note on nomenclature.—throughout the chapters dealing with the boer war i use the expression “cavalry” to mean british regular cavalry. i use the expression “mounted infantry” to mean regular british mounted infantry (i.e., drawn from infantry battalions). i use the general expression “mounted riflemen” to cover all mounted troops, boer or british, armed only with the rifle.
the campaign opened in natal with the attempt of general sir w. penn symons, with 4,000 men and 18 guns, to hold the untenable northern position at dundee against a greatly superior converging force of joubert’s transvaalers. sir george white, who only with reluctance had consented to this attempt, was concentrating at ladysmith, and facing the free staters; while midway between white and symons a detached boer force, 900 strong, under koch, war about to plant itself upon the railway connecting dundee and ladysmith. symons’s mounted troops were one regiment of cavalry, three companies of mounted infantry drawn from the three battalions which formed his infantry brigade, a squadron of natal carbineers, and a few picked guides. joubert’s southward advance from the frontier was excessively slow—seventy miles in a week. watched and reported by cavalry and other patrols, it nevertheless culminated in a complete surprise of the british camp at dawn on october 20, 1899, by meyer’s force of some 4,000 men and 8 guns. the general’s overconfidence was the principal cause of this surprise, and it is interesting to 62note that his reason for not establishing more cavalry pickets to supplement the inadequate system of defence in the heights above the dundee valley was that he wished to keep the cavalry fresh—fresh, that is, for shock action. the battle of talana was, from our point of view, an infantry fight, fought with splendid spirit and tenacity, and, for the moment, a victory. from the boer point of view, in this case, as in all others, it was a mounted rifleman’s fight. our own mounted troops were employed with an aggressive purpose, that of turning the boer right and intercepting the boer retreat. they consisted of cavalry and mounted infantry acting in concert, the latter, according to the regulations of that period, being regarded as a “valuable auxiliary to the former.” the movement began well. an admirable, but also a somewhat dangerous position, was gained well behind the main boer force, within range of its led horses and commanding its line of retreat, at a moment when retreat was just setting in. stratagem and fire-action combined might have produced great results. shock was preferred. a few boers were sabred, some thirty prisoners were taken, and then the movement collapsed. the boers took the offence. the commanding officer on our side lost his head, and, after much difficulty, half the cavalry got back without their prisoners to the british lines; the rest of the force, after a running fight, in which the rifles of the mounted infantry were the only effective means of defence, was surrounded and forced to surrender. it would be unjust and undiscerning to make too much of this opening episode. nevertheless, in so far as the value of the arme blanche was concerned, not merely as a weapon, but as an inspiration of resourceful and effective man?uvre, the incident was of bad augury.
the next day, october 21, came elandslaagte, fought on the line of communication connecting dundee and ladysmith between koch’s force of 900 men and 2 guns, 63planted astride the railway, and a mixed force of 3,500 men and 18 guns sent out by white from ladysmith under command of general french. our mounted troops were three squadrons of cavalry, five of the imperial light horse, and a few natal volunteers. the fighting, which ended brilliantly for ourselves, was highly honourable to both sides. from the boer point of view, it consisted in a magnificently stubborn defence of a strong position by an inferior force of mounted riflemen, fighting on foot up to the moment of actual contact, and under crushingly superior artillery fire. from our point of view, with one interesting novelty, to which i shall refer later, it was a plain, hard, straightforward fight with the three arms co-operating on thoroughly conventional lines: the infantry carrying through a well-planned frontal attack with remarkable dash; the artillery shelling the main position; the cavalry watching both flanks during the progress of the action, and, just at dusk, after the final repulse of the enemy from the main position, pursuing with the lance and sword. the pursuit, carried on for about a mile and a half with vigour and enthusiasm, touched only a portion of the retreating burghers, but, so far as it went, it was effective: it struck the “terror of cold steel” into the pursued with scarcely any loss to the two squadrons engaged; it caused casualties and surrenders, though precisely to what extent is difficult to say. no figures exist. in short, the cavalry had performed with considerable success the peculiar function traditionally assigned to their arm.
now let us turn to the unconventional feature of this fight. the imperial light horse, early on the same morning, had made the reconnaissance on which the battle scheme was founded, and had seized and held necessary tactical points. they had rushed the railway-station by a gallop in open order. together with the cavalry (who came out later with the main force from ladysmith) 64they had prepared the way for the infantry advance, and had helped to clear a flank during the early part of the action. but in addition to these duties they dismounted and joined with the infantry in the assault of the main position, took a prominent and, at one critical moment, a decisive share in the desperate fighting which wrested it from the boers, and suffered losses (including that of their brave colonel) heavier than most of the units engaged.
mr. goldman, in remarking on elandslaagte, makes the strange comment that the imperial light horse were “trained as cavalry,” and adduces their exploits on this occasion as an example of the value of that arm in south africa.[17] this is the first of many misinterpretations upon which i shall have to comment. for all practical purposes the imperial light horse were mounted riflemen, who used rifles, not carbines, and, as far as i know, never in all their history made or attempted to make an arme blanche charge, yet were very effective in action, and were very fair scouts. used for the bloody assault at elandslaagte, they could not also be used for the pursuit. if they had not joined in the assault, could they, or troops of their type, have been used in an equally effective way for the pursuit? the inquiry compels us to look back a little more closely at the conditions of the charge.
the following points should be noted:
1. for the troops engaged on both sides this was the first day of hostilities. steel-armed cavalry was a new fact to the boers. the steel had the best chance it ever was to have of inspiring “terror.”
2. there were no boer reserves left to cover the retreat.
3. the light was failing, a circumstance favourable to the steel, unfavourable to fire. (contrast the broad daylight at talana, when the boers rallied and outman?uvred 65the cavalry.) some light is necessary, of course, but, within obvious limits, the poorer the better.
4. the ground was as open and smooth as cavalry on the average can expect. dongas and rocks during the initial advance only; from within 300 yards of the enemy and onwards (according to the “official history”) not a lawn, but fair galloping ground.
5. horses and men fresh, not hitherto seriously engaged. why? because there had been no opportunity for the use of steel.
6. the enemy, already shaken and spent by a hard fire-fight on foot, were retreating at their usual ambling trot, in loose, formless groups; “raggedly streaming,” as the official historian correctly puts it. he adds that this objective, “a crowd in the loose disorder of defeat, seemed to offer an indefinite object for a charge,” but “that there was no likelihood of a better whilst sufficient light remained.” i must digress for a moment on that illuminating obiter dictum, because it gives a clue to the cavalry view of cavalry work. the historian is regretting the absence of a chance for “shock,” in its literal and in its only accurate meaning, of the collision of two massed bodies; two, and both massed. the boers were not massed; clearly, therefore, it was of no use for the cavalry to adopt mass, and in point of fact they charged with “extended files.” there could be no “shock,” therefore—that is, violent physical impact—and there was in fact none. the boers were ridden down individually. what the official commentator does not apprehend is that this absence of mass, in his view an unfortunate drawback, was in fact one of the very conditions which made the charge possible. it was a corollary to the beaten, spent state of the pursued. ragged streaming away is a characteristic of defeated troops in retreat. cohesion means morale, and morale means the will and power to retaliate. nor is it only a question of morale. the physical 66conditions of the preceding fire-fight determine the nature of the retreat. in this case some 900 boers, in widely extended order, had been defending a line nearly two miles long against an enemy proportionately extended, both extensions being truly normal—that is to say, dictated by the range and deadliness of the modern rifle. retreat from such a line, immediately after a failure to withstand a punishing assault, pressed in some quarters to the bayonet’s point, excludes cohesion in any troops, european or extra-european. boers, as i pointed out in the previous chapter, never troubled much about set formations at any time, whether or no there was time for them, not through incapacity, but simply because they did not need them, and not needing them were better without them. for them, therefore, this kind of ragged retreat was not solely the result of the beating they had suffered. normal in any troops, it was normal in a peculiar sense with them.
i dwell on this point at some length, not because of the intrinsic importance of this fight, or of the official historian’s comment upon the pursuit (for he may have written thoughtlessly), but because it directly raises the big issue dealt with in my analysis of the physical problem in chapter ii. i enumerated there the many crushing limitations which surround the use of real shock against riflemen, mounted or on foot, and i instanced the pursuit of beaten troops as one of those rare cases where the steel weapon has its best opening. but i also pointed out that this was a case where any well-mounted troops, however armed, have a good opening; and that brings us back to the point from which we started in comparing, for the sake of illustration, the work of the imperial light horse and the cavalry at elandslaagte. first, however, let us recapitulate the six favourable conditions of this cavalry pursuit:
(1) novelty of the steel. (2) no boer reserves. 67(3) bad light. (4) open and smooth ground. (5) fresh horses and men. (6) ragged retreat of beaten enemy.
this may be regarded as a rare combination of ideal conditions; how rare will be seen as the war proceeds.
now for the imperial light horse, whom, let me say, i am regarding, not as an individual regimental unit, but as a type of what good riflemen can do, just as the cavalry squadrons engaged were types of what cavalry, decidedly good according to the standard of their time, could do.
i asked, would the imperial light horse, if they had not been used for the fire-fight, have been capable of an equally effective pursuit without the use of steel weapons? the speculation, of course, though instructive, is largely academical, the crucial point being that they had been used for the preceding fire-fight. however, for the sake of argument, we must vest them with favourable condition no. 5, “fresh horses and men.” nos. 2, 4, and 6 would have been equally applicable to them; no. 1 is irrelevant. there remains no. 3, “failing light.” this would have been distinctly adverse to the accurate use of the rifle, but at the same time let us remember the fundamental distinction between the rifle and the steel—that is, range. posted, for the sake of argument, in the spot where the cavalry were posted (threatening the enemy’s right rear), the imperial light horse would at once have had the first bodies of retreating boers well within the range of vulnerability: 500 yards is the official estimate. yes, but fire at this moment would no doubt have meant delay, and caused less damage to the boers than the undelayed steel-armed cavalry. granted; a point to the cavalry. let us go on. after routing a first batch in a long gallop, the cavalry turned on their tracks, met a second batch, and scattered and harassed these men also. would not the imperial light horse meanwhile have had a good chance of intercepting these men? finally, picture the irregular 68corps as capable of fire from the saddle, and keep that point in your mind for future illustration.
all this is the veriest sketch, suggestive of the factors inherent in mounted combats, but utterly unreal, because it is utterly impossible to postulate identical circumstances for steel-action and fire-action. the essence of the matter is that the imperial light horse, by aptitude, training, and equipment, were capable of joining effectively in the infantry assault of the main position, and that the cavalry, by aptitude, training, and equipment (they carried the short carbine), were neither capable of, nor designed for, similar intervention. if the colonials had not been used for the main assault, the course of the battle might have been changed. the assault might have failed (in the penultimate phase there was an exceedingly critical revival on the boer left flank, checked by the gordons and imperial light horse combined), or the assault might have been consummated too late to give to the cavalry the margin of light necessary for their pursuit. or—and this is really the most pertinent and suggestive eventuality—the imperial light horse used as their capacity deserved, might have operated actively on the enemy’s rear at an earlier period, when the cavalry was still passive. result, a change of battle conditions, which defies speculation. on the other hand, we can, to a certain extent, isolate our view of the cavalry exploit. they did, under ideal conditions, exactly what they were trained to do, and i do not think they, or any other cavalry similarly trained, could have done it better.
in dwelling so long upon the topic of pursuit we must remember that there was no question at any moment of a charge by cavalry either upon unbroken riflemen or upon led horses. nor (save in the case of the rush upon the station by the imperial light horse) was there any attempt on the part of the mounted riflemen on either side, boer or british, to carry aggressive mobility to the 69point of charging on horseback into point-blank range of riflemen on foot.[18] developments of that sort were still a long way off.
i have enlarged so much on this small fight in order to focus the reader’s attention upon the principles it illustrates. let him study it in conjunction with the action of talana, which preceded it, and with all the multitude of fights which followed it, in the next two and a half years. let him begin at once to picture parallels in european warfare, on a bigger scale or smaller scale, and ask whether they tell for or against the arme blanche, and why? imagine the 900 boers as a german force, either of cavalry or of the three arms in normal proportion, and without anything in the least degree resembling either our imperial light horse or the militant burgher. should we have won more or less easily? or imagine 3,500 germans, constituted as before, tackling the 900 boers. instead of moderately open ground, suppose ground diversified with copses, walls, hedges, a sunk lane or two. make any permutations or suppositions that you please, and test each by south african facts.
finally, ask yourself at every step, on which method, that of the arme blanche or the rifle, will it pay best in the long-run to train mounted troops?
