As the New Year dawned the Germans had the initiative: Russia had collapsed, freeing most of the German and Austrian troops on the Eastern Front. But the US was gradually deploying troops to Europe, so the Germans had to move fast or these fresh troops and the new Allied resources would grind them down. Ludendorff told his staff ‘every thought must be turned away from trench warfare and towards the offensive … on the defensive, the army was bound to succumb little by little to the constantly increasing superiority of the enemy in men and materiel’.
The Germans pulled many divisions (and almost all men below 35 from the other units) out of the East. Many were retrained in Western Front tactics, and the thousands of guns were calibrated at firing ranges well in the rear,1 although that would be the cause of some controversy. Fifty-six divisions were well enough trained as stosstruppen, storm troops, to be considered attack forces, but some others simply got a couple of weeks’ training for mobile warfare, just shaking off the mud of the trenches, rather than full-scale training in infiltration, using minenwerfer, infantry guns, flamethrowers and the panoply of weapons. Stripping the East was not the limit of Germany’s reorganisation: they also cut battalion sizes to more closely approximate their actual strength, and they pulled most of the younger men out of ‘trench’ divisions in the West. There was certainly a cutting edge of elite units, but the rest of the German army became a low‐quality, line‐holding force.
Munitions were less of a problem with only one front, but tactical distribution would remain difficult. There were too few horses (and the cavalry units were in the East where they could forage in the countryside), and there was a shortage of rubber for tyres. Steel tyres meant that lorries had no cross-country mobility, and chewed up the roads. Germany had only some 30, 000 lorries (and it took 150 lorries to equal the cargo capacity of a single train) so the Germans were highly dependent on railways. Inevitably, they would have substantial problems supplying any breakthrough and sustaining momentum.
Austria was staggering. The economy was collapsing, and confidence in the government was fading; only the hope of Germany winning the war kept them lurching forwards. There were not even hopes of drawing resources from Rumania and Russia (the Germans made sure they got the lion’s share) and the public was beginning to wonder whether victory might make anything better. Bulgaria was also fading. Her economy was mainly agri-cultural, and with so many men mobilised for so long, there was no labour. Turkey was a mixed bag. The Russian collapse meant one front went away, but the Turks decided to take advantage of that and grab land in the Caucasus rather than switching troops to hold off the British in Palestine and Mesopotamia – where they found themselves facing commanders adept at using the advanced artillery planning systems developed on the Western Front. Mesopotamia was a long way from anything vital but the Palestine front was wearing thin.
The British and French were holding on militarily as well as politically. After years of attacking, the armies were now making defensive preParations. Manpower was short. The French disbanded divisions (and took a few regiments of black troops from the US), while the British maintained the number of divisions by reducing the amount of infantry in each by a quarter. Knowing the Germans would attack, the British kept their other fronts quiet and also transferred more and more Indian troops to Palestine and Salonika to free up British troops for the Western Front. Factories were humming, and munitions would be no problem in the coming year. The British would be able to replace all guns lost in battle and from wearing out, and upgrade their quality as well.
Italy also recuperated on the defensive after Caporetto. Men were mobilised to refill the ranks, units were re-equipped and discipline restored. There were even some doctrinal improvements as the lessons of the past few years filtered into the army after Cadorna’s dismissal. France and Britain each sent a corps of troops (the US sent a single regiment to show the flag), while the Italians sent a handful of divisions to the Western Front to reciprocate.
The US was mobilising as fast as possible, but going from an army of 100, 000 to one several million strong was hard: there were few officers and NCOs, and schools and infrastructure were structured for a small force. Factories were switching to military production, but it was easier to make olive-drab socks than tanks or artillery. Moreover, many of the factories producing munitions were already fulfilling contracts for Britain, France or Italy, so their new capacity was limited. The US was forming divisions and training individuals, but there were few guns to train on. The plan was to issue artillery units with their guns once they reached France and give them several months of phased training ‘over there’.
The Germans had been mulling where to attack – British or French; north, central or south— and started preindenttions in several sectors to deceive the Allies.2 Digging gun positions, dumping ammunition and improving roads were visible to Allied aerial reconnaissance, so while they were real preparations they also assisted in the deception plan. Lack of guns (and the ability to rapidly move the guns they had from place to place) meant there would be no diversionary attacks to mislead the Allies and draw in reserves. There were some minor operations (on the French front) and troops all along the line were allowed to think they were taking part in something much bigger so any prisoners would mislead the Allies.
Ludendorff ultimately chose the British Fifth Army’s sector because it had weak defences, not because of a key objective that could be reached; he famously wanted to break a hole and trusted that opportunities would develop as the battle evolved. The St Quentin/Somme area was selected not because it was particularly strategic, nor because it was where the French and British sectors met and it would be easier to split the two nations; instead, the ground would dry out here sooner than elsewhere and the attack could be launched weeks earlier than in the preferred sector, Flanders. While a strategic commander does have to work within tactical realities (something Haig should have borne in mind on the Somme), Ludendorff’s mistake was in completely subordinating strategy to tactics: ‘The whole matter depends on the advance of the artillery, and that depends on terrain.’3 The focus on fighting and ignorance of logistics were typical of the German General Staff. (Ironically the Somme could have been an incredibly important sector because of the BEF’s railway bottleneck at Amiens, but the Germans tended to analyse their plans at the tactical level and did not recognise the BEF’s logistical weak points.) Even if the Germans broke through they would quickly be in the area they had devastated in the spring of 1917 as they pulled back to the Hindenburg Line, and then in the Somme battlefield of 1916. Those areas would test the German logistical capabilities while also giving the Allies ready-made defensive positions.
With Ludendorff looking at the front lines rather than at a deeper objective, the Germans assembled 6, 473 guns and 3, 532 mortars – 48 per cent of their guns and 40 per cent of their mortars.4 Three armies would be attacking, but there was no single artillery commander: instead each army had a different chief. Among the Germans coming west was Oberst Georg Bruchmüller. While he was important in the upcoming offensive, he was not the key figure that he is sometimes made out to be. Bruchmüller certainly brought new techniques (including the ‘Pulkowski method’ of calibration and allowance for daily weather variations, which was essentially what the British had been doing for several months5) and his own gift for matching the right mix of munitions to targets. While he influenced Ludendorff (especially in getting the ‘Pulkowski method’ widely adopted so that the whole bombardment could be predicted instead of registered, thus preserving surprise), he was only in direct charge of the artillery for the Eighteenth Army. This meant that the impact of his talents was limited; the Second Army paid close attention to his suggestions but dedicated Westerners in the Seventeenth Army had to be forced to listen.
The Germans had not made a major attack on the Western Front since Verdun, over two years before. Bombardment methods had changed a great deal since Verdun, where areas had basically been drenched with shells; if the target area was trenches it was a ‘trench bombardment’ and if the area had French guns it was called ‘counter-battery fire’. In 1918 far more precision was used, along with different types of artillery and a range of specialist shells. Bruchmüller was extremely confident in the Pulkowski method and planned to use a wholly predicted barrage. He also mixed gun types and shell types based on the effect he was trying to achieve, not on tradition. For instance, in the past counter-battery fire was often done with howitzers because they fired larger shells that were more likely to destroy enemy guns; meanwhile, field guns would fire at infantry. Bruchmüller turned this assumption around, trying to break up the infantry’s positions (especially headquarters, observation posts and strong-points) with howitzer shells while field guns swamped the enemy artillery with gas shells. (He also mixed gas types, using a tearing/sneezing agent to make British gunners take off their masks and breathe in lethal phosgene.) At different phases of the bombardment Bruchmüller mixed field guns, field howitzers, gas and high explosive for counter-battery fire, and used various types of gun and shell to support the infantry. Trench mortars, minenwerfer, were used wherever possible, including during the barrage, but due to range limitations they could only reach the first British line. Bruchmüller kept the supportive role of artillery in mind all the time (he later wrote ‘the thanks of the infantry, in my opinion, must be treasured more by every artilleryman than all decorations and citations’6), but took a wide range of routes to the objective. The artillery commanders for the Second and Seventeenth Armies were far more traditional; they lacked Bruchmüller’s (uniformly successful) Eastern Front experience and confidence in his unorthodox methods. For instance Lieutenant-General Richard von Berendt of the Seventeenth Army would not fire a wholly predicted bombardment; there was a little registration fire before the attack, and he also had a pause in the fire on the morning of the attack to adjust fire. (Ludendorff had had to specifically order the Seventeenth Army to use the new principles and even suggested they use the same procedures, but the order came only two weeks before the attack, while Bruchmüller had used seven weeks.7) Another difference was in trusting the infantry. All the armies had a feuerwalze to cover the advancing infantry (including a creeping barrage advancing 200 metres every 4 minutes),8 but Bruchmüller ended the fire at the guns’ maximum range so the infantry could continue their advance, while von Berendt continued the barrage to protect against British counter-attacks.9
While there was a little registration fire, the Germans took strong measures to achieve secrecy,10 including deploying most artillery as late as possible. The large number of guns, and accepting neutralising fire rather than destructive fire (both for counter-battery and against the infantry) allowed lower accuracy, which in turn allowed more surprise. Camouflage was also emphasised, and work was done all along the line to dupe the Allies. (This worked quite well against the French, who were concerned about being attacked.) Where precise fire was needed, the target was usually assigned to a battery that had been in the sector a long time and had already registered.
There were two more sources of artillery support during the attack. The forward battalions had infantry guns that used direct fire; sometimes they were even hauled forwards by men with ropes. In addition, begleit (escort) batteries would move with reserve regiments (typically the sections of the battery would leapfrog so some fire support was always available) and use observed indirect fire. Both would be more responsive than the main barrage, although the Germans tried to have as much flexibility as possible for that as well.11 Some infantry units hauled small balloons forwards to mark their positions – obviously a double-edged sword since it would show the Allies where they were as well. Flares were also used, but as the Germans were counting on infiltration to break through somewhere and spread out, and the troops could not bring the barrage back – the timetable was sacrosanct – the flares could only tell the gunners to lengthen the range. Observers stayed in front-line observation posts, and others were sent forwards, but all would be at the mercy of circumstances, the proverbial fog of war. In those circumstances the infantry guns and begleit batteries were even more important.
It is easy for historians writing in warm, dry buildings to discuss numbers of guns and various planning factors but each of those guns was manned by very ordinary humans, who had to shove their weapon the last few feet, unload over 2 million rounds of ammunition from wagons and store them, dig gun-pits, and string up telephone and telegraph wire. Roads had to be mended: one of the British techniques to judge the sincerity of apparent attack preParations was to shell roads repeatedly; if the Germans doggedly mended the roads, an attack was likely. The staff, often maligned for working in comfort far behind the lines, had to devise and revise plans as circumstances developed. Every change of a battery or battalion, every alteration of a timetable, every move along a road had to be calculated. A unit’s march-speed would differ depending whether it was day or night, and unit strength affected how much road space was needed. With those thousands of guns to move, plus the hundreds of trainloads of ammunition, and hundreds of thousands of infantry as well, the staff may have been safer but their work was not necessarily easier. On 19 March Herbert Sulzbach noted he ‘could hardly get along the roads – it’s a mass concentration of troops completely impossible to describe’, and the next day ‘there are still troops, troops and more troops moving up to the front’.12 Ernst Junger’s unit was moving around as well: ‘All the roads were crowded with columns on the march, eagerly pressing forwards, with countless guns and endless transport… . Woe betide any unit whose movements were not up to sched-uled time! They were ruthlessly relegated to the ditch and had to wait hours before they found a gap into which they could squeeze.’13 This was no accident: German plans phased the arrival of new units.14 First echelons (including artillery staff and munitions columns) arrived over 1–5 March; 8–10 March saw air units, engineers and supply units arrive; 9–20 March was timetabled for infantry and artillery; these overlapped with the medical and bridging units over 17–20 March. Sulzbach gratefully noted the overcast and rainy weather that hid the assembling forces from British fliers.
Bruchmüller used the time for his own detailed planning. Typically a corps would be attacking from the sector of a single defensive division, and he made use of local knowledge. Senior officers from the stellungsdivision would take on counter-battery or bombardment-planning roles for the attacking corps. The soldiers of the stellungsdivision also started work on the ammunition dumps, observation posts, communications network, camouflage and survey duties. With the work started, it was easier for the reinforcing units to continue it rather than starting from scratch. Most of the planning took place before the troops arrived. Leutnant Kurt Fisher of the 464th Regiment noted: ‘Not a single gun was present – small yellow wooden stakes stuck in the ground instead, hundreds of them, under every bush and in the open. Two stood in line for the wheels of a gun, a bigger post behind for the trail. Every gun position was fixed geometrically [by survey].’15
The French had shortened their line by handing a sector over to the British Fifth Army; since the defences were in poor condition the British stayed generally quiet rather than indulging in ‘active trench warfare’ activities like bombardment and counter-battery fire. (Over the winter of 1916/17 the Fifth Army had fired over 2 million shells and never attacked in more than battalion strength; in the winter of 1917/18 they only fired in response to the Germans.) This quiet meant the Germans had little reason to conduct similar operations and allowed the BEF time to absorb replace-ments, reorganise divisions and build up defences. Lieutenant Arthur Behrend noted:
Since the Cambrai ‘show’, the whole brigade had suffered barely a dozen casualties. Indeed, since Christmas, life on the Bapaume front had been delightful – a succession of invigorating canters across the overgrown downs to the batteries, joy-rides to Amiens through the snowy wastes of the Somme, early morning partridge shoots over the fields around our headquarters.16
The British were not simply enjoying themselves; worse, much of their work was not directed at strengthening the trenches that were about to be attacked.17 Rear-area facilities (for instance stables, baths and billets) had to be built or improved. The longer British line meant a thinner line. Two things could have compensated for their precarious position: better defensive doctrine or better defences. The BEF had not faced a major German attack since 1915 at Ypres, and defensive doctrine for infantry and artillery alike was badly out of-date – the artillery pamphlet was titled ‘Artillery in Holding the Line’, suggesting how rigid the concept was. (The report on German techniques at Cambrai stayed at GHQ for far too long.) Haig’s operations staff wrote new manuals over the winter of 1917/18 but they were not published in time to make much difference, although their principles were being implemented before actual publication. They also ran against everything the British troops had been told for years: it is difficult to change any organisation quickly and the BEF was no different. Troops were accustomed to standing and holding (and commanders had been sacked for losing ground), and while the idea of falling back and then counter-attacking could be taught, it was not a concept that the British army readily embraced. For instance, artillery were told that ‘repulse of attacks’ was the third priority, behind ‘destruction of the enemy’s fighting power’ and ‘hitting communications’, but they still fired barrages whenever the infantry put up a SOS flare.
Moreover, there was not enough labour to dig the necessary defences for defence in depth. (W. H. F. Weber, an artillery brigade commander, noted that he had to dig positions not only for his own batteries in each line of defence, but also for a reinforcing brigade – so his four batteries had to prepare forty-eight battery positions, plus anti-tank gun-pits.) Britain was short of manpower, and the Cabinet made the army only its seventh priority: coalminers, farmers and munitions workers were more important not only to the British war effort but to the inter-connected Allied war effort. While that was theoretically prudent, it still left the BEF short-handed for digging defences and manning them. There were maps at divisional, corps and army headquarters with neatly coloured lines showing Forward, Battle and Rear zones of defences – while on the ground the Rear (and sometimes Battle) zones were only staked out, to be dug later. (The very open terrain made it hard to disguise positions from German air recon-naissance.) Haig was also substantially responsible for the St Quentin sector remaining weak: he looked at his line in much the same way that Ludendorff did and kept more troops in the centre and north. There was space to trade for time in the south, and if the British troops were thrown back, there was a good chance the French would come and help. So Hubert Gough’s Fifth Army was left far weaker than it should have been, with only nine infantry divisions and three cavalry divisions (with a total of 1, 566 guns) to cover 42 miles of front. (In contrast, the Third Army, which would take some of the German assault, had 14 infantry divisions for 28 miles of front.) With those sparse troops spread out over the three defensive zones (or even the front two), the artillery was also spread out (with around 10 per cent of the field guns scattered forwards as anti-tank guns), so concentrating firepower required even better communications than normal. And command was absurdly centralised: any counter-preParation fire required the approval of a division commander, and then brought down rather feeble fire: a field battery might be covering 200 yards – while in the attack it would cover at most 50 yards.
The regulations for the ‘Attack in Position Warfare’ of January 1918 noted that ‘in every offensive action it is of decisive importance to produce surprise’, and added that ‘rapid progression [of the attack] offers the maximum protection and secures success’. The German army was deter-mined to demonstrate these principles in the spring offensives. German morale was quite high; the troops believed they were winning the war, and were pleased to be finally attacking again after years of fending off the enemy, gradually losing ground and comrades. In contrast, the British troops were tired from digging all the time, and dispirited by the continual alerts. Weber commented derisively on ‘daily, even hourly, warnings of an overwhelming attack to be supported by tanks of fabulous speed, size, armament, pattern and number’.
At 05. 40 hrs on 21 March 1918 the German guns thundered. A German infantry officer described the sensation: ‘If you put your hand over your ears and then drum your fingers vigorously on the back of your head, then you get some idea of what the drumfire [a hurricane barrage] sounded like to us.’18 An artillery observer had a different perspective:
Within seconds of the bombardment opening, we could see sparks and columns of fire in the enemy’s trenches and their rear areas. A terrific roar, an immense noise greeted the young morning… . In the past the French and the Tommies had bombarded us for seven days without a pause; we would now do it for 5 hours. We laughed and looked happily at each other. Words were useless; the hell of the inferno outside saw to that. There was only lightning and noise.
The first salvoes of shells were largely gas (to catch men who were in the open or sleepy and slow to put their masks on) and high explosive, and focused on the front lines and headquarters. A British machine-gunner recalled: ‘It seemed as if the bowels of the earth had erupted, while beyond the ridge there was one long and continuous yellow flash. It was the sudden-ness of the thing that struck me most, there being no preliminary shelling but just one vast momentary upheaval.’19 After a few minutes everyone who was going to be caught by surprise was already caught, and the Germans shifted targets. Counter-battery fire was the next priority, once the first phase had drawn the gunners from their dugouts to their guns, and gas continued to play a major part because without gunners the guns were useless. Once the enemy guns were neutralised, counter-battery fire dwindled to a sustainable level and most guns switched back to pounding the enemy infantry. There were plenty of targets – wire to cut, strong-points to hammer, headquarters and communications links to disrupt, guns to silence – and the Germans shifted around among them. Locations would be shelled for a while then left alone, then the fire would return in case anyone had put their head up. Minenwerfer pounded the front lines and cut barbed wire, and infantry parties moved around to gather information about the progress of the attack and to snip wire where necessary. During the 5 hours of bombardment the Germans fired 2. 3 million rounds.
There is no way to calculate British casualties from the bombardment but that was not the point of this deluge of shell-fire. It sought to break British command and control, shattering the defenders into groups who could not coordinate their fire or even call for help. By and large it worked. Weber remarked that his gunners could cover their front, but could make no difference on the flanks. With the British trying to distribute in depth but weak and clumped up front (in strong-points that one old sergeant dismissed as ‘bird cages’), the infantry were very vulnerable to infiltration as fog and the rolling barrage covered the advancing stosstruppen. German troops moved around British strong-points and their light minenwerfer and infantry guns gave them an edge: they had weapons the British infantry generally lacked. There is nothing more demoralising than being hit by weapons to which you cannot reply, and plenty of British troops honourably surrendered when they found themselves out-flanked and out-gunned.
In the Fifth Army sector the British artillery was willing to fight (despite the gas and counter-battery fire) and did what it could, but communications were impossible: the telephone lines to forward headquarters and observation posts were wrecked, and the fog obscured the SOS flares. The forward anti-tank guns often fired off their ammunition (as did the ‘sniping’ guns that did much of the harassing work) because the Germans were almost on top of them and they could see targets. In contrast, ‘silent’ batteries were detached from their brigades and stationed in the rear to surprise the Germans. Many of them never received orders to fire and simply waited: such was the legacy of top-down control established by the years of trench warfare. One gun unit waited too long and surrendered: ‘We came upon a position for four field guns… . The infantry had all gone, leaving the artillery on their own. They gave up quietly, forty or fifty men including some officers. None of them had been killed at all.’20 At least the field artillery was mostly able to withdraw when the need arose; much of the heavy artillery was immobile (in winter the horses were frequently with-drawn to rear areas for better stabling) and had to be disabled before the Germans reached the positions. There was gallantry aplenty that day and in succeeding days, but it availed little at first. One gun, for example, fired 1, 000 rounds and held up a German regiment (with a subaltern covering the gun’s withdrawal with a machine gun),21 but the Germans kept coming and overwhelmed the Fifth Army. For instance, XIX Corps had eight battalions holding the Forward Zone; at the end of 21 March these units mustered only 50 men ready for duty.
In the Third Army matters were substantially better for the defenders. With more infantry per mile, there were fewer gaps for infiltration. Better defences (especially a very deep forward zone, so the battle zone was out of range of the minenwerfer) meant the Germans needed a stronger bombardment, but not only did the Germans have fewer guns – especially heavy guns – per kilometre (despite recognising this as a key sector), but their stodgier commander used them less effectively. Soggy patches of ground slowed the Germans, especially hampering the gunners moving forwards where the stormtroopers had made some progress. Yet with the Fifth Army cracking under the repeated hammer blows, the Third Army soldiers also had to fall back to avoid being out-flanked.
That day the Germans shattered the deadlock but they did not break through the British line: they broke it open and destroyed the command system, but this tactical success did not mean a strategic triumph was now inevitable. The British had taken heavy casualties on 21 March (about 38, 000, plus some 400 guns, while the Germans suffered roughly 35, 000 casualties), and it would be two weeks before the defences stabilised, largely because the defenders would have trouble coordinating themselves even as reserves poured into the sector. Units could not coordinate with their neighbours and were extremely cautious about their flanks – a minor penetration (or a rumour of one) between two units could cause both to pull back, and that would ripple out along the Allied line. (French reserves soon joined the battle, but many of the first units arrived without their artillery and had little firepower.) It was easy to identify communications as a problem – one battery commander said ‘the more [telephone] wire you give us the more Huns we shall kill’ – but it was hard to connect units that were on the move and harder still to keep them connected. Of course, conditions were difficult for both sides: on the 21st the fog helped the German infantry infiltrate, but it lasted into mid-afternoon which made it hard to keep track of where the infantry was – flares, balloons and aircraft were of little use. After the first day the fog was less helpful to the Germans. The infantry still appreciated the cover it gave, but it actively hampered the artillery: on 21 March the British positions were generally known, but after the first day the Germans were also groping in the fog.
For those two weeks it was mainly a soldiers’ battle, with battalions and batteries at times fighting semi-independent engagements. British batteries often engaged over open sights, and battery commanders spotted their own gunfire. They might well hold off a German advance, but with their flanks left ‘in the air’ units were falling back daily. Nobody can question the gunners’ individual valour, but there was little opportunity for the artillery to have more than a local effect on the battle. Ammunition was often a problem since the shells dumped with the guns were left behind with the first move, and the light railway networks were laid out for the entrenched front. (Keeping the men fed was also a problem; Weber’s group dined on the 26th on ‘a tin of pork and beans from an abandoned dump, eaten off an envelope with a pen-knife’.) When a load of ammunition arrived it might have to be abandoned (or blown up) because nobody knew it was available. Divisional artillery commanders sometimes found themselves simply working as senior liaison officers, trying to pass information around from the quartermaster’s staff to the guns.
The Germans were having problems too. Moving guns forwards through their old trench system was a simple exercise in planning, complex enough but predictable: routes could be prepared and bridges pre-positioned. Building roads across no-man’s-land and then through the British lines was harder, and every kilometre of forward progress was a kilometre further from the supply points (up to 69km by the end of the first week) and many kilometres of communications wire would be needed to not only go forwards but laterally as well. It could take a full day for orders to get from an army headquarters to an infantry division, and then more time for it to reach the combat units. Moving across the old Somme battlefield did not help. Officer Candidate Paul Knoch wrote home:
The area where the Battle of the Somme was fought, around Peronne etc., is a scene of terrible destruction. Crater after crater, some villages such as Bouchavesne have disappeared completely, only metre-high stone walls remain. A few splintered stumps is all that remains of the trees. The roads are littered with dead horses, corpses of Germans, Englishmen and Frenchmen, strewn with equipment, weapons, ammunition, here and there damaged or abandoned English guns. One gets used to such sights, but one tries not to think about it.22
The infantry guns and begleit batteries could help, and they carried extra ammunition. But even the double stocks of the begleit batteries could be fired off in only 24 minutes, so it was hardly a solution. The Germans were so short of horses that the begleit batteries had to borrow (for how long nobody knew) horses from other units for their extra caisson of ammunition. And the heavy guns and minenwerfers could not be moved easily –some of them had to be broken into several loads and were not even scheduled to advance (they formed a ‘battering train’ that would be moved sideways to support each of the German offensives) – and all suffered from Germany’s general lack of horses and lorries. By the 26th the German infantry was pulling ahead of the artillery and the artillery was having to cut back on shells because it was ahead of the supplies; III Korps had to cut its ammunition allotments in half that day.23 The Germans had not had to deal with this situation before, and they encountered some of the logistic problems the Russians had faced at Tannenberg: they were marching men and moving supplies by horsedrawn wagons while the Allies were moving men and supplies by railway and lorries.
Sulzbach’s unit was a begleit battery, sometimes engaging in direct fire and sometimes indirect, but the pace was draining.24 For the first week casualties were light, although air attacks were more of a problem than in trench warfare as there were no dugouts to jump into. He was able to manoeuvre through a doublebarrage, probably because it was fired blind, without an observer adjusting fire to catch the moving Germans. Nine days into the battle he had a close encounter as a German attack blundered into French troops, infantry and artillery:
I bring the battery up behind, and now we’ve got so much shrapnel rattling down on us that you can hardly hear or see anything. The machinegun fire, chattering away at us from only a few hundred metres distance, keeps on as heavy as ever. All hell has been let loose. The French seem transformed, they must have thrown completely fresh, properly rested, troops into the sector, and a large number of them too… . We pull up a steep track onto a plateau, and there is our No. 3 Battery next door to us. And up there it’s a witches’ cauldron, compared with which the business we had before was child’s play: the machinegun and small-arms fire so strong that it might have been thousands and thousands of enemy gun-barrels being trained on our one Battery. The concentration of fire is so heavy that all we can do is lie on the ground beside the guns, with the infantry hardly 300 metres in front of us; and we haven’t reached the peak yet, because suddenly we start being fired on from the right and left flanks as well, and it looks as if we are on a pointed wedge of ground offering a marvellous target to the French on all three sides… . Our own attack never gets off the ground … just in front of us, 2/Lt Mayer of No. 8 Battery gets killed in action; that’s the tenth officer our Regiment has lost since 21 March. [At this stage of the war German batteries might have only two or three officers.] Now it gets even ‘lovelier’. Our infantry start coming back, in groups or singly, because they can’t stand it any more up there at the front, and finally there they are lying between the guns.
German casualties mounted, and units simply got tired as they moved further and further from their supply points. (Looting Allied supply dumps was hardly a long-term solution, not least because the German troops were demoralised to see how lavishly supplied the British were. The landser had been told that the U-boats were sinking hundreds of ships, yet now they saw that the ordinary British Tommy ate better, was dressed better and had more ‘comforts’ of life.) Ludendorff also proved an indecisive commander, spreading his resources and reinforcing in areas that offered no strategic results and refusing to withdraw units as they wore out.25 By early April the Germans had inflicted just over 254, 000 casualties on the Allies (including around 75, 000 on the French) but had lost almost 240, 000 men of their own
– hardly the way to win the war, especially with Americans arriving every day; even capturing 1, 300 guns was unimportant because Allied munitions factories were already producing more than they needed.
The next real German attack (a one-day attack on 28 March at Arras was slapped back) took place in Flanders between 9 and 29 April. On the first day a weak Portuguese division broke. Although the 55th (West Lancashire) Division held firm against the stosstruppen, the Portuguese hole created flanks and the Germans gained ground, although they never threatened to break through unless the British collapsed. Haig was sufficiently alarmed to issue an Order of the Day beginning ‘With our backs against the wall and believing in the justice of our cause’, but again the German attacks weakened and French reserves bailed out the BEF. One thing Ludendorff had learned was that putting Bruchmüller in charge made the artillery run better but even the individual genius of Bruchmüller (with 2, 210 guns under his control) could not wipe out stout defences and determined defenders. The British had held firm against two fierce offensives but Ludendorff identified French reserves as the reason, so he aimed his next offensive against the French. Hindenburg wrote in his memoirs: ‘Twice had England been saved by France at a moment of extreme crisis’ through ‘massed attacks and skilful artillery.’26
The British were also looking for ways to improve their defensive performance. Infantry lost the ‘right’ to supporting firepower; SOS signals now were for information and not a demand for fire-support. Counter-preindenttion had to be fired as intensely as for an attack, not just desultory fire to hamper attacking infantry. Guns might be sacrificed if the gains – in dead Germans, or a sector held – outweighed the losses. And the training on mobile artillery operations that had seemed so useless over the winter of 1917/18 now had a relevance that even the lowliest soldier could see, and units out of the line trained as much as they could. With the Germans attacking the French, this training would pay off when the British finally attacked, and it would pay off handsomely.
After the May offensive ran out of steam, there was a pause on the Western Front. The Germans were largely spent; Ludendorff knew he lacked fresh troops to attack soon, but he could not pull back both for diplomatic reasons and because of domestic politics in Germany. Unable to decide what to do, he dithered while hoping that Germany could regroup faster than the Allies could. The German troops were not optimistic and aggressive (a result of concentrating the younger and more aggressive men in the stoss divisions earlier in the year), but they also were not mutinying nor surrendering. Many would not waste time entrenching newly gained ground but they would fight reasonably well if attacked. In some places the logistic system had problems bringing supplies forward (even basics like barbed wire) but Germany was also running out of food and bandages.
The Allies were feeling far more confident – they had weathered the storm so far – but were not certain who had the advantage. They had a supreme Commander, Ferdinand Foch, who was aggressive by nature and ordered the Allies to commence a series of small attacks. If these went well, they would restore confidence among the troops and on the home fronts, keep the Germans off balance, clear bulges in the line and exploit tactical advantages. If they did not go well, they could be broken off until there were (in Petain’s phrase) more tanks and more Americans. Foch was supreme Commander but he was more a coordinator of the various countries; he encouraged them and set the tone but could not fire his subordinates. The decision to launch a number of small attacks rather than try a major offensive was the right one. Typically the Germans were poorly fortified, and there could be surprise attacks even without tanks; there simply was not much barbed wire to cut. A small attack could be broken off once it ran out of steam, when losses mounted and gains slowed. The succession of small battles drained the German reserves and were demoralising, since almost no unit could be sure it would not be attacked the next dawn. (By late September the Germans were breaking up divisions and cutting battalion strengths further, but nothing could hide the haemorrhage of manpower.) Once the Allies had pushed the Germans back to the original (spring 1918) defences, the German troops were demoralised and the Allies could bash their way through the Hindenburg Line, as the British and Americans proved.
Pétain and his staff recognised that they were likely to face a major German offensive in the spring of 1918 but unlike Foch, Pétain failed to see the Western Front as a whole and focused on the direct threat to his own men. His defensive strategy was based upon an initial defence phase where casualties would be minimised while the ‘first shock’ of the German offensive was absorbed and then a second phase with powerful artillery reserves rapidly transferred to bolster the defence and support limited counterattacks. The German assault on the Chemin-des-Dames Ridge in the Aisne sector, made on 27 May, was supported by 4, 000 guns firing a night barrage that lasted 2 hours and 40 minutes. The density was roughly one gun for every 7-metre section of the front line. Ernst Jünger would later write:
A flaming curtain went up, followed by unprecedentedly brutal roaring. A wild thunder, capable of submerging even the loudest detonations in its rolling, made the earth shake. The gigantic roaring of the innumerable guns behind us was so atrocious that even the greatest of the battles we had experienced seemed like a tea party in comparison. What we hadn’t dared hope for happened: the enemy artillery was silenced; a prodigious blow had been laid out… we gasped at the colossal wall of flame over the English lines, gradually obscuring itself behind crimson, surging clouds.27
The advances were impressive, largely due to General Duchêne’s failure to reorganise (and deepen) the Sixth Army’s defensive system, but the failure to exploit the initial successes and the survival of major flanking positions at Soissons and Rheims prevented a full-scale breakthrough. The next offensive was delayed until 9 June but by this stage the French were adept at recognising German offensive preParations. Even though the Germans only moved their guns forwards under cover of darkness (many to as close as 1, 000 metres from the front line) on 7 June, the French intelligence service discerned their objectives and the Third Army belatedly began to redeploy their reserves for prompt counter-attacks. This time the French counter-preParation shattered the cohesion of the assault and although the Third Army’s defensive preParations were incomplete, the French reserves stemmed the German tide. The intuitive approach used by the German stormtroopers left them disorientated when faced by an organised opponent, particularly when their enemy was able to deploy large numbers of tanks in a properly supported counter-attack (163 were used on 11 June).
On 9 June the Germans did penetrate the primitive defensive system between Noyon and Montdidier. But the Allies had learned how to defend: echeloned in depth and supported by reserves of artillery and infantry. Delighted by the tactical success, Ludendorff was pleased with the progress of his stormtrooper assault teams, but recognised that the artillery was the real key to making a full-scale breakthrough. That day he noted that ‘a greater density [of infantry] serves only to increase losses. The most vigorous resistance is broken more readily by reinforcing the artillery than by adding to the infantry …’. If the Allies built up their reserves then the artillery would have to be deployed even more flexibly to support the attacks that were making progress. The problem was identifying these before the Allied reserves moved into position. The German General Staff and commanders were slowly coming to the view that the process of breaking through was an objective in itself but that the cost of such offensives could quickly wear down the attacking units in pointless assaults on objectives other than those which maintained the momentum of the attack. In midJune OHL noted that gaining artillery dominance in the first phase was essential to success. Pushing artillery forwards to support minor breakthroughs could disperse it before the real opportunities became apparent. As a result, planners were advised to allow organic divisional artillery to advance but to leave the allocation of heavier reserves to the judgement of the corps commander. The final assault was expected to combine all the lessons of the spring offensive in one final push to break through the French positions along the river Marne. However, the Germans had used the same ‘battering train’ for each of their offensives and their artillerymen were weary and the guns increasingly worn, while Allied intelligence could hardly miss the arrival of the battering train.28 Without tanks, the Germans depended wholly on artillery and stosstruppen to create any kind of momentum and found that the process of force-marching reserve units into the breach tended to exhaust the infantry before they even entered the battle. The Germans had 78 fresh divisions in reserve in March, 62 in May, 43 in July and only 21 in August. Thus the Allied commanders were able to gather and redeploy reserves in good time, and because Ludendorff’s objectives were created by tactical successes, Foch was able to secure and support the strategic keys to the campaign and focus on building an everexpanding reserve capable of shattering the momentum of any German offensives. The initial successes of March and April impressed the attacking infantry and the artillerymen that supported them, but the failure to win the ‘Peace Offensive’ before July wore down their morale and made a direct contribution to their second defeat on the Marne.
Foch hoped that the Germans would persist in using the same assault tactics they had used since March and was convinced that Ludendorff’s clumsy strategy had created an opportunity for a series of coordinated counter-offensives once the Germans exhausted themselves creating pockets they would find difficult to hold.29 If the pressure continued, GQG hoped that the defensive lessons of 1917 would be sufficient to frustrate the offensive. The German attack was launched towards Chalons and Epernay on a front of 90 kilometres. Although they still outnumbered the French, many reserve units were still being reconstituted after being burned up in the spring offensives. The French Fourth Army had time to reorganise its entire defensive system, using the lessons of 1917, but the Fifth and Sixth Armies had less time to prepare and fought in the early 1917 dispositions, thus leaving historians an opportunity to compare the two systems. It is important to note that the British and French adjusted their front lines several times during the spring and neither was entirely satisfied with their ally’s defensive doctrine.30 As early as 9 April Pétain outlined his views on the initial attacks. Counter-attacks were to be organised as quickly as possible; the local commander’s discretion was preferable to detailed fixed plans; infiltration and flanking movements were to be encouraged; strong-points and enemy batteries should be rapidly identified and subjected, respectively, to destructive or neutralising fire; and potential routes of reinforcement should be cut off with curtain barrages. In summary, Pétain expected ‘rapid and violent preParation’ and the ‘full use of the [potential] flexibility of the infantry and artillery’. As Mangin noted after the war, ‘the enemy was a good teacher, but his lessons cost us dearly’.31 Collating the views of various officers on the initial battles, the Instruction of 10 May 1918 added details on the most efficient allocation of assets when making methodical attacks in position warfare.
Foch’s insistence on counter-offensive preParations enabled the French to gather 27 divisions, and the key was in recognising the culminating point of Ludendorff’s offensive plan. Mangin was selected to conduct an initial operation towards Soissons on 28 June and identified an opportunity that might bear fruit if the Germans blundered forwards again without realising they were over-extended. On 15 July, after a 3 hour and 20 minute bombardment, the French Fourth Army was attacked by 50 German divisions, 27 of them in the first wave. The French had 8 divisions holding the 42 kilometre front line and 5 in reserve. The stormtroopers penetrated the main battle zone but after seven bitterly contested attempts the Germans finally realised that the French defensive preParations and artillery reserves had completely neutralised the German system. On the Marne itself the Germans attacked with 13 divisions, after a bombardment that lasted 3 hours and 40 minutes.32 Securing the bridges across the river delayed the main attack but the Germans still penetrated 5 kilometres into the Sixth Army’s defences. French artillery and aircraft focused on interdicting the bridges, so the Germans were unable to bring up their artillery to support further advances.
Against the Sixth Army the Germans made major gains and drove some 10 kilometres towards Epernay. Foch was confident of holding the sectors being defended by the Fourth and Sixth Armies and unleashed his counter-attack between Sisne and Belley, sending Mangin’s Tenth Army crashing into the German right flank on 18 July. No fewer than 18 infantry divisions, 470 batteries (comprising 1, 545 pieces), and 469 light and medium tanks joined the Sixth Army’s 9 divisions, 230 batteries and 273 light and medium tanks. Mangin’s Franco-American force used no artillery preParation, instead attacking with a powerful creeping barrage (one battery fired a round per gun per minute for almost 8 hours), while Degoutte’s Sixth Army attacked after a 1½ hour preParatory bombardment. The Germans were completely surprised by the strength of the counter-offensive and were flung back 8 kilometres on a front of 20 kilo-metres. The Sixth Army and Berthelot’s Fifth Army joined the offensive and the German army was forced to recross the Marne and to withdraw from part of the salient that had been created in the spring.33 While the Germans would refine their offensive doctrine, they had lost the initiative and would never attack again.
Foch noted, ‘if we attack it must be to beat him, to disorganise him to the greatest extent possible … a battle engaged by us for this purpose must be pushed with the greatest possible speed and to the greatest possible distance, with the utmost energy’. Directive No. 5 of 12 July 1918 summed up the new French approach to warfare. Obtaining strategic surprise was essential to success, with maintaining secrecy being seen as ‘a question of honour’. Tactical surprise would be exploited by sudden attacks with short but violent preParation, using gas barrages and aircraft to neutralise the defending batteries. Deep penetration was to be enabled by aiming for distant objectives but employing ‘simple and concise orders, which leave the greatest part to the initiative and temperament of each for the accomplishment of his mission’. Tanks were recognised as a vital innovation to assist but the French recognised that German gunners could rapidly neutralise unsupported tanks.34 The artillery batteries were no longer the conductors of the orchestra of battle but were still the primary factor in failure or success: ‘the capture, as rapidly as possible, of the mass of the hostile artillery, in every case, will be the surest guarantee of rapid and deep progress’. Like the Germans, the French now used the preParatory bombardment to neutralise enemy batteries, promptly moved their own guns forwards to support the advancing units, including placing ‘batteries or sections of the artillery at the immediate disposal of division infantry commanders, regimental, or battalion commanders’. Over time the French learned when it was best to decentralise command and when a piece-meal approach was risky and it was time to reconcentrate. As Colonel Lucas noted after the war, the French now believed that an ‘audacious infantry, knowing how to dispense with the aid of the artillery; artillery, whose constant and primary preoccupation is constantly to support the infantry … seemed to be the true formula for the exploitation of success’.35 The British and French learned valuable lessons about the value of initiative during an improvised defensive situation, lessons that helped prepare them for the counter-offensives that would follow. Mobile warfare, à la 1914, had to be relearned.36 Many division commanders intuitively realised that artillery had to be decentralised, and directly allocated to front-line units; the problem was that liaison was still difficult and effective communications took time to set up.
After the Second Battle of the Marne, Foch coordinated a series of successive Allied offensives designed to scatter and destroy the German strategic reserves and shatter the German defensive system. Foch made sure that these offensives penetrated deeply enough to force the Germans to commit reserves to the battle but halted them before they created salients that might give the Germans a chance to regain the initiative. When the Germans withdrew to preprepared positions (shortening their line by 200 kilometres), the Allied armies rapidly advanced and immediately commenced preParations for the next series of offensives, preventing Ludendorff from resting his men and reconstituting his reserves. The Germans had worn out their best units in the spring offensives and Foch’s hammer blows continued to rain down on their defensive system, switching between pugnacious British and Dominion troops, rapidly improving Americans and the weary French. In contrast, the Germans were declining in numbers and quality. In May, forced by Lloyd George’s increasingly petty intrigues to undermine GHQ, Haig had considered creating B class British units to do no more than hold the line but was urged to abandon this decision by Foch, who argued that units of uneven quality would create opportunities for the enemy. As soon as the Germans were forced to adopt such desperate measures, Foch was delighted, recognising the problems for units without experience of liaison and incapable of tactical flexibility. ‘This is a weakness of which we must take advantage by undertaking important offensives.’37
Each successive offensive drove the Germans back, then forced them into delaying actions as they tried to consolidate their next position faster than the Allies could exploit, while the Allies moved their guns forwards to begin the grinding process anew. Both sides had to develop methods of alternating positional and mobile warfare to suit this new Paradigm but at least the deadlock appeared broken. As Mangin confidently observed, ‘it is time to shake off the mud of the trenches’. General Debeney, of the First Army, agreed wholeheartedly, noting in his orders for 8 August 1918, ‘I approve in advance every act of initiative, no matter what its result may be.’ The German defensive system (especially as thin as it often was) could be shattered with brutal efficiency by the Allies and there is little evidence for the Germans managing to formulate a strategic solution to complement their otherwise impressively elastic tactical approach to defence. The key to success in the assault was the availability of reserves, with even the most formidable positions falling when they were poorly garrisoned. Even vigorous counter-attacks had lost much of their value in the defensive, Ludendorff informing commanders on 4 September that ‘counter-attacks should be delivered only when there is a chance of success’, arguing that such an attack, ‘under unfavourable conditions, leads solely to useless losses’.
The first deliberate Allied step was the British Fourth Army’s attack at Amiens.38 Sir Henry Rawlinson wanted to attack to clear the Germans away from the key railway junction at that city, and detailed why it was tactically feasible as well as strategically desirable. His top two reasons were weak German defences and weak German defenders – a far cry from 1916 when the Allies attacked without much regard to the German situation. (The German defences were weak because they wanted to attack Amiens themselves, and thus had little incentive to dig elaborate defences; as long as they held the initiative there was little risk in weak defences, but when circum-stances changed they did not react. Troop strengths were also reduced by the influenza epidemic.)
While Rawlinson wanted to attack, he was also cautious. He wanted all the tanks in the BEF, the best infantry (the Canadians and Australians), and strong artillery support, eventually totalling 60 train-loads of artillery. By 1918 the artillery units formed in 1916 and 1917 were in action, and there were finally enough guns to do all the tasks at one time, so destructive and neutralising fire could be one simultaneous deluge of shells. He also wanted surprise, which would keep the Germans weak, and would also have other tactical repercussions. For instance, surprise meant counter-battery fire would be far more effective: the Germans would not be moving their guns around to dodge increasing British fire, but would be stunned on Z-Day. Since the Germans would not be reinforcing their artillery, the British barrage guns could be further forwards and thus the creeping barrage could reach deeper into the German positions. (Without a preliminary counter-battery battle, more British guns were functional on Z-Day – 98. 5 per cent.) A bit of registration could be concealed through having newly arrived guns fire the routine ‘daily hate’ and harassment missions, but not all guns could be registered. Surprise was so important to Rawlinson that he gained approval for a BEF-wide deception plan, with a series of rumours, and also made sure every man in the Fourth Army received a leaflet headed KEEP YOUR MOUTH SHUT. Revealing the build-up would put so much at risk that he decided the reinforcing guns would only fire if there were a major German attack – a simple probe or raid would not trigger massive defensive firepower – and the decision about what to fire was decentralised to the CBSOs, presumably because they would have plenty of information and they controlled much of the heavy artillery. It was a remarkable display of how the BEF could be at once centralised and decentralised.
The battle began at 04. 20 hrs on 8 August. Surprise was total, and most of the artillery elements worked like clockwork; even when mist meant the infantry could not see the barrage, they could hear it and follow it. Twothirds of the heavy guns were firing counter-battery missions, and 95 per cent of the German artillery had been correctly located. Many batteries were abandoned without firing (some still had their muzzle-caps fitted) since German artillerists’ morale was no higher than their infantry’s. (The Germans had been reorganising their artillery, and were now mixing a bit of heavy artillery with the field artillery as ‘close-range groups’ to support the infantry regiments, while a ‘long-range group’ of heavy artillery was responsible for other missions. It hardly mattered on the first day at Amiens, but they were trying out new methods.39) Other batteries were found demolished with their horses dead and the gunners fled once there was no point in trying to continue the fight. The Canadians had somewhat poorer artillery support since they had been the last to arrive and were not permitted to fire any preliminary rounds; it did not help them much that the French First Army (on their right) was also attacking but at H+40. The French had no tanks and needed a short bombardment, but for the first few minutes the Canadian flank was only supported by fire instead of by advancing troops. The Canadians did something unusual for counter-battery fire: instead of one battery firing on a German battery, they assigned one gun from three or four different batteries to each German battery. They knew their accuracy would not be perfect, but by blending in several guns with different errors, on average the targets would be hit. It worked, and they continued the technique through to the end of the war – despite the extra paperwork involved.
The creeping barrage was fired as far as the field artillery could reach, and then the barrage was held on that line until H+240 as a protective barrage for the infantry to mop up and reserves to start moving. The barrage then stopped, and the infantry/cavalry/tank forces could fight their own way forwards. Some field artillery did advance (about a brigade per infantry division, and it advanced when ordered to by the supported infantry) and there were thorough plans to open lanes in the wire and have signallers to connect the guns to the infantry, but these guns fired concentrations where Germans were resisting rather than a barrage. While the problems of cavalry–tank coordination are better known, it was equally hard to coordinate guns (even the Royal Horse Artillery) and cavalry because mobility was so different. The RAF provided valuable support to both infantry and artillery. Many zone calls provided information and targeting, and there was some strafing of German artillery where it was beyond the range of counter-battery guns. Smoke barrages had formed part of the original plan (not just providing some cover to the advancing infantry but blinding specific German observation posts) and in later phases, when smoke was needed beyond artillery range, the RAF tried dropping large quantities of smoke bombs.
The results on 8 August were excellent: up to 8 kilometres gained across the 24-kilometre front, about 18, 000 prisoners (plus probably 10, 000 killed and wounded who retreated) and around 400 guns. In yet another way surprise paid off: the Royal Artillery suffered no casualties from German shelling. When the attack continued on 9 August, progress slowed markedly. The Germans flung in more reserves than the BEF had anticipated, and the Allied forces were fatigued and had suffered casualties. They were also far less organised (16 British brigades would attack on 9 August, but they had 13 different start times as coordination broke down; individual attacks also had far less artillery support), and it was harder to move supplies forwards and get information back. All these were familiar problems in the Great War, indeed in warfare in general. Still, the Allies had the advantage and fought their way forwards, gaining ground and taking prisoners. (The artillery had its sound-ranging specialists and observers moving forwards on the 9th to support the fight when the RAF could not.) But on the 10th Rawlinson was already suggesting to Haig that his attack be wound down and the attack widened to the flanks, with the British Third Army on the left and the French First Army on the right.
Rawlinson’s attack was continued for a few more days, partly to wring maximum effect from the initial success but partly to cover the preParations of Sir Julian Byng’s Third Army. (The attack would ultimately take over 25, 000 prisoners, inflict roughly 50, 000 more casualties on the Germans, take 600 guns, and take up to 19 kilometres of ground.) Intangibly, Amiens was proof of the momentum swing on the Western Front and in the whole war. The Germans were not to attack again, but the Allies would continue to rain down blows throughout the ‘Hundred Days’ until the German army was defeated. When Ludendorff referred to 8 August as a ‘Black Day’ for the Germans, he was not just referring to the casualties (although they were heavy, and the large number of prisoners showed how fragile German morale was); he had been shown that he was no longer calling the shots. The Allies were, and Foch’s plans for successive attacks would continue along the front.
Byng had been making phony attack preParations as part of the deception plan, and now the BEF could make Third Army’s phony preParations real while the Fourth Army’s real attack (which was becoming more cautious) became the deception. Byng had few tanks, which made protecting them from German artillery even more important, so the Third Army stressed its counter-battery efforts. This included integrating the RAF and artillery; aircraft were earmarked to hunt for German anti-tank guns and not only strafe them but call in the 60-pounder batteries which were on call – their sole mission was to hit anti-tank guns. At first Byng just cleared the German outpost zone, although in places the German artillery resisted stoutly and delayed the British; he then paused and brought up his guns and prepared (while beating off German counter-attacks) for the main attack. After only a day of preParations (and the Germans weakening them-selves through ineffective counter-attacks), he was ready and cracked the German main line with only modest resources. Even in the V Corps’ sector, where the German artillery had caused problems on 21 August, the attack on 23 August gained ground readily. The results (up to 4 miles of ground, over 10, 000 prisoners plus other casualties) were far less dramatic than in the surprise attack at Amiens, but they showed both sides that the Germans could be driven out of their positions by an ‘ordinary’ attack – tanks were not necessarily the key to an attack, nor were Dominion ‘shock troops’.
German reserve divisions were getting little time to rest before having to go back into battle. On the 23rd Herbert Sulzbach heard the battle over the horizon. His reserve positions in a pretty village were ‘invaded by the rumbling and rolling of the new large-scale battle between Noyon and Laon; it seems to be the fourth phase of the Foch offensive, directed against our Fourth Army’.40
Ludendorff recognised the weakness of his position and tried to pull back to a ‘winter’ line in front of the Hindenburg Line where he could reorganise. But he needed time to actually build the winter line and time to rest his troops and restore morale. He tried to buy time with rearguards, and both sides had to coordinate semi-mobile operations. There were few or no trenches, so it was not trench warfare, but there was a reasonable front line so it was not quite the mobile operations that had been taught before the war. The Germans found it hard to meld artillery into the rearguards for a range of reasons, not least because the gunners might simply not know where the infantry was, as units were moving every day or two. Meanwhile, the Germans had identified tanks as a key Allied weapon, not just for their firepower and mobility but because they demoralised the German infantry. New guidelines and methods came down from on high that looked rather like the British methods of the late spring of 1918: guns to be mobile, distributed in depth, withdraw if losing the battle, counter-battery fire in lulls in the overall fighting, counter-preParation fire instead of defensive barrages, and the like.41 The anti-tank instructions were unrealistic: it was all very well to say that the field artillery needed to be even more mobile and move to support the infantry, but where were the horses to come from?42 supplies might be a problem too, as resupply points changed erratically. But morale seems to have been the worst issue, as infantry frequently simply pulled back past batteries, or batteries fired off the ammunition they had (at whatever target they selected) and then pulled back themselves. At the end of August the British broke Ludendorff’s planned winter line in two places, the Canadians breaking through the heavy fortifications of the Drocourt–Queant Line (an extension of the Hindenburg Line) while the Australians fought their way up Mont St Quentin.
These actions set off another round of German withdrawals, this time back to the Hindenburg Line itself. For the BEF much of September would consist of day after day of semi-mobile operations, and they became fairly adept at it. Infantry divisions were typically rotated every two or three days to rest and reorganise, but because the artillery did not rotate as frequently, so there were 4 to 6 brigades of field artillery available to support each infantry division, and also a brigade of heavy artillery.43 No more than one-third would be moving at once, so plenty of fire support was available. These brigades may have been available, but they were not necessarily engaged; ammunition supplies were problematic, and if the Germans were not entrenched there was no need to fire much of a barrage. There was seldom a bombardment, just a creeping barrage in the morning to help the infantry get started and suppress whatever defences the Germans had organised overnight. When ammunition was short, the creeping barrage might do little more than guide the infantry forwards, and the few smoke (and incendiary) shells mixed in were very useful because they stood out. Some days the infantry did not even know where they were, and the barrage would start firing along a terrain feature (such as a road) and the infantry would advance to the start line before the barrage even began moving. German strong-points would receive concentrated fire, and there might also be enough of a bombardment plan to shell suspected strong-points ahead of the barrage. Much fire was observed rather than planned from the map; that this was a novelty shows how much things had changed since 1914. Against lighter defences the concentrated heavy artillery was enormously potent: ‘Sections of trench were flattened out in minutes, dugouts blown in, walls collapsed, machine guns and mortars were covered by earth and rubbish, and ammunition dumps exploded.’44 Artillery support did not stop with the barrage; typically a field battery moved forwards with each infantry brigade headquarters to provide on-call indirect fire, while the leading battalions could have a section (two guns) pushed forwards to blast machine-gun nests. A number of 6-inch mortars were also pushed forwards on wagons, with improvised mounts on the wagon bed so the mortar did not have to be dismounted.
This sort of fighting led to average gains of 1–3 miles per day, not much more than the range of a creeping barrage (showing how much the infantry wanted and needed support – which was understandable, with them fighting every day). Often the attack would peter out by midday, with the afternoon spent organising for the next day’s attack, scouting, laying telephone cable and bringing up supplies. With few tanks available and the evervariable weather limiting the RAF’s activities, artillery was the most reliable support for the infantry and the BEF recognised that artillery support reduced casualties. The senior gunner at GHQ reported: ‘All army commanders are at me not to reduce the artillery and say with the present state of the infantry they cannot do with a gun less …’.45
Where the German defences were thicker, as at the Hindenburg Line, the BEF slowed down and brought up the panoply of trench-warfare tools. More guns and ammunition were sent forwards, and for the artillery the sound-ranging and flash-spotting sections also moved up. Survey units needed only 40 hours to identify friendly positions to a reasonable standard, allowing good (if not perfect) shooting in only two days. It helped that the BEF did not face the same challenges as in previous years: German morale was so much lower that there was no need to demolish the defences, only to crack a way in and let the infantry fight their way through, albeit with artillery support.
Rawlinson’s Fourth Army would be tackling the Hindenburg Line, and as his men reached it and drove in the outposts, they had an inestimable advantage: on 8 August they had captured a detailed map of the sector they would be attacking. It marked the locations of all the key German installations, so the British could select exactly what they needed to hit and not waste shells from their 1, 600 guns. The pinpoint targeting probably increased the demoralising effect on the German troops, who would have seen an uncannily accurate bombardment. As the British noted at the time, ‘Prisoners state that our 48 hours’ bombardment prior to the attack was extremely effective, and that it was owing to this that the pioneers of the 2nd Division were unable to blow up the bridges over the canal at Bellenglise, as they did not receive food for two days and dared not leave their dugouts owing to the artillery fire.’ Increasing the demoralisation was the first British use of mustard gas. The Germans had used it during the Third Battle of Ypres, but it took British industry roughly a year to identify the chemical, develop a production process and make a substantial quantity of shells. Now they were being used and were fired early so that the Germans would experience the worst effects. In theory (after the fourday bombardment), there would be little residual effect for the attacking British troops.
Mustard gas certainly helped the bombardment, but there were plenty of problems to counter-balance. The weather was mediocre, with rain and cloud producing a murk that seriously reduced observation (from air or ground) and affected the counter-battery effort. The Germans also moved guns around as best they could, given their serious shortage of horses, and the net result was that the German artillery was not particularly well suppressed. Ammunition supply was also a weak point for Rawlinson. Only one main railway line was in working order and the roads were clogged with units and vehicles. Shells could be brought up, but often arrived late. With other operations under way or in preParation (Foch had Belgian, British, French and American attacks scheduled for the end of September), there were simply not enough guns to go around, and the Fourth Army had fewer than they would have liked. But the bombardment still involved 750, 000 shells (including 32, 000 mustard gas ones) over four days, starting on 26 September. Counter-battery fire was vigorous, while wire-cutting was absolutely vital since there were few tanks available and much wire; harassing and interdicting fire were also important across the depth of the German position, and there were of course the machine-gun nests and dugouts to hammer.
Rawlinson was attacking the toughest point of the Hindenburg Line in two sectors: the Americans and Australians faced the Bellicourt Tunnel, which was heavily fortified because the tunnel was an obvious weak point, while IX Corps was planning to cross the St Quentin Canal, a formidable feature in itself (35 feet across and up to 50 feet below the natural level of the ground) with lots of barbed wire and a good number of machine-gun positions. However, IX Corps was attacking the Bellenglise Salient, and it was simple to concentrate shells on the salient and get advantageous angles of fire.
On the 29th there was fog (in addition to the smoke shells mixed into the barrage) which substantially bothered the Germans. Their artillery observers could not see what was happening and infantry could not see very far. This fog was bad for the RAF but it was still a net benefit for the attacking Allies because air support was only a modest benefit. One brigade commander wrote later: ‘The night 28/29 had been an unpleasant one – very dark and wet, everyone on the move, heavy shelling and much gas, and quite a number of casualties; but the morning found us merrily firing our barrage, which lasted several hours (3, 000 rounds by Zero + 512) and left us in the immediate necessity to bring up more ammunition.’46
The attacks developed very differently. The US 27th Division attacked without a creeping barrage (because of fears that isolated groups of friendly infantry were still scattered around after a preliminary attack) and barely captured the German forward line. The 30th American Division did better, largely due to having better observation posts (so their preliminary bombardment had been more effective) and a creeping barrage for cover, but they pushed too hard without mopping up effectively. They punched through the German line but reserve troops got held up by the by-passed Germans and spent their strength mopping up and fending off the inevitable German counter-attacks. The American/Australian attack had pushed the Germans back and partly broken through, but hitting the Germans in this heavily fortified and defended sector had proven problematic.
Meanwhile IX Corps led its attack with the 46th (North Midland) Division crossing the canal. The terrain problems were formidable, but the bombardment had done much to help, for instance by knocking down sections of the brick walls to make it easier to get out of the canal. But the unit needed ladders to get down to the canal, boats to get across and ladders to climb back out. Good staff work provided all that and more, including life-jackets from cross-Channel steamers, to get more men across the canal quickly. On the 29th their careful planning was only part of the explanation for their success. The Germans were extremely surprised by the bold frontal attack, and that threw them off guard. Their reactions were also slowed by the fog. A heavy bombardment had played a role, with the men of the 46th covered by an exceptionally dense creeping barrage: no fewer than 54 18-pounders each firing two rounds per minute and 18 4. 5-inch howitzers each firing one round per minute covered the 500-yard bridge-head. The leading brigade cleared the German advanced positions, crossed the canal and reorganised; reserves crossed under a protective barrage and then turned laterally to widen the hole. When the fog cleared, the Germans were able to direct in more artillery fire, but thanks to the preliminary bombardment their communications were patchy and they simply lacked the strength. The corps’ reserve division moved through the hole and by the end of the day the corps had taken 5, 100 prisoners and 90 guns. It was a stunning accomplishment: the Main Hindenburg System and the Hindenburg support System had both been broken on a corps’ width, and the BEF had overwhelmed the Germans’ last prepared defensive line. It was made possible by a combination of good infantry training, effective artillery preParation and support, and intelligent leadership. As Charles Budworth, the Fourth Army’s senior artilleryman, boasted in his report, ‘The results of our artillery fire as a death-dealer and as a life-saver are written on the ground and in the trenches by German corpses and unused rifles and ammunition.’47
Bad news was pouring in to Hindenburg, Ludendorff and the Kaiser. The Turks had been routed in Palestine; the Bulgarians were breaking under attack around Salonika; repeated Allied attacks were hitting the Western Front. Their last hope had been to fall back to a winter defensive line and hope the Allies exhausted themselves in fruitless attacks, but now all they could do was retreat as slowly as possible and hope to gain better terms as armistice negotiations began. Ludendorff continued trying to tweak tactics to win, possibly trying to avoid thinking about the collapsing strategic position. The same day that Rawlinson broke through the Hindenburg Line and the Salonika front collapsed, Ludendorff was sending out new instructions: ‘The selection of a position is dependent on artillery observation, by which the ground in front of and behind the main line of resistance must be watched. This generally entails a reverse slope position. The possession of high ground is not of so great importance for the infantry defence.’ Moreover, artillery machine guns were to be integrated into the overall checkerboard of resistance and ‘the guns will fight to the last … their fire over open sights forms an extraordinarily effective support to the infantry’.48 He was at least realising that Allied attacks were routinely reaching the artillery positions, but paper instructions from the First Quartermaster General could not make the German artillerymen fight to the last: Germany was defeated.
From mid-September the US was an important factor on the Western Front. The American Expeditionary Forces (AEF) were over a million strong and growing as fast as the transports could bring over more men.49 Artillery, however, was not one of the AEF’s great strengths. The AEF was organised in enormous divisions of around 28, 000 men – about the strength of a European corps. (This was largely due to the Americans’ lack of trained staff officers: the few available could make more impact with fewer divisions. The huge divisions may have been one reason why the American creeping barrages were planned by divisions rather than by corps.) With four infantry regiments of around 16, 000 men, each division also had four artillery regiments, three light and one heavy. But the US lacked enough artillery of its own (at home, units trained on log cannons), and production was hampered by a series of bad decisions about what to produce – the US pre-war 3-inch gun, a 1916 variant, a blend of the 1916 gun and a French 75, or a blend of the British 18-pounder and the 1916 gun.50 It seems the desire to have ‘Made in USA’ stamped on the guns drove the decision for domestic manufacture, but in nineteen months of war the US built only 109 guns suitable for the front. Instead, the French offered them guns, the 75 for the light regiments and 155mm howitzers for the heavy regiments. (This made the Americans the only force to have 155mm weapons at divisional level.) But most American units only received their equipment in France, so training was greatly curtailed; trench-mortar units were even worse off, because there was not even obsolete or dummy equipment in the US for minimal training.51
The Americans had similar problems with doctrine. General John Pershing, their commander, was obsessed with mobile warfare and avoiding the ills of trench warfare; he held that after so many years in the trenches the Europeans lacked aggressiveness and dynamism and insisted that Americans should not learn too much about trench warfare or their native superiority would be diluted. American Field Service Regulations even held that ‘pre-concerted plans’ were ‘objectionable’, looking back to a simpler day when a general on horseback could run his battle.52 For artillery, this approach even included disdaining indirect fire as it was ‘tainted’ with the trenches and lacked the aggressiveness of moving forwards to conduct direct fire. However, given the shortage of horses in France, there was not much opportunity for mobile warfare training and so trench warfare training was the default, largely because the US lacked experienced officers (there were only 275 artillery officers with more than one year of experience) and European officers were the main instructors. There was also much publishing of European experience and methods, for example a full reprint of the British IX Corps’ artillery plans for attacking Messines Ridge, which was essentially a siege operation – about as far from mobile warfare as possible.53 Thus, whatever Pershing might think, his artillery crews had to learn about modern methods. Implementing them would be more difficult, as the Americans would have as much problem moving on from pre-war methods as the Europeans had done, and they were trying to do it faster.
The first major American battle was a mid-September attack on the St Mihiel Salient, as part of Foch’s series of limited attacks. For the American soldiers, it was their first battle as an army; always before they had been scattered in seParate divisions. (While a number of divisions had seen action, their infantry had suffered heavily, and even Pershing was now seeing the merits of artillery preParation; he had issued new instructions encouraging more fire support, albeit still emphasising deep penetrations.) It was also the Americans’ first battle with their own corps, and they had to borrow much of their non-divisional artillery from the French, including artillery-support units such as balloons and aircraft, and even borrow some corps–artillery commanders. Probably as a result, the Americans followed French doctrine in a number of ways, including concentrating counter-battery work at corps level. Pershing was determined that the US First Army would not falter in its first battle (and under his personal command), and he assembled overwhelming force and his most experienced units. After Verdun, the Germans had treated the St Mihiel sector as a quiet area, and rested divisions there. It was undermanned if well fortified, but units there had very limited transportation service to move supplies and equipment; they were even starting a withdrawal from the salient as the American preParations were finalised.) One change for the Americans was a strong emphasis on forward guns, a battalion of 75s being attached to each infantry brigade; one battery was to stay around the headquarters of each infantry regiment for quick-response fire while the third battery was broken up into single guns (dubbed ‘Pirate Guns’) that were to provide direct fire support to the infantry.54 These were a decent compromise: Pershing could be pleased with the aggressiveness of these guns, while the French advisers might not mind losing a small part of the field artillery.
There is little to say about the artillery at St Mihiel except to say it worked. Massive artillery support for a massive attack against defenders who were weak in numbers (and in spirit) was ample. An American infantryman remembered his battle: ‘Shrapnel bursting and sending down its deadly iron. High explosives bursting on the ground and sending bricks, mud and iron … A mad dash of 50 feet, then look for cover. A stop for a minute … then another mad dash!’55 The Germans were broken in two days (12–13 September) and the Americans took more prisoners than they suffered total casualties. Looking back, communications were identified as a problem (as they always were in the Great War), especially where the infantry advanced quickly and outran the telephone cables.
However, the next attack (actually a sustained offensive) was not so easy. Pershing had agreed to the St Mihiel attack but he had also promised to take the Meuse–Argonne sector for the Americans, and he was keen to launch an attack there as soon as possible. Unfortunately, he rushed the offensive. Road links were poor, delaying troops, guns and ammunition. With the experienced divisions busy disentangling themselves from St Mihiel, the Meuse–Argonne attack would open with inexperienced divisions to meet the timetable. And nobody had asked for gas shells to neutralise the German guns. Trying to advance 16 kilometres in two days with all these problems – plus no artillery support after the initial creeping barrage because the infantry would be too far ahead – was asking for trouble. Unsurprisingly, the opening attack on 26 September had mixed results, rather like an attack in 1916, and in subsequent days the Americans laboured mightily to even out the line. The Germans held on to key observation positions in the centre and on both flanks; continued battering would eventually take Montfaucon in the centre, but the AEF only gradually realised that a new phase was needed to clear the flanks. Each day casualties rose, cohesion fell and there was less and less progress despite Pershing’s hectoring: ‘we must push on, night and day, regardless of men or guns’ and his firing of commanders who were insufficiently aggressive.56 (In contrast, one of his corps commanders commented: ‘If we are to be economical with our men, we must be prodigal with guns and ammunition.’57) After a week they had indeed taken the German first line, at least in the centre, but there were a number of German lines – this area protected key rail junctions in Germany and was heavily fortified, even if there were too few troops to hold all the lines. (German troops fought almost fatalistically, knowing they were protecting the supply lines to their comrades further north. Machine-gun teams might fight to the death, but there were relatively few counter-attacks and artillery fire was often on a zone and not adjusted to where the American troops actually were.)
The Americans paused, reorganised and brought up fresh troops, and finally succeeded in clearing their flanks. By 14 October they had taken all the objectives set for 27 September. As elsewhere on the Western Front, the Germans were exhausted and the Allies gained ground in spurts: the Germans could easily be knocked back when an attack was well prepared and well supported, but it took time to organise such an attack and the Germans could resist poorly supported attacks. But Pershing’s insistence on always pressing home the attack, rather than accepting that attacks inevitably ran out of steam, ran up casualties and only burned divisions out sooner. Repeated American attacks in early October and then on 1 November (with ‘cannon enough to conquer Hell’58) each drove the Germans back but lost momentum. The attack on 1 November was especially well planned, with gas being used, German battery and reserve positions identified, and interdiction fire to isolate hills while trained infantry teams wormed through gaps. That attack kept its momentum as the Germans finally lost heart with the war, and the leading American troops reached the German border by the Armistice. An infantry regimental commander gushed:
The initial attack was marked by a brilliant coordination of arms, the artillery laying down an absolutely smothering barrage which the infantry followed closely … Attention must again be called to the wonderful work of our artillery in the first fighting. Our troops have never advanced behind a more perfect barrage.59
The Americans had plenty of guns for dense creeping barrages, and also experimented with deep ones, using 6-inch and even 8-inch guns firing ahead of the infantry. Counter-battery fire also improved, and at times the Americans had enough shells to hit areas where German batteries were suspected but not proven.60 (Ammunition supply was more a question of the roads and light railways; sometimes guns were held to the rear because supply links were too poor, and sometimes there were enough shells to fire on possible but unproven positions.)
The Americans did not live up to their own expectations, but they learned fast and they were certainly able to bash the Germans. However, their tactical clumsiness cost them more casualties than experienced troops would have suffered.
After 1915, when the Franco-British landing force at Salonika had failed to link up with the retreating Serbs, the Salonika front remained a backwater for years; the Germans sarcastically referred to it as an internment camp since the Allied troops were hardly fighting.61 It caused a certain amount of inter-Allied diplomatic and military friction: the French generally wanted to sustain the troops there (partly as a counter-balance to the British forces in Palestine that stood to expand British influence in the eastern Mediterranean after the war), while the British were less enthusiastic. The Greeks sat on the fence, with factions arguing the pros and cons of entering the war on either side, and the question marks about that also kept the Allies on tenterhooks. There was some fighting in 1916 (the Allies attacked just ahead of Rumania’s declaration of war, trying to draw Central Power forces away from Rumania) and 1917 (tied to Nivelle’s attack), but the battles gained little ground and none of what they gained was strategic. That is not to say units got off scot-free: a battalion or brigade could be savaged by a day’s fighting.
The Allies under-resourced the forces they had there, and their ideas on artillery employment were also outdated. Salonika was, unsurprisingly, a low priority for equipment – the first battery of 8-inch howitzers arrived in January 1918, and was actually discussed at Cabinet level.62 (Imagine if Western Front operations had been that tightly controlled!) Until the late summer of 1916 the British had no heavy artillery, and only gained ten batteries then – again very low quantities for multiple divisions. Trench mortars only began arriving in December 1916. There was also not much specialist equipment: the British army had only three batteries of mountain guns, and while they were sent to Salonika, they were, after all, only three batteries. (The Indian army had more mountain guns, but Indian troops were not sent to Salonika.) The British had problems with both high explosive and gas shells. The high explosive ‘sweated’ (or chemically seParated) in the summer heat and had a high dud rate, while of the few gas shells that were shipped, some must have leaked during shipment because they had no effect either. Then the British (at least) showed their low regard for Salonika by not shipping doctrinal manuals: for the April–May 1917 attacks they were using pre-Somme handbooks. The 1916 and 1917 attacks often had problems with counter-battery fire (Bulgarian guns were hard to locate and often well protected against the sparse Allied fire, and the first sound-ranging equipment only arrived in January 1918), barrages were hard to coordinate without much experience and in rough ground, and lengthy preParatory bombardments gave the Bulgarians warning of where to move their operational reserves.
The front was a backwater for the Central Powers as well. The Austrians never had enough troops, while the Germans sent only enough to stiffen to the Bulgarians. Instead, they built up the Bulgarian forces.63 The Bulgarians had decent artillery in the Balkan Wars, and had captured a fair number of modern guns from the Turks in the First Balkan War. Then they had captured a good number from the Serbs in 1915, and although many of those were obsolete fortress guns, they were useful on a semi-static front like Salonika. With all these guns, they not only increased the number of guns, but divisions were upgraded with quick-firing batteries. Trench mortars were also delivered, while the Germans supplied gas shells and sound-ranging equipment. The Bulgarians organised new artillery formations, including a seParate artillery brigade for counter-battery fire, drawing in the sound-rangers but also observers from the dominating heights such as the Grand Couronne – called the ‘Devil’s Eye’ by the British who could see the observation post on top – that looked out over Allied positions on the Salonika plains. Captain A. D. Thorburn, who commanded a British howitzer battery, commented, ‘Accurate counter-battery work they understood, and executed with unpleasant results to us on more than one occasion. They had most efficiently handled 8-inch and 5. 9-inch and 4. 2-inch howitzer batteries whose shooting was at least as good as that of German gunners we came across …’.64 It was lucky for the Allies that Bulgarian ammunition was often limited, and so was not wasted on counter-battery or harassing fire but saved for when there was an attack. Bulgarian troops were fierce in defence, and counter-attacked as much as the Germans did, and those counter-attacks presented some of the best artillery targets: 28th Division laconically recorded one action: ‘07. 50 hours. A counter-attack by the Bulgars was launched on Yenikoi from Topalova. 3rd Brigade RFA engaged and stopped it.’ A second counter-attack in the same spot at 08. 20 hrs ‘was pushed forwards with very heavy losses until about 09. 40 when it dwindled away and was crushed’.
Yet seizing a few farms or villages was not the result the Allies sought: nevertheless, there was no easy end-run around the Western Front. It was not somehow easier to coordinate infantry and artillery just because they were fighting the Bulgarians. The Salonika Front saw its own halting advances in tactical ideas, and steady progress in thickening defences and the build-up of troops.65
By 1918 the melange of Allied troops (not just French and British, but Serb, Greek and Italians, with some Russian troops disarmed after the Bolshevik revolution and used for labour) had an increasing advantage. The Germans had never put many troops on the Bulgarian front – at most about three divisions and 72 batteries – but to maximise their forces for the Kaiserschlacht they had withdrawn almost all the infantry and much of the artillery. The Bulgarians could not effectively use the guns that were left behind; they were already over-mobilised, with more men in uniform than the country could support – about one-third of the agricultural land was not under cultivation in 1918 because of a lack of farmers and draught animals. Bulgarian morale was dropping because they knew the conditions on their home front, they knew the Germans were expropriating some of what food was being produced, they knew their uniforms and boots were wearing out, and by September they knew the Germans were losing battles on the Western Front.66 As a result, while the Germans left some of their guns with the Bulgarians, it hardly strengthened the front. The lack of reserves (and the lack of draught animals to move guns and supplies) proved more disastrous as Foch spread his range of limited attacks beyond the Western Front. He had been urging the Italians to attack, but they were resisting. Since the troops at Salonika were under Anglo-French control, they could more easily be sent forwards, and an attack was approved for mid-September.
The tactical details are almost beside the point. Allied operational planning had misled the German commander (the Germans retained command, despite having only three infantry battalions in the theatre) and the reserves were in the wrong place.67 (This was helped by British preliminary attacks that took heavy casualties for limited gains, but pinned down more Bulgarians.) When the bombardment started on 14 September (in what a German officer called an ‘iron storm’ in ‘hurricane force’),68 it caught the Bulgarians off balance. The front-line troops resisted as best they could and counter-attacked as much as they could, but the Allies seized the front line at Dobropolje, which gave them observation over the second and third lines and meant that their interdiction fire was even more effective against the limited Bulgarian reserves. Their morale was already brittle, and once the front was broken the Bulgarians cracked and streamed for home. The Allied advance northwards was limited by their inability to repair the railway, but by November the Allies were well into Serbia and threatening central Austria-Hungary.
The Germans transferred their artillery and divisions to the Western Front in late 1917 and the Austro-Hungarian army was left to deal with the Allied armies on the Piave. General Armando Diaz set about rebuilding the Italian army and organising it for the battles of 1918. The artillery was a particular priority and Diaz was far keener to learn from his allies than Cadorna had been. The Austrians benefited from the transfer of troops from the Eastern Front but the Empire was already on the verge of collapse. Boroevic´ was ordered to plan a new summer offensive but received little of the vital equipment or reinforcements that might ensure success.69 Of particular concern was the absence of substantial numbers of heavy batteries on the Trentino and the empire’s critical weakness in the air (there were just 53 aircraft to take on the Allies’ 486). The Austrians would have to attack with limited ammunition and without detailed intelligence on the dispositions of Allied units, while blissfully unaware that ‘the most minute details of the coming bombardment and the assault were in the hand of the Italian command previous to the attack’.70 Operation Lawine (‘avalanche’) went in on 15 June. The bombardment phase, broadly based upon Bruchmüller’s approach to fire-planning,71 used large quantities of gas but the majority of Italian batteries remained unsilenced and their counter-barrage hit the Austrian assault divisions as they struggled across the river Piave and attempted to break through near the Asigo Plateau. The Austrian Official History bitterly noted that ‘our artillery fire had – as prisoners related – shot badly and inflicted hardly any loss on the enemy’.72 The Austrians made surprising progress in some sectors but overall the offensive faltered against Badoglio’s more ‘elastic’ defensive system. The Battle of the Solstice gave the Italians new hope and, although there were Allied units in their order of battle, they savoured the feeling that they had finally beaten their old enemy without being the handmaiden of more powerful Allied forces.
As Germany’s offensives also ground to a halt and Foch’s counter-offensives began, Diaz came under pressure from both his government and the Allies to go on the offensive. The reluctant Italian commander-in-chief decided on a breakthrough in the central sector, between Vittorio Veneto and Sacile, as his main objective and began the agonising process of plan-ning what he privately suspected would be another disastrous offensive. The Fourth, Eighth, Tenth and Twelfth Armies were to initiate the offen-sive, with an attack by the Fourth Army on Mount Grappa unhinging the entire Austrian position. Some 400 additional guns were transferred in to support the attack and Diaz sensibly assigned key objectives to some of his most experienced Allied units. The Fourth Army ran into trouble at Mount Grappa on 23 October, attacking clumsily after days of shelling, but the Tenth Army had more success, benefiting from the presence of two notably pugnacious British divisions under Lord Cavan’s command and a bombardment that one British veteran described as ‘thousands of gun flashes coalescing to form a continuous blaze of light along the bank’.73 The Piave line shattered after six days of fighting. The assaulting units quickly left their artillery support behind and only the headlong retreat of the demoralised Austrians prevented the Allies from suffering heavy casualties as the offensive gathered momentum in what Diaz gleefully described as ‘Caporetto in reverse’.
On 1 November 1918 the Canadian Corps would take Valenciennes. The small city was only 30 kilometres from Le Cateau but the artillery tactics and techniques were four years apart, and it made a world of difference.74
In late October Haig reckoned the Germans were on their last legs, with Turkey and Bulgaria knocked out of the war and Italy preparing to attack the tottering Austrians. With the Americans and French attacking, it was time for the BEF to launch one final blow. To get the British First Army in position for the anticipated Battle of the Scheldt, they first needed to take Valenciennes, which lay east of the Scheldt Canal. Because of the rain and German-controlled flooding, the low ground west of the canal was flooded for a distance of perhaps a thousand yards; in addition, there was barbed wire on the eastern bank and the German troops (and machine guns) were safely positioned in houses. A frontal assault across the canal was out of the question. However, the canal swung round the city and to the south XXII Corps had got across. If the Germans could be thrown off Mt Houy (which was only 150ft high, but about 50ft higher than the surrounding country-side, and blocking observation of German artillery to the east), they could be levered out of Valenciennes.
However, the Germans recognised the key ground and they had plenty of guns; in addition, troop morale was reasonably firm. From 24 to 28 October several British attacks were made, all rushed and poorly supported, more in hopes that the Germans were weak than in confidence that the attacks would succeed. But the British troops were at the limit of their supply lines (railheads were 30 miles back, and lorries were in short supply), casualties had thinned the ranks and everyone was tired. The Scots of the 51st (Highland) Division pushed up Mt Houy, but their last attack on 28 October was driven back from the crest by a German counter-attack, despite support from nine brigades of field artillery and fourteen batteries of heavies.
The Canadian Corps was now moved in to make the attack. The Canadians had been facing the canal, but since the main thrust could not be made there, they were available. The 4th Canadian Division relieved the 51st Highlanders, and moved up guns and shells; they took several days to plan their attack. Few infantry and plenty of support was a key element of their plan: ‘to pay the price of victory, so far as possible, in shells and not in the lives of men’.75 The delay also allowed time to coordinate infantry, machine guns and artillery. The Canadians knew there had been several failed efforts to take Mt Houy, and steadily increasing German artillery fire showed the enemy’s determination to hold the position; however, the Canadian gunners were just as determined to crush German resistance by weight of shell.
The attack would be some 2, 500 yards wide (about 1½ miles). One Canadian infantry brigade would attack (by this stage of the war, that meant about 1, 200 men). Generally speaking, about 10 per cent of any unit was left out of battle in case there were heavy casualties. For that one infantry brigade, there were eight brigades of field artillery and six of heavy artillery. The first objective was basically Mt Houy, and the second was 2, 000 yards beyond it, clearing a few villages and the suburbs of Valenciennes.
There was no preliminary bombardment, but most of the heavy artillery fired well ahead of the infantry, hitting the German defence in depth and the reserves. No fewer than 39 6-inch howitzers were assigned to fire one round per minute over the front of the attack, a ratio that equated to 1. 6 shells per 100 yards and the bursting radius was over 500 yards. McNaughton was putting ‘a practically continuous rain of chunks of steel across the whole front of the attack’. That was the first phase; when the Germans were pushed off Mt Houy and lost their observation posts there, more Canadian guns could fire, and the second phase of the attack narrowed to 1, 000 yards. Some 55 howitzers would fire 2 rounds every 3 minutes, so it became 3. 6 rounds per minute per 100 yards.
In all, 144 18-pounders and 48 4. 5-inch howitzers would fire a creeping barrage (effectively 7 tons of shells per minute), deliberately moving at only 100 yards in four minutes (later slowing to five minutes) so that the infantry would have no problem keeping up. The fieldfrom the foremost howitzers would fire some smoke shells but would also hit selected strong-points ahead of the 18-pounders. The infantry, in turn, pulled back from the foremost positions on the lower slopes of Mt Houy so the artillery would have a straight (and convenient) line for its starting barrage. Machine guns fired both forward and flanking barrages, taking advantage of the topography: Mt Houy was an exposed salient. The infantry would be attacking from the southwest with machine guns firing from the south and heavy artillery firing from the north. Additional machine-gun and heavy artillery barrages were planned for the right flank of the attack, covering the ground with fire instead of sending more infantry into battle. Planning also took into account where German reserves were likely to be and thus where counter-attacks were likely to start. Since the towns and villages were full of refugees, the French had forbidden unnecessary shelling. (The Germans were contin-uing to use gas shells, and the Canadian troops were upset about its use around unprotected civilians; they were prone to confiscate gas-masks from German prisoners and give them to civilians. They were also taking relatively few prisoners at this stage in the war.) The Canadians decided only to hit counter-attacks on the edge of towns; this meant that the Germans had a good night’s sleep in a building but they were easier to kill in the open. The half-circle of British positions allowed enfilade fire not only on the front line but on roads (for harassing fire) and on reserves. Counter-battery work was not neglected, with 49 guns assigned to oblit-erate the 26 known German battery positions. The gunners slept by their guns in case the Germans got wind of the attack.
One battery was assigned a particularly devious mission. It was deliberately sited where it could fire into the rear of the German positions, and shortened its range as the attack progressed. Not only did this prevent it from hitting the Canadian infantry, but the Germans would think their own artillery was shelling them and their morale would suffer accordingly.76
At dawn, 05. 15 hrs, on All Saints’ Day the bombardment crashed out and the infantry moved forwards. German artillery fired promptly and accurately but mainly at the British artillery, with little or no effect. (Gibbs called it a ‘fierce line of fire’ but noted that it quickly ended as counter-battery fire took effect.) The hapless German infantry soldiers, meanwhile, were deluged with shell-fire. Gibbs wrote, ‘our barrage rolled like a tide wiping them off the map of France’, and the New York Times headlined the story ‘British Gunfire Paralyses Foe’. Prisoners, ‘stupefied and demoralised’, surrendered freely, including a complete company that was trapped in the fog and smoke; perhaps the first thing they saw of the Canadians was their bayonet points. With these advantages, the first objective was reached on time. A few machine-gun nests and a single field gun held out during the advance to the second objective, inflicting casualties before being overrun by the experienced infantry. The heavy artillery fire stayed ahead of the barrage and deliberately smashed some rows of houses where the Germans were known to have positions (any refugees killed here were regarded as collateral damage). Once the objectives were secured, it was time to see what the Germans would do. Each of the infantry battalions moved a 6-inch trench mortar forwards, and three brigades of field artillery moved on to the slopes of Mt Houy. Their observers moved to the top, so they could quickly engage any target they saw. Shortly after noon German infantry was seen forming up and the planned protective barrage was employed: 11 batteries of 6-inch howitzers rolled a barrage over the Germans. The survivors lost all interest in attacking. Between 15. 00 and 16. 00 hrs more movement was seen on the right flank, and on-the-fly plans were made to hit the Germans once they had fully formed up. At 16. 35 hrs the situation was judged ripe, and 9 batteries of 6-inch howitzers obliterated another counter-attack.
The results were gratifying. Mt Houy was taken and the Germans were levered out of Valenciennes. (Another Canadian brigade had squelched forwards to the canal to test the German positions, and found almost no resistance. By mid-morning two Canadian battalions were solidly across. The German infantry had withdrawn very quickly, probably realising from the noise of the bombardment on their left rear flank that their comrades could not hold under such a maelstrom.) Over 800 dead Germans were found around Mt Houy alone, and 1, 800 prisoners taken. The 2, 149 tons of shells had done their work.77 But the Canadians also suffered 501 casualties,78 out of the 1, 200 infantry in the attack. Massive (and well handled) firepower could reduce casualties – not least by allowing fewer infantry to attack – but there was no avoiding a substantial percentage of casualties. The three British divisions attacking further to the south took over 1, 600 prisoners and counted 300 dead; their casualties were higher than the Canadians’, but by this stage of the war a well supported Allied attack could easily break any German line. The Canadians had used every trick in the Allied arsenal and noted a number of ideas for the future but their brutally effective use of artillery had not solved all the problems of the Great War.