The First World War

by John Keegan

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.

Measured from Memel on the Baltic to Czernowitz in the Carpathians and from Nieuport in Belgium to the Swiss border near Freiburg, the line of earthworks stretched for nearly 1,300 miles. Barbed wire, an invention of American cattle ranchers in the 1870s, had begun to appear, strung in belts between the opposing trenches by the spring. So, too, had underground shelters, “dugouts” to the British,

and support and reserve lines to the rear of the front. In essence, however, the new frontier was a ditch, dug deep enough to shelter a man, narrow enough to present a difficult target to plunging artillery fire and kinked at intervals into “traverses,” to diffuse blast, splinters or shrapnel and prevent attackers who entered a trench from commanding more than a short stretch with rifle fire. In wet or stony ground, trenches were shallow, with a higher parapet to the front, built of earth, usually sandbagged. The drier and more workable the soil, the less need for supporting “revetments” of timber or wattle along the internal trench walls, and the deeper the dugouts; these, which began as “scrapes” in the side of the trench nearest the enemy, excavated thus to protect the entrance from incoming shells, developed quite soon into deep shelters, approached down staircases; the “stollen,” thirty feet or more deep, eventually excavated by the Germans into the chalk of Artois and the Somme, would prove impervious to the heaviest bombardment.

Yet there was no standard trench system. The pattern varied from place to place, front to front, the design depending upon the nature of the terrain, the ratio of troops to space—high in the west, low in the east—tactical doctrine and the course of the fighting which had caused the line to re...

In the west, by contrast, no man’s land was usually two to three hundred yards wide, often less, in places only twenty-five. Intense trench fighting could even produce an “international” barbed wire barrier, mended by both sides. Barbed wire had become plentiful by the spring of 1915, though entanglements, strung on wooden posts, later on screw pickets which could be fixed without noisy hammering, were still quite narrow. The dense belts fifty yards deep were a development of later years. To the rear of the front line, the British made a practice of digging a “support” line, separated from the first by two hundred yards and usually a sketchier “reserve” line four hundred yards further back. Connecting these lines, and kinked also by traverses, ran “communication” trenches which allowed reliefs and ration parties to reach the front under cover, all the way from the rear. Diagrammatically, the layout would have appeared quite familiar to any siege engineer of the eighteenth century: “parallels,” connected by saps.2 Any diagrammatic neatness, however, quickly disappeared, as trenches were abandoned because of flooding, exposure to enemy view, or loss to the enemy in combat.

The British, hurried to Ypres in October 1914 to stop the open gap in the Western Front, had got below ground level wherever and as best they could. Shelter pits, which one man could dig at the rate of one cubic foot of earth removed in three minutes, or enough to give

him cover in half an hour, became trenches when joined up.4 More often, the first shelter was an existing ditch or field drain; when deepened, or as rain fell, these ready-made refuges filled with water and proved habitable only at the expense of great labour or not habitable at all,

Theirs had been a deliberate strategy of entrenchment, as reported by the commanders of the pursuing French formations which were stopped in sequence as they advanced from the Marne.

From Rheims to the Swiss frontier, therefore, the Germans had already succeeded in carrying out Moltke’s order of 10 September to “entrench and hold” the positions reached after the retreat from the Marne, while from the Aisne northwards towards the English Channel a line of entrenchments was being dug piecemeal as the series of short-range outflanking movements failed one after the other. The last of these stages of the “Race to the Sea” ended in episodes of ditch-deepening, scraping, scrabbling, pumping and rough field-carpentry, as described by the officers of the 2nd Royal Welch Fusiliers, all under the fire of an enemy dug into higher, drier ground on the ridges that overlook Ypres and its surroundings from the east.

The British, who had learnt recent and important lessons in South Africa, where the Boers had taught them at the Modder and Tugela rivers the value of complicating any trench system, compensated for the inferiority of their overlooked positions in Flanders by digging in duplicate and triplicate, an insurance both against sudden infantry assault and artillery damage. The Germans, who had last dug earthworks around Paris in 1871 and otherwise derived their knowledge of trench warfare from indirect studies of the Russo-Japanese War, had a different doctrine. In two instructions, issued on 7 and 25 January 1915, Falkenhayn ordered that the western armies were to fortify the front in a strength sufficient to assure that it could be held with small numbers against attack by superior forces for a long time.9 Falkenhayn’s insistence on this point derived from his pressing need to find reinforcements from France and Belgium for the campaign in the east, where the demands of the fighting in Masuria and the battles on the Vistula, together with the necessity to prop up the Austrians in Galicia, exerted a growing drain on his resources.

They represented over one-tenth of his western field army and a third of its peacetime Prussian formations, those most counted upon for their offensive qualities.

The front line was to be the main line of resistance, built in great strength, to be held at all costs and retaken by immediate counter-attacks if lost. Secondary positions were to be dug only as a precaution. Some German generals, including Prince Rupprecht, commanding Sixth Army opposite the British in Flanders, objected even so to the digging of a second line, believing that the front troops would hold less firmly if they knew there was a fall-back behind them. Not until 6 May 1915 was a binding order issued by OHL for the whole of the German front to be reinforced by a second line of trenches, two to three thousand yards to the rear.10 By then, however, the main line of resistance was becoming a formidable fortification. In the chalk of Artois and the Somme, on the heights of the Aisne and the Meuse, the German infantry were burrowing deep beneath

the surface to construct shell-proof shelters. Concrete machine-gun posts were appearing behind the trenches, which were heavily walled with timber and iron. Parapets were thick and high, trench interiors floored with wooden walkways. Militarily, the German front grew in strength week by week. Domestically, it was even becoming comfortable. Electric light was appearing in the deeper dugouts, together with fixed bedsteads, planked floors, panelled walls, even carpets and pictures. Rearward from their underground command posts ran telephone lines to their supporting artillery batteries. The Germans were settling in for the long stay.

The eighty French departments not directly touched by the war were largely agricultural. The ten occupied by the Germans contained much of French manufacturing industry and most of the country’s coal and iron ores. If only to prosecute the war, it was urgent that they

Whereas his German opponent, however, wanted to turn the whole of the Western Front into a passive sector, so as to find troops for the east, Joffre wanted to subdivide it into passive and active sectors, the former providing attack forces for the latter. Geography dictated where the subdivisions should fall. The wet and the hilly sectors—Flanders in the north, the heights of the Meuse and the Vosges in the south—should be passive. The active sectors should be those intervening, particularly those shouldering the great German salient in the Somme chalklands at Arras and in Champagne near Rheims.

Two offensives in those sectors in December proved premature. The First Battle of Artois, 14–24 December, ended without any result. The Winter Battle in Champagne, which began on 20 December, dragged on, with long pauses, until 17 March, costing the French 90,000 casualties and bringing them no gain in territory at all. There was also local and quite inconclusive fighting further south, in the Argonne, near Verdun, in the St. Mihiel salient, and around Hartmannweilerkopf in the Vosges, a dominant point to which both sides sent

their specialised mountain troops, Jäger and Chasseurs Alpins, to engage in fruitless as...

In the first, he ordered that the active sectors were to consist of strongpoints sited to cover the ground to the front and to the flanks with fire. The passive zones in between were to be garrisoned only with lookouts, and to be heavily wired but held by fire from the active zones. Across the whole front, active and passive, two belts of wire were to be constructed, twenty yards or so apart and about ten yards deep, with gaps for patrols to pass through. Behind the line of strongpoints there were to be secondary positions with shell-proof shelter for counter-attack companies.11 A survey of the fronts of the eight French armies revealed that most of the work Joffre required had already been done. In his second January letter he therefore stipulated that the front be strengthened by the digging of a second line some two miles to the rear, resembling the first, as a precaution against local break-ins. Such work had already been completed in the Verdun and Rheims sectors. Joffre added the general instruction that fronts were to be held as thinly as possible, to economise manpower and avoid casualties, and that local commanders should avoid pushing outposts too close to the enemy’s positions, a practice he thought wasteful of lives.

land” by redigging trenches closer to the enemy’s and staging frequent trench raids. The first trench raid appears to have been mounted on the night of 9/10 November 1914 near Ypres by the 39th Garwhal Rifles of the Indian Corps.

The event set a precedent of which the British were to make a habit and which the Germans were to copy. The French, despite their long experience of tribal warfare in North Africa, never found a similar enthusiasm for these barbaric flurries of slash and stab. Disposing of many more field guns in their corps reserves than either the British or Germans did, they preferred to dominate their defensive fronts from a distance with artillery fire, for which, after the solution of the shell shortage of the winter of 1914–15, they were amply supplied.

That remained apparently unalterable by the effort of the armies on either side until, in March 1917, the Germans voluntarily surrendered the central Somme sector and retired to shorter, stronger, previously prepared lines twenty miles to the rear. Until then the Western Front stood the same, month after month, for almost every yard of its length, running in a reversed S shape for 475 miles from the North Sea to the Swiss border.

There the Swiss militia army, fully mobilised for war, surveyed the termini of the opposed trench barriers from neutral territory.

The Vosges was such a front, and was accepted to be so by both French and Germans, who held it with second-rate divisions, reinforced by mountain infantry who occasionally disputed possession of the high

points. Indeed, south of Verdun, neither side was to make any major effort between September 1914 and September 1918 and this stretch, 160 miles long, became “inactive.” Elsewhere, the Argonne proved unsuited to offensives as, for different reasons, did the Flemish coastal zone. The former was too broken, stream-cut and tree-choked, the latter too waterlogged for the delivery of attacks that required firm, unobstructed avenues of advance for a successful conclusion. Shelling into the Argonne threw the woodland into a jungle of broken vegetation; in the sea-level fens of Flanders, shelling quickly reduced the soil to quagmire. In the centre, the heights of the Aisne and the Meuse, though they were both to be contested in great battles, too much favoured the defender for offensive effort to be profitable. It was therefore only on the dry chalklands of the Somme and Champagne that attacks offered the promise of decisive success. The former stood below the wet Flemish country, the latter above the mountainous forest zone of the Meurthe and Moselle. They were separated from each other by the high ground of the Aisne and Meuse, the bulge in the front to which they formed the shoulders. Military logic therefore required that it was at those shoulders that the attackers should make their major efforts and defenders be best prepared to withstand an assault.

The trace of the railway network explains how these outcomes came about. Early in the campaign of 1914 the Germans took possession of the Metz–Lille line, running north-south within their area of conquest. The French, on the other hand, retained control of the Nancy–Paris–Arras line facing it. The latter is closer to the line of engagement than the former, and that proximity explains why the French were better able than their enemy to deliver reserves to the crucial point in time to win one battle after another.

The real reasons were quite different. France, as the victim of Germany’s offensive of August 1914, and the major territorial loser in the outcome of the campaign, was bound to attack. National pride and national economic necessity required it. Germany, by contrast, was bound to stand on the defensive, since the setbacks she had suffered in the east, in its two-front war, demanded that troops be sent from France to Poland for an offensive in that region.

The real

outcome of 1914 was not the frustration of the Schlieffen Plan but the danger of a collapse of the Central Powers’ position in Eastern Europe.

Falkenhayn, resented the transfer of every one. He believed that the war had to be won by making the major effort in the west. There the French army was recovering from its losses of the opening campaign—thirty-three new divisions were forming—while French industry was gearing up for a war of material. The British were creating a whole new army of volunteers, while training their peace-time militia, the Territorial Force, for active service; together these would produce nearly sixty divisions, besides those from Canada and Australia which were hastening across the Atlantic and Pacific to the motherland’s aid. Of these figures Falkenhayn did not have exact intelligence but his impression of the gathering of a huge reinforcement was accurate enough. It would shortly double the force opposing the Germans on the Western Front, while they were already reaching the limits of expansion open to them from their manpower

potential. The number of their divisions could be increased by reducing the infantry strength of each, counting on artillery and machine guns to make good the consequent diminution of firepower, a measure already in hand. The absolute limit of troop availability already stood, nevertheless, in sight.

By early 1915 the BEF was large enough to be divided into two Armies, First and Second, the Territorials were reaching France in strength and the first of the volunteer divisions of Kitchener’s New Armies were beginning to appear also. Soon the British would be able to take over stretches of the line from their ally and find a striking force to mount offensives on their own initiative.

An early plan to make a major effort on the Belgian coast, with the Royal Navy supporting a combined Anglo-Belgian army, foundered on Admiralty warnings that its light ships could not stand up to German coastal artillery and that its battleships could not be risked in such confined waters.19 Plans to use troops against the Austrians were shown to be equally unrealistic. Militarily weak though Austria-Hungary was, geographically it was almost unapproachable by a maritime power.

It turned on the rail communications which supported the German armies in the field. There were three systems that led back across the Rhine into Germany. The southernmost was short and easily defended. That left the two systems that supplied the Germans holding the great salient between Flanders and Verdun. If either, or preferably both, could be cut, the Germans within the salient would be obliged to fall back, perhaps creating once again those conditions of “open warfare”

agreed during January that the correct strategy during 1915 was for offensives to be mounted at the “shoulders” of the salient, in the north against the Aubers and Vimy ridges which stood between the Allies and the German railways in the Douai plain behind, in the south against the Champagne heights which protected the Mézières-Hirson rail line. The attacks would, in theory, converge, thus threatening the Germans in the great salient with encirclement as well as disruption of their supplies.

There was to be a spring offensive, jointly British and French in Flanders and Artois, French alone in Champagne.21 Indeed, this first agreement was to set the pattern for much of the Allied effort on the Western Front throughout the war. The pattern was to be repeated in the coming autumn, during 1917 and, finally with success, in 1918. Only in 1916 would the Allies attempt something different, in the offensive against the centre of the great German salient to be known as the battle of the Somme.

All the contributing factors that were to bedevil success in trench offensives for much of the war were present, both the functional and structural. The functional were to be cured, in time, the structural persisted, even after the development and large-scale deployment of the tank in 1917. Among the functional were inadequacy of artillery support, rigidity of planning, mispositioning of reserves and lack of delegation in command. Among the structural were the relative immobility and total vulnerability to fire of advancing infantry and absence of means of speedy communication between front and rear, between infantry and artillery and between neighbouring units.

The plan was simple. Neuve-Chapelle, a ruined village, twenty miles south of Ypres in the Artois sector into which the British had been extending their position as fresh troops arrived in France during the winter, was to be attacked on 10 March

The front of attack was about 8,000 yards, behind which 500 guns had been assembled, to fire a stock of 200,000 shells, mainly light-calibre, into the enemy trenches, the barbed wire protecting them and certain strongpoints in the rear.

There was also to be a “barrage”—the term was French, meaning a dam or a barrier—of bursting shells fired behind the German trenches, parallel to the front of at...

reinforcements reaching their stric...

The bombardment, which opened at seven o’clock in the morning, took the Germans by complete surprise. That was an achievement, rarely to be repeated; even more of an achievement was First Army’s success in having assembled the leading waves of an attack force of sixty thousand men within a hundred yards of the enemy in complete secrecy, a fact scarcely ever to occur again. The defenders, belonging to two infantry regiments and a Jäger battalion, about one-seventh in strength to their assailants, were overwhelmed. Their wire had been extensively cut, their front trench destroyed. When the British infantry assaulted at five past eight, they were not opposed and within twenty minutes a breach 1,600 yards wide had been opened in the German line. The makings of a victory, local but significant, had been won.

Orders, however, now required that they should wait for a second time. The commander of the battalion in the centre, 2nd Rifle Brigade, managed to send back a message requesting permission to disregard the order and continue the advance. Surprisingly—there were no telephone lines and this was the pre-radio age—it was received; even more surprisingly an answer was returned from brigade headquarters speedily enough to affect the situation, wholly for the worse. Permission to move forward was refused.

Falkenhayn’s tactical instruction of 25 January had laid down that, in the event of an enemy break-in, the flanks of the gap were to be held and immediately reinforced, while reserves were to hurry forward and fill the hole. That was what was beginning to happen.

Meanwhile, according to plan, fresh British battalions were crowding into the gap opened by the leading waves. By ten o’clock, “roughly nine thousand men [were squeezed] into the narrow space between Neuve-Chapelle village and the original British breastwork [where] they lay, sat, or stood uselessly in the mud, packed like salmon in the bridge pool at Galway, waiting patiently to go forward.” Fortunately, the German artillery batteries within range had little ammunition available.

The British artillery, which had ample stocks, could not rapidly be informed of the deteriorating situation, one of the structural defects contributing to failure. Without radio, communication depended on flag signals or runners, the first usually obscured, the second slow and vulnerable. At half past eleven a bombardment was organised against the 11th Jäger’s machine-gun positions, and an officer and sixty-three men came out to surrender, having killed about a thousand British soldiers. Precise and timely bombardment of their and other strongpoints could not be attempted because the gunners could not be informed.

All the while the local German commanders, junior but determined and well-trained officers, were hurrying reserves to the flanks by bicycle or on foot. By contrast, and here the functional contribution to failure was at work, the British junior officers were passing their observations of the local situation, as the plan required, back up the chain of command so that authority could be granted for any alteration of the all-defining plan they requested. Behind the battle zone, telephone lines speeded communications but it was still painfully, indeed lethally, slow. “The Corps commander in some room five miles or more from the battle had to make a decision on the flimsiest and often false information, and the necessary orders had then to travel back, along the same chain, to be considered and written out in greater detail at each stage (divisional headquarters, brigade headquarters, battalion headquarters), till finally they reached the front-line companies.”

What all this meant, in terms of the actual rather than planned timetable in this particular trench battle, was that between nine o’clock in the morning, when the German line had been broken and a way forward lay open for the taking, and the writing of firm orders to exploit the success at ten to three in the afternoon, nearl...

and runner, another three hours were lost. The time the advance was resumed on the ground was bet...

Dark was drawing in and so were the German reserves. The flanks of the break-in had been secured before midday. By nightfall fresh German troops, hurried forward from battalions in rearward support, were filling the open gap and bending their flanks forward to join up with the positions at the edges which had never been lost.

When, on the morning of 12 March, the attack did go in, it was immediately stopped with heavy German losses. The British front-line commanders had used the pause imposed by the mist the day before to consolidate their foothold and site twenty machine guns in commanding positions.