The demands and preparations of the Military Authorities with regard to Mechanical Warfare for the fighting season of 1918 are entirely inadequate, and the changes which you have made in this Department at this critical time (and which involve reconsideration of design and consequent loss of production) will most seriously affect even the efficiency of this programme for next season’s fighting. This is also the considered opinion of my technical and commercial advisers.2
Stern got to work at once, contacting Maj James A. Draine and Maj Herbert W. Alden of the American Army. They were in London, detailed by the Chief of Ordnance in Washington to report on tank developments in Britain and France. There was no question for Bertie Stern of ‘eventual’ production. He rapidly secured agreement for an Allied programme of tank design and assembly. D’Eyncourt joined them in a follow-up meeting in Paris to confer first with M. Munich, a spokesman for M. Loucheur, the French Minister of Munitions, before reporting to Gen John J. Pershing, commander of the American Expeditionary Force in Europe. M. Munich emphasized that France had no interest in heavy tanks and could give no assistance to a joint programme, though there was no objection to an assembly plant in France. The party then proposed to Gen Pershing that the USA and Great Britain should urgently co-design and produce 1,500 heavy ‘Liberty’ tanks. Of the main components, Britain would contribute the guns, ammunition, armour and track assemblies; America the power train, track links and chains. Pershing gave his approval on 15 November and appointed Draine American Commissioner. Draine had been a general in the National Guard at home, and as fellow businessmen he and Stern would work well together.
D’Eyncourt went one further, calling for a ‘United Allied Board of Mechanical Warfare’ to unify control of design, production, strategy, tactics and development, with Stern as chairman and representation from the army, MWD, America, France and Italy. It was an unattainable proposal which the War Office would certainly refuse, but d’Eyncourt made some telling points. He deplored the scattergun approach to tank development, the multiplicity of types, the penny-packet numbers ordered and the loss of output resulting from design changes imposed during production. He wanted Britain and the Allies to adopt and build a single heavy type and to drive production forward without interruption, regardless of demands for refinements. He urged the immediate cancellation of the Mk IX infantry supply tank (Rackham had begun its design in September), and of Mk V and Medium B production after August 1918. No new design of tank or engine should be considered. MWD should concentrate all its efforts on building the 800 Mk V and the 650 Medium B to schedule, without distractions. On their completion d’Eyncourt wanted Britain and the Allies to standardize on a Liberty heavy tank with infantry supply and other variants as necessary. Not the least of the benefits of a common battle tank was the opportunity to pool production of parts and assemblies, assigning specialist work to individual countries. If this was agreed with the Allies, he believed there was a sporting chance of putting tanks in decisive numbers into the field in July 1918. He pressed for the formation of three tank armies, perhaps one each from Britain, America and France, each of 100,000 men with 500 machines as their primary weapon, and with infantry and artillery as auxiliary arms. Their very presence along an extended front would pin down enemy forces fearful of attack, preventing them massing elsewhere for an offensive. Their strategic potential was self-evident.3
It took another war for that concept to become reality. The DNC’s paper brought a considered response from the General Staff in London which disputed his assertion that one tank equated to some 400 infantry in attack, but conceded that the machines brought ‘economy in personnel’. This was mere hair-splitting. The War Office saw no strategic role for the Tank Corps, nor any reason to increase its manpower until tank strength rose in August 1918, ignoring the fact that such delay meant the fighting season would be over by the time a new intake had been trained.4
Churchill attended a conference with M. Loucheur in Paris on 3 December at which Loucheur was given the details of the joint project and asked if his government approved. He affirmed that it did, providing some of the resultant tanks were allocated to France. An outline specification for the Liberty tank was agreed the next day at a joint British/American conference chaired by Churchill at GHQ. The first 600 of the Mk VIII, as it was later designated, would go to the American Army. The machine superficially resembled the current heavy tanks with all-round track and sponsons, but there were to be many improvements including a 14ft trench-crossing capability as against the 11ft of the Mk IV.
For the American Army the power unit would be a suitably modified 300hp US Liberty aero engine, with a similarly rated and as yet unbuilt Ricardo engine for British use. The large fighting compartment with a crew of eight (11 in the US version) could accommodate a further 20 men, and was separated from the engine by a bulkhead. A short 6-pdr gun in each sponson would be supplemented by five machine guns mounted in an elongated roof turret above track height to give all-round fire. Overall length was just over 34ft, against the 26ft 5in of Mk IV. A design committee was appointed, to be led jointly by commissioners Stern and Draine, with Capt G.A. Green and Maj Alden representing the British and US Tank Corps respectively and d’Eyncourt as technical adviser. When Gen Capper was informed he lodged a formal objection, telling Duckham it would sideline his committee and leave the MWD and the fighting side of the Tank Corps in France without a voice. He considered Green a competent engineer but out of his depth on the wider issues. Stern’s department would effectively determine the design of the British Army’s next main battle tank. The argument was patched up with an undertaking that while detailed design would be decided by the Anglo-American team, matters of general design would require the approval of Capper’s committee. Capper remained hostile to the introduction of an Allied voice in design, and implacably so to Stern’s role.
Gen Capper’s Tank Committee did not meet until the end of November 1917 – the battle of Cambrai intervened. It opened on the 20th when for the first time the tanks attacked en masse as their pioneers intended, and over almost unshelled ground. A total of 476 machines were deployed, of which 378 were fighting tanks supported by gun carriers, supply Mk IVs, three wireless tanks and 32 special wire clearance tanks. The action was originally conceived in August by Col Fuller, by then Tank Corps GSO 1, as a great one-day tank raid to restore British prestige and strike ‘a theatrical blow’ after the disaster of Third Ypres. His objective was to penetrate the allegedly impregnable ‘Hindenburg Line’ just south of Cambrai in the area of Gen Sir Julian Byng’s Third Army, destroy the enemy’s guns, demoralize and disorganize the defenders, and demonstrate that the tanks could break the line at any time and at any point. Byng was receptive and saw possibilities in extending the raid to take and hold Cambrai. Haig agreed, but Kiggell, his Chief of Staff, strongly opposed the scheme, saying troops could not be spared from the Ypres front. Fuller’s plan was accordingly shelved until mid- October. Haig then approved a modest tank action limited to 48 hours. Byng inflated it to an offensive to push cavalry through the break and take Cambrai before moving on towards Valenciennes. He had six divisions of tired infantry and four of cavalry, three tank brigades and some 1,000 guns. After Ypres there were no reserves save his cavalry to exploit success, or even to secure the long flanks resulting from a deep penetration.
Elles and Fuller had barely five weeks to prepare. Success relied on surprise, on getting the tanks across the exceptionally wide German trenches, and on effective cooperation between tanks and infantry. Surprise was to be achieved through Byng’s insistence on avoiding all preparatory bombardment, which also ensured good going for the tanks. Instead of registering the guns on their targets and so forewarning the enemy, all gun-laying was plotted by survey.
Trench crossing was left to Col Frank Searle and Maj John Brockbank, his number two. The Hindenburg Line’s forward trenches had been considerably enlarged and deepened to serve as tank traps. Central Workshops at Erin worked eight-hour shifts round
the clock to produce 400 tightly bound large bundles of brushwood and release gear, each tank carrying one on its nose for tipping into a trench before passing over it. As the tanks fought in sections of three they could sequence the drops to get the section across all three trench lines, the crews marking the crossing points with a red/yellow flag for the following tanks. Each drum-like 1.75-ton bundle was made up of between 90 and 100 smaller bundles or ‘fascines’ cut to 10ft lengths and laid in an open cradle over a pair of wire ropes. The rope ends were thrown across the top of the heap and shackled to a tank at either extremity, before the machines drew apart, pulling the ‘super fascine’ tightly to a diameter of 4ft 6in. Two securing chains were fitted before the cables were relaxed. A soldier scrounging for firewood was filing through a chain when it snapped under tension, killing him. Rumour had it that the chain’s whiplash cut him in half. The fascine release gear was operated from inside the tank.
In the event most machines were able to cross the trenches without help from the fascines; the German firestep was cut so wide that it supported the rear of the machine and stopped it tail-diving. Erin also made up 110 timber tank sledges for bringing supplies to forward rallying points (‘tankodromes’). Their tow tanks were fitted with roof-mounted hitching gear, the protruding rear track horns preventing any lower coupling.
Fuller evolved an attack drill with infantry. The lead tank of each section would keep about 100yd ahead of the other two, raking German parapets and drawing fire away from the following tanks, behind which the infantry advanced in single file over the flattened wire. The troops would mop up trenches and, importantly, support the machines by spotting for enemy guns. When guns were sighted, tank commanders were alerted via an external bell pull at the rear, or the banging of a shovel or crowbar left hanging beside a sponson door. Signalling was primitive; a shovel waved through the tank’s roof indicated ‘Tank broken down. Don’t wait’. The infantry could call for assistance from tanks by raising a rifle topped by a helmet. Conversely, a red/yellow flag was waved from the tank. A wireless tank was attached to each brigade and was usually run up to a ‘sheltered’ position where messages could be relayed by company commanders to brigade HQ. The system was little used, tank commanders having two carrier pigeons as preferred options providing the birds were not asphyxiated by then, or their feathers oiled.
The wire-pulling detachment was led by Stuart Hastie, now a captain, whose progress through Flers had so electrified the press in 1916. His 32 tanks were to clear three 60yd-wide lanes for the cavalry and following troops and guns immediately after the first line of trenches had been taken. They would continue until every succeeding belt had been dealt with. The tanks each pulled a heavy four-fluke grapnel on a 30ft wire cable secured to a short roof-mast. A machine could peel the ground bare of 20 tons of wire and iron pickets, so engulfing it that sometimes the sponson doors could not be opened.
On Y/Z day Elles issued an inspiring Special Order to his men:
1. Tomorrow the Tank Corps will have the chance for which it has been waiting for many months – to operate in the van of the battle.
2. All that hard work and ingenuity can achieve has been done in the way of preparation. 3. It remains for unit commanders and for tank crews to complete the work by judgement and pluck in the battle itself.
4. In the light of past experience I leave the good name of the Corps with great confidence in their hands.
5. I propose leading the attack of the centre division.
Fuller tried to persuade Elles not to enter the battle, fearing the consequences to the Corps if he were badly wounded or killed, but he was adamant. Fuller later admitted he was wrong: ‘To lead his command was to give life and soul to all our preparations – it was spiritually the making of the Tank Corps, and in value it transcended all our work.’5
On 20 November, after heavy overnight rain, the tanks moved off in thick mist at 6.10am, the infantry following. Byng threw in every machine, keeping no reserve. Ten minutes later at zero hour his 1,000 guns opened fire for the first time, laying down a creeping barrage on a predicted shoot. The absence of a preliminary bombardment had avoided alerting the enemy and precipitating a counter-shelling to cut up infantry and tanks in their jumpingoff positions. It was a stunningly successful action. The advance had penetrated over 5 miles when it was broken off for the day as winter’s early dusk set in. Four immensely strong defence lines had been overcome and only an unfinished stop line lay between the tanks and open country. Some 48 tanks had been disabled or destroyed. Over 100 guns and 8,000 prisoners had been taken. Most of the German wire was sited on reverse slopes where artillery was unlikely to cut it, but Hastie’s detachment had cleared lanes in a few hours. Casualties in killed, wounded and missing among the tank crews were very heavy, totalling 118 officers and 530 other ranks. Casualties among the six attacking infantry divisions were estimated to total 76 officers and 2,508 other ranks. Thereafter, exhaustion among the crews and the infantry, and the absence of reserves of either, greatly reduced the scale of further combined actions. The enemy were overwhelmed and put up little resistance that first day, but their later reinforcement soon halted further advance and the cavalry served only to choke lines of communication. A well-prepared German counter-attack on 30 November wiped out almost all the gains, but the devastating effectiveness of the tanks had rung the church bells at home and shown the way forward for 1918. They had penetrated up to 10,000yd in 12 hours from a base of 13,000yd. At the Third Battle of Ypres the base was 25,000yd and it took three months to achieve a similar gain, with the loss of hundreds of thousands of men and millions of shells.
Cambrai was the world’s first great tank battle. It was a personal triumph for Fuller whose meticulous planning of tank tactics and organization became a Staff training study for many years. Above all, it was a most gallant victory for the crews and their machines, regardless of the wider outcome of the offensive. They had begun to win the confidence of the army, but Elles considered it was well into mid-1918 before the void between the pro- and anti-tank factions among the staffs was bridged with those willing to assess the weapon dispassionately. Gen Capper observed after Cambrai, ‘I do not think that the battle confirms the view of those who hold that Cavalry are out of date. They should be invaluable to obtain speedy and decisive results.’6 He was, nevertheless, convinced of the tank’s growing role, writing to Elles: ‘Gen Furse [Master General of Ordnance] is a strong supporter, and cannot understand the attitude that is taken up by many still as to the purely accessory nature of the new arm, which I think will more and more develop into a main feature of the battle.’7
Stern was able to remind Churchill of the charge that he had lumbered up the front with useless tanks and wasted a year of development – Cambrai told a different story. He warned the Minister that the MWD under Moore would never produce tanks in sufficient quantity or speedily enough to win the war in 1918. Stern also criticized the creation of his own parallel O&A department alongside the MWD, and his exclusion from the New Tank Committee. Churchill spoke of difficulties but promised to get him into membership. (He tried and failed; the committee agreed to invite Stern to ‘relevant’ meetings but not to participate as a member.) The interview brought a private rebuke from Duckham who warned Stern that he could not continue to ignore his negative attitude.
The New Tank Committee (henceforth ‘the Tank Committee’) heard Adm Moore’s first situation report as Controller on 26 November. His delivery forecasts were based on current materials and labour priorities.
Type Initial Order Completion in 1918: forecast (actual)
Mk IV fighter 950 January (late April)
MK IV converted to tank supply tenders 205 End of January (July)
Medium A (Whippet) 200 April (October)
New Types
Mk V 600 June (400th delivered June. Production then ceased)
Medium B 380 175 by June (Priority reduced. 26th tank delivered Armistice week, 11 November 1918)
Mk IX infa
ntry supply 200 Contract about to be placed. Completion expected May (23 delivered by Armistice)
Mk II Gun carriers, etc.
Gun carriers 120 Just contracted, no forecast possible (None built)
Salvage 30 As for GC above (None built)
The War Office had earlier given notice that it was prepared to accept 216 of Stern’s 400 unauthorized Mk IV overproduction if they were converted to tank tenders, but it refused to take delivery of the remaining machines. It was also clear that the heavy tanks would have to cross much wider trenches. Mk V production was accordingly halved in December to 400 machines as the type was unable to cross gaps greater than 10ft. Deliveries began in February 1918.
At the same time the WO ordered 300 of a male variant, the Mk V Star. It had been lengthened by 6ft, increasing trench crossing performance to 13ft. The order was raised early in 1918 to a total of 500 male and 200 female machines. Delivery began in June and by the Armistice 579 had been built. Instead of the box-like rear turret with two revolver ports as on the Mk V, the fore and aft plates of the Star’s turret were each fitted with a ball mounting for a machine gun and sloped to give them greater elevation. David Fletcher of the Tank Museum believes this modification was made to aid street fighting – the Germans knew that armament on tanks could not be elevated to sweep Maxim and bombing positions on the upper floors of adjacent buildings.8 It was intended that the Star should have Ricardo’s new 225hp engine to compensate for a 4-ton weight penalty but these fine motors were not available in time. With the Ricardo 150hp engine of the Mk V the Star was under-powered, and a pig to steer by virtue of the longer run of track in ground contact. A Two Star version with the 300hp engine was ordered in January; the 300 machines were to be delivered by 15 November. The length of track in ground contact on a hard surface was cut to only 6ft to improve steerability, and the fighting compartment was moved from the waist to the front of the tank. Shortage of steel delayed production. The first Two Star machine was completed in December 1918, and a total of 25 were built.
The Devil's Chariots Page 29