille, lasted nine days before demotion to a reserve division, and must have thought himself lucky not to disappear altogether. The majority did. Only seven of the forty-eight commanders of peacetime infantry divisions were still en poste in January 1915. One, Raffenet, of the 3rd Colonial, had been killed, another, Boë of the 20th, had been severely wounded. A few, Deligny, Hache, Humbert, had been advanced to command corps; so, too, had Pétain, who started the war as a mere brigadier. The rest had gone for good. ‘My mind was made up on this subject’, Joffre would write later. ‘I would get rid of incapable generals and replace them with those who were younger and more energetic.’ Right was on his side. French generals were too old – in 1903 their average age had been sixty-one against fifty-four in Germany – or, if younger, often unfit.⁴⁸ Joffre, admittedly, set no example. Heavily overweight, he was devoted to the table and allowed nothing, even at the height of the crisis in 1914, to interrupt lunch. He was, for all that, shrewd, imperturbable and a keen judge of character, the qualities that would see the French army through the coming campaign as the crisis deepened.

THE BATTLE OF THE FRONTIERS

A curious interval of calm had followed the upheaval of mobilisation and the subsequent mass migration to the areas of concentration. Both French and German divisional histories record an interlude of a week or even ten days between detraining behind the frontier and the onset of action. It was spent in distributing stores, hurried exercises and deployment on foot towards the front. There was, for some very senior officers on both sides and for others who had read their history, a certain familiarity about the preliminary events. They resembled those of the first days of the Franco-Prussian War forty-four years earlier, with the difference that everything was working with greater efficiency. Otherwise, the troop trains looked the same, the long columns of horse, foot and guns looked the same, on the French side the uniform looked the same, on both sides even the weapons looked the same; the revolutionary power of quick-firing artillery and magazine-rifles had yet to reveal itself.

The battlefront chosen by the French high command was, for much of its length, almost exactly the same also. True, in 1870, there had been no operations north of the point where the French met the Luxembourg frontier, while in 1914 the deployment areas of the French Third, Fourth and Fifth Armies reached from there towards Belgium. In Lorraine, however, the soldiers of the First Army found themselves treading the same roads as their grandfathers had done under the command of Napoleon III. The lines of departure were further to the west, transposed thence by the German seizure of territory that had been the price of defeat in 1871, but the avenues of advance were the same and so were the objectives: the line of the River Saar, Saarbrücken and the country beyond on the way to the Rhine. These had been given in Joffre’s General Instruction No. 1 of 8 August.⁴⁹

The Lorraine offensive opened on 14 August, when Dubail’s First Army, with de Castelnau’s Second echeloned to its left, crossed the frontier and advanced towards Sarrebourg. Bonneau’s setback at Mulhouse seemed forgotten. The French advanced as liberators and conquerors, bands playing, colours unfurled. The thought that the Germans might have plans of their own for victory in the lost provinces – to them ‘Reich territory’ – appears to have crossed no mind in the French high command. Its intelligence underestimated the Germans’ strength and its judgement was that they would stand on the defensive. In fact the German Sixth and Seventh Armies, commanded by Crown Prince Rupprecht of Bavaria and General Josias von Heeringen, a Prussian ex-War Minister, comprised eight, not six, corps and were preparing to strike the French a weighty counterblow as soon as they overreached themselves.

They were shortly to do so. For four days the Germans fell back, contesting but not firmly opposing the French advance, which in places reached twenty-five miles into Reich territory. A German regimental colour was captured and sent for presentation to Joffre at Vitry-le-François, where he had established General Headquarters (GQG). Château-Salins was taken, then Dieuze, finally on 18 August, Sarrebourg, all places that had been French since Louis XIV’s wars against the Habsburgs in the seventeenth century. Then the front lost its sponginess. The French infantry found German resistance stiffening. The small Army of Alsace, advancing continuously on the First’s right, recaptured Mulhouse next day, but its success lent no support, for a wide gap yawned between it and Dubail’s positions. It was not the only gap. First Army was not firmly in contact with Second; west of the Saar Valley, Dubail and Castelnau were not in operational touch at all. Dubail was conscious of the weakness and intended on 20 August to mend it by launching an attack that would both restore contact and open a way through for Conneau’s Cavalry Corps (2nd, 6th and 10th Divisions) to debouch into the enemy’s rear and roll up his flank; but even as he set the attack in motion on the night of 19/20 August, the Germans were preparing to unleash their planned counter-offensive.⁵⁰

Rupprecht’s and Heeringen’s Armies had been temporarily subordinated to a single staff, headed by General Krafft von Delmensingen. Thus, while the French Second and First Armies co-ordinated their actions only as well as sporadic telephoning could arrange, the German Sixth and Seventh fought as a single entity. Here was the anticipation of a new trend in command, which would bring into being formations as large as existing communication systems could control. On 20 August its worth was swiftly demonstrated. Dubail’s night attack was checked as soon as begun. The setback was followed by a simultaneous offensive along the whole line of battle by the eight German corps against the French six. The French VIII Corps, which had reached the Saar at Sarrebourg, was overwhelmed; its artillery was outmetalled by the heavier German guns, under the fire of which the German infantry drove the French from one position after another.

Heavy artillery did even worse damage to Second Army, which was struck by a concentrated bombardment along its whole front as day broke on 20 August. The XV and XVI Corps abandoned their positions under the infantry attacks that followed. Only the XX, on the extreme left, held firm. It was fighting on home ground and was commanded by General Ferdinand Foch, of exceptional talent and determination. While his soldiers clung on, the rest of the Army was ordered by Castelnau to break contact and retreat behind the River Meurthe, the line from which it had begun its advance six days earlier. It had very nearly been enveloped on both flanks, which would have resulted in irretrievable disaster to the whole French army, and had completely lost touch with the First Army, which Dubail was therefore obliged to disengage from battle also. By 23 August it, too, had returned to the Meurthe and was preparing to defend the river, hinging its defence on strong positions which Foch had established on the high ground of the Grand Couronné de Nancy. There the two armies entrenched to await further German assaults. Schlieffen had warned such assaults must not be attempted if the victory he had rightly anticipated would follow a French offensive in Lorraine. The temptation to exploit the victory proved, however, too strong to resist. Von Moltke yielded to the demands of Rupprecht and Delmensingen and sanctioned their renewal of the offensive which, between 25 August and 7 September, broke on the stout defences the French unexpectedly established along the Meurthe.⁵¹

The significance of the French recovery on the right of their enormous front would take time to emerge. Elsewhere disaster persisted. Next above the First and Second Armies stood the Third and Fourth, given by Joffre the mission of penetrating the forest zone of the Ardennes and striking towards the towns of Arlon and Neufchâteau in southern Belgium. Their front of attack was twenty-five miles, the depth of forest to be penetrated about eight. Two considerations argued against Joffre’s offensive instructions. The first was that the terrain of the Ardennes – tangled woods, steep hillsides, wet valleys – impedes military movements, confining marching troops to the infrequent roads. The second was that the German armies, Fourth, commanded by the Duke of Württemberg, and Fifth, commanded by the German Crown Prince, were deployed to attack to the east on a collision course with the approaching French, and in exactly equal strength, eight corps against eight. Of this equality Joffre’s headquarters were quite unaware. The main French reconnaissance force, Sordet’s Cavalry Corps, had criss-crossed the Ardennes between 6 and 15 August without detecting the enemy’s presence. The troopers had ridden bare their horses’ backs – French cavalry had the bad habit of not dismounting on the march – but seen neither hide nor hair of the enemy. As a result, GQG had assured both de Langle, Fourth Army, and Ruffey, Third Army, on 22 August, that ‘no serious opposition need be anticipated’.⁵² Reports from French aviators had confirmed this wholly false judgement throughout the previous week.⁵³

The Germans were better informed than the French. Their aviators had reported significant enemy movements on the front of Fourth Army and, though what had been observed was the northward march of elements of Lanrezac’s Fifth Army towards the Meuse, the mistaken interpretation alerted the Germans to Joffre’s real intentions.⁵⁴ On 20 August the Crown Prince’s army had remained in its positions while its heavy artillery had brought the French frontier fortresses of Montmédy and Longwy – both old and ill-defended – under bombardment, but on the morning of 22 August both it and Fourth Army were on the march.⁵⁵ Fourth Army was particularly concerned with the danger of being outflanked and its headquarters issued orders for the corps on its left to take particular care to maintain contact with its neighbour.⁵⁶

In fact, it was the French, not the Germans, who risked being unhinged. Their formations were disposed ‘en echelon’, like a flight of steps descending in a shallow easterly direction from north to south, so that the flank of each corps was exposed on its left. Were the Germans to push hard against the top of the French front, there was a danger that the steps of the French line would separate in sequence, leading to the wholesale collapse of Fourth and Third Armies. That, on 22 August, was exactly what happened. In practice, it was Third Army which collapsed first. Advancing at daybreak, its vanguard ran into unexpected German resistance and, when a sudden bombardment overwhelmed its supporting artillery, the infantry were panicked into flight. The rest of the Army, with a gap yawning in its centre, was stopped in its tracks and had to fight hard to hold its position. Fourth Army, thus unsupported to its south, also failed to advance, except in the centre, a position held by the Colonial Corps. This, the only truly regular element of the French army, was composed of white regiments which in peacetime garrisoned the empire in North and West Africa and Indo-China. Its soldiers were hardened and experienced veterans. That was to be their undoing. Pressing forward with a determination the unblooded conscripts of the metropolitan army could not match, it rapidly became embedded in a far larger mass of Germans. Five of its battalions, advancing one behind the other on a front only 600 yards wide, launched repeated bayonet attacks through dense woodland, only to be thrown back by concentrated rifle and machine-gun fire. The harder the Colonials pressed, the higher their casualties mounted. By the evening of 22 August, the 3rd Colonial Division had lost 11,000 men killed or wounded, out of a strength of 15,000, theworst casualties to be suffered by any French formation in the Battle of the Frontiers.⁵⁷ Its effective destruction spelt an end to Fourth Army’s efforts to take ground forward, just as V Corps’ collapse had halted Third Army’s offensive further to the south.

Plan XVII had thus been brought to a standstill along a crucial section of front, seventy-five miles wide, between Givet and Verdun. Joffre at first refused to credit the outcome. On the morning of 23 August he signalled de Langle de Cary to say that there were ‘only . . . three [enemy] corps before [you]. Consequently you must resume your offensive as soon as possible’.⁵⁸ De Langle de Cary obediently attempted to do as ordered, but his army was only driven further back that day. Unsuccessful, too, were the Third and the recently assembled Army of Lorraine. On 24 August, the Fourth Army retired behind the protection of the River Meuse and Third Army shortly followed. Much of Maunoury’s Army of Lorraine was meanwhile withdrawn to Amiens, where a new Army, the Sixth, was to be created around its complement of reserve divisions.

The Battle of the Sambre

On two sectors of the French frontier, Alsace-Lorraine and the Ardennes, the Germans had, by the end of the war’s third week, achieved significant victories. The scene of action was now to shift to the only sector as yet untouched by major operations, the frontier with Belgium. It was there that Germany’s offensive plan must succeed if Schlieffen’s dream of a six-week war were to be realised. The seizure of Liège had laid the ground. The consequent retreat of the Belgian field army to the entrenched camp at Antwerp had opened the way. The fall of Namur, clearly imminent by 24 August, would complete the clearing of the theatre of major obstacles. Most important of all, the French high command, despite the weight of warning given by the German invasion of eastern Belgium, remained apparently and obstinately blind to the danger that threatened. Lanrezac, commander of Fifth Army deployed at the northern end of the line, had begun to warn GQG, even before war was declared, that he feared an envelopment of his left – northern – flank by a German march into Belgium. Joffre, whose thoughts were fixed on his own offensive into Germany, dismissed these anxieties. As late as 14 August, when Lanrezac brought his concerns to GQG at Vitry-le-François on the Marne, east of Paris, and soon to lie within earshot of the guns, the Commander-in-Chief continued to insist that the Germans would not deploy any major force inside Belgium north of the Meuse.

Over the next six days, Joffre began to reconsider, issuing orders that first directed Lanrezac’s Fifth Army into the angle between the Meuse and the Sambre, as a precautionary measure, then that instructed Lanrezac to join with the British Expeditionary Force in operations against the left wing of the German battle line, whose appearance in great strength in Belgium could no longer be denied.⁵⁹ By that date the battle with von Kluck’s, von Bülow’s and von Hausen’s Armies – the battle of the Sambre to the French, Mons to the British – was already about to begin. It was in its opening stages what military theorists call a ‘battle of encounter’, the nature of which is decided by the actions of the troops engaged rather than by orders received from the top. Orders, indeed, discouraged engagement. Lanrezac, in a conference held at Chimay on the afternoon of 21 August, told the Chiefs of Staff of his subordinate corps that the plan was for Fifth Army to hold the high ground on the south bank of the Sambre.⁶⁰ He feared that if he committed his soldiers to hold the dense belt of little industrial buildings and cottages – le Borinage – that line the bank between Charleroi and Namur, they would become involved in small-scale street fighting and be lost to his control. The Germans received similar orders from von Bülow, who was co-ordinating the movements of First and Third Armies as well as of his own Second, though given for different reasons. On 20 August Moltke had warned Bülow that the French were present in strength in front of him and the British were to his right, but in unlocated positions, and that he should in consequence attack across the Sambre only when Second and Third Armies could co-ordinate a pincer movement. On the morning of 21 August, Bülow accordingly wirelessed von Hausen that he was postponing Second Army’s advance, which meant that Third was to pause also.

Events at a lower level then took charge. Rivers, unless wide, are always difficult to defend. Meanders create pockets that soak up troops and cause misunderstandings between neighbouring units as to where responsibilities start and end. Bridges are a particular problem: does a bridge which marks a boundary between units lie in one sector or another? Buildings and vegetation compound the problems, breaking lines of sight and impeding easy lateral movement along the river when local crises, requiring rapid reinforcement, arise. Long experience has taught soldiers that it is easier to defend a river on the far, rather than the near, bank but, if the near bank is to be defended, then it is better done behind it than at the water’s edge.⁶¹ All these truths were to be proved again in the battle that developed on the Sambre during 21 August.

Lanrezac, with perfect orthodoxy, had ordered the bridges to be held only by outposts, while the bulk of the Fifth Army waited on higher ground, whence it could advance to repel a German crossing or mount its own offensive across the bridges into Belgium. The outposts at the bridges, however, found themselves in a dilemma. At Auvelais, halfway between Namur and Charleroi, for example, they were overlooked from the far bank, and requested permission either to cross or to fall back. Their regimental commander, bound by Lanrezac’s instructions, refused but sent more troops to support them. The reinforcements discovered more bridges than their orders indicated had to be defended. While they were making their dispositions, German patrols of Second Army appeared opposite, sensed an opportunity and requested permission to chance a crossing from corps headquarters. It was that of the Imperial Guard, which, fortuitously, Ludendorff happened to be visiting when the message arrived. Showing the same initiative as he had done fifteen days earlier at Liège, he took personal responsibility for approving the venture. The 2nd Guard Division attacked, found an undefended bridge – there were eight in a sector where the French troops had thought there was but one – and established a foothold. To the west of Auvelais, at Tergné, a patrol of the German 19th Division found another unguarded bridge and crossed without asking for orders. Responding to opportunity, the divisional commander sent a whole regiment to follow and drove the French defenders away. By the afternoon of 21 August, therefore, two large meanders of the Sambre were in German hands and a gap four miles wide had been opened across the river front.

The results were characteristic of an encounter battle and greatly to the credit of the German front-line troops and their local commanders. Yet Lanrezac might still have retrieved the situation had he stuck to his original plan of holding the high ground south of the Sambre as his main position. Inexplicably, however, he now acquiesced in the decision of his two subordinates commanding III and X Corps to counter-attack, in an attempt to retake the meanders of the Sambre already lost. They tried and on the morning of 22 August their troops were repelled with heavy loss.

The French infantry made a gallant show, advancing across the Belgian beet fields with colours unfurled and bugles sounding the shrill notes of the ‘charge’. As the ranks drew near to the German lines . . . rifles and machine guns pounded forth a rapid-fire of death from behind walls and hummocks and the windows of houses. Before it the attack wilted. Running, stumbling, crawling, the French sought cover as best they could, and the attack ended leaving the German Guard undisputed masters of the field.⁶²

That night both corps had taken positions on Lanrezac’s original and preferred line on the high ground with nothing to show for the day’s brave effort but yet more casualties. They were very heavy. Of the regiments engaged, each beginning with a strength of some 2,500 men, the 24th had lost 800, the 25th, a Cherbourg regiment, 1,200, the 25th (Caen) 1,000, the 49th (Bordeaux) 700, the 74th (Rouen) 800, the 129th (le Havre) 650.⁶³ Strategically the result was even worse. Nine French divisions had been defeated by three German, and forced to retreat seven miles, contact with the Fourth Army, on the Meuse, had been broken, contact with the British Expeditionary Force at Mons had not been established and Sordet’s Cavalry Corps, which had wholly failed in its mission of finding the Germans before they fell on the French along the Sambre, was drawing back through Fifth Army’s positions, its men exhausted and its horses worn out. The situation did not improve during 23 August. Though parts of the Fifth Army tried to resume the offensive, it was the Germans who made ground, particularly on the right, where they got across the water obstacle of the Sambre-Meuse confluence in strength; that despite a counter-attack organised by General Mangin, thenceforth to be recognised as one of the French army’s most ferocious warriors. An hour before midnight Lanrezac concluded he was beaten and telegraphed Joffre that as the ‘enemy is threatening my right on the Meuse . . . Givet is threatened, Namur taken . . . I have decided to withdraw the Army tomorrow’.⁶⁴

The Battle of Mons

Lanrezac made no mention of the situation on his left, though there his British allies had also been locked in combat with the Germans throughout 23 August, showing considerably more effectiveness in defence of a water obstacle – the Mons-Condé Canal – than his own troops had done on the Sambre. The British Expeditionary Force, of one cavalry and four infantry divisions, had begun landing at Le Havre, Boulogne and Rouen eleven days before and had arrived on the canal on 22 August. By the morning of 23 August they were deployed on a front of twenty miles, II Corps to the west, I Corps, commanded by General Douglas Haig, to the east, with the whole of von Kluck’s First Army, fourteen divisions strong, bearing down on them from the north. General Sir John French, the BEF Commander, had expected to march level with Lanrezac in an advance into Belgium. News of Lanrezac’s defeat on the Sambre ruled that out but, when a message arrived from French Fifth Army headquarters just before midnight on 22 August, asking for assistance, he agreed to defend the canal for twenty-four hours. It was evidence of how poorly the French grasped the nature of the German onset that the request was actually for an attack into von Kluck’s flank; von Kluck’s flank already extended beyond both Fifth Army’s and the BEF’s positions. The British, if only for a moment, were to be cast into the role of opposing both the concept and the substance of the Schlieffen Plan – ‘Keep the right wing strong’ were allegedly Schlieffen’s dying words – at the crucial point.

The BEF was equal to the task. Alone among those of Europe, the British army was an all-regular force, composed of professional soldiers whom the small wars of empire had hardened to the realities of combat. Many of them had fought in the Boer War fifteen years earlier, against skilled marksmen who entrenched to defend their positions, and they had learnt from them the power of the magazine-rifle and the necessity of digging deep to escape its effects. Russian veterans of the war against Japan remembered those lessons. The British were the only soldiers in Western Europe who knew them by heart. Ordered to hold the Mons-Condé Canal, they began to dig at once and by the morning of 23 August were firmly entrenched along its length. At the heart of a mining area, the canal offered excellent defensive positions, mine buildings and cottages providing strong-points and the spoil heaps observation posts from which the supporting artillery could be directed onto the advancing enemy masses.⁶⁵

The Germans, who outnumbered them by six divisions to four, were unprepared for the storm of fire that would sweep their ranks. ‘The dominating German impression was of facing an invisible enemy’, hidden behind freshly turned earth in trenches much deeper than the inexperienced French or amateur Belgians thought to dig.⁶⁶ On the Tugela and the Modder rivers, at Spion Kop, the Boers had taught British infantry the cost of assaulting skilled riflemen in deep earthworks and on 23 August the British found the opportunity to teach the lesson themselves. The British Lee-Enfield rifle, with its ten-round magazine, was a superior weapon to the German Mauser, and the British soldier a superior shot. ‘Fifteen rounds a minute’ has become a catchphrase, but it was the standard most British infantrymen met, encouraged by extra pay for marksmanship and an issue of free ammunition to win the badge in their spare time.⁶⁷ A German officer of the 12th Brandenburg Grenadiers was among the first to experience the effect of long-range, well-aimed rifle fire. ‘In front [of my company position] lay an extremely long, flat marshy-looking meadow. Its left side was broken into by scattered buildings and sheds, and on the right a narrow strip of wood jutted into it. At the far end, about 1,500 yards straight ahead, were more scattered groups of buildings. Between the near and far buildings a number of cows were peacefully grazing.’⁶⁸ The peace of the bucolic scene was illusory. On the day following, Captain Bloem would discover how the British ‘had converted every house, every wall into a little fortress; the experience, no doubt, of old soldiers gained in a dozen colonial wars’.⁶⁹ On the morning of Mons, as his company stepped out into the void, the danger the empty vista held suddenly became reality. ‘No sooner had we left the edge of the wood than a volley of bullets whistled past our noses and cracked into the trees behind. Five or six cries near me, five or six of my grey lads collapsed in the grass . . . The firing seemed at long range and half-left . . . Here we were as if advancing on a parade ground . . . away in front a sharp, hammering sound, then a pause, then a more rapid hammering – machine guns!’⁷⁰

The soldiers opposite the Brandenburg Grenadiers belonged to the 1st Battalion Queen’s Own Royal West Kent Regiment and it was their rifles, rather than the battalion’s two machine guns, that were causing the casualties. By the end of the day, Bloem’s regiment was ‘all to pieces’. Many of the men had lost contact with their officers during the fighting and, shamefaced and full of explanations, rejoined only in the evening; 500 had been killed or wounded, including three out of four of his battalion’s company commanders. Bloem was lucky to be untouched. The results were the same in many other units, for every British battalion held its ground and the supporting artillery, including the 60-pounders of 48th and 108th Heavy Batteries, had kept up a steady supporting fire throughout the action. Total British casualties were 1,600 killed, wounded and missing. German casualties, never fully disclosed, must have reached nearly 5,000; the 75th Regiment, of infantry from Bremen, lost 381 men attacking the Royal Scots and King’s Royal Rifle Corps, without making any dent in their line.

That evening the Germans of von Kluck’s army slept where they tumbled down, exhausted, on the north bank of the canal, with the day’s work of carrying crossings over it to do all over again on the morrow; only one foothold had been gained. The British, exhausted too, prepared to fall back on positions a little to the canal’s south. They were flushed with the emotion of a fight well fought; the German official historian’s judgement that ‘the Battle of Mons had ended in failure for the British’ would not have rung true with them.⁷¹ They expected to sustain their defence of the Allies’ left flank the following day. Even as they began to retire to their night positions, however, new orders came in. They were for retreat.

Late on the evening of 23 August, the British liaison officer with the French Fifth Army, Lieutenant Edward Spears, arrived at General Sir John French’s headquarters with alarming news. General Lanrezac had warned Joffre that, as a result of the German success on the Sambre, he was giving orders for Fifth Army to retreat southwards the following day. French, who had announced only a few hours before, that ‘I will stand . . . on the ground now occupied’ and that positions were to be strengthened ‘by every possible means during the night’ was forced to recognise that, as his allies intended to fall back, he must do likewise.⁷² On the morning of 24 August, the BEF began a general retirement. At 9.35 Joffre explained in a message to the Minister of War why the whole front must be withdrawn.

In the north, our Army operating between the Sambre, the Meuse and the British Army, appears to have suffered checks of which I still do not know the full extent, but which have forced it to retire . . . One must face facts . . . Our army corps . . . have not shown on the battlefield those offensive qualities for which we had hoped . . . We are therefore compelled to resort to the defensive, using our fortresses and great topographical obstacles to enable us to yield as little ground as possible. Our object must be to last out, trying to wear the enemy down, and to resume the offensive when the time comes.⁷³

THE GREAT RETREAT

The great retreat had begun, a retreat which would carry the French armies, and the BEF on their left, back to the outskirts of Paris during the next fourteen days. GQG, Joffre’s headquarters at Vitry-le-François, would be abandoned on 21 August, first to roost at Bar-sur-Aube, then to establish itself on 5 September at Châtillon-sur-Seine, the river on which Paris itself stands. Yet Joffre’s despatch, doleful as it must have read to Messimy, Minister of War, remains one of the great documents of the war. In its few sentences it sketched out a plan of recovery, even of eventual victory. The great fortresses, Verdun foremost, were indeed still in French hands. The topography which defends France against Germany from the east, the mountains of the Vosges, the waterways of the Seine river system, were unviolated. The spirit of the French army, unwisely committed in peace to a maniac offensive, survived unbroken by war. Could the army but retain its cohesion as it fell back on the capital, the opportunity for a counterstroke remained. With every mile marched, the German army’s links with its base of support on and beyond the Rhine attenuated, while the French army’s were shortened and strengthened. ‘Future operations’, Joffre wrote in his General Instruction No. 2 of 25 August, ‘will have as their object to reform on our left a mass capable of resuming the offensive. This will consist of the Fourth, Fifth and British Armies, together with new forces drawn from the eastern front, while the other armies contain the enemy for as long as possible.’⁷⁴

The location indicated by Joffre for the positioning of the ‘new offensive mass’ (comprising the Sixth Army, under General Maunoury, and Ninth, under General Foch) was the line of the River Somme near Amiens, seventy-five miles south-west of Mons. Thus Joffre already envisaged a long retreat before his redeployment of forces could permit a resumption of the attack. There was a grim realism to his appreciation of the French army’s situation. Even in Lorraine, where it had suffered the worst of its setbacks, thirty miles was the longest retreat it had yet made. The reality of the coming retreat was to be grimmer by far than anything Joffre anticipated. The German infantry of the right wing, despite twelve days of fighting and marching through Belgium, remained fresh. Buoyed up by victories already gained, hardened by days on the road, hearts high with the expectation of final victory soon to come, they were ready to forget sore feet, lean on their chinstraps and step out with a will if the demands of distance would defeat the French army. ‘This frantic, everlasting rush’, Bloem’s battalion commander told him on the seventh day after Mons, ‘is absolutely essential . . . use all your powers to keep up spirits at any price. Make it clear that we must allow the enemy no rest until we have utterly defeated him on the whole front. Tell them that sweat is saving blood.’ Bloem’s Brandenburgers needed little encouragement. Despite ‘inflamed heels, soles and toes . . . whole patches of skin rubbed off to the raw flesh’, they kept up the pace under the grilling sun of one of the century’s most brilliant summers for day after day.⁷⁵ Falling back before them, the 1st Battalion the Gloucestershire Regiment, for example, recorded a distance covered of 244 miles in thirteen days, with only one of rest (29 August) and two successive marches of over twenty miles on 27 and 28 August.⁷⁶ What the British and French endured, the Germans did likewise.

Both sides fought as well as marched, the French and British to delay the German advance or to escape from danger, the Germans to force a way through any resistance they met. The British I Corps had to fight at Landrecies and Maroilles on 26 August but, since it had suffered very little at Mons, disengaged easily and resumed its retreat; II Corps, battered by Mons, was forced to fight at Le Câteau on the same day an even bigger battle in order to get away. General Smith-Dorrien, commanding II Corps, had three infantry divisions under command, supported by the Cavalry Division. His tired men were assaulted on the morning of 26 August by three German infantry and three cavalry divisions, reinforced during the day by two more infantry divisions, a total of eight against four. Such an inequality of force offered the Germans the opportunity to overlap the ends of the British line and that, as the day developed, was what they achieved. The front ran along the ancient Roman road between Le Câteau and Cambrai where, three years and three months later, the British would launch the first massed attack with tanks, a weapon of war not then invented or even envisaged. At first the British infantry held the line by their usual outpouring of aimed and rapid rifle fire, supported by salvoes from the field artillery. Then, as enemy numbers mounted during the afternoon, the flanks began to crumble, units to break up and batteries to lose their gun crews under the weight of opposing bombardments. As evening approached, II Corps stared dismemberment in the face. It was saved partly by German mistakes but, as much as anything, by the intervention of Sordet’s Cavalry Corps, which at Le Câteau retrieved much of the reputation it had lost by its failure to find the Germans in their advance through Belgium, and by one of the despised French Territorial divisions, whose over-age reservists fought valiantly outside Cambrai to delay the arrival of the German II Corps. As dusk fell, II Corps, which had lost 8,000 killed, wounded and missing during the battle – more than Wellington’s army at Waterloo – summoned its reserves of strength to slip away and resume the retreat.⁷⁷ Thirty-eight guns, half a divisional artillery, were lost nonetheless, despite desperate attempts to save them. At the position of 122nd Battery, Royal Artillery, efforts at rescue by a gallant officer and his team to extricate their equipment left ‘an extraordinary sight: a short wild scene of galloping and falling horses, and then four guns standing derelict, a few limbers lying about, one on the skyline with its pole vertical, and dead men and dead horses everywhere’.⁷⁸

On the day of Le Câteau, Joffre met Sir John French, the BEF’s commander, at St Quentin, together with Lanrezac and General d’Amade, Commander of the Territorial Group which had fought so unexpectedly well on the BEF’s left. It was not a happy meeting. Lanrezac and French had got on badly since their first encounter ten days earlier, while Joffre was already beginning to doubt the capacities of the Fifth Army commander, who had long been his protégé. The atmosphere of the conference, held in a darkened room in a private house, was uneasy. French denied having received Joffre’s General Instruction No. 2, for a future counter-offensive. All he could talk of were his own difficulties and, by implication, of Lanrezac’s failure to support him. Lanrezac’s manner implied that the BEF was an embarrassment rather than a support. There was a language difficulty. The French did not speak English, French scarcely any French; General Henry Wilson, Deputy Chief of Staff, translated. There were also personal differences. Joffre and Lanrezac, big, heavy men in dark blue, gold-buttoned uniforms, looked like station masters, the vulpine Wilson and the peppery French, in their whipcord breeches and glittering riding boots, like masters of foxhounds. It was also confusing to the French that the commander of the BEF was a field marshal. Maréchal was, in the French army, not a rank but a ‘dignity of state’, conferred on victors. The republican soldiers, none of higher rank than general, looked askance at a titular superior whose successes had been won against South African farmers.

The conference came to no clear decision and, when it ended, Lanrezac declined to lunch with French.⁷⁹ Joffre, however, accepted and, when he left, returned to GQG with the intention of stiffening Lanrezac’s backbone. He perceived that the British needed a breathing space, for he was aware of the risk that a beaten BEF might disengage and head for safety in the Channel ports, and so sent orders to Lanrezac to check his retreat next day, 27 August, and counter-attack the German Second Army, treading close on his heels in its path towards Paris. Lanrezac complained but obeyed. His instructions were to align Fifth Army along the upper course of the River Oise, which Bülow’s divisions would have to cross to reach their objectives, with two corps, X and III, facing north in defence and another, XVIII, to attack to the west where the river turned south to join the Seine at Pontoise above Paris. A fourth Corps, the I, commanded by the very determined Franchet d’Esperey, was to stand in reserve behind the right angle formed by Fifth Army’s two wings. The battle – known to the French as Guise, to the Germans as St Quentin – opened on the morning of 29 August in thick mist. The Imperial Guard Corps and Plattenburg’s X Corps stepped out with a will, their commanders believing that no serious French resistance was to be met before the River Aisne, thirty-five miles distant. They were surprised by the strength of X and III Corps’ opposition, against which they began to suffer heavy casualties. Plattenburg, the Guard Corps commander, lost a son killed in the fighting and at one stage Prince Eitel Friedrich, the Kaiser’s second son, had to put himself at the head of the 1st Foot Guards, Germany’s premier regiment, and lead it forward beating on a drum.⁸⁰

During the course of the day, however, the Guard and the Hanoverians of X Corps advanced some three miles and, as evening approached, were preparing to consolidate the ground won. At that moment the character of the battle was transformed. Franchet d’Esperey had been ordered shortly after noon to engage in support and at six o’clock, having spent the intervening hours positioning his artillery to achieve maximum firepower effect, he did so in person. Riding a chestnut charger at the head of regiments advancing behind their unfurled colours and the braying brass of their bands, with the corps artillery thundering overhead, he led his soldiers forward in counter-attack. The effect galvanised III and X Corps to join in and, as darkness fell, villages lost in the morning were retaken and the victorious French took up positions from which they intended to resume the counter-attack next day. Their success was all the more surprising since their orders had been merely to hold ground, while de Mas Latrie’s XVIII Corps relieved pressure on the British by attacking towards St Quentin. The result of 29 August on his front was disappointing and he would shortly be relieved of command. Franchet d’Esperey, by contrast, made his reputation at Guise. ‘Desperate Frankie’, as his British admirers christened the fire-eater, would soon succeed Lanrezac at the head of the Fifth Army. It would be a just reward, for his spectacular intervention had halted the Germans in their tracks and won an extra day and a half for the army to reposition itself for the counterstroke which Joffre remained determined to deliver.

Whether he could or not now depended more on the movements of the German armies than his own. Were they to persist in their march south-westward, aiming to pass Paris to the right, Joffre’s scheme of forming an offensive mass to drive into their flank might be defeated by distance and logistic difficulty. Were they, on the other hand, to press to the south-east, leaving Paris on the left, they would be doing the French what Schlieffen, in another context, had called a ‘willing favour’. Schlieffen had, as his Great Memorandum reveals, come to fear that whichever decision was taken, it would favour the French. To aim to pass Paris to the right would expose the German outer wing to a thrust launched from the Paris fortified zone by its strong garrison; to pass Paris to the left would open a gap between the outer German force and those with which they should keep station, for Paris, like a breakwater, would then divide the tide of the German onset, open a gap in the line and expose the force the wrong side of it to an alternative thrust from Paris in the opposite direction. This ‘problem of Paris’ had driven Schlieffen to ‘the conclusion that we are too weak to continue operations in this direction’.⁸¹ The fault of conception Schlieffen had recognised in his study the German General Staff was now discovering in the field, as its troops marched southward, while their chiefs dithered about their eventual destination.

The difficulty of choice had revealed itself soon after the Kaiser and the Great General Staff – the Supreme Army Command, Oberste Heersleitung or OHL, as it became in war – had displaced from Berlin to Coblenz, on the Rhine, on 17 August (its next location would be in Luxembourg and its final station the little resort town of Spa, in Belgium). Moltke’s decision to allow von Bülow, of Second Army, to oversee the operations of First and Third, understandable in the early stages of the campaign while the need to overwhelm Belgium was paramount, began to have unfortunate consequences soon after the move to Coblenz had been completed. Bülow’s anxiety to assure mutual support between the armies of the right wing deprived Hausen, Third Army, of the chance to strike into Lanrezac’s rear as he disengaged from the Sambre on 24 August. Then, as the line of battle descended to the River Somme, Moltke allowed anxieties of his own about the predicament of the Eighth Army, defending East Prussia against the Russians, to distort his control of the larger and more critical operations in the west. Seeing in the fall of Namur a chance to economise force, he decided to redirect the troops thus released not to their parent formations but across Germany to the eastern frontier.⁸²

Eighth Army did not want the reinforcement of the Guard Reserve and XI Corps, as Ludendorff, newly appointed its Chief of Staff, told OHL on 28 August. They were sent all the same. Meanwhile the marching armies had been further weakened by the detachment of III Reserve Corps to contain the Belgian army in the Antwerp entrenched camp, of IV Reserve Corps to garrison Brussels and of VII Reserve Corps to besiege Maubeuge on the Sambre, where a large French garrison bravely held out behind enemy lines. The loss of five corps from the fighting line – one-seventh of the western army – actually eased Moltke’s logistical difficulties, which grew as the armies drew further away from Germany but closer together as they approached Paris on the overcrowded road network. Nevertheless, preponderance of force at the decisive point is a key to victory and Moltke’s dispersions made preponderance less rather than more likely of achievement. On 27 August, moreover, he further diminished his chance to secure a concentration of superior force by ordering the outer armies, von Kluck’s First, von Bülow’s Second, to fan out. First Army was to pass west of Paris, Second to aim directly for the fortified city, while Third was to pass to the east and Fourth and Fifth, still battling with the French armies defending the lower Meuse, to press westward to join them. Sixth and Seventh, operating on the front where the French had launched their opening offensive of the war, were to attempt to reach and cross the River Moselle.

The march west of Paris was the manoeuvre which Schlieffen had deemed the German army ‘too weak’ to realise. If attempted, it might have proved so but the practicability of Moltke’s directive was not put to the test. The day after its issue, 28 August, von Kluck independently decided to change his line of march and move south-eastward, inside Paris, giving as his reasons the disappearance of any threat from the BEF, seemingly incapacitated by Le Câteau, and the desirability of finally disabling Fifth Army by a drive into its flank. Moltke, despite his quite precise order of 27 August that Kluck should go west of Paris, acquiesced and on 2 September went further. In a message to First and Second Armies, wirelessed from OHL’s temporary headquarters in Luxembourg, he announced that it was ‘the intention of the High Command to drive the French back in a south-easterly direction, cutting them off from Paris [italics supplied]. The First Army will follow the Second in echelon and will also cover the right flank of the armies.’ This was an acceptance of events rather than an effort to determine them. Second Army had halted to recuperate from the effects of fighting and the long march, so that for First to echelon itself with it would entail a pause also. The French Fifth Army meanwhile was slipping away to the east, thus eroding the danger of an attack into its flank and distancing itself from Paris in so doing. The BEF was not disabled but had merely disappeared into the countryside, unhampered by the German cavalry as the advancing Germans had been by the French in Belgium in the opening weeks of the campaign, while the growing assemblage of Joffre’s new striking force in and around Paris remained undiscovered by the enemy altogether.⁸³

Meanwhile the marching armies tramped on, fifteen and twenty miles a day in the heat of a brilliant late summer. ‘Soon we were crossing the last ridge that separated us from the Marne valley’, recorded Bloem. ‘It was another grilling, exhausting day. Twenty-five miles up hill and down dale under a blazing sun. To our left we could hear the guns of Bülow’s army with which we seemed nearly in touch again.’ There were flashes of action, engagements between advance and rear guards, short, bitter little battles, such as that at Néry on 1 September, where the British 1st Cavalry Brigade and L Battery, Royal Horse Artillery, held up the progress of the German 4th Cavalry Division for a morning. L Battery’s gunners won three Victoria Crosses in their unequal contest with the enemy, which ended, a German historian recorded, ‘decidedly to the disadvantages of the German cavalry’.⁸⁴ There was a great deal of bridge-blowing and re-bridging, as the armies negotiated the many-branched river system of the Paris basin, of contested delays at obstacles, of artillery exchanges, of brief outbursts of rifle fire, as scouts ran into outposts or the tail of a retreating column was overtaken by pursuers. For the vast majority on both sides, however, the last week of August and the first of September was an ordeal of day-long marches, begun before the sun rose, ended in the twilight. A trooper of the 4th Dragoon Guards, Ben Clouting, recorded that his regiment was roused at 4.30 on the morning of 1 September, 2 on the morning of the 2nd, 4.20 a.m. on the 3rd and 5th and 5 a.m. on the 6th. He remembered that the horses, beside which they often walked to spare their backs, ‘soon began to drop their heads and wouldn’t shake themselves like they normally did . . . they fell asleep standing up, their legs buckling. As they stumbled forward . . . they lost their balance completely, falling forward and taking the skin off their knees.’ For the men, ‘the greatest strain . . . worse than any physical discomfort or even hunger was . . . fatigue. Pain could be endured, food scrounged, but the desire for rest was never-ending . . . I fell off my horse more than once, and watched others do the same, slowly slumping forward, grabbing for their horse’s neck, in a dazed, barely conscious way. At any halt men fell asleep instantaneously.’⁸⁵

The infantry, who got no chance to ride, dropped behind the column of route in scores and these stragglers, ‘in grim determination . . . hobbled along in ones or twos . . . as [they] sought desperately to stay in touch with their regiments . . . Food came up from Army Service Corps ration dumps, which were just boxes of biscuits [and] tins of bully beef . . . Very occasionally, a chalk notice marked the food up for a particular regiment, but more often than not we just helped ourselves, stuffing what we could into every pocket.’⁸⁶ Joffre, out on inspection of the French armies on 30 August, passed ‘retreating columns . . . Red trousers had faded to the colour of pale brick, coats were ragged and torn, shoes caked with mud, eyes cavernous in faces dulled by exhaustion and dark with many days’ growth of beard. Twenty days of campaigning seemed to have aged the soldiers as many years.’ The French and British, long though their daily marches, were at least falling back on their lines of supply. The Germans marched ahead of theirs and often went without food, though, like the British, their need was for rest rather than rations. A French witness noticed on 3 September, when a unit of the invaders reached their billets for the night, ‘they fell down exhausted, muttering in a dazed way, “forty kilometres! forty kilometres!” That was all they could say.’⁸⁷

On 3 September von Kluck’s headquarters were installed in Louis XV’s chateau at Compiègne. It was there that he received Moltke’s wireless message of 2 September directing his First Army to follow Bülow’s Second ‘in echelon’ to the south-east, in order to cut the French off from Paris.⁸⁸Kluck decided to interpret the order literally, as giving him freedom to veer further eastward still in pursuit of Lanrezac’s Fifth Army, to cross the River Marne and to initiate the decisive battle that Moltke actually intended to be delivered by the armies of the centre, coming west from the Meuse. The German strategic effort, though neither Moltke nor Kluck perceived it, was beginning to fall apart. ‘Moltke’, a French historian comments, ‘had never much believed in the possibility of manoeuvring masses . . . like his uncle [the elder Moltke], he thought it necessary to leave each army commander a wide freedom of movement.’⁸⁹ Laxity of control had not mattered in 1870, when the front of battle was narrow and the opportunity for armies to diverge from the critical axis of advance correspondingly slight. Moltke the Younger’s easy reign over the far wider battlefront of 1914 had resulted in his right-hand army, the army on which all depended, first slipping to the south when it should have been marching south-westward, then turning south-eastward, at right angles to the direction which the plan of campaign laid down it must maintain for victory to be achieved.

Critics would later point out Schlieffen’s own inability to decide what track the right wing should take, apologists argue that Kluck was doing the right thing by keeping on Lanrezac’s heels. The truth is that he was being led by the nose. Every mile he marched in pursuit of the Fifth Army, once he had crossed the Oise and headed towards the Marne, served Joffre’s purpose. The line on which Joffre wished to fight may have receded southward from the Somme to the Oise to the Marne as the situation map shifted and August drew into September, the opportunity to deliver the disabling blow improved proportionately. For the further Kluck widened the gap between army and Paris to his right, without achieving the crucial overlap which would allow him to begin an encirclement of Lanrezac from the west, the more space he created for Joffre to position the ‘mass of manoeuvre’ against the German flank. That mass, with the existing garrison of Paris, menaced a fiercer strike against Kluck than he could now hope to deliver at the enemy.

The creation of this ‘mass of manoeuvre’ had been foreshadowed in Joffre’s General Instruction No. 2 of 25 August. Then he had said that it was to consist of VII Corps, of four Reserve divisions and perhaps another Active corps, which were to be transported to the west by rail. By 1 September it consisted of VII and IV Corps, taken from First and Third Armies, and the 55th, 56th, 61st and 62nd Reserve Divisions, the whole forming Sixth Army under General Maunoury; with it was associated the garrison of Paris, including the 45th Division, from Algeria, five Territorial Divisions, 83rd, 85th, 86th, 89th and 92nd, a brigade of Spahis and a brigade of fusiliers-marins.⁹⁰ Together they constituted the Armies of Paris, under the overall command of General Gallieni. Gallieni, a veteran of the French wars of empire, was sixty-five in 1914, Maunoury sixty-seven; even in a war of old generals – Moltke was sixty-six, Joffre sixty-two – they might have appeared too elderly to find energy sufficient to mastermind a counterstroke against the largest army ever deployed in the field. Maunoury and Gallieni, however, were men of vitality, Gallieni exceptionally so. Recalled from retirement on 25 August to replace the ineffective General Michel as Military Governor of Paris, he had at once warned Messimy, Minister of War, that the enemy would be at the gates in twelve days to lay a siege the capital could not withstand. He demanded reinforcements, which could only be got from Joffre who was unwilling to release any and, as supreme commander with war powers, could not be overruled by ministers or even the President. Gallieni’s demands provoked a government crisis. Messimy, finding himself blamed for the dangers of which Gallieni was now warning, insisted on being dismissed rather than accept a new appointment and by so doing brought about the resignation of the whole ministry. Messimy was replaced by the tough and taciturn Millerand and departed to join the armies at the front as a major of reserve.⁹¹

The political upheaval shook Joffre’s imperturbability no more than military setback. He adhered to his routine of the long lunch, of a solid dinner and regular hours of sleep. Nevertheless, unlike Moltke, who remained secluded in his Luxembourg headquarters far from the scene of action, he also visited subordinate commanders and troops almost every day. He saw Lanrezac on 26, 28 and 29 August, visited the commanders of Third and Fourth Armies on 30 August and Lanrezac again on 3 September. He also saw Sir John French on 26 August and 3 September. The British were causing anxiety. French had been shaken by the intensity of the fighting at Mons, even more by that at Le Câteau, and had convinced himself that his army needed several days of rest before it could re-enter the line. As the retreat lengthened, he and his staff officers began to consider the eventuality of retiring to base, leaving France altogether and returning only when the troops had rested and re-equipped in England. He had come to believe that the French to left and right of him were retreating without warning, leaving him exposed to attacks by the advancing Germans. He next announced his intention of retiring below the Seine, in eight days of easy marching, and of transferring his stores from Rouen and Le Havre, on the English Channel, to St Nazaire or even La Rochelle on the Atlantic coast. Kitchener, Secretary of State for War, demanded clarification in a series of telegrams. When none came, he took a destroyer to France, summoned French to the British embassy in Paris and left him in no doubt that his task was to co-operate with Joffre even at extreme risk to his own army.⁹²

That meant its taking its place in the ‘mass of manoeuvre’ which, by 3 September, was gathered north-west and west of Paris: the new Sixth Army, the Paris garrison, the BEF, the Fifth Army and, on its right, the Ninth Army, also new and commanded by General Ferdinand Foch. Foch, promoted from command of XX Corps, was a star in the ascendant. Lanrezac’s star fell on 3 September; Joffre motored to his headquarters at Sézanne that day to tell him he was replaced by Franchet d’Esperey. It was a painful meeting. They were friends and Lanrezac had been Joffre’s protégé. Now he was a man worn-out by the burden of confronting the danger he, almost alone, had foreseen of the German attack through Belgium. The two generals walked around the playground of the school where Fifth Army had its headquarters while Joffre explained that he judged his subordinate to have lost the power of decision. Then Lanrezac departed, accompanied by a single non-commissioned officer, not to be seen again in uniform.⁹³

Gallieni, also a star in the ascendant, was meanwhile terrorising the municipality of Paris with his orders to put the city into a state of defence. On 2 September the government had, as in 1870, transferred its seat to Bordeaux. Joffre had incorporated the capital into the Zone of the Armies, where he ruled with total power, on 31 August. With constitutional authority, therefore, the Military Governor issued instructions to prepare the Eiffel Tower for destruction (it was the transmitting station for general staff radio communications), to lay demolition charges under the Seine bridges, to send all rolling stock useful to the enemy out of the Paris rail system, to provision the 2,924 guns of the fortifications with ammunition, to clear fields of fire for the artillery of trees and houses and to conscript the labourers to do the work. Paris, in 1914, was still a fortified city, surrounded by walls and a girdle of forts. It was also, under Gallieni’s command, constituted an Entrenched Camp, with improvised defences stretching out into its surrounding countryside, further to enhance the ‘obstacle of Paris’ which had so troubled Schlieffen in the long years while he had been devising his plan.

Yet the obstacle had already done its work. On 3 September Schlieffen’s ‘strong right wing’ represented by Kluck’s First Army, had drifted forty miles to the east of Paris and was aligned to the south, with the Sixth Army and the Paris garrison behind it, the BEF on its right flank, the Fifth Army to its front and Foch’s Ninth Army menacing its left and threatening an irruption into the gap which had opened between it and Bülow’s Second Army. It was the existence of Paris and Lanrezac’s evasive manoeuvring that had brought about this result.

Meanwhile the French railway system was hurrying to the front the forces with which Joffre planned to deliver his counterstroke. Since it centred on Paris, its network brought troops rapidly from the increasingly stabilised eastern sector to the critical points. By 5 September the Sixth Army consisted, besides Sordet’s Cavalry Corps and the 45th (Algerian) Division, of the VII Corps, brought from Alsace, and the 55th and 56th Reserve Divisions from Lorraine; the IV Corps was on route from Fourth Army. The Ninth Army, originally constituted as the Foch Detachment, comprised the IX and XI Corps transferred from Fourth Army, together with the 52nd and 60th Reserve Divisions and 9th Cavalry Division, the 42nd from Third Army and the 18th Division from Third Army. Between the Paris Entrenched Camp and the Marne, Joffre therefore disposed, at the opening of the great battle named after the river, of thirty-six divisions, including the BEF, strengthened by the arrival of four fresh brigades from England, while the German First, Second, Third, Fourth and Fifth Armies opposing totalled just under thirty. Schlieffen’s ‘strong right wing’ was now outnumbered, the result of Moltke’s failure to control his subordinates and of Joffre’s refusal to be panicked by early defeat. Much else had contributed to the mismatch, notably the logistic difficulties imposed on the Germans as their lines of communication lengthened, and the consequent easing of the problems of reinforcement and supply enjoyed by the French as they fell back on the centre. Nevertheless, the opening circumstances of the Battle of the Marne betrayed a failure of German generalship. It remained to be seen whether French generalship might yet pluck victory from the jaws of defeat.

THE BATTLE OF THE MARNE

‘It is the thirty-fifth day’, the Kaiser exulted to a delegation of ministers to his Luxembourg headquarters on 4 September, ‘we are besieging Rheims, we are thirty miles from Paris.’⁹⁴ The thirty-fifth day had an acute significance to the German General Staff of 1914. It lay halfway between the thirty-first day since mobilisation, when a map drawn by Schlieffen himself showed the German armies poised on the Somme to begin their descent on Paris, and the fortieth, when his calculations determined that there would have been a decisive battle.⁹⁵ That battle’s outcome was critical. Schlieffen, and his successors, had calculated that the deficiencies of the Russian railways would ensure that not until the fortieth day would the Tsar’s armies be assembled in sufficient strength to launch an offensive in the east. Between the thirty-fifth and the fortieth day, therefore, the outcome of the war was to be decided.

On 4 and 5 September, the commanders issued the orders which would set the engagement in motion. ‘The enemy’, von Moltke admitted, on 5 September, ‘has eluded the enveloping attack of First and Second Armies and has succeeded, with part of his forces, in gaining contact with Paris.’⁹⁶First and Second Armies were therefore to stand on the defensive outside Paris, while Third Army was to advance towards the upper Seine and Fourth and Fifth Armies were to attack to the south-east, with the object of opening a way for the Sixth and Seventh to cross the River Moselle and complete the encirclement of the enemy. This was the opposite of what Schlieffen had intended; his plan was for the First and Second Armies to drive the French into the arms of the left wing. On 4 September, Joffre had issued General Instruction No. 6 which exactly anticipated Moltke’s recognition of his predicament and proposed means to exploit it. ‘It is desirable to take advantage of the exposed position of the German First Army to concentrate against it the strength of the Allied armies [opposite]’.⁹⁷ Accordingly, the Sixth Army, at the outermost extremity, was to cross the Ourcq, a tributary of the Marne, and advance round the Germans’ flank, while the BEF, the Fifth Army and Foch’s Ninth Army were to make a fighting advance northward; effective date of the order, 6 September. The biter was to be bit. The German, not the French, army was to be the target of an encirclement.

What stood between conception of the order and its realisation were water barriers, not the Marne itself, but its tributaries, the Ourcq, which flows north to south athwart the line of advance of Maunoury’s Sixth Army, and the Morins – the Grand and the Petit – which run east to west, and so across the front of the BEF and the Sixth and the Ninth Armies; the latter’s room for manoeuvre was further impeded by the marshes of the St Gond which formed part of the riverine system. None of the waterways was a serious obstacle. They defined, nevertheless, the lines on which action was to be joined and required preparation for deliberate attack. That necessity, as it proved, was to favour the Germans rather than the French, thanks to tactical quick-thinking by a commander on the spot at a critical point. The man was General von Gronau, an artillery officer commanding IV Reserve Corps. His formation had played little part in the campaign thus far, had indeed been much weakened by transfers of units to act as flank guards for the main body of First Army. Von Gronau, nevertheless, remained alert to his responsibilities. His Corps held station on the outermost edge of the German invasive swathe and was therefore not only in a vulnerable position itself but stood security on the right for the whole offensive deployment. On the morning of 5 September, as Maunoury’s Sixth Army probed forward to take up attacking positions for the following day, he was seized by disquiet at the reports sent back by his attached cavalry division. Its patrols found advancing French troops all across its front. As the IV Reserve Corps was aligned at right angles to and to the rear of von Kluck’s First Army, that meant that the enemy was manoeuvring to take First Army in flank and roll it up. His response was instantaneous and courageous. He decided to attack.

The German advance, 1914

As Maunoury’s advance guard, the 55th and 56th Reserve Divisions and the Moroccan Brigade, breasted forward towards the Ourcq in the mid-morning of 5 September, they were suddenly brought under fire by the rifles, machine guns and artillery of Germans who were occupying terrain supposed empty. The French went to ground and a fierce firefight broke out that lasted the rest of the day. As darkness fell, von Gronau wisely judged he had won the time necessary to save First Army from surprise attack and disengaged his troops, who slipped away to the line the French had intended to assault on 6 September. In bright moonlight the French followed, launching attacks against positions the Germans had already abandoned.

The battle of the Marne had therefore opened a day earlier than Joffre had intended, and on terms dictated by the enemy. Thanks to von Gronau’s independent action, the beckoning open flank which offered the opportunity for an encirclement had been covered and von Kluck given the warning necessary to hurry reinforcements from his centre to his right before the danger heightened. Kluck reacted with an energy and decisiveness he had not shown during the days when he had let his army drift reactively eastward in the footsteps of Lanrezac’s defeat. By the morning of 6 September, he had transferred his II Corps from south of the Marne to west of the Ourcq, to form a line north of von Gronau’s position and he would successively transfer northward the IV Corps on 7 September, the III Corps on 8 September and the IX Corps on 9 September. What strategists call ‘interior lines’ were now working in von Kluck’s favour, as they had worked for Joffre in the last week of August and first week of September, when he had brought the constituents of Sixth and Ninth Armies behind the fighting front from the armies that were holding their ground in Alsace and Lorraine.

There was this difference. It was critical. Joffre’s transfers had not altered the strategic situation on the eastern front, which had stabilised as soon as the French ceased to attack and found strong defensive positions behind the Meuse and Moselle. Kluck’s withdrawals, by contrast, weakened his principal front at the point where his mission was still to deliver a decisive, war-winning blow and at a moment, in the very last of the forty days which were expected to bring victory, when the French were gathering to deliver their counter-offensive over the same ground. Indeed, by 9 September, the fortieth day itself, the German First Army, instrument and hope of Schlieffen’s vision, was not on the Marne at all, but had been withdrawn in its entirety to the Ourcq, where it faced not Paris, in popular imagination the object of the whole campaign, nor the mass of the French army, its strategic target, but Maunoury’s detached manoeuvre force. Between the German First Army and Second an enormous gap had opened, thirty-five miles wide, which the Germans could disregard only because they believed that the enemy troops opposite, the British Expeditionary Force, lacked the strength and had demonstrated the disinclination to penetrate.⁹⁸

The high command of the BEF, though not its brave soldiers, had given von Moltke, Kluck and Bülow reason for so believing. Sir John French, ‘the little Field Marshal’, stout, florid, peppery, had proved a dashing cavalry leader in the British Army’s small wars. At the head of his country’s only field army in the largest war ever to involve it, he displayed an increasing tendency to nerves. The losses at Mons had unsettled him, the far heavier losses at Le Câteau had shaken his resolve altogether. He feared that the BEF would fall to pieces unless given a respite to rest and re-equip. What heightened his anxieties was his fixed conviction that Lanrezac had let him down, retreating from the Sambre without warning and leaving the BEF to cover the withdrawal. Before August was out, he had come to hate Lanrezac and to distrust the French generally. For Joffre he retained a personal regard but, as he told Kitchener on 30 August, ‘my confidence in the ability of the leaders of the French Army to carry this campaign to a successful conclusion is fast waning’.⁹⁹ During the next days, he spoke of transferring his base from the Channel ports to Brittany, of the impossibility of allowing the BEF to ‘take up a position in the front line for at least ten days’, of retiring behind the Seine by ‘marching for some eight days . . . at a considerable distance from the enemy’.¹⁰⁰ It took Kitchener’s visit to Paris on 2 September to check this defeatism but he remained unwilling to rejoin battle. As late as 5 September, when it had been made clear to him that the participation of the BEF in the counter-offensive prescribed by Joffre’s General Instruction No. 6 was essential to its success, he continued to prevaricate. Only when Joffre found the time, at this moment of acute crisis, to visit his headquarters and make a personal appeal did he stiffen. French was an emotional man. Joffre’s clutching of his hands and supplication in the name of ‘France’ set tears running down his cheeks. He tried his ally’s language, fell tongue-tied, then blurted at a staff officer who spoke French better, ‘Damn it, I can’t explain. Tell him that all man can do our fellows will do.’¹⁰¹

Difficulties remained. The BEF had fallen too far to the rear to join at once with Sixth and Fifth Armies in the general offensive. ‘Desperate Frankie’, the new commander of Fifth Army whom all his British collaborators admired, fell into a rage at his ally’s apparent unco-operativeness. Sixth Army, marching up in echelon to bar its advance into the German rear, but opposed progressively by the whole of Kluck’s strength, faltered under one counter-attack after another. It would have been surprising if it had not. In any case an improvised force, its components – four Reserve divisions, only two Active divisions, and a collection of cavalry and Active North African formations – lacked both the quality and numbers to stand up to Kluck’s First Army, which contained eight Active divisions, besides Reserve and cavalry formations. The distances over which the arriving German divisions had to travel, compared to those Sixth Army’s had covered from the eastern frontier, were quite short. The IX Corps, which appeared opposite Maunoury’s left flank on the morning of 9 September, had made the longest march, but it was one of only forty miles. It deployed intact and in vigour. The corps which had arrived earlier had blunted all Maunoury’s efforts to take ground, and had continually counter-attacked. One critical situation had been saved for the French only by a dashing intervention of the 45th Division’s artillery, led by Colonel Nivelle, a future commander of the French army, another by the arrival from Paris of a portion of the city’s garrison mounted in commandeered taxicabs, an episode of future legend. The battle of the Ourcq, between 5 and 8 September, nevertheless tended Kluck’s way. On the evening of 8 September, he felt confident enough to signal his subordinates that ‘the decision will be obtained tomorrow by an enveloping attack’. The Schlieffen Plan, in short, might be about to work after all.¹⁰²

Geography spoke otherwise. The aggressiveness Kluck’s army had shown against Maunoury’s had actually worked to enlarge the gap that now loomed between it and Second Army, a gap too wide for the only German troops not engaged elsewhere, those of 2nd and 9th Cavalry Divisions, to fill. They were, moreover, too weak to oppose the force marching up to exploit the weakness in the German line the gap presented. True to his reluctant word, Field Marshal French started the whole of the BEF forward on 6 September and, though it had ten miles to make up before it reached Joffre’s intended point of departure, it soon covered the distance, bringing with it the elements of a new, third, corps, formed in France on 21 August. The intervention of the British, who fought a sharp encounter action at Rozoy, alarmed von Kluck. Even more alarmed was von Bülow, whose Second Army was heavily engaged throughout the day against the French Fifth Army, galvanised by the leadership of its new commander, Franchet d’Esperey. On 7 September Bülow radioed the high command to warn that he was withdrawing the troops on the east of the gap, into which the BEF was marching, behind the Petit Morin river, for safety, a retreat of ten miles and more. Worse, under pressure during the day, he was obliged to swing his right wing northward, thereby further widening the gap between his army and von Kluck’s and leaving the way open for a full-scale Allied advance to the Marne.

The right wing of the German army was now divided effectively into three sections, with Kluck’s First Army north of the Marne, the right of von Bülow’s Second Army south of the Marne but falling back towards it across the waterways of the Grand and Petit Morin, and his left, which connected only weakly with von Hausen’s Third Army, positioned on the Petit Morin itself, in the Marshes of the St Gond, where that river rose. The whole region ‘is a country of great open spaces; highly cultivated, dotted with woods and villages, but with no great forests, except [those to the south]. It is cut from east to west by the deep valleys, almost ravines, of the Grand Morin, Petit Morin, the Marne, the upper course of the Ourcq, the Vesle, the Aisne and the Ailette.’ The Marshes of the St Gond are a topographical exception, ‘a broad belt of swamp land . . . [extending] from east to west nineteen kilometres, with an average width of three kilometres . . . five lesser roads and three foot-paths cross [the marshes] from north to south, but they are otherwise impassable, forming a military obstacle of the first importance’.¹⁰³ Von Bülow’s left, and the right of von Hausen’s Third Army were, on 6 September, firmly embedded on the northern edge of the marshes, with Foch’s new Ninth Army positioned on the other side. The mission given him by Joffre was to protect the flank of Fifth Army, battling to drive von Bülow beyond the Marne. It was in character that he chose to interpret it offensively. While his centre and right stood fast, he ordered his left, the 42nd Division, to advance, supported by the Moroccan Division and part of IX Corps. During 6 and 7 September they battled valiantly to work their way round the western end of the marshes, while the rest of Ninth Army and the Germans opposite conducted artillery duels over the sodden ground of the marshes themselves.

The battle of the marshes threatened to descend into a stalemate, as that on the eastern frontier had become. Then it was transformed by the uncharacteristic boldness of von Hausen. This Saxon general has been described as too deferential to the wishes of the Prussian Kluck and Bülow on his right, too overawed by the German Crown Prince who commanded on his left, to take forthright decisions in the handling of his own army. On 7 September he displayed an independence that contradicted both judgements. Persuading himself that the ferocity of the two previous days’ fighting had blunted the enemy’s alertness, he decided to launch a surprise night attack. In the moonlit early morning of 8 September, the Saxon 32nd and 23rd Reserve Divisions and the 1st and 2nd Guard Divisions advanced through the marshes and across the dry ground further east, fell on the French with the bayonet and drove them back three miles. This was a local victory that shook the confidence of Foch’s Ninth Army, which lost further ground on its right during the day and merely held its own on the left.

The events of 8 September prompted Foch to draft the later legendary signal: ‘My centre is giving way, my right is in retreat, situation excellent. I attack.’¹⁰⁴ It was probably never sent. Nevertheless, the general’s actions bore out the spirit of those words. During 9 September, using reinforcements lent by Franchet d’Esperey, and in expectation of the arrival of XXI Corps from Lorraine, Foch succeeded in plugging every gap in the line opened by Hausen’s continuing offensive and did, at the day’s end, actually manage to organise a counter-attack at the right hand extremity of his army’s position. Merely by holding his front, Foch achieved a sort of victory.

On the Ourcq, meanwhile, 9 September was also a day of crisis. Kluck’s First Army was now fighting as an independent entity, separated from Bülow’s Second Army by a forty-mile gap into which the BEF was pushing northward towards the Marne almost unopposed, but still formidably strong and still committed to attack. With four corps in line, it still outnumbered Maunoury’s Sixth Army and, overlapping the flanks of the French to north and south, still retained the chance of winning an encirclement battle and so reversing the increasingly dangerous situation on the critical right wing. The weight of his deployment was in the north, where von Quast’s IX Corps, supported by von Arnim’s III, was positioned and prepared to fall on the French 61st Reserve Division, turn its flank and drive into the rear of the defenders of Paris. On the morning of 9 September von Quast began his attack opposed initially only by the weak artillery of the French 1st and 3rd Cavalry Divisions. When his troops came up against the positions of the 61st Reserve Division, they drove the French infantry to flight, so that by early afternoon they were poised to sweep forward into undefended territory. The balance of advantage on the Marne seemed once more to have tilted the Germans’ way.

The Mission of Lieutenant-Colonel Hentsch

That was the local reality. Von Quast sensed no resistance to his front. His soldiers were elated by success. Paris, only thirty miles distant, beckoned. The way to the French capital seemed to lie open and victory therefore to promise. Then, at two o’clock in the afternoon, Quast received a telephone call from Kluck’s headquarters. The offensive was to be discontinued. An order for retreat had been received. The First Army was to retire northward towards the Marne and it appeared not the First Army only but the whole of the right wing. Local reality was dissolved in a larger reality. The great advance, the sweep through Belgium and northern France, the master stroke that was to end the war in the west before the fortieth day, had failed. Schlieffen’s vision had evaporated in the heat of battle.

Not just the heat of battle. The cool appraisal of a military technician had decided that the position of First, Second and Third German Armies was untenable. The technician was a middle-ranking officer of the General Staff, Lieutenant-Colonel Richard Hentsch, the peacetime head of the Operations Section of the Great General Staff, since mobilisation the head of the Intelligence Section at Supreme Headquarters. In the war’s aftermath Allied historians expressed surprise that an officer of such junior rank should have been devolved the authority to nullify Schlieffen’s great plan. The German high command itself, at Hentsch’s request, held an official inquiry in 1917 to examine the probity of his intervention. Even today the scope of the powers delegated to him seems remarkably wide, and all the more so because Hentsch was a Saxon, not a Prussian officer, in an army the Prussians dominated. Moreover, he was an intelligence, not an operations, officer, on a general staff whose operations section treated the intelligence section as a handmaiden. Nevertheless, Hentsch was a considerable figure. He had shone as a student at the War Academy, won the high opinion of his contemporaries and superiors and was on intimate terms with both Moltke and Bülow.¹⁰⁵ He was therefore an obvious person to choose as an intermediary between Supreme Headquarters and the right wing, at a moment when the distance separating them had increased to 150 miles. Moltke felt unable to make what would be a time-consuming journey himself. He judged signal communications to be both unsatisfactory and insecure. His well-informed intelligence section chief was perfectly qualified to bridge the gap. It was unfortunate, and would continue to appear so, that Moltke wrote nothing down but despatched Hentsch on his mission with nothing more substantial to validate his plenipotentiary authority than a verbal instruction.¹⁰⁶

Hentsch set off by motor car from Luxembourg at eleven o’clock on the morning of 8 September. He was accompanied by two captains, Köppen and Kochip, and visited in succession the headquarters of Fifth, Fourth and Third Armies. With each he discussed its situation and concluded that no withdrawal from its front was necessary, with the possible exception of Third Army’s right wing; he nevertheless radioed Luxembourg that the ‘situation and outlook entirely favourable at Third Army’.¹⁰⁷ In the evening he arrived at Second Army’s headquarters, from which Bülow was temporarily absent. When Bülow returned, he, his two principal staff officers and the Hentsch party settled to survey the situation. The result of their discussion was to be decisive for the outcome of the campaign in the west. Bülow dominated. He represented his army’s predicament as one the enemy might exploit in two ways, either by turning the right wing of his own army or by massing against the left wing of First Army. Since the gap between the two was in the hands of the French and British, they enjoyed freedom of action and could use it with ‘catastrophic’ results. Bülow proposed to avert disaster by a ‘voluntary concentric retreat’.¹⁰⁸ That meant a withdrawal from the positions from which the German offensive threatened Paris to safer but defensive lines beyond the Marne. On that note, towards midnight, the meeting dispersed. Next morning, 9 September, Hentsch conferred again with Bülow’s staff officers, though not the General himself, and agreed that he would visit Kluck at First Army to advise a retirement, which would close the menacing gap. He left at once. While he was covering the fifty miles to First Army headquarters, Bülow decided to act on the conclusions arrived at by his juniors. He signalled Kluck and Hausen that ‘aviator reports four long columns marching towards the Marne’ (the aviator was Lieutenant Berthold, the columns those of the BEF) and that consequently, ‘Second Army is beginning retreat’.¹⁰⁹

The retreat that followed was orderly but precipitate. Once Second moved, First and Third were obliged to conform, as by the working of interlocking parts. Mechanistically, Fourth, Fifth and Sixth fell in with the retrogression. Along a front of nearly 250 miles, the German infantry faced about and began to retrace its steps over the ground won in bitter combat during the last two weeks. Moltke gave the orders himself for the retreat of the left wing, and in person. When Hentsch at last returned to Supreme Headquarters at two in the afternoon of 10 September, bringing the first comprehensive account of the situation at the front to amplify the few brief signals Moltke had received from him and Bülow in the previous two days, the Chief of Staff decided that he must do what he might have done in the first place and visit his subordinate army commanders himself. On the morning of 11 September he departed by road from Luxembourg, first for the headquarters of Fifth Army, where he saw the Crown Prince, next to Third Army, where he found Hausen stricken with dysentery, then to Fourth Army. While there he received a message from Bülow warning of a new danger to Third Army, posed by a fresh French attack, and decided that Fourth and Fifth must follow Third, Second and First in retirement. The positions to which he directed them were those on the river system next above the Marne, that of the Aisne and its tributaries. ‘The lines so reached’, he stipulated ‘will be fortified and defended.’¹¹⁰

Those were the last general orders he issued to the German armies; on 14 September he was relieved of command and replaced by General Erich von Falkenhayn, the Minister of War. They were also the most crucial orders to be given since those for general mobilisation and until those initiating the armistice four years and two months later. For the ‘fortification and defence’ of the Aisne, which the German First and Second Armies reached on 14 September, initiated trench warfare. Whatever the technical factors limiting the German army’s capability to manoeuvre with flexibility and at long range from railhead in 1914 – lack of mechanical transport, rigidity of signal networks working along telephone and telegraph lines – none constrained its power to dig. It was better provided with field engineer units than any army in Europe – thirty-six battalions, against twenty-six French – and better trained in rapid entrenchment.¹¹¹ The entrenching tool had become, by 1914, part of the equipment of the infantryman in every army. However, while the British cavalry took pride in avoiding entrenchment exercises, and the French disregarded ‘the most demanding notions of cover’, the German soldier had been obliged to use the spade on manoeuvre since at least 1904. ‘From 1906 onward, foreign observers [of German manoeuvres] noted that German defensive positions frequently consisted of several successive trench lines linked by communication saps, often with barbed wire entanglements strung in front of them.’ The Germans had not only noted the significance of entrenchments in the Boer and Russo-Japanese Wars but, unlike others, had drawn the lesson.¹¹²

When, at the end of the second week of September, therefore, the French and British troops pursuing the enemy came up against the positions on which the Germans had halted, they found their counter-offensive halted by entrenchments which ran in a continuous line along the crest of the high ground behind the Aisne, and its tributary the Vesle, between Noyon and Rheims. The line ran on beyond, turning south-west at Verdun and following the River Meurthe until it climbed away through the precipitate Vosges to reach the Swiss frontier near Basle. Beyond Rheims, however, the opposing armies – the German Fifth and Sixth, the French First and Second – had been so weakened by combat and by withdrawals to reinforce the crucial western sector that active operations had attenuated. The Aisne had now become the critical front and there, between 13 and 27 September, both sides mounted a succession of attacks, as troops became available, the Allies in the hope of pressing their pursuit further, the Germans with that of holding their line or even going over again to the offensive. The Allies began in optimistic mood. Wilson, British Deputy Chief of Staff, had discussed with Berthelot, his French equivalent, during the advance to the Aisne how soon their armies would be on the Belgian frontier with Germany. He thought a month, Berthelot three weeks. They were shortly to discover that the days of ‘open warfare’ were over.¹¹³

The Western Front in outline, 1914–18

The Aisne is a deep, wide river, passable only by bridging. At the outset of the battle not all the bridges had been destroyed, while others were improvised; none was safe while within range of German artillery fire. Beyond the Aisne the ground rises some 500 feet above the valley to form a long massif, indented by re-entrants between bluffs and in places heavily wooded. The feature, some twenty-five miles long, affords excellent points of observation and dominating fire positions, while the road that traverses it, the Chemin des Dames, laid out for the daughters of Louis XV, provides easy lateral communication from left to right.¹¹⁴ A British formation, the 11th Infantry Brigade, was the first to attempt an assault. It had found an unbroken bridge at Venizel and managed to establish itself on the crest on 12 September, after a thirty-mile approach march in pouring rain.¹¹⁵ Thereafter the difficulties increased. The French Sixth Army tried on 13 September to get round the flank of the Chemin des Dames ridge near Compiègne but met German resistance across its whole front. The BEF was also held up under the centre of the Chemin des Dames that day and the only success was achieved on the right where the French Fifth Army found the gap that still existed between von Kluck’s and von Bülow’s armies and reached Berry-au-Bac on the Aisne’s north bank.

The gap was rapidly being filled, however, by troops hurrying down from Maubeuge, where the valiant French garrison had at last been compelled to surrender the fortress on 8 September, and by others brought from Alsace and Lorraine to form a new German Seventh Army between First and Second. Moreover, with the Germans digging furiously – the first load of ‘trench stores’ to reach what was becoming the Western Front arrived from Germany on 14 September – the enemy line thickened almost by the hour.¹¹⁶ The French ability to find reserves was meanwhile hindered by their need to hold Rheims, recaptured on 12 September, but subjected to devastating bombardment in the days that followed; the damage done to its famous cathedral, outside which stands the statue of Joan of Arc, would cause as much discredit to the invaders as the sack of Louvain a month earlier. What troops were available Joffre was forming into a new Second Army on his outer wing, under the fiery General de Castelnau. It was composed at the outset of corps taken from the Sixth, First and former Second Armies, most released by the stabilisation of the front in Lorraine and Alsace.

Joffre’s object, not yet fully formulated, was to deploy across the rear of the Germans’ thickening front on the Chemin des Dames and so to regain possession of the northern departments, rich in agriculture and industry, lost to France during August. While from 14 September Sir John French was ordering his troops to entrench wherever they occupied ground on or above the Aisne, Joffre was seeking means for this new manoeuvre. On 17 September he instructed his armies to ‘keep the enemy under threat of attack and thus prevent him from disengaging and transferring portions of his forces from one point to another’.¹¹⁷ Three days earlier, Falkenhayn, the new German Chief of Staff, had likewise ordered counter-attacks along the whole front with a similar object. Both commanders had grasped that opportunity in the campaign in the west now lay north of the active battlefront, in the hundred-mile sweep of territory standing, denuded of troops, between the Aisne and the sea. Whoever could find an army to operate there, without weakening his grip on the entrenched zone, might still outflank the enemy and so triumph.

There was an army in the region. It was the Belgian, hanging grimly on to the ‘national redoubt’ in the entrenched camp at Antwerp, to which it had retreated in the third week of August. King Albert, acting as Commander-in-Chief, was keenly aware of the damage he might do to the invader’s strategic position by operating against his rear and on 24 August had mounted a large-scale sortie from Antwerp towards Malines. The scratch force, III Reserve Corps and the Naval Division, left by German Supreme Command to contain the Belgians, proved just strong enough to block their advance and turn them back on the third day. On 9 September Albert tried again and his men advanced as far as Vilvoorde, ten miles from the outer lines of the fortresses, before being halted.¹¹⁸ There was a third, equally fruitless, attempt at an offensive on 27 September, which was also the last day of active operations between the Allies and Germans on the Aisne. Thereafter the German besiegers of Antwerp, who had been reinforced, were able to begin a deliberate reduction of the fortress, while the campaign between the Aisne and the sea took on the character of a frenzied search for the ‘open flank’ by the Allies and Germans in succession.

This passage has come to be called ‘the Race for the Sea’. A race it was; not for the sea, however, but to find a gap between the sea and the Aisne position before it was exploited by the other side. Both sides, with the line stabilising along its whole length, could economise force in the burgeoning entrenchments to send formations northward. The largest was the new French Tenth Army, commanded by General de Maud’huy and comprising the X and XVI Corps, which from 25 September onwards began to deploy beyond the River Somme on the great stretch of open chalk downland that sweeps northward above the steeper countryside of the Aisne. The army arrived in the nick of time, for the only French troops thereabouts were a scattering of Territorials and cavalry. Even as it began to deploy, however, with the object of pushing south-eastward behind the German front, an equivalent German mass was marching forward to oppose it. It consisted of three corps, the IV, the Guard and the I Bavarian Reserve, which together were to compose a new Sixth Army, some of which had marched cross-country from the Aisne, other parts having been transferred by rail to Belgium first.¹¹⁹ Falkenhayn’s plan, agreed with Bülow, was to use Sixth Army to mount an offensive westward towards the Channel, while eight of the eleven German cavalry divisions swept the Flanders coast and the besiegers of Antwerp brought Belgian resistance to a peremptory end. The outcome Falkenhayn intended was a new drive through northern France, leaving the Germans in possession of all the territory above the Somme and thus positioned to march down towards Paris from lines that outflanked the French entrenched zone between the Aisne and Switzerland.

Part of the Falkenhayn plan succeeded. At Antwerp, General von Beseler, an engineer by training, had by 27 September devised an effective scheme to crack the entrenched camp’s three lines of defences. The siege train of super-heavy guns that had reduced Liège and Namur having been transferred to his command, he began by bombarding the outermost and newest ring and then launched his infantry through the breach gained on 3 October. A British intervention temporarily stayed the crisis. On 4 October an advance guard of the Royal Naval Division, which had landed at Dunkirk on 19 September and had meanwhile roamed western Belgium, arrived in Antwerp by train.¹²⁰ In its wake appeared the First Lord of the Admiralty, Winston Churchill, thirsting for action and glory. The Royal Marines and sailors who composed the division temporarily halted the German advance. On the night of 5 October, however, Beseler’s men managed to penetrate the second ring of forts at an unguarded point and advance to the first, a cordon of obsolete redoubts erected in 1859. The German artillery quickly began to break up their antiquated masonry, forcing the Royal Naval Division and what remained of the Belgian field army to evacuate towards the westernmost corner of Belgium on the River Yser. On 10 October General Deguise, the heroic Belgian commander of Antwerp, delivered up his sword to a German colonel. He was accompanied by a sergeant and a private soldier, all that remained of the garrison still under his command.¹²¹

The two other elements of Falkenhayn’s plan foundered. Between 1 and 6 October the offensive of the new Sixth Army, whose mission was to ‘break down the weakening resistance of the enemy’ between the Somme and Flanders, was checked and defeated by the French Tenth Army; it was then and there that Foch, acting as Joffre’s deputy on the critical front, issued the celebrated order, ‘No retirement. Every man to the battle.’¹²² Finally, the great sweep of the eight German cavalry divisions, the largest body of horsemen ever to be collected in Western Europe before or since, was rapidly blunted by the appearance, west of Lille, of the French XXI Corps and its own supporting cavalry.

THE FIRST BATTLE OF YPRES

Thus, by the end of the second week of October, the gap in the Western Front through which a decisive thrust might be launched by one side or the other had been reduced to a narrow corridor in Belgian Flanders. There is one of the dreariest landscapes in western Europe, a sodden plain of wide, unfenced fields, pasture and plough intermixed, overlying a water table that floods on excavation more than a few spadefuls deep. There are patches of woodland scattered between the villages and isolated farmsteads and a few points of high ground that loom in the distance behind the ancient walled city of Ypres. The pervading impression, however, is of long unimpeded fields of view, too mournful to be called vistas, interrupted only by the occasional church steeple and leading in all directions to distant, hazy horizons which promise nothing but the region’s copious and frequent rainfall.

It was here, between 8 and 19 October, that the five corps now comprising the British Expeditionary Force arrived by train and road to sustain the Allied defence. To the BEF’s north the remnants of the Belgian army, which had managed to escape from Antwerp, had made their way along the coast to Nieuport, the town at the mouth of the Yser river that there flows into the sea; most of the marines and sailors of the Royal Naval Division had already got away to Ostend, where the British 7th Division, landed earlier, held a bridgehead until it joined the main body of the BEF near Ypres on 14 October.¹²³ On the Yser, a narrow but embanked river that forms a major military obstacle in the waterlogged coastal zone, the Belgians quickly erected barricades and laid plans to inundate the surrounding countryside if the river line were breached. Though they had arrived from Antwerp a broken army, their recovery was quick and their resistance on the Yser was to win the admiration of their Allies and the respect of the Germans. Their six divisions had been reduced in strength to 60,000 men, but they succeeded in garrisoning ten miles of utterly flat and featureless terrain and in holding most of their positions until, after the loss of another 20,000 men, King Albert decided, on 27 October, to open the sluices at the mouth of the Yser and let in the sea and flood the area. The resulting inundation created an impassable zone ten miles long between Nieuport and Dixmude.¹²⁴

South of Dixmude the line of the Yser and the Ypres canal was held by a brigade of French sailors, stalwart regulars of the Fusiliers-marins, then by Territorials and cavalry as far as Langemarck, on the outskirts of Ypres. From Langemarck southward the arriving British had pegged out an advanced line that ran in a circle around Ypres towards the low ridge of higher ground at Passchendaele and then southward again across the River Lys to the La Bassée canal. The length of their line was thirty-five miles, to hold which Sir John French had available six infantry divisions, with one in reserve, and three cavalry divisions, unsuitable as cavalry was, given its low scales of artillery and machine-gun equipment, for defensive operations. The only reinforcements on which he could count were another infantry division, the 8th, some additional regular cavalry and volunteer horsed yeomanry and, en route from India, the advance guard of four infantry and two cavalry divisions of the Indian Army. These, composed of British and Indian units in a ratio of one to three, though they included a high proportion of hardy Gurkhas, were scarcely suitable for warfare in a European winter climate against a German army.¹²⁵ Weak in artillery and without experience of high-intensity operations, their arrival did not promise any enhancement of the BEF’s offensive capacity.

Yet, at the outset of what would swell into the First Battle of Ypres – in which Indian units would fight gallantly and effectively both in defence and attack – Field-Marshal French still preserved hopes of mounting an attack that, in company with the French armies, would carry the Allies to the great industrial centre of Lille and thence to Brussels.¹²⁶ His hope was shared by Foch, who now commanded the northern wing of the French armies and had convinced himself that the enemy could not find the strength to hold what he still believed was an open front on the coastal plain. Both deluded themselves. Falkenhayn, the new head of OHL, not only disposed of the relocated Sixth Army, with its eleven regular divisions, and of Beseler’s III Reserve Corps, which had conquered Antwerp, but of an entirely new collection of war-raised formations, eight divisions strong.

These belonged to a group of seven Reserve Corps, numbered XXII–XXVII, raised from volunteers who had not previously undergone military training. Because Germany had needed to conscript only 50 per cent of the annual class of men of military age to fill the ranks of the peacetime army, (France had conscripted 86 per cent), a pool of five million men aged from twenty to forty-five was available to Germany for war service.¹²⁷ Of those the best were students exempted while they pursued their studies. They had responded to the outbreak by volunteering in huge numbers, together with high-school boys preparing for university, and other young men ineligible for the draft. The later-to-be-famous writer Ernst Jünger, who had just completed his school-leaving certificate, fell into the second category; Adolf Hitler, an Austrian citizen living in Munich, into the third. Jünger, after waiting for three days at the recruiting office, managed to find a place in the 44th Reserve Division.¹²⁸ Hitler, who had written a personal appeal to the King of Bavaria, was eventually embodied in the 6th Bavarian Reserve Division.¹²⁹ The recruits received two months’ training, under sergeants who were mostly schoolmasters recalled to the colours, and then left for the front.¹³⁰ Of these thirteen new divisions, two went to Russia, one to the front in Lorraine, ten to Flanders. It was those which, in the third week of October, would open the assault on the BEF between Langemarck and Ypres.

The battle that ensued raged almost continuously from early October, while the British and French were still attempting to push forward round the imagined German flank, until late November, when both sides accepted the onset of winter and their own exhaustion. Geographically it divided into four: a renewed offensive by Beseler’s corps against the Belgians on the coast, nullified by the inundations; an attempt by the French under Foch to drive north of Ypres towards Ghent, deep inside Belgium, an over-optimistic project checked by the Germans’ own offensive; the battle of Ypres itself, between the BEF and the German volunteers; and, to the south, a defensive battle conducted by the right wing of the BEF against the regular divisions of the German Sixth Army. Fighting on the three latter sectors merged effectively into one battle, so confused was combat and so unrelenting the German effort. British survivors were content to say they had been at ‘First Ypres’, a battle honour that denoted both a crucial success and the destruction of the old regular army.

Arriving in stages from the Aisne, II Corps on 10 October, III Corps on 13 October, the BEF began by pressing forward east of Ypres towards the ridges that swell some five miles beyond. The names of these low heights – Passchendaele, Broodseinde, Gheluvelt, Messines – were to recur, had the first assailants but known it, throughout the four coming years of war and to resound with menace. As the British arrived, so did fresh German corps to meet them, the XIV on 15 October, then the VII and XIX, on 19 October the XIII Corps. Under pressure, the British fell back. The British IV Corps, composed of the 7th Division and 3rd Cavalry Division, was driven close to the ancient ramparts of Ypres. The arrival of I Corps, commanded by General Douglas Haig, on 20 October secured Ypres itself, but that exhausted the army’s strength on hand; reinforcements from the empire, including the Indians, were all that were promised, and they were as yet only on their way. It was on 20 October that a general German offensive began against the whole front from the La Bassée canal in the south to the estuary of the Yser in the north, twenty-four divisions against nineteen, though the latter total included the six terribly weakened Belgian. The real contest was between fourteen German infantry divisions against seven British, with three British cavalry divisions fighting as infantry, and a collection of French sailors, Territorials and cavalry holding the river line between the British and the Belgians on the sea.

The line was held by the superiority of the British in rapid rifle fire. In artillery they were outgunned more than two to one, and in heavy artillery ten to one. In machine guns, two per battalion, they were equal with the enemy. In musketry, still quaintly so called in the BEF, they consistently prevailed. Trained to fire fifteen aimed rounds a minute, the British riflemen, of the infantry and cavalry alike, easily overcame the counter-fire of the attacking Germans who, coming forward in closely ranked masses, presented unmissable targets.¹³¹ That the British were defending, the Germans attacking might be thought to explain the extraordinary disparity in casualties suffered during the October and November fighting around Ypres – 24,000 British dead to 50,000 German – but it does not. The BEF’s trenches, at best hasty scratchings three feet deep, at worst field ditches, both frequently knee-deep in rain- or groundwater, were as yet unprotected by barbed wire. At the wettest places the defenders crouched behind sandbag mounds or brushwood barricades. In the absence of strong physical barriers to hold the enemy at a distance, it was the curtain of rifle bullets, crashing out in a density the Germans often mistook for machine-gun fire, that broke up attacks and drove the survivors of an assault to ground or sent them crawling back to cover on their start lines. ‘Over every bush, hedge or fragment of wall floated a thin film of smoke’, wrote the German official historian, ‘betraying a machine gun rattling out bullets’ – mistakenly.¹³² The smoke was the signature of individual British soldier’s marksmanship.

By the end of October the wider German offensive had failed, at enormous cost, particularly to the German volunteer corps. At their cemetery at Langemarck today, beyond a gateway decorated with the insignia of every German university, the bodies of 25,000 student soldiers lie in a mass grave; others lie in threes and fours under headstones inscribed to Volunteer Schmidt and Musketeer Braun. Dominating the hecatomb are sculptures by Käthe Kollwitz, herself a bereaved parent of 1914, of a mother and father mourning their lost son.¹³³ They represent tens of thousands of bourgeois Germans whom this phase of the battle, the ‘Kindermord bei Ypern’, the Massacre of the Innocents at Ypres, disabused of the belief that the war would be short or cheap or glorious, and introduced to the reality of attrition, of mass death and of receding hope of victory.

This brutal disillusionment was the work of the last Tommy Atkinses, working-class, long-service regulars, shilling-a-day men of no birth and scanty education. They shared nothing of the mystical patriotism of their German enemies who ‘had left lecture rooms and school benches [to be] melted into a great inspired body, [longing] for the unusual, for great danger . . . [and] gripped [by war] like an intoxicant’.¹³⁴ Their patriotism was to the little homeland of the regiment, their first loyalty to barrack-room friends. ‘After a while’, recalled Corporal William Holbrook of the Royal Fusiliers, separated from his platoon during confused fighting, ‘I came across some more of our fellows and one officer . . . Once we’d got together and were deciding what to do, a German officer came crawling through the bushes. When he saw us he said, “I am wounded” – perfect English . . . [our officer] said to him, “You shouldn’t make those bloody attacks, then you wouldn’t get wounded.” It gave us a laugh! Anyway we bandaged him up, waited on there and shortly afterwards [our officer] was killed by a stray bullet, so we had no officers then. All you could hear was some firing going on, but I didn’t know where the devil I was really.’ Holbrook found a friend, ‘name of Cainici, he was a London Italian, he was a real Cockney he was, I used to like him’, took shelter with him from shelling, dug a shrapnel ball out of his friend’s knee when he was hit, saw him off to the rear, then crawled off to look for ‘a better place’, found a dying German, tended him, saw him die, ‘covered him over with leaves and twigs, anything I could scoop up just there’ until, eventually, when he could ‘hear where the firing was [and] knew which direction [to go], crawled back’ to rejoin his unit.¹³⁵ Holbrook’s Cockney matter-of-factness – the Royal Fusiliers was a London regiment – epitomises the spirit of the old British Expeditionary Force, whose soldiers died in their thousands at Ypres not because of an ideal of self-sacrifice but because it was expected of them and, in any case, there was no alternative.

On 31 October Falkenhayn renewed the offensive on a narrower front, astride the road that leads from Menin, on the higher ground the Germans occupied, to Ypres. The attack was mounted by the specially assembled Group Fabeck, named after its commander,which consisted of a mixture of regular and volunteer corps, six divisions in all. Pressing down into the low ground, through a belt of vegetation the British would continue to call ‘woods’ – Polygon, Shrewsbury, Nuns’ Wood – long after the trees had disappeared, the Germans secured territory everywhere, and at the height of the attack broke through at Gheluvelt. Their thrust was repelled by the hasty assembly of bits and pieces of broken and exhausted battalions, Worcesters, Gloucestershires, Welch, Queen’s, 60th Rifles, Loyals, Sussex, Northamptonshires, Gordon Highlanders, Oxfordshire and Buckinghamshire Light Infantrymen, and some Royal Dragoons fighting on foot. The German history speaks of ‘the enemy reserves being too strong’ and of the British ‘bringing up two new divisions’.¹³⁶ The reality was of tiny parcels of tired men stopping gaps in the line, stuffing fresh clips into their Lee-Enfields and firing ‘mad minutes’ at the onpressing field-grey ranks. The arrival of some French units, begged by French from Foch, thickened the defence, but the crucial sector was held by British rifle fire.

The Germans renewed their offensive on 11 November, by their calculation the twenty-second day of a battle ‘in which death had become a familiar comrade’.¹³⁷ The point of pressure was Nuns’ Wood (Nonnenboschen), just north of the Menin road and only four miles from Ypres itself. Already the magnificent Gothic buildings of the ancient wool town, the Cloth Hall, the Cathedral, the merchant weavers’ houses, were falling into ruin under the weight of heavy German artillery fire. The outlying country, too, was taking on the pockmarked, denuded look that would characterise its landscape for years to come. Its villages and farmsteads were broken by shelling, the little chateaux of the Flemish nobility already stood roofless and forlorn; a direct hit on Hooge chateau, two miles from Ypres, had killed many of the staff of the British 1st and 2nd Divisions on 31 October.¹³⁸ Hooge was the target of a concerted attack by the Prussian Guard and the German 4th Division on 11 November and the battle raged all day. The initial assault by the 1st Foot Guards, premier regiment of the German army, was stemmed by a collection of cooks and officers’ servants from the 5th Field Company, Royal Engineers. Later part of the 2nd Battalion Oxfordshire and Buckinghamshire Light Infantry, a few dozen strong, counter-attacked and drove the 1st and 3rd Foot Guards back whence they came.

Fighting around Ypres would flicker on until 22 November, the date chosen by the official historians to denote the First Battle’s termination. The British survivors, whose unwounded numbers were less than half of the 160,000 which the BEF had sent to France, were by then stolidly digging and embanking to solidify the line their desperate resistance over the preceding five weeks had established in the face of the enemy. The French, too, were digging in to secure the territory for which they had fought both north and south of the city. At best the line ran a little more than five miles to the eastward; elsewhere it stood much closer. Everywhere the Germans held the high ground, dominating the shallow crescent of trenches the British, who were to be its guardians for most of the coming war of attack and defence, would call ‘the Salient’. Its winning had cost uncountable lives, French as well as British. The Germans, ‘whose vanguards had known in the plains of Flanders life and purpose for the last time’, had lost even more heavily.¹³⁹ At least 41,000 of the German volunteers, the Innocents of Ypres, had fallen outside its walls.

They represented but a fraction of all the dead of the Battles of the Frontiers, of the Great Retreat, of the Marne, of the Aisne, of the ‘Race to the Sea’ and of First Ypres itself. The French army, with a mobilised strength of two million, had suffered by far the worst. Its losses in September, killed, wounded, missing and prisoners, exceeded 200,000, in October 80,000 and in November 70,000; the August losses, never officially revealed, may have exceeded 160,000. Fatalities reached the extraordinary total of 306,000, representing a tenfold increase in normal mortality among those aged between twenty and thirty; 45,000 of those under twenty had died, 92,000 of those between twenty and twenty-four, 70,000 of those between twenty-five and twenty-nine.¹⁴⁰ Among those in their thirties, the death toll exceeded 80,000. All deaths had fallen on a male population of twenty million and more particularly on the ten million of military age. Germany had lost 241,000, including 99,000 in the 20–24 age group, out of a male population of thirty-two million.¹⁴¹ Belgium, out of 1,800,000 men of military age, had suffered 30,000 dead, a figure that was to recur with gruesome consistency in each succeeding year of the war.¹⁴² The total was the same as that for British deaths, with the difference that the British dead had almost all belonged to the regular army and its reserve of time-expired volunteer soldiers; those outside its ranks who had died, citizen soldiers of the few Territorial Force regiments, such as the London Scottish, which had reached Ypres before the end of the battle, and sepoys of the Lahore and Meerut Divisions, were few in number.¹⁴³ Their casualties would soon rise grievously, for the Indians held long stretches of the line throughout the coming winter, suffering casualties of a hundred per cent in some battalions before the year of 1915 was out, while it was the arrival of the Territorial Force in strength in 1915 that alone allowed the British army to sustain its share of the war effort in France and to participate in the offensives mounted by Joffre in the Artois and Champagne sectors of the Western Front.¹⁴⁴

The prospect of any offensive, either by the Allies or the Germans, looked far away as winter fell in France at the end of 1914. A continuous line of trenches, 475 miles long, ran from the North Sea to the mountain frontier of neutral Switzerland. Behind it the opposing combatants, equally exhausted by human loss, equally bereft of resupplies to replace the peacetime stocks of munitions they had expended in the previous four months of violent and extravagant fighting, crouched in confrontation across a narrow and empty zone of no man’s land. The room for manoeuvre each had sought in order to deliver a decisive attack at the enemy’s vulnerable flank had disappeared, as flanks themselves had been eaten away by digging and inundation. The hope of success in frontal attack had temporarily disappeared also. The experience of the French in Alsace and Lorraine in August, of the British in the Aisne in September, of the Germans in Flanders in October and November had persuaded even the most bellicose commanders that offensives unsupported by preponderant artillery would not overcome and, for the meanwhile, the artillery of all armies was short of guns and almost wholly without ammunition; at the end of the First Battle of Ypres British batteries were limited to firing six rounds per gun per day, scarcely enough to disturb the parapets of trenches opposite and wholly inadequate to support infantry in an advance against machine guns.¹⁴⁵ A sort of peace prevailed.

The war in the west had come full circle. In the four months between mobilisation and the stabilisation of the front, it had moved from hostility without action to hostility with quietus, with an intervening passage of intense aggression. Hindsight supplies a strong sense of similarity between the campaign of 1914 and that of 1870. Both had begun with French attacks in Lorraine towards the Rhine. Both had developed as German counter-offensives which resulted in grievous French defeats. Both had continued with a German advance to the outskirts of Paris, which failed to secure victory in the face of revived French resistance. Both had culminated in each side constructing entrenched positions too strong to be carried by sudden assault and in the attacker’s decision to wait out events until the defender’s powers were overcome by the pressure of events. The comparison fails beyond that point. In 1870 the Germans succeeded in surrounding the capital and consigning the French field armies of the interior to haphazard and uncoordinated local operations. In 1914 the French army had ridden out defeat in the field, sustained its cohesion, driven the invader from the environs of the capital, won a sensational defensive victory and dictated that a war of entrenchment would be fought not in the heart of the country but at the periphery of the national territory. In 1870 German armies roamed northern, central and western France at will. At the end of 1914 the French army still controlled seventy-seven of the republic’s ninety departments, remained firm in spirit, potentially strong in material force, and was supported by a great imperial and maritime power determined to see through the ordeal of an alliance war until the invader was defeated. These were the conditions that would assure Germany would enjoy no repetition of the quick and easy victory it had won forty-three years earlier.