such are the facts of the south african war, our only great war since the crimea, and the first serious test for the whole world of the smokeless magazine rifle. what results can we place to the credit of the arme blanche?

1. the pursuit at elandslaagte (october 21, 1899), on the second day of hostilities: boers killed, wounded, and prisoners, say fifty. (no figures are forthcoming, but i think fifty is on the safe side.)

2. klip drift (february 15, 1900): a “penetrating,” semi-aggressive charge, in widely extended order, by a very large force, with a big backing of infantry and artillery, through a gap in a small hostile skirmishing screen. boer casualties about fifteen.

3. diamond hill (june 11, 1900): two brave but insignificant little charges, which received as much punishment as they gave. boer casualties about seventeen.

4. welgevonden (february 12, 1901): a small charge in the open. boer casualties and prisoners about twenty.

not a single example of true shock.

this gives a record of about a hundred casualties and prisoners due directly to the arme blanche. there may, no doubt, have been a few others in unrecorded episodes. to be well on the safe side, let us put the total at 200. all the other damage inflicted by the cavalry, whether in offence or defence, was inflicted through the agency of the carbine or rifle. the opportunities lost through 262over-training in the steel and inexperience in the firearm are beyond computation.

with the exception of an unknown, but certainly small, proportion of casualties caused by artillery, all the other losses in action, british and boer, during the war were caused by the rifle, and all of our own casualties, close upon 30,000 in number, were (with the same exception) inflicted by mounted riflemen.

from the first to the last day of the war the rifle dominated every encounter, small or great, elandslaagte and the rest included. awaking finally to this fact, but at least a year too late, we converted our cavalry into mounted riflemen. every possible function and every possible species of encounter which mounted men can conceivably undertake in any war was illustrated again and again. in reconnaissance, in raids, in protective work and independent work, in pursuit and retreat, in battle and out of battle, acting as divisions, brigades, regiments, squadrons, troops, patrols, or as single scouts, the cavalry were submitted to every sort of test during more than two and a half years. all our other mounted troops and all the boer troops were submitted to similar tests. out of it all emerges the single type of mounted rifleman, competent to do all duties alike, and incapable of doing any of them well unless he is as skilled in the rifle as he is on the horse—competent, too, if required, to perform functions never before dreamt of by any european cavalry—to make, hold, and storm entrenchments, and to take his place in the main line of battle.

here is a mass of evidence, vast, various, cogent. for the last time, i ask, was the war “peculiar”? of course it was peculiar. every war is peculiar. terrain differs, races differ, degrees of civilization and stamps of military organization differ, quarrels and aims differ, aptitudes and temperaments differ, and, lastly, with the progress of science, weapons differ. that brings us to the point—the 263only point relevant to our inquiry: were the peculiarities of the boer war such as to invalidate the conclusions developed in its course as to the armament and tactics of mounted troops?

even that way of putting the question is a little too wide. in great britain, at any rate, one big conclusion is admittedly valid for all future wars—namely, that the cavalry must carry a good rifle, not a bad carbine, and must be able to use it with far more freedom and skill than they ever dreamed of before the war. we have got that far, and stopped. shrinking from anything radical, taking refuge in compromise, we have fashioned in theory, and only in theory, an ideal hybrid, perfect both in shock and the rifle, and given him the formula for a hybrid “cavalry spirit,” which is quoted at the beginning of this volume. but—and this reservation is vital—we have taught him in “cavalry training” to rely mainly on shock and the “terror of cold steel,” which “nothing can replace.”

that settles the final form of our question: were the peculiarities of the war such as to justify the re-establishment of the lance and sword in their old position as the dominant weapons of cavalry? remember the proved penalties of error, if error there be—the extra weight and extra visibility of equipment, when every additional ounce of weight and every additional inch of vulnerable and visible surface tells, to say nothing of the complications, moral and physical, caused by allegiance to two diametrically opposite tactical ideals and tactical systems.

the answer we shall give to the question carries with it answers to many more. are we justified in reverting to exactly the same old view of “mounted infantry” as existed before the war, and which the war, regarded as an episode by itself, reduced to ridicule? was the war so abnormal that we are still in our handbook of “mounted infantry training” to lay down, foremost among the 264purposes for which that arm is to be employed, the purpose of “forming a pivot of man?uvre for cavalry, of supporting them generally with their fire, and ... of giving to the cavalry such infantry support, when they are acting at a distance from other troops, as will prevent the necessity of the cavalry regiments being employed in any other capacity than that of their purely cavalry r?le.”[66] prodigious indeed must be the abnormalities which would warrant the fresh enunciation of such a "general principle"! note the words “infantry support,” both in their context and in connection with the opening paragraph of the handbook, to the effect that “mounted infantry are infantry soldiers governed in their tactical employment by the principles of infantry training.” substitute the synonymous word “riflemen” for “infantry” in the three cases where the latter word is used, and there is, indeed, a substratum of very sound truth in the proposition. but it is truth which would be heresy to the authorities. for them, apparently, it was infantry who, under british leading, relieved mafeking, charged at bothaville and roodekraal, pursued at the biggarsberg and wildfontein, saved the guns at sannah’s post, and scouted, raided, and screened everywhere. it must have been infantry, moreover, disguised as cavalry, who held the colesberg lines, intercepted cronje on the modder, and ran to earth lotter; infantry, under boer leading, who captured a third of the main army’s transport at waterval, intervened brilliantly at the climax of the battle of paardeberg, ambuscaded and pursued at sannah’s post, raided cape colony, natal, and 265the railway communications, and charged at bakenlaagte and roodewal. was the war really so peculiar as to warrant such grotesque inferences as these? was a war which produced not a single example of true shock so peculiar as to justify the vague and unintelligible instructions to yeomanry—namely, that they are to be “so trained as to be capable of performing all the duties allotted to cavalry except those connected with shock action”? and what of our mounted forces overseas? suppose a war on colonial soil against a european army—to my mind a far more likely contingency than a war on european soil—suppose (merely for the sake of argument) such a war in south africa, where we should be aided both by dutch and british mounted troops. was the great war of 1899–1902 so peculiar as to warrant our telling the boer troops or the imperial light horse that they are not fit “to discharge cavalry duties”?

there is a big case, an authoritative case, an overwhelmingly convincing case, founded on a reasoned analysis of the campaign, to be made out here by the advocates of the arme blanche if they are to justify existing practice. when, where, and by whom has this authoritative case been presented? i am at a loss to say. directly we begin to grapple with this allegation of abnormality we find we are fighting with phantoms, with nebulous, elusive, and often mutually contradictory arguments held, some by one person, some by another. i scarcely know how far i need engage in this ghostly conflict. i have exhorted the reader from the first, in following my review of the war, to picture for himself parallel situations in a european war, distinguishing relevant from irrelevant peculiarities, and, without being led astray by mere names and labels, to test weapons and the tactical theories based on them by facts. i have endeavoured to assist him by analysis and comment, and i believe at one time or another i have dealt with every 266abnormality which is alleged to quash the verdict against the arme blanche. but i am not sanguine enough to hope that i have carried conviction, and i venture now to deal once more in a separate chapter with the allegation as a whole. in order to narrow the controversy within incontestably sound and fair limits, i will take the three powerful advocates of the arme blanche to whom i alluded in my first chapter, and from whom i have since frequently quoted—general sir john french, mr. goldman, and general von bernhardi. the last we may regard as the most powerful of all, since his book, “cavalry in future wars,” translated by mr. goldman, and furnished with an introduction by general french, is not only described by the latter officer as being the last word of logic and wisdom on all cavalry matters without exception, but has been largely drawn upon in practice by the compilers of our own “cavalry training.”

in general french’s long and warmly written introduction, levelled avowedly against the “misleading conclusions” of those who criticize shock, only one short passage is to be found in which the south african war—our own great war—is so much as alluded to, and then only to be dismissed with a shrug of the shoulders as almost irrelevant to the controversy. both the allusion and its context are, i am afraid, rather obscure, so i give the paragraph in full:

“in dwelling so persistently upon the necessity for cavalry being trained to the highest possible pitch to meet the enemy’s cavalry, i do not wish to be misunderstood. i agree absolutely with the author in the principle he lays down that the cavalry fight is only a means to an end, but it is the most important means, and i have thought it right to comment upon this because it is a principle which in this country, since the south african war, we have been very much inclined to overlook. to place a force of cavalry in the field in support of a great army which is deficient in the power to overcome the 267opposing cavalry, is to act like one who would despatch a squadron of war-vessels badly armed, badly trained, and ill-found, to blockade a distant coast-line defended by a powerful fleet. what is the naval fight in the open sea but a means to an end? it would be as sensible to dwell on the inutility and waste of a duel between hostile fleets as to lay down the principle that the ‘cavalry battle’ in no way affects the mutual situation of hostile armies” (p. 26).

sincerely desirous of understanding the general’s meaning, i confess that this passage baffles me, and i scarcely know what it would convey to a reader fresh from the study of our war. do the words which i have italicized imply that the non-cavalry portion of our “great army,” the infantry and artillery, were not worthy of the “support” of our cavalry, and denied that arm a chance of distinguishing itself in the “cavalry fight”—that is, presumably the shock fight? that cannot be the sense intended, for the imputation not only would never be made by general french, but is in itself indefensible. i need not argue that proposition again. if any narrative of the war does not disprove it to the most cursory reader, my previous narrative and comments would add no further conviction.

we must arrive at some other interpretation; and yet there seems to be no other that does not involve the writer in self-refutation. read literally, the sentence compares “a force of cavalry” (sent out under the circumstances described) to a squadron of war-vessels badly armed, badly trained, and ill-found, while the unequal naval fight with the “great fleet,” which results, is intended surely to be analogous to the “cavalry fight.” both are “means to an end”—in the one case to landing and invasion, in the other to the destruction of a hostile army. in the last sentence the simile becomes more precise, the “duel” between hostile fleets being expressly likened to the “cavalry battle,” and very 268aptly likened, if we do not assume with general french that the cavalry battle must inevitably be a shock battle. it is true that in the case of the south african war the simile is impaired by the fact that the “opposing cavalry,” constituting as they did the entire hostile force, cannot be regarded as the counterpart of our cavalry. but, disregarding that material point, where does the simile lead us? to the conclusion that our cavalry were badly armed, badly trained, and ill-found. that is admittedly true of armament and training; for the rifle has been permanently substituted for the carbine, and “thorough efficiency” in its use officially enjoined ever since, while the steel weapon, during the war, failed. “ill-found” might refer to horses. but the general, as the context shows, does not mean to take this dangerous line of argument. who, then, were the troops referred to? no part of the army was “ill-found” by comparison with the boers, who in most of the resources possessed by great and wealthy nations were miserably ill-found, and were reduced for the last year to destitution. “badly armed,” except in the case of the cavalry, is another misnomer. the infantry were armed with the best modern rifle, and although the artillery at first found their guns outranged, they soon received the aid of naval and other heavy guns, and always had an overwhelming numerical superiority over the boers. “badly trained” does, indeed, apply in a certain sense to the whole army, particularly to its practical organization for war. but it applies, too, to the boers, and in the latter respect far more pertinently.

i have no desire idly to split straws. if the passage i have quoted formed part of a reasoned argument for the abnormality of our great war, i agree that it would be unfair to make too much of a case of obscure exposition. but it stands alone, and i am justified in criticizing the attitude of mind which permits so high a cavalry authority 269as general french, in an essay part of which is explicitly directed against the advocates of mounted riflemen, to treat so vaguely and superficially the great national struggle which, for the time being, at any rate, did confirm their views. my justification is the greater in that such an attitude of mind is strictly typical of a great number of the adherents of the shock system. pressed, they are altogether unable to put into precise language their reasons for disregarding the boer war. in a later chapter, when dealing with the manchurian war, i shall have to refer to general french’s equally inadequate treatment of the theme of another case of abnormality.

in the meantime i can do no better than take two propositions from the paragraph quoted above, about which there can be no doubt. (1) a “cavalry battle” without shock is inconceivable to general french. there must be either shock or no battle, for surely no opponent of shock would go so far as to argue that, shock being a thing of the past, it was “inutility and waste” for opposing sets of mounted troops to fight with one another at all, in any way? we have here, in an unusually extreme form, that theory of the inevitable shock duel between opposing cavalries to which i alluded in my second chapter. it occurs again on page xxii of the same introduction.

“how, i ask, can the cavalry perform its r?le in war until the enemy’s cavalry is defeated and paralyzed? i challenge any cavalry officer, british or foreign, to deny the principle that cavalry, acting as such against its own arm, can never attain complete success unless it is proficient in shock tactics.”

here is the case complete, but, alas! strangely qualified by the words i have italicized. is there some arrière-pensée here? what if the hostile cavalry, like the boers, do not believe in shock? surely, the case thus 270stated begs the whole question at issue. observe that the underlying axiom is that the steel can always impose tactics on the firearm. contrast this axiom with the facts of the boer war, where the boers were the “opposing cavalry,” and were admittedly strong enough, though in what way we are not told, to throw into prominence the many defects of the great army sent to overcome them. and, by the way, we may remind the general that it did overcome them in the end, mainly through the improvisation of mounted riflemen (whom he ignores altogether), and through the assimilation of the cavalry to that type.

(2) the other clear deduction from the paragraph is this, that the boers were, on the whole, from whatever cause, a formidable enemy. they are compared to a powerful fleet, and we are represented, in whatever capacity, as suffering from certain weaknesses. that is the general colour of the argument, and i draw the reader’s attention to it, because the gist of mr. goldman’s argument is of a precisely opposite character; and this contradiction, in one form or another, runs through all the hazy generalizations that one hears expressed in public or private on the topic of abnormality.

to the best of my belief, mr. goldman is the only writer who has had the courage to set forth categorically, in the form of a reasoned argument designed expressly to prove the superiority of cavalry over mounted riflemen, the various grounds for regarding the south african war as abnormal. he does this in his appendix a. to “with french in south africa” (1903), and again in his preface to von bernhardi’s “cavalry in future wars.”

before examining these grounds it is essential to know what mr. goldman means by a “mounted rifleman.” here is his appreciation, on page 408 of the former book: “... the horseman armed only with a rifle. we may assume that he has received the special cavalry training 271aforesaid, and that in every way he is qualified to perform the duties of cavalry.” (i do not know what to make of this curious admission.) “but he is equipped solely to fight on foot. hence, no sooner does it become necessary for him to assume the offensive than he is forced to dismount, and from that moment his rate of progress depends solely upon the pace he can walk.”

truly, a poor creature! but we think of south africa and rub our eyes. was this the figure cut by mounted riflemen, boer and british, in south africa? it may be said without exaggeration that all the “offensive” mounted work was done by mounted riflemen or by cavalry acting as such. and think of the charges—bakenlaagte, for example. at what moment did botha’s men begin to “assume the offensive”? according to mr. goldman, when they “dismounted.” and when was that? within point-blank range of our guns, after a charge of a mile and a half.

to proceed with the quotation: “moreover, he has given hostages to fortune. his led horses being an easy prey to a handful of mounted horsemen, he cannot leave them far behind, for, should he lose them, his usefulness for reconnoitring purposes is gone; the opposing cavalry will merely push on and through the gap he has left in his screen.” we rub our eyes again. when did boer led horses fall a prey to cavalry, acting as “cavalry”? not in a single instance. as for the idea that the mounted rifleman is handicapped for “reconnoitring purposes”—after all the bitter losses and humiliations from which we suffered in south africa through imperfect reconnaissance, one can only regard the suggestion with respectful amazement. similarly with the suggestion about “pushing through gaps in screens.” this, as i have repeatedly pointed out, is what the cavalry could not do. their inability to do it was the predominant characteristic of all the fighting in which they were engaged—with 272one only apparent exception—the klip drift charge, where the screen was not a screen, but an isolated skirmishing line of 900 men and 2 guns, which was pierced without shock, and almost without bloodshed, by 5,000 horsemen, covered by the fire of 56 guns, and supported by a division of infantry.

to proceed: “it must be remembered that the mounted rifleman cannot fight on horseback. he has no weapon for that purpose, so that his only means of taking the offensive is to act on foot.... if in open country, the mounted rifleman cannot hope to meet the cavalryman mounted. in these circumstances he is practically unarmed; for the firmest believer in the rifle will scarcely maintain that the rifle-fire of mounted men is a serious quantity; anyone who has experienced it knows how perfectly ineffective it is.” well, i leave the reader to judge of the soundness of all this, in view of our experiences in south africa. it reads like a dream. is it, to say the least, an adequate treatment of the theme? surely it would be wiser to make some overt reference to the fine examples of aggressive mobility shown by our colonial irregulars, or to the boer charges, if only for the sake of proving their negligibility. this particular passage may have been written before mr. goldman (whose narrative of the war ends at komati poort) had had full opportunity to study final developments, but his book was published in 1903; he was cognizant when he wrote, at any rate, of sannah’s post, and in his preface to, and notes upon bernhardi (1906 and 1909), he maintains an equally icy silence upon the achievements of mounted riflemen in south africa, until a passage of warm praise from bernhardi himself extorts from him the footnote, inaccurate as to facts and mistaken in criticism, which i quoted in the last chapter (p. 254).

i need not pursue this quotation further. the writer eventually admits that in an “enclosed country” (what 273of the south african terrain?) the mounted rifleman has a certain value, but the most he will yield is that here the “mounted rifleman and the cavalryman are on an equality.” truly, an astonishing conclusion! surely part of this appendix must have been written before the war and left unrevised? even then the writer was old-fashioned, for the mounted infantry regulations of 1899, while warning that arm in a general way that they “needed the assistance of cavalry,” told them that when they cannot get this assistance, their “best security was to be keeping in broken, intersected, woody, or marshy ground, where they would have a great advantage over cavalry.” it is indisputable that men who spend their whole time in practising rifle-tactics, must be more efficient than men who spend half or more than half their time on shock-tactics. the strange thing is, that mr. goldman, in another connection, himself quotes the official warning with approval, as putting the mounted riflemen in their right place. yet, we may well ask, when in south africa did mounted riflemen ask for the assistance of cavalry—that is to say, of cavalry “as such,” to use general french’s expression? how often, on the other hand, did cavalry, as such, ask for the support of mounted riflemen, as such?

and these mounted riflemen of ours, who came in so many thousands from so many lands, to do such splendid and such absolutely indispensable work for the empire? not a single allusion to them either in this essay or in the preface to bernhardi. boers alone are used for illustration. anyone without knowledge of the war would infer that the whole of the mounted work on our side had been done by cavalry. nor is the conversion of the cavalry themselves into mounted riflemen mentioned.

one further question of definition before i proceed to the “peculiarities” of the war. what does mr. goldman mean by “shock”? he does not define it, nor does 274“cavalry training,” wisely enough, attempt a definition; but under the heading “shock action” (p. 410), he adduces as an example of shock the klip drift charge, where the cavalry files were eight yards apart, and the immediate objective of the charge was a sprinkling of extended skirmishers. i should weary the reader if i again exposed this fallacy at length. shock means impact. this charge was not shock, by any interpretation of that word, nor in the sense in which bernhardi or any european cavalry understands it. it was the right pattern of charge, but, as after experience proved, it was essentially the pattern of charge appropriate to mounted riflemen, and it was through blindness to this fundamental difference that the cavalry never made another like it.

now for mr. goldman’s “abnormalities.”

1. terrain.—to take this point first, as the least important. indeed, i scarcely know whether to take it seriously or not. it is rarely expressed elsewhere, and i think mr. goldman himself regards it as a desperate resource. after saying, broadly, that “certain physical and local conditions go far to explain why the cavalry were not more effective with the lance and sabre,”[67] he complains that the “boundless plains” were “seamed with ridges and watercourses,” while “the shock-tactics of cavalry require open ground free from large obstructions like rocky kopjes, thick bush, and strong fences” (i.e., wire fences, which, as he admits, were easily cut, and in time became no hindrance). but, while condemning, apparently, the whole of the transvaal, he cautiously admits that in the free state “the conditions were favourable.” was there ever a more remarkable example of under-statement? what does he expect? where is his ideal battle-ground of the future? taken as a whole, south africa, though its rolling plains were not quite so flat or so free from fences and dongas as the 275plains of northern manchuria, may be regarded as one of the most perfect man?uvring grounds for cavalry which the civilized world contains. of course, there were “obstructions” even in the most favourable areas, and, of course, these obstructions had a way of coming into prominence when fighting was afoot. battles are not fought on billiard-tables. one side or the other usually seeks defensible positions. and why should cavalry complain of irregularities? how effect surprise on a dead-level plain? it was by using irregularities that mounted riflemen won their most brilliant successes in south africa. shock is extinct, precisely because the ground which it imperatively demands makes cavalry most vulnerable to fire and least capable of surprise.

2. bad condition of horses and poor remounts.—i dealt with this point in chapters vi. and vii. the difficulties of the long voyage and acclimatization, and the imperfections of the remount system, are well known. a preventable cause of wastage, careless management and riding, is also scarcely disputed. on the debatable point of over-weight, mr. goldman, in a separate appendix, contends that the horses were needlessly over-loaded. all causes together do not explain away tactical facts covering two years. the more closely these facts are scrutinized—even those of actions like poplar grove, where the excuse has been most loudly raised—the less adequate the explanation. on inspection it always turns out that the enemy’s skill with the firearm, and our own deficiencies in that respect, are the principal cause of imperfect achievement. where the cavalry showed skill with the firearm there they obtained their tactical successes, irrespective of the condition of their horses. in the excellent colesberg operations no complaint was raised about the horses. when were sabres drawn? once, but without result, owing to delaying rifle-fire. in the arduous operations for the relief of kimberley, 276when the horses were at their worst, the cavalry achieved their most important success, by intercepting and containing cronje. on the strategical aspect of these operations, the use of lance and sabre, as combatant weapons, had no bearing whatever. men do not ride better or quicker for carrying steel weapons; on the contrary, the extra weight and the habits instilled by the shock theory are a hindrance to mobility. tactically, the cavalry succeeded or failed in proportion to their skilful or unskilful use of carbine and horse combined; succeeded signally at the drifts, where they held up cronje; failed signally in the pursuit north from kimberley. on the natal side, the cavalry horses were as fresh at talana, a case of failure, as at elandslaagte, the solitary case of a successful charge. as for poplar grove, which mr. goldman singles out for illustration, let me give his own words: “how could horses pursue a fleet and mobile enemy after a long day’s engagement, in which they had covered forty miles, and had turned the boers out of position after position?” how indeed? does mr. goldman seriously expect or demand that in our next war, after four months of hostilities we shall be provided with super-horses capable of the kind of feat suggested—that is, of beginning a galloping pursuit after fighting over forty miles of country? but this is a case where the reading of facts makes such a difference. in point of fact the conditions of pursuit began to be present after twelve miles. the full forty miles can only be arrived at (as i pointed out in my narrative) by counting the unnecessary détours and countermarches caused by failure to break down or ride past trivial flank and rear-guards. in these and many other later operations i have pointed out the intimate connection between horse-wastage and deficiency in direct aggressive power.

from the capture of bloemfontein to the end of the war the complaint about horses has less and less force. 277the remount system, of course, was greatly improved. although the difficulty of acclimatizing foreign animals was never properly overcome, owing to the ceaselessly voracious demands of the field-columns, horses poured into south africa from all quarters of the globe at an enormous rate, while no less than 158,000 native south african ponies (exclusive of large numbers captured on the veld) were supplied by the remount department. whatever the condition of the horses from time to time, the tactical incidents are of the same general character. nor, it need scarcely be added, was the disability confined to the cavalry. all our mounted troops were similarly affected, and the boers, in spite of their possession of the hardy native pony, must be regarded as being on the whole in a worse position. from first to last they were confined to their domestic supply, and, as i have pointed out, from paardeberg onwards they suffered considerably from shortage of horses.[68] their advantages were a light load and good horsemanship.

lastly, let me remind the reader of what i believe to be the real gist and essence of this complaint about horses. the theory of shock among several other rigorous conditions presupposes the presence at any and every moment of fresh horses capable of bearing down upon their objective at a gallop, and during the last fifty yards at the “charge,” and in perfect formation. this condition alone is enough to make shock a negligible factor in future wars—if, that is, cavalry are going to play the great part in war which they should play, but which they have not played for the last forty years. mounted riflemen 278are subject to no such conditions, and would lose half their value if they were. picture a boer charge—the little grass-fed ponies breaking from their “trippling” trot to what would correspond in european cavalry to a moderate canter.

3. lack of opportunity.—from ground and horses we pass to the more important part of mr. goldman’s case for “abnormality.”[69] though he never admits that cavalry work fell short in any respect in south africa, he is evidently conscious that this perfection needs much special proof, and he falls back on the proposition that they did not have a proper chance of distinguishing themselves in their own special line. two absolutely distinct causes—the one domestic, the other external—are represented as having produced this lack of opportunity:

(a) they were not employed properly—i.e., as the context shows, by lord roberts in particular, though he is not named.

(b) that the boers, owing to their habit of retiring without “fighting to a finish,” did not permit the cavalry to “discharge cavalry duties.”

i have alluded in previous chapters to both these points. let me add a word more.

as to (a), mr. goldman’s argument is vitiated from beginning to end by that old confusion between strategy and tactics, between mobility and combat, which lies at the root of arme blanche doctrine. the express point he is arguing, remember, is the relative value of cavalry and mounted riflemen, of the steel weapon and the firearm, or, to be more accurate, the steel weapon plus the firearm, and the firearm alone. now, the horse, whether 279used strategically or tactically, is common to both types. weapons are used only in combat. we are concerned, then, purely, with a question of weapons and of combat. strategy only concerns us in that the ultimate end of all strategy is combat. if there were to be no combat, equipment for a strategical errand would be vastly simplified. we should discard all weapons and all ammunition, and use the lightest men we could possibly find. in defending the steel weapon, therefore, and in showing that it had not its proper opportunities, we should expect to find mr. goldman dwelling on tactical opportunities. quite the reverse. his complaint—both in the appendix to his own book and in his preface to bernhardi—is that the cavalry were denied strategical opportunities. if he proved this up to the hilt, he would not have furthered the arme blanche theory one whit. but does he prove it? “strategical” is, of course, ambiguous, but let us follow his loose employment of the word in calling the kimberley raid “strategical.” he would not—and, indeed, does not—contend that at that period roberts denied the cavalry independent opportunities. he begins with the general advance of may, 1900.

but, again, we must pause to define the terms we are using. mr. goldman’s definition of the “strategical use of cavalry” is on page 412 of “with french in south africa”: “the use of that arm in such a fashion that, without of necessity engaging in any tactical action, certain well-defined effects are produced.” note the words i have italicized, for they prepare the way in advance for mr. goldman’s appreciation of the action of zand river, which he gives as a “concrete case to explain his argument.” “at that action french’s cavalry division was employed on the extreme left flank of the army to produce a purely tactical effect.... his operations could only, and did only, have the effect of causing the enemy gradually and in perfect order to withdraw from the 280position commanding the river.... the effect was purely tactical, for the early withdrawal of the enemy, unbeaten, undemoraiized, gave no chance to cavalry shock action.” what is the inner meaning of this contempt for “purely tactical effects”? simply this, that our cavalry, owing to their armament and methods, were outmatched in combat by the boers. let the reader examine once more the facts and maps of this action in the “official history,” the times history, mr. goldman’s own narrative, or any other. he will see that “could only” and “did only” are synonymous terms to mr. goldman. eight thousand boers held a twenty-five mile front, with their main strength in the centre and left, against nearly 40,000 british troops, of whom 13,000 were mounted. aiming at envelopment and destruction, roberts gave the cavalry a supremely important tactical object. of the two turning forces employed, two brigades of cavalry, supported by 2,200 mounted riflemen, were to make an extensive sweep round the boer right flank, and gain an intercepting position at ventersburg road station. the cavalry got well round to the rear in very good time (for the movement was a complete surprise to botha), but were subsequently checked by small flank-guards. one brigade was badly mauled, and the whole division was delayed long enough to enable botha to withdraw his whole force in safety. the lancer brigade, near the railway, and next in line to french’s division, though lightly opposed, showed no greater aggressive capacity (see “official history,” vol. iii., p. 56, and map), and the same applies to the remaining cavalry brigade on our extreme right. mr. goldman is content: there could not have been any “tactical effect.” the logical inference is that cavalry can have no tactical value at all.[70] for he does not 281suggest any tactical alternative for them. one tactical retort to these immense boer extensions was, as i indicated, a piercing movement; but mr. goldman makes no such suggestion, although in the same appendix (under “security and information”) he expressly gives as one of many normal cavalry functions that of “piercing the opposing cavalry screen with a division or divisions cut loose from the main army.” as i have pointed out, for purposes of analogy, the boer army on this, as on so many other occasions, did represent a cavalry screen.

and what is his suggested strategical alternative? this, that the cavalry division should have been, say, “100 miles to the north of the main army, moving south, while our main army moved north.” the effect on the “ill-disciplined boer troops” would have been “incalculable,” and then, in some unexplained way, would have come the “opportunity for the shock tactics of cavalry.” how wonderfully simple war seems to mr. goldman, and how carelessly he must have read his master, bernhardi, who makes short work of this conception of miraculously easy and effective raids in modern war! but let us look a little closer. the cavalry had arrived from bloemfontein at smalldeel, freshly remounted, on the 8th. on the 9th french’s two brigades covered twenty miles of their turning movement. on the 10th, the day of the battle, they covered upwards of thirty miles, and their horses were too tired for them to be able to act on the suggestion of roberts for an enveloping march that same night round the boer army and to the north of kroonstad. starting at 6.30 a.m. next day, they were too late to produce any important results.

the tasks set the cavalry, whether we call them strategical or tactical, were as heavy and responsible as 282the most ardent leader could desire. this craving for grandiose strategical “effects” without combat is thoroughly unhealthy and distorted. i venture to lay down the proposition that no cavalry has a right to complain of strategical mishandling until it has proved beyond question high combative capacity. with carbines and inadequate fire-training this high standard was beyond the reach of the cavalry. it has been said that roberts misused them in the middelburg operations of july 23–25, 1900. study the facts. french had planned a very extensive circuit. roberts, who had no spare mounted troops for his main columns, prescribed a shorter curve. on the 24th both cavalry brigades, even with the help of hutton’s mixed force of 3,000 men, were held up for four hours by a small rear-guard. casualties, two men wounded. it is impossible to assume that a wider circuit against so mobile an enemy would have produced important results.

genuine strategical independence for a purely cavalry force, on the lines of the great civil war raids, was never in question during this period, and would have been useless if feasible.[71] the nearest approach to such an expedition was the futile divisional march of 173 miles across the eastern transvaal in october, 1900, where some infantry and a few mounted riflemen, besides masses of ox transport, accompanied the column. there was no mobility worth the name; the column became nothing more than an escort to its own transport. the kimberley raid was not, of course, one of strategical independence. the division as a whole was never more than twenty miles from large portions of the main army, and was not rationed independently for a longer period than three days. kimberley was a friendly town, and after the return of the main army, on which the force was dependent for all but temporary supplies, forage ran out 283owing to de wet’s stroke at waterval. mounted riflemen were associated with cavalry, and the cavalry themselves won success by acting well as mounted riflemen.

mr. goldman’s idea that hundred-mile circuits would end in “opportunities for shock” is utterly chimerical. it is against all evidence, from this war and others, european, american, or asiatic, and bernhardi scouts it.[72] the ride from kimberley on february 15, selected by mr. goldman as a case where for once “cavalry” were used in a proper “strategical” manner, did not end in shock; on the contrary, it ended in tactical fire-action pure and simple. the chance for interception was of the same tactical character at zand river. in point of fact, at one moment during the latter action an attempt was actually made at an open-order cavalry charge, by a brigade against about 200 boers.[73] it came to nothing. and the reason, as given by mr. goldman and the official historian? horses too much blown. and yet mr. goldman cries out for hundred-mile expeditions which are to culminate—with fresh horses—in shock.

no one, of course, would go so far as to assert positively either of roberts or of any commander-in-chief in any war that he never once missed an opportunity for the strategical use for mounted troops. that is a different matter altogether. the issue lies between steel-armed troops and the mounted riflemen, whom mr. goldman ignores. why not a bare allusion to plumer’s brilliant defence of rhodesia, or to the relief of mafeking—a strategical march of 250 miles in fourteen days, with fire-fights en route, by irregulars?

with regard to general buller’s use of cavalry i need add nothing to my criticisms in chapters viii. and ix. 284his fault was to carry disbelief in the steel for the boer war to the extent of disbelieving in cavalry altogether for that war, a wholly unwarrantable point of view, derived from an equally distorted conception of the utility of cavalry.

(b) refusal of the boers to stand.—the facts speak for themselves. only by avoiding the whole topic of boer aggression, and by treating boer rear-guard skill as a non-cavalry quality which “made pursuit practically impossible,” is the point even arguable. indeed, i approach it again with the utmost reluctance; for mr. goldman’s idée fixe that the boers were from first to last mortally afraid of the lance and sword carries him to lengths where no upholder of mounted riflemen who respects and admires the cavalry and attacks only their weapons and methods can consent to follow him. i shall refrain from making controversial use of these passages, and shall confine myself, briefly, to less difficult ground.

mr. goldman is probably thinking mainly of the operations of lord roberts, though his proposition is general (p. 420). he would scarcely contend that the boers did not “stand” from november, 1899, to march, 1900, on the tugela heights, or that they did not show positively aggressive qualities and outmatch our cavalry at talana and the battle of ladysmith. with all his belief in the steel, he would scarcely in set terms allege that regular cavalry would have defended or attacked spion kop or pieter’s hill better than they were in fact defended and attacked. but these were tactical occasion, presumably with no “tactical effects” to be produced. what, then, of elandslaagte?

as for the main operations under lord roberts, has mr. goldman ever seriously reflected upon the relative numbers engaged? of course, the boers frequently showed moral weakness—we ourselves were not exempt—but they did not fear the sword. assuredly they 285“stood” at paardeberg to their ruin; but was there shock at paardeberg? assuredly they may be said to have stood at dornkop and at the two days’ battle of diamond hill, where cavalry were hotly engaged, and at bergendal, where seventy-four boer cavalry (though mr. goldman would never admit they were “cavalry”) delayed an army and were ejected by infantry. in the other actions of this period, as i have pointed out, their retreats were conducted in an orderly manner and with small loss.

let me lay down another proposition, which i believe all cavalrymen will agree to. no one on behalf of cavalry has a right to make a general complaint of pusillanimity or insufficient resistance on the part of the enemy, unless (a) that enemy has had something approaching numerical equality; or (b) has been forced into disastrous retreats, with loss of guns, transport, etc.; or unless (c) the infantry and mounted riflemen associated with the cavalry have not been seriously engaged. on this latter point the facts of the war and statistics of losses are decisive. there is something that makes the brain a little dizzy about the first two conditions, but the whole case for the arme blanche teems with paradoxes which can only be met by the method of reductio ad absurdum.

finally, i ask again, as i asked above, what is the real meaning of this complaint about lack of resistance? simply this, that the boers would not engage in shock and imposed fire-tactics on the cavalry. in his remarks on terrain (p. 423) mr. goldman reveals the truth. “favourable on the whole as the ground was in the free state, in the presence of cavalry operating on favourable ground the boers refused to give battle.” well, i can only ask the reader to study as one example among scores mr. goldman’s own example, zand river, noting (1) that we were nearly five times superior in total strength, and in 286guns, and that the regular cavalry, reckoned apart from mounted riflemen and infantry, amounted to five-eighths of the whole boer army; (2) that the terrain was as suitable for shock man?uvre as any cavalry could expect to obtain, and such as they very rarely would obtain in any probable european battle-field; (3) the tactical incidents of the cavalry turning movement, the offensive strokes by the boers, and the failure of our charge. how could the cavalry lose 224 horses and 161 men in casualties and prisoners and fail in their tactical task, unless someone “gave battle”? in other words, “battle” is synonymous with “shock.” nothing but shock counts.

time has convinced mr. goldman more and more strongly of this truth. in his preface to bernhardi he lectures the boers in a vein of compassionate condescension on their ignorance of the “art of war.” it is true enough that there was much in the art of war which the boers did not understand, or understood fatally late. but what does their mentor, for the purposes of his argument in this preface, mean by the “art of war”? he means shock, though he gives it the customary name of “mounted action.”

“had the boers understood the art of war and taken advantage of the openings which their superior mobility gave them, or had they been possessed of a body of cavalry capable of mounted action, say at magersfontein, they might repeatedly have wrought confusion in our ranks.”

this passage sets the crown upon the case for “peculiarity.” i leave it as it stands without further comment.

such are mr. goldman’s reasons for regarding his south african war as a vindication of the arme blanche. i have not discussed them at excessive length. they are extreme views, but such views, if honestly expounded, as mr. goldman expounds them, must be extreme. many people vaguely entertain similar ideas, but if they take 287the pains to work them out with facts and maps, they will either be forced to similar extremities or will abandon them altogether. in my next two chapters i shall give further proof of the astounding contradictions in which arme blanche doctrine abounds.

i come to the last of the triumvirate, general von bernhardi himself. it is a relief. we begin to breathe fresh air after an atmosphere which, i believe the reader will agree with me, becomes sometimes almost unbearably close and enervating. when censure of the commander-in-chief, depreciation of a brave enemy, implied depreciation of our own mounted riflemen, complaints about ground, complaints about horses, complaints about anything and everything but the one thing which really merited complaint, when apology and insinuation are carried so far, we begin to long for something stimulating and straightforward, and in bernhardi we get it. on his work i shall have to write more fully in the next chapter. at this moment i wish only to call attention to his view of the “peculiarity” of the boer war. it is contained in half a dozen lines on page 56 of “cavalry in future wars.” he has just been praising the boer charges as having achieved “brilliant results,” in spite of “any kind of tactical training for this particular purpose.” (what a curious sidelight that latter remark throws on official views of "training"!) he adds:

“certainly weapons and numbers have altered materially since the days of the american civil war, and the experiences of south africa, largely conditioned by the peculiar topographical conditions and the out-of-door habits and sporting instincts of the boers, cannot be transferred to european circumstances without important modifications.”

that is all he explicitly says about the boer war. but the reader will see at once that here is a totally different point of view from that of mr. goldman, whose thesis is 288that the boers were not formidable enough to be fit adversaries for our cavalry, that they would not “stand,” and that their great deficiency was lack of a steel weapon and shock power. the idea underlying bernhardi’s vague words is much more akin to that contained in the passage quoted from sir john french, and, of course, it is essentially the right idea. i pass by the “peculiar topographical conditions.” without further elaboration, we need not take the words to mean in set terms that south africa was less favourable for shock man?uvre than europe. the kernel lies in the “outdoor habits and sporting instincts,” creating conditions which “cannot be transferred to european circumstances without important modification.” these words, read in connection with the “brilliant results” of the boer charges, can only signify that town-bred europeans cannot hope to imitate methods, excellent in themselves, but demanding outdoor habits and sporting instincts.

this idea, expressed in one shadowy form or another, of an element of superiority in the boers, is very common; commoner, i think, on the whole than its antitype, the idea of inferiority, though i have more than once heard both propounded, unconsciously, in the same breath by the same person. but it is never in this country voiced authoritatively; and with good reason, for it shakes to its foundations the whole fabric of the shock system and opens up a line of thought which can end only in one way. mr. goldman does not even hint at it, except in connection with that strange faculty for fighting defensive rear-guard actions which he regards as quite outside the topic of cavalry. general french, while implying that the boers were formidable, is silent about the reason.

let us face this shadowy argument for what it is worth. what does it mean? that we cannot train our cavalry to become genuinely mobile mounted riflemen, with the 289rifle charge as their tactical climax instead of the shock charge, which is not a climax at all, but an isolated species of encounter in glaring conflict with real battle conditions? the contention, if it is really made, is absurd. if we cannot artificially create inbred instincts and habits so strong as those of the boers, we have the advantage of moral and tactical discipline, acquired only too late by the boers. we can work in the right direction on the magnificent material we have, and instead of imbuing the wrong spirit deliberately imbue the right spirit. we can teach our men to “fight up” to the charge and rely on one and the same weapon both for the process of “fighting up” and for the charge itself, when and where the actual mounted charge is necessary. the tactical form of the boer and british rifle charge—that is, in successive lines and with wide intervals—was precisely the same as our own open-order steel charge as practised at klip drift and at the present moment. the crucial difference lay in spirit and object; the spirit leading up to the charge was that of the rifle, and the object was that of overcoming the enemy with the rifle, not necessarily in a mêlée unless by way of pursuit, but at decisive range.

is saddle-fire an accomplishment our cavalry cannot acquire—an accomplishment which at this very moment we inculcate for “picked men and scouts” of the mounted infantry, a force with not a quarter of the mounted training that the cavalry receive? for professional troops it cannot be more difficult to acquire than skill with the steel weapon on horseback. that is an art which, as everybody knows, demands long and continuous drill and practice. indeed, it must demand longer drill and practice, because true shock—that is, heavy impact—involves close, knee-to-knee formations, rigid, mechanical, symmetrical, not only difficult to attain in themselves, but exceedingly difficult to combine with the free and effective use of steel weapons. obviously, neither saddle-fire nor 290the use of steel weapons can safely be enjoined in times of peace for volunteer troops like our yeomanry, for example, who obtain at the most a fortnight’s continuous field training in the year.

i ask the reader seriously to follow out the train of thought suggested by those pregnant words “outdoor habits and sporting instincts.” is it not common sense that these habits and instincts, fortified by drill and discipline, must be the very foundation of mounted success in war, and is not a system of tactics founded upon them likely to be a good system? should it not be the aim of a highly-civilized industrial people to aim at approximating as far as possible to such a system? or, taking as their starting-point indoor habits and urban instincts, are they to persist in working in the opposite direction? was it not the possession of these habits and instincts by such a large number of americans at the time of the civil war that led to the brilliant achievements of cavalry in that war, mainly through trained reliance on the firearm, imperfect weapon as it was? was not our own possession of sporting and hunting aptitudes, embodied in englishmen and colonials alike, our very salvation in south africa? of course it was. wherever these natural instincts were strong enough to burst the bonds of ancient tradition, there we obtained enterprising cavalry leaders. the same instincts called into being many good leaders among infantrymen, gunners, and sappers, and among ordinary civilians from every quarter of the empire.

do we not pride ourselves on this fact? is it not a commonplace in every englishman’s mouth that, hard and bitter as the struggle was, “no other nation”—and among other nations germany is often instanced—could have engaged in it so successfully as ourselves? there is sound truth in the boast. but it is the emptiest and silliest of boasts if we do not recognize the meaning behind 291it, which is nothing but this—that we have a greater proportion of men in our empire who possess those outdoor habits and sporting instincts which take shape in skilled mounted riflemen. and when we envisage a european war, are we to forget this boast and, ignoring not only our own priceless experience but our own innate capacities, revert to the antiquated european system?

if there are other arguments for “peculiarity,” i do not know them. but if i have carried my readers with me, they will agree that in this chapter and every other, in investigating and combating alleged peculiarities i have, in fact, been pursuing phantom arguments round the circumference of a vicious circle. disguise it as we may, the real peculiarity of the boer war was that the boer horsemen did not carry steel weapons. european cavalries do. let us turn to europe.