End of the Battle of Pilckem Ridge

On 2 August 1917, for the third day running, it rained over the Ypres salient. The first great attack of the Third Battle of Ypres, which on 31 July had given the British and French Pilckem Ridge, had turned into a struggle with mud. Both sides still tried to improve their positions: in the afternoon British artillery broke up two German attacks near Pommern Redoubt, and north of Ypres the British again occupied the ruins of Sint-Juliaan (St. Julien) and pushed posts across the Steenbeek stream. On the same day the commander of the Fifth Army, General Hubert Gough, postponed the resumption of the offensive. The British Battles Nomenclature Committee later recognised 2 August as the last day of the Battle of Pilckem Ridge.

Rain and mud

According to measurements cited by historians, more than 20 millimetres of rain fell on 31 July, and 5–10 millimetres a day over the next four days. The rainfall itself was not a record, but the sky stayed overcast, there was no wind and the water hardly evaporated. Two weeks of shelling had moreover destroyed the Flanders drainage ditches, and the Steenbeek, Hanebeek and Bassevillebeek streams overflowed their shattered banks. A belt of land several kilometres wide behind the new front line turned into a quagmire full of water-filled shell holes. Soldiers carrying ammunition and food walked on duckboards, and anyone who stepped off them could drown. The wounded were carried to dressing stations for hours. Airmen who were supposed to direct the artillery fire could hardly get into the air.

Gough’s and Haig’s decisions

Gough wanted to resume the attack on the most important sector, the Gheluvelt plateau, as early as 2 August, and planned a second great blow by the whole army for 4 August. On 1 August the head of operations on the staff of Field Marshal Douglas Haig, General John Davidson, warned against haste. He advised that II Corps should attack only after two or three fine days, when the artillery would have had time to destroy the German shelters marked on the maps captured on 31 July, and that fresh divisions should be used for it. On 2 August Gough called off the attack, and the commander of the British tanks, General Hugh Elles, obtained permission to hold back the surviving Mark IV machines until the ground dried. Of the roughly 140 tanks that had set off on 31 July, about half had got stuck or been destroyed.

Haig remained optimistic. On 4 August he assured the Cabinet in London that the attack had been a success and that the losses, about 32,000 killed, wounded and missing from 31 July to 2–3 August, were small for so great a battle. According to the British official history, the Fifth Army alone lost 27,001 men from 31 July to 3 August, 3,697 of them killed. Haig also claimed that nine German divisions had been shattered and had had to be hastily relieved. Further dates were set and moved back for another two weeks: the attack on the plateau did not begin until 10 August, and the second general attack, at Langemarck, on 16 August.

On the other side

The Germans were suffering too. On 2 August the staff of Group “Ypern”, which held the sector from Pilckem to the Menin road, correctly judged that the British would strike again at the plateau and to the north, towards Langemarck. It also reported that of its six divisions only two were still fit for battle, while the other four had lost about 10,000 men between them. A day later the army group commander, the Crown Prince of Bavaria, Rupprecht, wrote in his diary that the weather was rapidly wearing out the infantry. Divisions lost 1,500–2,000 men each in the line, fewer than on the Somme in 1916, but only because they were relieved much more often. British gas bombardments surprised the units carrying supplies to the front line: the 6th Bavarian Reserve Division alone lost about 1,200 men to gas.

The weather in historians’ dispute

August 1917 was exceptionally wet in Flanders. About 127 millimetres of rain fell in the whole month, and the cloudy sky and lack of wind meant that the ground did not dry. The historian John Lee called it the worst August in 75 years. In 1929 Haig’s chief of intelligence, General John Charteris, wrote that a study of records covering more than eighty years had shown that the weather in Flanders broke at the beginning of every August with the regularity of the Indian monsoon. His words were later quoted by Lloyd George and many historians as proof that the offensive had been planned recklessly. The British official history rejected them, and the army’s chief meteorologist, Ernest Gold, wrote that they were contradicted by the facts: measurements from Lille for 1867–1916 show that August there was more often dry than wet. The dispute over whether the rain could have been foreseen has not died down to this day. One thing is certain: the downpours that began on the afternoon of 31 July robbed the offensive of its momentum and gave the Germans time to bring up reserves.