Another problem with Intelligence at the time was that although aerial photography could show enemy trenches and even individual dugouts, it could not discern how strongly the lines were held. Cavalry could no longer carry out reconnaissance in the age of trench warfare, so aeroplanes had to do the job, as well as help find enemy guns and help guide fire against them. In 1915 aerial photography and ground-to-air communications were invented, revolutionizing the efficacy of the air arm. In February 1916 at Verdun, fighter aircraft were employed by both sides to protect photo-reconnaissance planes, and of course to shoot down enemy observers. Haig liked and trusted Maj.-Gen. Hugh ‘Boom’ Trenchard, commander of the Royal Flying Corps (RFC), and supported his aggressive air strategy. His airmen paid a heavy price on the Somme, however: from 1 July to the end of the battle on 18 November the RFC lost 782 aircraft and 576 pilots.29

Other classic sources of intelligence were similarly limited; civilians were moved out of the combat areas, prisoners were hard to capture, newspapers were censored. So trench raids, sometimes with diversions, were staged to bring back information. The evidence collected in them could be as slight as a canteen cover with a regimental number stenciled onto it, but of course documents, orders, letters and prisoners were far better. In the days before the Somme Offensive, however, only about twenty prisoners were taken by the 4th Army because of the bombardment, not enough to deduce anything very useful. ‘Although the raiders did not succeed in entering the enemy’s trenches,’ one regimental war diary noted, ‘the Corps Commander considered that they attained most useful results, as it has been established that the enemy’s front-line trenches are strongly held.’30 Part of the reason why the British High Command believed so implicitly that the bombardment would destroy the German defences, barbed wire and morale was because General Joffre had said that that was the effect the German bombardment had had on the defenders of Verdun.31 German prisoners captured on the Somme seemed to indicate that their morale had been broken. But the prisoners—even supposing they were telling the truth rather than what they thought their captors wanted to hear—were an unrepresentative group. They had been captured, after all.

There was thus a collective assumption, especially in VIII Corps, that the preliminary bombardment, and especially its final sixty-five minutes, would destroy and demoralize the Germans in their forward positions, yet intelligence contradicting that theory was gleaned throughout the previous week by trench raiders who found uncut wire, or were vigorously repulsed, which should have given planners pause for thought. ‘Critical evaluation of the intelligence gathered over the previous seven days’, one historian states, ‘seems to have been non-existent.’32 Here was group-think at its most pernicious.

It was not until 1917 and 1918 that aerial photography and signals intelligence provided much help to Charteris and his successors in trying to work out the whereabouts of German theatre reserves. The agent network in Belgium did not provide anything useful until 1917 either, and the situation was not helped by a turf war between GHQ’s Intelligence operation and the Secret Intelligence Service.33 The Allies had no moles in the German High Command or the senior reaches of government and had access to nothing like the Ultra decrypts that led to such brilliant results in the Second World War. Charteris was perhaps doing his best in very difficult circumstances, but if he was often flying blind it seems he did not confess his relative ignorance to his commander-in-chief. Few Intelligence chiefs ever do.

The surprising thing is that the general paucity of information induced such over-confidence in Charteris. In February 1917 he was still warning audiences, for example, not to believe that the enemy was in good shape, whereas in fact the German army really was a formidable machine almost up to the moment when it cracked in the summer of 1918. Charteris must not be used as a scapegoat for all the High Command’s errors, however; every commander needs to be aware of the limitations of Intelligence. Besides, as the government insider Lord Esher told Haig’s private secretary, Philip Sassoon, Charteris ‘had no more real influence with Haig than his chauffeur’.34

By April 1916, British Intelligence had recognized that the Germans had improved their front-line defences on the Somme, having long perfected their first and second lines and now starting on a third. Yet 4th Army planners did not know about the completion of the third line until the preliminary bombardment began.35 They did know, and had since January, that the Germans were building significantly deeper dugouts and that the openings to these were placed at the front wall of the trenches rather than the back, making them harder for the artillery to hit.36