The chief Boer successes of the same type were at Wilmansrust (June 12, 1901), Quaggafontein (September 20, 1901), and Tweefontein (December 24–25, 1901). Careless outpost work by irregular troops was responsible for all three reverses. On the first two occasions camps on the level were rushed and overpowered instantaneously; but Tweefontein, besides illustrating stratagem and stalking skill, is also suggestive of the risks taken by a force which attacks in the dark. De Wet’s men scaled a precipitous cliff to storm the British camp, and, in doing so, overlooked a strong picket ensconced below the crest on the opposite side. It is possible that if reinforcements to the hill had come as promptly as they might have come, this picket, which was eventually discovered and overpowered, might have served as a useful point d’appui for a counter-stroke. At night, in the confusion of a sudden assault, the slightest stand made by a handful of determined men is likely to bewilder and daunt the enemy.
Lake Chrissie (February 5, 1901) and Moedwil (September 30, 1901) were finely conceived and finely executed night attacks by Botha and De la Rey respectively against columns under Smith-Dorrien in the one case and Kekewich in the other. Both were repelled in the most spirited fashion, but in both there were moments of extreme danger. At Langerwacht (February 23, 1902) there was a very dramatic and exciting night combat, when De Wet, to avoid envelopment in one of our great drives of the latest model type, burst through the cordon of entrenched pickets with a horde of waggons, carts, cattle, and non-combatants. There were several other episodes of the same type at that period.
Nooitgedacht (December 13, 1900) may also be placed in the category of night attacks. De la Rey’s first and unsuccessful attack was delivered in pitch darkness; the subsequent assault of Beyers in the grey of early dawn.
All the above night attacks were upon the camps of mobile forces, but there were many others upon fortified posts and towns. Helvetia (December 29, 1900) and the small post at Modderfontein (January 30, 1901) were stormed in darkness. At Vryheid (December 11, 1900) an outlying post and the Mounted Infantry camp were rushed under the same circumstances, though the main position held out gallantly. Belfast (January 7, 1901) had a similar, but a more dangerous, experience, losing a strongly held outlying post and two entrenched posts, all defended with great tenacity, shortly after midnight and in misty weather. But the mist and darkness eventually favoured the defence. Viljoen and Botha, in endeavouring to unite their forces against the inner defences, lost their way, and had to retire baffled. The six other attacks on the garrisons of a section of railway forty miles in extent, made simultaneously on this same night, were carried out with marvellous punctuality, but were all gallantly repulsed. In the Western Transvaal, at a later date, De la Rey’s unsuccessful attack on Lichtenburg (March 2, 1901) was begun and carried on for several hours in the dark.
One of the most thrilling episodes of this class was at Itala (September 25, 1901), the frontier post under Colonel Chapman, which Botha struck at when he was trying to raid Natal. An outlying post on the peak of Itala Mountain was taken by a sudden coup de main at midnight, and the fight, fiercely contested on both sides, raged round the central position until dawn and throughout the following day. At nightfall there was a lull, during which each side concluded that the other was irresistible, and both retired! Prospect, a neighbouring frontier fort, was also attacked on the night of the 25th, but held its own with ease.
Columns on the march were very rarely attacked in complete darkness. The only case I know of is that of Yzer Spruit (February 24, 1902), where De la Rey ambushed a convoy, beginning his attack before the dawn. Attacks in twilight were common.
Scrutinizing these incidents with a view to our special inquiry, let us note three points:
1. This is of general application—that is, to day or night attacks. All mounted troops should, in the art of entrenchment, be as nearly as possible the equals of Infantry. Though regular Cavalry were not, I think, concerned in any of the above incidents, the kind of work involved, whether in attack or defence, was work which normally falls to Cavalry in all modern war. Troops who cannot make entrenchments will never be able to storm them.[[51]]
At this moment the regular Cavalry are supposed to be able both to attack and defend entrenched positions. “There are certain difficulties in modern war,” admits “Cavalry Training” on page 186, “which cannot be overcome by mounted action”—that is, by shock action. This action, it is explained, “is precluded against an enemy posted behind entrenchments or occupying intersected or broken ground,” or “an extended position,” etc. In other words, the Cavalry are expected to be able to do the same offensive work as Infantry. Can they do it? How far could they do it in South Africa? Similarly in defence. They are “to deny important points to the enemy” by fire-action (and presumably to deny them effectively), and on page 215 (“The Defence”) they are “often to be called upon to occupy localities for defence, especially in small bodies.... Whenever time and means permit, the position should be put into a state of defence; the preparations, however, should be limited to those of the simplest kind.” The italics are mine. It is thus that, after South Africa and Manchuria, we persist in ruinous error. One thinks of Majuba, of Spion Kop, of Nicholson’s Nek, Dewetsdorp, Nooitgedacht, and only too many other examples of the Nemesis which attends “defences of the simplest kind,” no matter by what class of troops they are made and used. The compilers of the section entitled “Dismounted Action” should have taken to heart the lesson of Zilikat’s Nek (July 11, 1900), where regular Cavalry were concerned, both in defence and in attack. Of course, behind all the compromise which pervades the section there lies the fatal obsession that openings for shock action must at all costs be allowed for, and that, in defence, entrenchments should not be so good as to encourage Cavalry to rely on them, to the prejudice of “mounted action,” which in Cavalry language means shock. This is to disregard the facts of war. Why did not the Cavalry execute shock charges at or after the Boer assaults on Wagon Hill? They were there, fighting bravely enough on foot in defence, but the counter-charges were made by Infantry and irregular horsemen acting dismounted.
2. Nothing, not even the strongest entrenchments, can replace vigilance. Here the Cavalry showed an excellent example to their irregular comrades. Cavalry outposts were rarely surprised, and, I think, there was only one case of any consequence of a homogeneous Cavalry force being completely surprised in daylight.