EIGHT
WAR AT SEA
IF THE WAR OF 1914 was not a war which the armies of Europe were ready to fight, that was not so with Europe’s great navies. The armies, as the opening campaigns had proved, were technically equipped to solve certain easily perceived problems, in particular how to overcome the defences of modern fortresses, how to move vast numbers of men from home bases to the frontiers and how to create impassable storms of rifle and field-artillery fire when those masses came into contact with each other. They were quite unequipped to deal with the unperceived and much more critical problems of how to protect soldiers from such fire storms, how to move them, under protection, about the battlefield, indeed how to move them at all beyond railhead unless on their feet, and how to signal quickly and unambiguously between headquarters and units, between unit and unit, between infantry and artillery, between ground and the aircraft with which, almost fortuitously, the armies had so recently provided themselves.
The failure of the generals of 1914 had largely been a pre-war failure. They had had the wit to adapt the technologies ready to hand, particularly that of Europe’s many-branched rail network, to their purposes. They had lacked the wit to perceive the importance or potentialities of new technologies, among which the internal combustion engine and wireless-telegraphy, as radio was then called, would prove the most important; they had, indeed, lacked altogether the wit to perceive the problems to which such new technologies would be the solution. No such charge could be laid against the admirals of the years before 1914. With foresight they had divined the significance of the developing technologies likely to affect their service and had applied them to it with exactitude. Admirals have traditionally had a reputation as seadogs and salthorses, with little ability to see far beyond the bulwarks of their ships and little desire to change anything within them. Nineteenth-century admirals are commonly thought to have opposed transition from sail to steam as fiercely as generals opposed the abolition of scarlet coats. Nothing could be further from the truth. When the admirals of the Royal Navy were persuaded that sail had had its day, they displayed a ruthless lack of sentimentality for the beauty of pyramids of canvas. The sailing navy was abolished almost overnight after the Crimean War, in which steam gunboats had devastated wooden walls.Warrior, the Royal Navy’s first steam ironclad of 1861, was not an experimental but a revolutionary ship, which surpassed several intermediate stages of naval design in a single leap.1 Palmerston, seeing her at anchor among the old men-of-war in Portsmouth harbour, described her as a ‘snake among the rabbits’ and the successors of the admirals who had commissioned her would build new snakes whenever they judged the old had lapsed into rabbit status. Naval design changed with almost bewildering rapidity between 1860 and 1914, from broadside to central battery to turret arrangement of guns, from all-round to ‘citadel’ to ‘armoured deck’ arrangement of protection, from wrought-iron to case-hardened to composite quality of armour, from piston to turbine engine power, from coal propulsion to oil.
The changes came faster and faster, as admirals accepted the significance of the new technologies civilian industry was creating and took stock of the evidence presented by the clash of such technologies in engagements between navies in non-European waters: the Spanish-American War of 1898, the Russo-Japanese War of 1904. In 1896 the Royal Navy, still the world leader, was launching battleships of 13,000 tons, armed with four 12-inch guns and capable of a speed of eighteen knots from piston engines fired by coal. By 1913 its most modern battleships, of the Queen Elizabeth class, displaced 26,000 tons, mounted eight 15-inch guns and achieved speeds of twenty-five knots from turbine engines fired by oil.2 The key intermediate ship between these two designs had been theDreadnought of 1906, which gave its name to all subsequent classes of ‘all-big-gun-ships’, so called because they dispensed with the previous clutter of secondary, small-calibre weapons and concentrated their armour around the ship-killing main armament, their magazines and their turbine engines. Dreadnought, the brainchild of Admiral Sir John Fisher, was as revolutionary as Warrior had been and the decision to build her as brave, for, like Warrior, she made all contemporary battleships obsolete, including the Royal Navy’s own. Only a nation as rich, as fiscally efficient and as committed to its maintenance of maritime predominance as Britain could have taken such a risk and only a navy as technically adaptable as the Royal Navy could have seen the need to do so. The inspiration was not wholly British. Italian naval architects, always at the forefront of their profession, had anticipated the conception of the all-big-gunship. They did not nerve themselves to put conception into practice. The appearance of Dreadnought, and of a stream of similar and improved sister-ships appearing in rapid succession after her launch, forced all advanced navies – the French, the Italian, the Austrian, the Russian, the United States, the Japanese, the German – to do so. Between 1906 and 1914, Dreadnoughts went down the ways of the world’s shipyards in ever-increasing numbers, to fly the flags of every major country, and of many which had not before aspired to maritime position. Turkey placed orders for Dreadnoughts in Britain and a Latin American naval race broke out between Argentina, Brazil and Chile which, lacking the resources to build large warships themselves, distributed their commissions between American and British yards. The Dreadnought in those years became a symbol of a state’s international standing, whether or not it served an objective national purpose.
Competition – and competition was fierce between the British and American yards, which operated in the free market and sold abroad whenever they could – ensured that design met the highest standards and followed the most recent innovation. The ships building in Britain for foreign navies in 1914 – Almirante Latorre for Chile, Reshedieh for Turkey, Rio de Janeiro for Brazil – were of the most advanced class. The Admiralty had no hesitation in buying all three into British service in August 1914, when, asCanada, Erin and Agincourt, they immediately joined the Grand Fleet.Agincourt, which mounted twelve 14-inch guns, was the most heavily armed ship in any European navy. German Dreadnoughts were better protected than their British equivalents, having thicker armour and more elaborate internal division into small, water-tight spaces, which limited the danger of flooding, but mounted guns of smaller calibre. The latest class of the neutral United States’ Dreadnoughts, Oklahoma and Nevada, achieved a remarkable compromise between speed, hitting-power and protection, while Britain’s two Queen Elizabeth class (three more were building) clearly represented the newer generation of even faster, better armed and armoured battleships.
Marginal differences in design between Dreadnoughts would prove significant in battle, often startlingly so, for a chink in the armour might be lethal. Modern naval warfare was unforgiving. Steel ships, unlike wooden walls, could not be repaired in action (trivial damage excepted), while the huge loads of volatile high-explosive they carried in their magazines threatened them with disintegration if they received a deep hit. What, nevertheless, is striking about the Dreadnoughts is, first, their similarity to each other, second, their ‘state-of-the-art’ modernity. Admirals supported naval architects in striving to provide their ships with the very latest equipment on offer, from range-finding equipment (in which the German optical industry gave the High Seas Fleet a distinct advantage) to mechanical computers for calculating bearing and elevation in directing guns.3 The armies of 1914 may not have been very efficient battle-winning organisations; the Dreadnought fleets were as efficient as they could be made within the constraints of available technology.
If there were any major technical deficiency in the equipment of fleets, it lay in their signalling arrangements.4 Navies had enthusiastically embraced the new science of wireless-telegraphy (radio) and its introduction had enormously enhanced their ability to communicate, both strategically and tactically. It allowed the disposition of fleets to be altered over very long distances and, by radio direction-finding, for the position of enemy ships which broke wireless silence to be established with a high degree of accuracy. It also revolutionised the business of scouting and reconnaissance, by a battlefleet’s attendant minor warships. Before the advent of wireless, signalling between scouts and scouts, and scouts and fleet, was limited by the height of masts above the visual horizon and by conditions of visibility within the radius thus defined, in practice twenty miles at most. After the introduction of wireless, scouts could communicate for hundreds, sometimes thousands of miles, and flagships directly and instantaneously with the humblest reconnaissance vessel and vice versa. It was the light cruiser Glasgow, the only survivor of the disaster of Coronel, that saved the latecomer Canopus from destruction and it was its wireless transmissions that set in motion the trans-equatorial chase which eventually brought Spee’s squadron to defeat at the Falklands.
Naval wireless telegraphy in 1914 had, however, one critical drawback. It as yet did not transmit voice signals, only morse code. As a result, there was ‘a period which includes the time taken to write out the [message], to transmit it to the wireless office, to code it, to signal it, de-code it aboard the receiving ship, write it out and transmit it to the bridge’, a period estimated by Admiral Jellicoe, commanding the Grand Fleet, to be ‘ten minutes to a quarter of an hour’.5 This lapse in ‘real time’ was unimportant when strategic signals were being transmitted and received. It was crucial in action, when densely ranked fleets had to manoeuvre simultaneously at the admiral’s command. Wireless was therefore judged to be ineffective as a means of tactical signalling, which continued to be done, just as in Nelson’s time, by flag hoist. An admiral, wishing to turn his battleline towards or away from the enemy, would instruct the flag lieutenant to ‘make’ the appropriate flag hoist, which the yeomen of signals on the bridge of each of his subordinate ships was expected to identify by naked eye or telescope and announce to the captain. The procedure required first the hoist, perhaps copied by a ‘repeater’ ship nearer the front or rear of the line, and then the display of an ‘executive’ flag, which ordered the manoeuvre the hoist specified when dropped. The system had worked admirably at Trafalgar, when the speed of the British approach towards the Franco-Spanish line was five knots and the distance between the leading and last ship of a formation was two miles at most. Dreadnought fleets, manoeuvring at twenty knots in formations six miles long, were controlled by flag hoist only with great difficulty, as signallers struggled to identify tiny squares of coloured cloth, obscured by the smoke of funnels and guns, at distances of a thousand yards or more.
In retrospect, it seems that it might have been possible to simplify wireless-telegraphic procedure, by dispensing with encoding and by locating a receiver on the bridge, to be used in tactical circumstances when the dangers stemming from interception, since they must occur in ‘real time’, would be minimised. It was not done, perhaps because, through one of those lapses into ‘backwardness’ so characteristic of the armies of 1914, the ‘culture’ of the signal flag had fleets in its grip. The lapse was common to all navies. Unfortunately for the Royal Navy, the High Seas Fleet had overcome the signalling difficulty to a degree by simplifying its system of manoeuvre, allowing large changes of direction and alignment to be achieved by fewer hoists than the Grand Fleet employed. That would prove greatly to its advantage during the battle of Jutland.
Otherwise, in technical circumstances as remarkable for their modernity as for their similarity, only one shortcoming was notable, and that affected both the navies locked in the war’s critical confrontation, the British and the German. Neither had adequate reconnaissance resources. Traditionally, fleets had deployed, forward of what were now known as their ‘capital’ units, the battleships, and their attendant light craft, a screen of intermediate ships fast enough to find the enemy and strong enough then to disengage before suffering crippling damage. In the decades before the First World War, they had acquired the name of ‘cruisers’. Admiral Fisher, the sponsor of the Dreadnought concept, had conceived the idea that the function of the cruiser would in future best be served by a vessel as large as the battleship and as well armed, but faster, its superior speed being achieved by dispensing with much of the battleship’s armour. By 1916 the Grand Fleet included nine of these ‘battlecruisers’ and the High Seas Fleet, since the Germans had followed the British initiative, five. Of traditional cruisers neither had any number and those in service were old, slow and weak in armament and armour. That would not have mattered had the admirals restricted their employment to the appropriate reconnaissance role and deterred the battlecruiser squadron commanders from exposing their ships to punishment they were not built to withstand. Unfortunately for both navies, the belief had arisen that battlecruisers should, in extension of their scouting function, engage in action with the enemy’s battleships when found, using their main armament to ‘fix’ them while their own supporting battleships came up, and trusting to their superior speed to escape damage in the interim. ‘Speed is protection’, Fisher had argued. His battlecruisers were indeed faster than any battleship then afloat by a margin of as much as ten knots (British battlecruiser Queen Mary = 33 knots, German battleship Kaiser = 23.6 knots). As battle would prove, however, speed was not protection against modern naval guns, firing 12-inch or heavier shells out to ranges of 17,000 yards. The illusion that it might prove so had caused navies to spend the money that could have bought dozens of smaller but effective cruisers on a handful of battlecruisers no better at doing their work and wholly unsuitable to challenge battleships even in the preliminaries of fleet action. The Royal Navy went into battle at Jutland in 1916 with but a handful of traditional cruisers, none up to their work, swarms of light cruisers too weak even to show themselves to the enemy’s heavier ships and an advanced guard of battlecruisers which would suffer terrible and pointless loss before the main action was joined.
The clash of battlefleets at Jutland took place on 31 May 1916 and the following night. There had been two earlier engagements, near Heligoland and the Dogger Bank in August 1914 and January 1915, but neither had engaged the main battlefleets against each other. The battle of the Heligoland Bight, at the entrance to Germany’s North Sea naval bases, came about through the determination of the commanders of the destroyers and submarines at Harwich, the British port nearest the German bases, to intercept the enemy’s offshore patrols and inflict damage. Tyrwhitt, commanding what came to be called Harwich Force, and Keyes, commanding the Eighth Submarine Flotilla, were aggressive officers whose thirst for action won the support of Churchill, First Lord of the Admiralty, and, through him, the promise of intervention by three of Admiral Sir David Beatty’s battlecruisers if opportunity for success offered. In a confused daylight encounter on 28 August, a misty day in the Heligoland Bight, the British at first succeeded in sinking only one destroyer. When German reinforcements appeared, however, Beatty’s battlecruisers came forward and sank three enemy cruisers before safely disengaging.6
This small victory greatly heartened the British but, while prompting the Germans to thicken the defences of the Heligoland Bight with minefields and standing patrols of heavy and light vessels, including submarines, it did not deter them from further action. In an effort to treat the British as they had been treated themselves, they sent fast ships to bombard the North Sea port of Yarmouth on 3 November and, on 16 December, Scarborough, Whitby and Hartlepool, on this second occasion with most of the High Seas Fleet’s Dreadnoughts following. The Grand Fleet sent a squadron to intercept but failure of intelligence prevented its making contact, fortunately, for it would have been outnumbered. In the second of the early naval encounters of the war, at the Dogger Bank, intelligence served the Royal Navy better. Its interception and cryptologic services, the latter accommodated in the Admiralty Old Building (Room 40 or 40 OB), was far superior to the German and those who worked there had benefited at the war’s outset from three extraordinary pieces of luck. In August the German light cruiser Magdeburg grounded in Russian waters and its signal books, with the current key, were recovered and sent to England. In October, the merchantman code, seized from a German steamer interned in Australia, also reached London. Later the same month a third codebook, used by German admirals at sea and jettisoned by the senior officer of a group of German destroyers recently sunk in a small action off the Dutch coast, was dredged up accidentally in the nets of a British fishing boat and brought to the Admiralty.7 These three documents opened the secrets of most German naval signalling to the officers of 40 OB, allowing them to read enemy transmissions often in ‘real time’, that is, as quickly as they were decoded by the intended recipients. In an uncanny foreshadowing of the cryptological history of the Second World War, the German naval staff swiftly recognised that the movements of their ships were becoming known to the enemy but ascribed that intelligence success not to signal insecurity but to espionage. Their suspicions fixed on the Dutch fishing boats trawling the shallow waters of the Dogger Bank, in the central North Sea, which they decided were British-manned, flying false flags and wirelessing their observations to the Admiralty.
Believing that they could turn such reports to their advantage while revenging themselves for the Heligoland defeat, the German naval staff decided to sail the battlecruisers of the High Seas Fleet to the Dogger Bank and lay a trap for their opposite numbers. On 23 January, the First and Second Scouting Groups sortied, only to encounter heavy opposition as they approached the Dogger Bank at dawn next morning. Beatty’s battlecruiser squadrons, alerted by 40 OB, were in position and, as the weaker and less numerous German formations emerged into visibility, they found themselves assailed by armour-piercing salvoes. The semi-battlecruiser Blücher was overwhelmed and capsized, the Seydlitz almost suffered a fatal internal explosion, averted only by flooding the magazines, and the two scouting groups, turning tail, escaped by the skin of their teeth. Examination of the damage caused to Seydlitz after she limped home revealed that far too much high-explosive, in the form of bags of propellant for the main armament, had been taken out of its flash-proof cases in the ammunition-handling chambers under the turret than was safe or necessary. Damage to the turret had ignited the charges there and the flash, travelling down the turret-trunk, detonated the loose charges below and started a fire next to the magazine. Warned in time of the dangers of bad practice, the German navy instituted much stricter procedures for the handling of its propellant, which was in any case more stable than the British equivalent. Beatty’s Battle Cruiser Fleet, as it became known immediately after the Dogger Bank, continued to keep loose propellant ready in quantity between magazine and turret, with results that would prove disastrous at Jutland.8
After January 1915, the High Seas Fleet kept close to its home bases for most of the next eighteen months and pondered its strategy. The Fleet’s submarine operations could bring returns, and so could minelaying, by U-boat or surface ship. The sinking of HMSAudacious, a brand-new Dreadnought, by a mine laid by an armed merchant cruiser in October 1914, caused the British Admiralty even greater anguish than the torpedoing of the ancient cruisers, Aboukir, Hogue and Cressy by U-9 in the ‘Broad Fourteens’ off Holland in September. Submarine warfare, however, by the rules of commerce-raiding, which stipulated that an attacker must give a merchant ship warning before sinking and make provision for the escape of the crew and passengers, could cause little interruption to trade, while exposing U-boats to rapid retaliation; ‘unrestricted’ submarine warfare, on the other hand, when U-boats torpedoed without surfacing, could all too easily lead to diplomatic incidents, through the sinking of wrongly identified neutrals, or to diplomatic disaster, as it did in May 1915, when U-20 sank the Lusitania. The loss of this huge British liner, and that of the lives of 1,201 passengers, of whom 128 were American, almost caused the United States to break off relations with Germany. Negotiations smoothed over the repercussions of the atrocity but the German naval staff imposed strict limitation on the operations of its submarines in the aftermath. The British merchant fleet continued to lose between fifty and a hundred ships a month to submarine attack during 1915 but could maintain supply to the home country nonetheless.9 Meanwhile, the Grand Fleet and its subordinate squadrons and flotillas of cruisers, destroyers and submarines sustained a blockade of Germany that denied it all trade with the world beyond Europe and which was extended, by British, French and Italian naval dominance in the Mediterranean, against Austria and Turkey. The ‘central position’ of the Central Powers, a strategic posture ordained by military theorists to be one of great strength, had been reduced to one of infirmity, perhaps disabling weakness, by the constriction of an all-encircling blockade. Germany’s sailors, during 1915, racked their brains to think of a way out.
They had brought their predicament upon themselves, aided and abetted by political and dynastic leaders who should have known better. The geography of the German-speaking lands, however configured into states, denies the Germans maritime power. The geography of the German empire of 1914 narrowed its access to the high seas to the short North Sea coastline between Denmark and Holland. From it, the way to the nearest ocean, the Atlantic, lay through waters easily choked by an enemy. Westward, the English Channel, only nineteen miles across at the narrows, had long lain under threat of closure by the Royal Navy; in more recent years the threat of closure by mine-barriers, though the British did not densely mine the Channel narrows until 1916, promised to render the western route impermeable. Northward, from the estuaries of the Ems, Jade, Weser and Elbe, the High Seas Fleet had a clear run into the North Sea from ports easily protected against a close British blockade. Once at sea, however, it faced a passage of 600 miles up the North Sea, between Great Britain and Norway, before it could break out into the ocean, and then only through a series of gaps, between the Faroes, Iceland and Greenland, easily kept under surveillance by light cruiser squadrons. The likelihood of the High Seas Fleet clearing the North Sea undetected or unassailed diminished, moreover, with every mile it steamed, because early in the century it had become the Royal Navy’s war plan to transfer its capital units on mobilisation from its English to the Scottish ports, Rosyth near Edinburgh and Scapa Flow in the Orkney Islands, leaving its light units of cruisers, destroyers and submarines to maintain an intermediate blockade off the Heligoland Bight, which would give early warning of a German sortie. On that warning the Grand Fleet would sail south at speed, making it an operational probability that a major fleet action would be joined long before the enemy had neared the waters from which it could stage an oceanic break-out. Admiral Fisher summed up the German predicament in an exultant epitome of the critical maritime geography to King George V: ‘with the great harbour of Scapa Flow in the North and the narrow straits of Dover in the south, there is no doubt, Sir, that we are God’s chosen people’.10
The Germans had never blinded themselves to the intrinsic geographical weakness of their position or the strengths of the British. They had toyed unrealistically with means of widening their access to the North Sea, by persuading or forcing their Dutch, Danish and Norwegian neighbours to grant them bases, and continued to consider means of doing so even after the war had begun; during 1915, Commander Wolfgang Wegoner of the German naval staff wrote a series of papers advocating the occupation of Denmark, the establishment of a protectorate over Norway and, at some future date, the acquisition of ports in France and Portugal.11 Perception of the value of the submarine, as a carrier either of mines or torpedoes, was also amplified after the war’s outbreak, by the success of the very small U-boat force against both warships and merchantmen. In the main, however, the German Admiralty, having early taken a view of the nature of the fleet it should build and operate best to serve its maritime purpose, persisted with its long-laid strategic policy. That is simply stated. Germany, within the fiscal limits imposed by the maintenance of a very large army, could not outbuild Britain in capital ships. It should, therefore, confine itself to confronting the Royal Navy with ‘risk’ – the risk that its traditional determination to command the seas might lead to a wearing down of its preponderant strength through small actions, and by mine and submarine, which would heighten the danger that, in unforeseeable conditions, the Grand Fleet might find itself at a disadvantage to the High Seas Fleet during one of its offensive sorties. After much debate about ‘risk’ strategy, the Kaiser issued a final war directive to the German navy on 3 December 1912, which stipulated that its ‘chief war task’ should be ‘to damage the blockading forces of the enemy as far as possible through numerous and repeated attacks day and night, and under favourable circumstances to give battle with all the forces at your disposal’.12
German naval operations in home waters during 1914 and 1915 had adhered strictly to the 1912 directive, and achieved some of its purposes. Heligoland and the Dogger Bank had been defeats but had indeed damaged the blockading force, since Tiger and Lionwere hit at the Dogger Bank, Lionhard enough to have to be towed back to harbour. The sinking of Audacious had been achieved at the cost of a single mine. The potentiality of the U-boat in fleet warfare had also been demonstrated by the sinking of the pre-Dreadnought Formidable in the Channel on 1 January 1915 by U-24. In early 1915 Sir John Jellicoe, commanding the Grand Fleet, was seriously concerned that German Kleinkrieg (small war) successes, combined with the need to disperse units of the Grand Fleet to secondary theatres, was whittling away its superiority. In November the ratio of British to German Dreadnoughts had fallen to 17:15 (it had been 20:13 in August) and of battlecruisers to 5:4.13 Germany was continuing to launch capital ships, moreover, and though Britain was doing likewise, it had calls on its resources, particularly in the Mediterranean, that Germany did not have to meet.
By the spring of 1916, the balance had swung back in Britain’s favour. The situation in distant waters, thanks to the destruction of the German raiding cruisers, the termination of the Gallipoli campaign and the addition of the Italian to the Franco-British fleet in the Mediterranean, no longer exerted a drain on the forces at home. New classes of Dreadnoughts had come into service, particularly the fast Queen Elizabeths and, while Germany had also added to the High Seas Fleet, the Grand Fleet had recovered a clear superiority. In April 1916 it comprised thirty-one Dreadnoughts and ten battlecruisers, the High Seas Fleet only eighteen Dreadnoughts and five battlecruisers. British superiority in light cruisers and destroyers was also large (113:72) and, while the Grand Fleet’s lack of effective heavy cruisers persisted, it was unencumbered with any of the pre-Dreadnoughts which, for want of weight in the battle line, the Germans continued to count as part of their capital strength.14
On paper, therefore, the risk in an active implementation of Germany’s ‘risk’ strategy was too large to be accepted; prudence counselled passivity and a reversion to traditional ‘fleet in being’ policy, by which a navy justified its existence simply by causing an opponent to watch its harbours. German naval pride forbade such inactivity. The navy was in Germany the junior, not, as in Britain, the senior service, and many of its officers felt it had to fight whatever the odds, if it were to keep the esteem of the German people, particularly at a time when the German army was pouring out its blood for the nation. A new and aggressive admiral, Reinhard Scheer, had taken command of the High Seas Fleet in January 1916 and a memorandum written to him by one of his captains, Adolf von Trotha, epitomises the attitude of the offensive school to which both belonged: there can be, he wrote, ‘no faith in a fleet which has been brought through the war intact . . . we are at present fighting for our existence . . . In this life and death struggle, I cannot understand how anyone can think of allowing any weapon which could be used against the enemy to rust in its sheath.’15
Scheer quickly resumed the policy of taking the fleet to sea in the search for action. He made two sorties in February and March 1916 and four in April and May; in the April sortie he succeeded in reaching the English east coast and, in a repetition of the raids of 1914, bombarding Lowestoft. The demonstration, timed to coincide with the Irish nationalist Easter Rising, of which Germany had foreknowledge, caused dismay in Britain but emphasised once again that, while the Grand Fleet at Scapa Flow closed the exit from the North Sea, High Seas Fleet operations must be limited to tip-and-run against targets close enough to home for it to beat a retreat before the Royal Navy’s heavy units could steam south and intervene. Even the Battle Cruiser Fleet, which was now located at Rosyth, one of the ports of Edinburgh, was anchored too far north to catch German raiders without considerable forewarning.
At the end of May, however, such forewarning it and Jellicoe’s battleship squadrons got. Scheer had been preparing another sortie for some time, on a scale elaborate enough to surprise Beatty’s battlecruisers if they came sufficiently far south. An encounter with the British Dreadnoughts he did not plan. Room 40’s decryption of his signals, however, gave Jellicoe word of his movements, so that, by the time Scheer had cleared the Heligoland Bight, not only were Beatty’s battlecruisers at sea and heading south from Rosyth, so too were the Scapa Flow battleships. On the morning of 31 May, over 250 British and German warships were steaming on convergent courses to a rendezvous, unanticipated by the Germans, off the Jutland coast of Denmark. Among the host of light cruisers, destroyers and submarines which made up the bulk of each side’s forces, it was the presence of the big ships which promised decision. They included, on the British side, twenty-eight Dreadnoughts and nine battlecruisers, on the German sixteen Dreadnoughts and five battlecruisers. Jellicoe’s arrangement of his fleet attached his four newest battleships, of the fast Queen Elizabeth class, to the six battlecruisers of Beatty’s Battle Cruiser Fleet, deployed ahead of the Grand Fleet’s Dreadnoughts as an advance guard, with orders to bring the Germans to action. Scheer’s fleet, advancing fifty miles behind the First Scouting Group of five battlecruisers, included six Deutschland-class pre-Dreadnoughts, which he appears to have brought with him for sentimental rather than military reasons.16Their lack of speed, five knots less than that of his Kaiser-class battleships, made them a liability in a contest of rapid closing and opening of the range at which main armament took effect.
Scheer’s decision to take the whole of the High Seas Fleet into the North Sea, something never ventured before, was predicated on the belief that the British would not have foreknowledge of his movements. Room 40’s success in penetrating his signals therefore laid the basis for a great victory, since, with Jellicoe’s and Beatty’s ships advancing to an encounter likely to occur too far from port for Scheer to escape to safety during daylight hours, he risked the danger of being overwhelmed, or cut off from his line of retreat by superior force. Jellicoe’s initial advantage was compromised at an early stage, however, by a procedural failure at the Admiralty in London. Mistrusting Room 40’s ability to make operational judgements, the responsible staff officer asked a veiled question and concluded from the answer that Scheer’s battleships were still in harbour. He transmitted that false information to Jellicoe who, in consequence, and in order to conserve fuel, limited his speed southward while allowing Beatty and the battlecruisers to forge ahead. Room 40 had correctly informed the naval staff that Scheer’s wireless call sign could still be located in his home port; since the question had not been asked, however, the intelligence officers did not say that, on going to sea, it left its harbour call sign behind and adopted another. At the critical stage of the preliminaries of what would prove the largest naval encounter of the war, therefore, Jellicoe was making less than best speed to a junction with the enemy, while his reconnaissance fleet of battlecruisers was hurtling to an early and potentially disastrous encounter with a superior force.
Jutland, as the impending battle would be called (by the British; to the Germans it would, contentiously, be known as ‘the victory of the Skaggerak’) promised not only to be the largest naval encounter of the war but of naval history thus far. No sea had ever seen such a large concentration of ships or of ships so large, so fast and so heavily armoured. The High Seas Fleet, which had cleared the Heligoland Bight in the early morning of 31 May, consisted of sixteen Dreadnoughts, six pre-Dreadnoughts, five battlecruisers, eleven light cruisers and sixty-one destroyers. The Grand Fleet and the Battle Cruiser Fleet, which had left Scapa Flow and Rosyth the previous evening, included twenty-eight Dreadnoughts, nine battlecruisers, eight armoured cruisers, twenty-six light cruisers, seventy-eight destroyers, a seaplane-carrier and a minesweeper.17 Both sides also had submarines at sea, in the hope that the enemy might present a target to a lucky shot. Scheer’s plan, indeed, was predicated on the chance of drawing the British into a U-boat trap by showing his battlecruisers off Jutland. No such chance came, however, nor were any of the navies’ associated aircraft or airships able to play a role.18 Jutland, in consequence, was to be both the biggest and the last purely surface encounter of main fleets in naval history. The spectacle they presented never left the memory of those who took part, the densely ranked columns of battleships, grey against the grey water and sky of the North Sea, belching clouds of grey smoke from their coal-stoked boilers, the flash of white from the bows of the faster light cruisers and destroyers in attendance, as all pressed onward to action. So large was the number of ships hurrying forward that the more distant formations blurred into the horizon or were lost to sight in the play of cloud and rain squall on the observer’s field of vision.
Jutland and the war in the North Sea
Jutland is the most written about battle of naval history and the most disputed between scholars. Each segment, almost every minute of the two fleets’ engagement have been described and analysed by historians, official and unofficial, without any reaching agreement about exactly what happened or why, or whether, indeed, the outcome was a British or a German victory. That it was a British victory of some sort is not now denied. That it was less than a decisive victory is not denied either. It was the disparity between British expectations of victory and the success actually achieved that led to the detailed dissection of the battle’s events and the controversy that persists to this day. The Royal Navy, undefeated in a major fleet action since Trafalgar, sailed for Jutland in the sure belief that, should a junction of battlefleets ensue, another Trafalgar would occur. The inconclusiveness of the event has continued to haunt the mind of the Royal Navy ever since.
Yet Jutland is not, in outline, complex at all. It falls into five phases: in the first Beatty’s Battle Cruiser Fleet made a ‘run to the south’ on encountering the weaker German battlecruiser force; then a ‘run to the north’ when, on meeting the German Dreadnoughts, it turned back to draw them into Jellicoe’s Grand Fleet; then two encounters between the Dreadnoughts, broken by a German ‘turning away’ as heavier British firepower told; and finally, after the German Dreadnoughts had sought escape from destruction, a night action in which the light forces of both sides sought to inflict crippling damage by torpedo attack.19
In the first phase Beatty’s Battle Cruiser Fleet passed through Scheer’s patrol line of U-boats without loss to arrive within fifty miles of his opposite number Hipper’s First Scouting Group, undetected. Chance then directed them towards each other. Their light forces diverted to investigate a neutral merchant ship, found each other and brought the two groups of battlecruisers into contact. Fire was opened and, because of bad British signalling, the German told more heavily. It fell, moreover, on ships defective in armoured protection and in prudent ammunition-handling. FirstIndefatigable, then Queen Mary suffered penetrations, which set off fires in handling-rooms where too many intrinsically unstable propellant charges were lying ready to be sent into the turrets. Both blew up and sank. Beatty’s superiority in numbers instantly disappeared.
The appearance of his four supporting fast battleships reversed the imbalance but then they and the surviving battlecruisers of the Battle Cruiser Fleet found that they had run down on to the main body of German Dreadnoughts. When they turned back towards Jellicoe’s Grand Fleet, the ‘run to the north’ began. During it the 15-inch gunfire of the fast battleships inflicted heavy damage on the following Germans – the unlucky Seydlitz, so heavily hit at the Dogger Bank, was hit again – so that Scheer’s battleline was in disarray when his Dreadnoughts unwittingly fell under the fire of Jellicoe’s a little after six o’clock in the evening. They were to inflict one more act of destruction, when Invincible was blown up, through the same causes that had devastated Indefatigable and Queen Mary. Then the concentration of British superior weight of shell proved so overwhelming that Scheer hastily ordered a retreat and disappeared into the gathering gloom of a misty North Sea evening.
There might have ended, inconclusively, an already unsatisfactory encounter. Scheer, however, then decided to turn back, perhaps to come to the assistance of the damaged light cruiser Wiesbaden which had been left behind, perhaps because he judged that he could pass astern of Jellicoe’s fleet as it continued its advance towards the Heligoland Bight, while he made his escape through the Skaggerak into the Baltic. Jellicoe, however, once again reduced speed, with the result that the German Dreadnoughts, heading north-east, encountered the British heading south-east, and steering to pass their rear so as to cut them off from safety. At the moment of encounter moreover, the British were deployed in line abreast, the Germans in line ahead, a relative position, known as ‘crossing the enemy’s T’, that greatly favoured the British. More of their guns could be brought to bear than could those of the German fleet, ranked one ship behind the other, which thus also presented an easier target. Ten minutes of gunnery, in which the Germans suffered twenty-seven hits by large-calibre shells, the British only two, persuaded Scheer to turn away again into the dark eastern horizon, leaving his battlecruisers and lighter ships to cover his retreat in a ‘death ride’. The torpedo threat they presented caused Jellicoe to turn away also – for which he has ever afterwards been reproached – and, by the time he turned back, Scheer had put ten miles between his Dreadnoughts and the pursuit. Many German ships remained to cover Scheer’s flight, including his squadron of vulnerable pre-Dreadnoughts, and in a series of dusk and night actions they suffered losses. So, too, did the British cruisers and destroyers that remained in contact. By the morning of 1 June, when Scheer had his fleet home, he had lost a battlecruiser, a pre-Dreadnought, four light cruisers and five destroyers. Jellicoe, though remaining in command of the North Sea, had lost three battlecruisers, four armoured cruisers and eight destroyers; 6,094 British sailors had died, 2,551 German.
The disparity in losses caused the Kaiser to claim a victory. Scheer, his sailors and ships had undoubtedly acquitted themselves well, while the battle had revealed serious defects in British ship design and tactical practice, particularly in inter-ship and inter-squadron signalling. Beatty had failed to report promptly and accurately in the encounter stage, gunnery had not been directed effectively during the Dreadnought engagements.20 Nevertheless, Jutland was not a German victory. Though the High Seas Fleet had lost fewer ships than the Grand Fleet, it had suffered more damage to those that survived, so that in the aftermath its relative strength in heavy units fell from 16:28 to 10:24. In those circumstances it could not risk challenging the Grand Fleet for several months nor, when it resumed its sallies from port, dared it venture outside coastal waters.21 Contrary to conventional belief, Jutland was not the German fleet’s last sortie, nor its last action. There was an encounter between German Dreadnoughts and British battlecruisers near Heligoland on 17 November 1917, while the High Seas Fleet steamed as far as southern Norway on 24 April 1918. It had accepted the verdict of Jutland nevertheless, pithily summarised by a German journalist as an assault on the gaoler, followed by a return to gaol.22 Inactivity and discontent would eventually lead to serious disorder among the crews of Scheer’s surface ships, beginning in August 1917 and culminating in full-scale mutiny in the last November of the war. After 1 June 1916, Germany’s attempt to win a decision at sea would be conducted exclusively through the submarine arm.
OFFENSIVES ON THREE FRONTS
In the early summer of 1916, Germany saw as yet no need to reverse the policy of restricting U-boat operations it had adopted, for diplomatic reasons, the previous year, nor did the Allies apprehend the deadly danger that such a reversal would bring. Their thoughts were concentrated on the great offensives they jointly planned to deliver in the west and east, offensives which they believed would, after eighteen months of stalemate in France and Belgium, a year of defeats in Poland, and six months of frustration in Italy, bring them decisive victories. On 6 December 1915, representatives of the Allied powers met at French headquarters at Chantilly to agree plans. Joffre presided but had no power to impose a single strategy, only to encourage co-ordination. In that he succeeded. It was easily decided that the minor fronts, in Salonika, Egypt and Mesopotamia (though there events were suddenly to take a turn for the worse), should not be reinforced. On the major fronts, by contrast, the Russians, the Italians and the British and French bound themselves to mount attacks so timed as to prevent the Central Powers from transferring reserves between theatres and with all the forces available to each army.
The Allied forces had grown considerably since the beginning of trench warfare. Italy, industrially and demographically the weakest of the major allies, had succeeded by early 1916 in raising its number of infantry battalions from 560 to 693 and of field artillery pieces from 1,788 to 2,068; the army in the zone of combat had grown in strength since 1915 from a million to a million and a half.23 Russia, despite the terrible fatalities of 1914–15 and the large loss of soldiers to captivity after Gorlice-Tarnow, had been able to fill the gaps with new conscripts, so that by the spring of 1916 it would have two million men in the field army. Almost all, moreover, would be properly equipped, thanks to a striking expansion of Russian industry. Engineering output increased fourfold between the last year of peace and 1916, chemical output, essential to shell-filling, doubled. As a result there was a 2,000 per cent increase in the production of shells, 1,000 per cent in that of artillery, 1,100 per cent in that of rifles. Output of the standard field-artillery shell had risen from 358,000 per month in January 1915 to 1,512,000 in November. The Russian armies would in future attack with a thousand rounds of shell available per gun, a stock equivalent to that current in the German and French armies, and its formations were acquiring plentiful quantities of all the other equipment – trucks, telephones and aircraft (as many as 222 per month) – essential to modern armies.24
In France, too, there had been a war-industrial revolution. Thanks in part to the mobilisation of women for factory work – the number employed in the metal industries rose from 17,731 in 1914 to 104,641 in July 1916 – shell output reached 100,000 per day in the autumn of 1915. Between August and December 1915, production of field guns rose from 300 to 600, while daily production of rifles in that month totalled 1,500; output of explosive had increased sixfold since the beginning of the war.25 There had not been a comparable expansion of the fighting force. Because of the small size of the national demographic base, relative to Germany’s, and high proportion of men conscripted and held in the reserve in peacetime, over 80 per cent of those of military age, France lacked the capacity to expand its field army to the extent possible in Germany or Russia, where the pre-war military intake was less than half the age group. Nevertheless, by skilful reorganisation and redeployment of soldiers to the front from employment in the rear, twenty-five new infantry divisions were formed between February 1915 and the spring of 1916. The French army of 1916 was stronger than that of 1914 by more than 25 per cent.26
The major addition to the fighting strength of the Allies, however, was British. On 7 August 1914, Lord Kitchener, on appointment as Secretary of State for War, had issued an appeal for a hundred thousand men to enlist for three years, or the duration of the war, which he believed would be long. Further appeals for ‘hundred thousands’ followed, and were met with an overwhelmingly enthusiastic response, in part because the promise was given that ‘those who joined together would serve together’. As a result men from the same small locality, workplace or trade went to the recruiting offices in groups, were attested and then went forward to training and eventually active service in the same unit.27 Many called themselves ‘Chums’ or ‘Pals’ battalions, among which the largest group was the Liverpool Pals of four battalions, largely raised from the shipping and broking offices of the city. Smaller towns supplied single battalions, like the Accrington Pals, the Grimsby Chums and the Oldham Comrades; others were raised by occupation, the Glasgow Tramways Battalion, or nationality; Newcastle-on-Tyne, the English industrial city, produced four battalions each of Tyneside Scottish and Tyneside Irish. The ‘first hundred thousand’ had included many of the pre-war unemployed. Subsequent hundred thousands – there were to be five altogether – were formed of genuine volunteers including, by January 1915, 10,000 skilled engineers, and over 100,000 each from the coal mining and the building trades. From this magnificent human resource, Kitchener was able eventually to form six ‘New’ or ‘Kitchener’ Armies, each five divisions strong, to join the army’s eleven regular divisions and the twenty-eight infantry divisions of the part-time, voluntary Territorial Force. By the spring of 1916, Britain had seventy divisions under arms, a tenfold expansion since peace, and of those twenty-four were New Army divisions on or waiting to go to the Western Front.28
It was this enormous increment in the striking power of the Anglo-French concentration in France and Belgium that allowed the French and British to promise their allies at Chantilly a continuation of their joint offensive efforts in 1916. It would, Joffre agreed on 29 December with General Sir Douglas Haig, the new commander of the BEF, take the form of a combined offensive in the centre of the Western Front. Joffre argued initially for mounting a series of preliminary assaults, in continuation of his policy of attrition. Haig, who feared that forces would be frittered away in such operations, counter-proposed an attack by the British in Flanders, to be matched by a French offensive further south, as had been tried in 1915. As a compromise, Joffre secured his agreement to a drive along the line of the River Somme, to which the British were to extend their line. As the movement would allow the French units north of the Somme to rejoin the main concentration of Joffre’s armies to the south, the two armies would then share a clear-cut boundary which, Joffre argued, should be the axis of their great offensive in the coming year. Haig, who doubted the military logic of an operation that seemed likely at best to dent the huge salient left by the failed German advance on Paris in 1914, demurred but, in the interests of Anglo-French harmony, eventually concurred.
Plans made without allowance for the intentions of the enemy are liable to miscarry. So it was to prove in 1916. While Joffre and Haig were making their dispositions for the Somme, the Italians preparing to persist in the struggle for the heights above the Isonzo and the Russians contemplating retaliation for the loss of Poland, Conrad von Hötzendorf was laying the basis for an Austrian ‘punishment expedition’ against the hated Italians from the unexpected direction of the Trentino, while Falkenhayn, who had wrongly concluded that the Russians had been beaten into submission by the series of victories beginning with Tannenberg and culminating in Gorlice-Tarnow, was devising a vast punishment expedition of his own against the French at Verdun.
Falkenhayn outlined his reasoning in a letter written to the Kaiser on Christmas Day 1915. Germany’s object, he insisted, must be to dishearten Britain on whose industrial and maritime power the Alliance rested. He therefore argued for a resumption of the unrestricted U-boat campaign. At the same time – he perhaps and rightly surmised that his call for a U-boat offensive would be refused – Britain’s continental partners should be destroyed. Italy was too unimportant to deserve a major effort against her. Russia, on the other hand, tied up German troops which could be better used elsewhere, without presenting the opportunity to strike against her a success decisive to the outcome of the war. His assessment was that, ‘Even if we cannot perhaps expect a revolution in the grand style, we are entitled to believe that Russia’s internal troubles will compel her to give in within a relatively short period. In this connection it may be granted she will not revive her military reputation meanwhile.’ What made even a weakened Russia too difficult to knock out of the war was the lack of a strategic objective: the capture of St Petersburg would have merely symbolic results; an advance on Moscow led towards the vast emptiness of the interior; while the Ukraine, though a prize of great value, was inaccessible except through Romania, whose neutrality Germany would be ill-advised to violate. Dismissing involvement in the Egyptian, Mesopotamian and Salonika fronts as irrelevancies, and accepting the British portion of the Western Front as too strong to attack, he therefore concluded that, since an offensive somewhere was necessary, because ‘Germany and her allies could not hold out indefinitely’, it must be made against France. ‘The strain on France’, he wrote, ‘has reached breaking point – though it is certainly borne with the most remarkable devotion. If we succeed in opening the eyes of her people to the fact that in a military sense they have nothing more to hope for, that breaking point would be reached and England’s best sword knocked out of her hand.’ The operational solution to his analysis was for a limited offensive at a vital point that would ‘compel the French to throw in every man they have. If they do so the forces of France will bleed to death.’29
He already had ‘the vital point’ in mind, the fortress of Verdun in a loop of the Meuse, isolated during the operations of 1914, exposed to attack from three sides, badly provided with communications to the French rear area but lying only twelve miles from a major railhead in German hands. He quickly secured the Kaiser’s agreement to what would be called Operation Gericht (Judgement) and, while a dissenting Hötzendorf proceeded to prepare his own offensive against the Italians, began to mass the divisions that would try ‘the remarkable devotion’ of the French to its limit.
1. Offensive at Verdun
Verdun had been a fortress in Roman times and its defences had been renewed many times, by Vauban in the seventeenth century, by Napoleon III and most recently in 1885, when its circle of detached forts had been duplicated with another at five miles’ distance from the small city’s centre. The new forts had subsequently been strengthened with concrete and armour but, following the collapse of Liège and Namur to German heavy artillery in August 1914, the French had lost faith in all fortifications and Verdun’s fortress guns had been dismounted and sent away for use in the field. The battle of 1914 had flowed around it but its value as a point of pivot had been forgotten in the aftermath. Verdun had become a ‘quiet sector’ and its garrison had been whittled down until, in February 1916, it consisted of only the three divisions of XXX Corps, the 72nd, a local reserve division, the 51st, also a reserve division, from Lille, and the 14th, a regular division from Besançon; the 37th Division, from Algeria, lay in reserve. Among the units of the divisions that formed the garrison the most notable were the 56th and 59th Battalions of Chasseurs à pied, notable because they had cleared the Bois des Caures north of Verdun of Germans in 1914 and had been there ever since and because they were commanded by Lieutenant Colonel Emile Driant, a local member of parliament, a constitutionally insubordinate soldier and the author of numerous sensationalist books on future warfare, of which the best known, The War of Tomorrow, foretold a great victory by France overGermany and had been crowned by the French Academy. Driant, in the Bois des Caures, commanded the foremost sector of Verdun’s defences on the east bank of the Meuse.30
The battle of Verdun
Opposite his and its neighbouring positions, Falkenhayn had assembled, during January and February 1916, a reinforcement to Fifth Army, the German Crown Prince’s, of ten divisions, including six regular, supported by an enormous concentration of artillery. Among the 542 heavy guns were thirteen of the 420 mm and seventeen of the 305 mm howitzers that had devastated the Belgian forts eighteen months earlier, and to supply them and the field and medium artillery a stock of two and a half million shells had been accumulated. The whole of the French defensive zone on a front of eight miles – one German division and 150 guns to each mile – was to be deluged with preparatory fire, so that ‘no line is to remain unbombarded, no possibilities of supply unmolested, nowhere should the enemy feel himself safe’. Falkenhayn’s plan was brutally simple. The French, forced to fight in a crucial but narrowly constricted corner of the Western Front, would be compelled to feed reinforcements into a battle of attrition where the material circumstances so favoured the Germans that defeat was inevitable. If the French gave up the struggle, they would lose Verdun; if they persisted, they would lose their army.
Operation Judgement was scheduled to begin on 10 February. Bad weather postponed it from day to day, during which growing intelligence of an impending German attack gradually brought the defenders to a better state of readiness, insufficient, nevertheless, without substantial reinforcements of guns and men, to guarantee successful resistance. On 19 February the rains stopped, next day a warm sun dried the ground, and early in the morning of 21 February the bombardment opened. All morning it raged and on into the afternoon; in the Bois des Caures, 500 by 1,000 yards square, it is estimated that 80,000 shells fell before the German infantry appeared. Only Driant’s meticulous preparation of his position left any of his men alive to fight.31
Had the Germans attacked in strength they must have overrun the devastated enemy positions on the eight-mile front, but they did not. The philosophy of the operation was that artillery would destroy the French defences, which would then be occupied by the infantry in follow-up. Driant and half of his soldiers survived until the next day, when stronger waves of Germany infantry appeared to overwhelm them. There were equivalent advances either side of the Bois des Caures. The French outer trench lines were crumbling and the defenders began to fall back, overwhelmed by fire and numbers, towards the old forts of Vaux and Douaumont. On 23 February a surviving lieutenant of the 72nd Division signalled to higher command, ‘The commanding officer and all company commanders have been killed. My battalion is reduced to approximately 180 men (from 600). I have neither ammunition nor food. What am I to do?’32 There was little that could be done in the absence of reinforcements. On 24 February, the whole of the outer trench zone was overrun, many of the defenders abandoning their positions in terror and fleeing to the rear. Only Forts Vaux and Douaumont stood as points of resistance on the forward slopes of the heights above the Meuse which, if taken, would allow German artillery observers to direct fire on to Verdun itself and the bridges across the Meuse which sustained the resistance. Then, on 25 February, Douaumont fell, taken by a lone German sergeant of the 24th Brandenburg Regiment who, blown into the fort’s moat by a near-miss, decided to explore the interior, found it occupied by only a handful of French troops and bluffed them into surrender. The news of the fort’s capture spread panic among the troops in Verdun and even the first of the reinforcements arriving to strengthen the front. Depots of food were pillaged on the word that the Meuse bridges had been prepared for demolition and retreat was imminent. Verdun seemed on the point of falling.
Had it fallen the results might have been beneficial to the French conduct of the war, for it was indeed a death trap, while the broken and wooded terrain to its rear was perfectly defensible at a cost in life much lower than the French were to suffer in and around the sacrificial city in the months to come. On the morning of 25 February, however, Joffre’s deputy, de Castelnau, who had commanded the Second Army at the Marne, arrived at Verdun, assessed the situation, and decided that the forward positions must be held. A ‘fighting general’, a romantic, a devout Catholic and a member of an ancient French military family, de Castelnau saw the fight for Verdun as a test of his country’s capacity to sustain the defence of the national territory and keep alive the hope of ultimate victory. The decision he took on 25 February was the one for which Falkenhayn might have hoped and the soldier chosen to implement it, Philippe Pétain, the opponent Falkenhayn might himself have chosen. Pétain was not a man for giving up. Taciturn and charmless, his disbelief in the doctrine of the offensive had denied him promotion in the pre-war army. At the war’s outbreak, however, his refusal to be deterred by losses had won him rapid advancement, from the colonelcy of the 33rd Regiment, in which Charles de Gaulle served as a subaltern, to, by 1916, command of Second Army. On his arrival at Verdun he telephoned the commander of XX Corps, newly arrived in reinforcement, to say, ‘I have taken command. Tell your troops. Hold fast.’
Pétain at once identified two essentials for the defence: to coordinate the artillery, of which he took personal control, and to open a line of supply. Henceforth it would be the Germans on whom fell a constant deluge of shells as they clung to the front line or made their way forward to battle through the narrow valleys beyond the Meuse. Behind Verdun, the single road that led to Bar-le-Duc fifty miles away was designated a supply route for trucks alone; 3,500 were assembled to bring forward the 2,000 tons of stores the garrison needed daily, the troops being ordered to march up and down the roadside fields. Any truck that broke down was pushed off the road, lest it interrupt the day-and-night flow of traffic. A whole division of Territorials was employed in road repairs and France was scoured for additional transport. Eventually 12,000 trucks would be used on what became known as the Voie sacrée.
A sanctified battle was what Falkenhayn had wanted France to fight. He had not counted upon the fervour the French would show. Already on 27 February, the Germans recorded ‘no success anywhere’.33 The XX ‘Iron’ Corps had come into line and its soldiers were sacrificing themselves in a desperate effort to defend every foot of ground held; among those of XX Corps wounded – and captured – that day was Charles de Gaulle. The Germans sought to overcome the resistance of the French infantry by pushing their artillery ever closer to the front, through saturated ground that demanded ever larger teams of horses to move a single gun. An immediate result was appalling casualties among the gun-teams – 7,000 horses are said to have been killed in one day – yet, despite the growing weight of bombardment, the French line would not shift. By 27 February, the Germans had advanced four miles and were within four miles of the city but no increase in offensive effort could push their front forward.
On the last day of February, Falkenhayn and the Crown Prince conferred and agreed on a new strategy. Since the narrow-front attack on the east bank of the Meuse had not achieved success, the offensive must be broadened to the west bank where, behind the heights of the Mort Homme and Côte 304, the French were hiding the artillery that flailed the German infantry struggling to reach the positions from which they could look down into Verdun itself. The terrain on the west bank was different from that on the east, open and rolling instead of broken and wooded. Falkenhayn had been advised to include it in his original assault plan, for the reason that advances there could be easily gained. So they were on the first day of the assault, 6 March, when the French 67th Division collapsed. The Germans were swiftly counter-attacked, however, the ground was regained and once again the line stuck fast. Simultaneous efforts on the east bank, in the direction of Fort Vaux, Douaumont’s neighbour, were equally ineffectual. The ruins of the village of Vaux changed hands thirteen times during March, and yet the fort itself still lay tantalisingly beyond German reach. It was, moreover, defending itself resolutely. Both the French and Germans were learning that the lessons of Liège and Namur were not as conclusive as had seemed. Fortifications, even quite antiquated fortifications, could stand up to intense and prolonged artillery bombardment and buttress trench lines, if occupied by garrisons prepared to sit out heavy shellfire and wait for assault by unprotected infantry. It was inexperience that had caused the Belgians, whom the Germans later came to respect as dogged defenders of any position they occupied, to give in; by 1916, the French had discovered that shellfire often sounded much worse than it was, had nerved themselves to sit it out and to repay the infantry attacks that followed with murderous small-arms fire.
By the beginning of April, Falkenhayn’s belief that he could win a victory of attrition without exposing his own army to comparable loss was failing. The opening attack on the narrow front east of the Meuse had been checked at the outer line of fortification. The second offensive on the west bank had faltered under fire from the heights of the Mort Homme and Côte 304. At the beginning of April it was decided to abandon the strategy of limited offensive and attack across the extent of the whole front, now nearly twenty miles wide. The operation began on 9 April and lasted four days, until the descent of drenching rain stalled all activity for the rest of the month. On the first day the Germans reached what they thought was the crest of the Mort Homme, only to find that the real summit lay just beyond their reach. The fight for the feature then resolved itself into an artillery combat. An officer of the French 146th Regiment, Augustin Cochin, spent from 9–14 April in the Mort Homme trenches without seeing a single German, ‘the last two days soaked in icy mud, under terrible bombardment, without any shelter other than the narrowness of the trench . . . The Boche did not attack, naturally, it would have been too stupid . . . result: I arrived there with 175 men, I returned with 34, several half mad . . . not replying any more when I spoke to them.’34
During May, after the bad weather relented, it was the Mort Homme that absorbed German efforts. On 8 May the French lost the true crest but clung on to the neighbouring slopes, against which the Germans picked step by step throughout the rest of the month. The final line of resistance delineated by Pétain on taking command was breached as they continued their advance but their progress was too slow to threaten the integrity of the Verdun position. Their casualties had now exceeded 100,000, killed and wounded and, though the French had suffered equally, most of the losses borne by the Germans had fallen on the same formations. While the French rotated divisions through Verdun, the Germans kept divisions in the line, making good casualties with replacements. By the end of April, forty-two French divisions had already passed through the Verdun sector, but only thirty German, and the disparity would persist.35 The German 5th Division, which attacked on the first day, was in line until the end of February, returned between 8 and 15 March and then again from 22 April until the end of May. The 25th Division was engaged from 27 February to 16 March, from 10 to 25 April and then again until 19 May. Between March and May the casualties in its infantry regiments amounted to 8,549, or over a hundred per cent of their strength.
A high proportion of losses on both sides was the result of the French policy of conducting an ‘active defence’, counter-attacking whenever possible. One opportunity that offered was at Douaumont, where carelessness detonated a German ammunition store inside the captured fort on 8 May. The vast explosion persuaded the French to venture an attempt at recapture on 22 May, and the assaulting parties succeeded in storming the fort’s outworks and scaling the exterior before they were repulsed a day later. The initiative, nevertheless, rested with the Germans who continued to attack wherever they could and at the beginning of June gathered forces for a decisive effort. They consisted of the divisions of I Bavarian, X Reserve and XV Corps, attacking side by side on a front of three miles, with one man per yard of front, supported by 600 guns. The objective was Fort Vaux, which between 1 and 7 June, the Germans first surrounded, cutting off the garrison from contact with the French rear area, and then blew up section by section. Ultimately the commander of the garrison, Major Raynal, was forced to surrender for lack of water. The attackers paid this man the honours of war and the German Crown Prince, to whom Raynal was taken, presented him with a sword to replace the one he had left behind.36
Direct command of the Verdun sector had now passed from Pétain, whose disregard for casualties troubled even Joffre, to Nivelle, an artillery expert, fluent and persuasive in manner, who had risen rapidly since the beginning of the war, due to his perfect English and winning ways with politicians. He was already improving control of the French guns, which were beginning to achieve dominance over those of the enemy and eventually swung the balance of advantage in the French favour. Meanwhile, however, the Germans sustained the offensive, gaining pockets of ground on the east bank and pushing forward to the surviving French forts of Souville and Tavannes. From Souville, ‘it was downhill all the way to Verdun, less than two and a half miles away . . . and once the fort fell into enemy hands it would be but a matter of time before the city fell into enemy hands’.37 German pressure continued unrelentingly after the fall of Vaux until on 22 June a new assault was preceded by a bombardment of ‘Green Cross’ gas – an improved form of chlorine – on the French artillery lines, which contained 600 of the 1,800 French guns at Verdun. Temporarily robbed of artillery protection, the French defence faltered before an attack by the Alpenkorps, an élite mountain division of Bavarian guard and German light infantry; among the light infantry officers was Lieutenant Paulus, the future commander of Sixth Army at Stalingrad.38 A soldier of theAlpenkorps recorded that, during the course of the successful advance that followed the bombardment, he glimpsed the roofs of Verdun from the Souville heights. He was, however, probably mistaken. In the afternoon, the German advance petered out in the broken ground around the fort and, in the summer heat, thirst attacked the soldiers in the foremost positions gained. No water could be got up from the rear and, as night fell, the Alpenkorps gave up its effort to press onward.
That day, 23 June, marked both the high point and crisis of the Verdun offensive. About twenty million shells had been fired into the battle zone since 21 February, the shape of the landscape had been permanently altered, forests had been reduced to splinters, villages had disappeared, the surface of the ground had been so pockmarked by explosion that shell hole overlapped shell hole and had been overlapped again. Worse by far was the destruction of human life. By the end of June over 200,000 men had been killed and wounded on each side. The losses had fallen more heavily on the French, since they had begun the war with a third fewer men than the Germans, but to both armies Verdun had become a place of terror and death that could not yield victory. The Germans made a final effort on 11 July, which reached Fort Souville, but it was beaten off. Thereafter the Germans ceased their attempt to destroy the French army at Verdun and relapsed into the defensive. For a while it became a quiet sector until, in October, the French moved to recover the ground lost. On 24 October Douaumont was recaptured, on 15 December a wider offensive regained much of the ground lost on the east bank since the beginning of the battle. By then, however, another battle altogether, raging since 1 July, had shifted the crux of the Western Front from Verdun to the Somme.
2. Offensive on the Somme
Verdun had been planned by Falkenhayn as an operation to ‘bleed white’ the French army and knock Britain’s ‘best sword’ out of her hand. Even by June, when the battle still had six months to run, it had failed in both its purposes and, as it failed, Falkenhayn’s credibility as Chief of Staff had waned also. Dominating though he was in personality and intellect, handsome and forthright, self-assured to the point of arrogance and of proven ability as a staff officer and Minister of War, he suffered from the disadvantage of association, in the popular mind, with defeat rather than victory.39Responsibility for the failure of the Schlieffen Plan – intrinsic though failure was in the plan’s defects – and for the entrenchment of the Western Front, though it properly lay in both cases at Moltke’s door, nevertheless attached in practice to him as Moltke’s immediate successor. The victories of the Eastern Front, Tannenberg, even Gorlice-Tarnow, seemed the achievement of Hindenburg, and of his alter ego, Ludendorff. Falkenhayn’s confederality with the Austrian Chief of Staff, Conrad von Hötzendorf, landed him with shared culpability for the poor showing of the Austro-Hungarian army against the Serbs and Russians and even for the entry of Italy into the war, since Italy’s motivation was essentially anti-Austrian. The only initiative undoubtedly his own, and for which he might have taken credit were it a success, was Verdun, which, by midsummer, was palpably a terrible failure. Even before the great bombardment that would usher in the Anglo-French offensive on the Somme had opened, Falkenhayn’s grip on high command was weakening, the star of his ascent and zenith already passing to the eastern titan, Hindenburg, who would replace him in August.
The battle of the Somme
The Somme was to be the enterprise of another ascendant general, Douglas Haig. John French, ‘the little field marshal’ who had taken the BEF to France, had been worn down by the attrition of his beloved army of regulars, the old sweats of his Boer War glory days, the keen young troopers of the cavalry in which he had been raised, the eager Sandhurst subalterns, the generation of decent, dutiful majors and colonels who had been his companions on the veldt and in the hunting field.40 The death of so many of them – there had been 90,000 casualties among the original seven infantry divisions by November 1914, rather more than a hundred per cent of mobilised strength – afflicted him, and he added to the pain he felt by his apparently compulsive need to tour the military hospitals and talk to the wounded. ‘Horribly sad and very pathetic to see how good and cheery and patient the dear fellows are . . . I hate it all so! . . . such horrible sadness and depression.’41 French was not made for modern war or for the politics of a national conflict. He could not feel for the citizen soldiers coming forward in their hundreds of thousands as he did instinctively for the vanishing seven-year men of the feudal order he had known as a young officer, nor could he play the ministerial game at which his War Office equals and younger subordinates were adept. Douglas Haig, commander of the BEF’s First Army, was sinuous in his relationships with the great, particularly at court. He had precipitately married a royal lady-in-waiting after the briefest of introductions and had accepted an invitation to correspond privately with King George V soon after the Western Front had relapsed into stalemate. Others in the BEF hierarchy shared by the end of 1915 the belief that French had proved his incapacity to continue in supreme command, and their views were made known to the government. It was Haig, however, who wielded the dagger. During a visit by the King to France at the end of October, he told him directly that French was ‘a source of great weakness to the army, and no one had confidence in him any more’. All that was true, but it would have come better from Haig had he not added that he himself was ready to do his duty in any capacity. ‘Any capacity’ clearly meant as French’s successor which, after further consultations between the King, the Prime Minister and Kitchener, still Secretary of State for War, though his perch was creaking also, he became on 16 December 1915.42
Haig, whom his contemporaries found difficult to know, has become today an enigma. The successful generals of the First World War, those who did not crack outright or decline gradually into pessimism, were a hard lot, as they had to be with the casualty figures accumulating on their desks. Some, nevertheless, managed to combine toughness of mind with some striking human characteristic: Joffre, imperturbability; Hindenburg, gravity; Foch, fire; Kemal, certainty. Haig, in whose public manner and private diaries no concern for human suffering was or is discernible, compensated for his aloofness with nothing whatsoever of the common touch. He seemed to move through the horrors of the First World War as if guided by some inner voice, speaking of a higher purpose and a personal destiny. That, we now know, was not just appearance. Haig was a devotee both of spiritualist practices and of fundamentalist religion.43 As a young officer he had taken to attending seances, where a medium put him in touch with Napoleon; as Commander-in-Chief he fell under the influence of a Presbyterian chaplain whose sermons confirmed him in his belief that he was in direct communication with God and had a major part to play in a divine plan for the world. His own simple religion, he was convinced, was shared by his soldiers, who were inspired thereby to bear the dangers and sufferings which were their part of the war he was directing.44
Despite his strangeness, Haig was an efficient soldier, the superior to French in every branch of modern military practice, and his skills were not better shown than in his preparations for the Somme. That high and empty battlefield had not been contested since the first weeks of the war. On the enemy side, the Germans had profited from the peace in which they had been left since 1914 to construct the strongest position on the Western Front. The hard, dry, chalky soil was easily mined and they had driven dugouts thirty feet below ground, impervious to artillery fire, provisioned to withstand siege and linked to the rear by buried telephone cable and deep communication trenches. On the surface they had constructed a network of machine-gun posts, covering all angles of approach across the treeless downs, and in front of their fire trenches laid dense entanglements of barbed wire. They had time to do so. Among the half-dozen divisions garrisoning the Somme sector, the 52nd had been there since April 1915, the 12th since October and the 26th and 28th Reserve Divisions since September 1914. They had made themselves secure.45
On the other side of no man’s land, little had been done since 1914. The French, who had occupied the sector until the extension of the British line southward in August 1915, held it as a ‘quiet front’, defended by artillery with few infantry in the front line. The British had introduced a more aggressive mood but the infrastructure for a great offensive was still not in place when Haig took command. Under his direction, the back area of the Somme, from the little market town of Albert to the departmental capital of Amiens twenty-five miles behind, was transformed into an enormous military encampment, cut by new roads leading towards the front and covered with shell dumps, gun positions and encampments for the army that would launch the attack. As a military technician, Haig could not be faulted. His talents as a tactician remained to be proved.
The army assembling on the Somme had no doubts in the high command or in itself. It consisted of twenty divisions, most grouped, under the command of General Sir Henry Rawlinson, in the new Fourth Army. The majority of the divisions were new to the war also. A handful were old regular formations, the 4th, 7th and 8th, and the 29th, all greatly changed since their ordeals in the original BEF and at Gallipoli. Four were Territorial, the 46th, 56th, 48th and 49th, which had been in France since the spring of 1915. The rest were ‘Kitchener’ formations of citizen volunteers, many organised round ‘Pals’ or ‘Chums’ battalions, for which the Somme would be their first battle. There were ten of these Kitchener divisions, of which the senior, the 9th Scottish, had arrived in France in May 1915 but the 34th only in January 1916.46 Perhaps the most unusual was the 36th (Ulster) Division, a wholesale embodiment in khaki of the Ulster Volunteer Force of Irish Protestants opposed to Irish Home Rule who, on the outbreak of war, had collectively volunteered. The Ulstermen differed from their other Kitchener comrades only in their pre-war experience of military drill. With the reality of battle they had no more familiarity than the rest. Their infantry battalions were wholly inexperienced; so, too, and more critically, were their batteries of supporting artillery, on whose accurate shooting and prompt changing of target the success of the coming offensive depended.
Haig’s plan for the Somme was simple, akin in outline to Falkenhayn’s for Verdun, with the difference that he hoped to break the enemy’s line rather than force him to stand and fight for it in a struggle of attrition. An enormous bombardment, to last a week and consume a million shells, was to precede the attack. As it died away on the date chosen for assault, 1 July, nineteen British divisions and, south of the River Somme, three French, all that could be spared while Verdun still raged, were to move forward across no man’s land and, in the expectation that the enemy surviving the shelling would have been stunned into inactivity, pass through the broken wire entanglements, enter the trenches, take possession and move on to the open country in the rear. So certain were Haig and most of his subordinates of the crushing effect the artillery would produce, that they had decided not to allow the inexperienced infantry to advance by the tried and tested means of ‘fire and movement’, when some lay down to cover with rifle volleys the advance of the rest, but to keep them moving forward upright and in straight lines. At the battle of Loos the preoccupation of the General Staff had been to ‘keep the troops in hand’, with the result that the reserves had been kept too far behind the lines and, when sent forward too late, deployed in dense masses.47 The preoccupation before the Somme was with the danger of the troops taking cover and not restarting the advance once they had lain down. The tactical instruction for the battle, ‘Training Divisions for Offensive Action’ (SS 109), and the associated instruction issued by Fourth Army, ‘Tactical Notes’, both prescribe an advance by successive waves or lines of troops and a continuous movement forward by all involved. ‘The assaulting troops must push forward at a steady pace in successive lines, each line adding fresh impetus to the preceding line.’48
Haig, as Commander-in-Chief, and Rawlinson, commanding the attacking troops, though agreeing on the tactics to be followed, differed over the offensive’s objects. Haig expected a breakthrough, to as far as Bapaume, the little market town on the far side of the Somme uplands seven miles from the start line. Rawlinson foresaw a more limited result, a ‘bite’ into the German trench system, to be followed by further bites to gain more territory. Rawlinson, as events would prove, was the more realistic. Both generals, however, were equally unrealistic in their expectations of what the preliminaries would achieve. Nearly three million shells had been dumped forward for the preparatory bombardment, to feed 1,000 field guns, 180 heavy guns and 245 heavy howitzers, giving a density of one field gun per twenty yards of front and one heavy gun or howitzer to fifty-eight yards.49 The artillery plan was for the field guns to concentrate, before the battle, on cutting the enemy’s wire in front of his trenches, while the heavy guns were to attack the enemy’s artillery with ‘counter-battery’ fire and destroy his trenches and strongpoints. At the moment of assault, as the British infantry left their trenches to advance across no man’s land, the field artillery was to lay a ‘creeping barrage’ ahead of the leading wave, which was intended to prevent the German defenders from manning the parapet opposite so that, in theory, the German trenches would be empty when the British arrived.
Almost everything that Haig and Rawlinson expected of the enormous artillery effort they had prepared was not to occur. The German position, for one thing, was far stronger than British intelligence had estimated. The thirty-foot dugouts in which the German front-line garrison sheltered were almost impervious to any shell the British could fire and had survived intact right up to the last days before the attack. A trench raid launched on the night of 26/27 June revealed, for example, that ‘the dugouts are still good. The [Germans] appear to remain in these dugouts all the time and are completely sheltered.’50 So it was to prove on the day. Even more ominous was the failure to cut wire. Later in the war a sensitive ‘graze’ fuse would come into use, which exploded a shell when it touched something as slender even as a single wire strand. In 1916 shells only detonated on hitting the ground and bombardments fired at wire entanglements therefore merely tossed them about, creating a barrier yet more dense than that laid by the enemy in the first place. The general commanding the British VIII Corps, Hunter-Weston, who had been at Gallipoli and should have known how tough wire was, reported before 1 July that the enemy wire on his front was blown away and ‘the troops could walk in’ but one of his junior officers ‘could see it standing strong and well’.51 Since uncut wire in front of defended trenches was death to attacking infantry, this complacent misappreciation by the staff was literally lethal.
Finally, the confidence shown in the artillery to lay a creeping barrage was misplaced. The movement of a line of exploding shells just in front of a line of advancing infantry, ideally fifty yards in front or less, was a new technique and demanded high gunnery skills. Without communication between infantry battalions and artillery batteries – and there could be none without tactical radio, a development of the future – the artillery had to fire by timetable, calculated by the speed at which the infantry was expected to advance, roughly fifty yards a minute. The guns would lay a barrage on an identified trench line, then ‘lift’ to the next at a moment when the infantry was deemed to have arrived. In practice, because the artillery feared killing its own infantry, the intervals in distance between ‘lifts’ was made too long, in time too short, with the result that the experience of attacking waves would be, too often, to see the barrage creeping away in front of them, beyond trenches still strongly held by the enemy, without any means of recalling it. The corrective, to be adopted by some corps, of bringing the barrage back and then forward, would not work either, since its return frightened the infantry into taking cover from ‘friendly fire’, the protection being lost when the barrage crept off again without warning. The worst feature of artillery precaution was to be the lifting of the barrage from the enemy’s front line too soon before the assault, while the infantry was still in no man’s land and the wrong side of often uncut wire. A veteran of Gallipoli, commanding a heavy battery in Hunter-Weston’s III Corps ‘knew that the attack . . . in his sector was doomed when [the corps commander] ordered the heavy artillery to lift off the enemy front line trenches ten minutes before zero, and the field artillery two minutes before zero hour’.52 It was not only in his sector that the barrage was to lift too soon. Almost everywhere on the front of Fourth Army on 1 July the artillery fire was to depart prematurely from the infantry, who were to advance against wire badly cut or not cut at all, against trenches filled with Germans fighting for their lives.
What the infantry should have done in such circumstances has generated an enormous literature, much of it quite recent. A new generation of young military historians has taken to re-fighting the battles of the British Expeditionary Force with a passion more understandable in survivors of the trench warfare disasters than in posthumous academic analysts. An underlying theme is that, dreadful as the experience of the early offensives was, it provided a learning process through which the survivors and their successors won the eventual victories of 1918, an argument akin to the thought that Dunkirk was a valuable rehearsal in amphibious operations for D-Day. At a more detailed technical level, the new Western Front historians explore such issues as what was the proper relationship between rifleman, light-machine gunners and grenadiers, how the potentialities of improved infantry weapons might have best been exploited and which was the ideal infantry formation, column, line or infiltrating ‘blob’.53 The energy expended in such reconsiderations seems, to this author at any rate, a pointless waste. The simple truth of 1914–18 trench warfare is that the massing of large numbers of soldiers unprotected by anything but cloth uniforms, however they were trained, however equipped, against large masses of other soldiers, protected by earthworks and barbed wire and provided with rapid-fire weapons, was bound to result in very heavy casualties among the attackers. That was proved to be the case, whatever the variation in tactics and equipment, and there was much variation, from the beginning on the Aisne in 1914 to the end on the Sambre and Meuse in 1918. The effect of artillery added to the slaughter, as did that of bayonets and grenades when fighting came to close quarters in the trench labyrinths. The basic and stark fact, nevertheless, was that the conditions of warfare between 1914 and 1918 predisposed towards slaughter and that only an entirely different technology, one not available until a generation later, could have averted such an outcome.
The first day of the battle of the Somme, 1 July 1916, was to be an awful demonstration of that truth. Its reality remains evident even today to anyone who returns to the centre of the Somme battlefield at Thiepval, near the memorial to the 36th Ulster Division, and glances north and south down the old front line. The view northward is particularly poignant. Along it, at intervals of a few hundred yards, run a line of the Commonwealth War Grave Commission’s beautiful garden cemeteries, ablaze near the anniversary of the battle with rose and wisteria blossom, the white Portland stone of headstones and memorial crosses gleaming in the sun. The farthest, on the ridge near Beaumont Hamel, contains graves of the regular 4th Division, the nearest, in the valley of the Ancre, the Somme’s little tributary, those of the Kitchener 32nd Division. A few, like those of the Ulster Division, stand a little forward of the rest, and mark the furthest limit of advance. The majority stand on the front line or in no man’s land just outside the German wire. The soldiers who died there were later buried where they had fallen. Thus the cemeteries are a map of the battle. The map tells a simple and terrible story. The men of the Fourth Army, the majority citizen volunteers going into action for the first time, rose from their trenches at zero hour, advanced in steady formation, were almost everywhere checked by uncut barbed wire and were shot down. Five divisions of the seventeen attacking entered the German positions. The infantry of the remainder were stopped in no man’s land.
Descriptions of zero hour on 1 July abound, of the long lines of young men, burdened by the sixty pounds of equipment judged necessary to sustain them in a long struggle inside the German trenches, plodding off almost shoulder to shoulder; of their good cheerand certainty of success; of individual displays of bravado, as in the battalions which kicked a football ahead of the ranks; of the bright sunshine breaking through the thin morning mist; of the illusion of an empty battlefield, denuded of opponents by the weight of the bombardment and the explosion of twenty-one mine chambers, laboriously driven under the German front lines, as the attack began. Descriptions of what followed zero hour abound also: of the discovery of uncut wire, of the appearance of the German defenders, manning the parapet at the moment the British creeping barrage passed beyond, to fire frenziedly into the approaching ranks, of the opening of gaps in the attacking waves, of massacre in the wire entanglements, of the advance checked, halted and eventually stopped literally dead.
The Germans (who were fighting for their lives) had practised bringing their machine guns up the steps from their deep dugouts hundreds of times. F. L. Cassell, a German survivor, recalled ‘the shout of the sentry, “They are coming” . . . Helmet, belt and rifle and up the steps . . . in the trench a headless body. The sentry had lost his life to a last shell . . . there they come, the khaki-yellows, they are not twenty metres in front of our trench . . . They advance slowly fully equipped . . . machine-gun fire tears holes in their ranks.’54 The machine guns reached even inside the British front line in places, to hit troops who had not reached no man’s land. A sergeant of the 3rd Tyneside Irish recalled seeing ‘away to my left and right, long lines of men. Then I heard the “patter, patter” of machine guns in the distance. By the time I’d gone another ten yards there seemed to be only a few men left around me; by the time I had gone twenty yards, I seemed to be on my own. Then I was hit myself.’55 The whole of the Tyneside Irish Brigade, of four battalions, nearly three thousand men, was brought to a halt inside British lines, with appalling loss of life. One of its battalions lost 500 men killed or wounded, another 600. In offensive terms, the advance had achieved nothing. Most of the dead were killed on ground the British held before the advance began.
Appalling loss of life was the result of the first day of the Somme along the whole front of the attack. When, in the days that followed, the 200 British battalions that had attacked began to count the gaps in their ranks, the realisation came that, of the 100,000 men who had entered no man’s land, 20,000 had not returned; another 40,000 who had been got back were wounded. In summary, a fifth of the attacking force was dead, and some battalions, such as the 1st Newfoundland Regiment, had ceased to exist. The magnitude of the catastrophe, the greatest loss of life in British military history, took time to sink in. The day following the opening of the attack, Haig, conferring with Rawlinson and his staff at Fourth Army headquarters, was clearly still uninformed of how great the casualties had been and discussed, as a serious proposition, how the offensive was to be continued, as if it were a possibility for the morrow or the day after. He believed that the enemy ‘has undoubtedly been severely shaken and he has few reserves in hand’.56 In fact, the Germans had brought up several reserve divisions during the day, while the losses suffered by their troops in line – about six thousand altogether – were a tenth of those of the British. The German 180th Regiment, for example, lost only 180 men out of 3,000 on 1 July; the British 4th Division, which attacked it, lost 5,121 out of 12,000. If the Germans had been shaken, it was by the ‘amazing spectacle of unexampled gallantry courage and bulldog determination’ and by their eventual revulsion from the slaughter inflicted; in many places, when they realised their own lives were no longer at risk, they ceased firing, so that the more lightly British wounded could make their way back as best they could to their own front line. There was, for the worse wounded, no early rescue. Some were not got in until 4 July, some never. A young British officer, Gerald Brenan, crossing subsequently captured ground in the fourth week of July, found the bodies of soldiers wounded on 1 July who had ‘crawled into shell holes, wrapped their waterproof sheets round them, taken out their bibles and died like that’. They were among thousands whose bullet-riddled bodies gave up life that day or afterwards, beyond the reach of stretcher bearers or simply lost in the wilderness of no man’s land. Even among those found and carried back, many died as they lay waiting for treatment outside the field hospitals, which were overwhelmed by the flood of cases.
If there were any exception to the unrelievedly disastrous results of 1 July, it was that the German high command, as opposed to their front-line troops, had been gravely alarmed by the scale of the British attack, particularly because in one sector, astride the River Somme itself, ground had been lost. Unknown, naturally, to Haig and Rawlinson, Falkenhayn reacted to that loss in peremptory fashion, relieving the Chief of Staff of Second Army, in whose sector it had occurred, and replacing him with his own operations officer, Colonel von Lossberg, the main architect of German defensive methods on the Western Front.57 Von Lossberg imposed the condition, on accepting the appointment, that the attacks at Verdun be given up at once, which they were not. Falkenhayn broke his undertaking and the offensive continued until his own dismissal at the end of August. Lossberg’s arrival was nevertheless significant, for his reorganisation of the Somme front ensured that the results of the first day, the outcome of British over-optimism and German hyper-readiness, would be sustained in the later stages of a battle which blunted the German edge relentlessly even as it taught the British a realism their inexperienced soldiers lacked at the outset. Lossberg’s intervention caused the defenders to abandon the practice of concentrating on the defence of the front line and to construct a ‘defence in depth’, based not on trenches but on lines of shell holes, which the British artillery created in profusion. The forward zone was to be thinly held, to minimise casualties, but ground lost was to be speedily retaken by deliberate counter-attacks launched by organised reserves held in the rear.58
This German technique defied all Haig’s efforts to exploit such success as had been achieved on 1 July. Not until 14 July, in the sector astride the Somme where the more experienced French had assisted the British to make a clear break-in to the German positions, was further ground gained. Haig’s suspicion of night attacks was overcome by his subordinates and, in an attack at half-light, four British divisions rolled forward to take Bazentin Ridge, Mametz Wood and Contalmaison. On the map the advance looks impressive; on the ground, where the visitor covers the distance in a few minutes of motoring, less so, though the atmosphere of menace clinging to the sector’s small valleys and re-entrants oppresses. Some of the BEF’s cavalry, still Haig’s preferred arm of decision, was brought up during the day but, after a skirmishing near High Wood at one of the Somme battlefield’s most commanding points, was forced to withdraw. Imperial troops, the 1st and 2nd Australian Divisions, veterans of Gallipoli, and the South African Brigade, renewed the advance during the second half of the month, taking Pozières and Delville Wood, the latter the site of a South African epic, but no opportunity for the cavalry to intervene recurred. Like Verdun, the Somme was becoming an arena of attrition, to which fresh divisions were sent in monotonous succession – forty-two by the Germans during July and August – only to waste their energy in bloody struggles for tiny patches of ground, at Guillemont, Ginchy, Morval, Flers, Martinpuich. By 31 July, the Germans on the Somme had lost 160,000, the British and French over 200,000, yet the line had moved scarcely three miles since 1 July. North of the Ancre, or along half the original front, it had scarcely moved at all.
The offensive on the Somme might have been doomed to drift away into an autumn of frustration and a winter of stalemate had it not been for the appearance in mid-September of a new weapon, the tank. As early as December 1914 a visionary young officer of the Royal Engineers, Ernest Swinton, having recognised that only a revolutionary means could break what was already the stalemate of barbed wire and trench on the Western Front, had proposed the construction of a cross-country vehicle, armoured against bullets, that could bring firepower to the point of assault. The idea was not wholly new – it had been anticipated, for example, in H.G. Wells’s short story The Land Ironclads of 1903, and in an imprecise form by Leonardo da Vinci – nor was the technology: an all-terrain vehicle, using a ‘footed wheel’, had been built in 1899 and by 1905 caterpillar track vehicles were in agricultural use.59 It was the crisis of war that brought together technology and vision and, through that of Swinton and collaborators, Albert Stern, Murray Sueter, backed by the enthusiasm of Winston Churchill, whose Royal Naval Division’s armoured cars had cut a dash in Belgium in 1914, bore fruit in the tank’s prototype, ‘Little Willie’ of December 1915. In January 1916 a larger and gun-equipped development, ‘Mother’, had been produced and by September a fleet of forty-nine similar Mark I ‘Tanks’, as they had been named for deceptive purposes, were in France and ready to enter battle.60
The tanks were assigned to the Heavy Branch of the Machine Gun Corps, a war-raised formation controlling the BEF’s medium machine guns. Following the attrition battles of August, a new effort was planned to open up the Somme front and the tanks, some armed with machine guns, some with 6-pounder cannon, were allotted to the Fourth and Reserve (future Fifth) Armies, to lead an assault along the line of the old Roman road that leads from Albert to Bapaume between the villages of Flers and Courcelette. The appearance of the tanks terrified the German infantry defending the sector and the armoured monsters led the British infantry onward for 3,500 yards before mechanical breakdowns and ditchings in rough ground brought the advance to a halt; a number, caught in artillery fire, were knocked out. The event brought one of the cheapest and most spectacular local victories of the war on the Western Front thus far, but its efforts were to be frustrated immediately by the disablement of almost all the thirty-six tanks that had crossed the start line. Though the infantry plugged away at the gains the tanks had made, the usual German stubbornness in manning shell holes and reserve lines blocked the potential avenue of advance and restored the stalemate.
October and November brought no change. Both the British and French attacked repetitively, at Thiepval, Transloy and in the sodden valley of the Ancre, in increasingly wet weather that turned the chalky surface of the Somme battlefield into glutinous slime. By 19 November, when the Allied offensive was brought officially to a halt, the furthest line of advance, at Les Boeufs, lay only seven miles forward of the front attacked on 1 July. The Germans may have lost over 600,000 killed and wounded in their effort to keep their Somme positions. The Allies had certainly lost over 600,000, the French casualty figure being 194,451, the British 419,654. The holocaust of the Somme was subsumed for the French in that of Verdun. To the British, it was and would remain their greatest military tragedy of the twentieth century, indeed of their national military history. A nation that goes to war must expect deaths among the young men it sends and there was a willingness for sacrifice before and during the Somme that explains, in part at least, its horror. The sacrificial impulse cannot, however, alleviate its outcome. The regiments of Pals and Chums which had their first experience of war on the Somme have been called an army of innocents and that, in their readiness to offer up their lives in circumstances none anticipated in the heady days of volunteering, it undoubtedly was. Whatever harm Kitchener’s volunteers wished the Germans, it is the harm they thereby suffered that remains in British memory, collectively but also among the families of those who did not return. There is nothing more poignant in British life than to visit the ribbon of cemeteries that marks the front line of 1 July 1916 and to find, on gravestone after gravestone, the fresh wreath, the face of a Pal or Chum above a khaki serge collar staring gravely back from a dim photograph, the pinned poppy and the inscription to ‘a father, a grandfather and a great-grandfather’. The Somme marked the end of an age of vital optimism in British life that has never been recovered.61
3. The Wider War and the Brusilov Offensive
While the great dramas of Verdun and the Somme were being played out in France, the war on the fronts elsewhere took a very varied form. In German East Africa, where Jan Smuts, the brilliant guerrilla opponent of the British during the Boer War, had arrived to take command in 1915, four columns set out in 1916 – two British from Kenya and Nyasaland respectively, one Portuguese from Mozambique, one Belgian from the Congo – to make a concentric advance on von Lettow-Vorbeck’s black army, encircle it and bring the campaign to a close. The Allied fighting troops numbered nearly 40,000, Lettow’s about 16,000. Dividing his force, he had no difficulty in eluding Smuts with his main body and in beating a fighting retreat southward from Mount Kilimanjaro towards Tanga and Dar es-Salaam, keeping parallel to the coast and retreating slowly southward across the grain of the country. He fought when obliged to do so but always disengaged before defeat and, destroying bridges and railway lines behind him, evaded encirclement and kept his force intact. His African askaris, moreover, were resistant to most of the parasitic diseases that attack humans in the interior. His enemies, who included large numbers of Europeans and Indians, were not. Their enormously high toll of sickness – thirty-one non-battle casualties to one battle casualty – was the real cause of their failure to run Lettow to earth. At the end of 1916 his little army was as fit, capable and elusive as at the start of the war.62
The Turks, so underestimated by the Allies at the outset, sustained the success they had achieved at Gallipoli. Though their efforts to revive their offensive against the Suez Canal were repulsed, in a limited campaign that took British forces to the Sinai border of Palestine, and though their army in the Caucasus suffered further defeats at the hands of the Russians, who pushed forward a perimeter from Lake Van to Trebizond on the Black Sea coast by August, in Mesopotamia they inflicted a wholly humiliating defeat on the Anglo-Indian force that had landed at the mouth of the Shatt el-Arab in 1914. During 1915 Expeditionary Force D, as it was known, pushed up the River Tigris towards Baghdad, part of the force moving by land, part by water, until in November 1915 its advance guard was at Ctesiphon. Its situation looked promising, since it was established in the heart of the Ottoman empire at a moment when the nearest Turkish reserves, according to British intelligence, were 400 miles distant in the Caucasus or 350 miles away at Aleppo in Syria. Somehow, however, the Turks managed to scrape together enough reinforcements to send troops down the Tigris and confront the Expeditionary Force. Its commander, Major General Townshend, decided, though he had not been defeated, that he was overextended and accordingly ordered a retreat to Kut al-Amara, a hundred miles down river. There the Force entrenched itself in a loop of the Tigris to await support and the recovery of its soldiers from their ordeal of the long advance and retreat.
Townshend had supplies for two months and personal experience of conducting defence; in 1896 he had successfully commanded the little North-West Frontier fort at Chitral during a siege that became celebrated throughout the empire.63 The Turks, masters of entrenchment warfare, proved far more dangerous opponents than the Chitrali tribesmen. Having encircled Townshend’s encampment with earthworks, they settled down to repel attacks both by the garrison and by the relieving force which, between January and March, four times attempted to break through their lines. Each effort was unsuccessful and the last, known as the battle of the Dujaila Redoubt, left a thousand dead at the scene of action. Townshend’s headquarters was only seven miles distant from the furthest advance but, immediately after the defeat, the annual floods, fed from the snow-melt off the Zagros Mountains, swelled the rivers and put the surface of the Mesopotamian plain under water. Kut was completely cut off from outside help and on 29 April surrendered. Townshend and 10,000 survivors of the Expeditionary Force went into captivity, harsh for the common soldiers, 4,000 of whom died in enemy hands. Kut was not retaken until the end of the year, when nearly 200,000 British and Indian troops and followers had been assembled, to oppose 10,000 Turks and a handful of Germans. Like Salonika, where the Allies continued to wage an unsuccessful campaign against very inferior forces throughout 1916, Mesopotamia had become a drain on resources instead of a threat to the enemy.
On the Italian front, though there the defenders were also heavily outnumbered by the attackers, the disparity was not so great. The strength of the Italian army was increasing, and would eventually almost double, from thirty-six peacetime divisions to sixty-five, and during 1916 the Italians would attract thirty-five of the sixty-five mobilised Austrian divisions to their mountains. The consequent weakening of Austria’s capacity to bear a fair share of the burden in the east would largely facilitate Russia’s successful resumption of the offensive in this year. Outnumbered though they were, however, the Austrians both frustrated Italy’s continuing attempts to break into the Austro-Hungarian heartland via the Isonzo route and launched a counter-offensive of their own directed towards the rich industrial and agricultural region in the plains of River Po. Conrad, the Habsburg Chief of Staff, nurtured an almost personal animus against Austria’s former partner in the Triple Alliance and had fallen out with Falkenhayn over his determination to punish them at the expense of sustaining the joint Austro-German success against the Tsar’s armies which had begun at Gorlice-Tarnow. On 15 May 1916, almost on the anniversary of that victory, Conrad unleashed his own ‘punishment expedition’ (Strafexpedition) from the northern mountain chain of the Trentino, between Lake Garda, the Alpine beauty resort, and the headwaters of the River Brenta, which leads towards the lagoons of Venice. The preliminary bombardment, which opposed 2,000 Austrian guns to 850 Italian, was powerful, but the defenders were forewarned by the evidence of Austrian preparations and then fought with heroic self-sacrifice to hold the invaders at bay. The Rome Brigade was almost wiped out in its defence of Piazza. As a result, the Austrians nowhere advanced more than ten miles and, though their losses were fewer than the Italian – 80,000 to 147,000 – the punishment expedition neither threatened a breakthrough nor deflected Cadorna, the Italian Commander-in-Chief, from pursuing his relentless offensive on the Isonzo. The Sixth Battle opened in August and secured the frontier town of Gorizia, the Seventh, Eighth and Ninth followed in September, October and November. The bridgehead across the Isonzo at Gorizia was enlarged and a foothold on the harsh Carso upland secured. The Italian infantry, despite heavy losses and the frustration of their offensive efforts, still seemed ready to return to the attack, even under Cadorna’s aloof and heartless direction of the war.
The course of operations in Italy during 1916 had one positive result: by attracting Austrian divisions from the Russians’ southern front, it allowed the Tsar’s armies to organise a successful counter-offensive against their weakened enemy. The Russians were committed to such an offensive by the Chantilly agreement of December 1915, while intelligence of Conrad’s punishment expedition had caused Cadorna to request its mounting as a matter of urgency. Its results exceeded what had been promised or expected, most of all by the Stavka, whose plans for 1916 were for a resumption of an offensive against the Germans on Russia’s northern front rather than the Austrians in the south. The German advanced positions in the north threatened Petrograd, the capital, and had brought under enemy occupation the productive Baltic States, where Ludendorff had created a full-blown occupation economy. In an anticipation of what Hitler would less imaginatively attempt after 1941, he divided the region into six administrative areas, under a German military governor, and set about harnessing its agricultural and industrial resources to the German war effort. Ludendorff’s plans went beyond the purely economic. ‘I determined to resume in the occupied territory that work of civilisation at which German hands had laboured in those lands for many centuries. The population, made up as it is of such a mixture of races, has never produced a culture of its own and, left to itself, would succumb to Polish domination.’ Ludendorff foresaw the transformation of Poland into ‘a more or less independent state under German sovereignty’ and by the spring of 1916 was planning to settle much of the Baltic States with Germans, who would take the land of the expropriated inhabitants. They did not include the Jews who, being often German-speakers, were regarded as useful instruments of occupation policy.64
Ludendorff’s scheme to Germanise the Tsar’s possessions in Poland and the Baltic regions was one reason for the Stavka to choose a resumption of the offensive in the north as its main strategy for 1916. It began, in response to a French appeal to relieve the pressure at Verdun, with an attack on each side of Lake Naroch, aimed at Vilna, the chief town of eastern Poland, on 18 March. Thanks to the mobilisation of Russia’s industry for war, and to the call-up of new classes of conscripts, the Russian armies now outnumbered their opponents, by 300,000 to 180,000 in the north and 700,000 to 360,000 in the centre; only in the southern sector, commanded by Brusilov, did numbers remain equal at about half a million men on each side. In the north the Russians for the first time had a large superiority in guns and stocks of shell, with 5,000 guns and a thousand rounds per gun, considerably more than assembled by the Germans for the Gorlice-Tarnow breakthrough.65
Somehow, however, the advantage was cast away. The artillery preparation was not co-ordinated with the assault by the infantry of Second Army which, attacking on a very narrow front, ran into its own fire and then, in the salient it had won, came under bombardment by German guns from three sides. Three-quarters of the infantry, 15,000 men, were lost in the first eight hours; yet 350,000 men were theoretically available for the offensive, had it been launched on a wider front. Reinforcement merely increased the casualty list without the gain of more ground. By 31 March, when the offensive ended, Russian losses totalled 100,000, including 12,000 men who had died of exposure in the harsh late-winter weather. In April a counter-attack by the Germans, who had lost 20,000, recovered all the ground the Russians had gained.66
Prospects for the general offensive promised in June did not therefore augur well, since the Stavka again wished to attack in the north, above the Pripet Marshes that divided the front into two. In fact, Evert, commander of the army group that had failed at Lake Naroch, did not want to attack at all. Alexeyev, the Chief of Staff, was nevertheless insistent and secured the reluctant co-operation of Evert and Kuropatkin, the other army group commander on the northern sector, on the understanding that there would be copious reinforcements of men and material. To the surprise of those present at the conference on 14 April, the new commander of the southern front, Alexei Brusilov, who had succeeded Ivanov in March, was not reluctant at all. He believed victory was possible, by careful preparation, against the weakened Austrians and, as he requested no reinforcements, he was given permission to make his attempt. He had proved his ability at lower levels of command and he had also found the time to consider the problems of attacking entrenched positions covered by defending artillery with reserves in rear ready to stem a break-in. The solution, he had concluded, was to attack on a wide front, thus depriving the enemy of the chance to mass reserves at a predeterminably critical point, to protect the assaulting infantry in deep dugouts while they were waiting to jump off, and to advance the line as near as possible to the Austrian, by digging saps forward as close as seventy-five yards to the enemy trenches. These were great improvements. In the past the Russians had often left no man’s land a mile or more wide, thus condemning the attacking infantry to heavy casualties during the approach, following equally heavy casualties suffered in trenches unprotected against enemy bombardment before the attack began.
Brusilov’s preparations worked admirably. Though his superiority of numbers over the Austrians on the twenty miles of chosen front was only 200,000 to 150,000, with 904 to 600 guns, the enemy was genuinely surprised when the attack opened on 4 June. The Russian Eighth Army overwhelmed the Austrian Fourth and pushed on to take the communication centre of Lutsk, and to advance forty miles beyond the start line. Huge numbers of prisoners were taken, as the shaken Austrians surrendered to anyone who would take them prisoner. Eighth Army’s neighbours also advanced but the greatest success was achieved in the south, between the River Dniester and the Carpathians, where the Austrian Seventh Army was split in two, lost 100,000 men, mainly taken prisoner, and by mid-June was in full retreat.
The Brusilov Offensive
At the beginning of July the Russian armies north of the Pripet Marshes also went on to the offensive, profiting from Brusilov’s success and the confusion reigning in the Austro-German high command as to where best to deploy its very scanty reserves, to press forward towards Baronovitchi, the old Russian headquarters town. Evert’s offensive, opposed by German troops, was soon stopped but Brusilov’s army group sustained its success over the Austrians throughout July and August and into September, by which time it had taken 400,000 Austrians prisoner and inflicted losses of 600,000. The German forces involved in opposing the Russian advance had lost 350,000, and a belt of Russian territory sixty miles deep had been taken back from the invaders. Had Brusilov possessed the means to follow up his victory and to bring reserves and supplies forward at speed, he might have recovered more of the ground lost in the great retreat of 1915, perhaps even to reach Lemberg and Przemysl once again. He possessed no such means. The rail system, which in any case favoured the Austrians rather than the Russians, could not provide tactical transport across the battle zone, while the roads, even had he had adequate motor transport, were unsuitable for heavy traffic. Nevertheless, the Brusilov offensive was, on the scale by which success was measured in the foot-by-foot fighting of the First World War, the greatest victory seen on any front since the trench lines had been dug on the Aisne two years before.67
The Russian victory, though it also cost a million casualties, sealed the fate of Falkenhayn, whose security of tenure as Chief of Staff had been weakening as the battle for Verdun protracted. His dismissal, and replacement by Hindenburg, was disguised by appointing him to command in the new campaign against Romania. Romania, long courted by both the Allies and the Central Powers, had thus far prudently avoided choosing sides. Its neighbour, Bulgaria, had thrown in its lot with Germany and Austria in October 1915 but Romania, which had acquired Bulgarian territory at the end of the Second Balkan War in 1913, continued to hold aloof. Its chief national interest was in the addition to its territory of Transylvania, where three million ethnic Romanians lived under Austro-Hungarian rule. As Brusilov’s advance pushed westward, widening the common border of military contact between Russia and Romania and apparently promising not only Russian support but Austrian collapse, the Romanian government’s indecision diminished. The Allies had long been offering an enlargement of Romania’s territory at Austrian expense, following Allied victory, and Romania, unwisely, now decided to take the plunge. On 17 August, a convention was signed by which France and Russia bound themselves to reward Romania, at the peace, with Transylvania, the Bukovina, the southern tail of Galicia, and the Banat, the south-west corner of Hungary; in secret, the two great powers had previously agreed not to honour the convention when the time came. That the Romanians could not have known the treaty was made in bad faith does not excuse their entering into it. Good sense should have told them that their strategic situation, pinioned between a hostile Bulgaria to the south and a hostile Austria-Hungary to their west and north, was too precarious to be offset by the putative support of a Russian army which had only belatedly returned to the offensive. It was Brusilov’s success that had persuaded the Romanians to take the plunge from neutrality into war, but his success was not great enough to guarantee the security of their flanks against a German intervention or an Austrian repositioning of divisions; against a Bulgarian attack it could offer no assistance at all.
The Romanians nevertheless went to war on 27 August in apparent high confidence in their army of twenty-three divisions, formed from their stolid peasantry, and in the belief that the Russian offensive north of the Pripet Marshes, towards Kovel, would prevent the transfer of German reserves towards Hungary, while Brusilov’s continuing offensive would hold the Austrians in place. They appear to have made little allowance for the eventuality of Bulgarian or, as came to pass, Turkish intervention and they overestimated the military potentiality of their armed forces, which were poorly equipped and owed their reputation for fighting power to their success in the Second Balkan War at a moment when Bulgaria was hard pressed also by the Serbs, Greeks and Turks. Alexeyev, the Russian Commander-in-Chief, in a rare flash of realism, actively discounted the value of the Romanians as allies, rightly reckoning that they would drain rather than add to Russian reserves. He certainly did little to assist them. Nor did the French and British at Salonika, whose undertaking to mount a diversionary offensive had been a major consideration in bringing Romania to declare war. In the event their attack was pre-empted by the Bulgarians who, forewarned by evidence of Allied preparations and with the assistance of the German and the Turkish divisions, took the Allies by surprise, on 17 August, defeated the refugee Serbian army at Florina and succeeded in postponing the main Franco-British offensive until mid-September.
The Romanians, in these deteriorating circumstances, opened an offensive all the same, not, as the commanders in Salonika had expected, against Bulgaria, where it might have lent support to and been supported by their own, but into Hungary through the passes of the Transylvanian Alps. Retribution was quick to come. The Austrians quickly organised the local defence forces into a First Army, under General Arz von Straussenberg, while the Germans found the troops, some Bulgarian, to position two armies, the Ninth, under the ex-Chief of Staff, Falkenhayn, and the Eleventh, under the old Eastern Front veteran, Mackensen, in Transylvania and Bulgaria. While the Romanians, having occupied eastern Transylvania, then did nothing, their enemies made their preparations and struck. On 2 September the Bulgarians invaded the Dobruja, the Romanian province lying south of the Danube delta. On 25 September Falkenhayn, whose troops included the formidable mountain division known as the Alpenkorps, in which the young Rommel was serving, made his move in Transylvania and began to push the Romanians back through the passes towards the central plain and the capital, Bucharest, which fell on 5 December. By then Mackensen’s army had crossed the Danube and was approaching Bucharest also. Assailed on three sides by four enemies, for the Turks had sent the 15th and 25th Divisions by sea to the Dobruja, the Romanians had been thrown into full retreat towards their remote eastern province of Moldavia, between the Sereth river and the Russian border. There, as winter closed in, and with support from the Russian Fourth and Sixth Armies, they entrenched themselves on the Sereth to sit out the bad weather.
Their decision for war had been disastrous. They had lost 310,000 men, nearly half as prisoners, and almost the whole of their country. Their most important material asset, the Ploesti oilfields, at the time the only significant source of oil in Europe west of the Black Sea, had been extensively sabotaged by British demolition teams before they were abandoned to the enemy. The Allies’ decision to entice Romania into the war had been ill-judged also. The addition of the nominal fighting power of lesser states – Portugal (which became a combatant in March 1916), Romania and even Italy – did not enhance the strength of the Allies but, on the contrary, diminished it, once the inevitable setbacks they underwent came to require the diversion of resources to shore them up. The defeat of Romania not only necessitated, as Alexeyev had foreseen, the commitment of the Russian armies to rescue them from total collapse. It also delivered into German hands, over the next eighteen months, a million tons of oil and two million tons of grain, the resources that ‘made possible the . . . continuation of the war into 1918’.68 The accession of Greece to the Allied side, through a coup stage-managed by Venizelos but engineered by the Allies in June 1917, equally brought the Allies no advantage at all and, by its installation of a violently nationalist and anti-Turkish government in Athens, led to Greek mobilisation in the cause of the ‘Great Idea’ – the recovery of the Greek empire in the east – which would complicate the Allied effort to resettle the peace of Europe for yearsafter the war had ended.