The selection of a theatre of operations depends on the objective to be gained; the gaining of the objective on the breaking down of the enemy’s resistance. Consequently the weapon which will most speedily overcome this resistance must be considered first, and the area of attack in the theatre of operations chosen must be selected as far as possible with reference to its powers.

In the present instance we find that the chief resistance to our infantry advance comes from the enemy’s machine guns. We dare not concentrate all our artillery on these, for if we do, we shall release his guns, which, free, can put up a stronger resistance than his machine guns on account of their superior range. Further, whilst by sound and flash ranging and aeroplane observation we can discover his main gun positions, no means have yet been discovered whereby his machine guns can be located other than by advancing on them and risking casualties. Tanks must, therefore, be employed to do this in order to clear the way for the infantry advance. Consequently, if sufficient tanks are forthcoming, in order to guarantee a decisive success, it is no longer a question of the tank as a spare wheel to the car, in case of an unforeseen puncture in our operations, but as the motive force of the car itself, the infantry being merely its armed occupants; without which the car is valueless.

The area of operations selected must firstly be suitable to the rapid movement of tanks, and secondly, unsuitable to hostile anti-tank defences. Further, it should be chosen with reference to the tactical characteristics of this arm. Once chosen, all other weapons should be deployed and employed to facilitate the advance of the tank, because it is to be used as the chief maintainer of infantry endurance, and it is the infantry man with his machine gun and bayonet who is going, for some time to come, to decide the battle.

Such were the views held in the Tank Corps at the opening of the Third Battle of Ypres, and the following extract taken from a paper written on June 11, 1917, is not only of interest but prophetic of future events:

“If we look at a layered map of France we can at once put our finger on the area to select. It lies between the Scarpe and the Oise, the Flanders swamps in the north and the Ardennes in the south-east. It was down this funnel of undulating country that the Germans advanced in 1914, and it is up it that they will most likely be driven if strategy is governed by ground and tactics by weapons.”

The main area suitable for tank operations having been fixed upon by the Tank Corps, the next requirement was to select a definite objective, the attack against which would draw the enemy’s reserves towards it and so relieve the pressure which was being exerted against the Fifth Army at Ypres. Two localities were considered, St. Quentin and Cambrai. The first was opposite the French area, the second opposite the British.

The suggestion put forward as regards the St. Quentin operation was abandoned on account of difficulties arising out of a British force operating in the French area; it must be remembered that at this time no real unity of command existed in France.

The Cambrai project consisted in a surprise raid, the duration of which would be about twenty-four hours. The whole operation may be summed up in three words, “Advance, Hit, Retire.” Its object was to destroy the enemy’s personnel and guns, to demoralise and disorganise his fighting troops and reserves, and not to capture ground or to hold trenches. It was further considered that such an operation would interrupt his roulement of reserves and make the enemy think twice as to replacing fresh divisions by exhausted and demoralised units in those parts of his line which were not included in his battle front. Further, it would confuse him as to the decisive point of attack; for any day one of these raids might be followed by a strong offensive.

The actual area of operations selected was the re-entrant formed by the L’Escaut or St. Quentin Canal between the villages of Ribecourt, Crèvecœur, and Banteux. The going in this area was excellent; further, the area to be raided contained several fair-sized villages and important ground, and was well limited by the canal, which not only made a rapid reinforcing of the area in the bend difficult, but completely limited the tank objectives.