Given, however, the impossibility of another combined strategical plan for 1917, there were urgent motives and sound reasons for the extension of Haig's offensive northwards from Arras to the coast. If it were successful beyond expectation, it would achieve all that Nivelle had hoped to do by a frontal attack, and would compel a general German retreat by turning the enemy's flank as Joffre had tried to turn it in October 1914. But short of such extravagant anticipations it might materially help to win the war by defeating the real German offensive for 1917. That was not a campaign on land at all, but on sea by means of the submarine, and the chief basis of operations was the Belgian coast. Submarines emerged from other lairs, but the German command of the Belgian coast shortened their distance from their objectives by hundreds of miles and correspondingly lengthened their range of operations. Bruges was their headquarters; situated inland, but connected by canal with Zeebrugge and Ostend, it afforded a base immune from any attack save those of aircraft, and Bruges was the real objective of our Flanders campaign. Incidentally, too, the Belgian coast provided harbours whence light German surface craft made occasional raids on British coasts, commerce, and communications, and also for those aeroplane attacks which became a serious nuisance as the year wore on. Apart from these considerations the German hold on Flanders was the bastion of their whole position west of the Meuse; and, but for the natural feelings of Paris, a more strenuous attempt might well have been made earlier in the war to deprive the enemy of its advantages. Obviously in the summer of 1917, if the two Allies were to be left to their own devices, there was none which suited us better than the Flanders campaign, and the official American commentator opined that it held out more fruitful prospects than the battle of the Somme. The drawback was that campaigning in Flanders depended upon the weather: a rainy season turned its flats into seas of mud, and the third quarter of 1917 was one of the wettest on record.
A preliminary obstacle to be overcome was the Messines-Wytschaete Ridge which dominated Ypres and the whole of the line from which an offensive in Flanders could start. Preparations to deal with it had been in progress since early in the year, and heavy guns had also been mounted on our positions near Nieuport. The plan indeed had been in Haig's mind since November 1916, and even earlier than that Sir Herbert Plumer had been training the Second Army for its task; it had had no serious fighting since the second battle of Ypres in April 1916, the battle of the Somme having been fought by the Fourth and Fifth, and that of Arras by the First and Third. The victory, however, was to be largely a triumph of engineering science. For nearly a year and a half tunnelling had been in progress under the ridge, and at dawn on 7 June nineteen huge mines were exploded beneath the enemy's lines in the greatest artificial eruption that had ever shattered the earth's crust. Ten days' surface bombardment had already obliterated much of the German defences, and it says something for the German moral that any resistance was offered at all when our troops advanced over the ruins of the soil. Messines was cleared by New Zealanders by 7 a.m., Wytschaete fell by noon before Ulstermen and Irish Nationalists fighting side by side, and Welshmen captured Oosttaverne a few hours later. The battle could not have have been better staged to exhibit the co-operation of the British Empire and of mechanical science and human valour. A few days later Australians pushed on to Gapaard and La Potterie in the direction of Warneton, and the Germans withdrew from all their positions in the salient. The danger to Ypres which had threatened for over two years and a half and had cost so much in British blood, had at last been exorcised, and from being an almost forlorn hope of defence the Ypres salient became the base of a promising advance (see Map, p. 288).