i.—german views.

we have now come to the exposition of the part cavalry will play in the great battle of all arms, which, says von bernhardi, is always "pre-arranged." but it will occur to the reader at once that, so far as our inquiry about fire and the steel in combat is concerned, there can be nothing new to be said. there are firearms in all warfare, and the tactical principles they enforce will be approximately constant. every great battle takes the form of a series of "attacks on localities," or "battles of encounter," however we interpret those phrases. if an enemy, to whatever arm belonging, who takes up a "defensive attitude" can only be attacked by fire in a fight of the independent cavalry, he can only be attacked by fire in a pre-arranged battle; and if the led horses are a paralyzing encumbrance in the one case, they are equally so in the other. the great battle, it is true, presents a[pg 134] more positive and obvious example of the co-operation of the various arms; but, as we have seen, the co-operation "of other arms" has been regarded by the author as a normal incident of the combats he has already described, and the "purely cavalry fight" as an altogether exceptional incident. and since even the purest cavalry carry the rifle, they can at any moment sully the purity of the said fight by resort to that sordid but formidable weapon.

the author, as we might expect, only dimly appreciates the universality of his own principles—if the mutually destructive propositions which he alternately lays down can be properly termed principles. he constantly confuses tactics with combat. different rules, of course, must always govern the action of mounted troops and horseless troops, because the one class is more mobile than the other; but it is impossible to lay down any lucid and intelligible principles for modern war until we realize the ubiquity and the supremacy of the missile weapon, rifle or gun.

the army cavalry, he tells us, as distinct from the divisional cavalry, "must be engaged en masse, and not in detail." "it must simultaneously engage its whole fighting strength," as an undivided entity (p. 190 et seq.), and its proper position is forward of one of the flanks.[pg 135] we have no sooner grasped this principle than we find a separate chapter devoted to the action of "those portions of the cavalry which find themselves behind the fighting-line, not on the exposed flank." this subdivision, we are vaguely told, "may be the result of circumstances," but there is no indication of what those circumstances are. but this is only one infraction of the principle of unity. in spite of the distractingly vague use of terms such as "front" and "flank," "enemy," "hostile forces," "troops within hostile reach," we are able to distinguish the following functions for the cavalry mass during the battle: it must conduct (1) a "far-reaching exploration" on the enemy's extreme rear and "probable lines of approach and communication," so as to give warning of the approach of fresh reserves; (2) an "immediate tactical reconnaissance," evidently of the whole battle-front—though the vague expression "against such hostile troops as may be within tactical reach" might mean almost anything. but we are told explicitly later that during the whole course of the battle the cavalry mass "must in all cases prevent the enemy's patrols from making observations as to the disposition of our own army, while, on the other hand, its own reconnaissance should never cease" (p. 199). we receive a[pg 136] sort of mental dislocation, therefore, when the author resumes: "screened by these various measures, the cavalry mass now advances fully deployed for the fight." were "these measures," then, only to screen the cavalry mass? but how can detachments, perhaps twenty miles away on the other flank, be said to screen the cavalry mass? (3) the mass is to provide for the occupation of "defiles and other important places to the flank and front of the main body" (i.e., of the main army).

let us pause and think. supposing the initial battle-front is thirty or forty miles in extent. even in the boer war it was frequently thirty miles, while in manchuria the fronts were sometimes enormously more extensive—at mukden nearly 100 miles. how in the world is the entire cavalry mass, posted outside one flank, to provide for the continuous reconnaissance, close and distant, of such a front, the occupation of advanced points, and for the maintenance of a reserve behind the front, while remaining a practically undivided force for united action? what is the enemy's cavalry supposed to be doing? in theory, we are told, they will do the right thing, that is, post themselves by instinct outside one flank exactly opposite our own mass. but supposing they do not. whatever they do, they[pg 137] have got (4) to be "driven from the field" (the reader will recollect the well-known formula), which will involve dispersion, if they disperse. but the author is not nearly so strong on the formula as sir john french. it is a very small matter (p. 191), this driving of the hostile cavalry from the field. "it has a certain value, but is comparatively useless for the main issue of the battle, unless, further, the possibility is gained of intervening in the decisive battle of all arms."

is not the reader conscious of an extraordinary artificiality and unreality in the terms employed? why speak of cavalry driving the hostile cavalry off the field, with more emphasis than of infantry doing the same to infantry? presumably, because cavalry, as we have already learnt, cannot break off the fight either in their pure or debased capacity. but on page 198 the beaten cavalry is to "seek shelter behind occupied points of support," where it is to be attacked by the greatest possible fire-power, words which seem to imply that hitherto the attack has been by shock. yet we have had it laid down as an axiom that neither party to a shock-combat can be used as a manageable unit for an indefinite time.

(5) the indivisible mass is now subject to fresh disintegration. "all portions of it not required for the pursuit" just described are to "regain their[pg 138] tactical cohesion" (an admission that the whole has lost its tactical cohesion), and, leaving their comrades to carry on the fire-fight, which may, of course, last for a week or more, are "to prepare for fresh effort." they are to occupy "localities" near the ground won, and "garrison" them with dismounted men—a direction we can scarcely take seriously when we recollect the crushing disabilities under which cavalry acting in passive defence have been supposed by the author to labour (see supra, pp. 122-123).

(6) what is left of the mass now "takes up a position of readiness" secure from the view and fire of the enemy, and disposed in what the author calls "groups of units." the expression seems to lack precision, but "this is the most suitable formation." subsequent action is to be according to the "circumstances of the various cases," and it is here that the reminder is casually interpolated that a protective and offensive reconnaissance along the whole battle-line is to be a continuous duty of the mass. but this action is "not to be regarded as sufficient." "the mass is to insure its own advance to that portion of the field where the decisive battle will probably take place, so that the charge will not meet with unexpected resistance and obstacles when the moment comes to ride it home. when this[pg 139] crisis of the battle approaches, the cavalry must be ready to intervene.... as the crisis approaches, endeavours must be made to get as close to the enemy as possible, in order to shorten the distance that will have to be covered in the charge." observe how naturally, how mechanically, the author associates the "crisis" with a gigantic cavalry charge, and with what simple trustfulness he believes that unexpected resistance and obstacles will melt away, if only the mass can insure its advance to the right spot in time.

as i shall show, he ruthlessly shatters his own hypothesis in the next breath; but consider, in the light of "real war," the utter futility of all this so-called instruction for the "pre-arranged battle," with its pre-arranged crisis. note the complete neglect of all the really important factors, the tremendous power of modern rifles and guns, and the vast extent and duration of modern battles, as contrasted with the limited physical powers of the horse and the small proportion which cavalry in all armies bears to other arms. take liao-yang, the sha-ho, mukden, battles which lasted ten days, two weeks, and three weeks, and try and find from the author's remarks any practical, tangible guidance for such situations. fancy one indivisible mass maintaining a con[pg 140]tinuous reconnaissance over such distances, occupying defiles and "localities" to the front, leaving a reserve behind the battle-front, driving the entire hostile cavalry from the field, and utterly destroying its power of further action; garrisoning points in the ground won, and at the same time advancing towards the "probable" point of crisis. but this point may be two days' march from the flank, where the mass—or what remains of it—was posted, and when it gets there it will certainly find that the crisis is centring round some strong, defensible position where lances and swords will be less useful than bows and arrows. no such picture as the author draws occurred in the franco-german, austro-prussian, or russo-turkish wars. it did not occur at vionville, the only battle in which a situation came about even approximately resembling the circumstances he outlines. so far as there was a crisis there, and so far as it was dealt with by a cavalry charge, the circumstances have radically altered, and there is a "total absence of analogy," as the author himself expressly states. bredow's steel-charge was made against unbroken infantry and artillery, flushed with the hope of victory. such charges, he has told us with truth, are utterly impossible in modern war. "i cannot conceive any real case in which cavalry[pg 141] can break through detachments of all arms" (p. 160). "nowadays, when infantry can cover the ground to a distance of 1,500 or even 2,000 yards with a hot and rapid fire, and offer in their wide extension no sort of objective for shock-action, an attack on unshaken, steadily-firing infantry, which has any sort of adequate field of fire, is quite out of the question" (p. 127).

it seems odd to have to recall these matters, for the author, as i said before, shatters his own hypothesis in the paragraphs immediately following his pages on the crisis and the charge. "however important and desirable it may be to contribute to the great decision by a glorious cavalry charge, it should be borne in mind that the possibility of this will occur in very rare cases." he goes on to insist emphatically on this point, saying nothing here about the vastly enhanced effect of the modern rifle, but basing his argument on terrain. great charges, he says, were almost impracticable in the franco-prussian, russo-turkish, and manchurian wars, and "possible european theatres of war are but little suitable for charges, owing to the extent to which they have been cultivated." peace operations are of no practical significance, because uncultivated country is expressly chosen. and so on.

[pg 142]

then, why, we ask, all this reasoned instruction about cavalry making its way to the crisis and delivering its charge? why not have said at the outset that their normal action must be something quite different? instruction for remote improbabilities is practically useless. what the commander wants to know is what to do as a general rule, especially when a wrong decision may, owing to the extent of the battle-field, involve him in ignominious impotence. such is cavalry literature. serious men in any other walk of life would not tolerate exposition of this sort.

we discover now that the cavalry are not, after all, to make their way to the crisis and charge. that was conventional rhetoric. in reality they are to act on the rear of the hostile army, "upon the reserves, the column of supply, the heavy artillery, etc." "it is here that opportunities for decisive action must be sought." well, obviously that is a different proposition altogether. why not have begun with it? habit—just the irresistible habit of associating cavalry with shock, and of calling shock their "proper r?le," although it is only their "exceptional" r?le. for, of course, such action as the author now indicates is purely a matter of fire. that is why no such decisive attack upon[pg 143] the rear of a great army has ever in recent times been accomplished by european cavalry. the cavalries of the sixties and seventies in the last century were absolutely incapable of such action, owing to their lack of fire-power. he is no doubt thinking of his model war, the american struggle of 1861-1865, and if he were truly candid, he would tell his countrymen that the brilliant exploits of the civil war leaders in raiding communications and "hostile reserves" were performed solely through the rifle.

the author is perfectly aware that the modern rifle has five times the power of the rifle of 1865, but he has not the courage of his own opinions, and descends to misty compromise. "such action must, of course, be conducted with a due co-operation between mounted and dismounted action." what is the use of a rule like that? "against intact hostile reserves the firearm will be principally used." why "principally"? will not these intact reserves, to say the least, "take up a defensive attitude," and therefore render a fire-attack, according to his own repeatedly formulated rule, absolutely indispensable? "against columns of waggons it will be well to commence by fire-action." why "commence" only? is there no lesson from south africa here? on what single occasion[pg 144] were lances and swords of the smallest value in attacks on transport? not on one. and on how many occasions did mounted riflemen, destitute of these weapons, capture transport and guns and rout reserves? we all know—sir john french knows—what our troops suffered in this way. why does he not warn his countrymen, instead of telling them that these german speculations are brilliant, logical, conclusive, complete?

look once more at the great manchurian battles. observe, for example, the great battle of mukden, (with its awful record of massacre by firearms), when a japanese cavalry brigade, acting with nogi's turning force, endeavoured to operate on the russian rear. it was miserably weak numerically, and it failed to accomplish anything "decisive"; but it did wonders, as it was, purely through fire. has any critic, however enamoured of the arme blanche, ever suggested that, however strong, it could have accomplished anything with the lance and sword? the very suggestion is preposterous. fire ruled that terrific struggle from first to last. look at mishchenko's pitiful cavalry raid on the japanese communications in january, 1905; and observe the shame which overtakes cavalry who cannot fight on foot: whole brigades paralyzed by squads of isolated riflemen, remind[pg 145]ing us only too painfully of dronfield and poplar grove; cossacks pathetically charging stone walls with drawn swords; disaster and humiliation clouding the whole sordid drama. sir john french's contribution to our enlightenment on the manchurian war, in his introduction to bernhardi's first book, "cavalry in future wars," was that the cossacks failed through excess of training as riflemen. he has not repeated that statement in his introduction to the second book. he scarcely could.

all the world knows the truth now—namely, that the cossacks, as one who rode with them said, "once dismounted, were lost." they did not know how to handle rifles, and all their humiliations may be traced to that fact. nor did the japanese cavalry at first, and they were equally impotent. but they learnt, and learnt to admirable purpose, as the records show. if he cannot repeat and confirm what he said in his first introduction, why is sir john french altogether silent on the point in his second introduction? well, it was an awkward dilemma for him; for bernhardi himself (p. 97), in his chapter on raids, alludes to mishchenko's raid in highly significant, though characteristically obscure, language. and if he follows up the clue, the reader may understand why it is that only on this[pg 146] one solitary question of raids, out of all the multitude of topics dealt with in the two books, sir john french "ventures to differ" from the german author, pronouncing, for his own part, against them. von bernhardi expressly founds his advocacy of the raid on the american civil war. "the idea," he says na?vely, "is taken" from that war. as though the boers who made the raids of 1901, of which he never seems to have heard, took their ideas from that war or any other! their ideas were the fruit of their own common sense. now, the civil war is particularly dangerous ground in england for advocates of the arme blanche, although it is safe enough ground in germany, where nobody studies it, and where there has been no henderson to immortalize the exploits of the great cavalry leaders. fire, and fire alone, rendered the american raids possible.

i need scarcely say that there is no incongruity in discussing together the raid proper and the attack on the reserves and communications of a great army from which my digression originated. the weapon factor is precisely the same in both. rifles are rifles and lances are lances, whatever the strategical or tactical scheme which bring them into play.

we turn lastly to the r?le of that portion of[pg 147] theoretically indivisible cavalry mass which is maintained as a "reserve behind the front" (p. 204). the author's method is the same: first, to expound at length the duties and powers of this body as though they were its normal duties and powers, and then to state that these normal duties and powers—in other words, the "proper r?le"—of the force concerned are, in nine cases out of ten, impracticable and visionary. he first represents the great mounted charge as the primary object, the great mounted charge, moreover, against infantry; for in this case there will be little chance, he says, of having "to deal with the hostile cavalry." he proceeds to lay down the truly delightful maxim that the force is to mass behind "that part of the fighting line where the ground is adapted for a charge of large masses," though he has taken great trouble to show in the previous chapter, quite correctly, that this is precisely the kind of ground upon which important struggles will not centre. then, in flat defiance of all he has said about charges against infantry, he advocates what in effect is our old discredited friend the "death ride" against unshaken and victorious infantry (p. 208), "in order to relieve our own exhausted infantry," etc. the cavalry are to "ride through the hostile infantry, and fall upon the artillery," although[pg 148] we know already that the author "can conceive no case in which cavalry can break through detachments of all arms," and that an enemy who takes up even a defensive attitude can only be attacked by dismounted action. but in a flash of recollection of a prior maxim, he enjoins that not only the preliminary deployment, but the formation for attack in widely extended order, must take place "beyond the effective range of the enemy's fire"; for "once outside this zone ... nothing else can be done but to gallop straight for the front." beyond the effective range of the enemy's fire! what is that range? he has told us before that it must, for average purposes, be reckoned 6,500 yards, or nearly four miles. conceive a charge of four miles, begun out of sight of the enemy, and in the blissful confidence that at the end of it the "ground will be suitable" for fighting on horseback with steel weapons! he proceeds in this strain for four pages, elaborating his topic with detailed tactical instructions, and then comes the usual nullifying paragraph:

"it must be clearly understood that in this case, as in the other where the cavalry is on the flank of the army, there will seldom be an opportunity for a charge." what, then, if not a charge? half a page of fervid generalization. "the first[pg 149] essential is that victory shall be won.... the cavalry must not shrink from employing its whole force on the fire-fight." we are bidden, rightly enough, to study the ancient lesson of fredericksburg. but it is now 1911. and we know what the author's views of the fire-fight for cavalry are—that, owing to the burden of led horses, it is never on any account to be attempted, unless there is an assurance of complete moral, tactical, and numerical superiority. cadit qu?stio once more. our reserve becomes a dummy.

there remain two topics in connection with the great pre-arranged battle of all arms—"pursuit and retreat" and the "r?le of the divisional cavalry." i shall take the latter first, and, with little comment, merely appeal to the reader's sense of humour. "in the battle of all arms," says the general, "as soon as fighting contact has been established with the enemy, and the close and combat reconnaissance is then probably at an end, the divisional cavalry must endeavour to gain touch with the army cavalry in order to strengthen the latter for the battle. in so doing it must not, of course, lose all connection with its own infantry division." remember that the army cavalry is, by hypothesis, well outside our flank of a battle area which may be of any extent from ten to seventy miles. picture the various[pg 150] divisional cavalries along this front endeavouring to "gain touch" with the army cavalry, while not losing connection with their own respective divisions.

it may be that this particular injunction has aroused merriment in germany. that is not our business. but that sir john french, with undisturbed gravity, should solemnly pass it on to englishmen as the last word of military wisdom—that is extraordinary. observe that, as usual, the arme blanche is responsible for the aberrations of the german writer. in the succeeding sentence this becomes clear. "when this cannot be done, and when no other chance of mounted action offers, the divisional cavalry must seize the rifle, and act as an immediate support for the infantry." the words i have italicized show that the physical feats contemplated in the original injunction are to be performed in the interests of shock, and that, if in the cold prosaic light of day they daunt the imagination of the leaders on the field, there is nothing left but to "seize the rifle."

"pursuit and retreat" is a chapter which almost defies any brief analysis. only those who are thoroughly acquainted with the curiously ambiguous vocabulary which hampers cavalry writers at every turn can fully appreciate the[pg 151] bankruptcy of the steel weapons as disclosed in these pages, and, at the same time, the disastrous effect of these useless bits of steel upon the reasoning faculties of those who still believe in them. the first few pages leave us only the impression that both pursuit and retreat are very dubious topics for cavalry. we approach the kernel of the matter at p. 215, where the writer deprecates "direct frontal pursuits," which "will generally yield but meagre results against the masses of the modern army and the firearm of the present day." the enemy will occupy "localities, woods, and the like," and "bring the cavalry pursuit to a standstill." "only when completely demoralized troops are retreating in the open, and cannot be reached by fire" (what this last clause means i cannot conceive), "will a charge be feasible." very good; but why not have followed the same principle in earlier chapters, instead of talking of cavalry charging infantry under cover, etc.? "frontal pursuit is essentially a matter for the infantry, who must press the retreating enemy to the utmost." this seems a fairly definite rule, but we have no sooner grasped it than it is cancelled.

"on the other hand, it is, of course, the duty of the cavalry to maintain touch with the enemy under all circumstances. with this object in[pg 152] view, it must continue the frontal pursuit, sometimes even without seeking to draw on a fight, by day and night." how one can continue a frontal pursuit by day and night without seeking to draw on a fight i leave the reader to guess. we turn to "retreat," which is, of course, the counterpart of pursuit, only to be involved in a fresh tangle. whether the enemy's cavalry is assumed to be conducting a frontal pursuit by day and night in spite of its "meagre results," or whether our own infantry are bearing the brunt of the retreat in the face of the frontal pursuit of the enemy's infantry—a pursuit which is "essentially" their business—we are left in uncertainty. all we have are vague heroics about the "maintenance of morale" (the writer seems to be very nervous about the morale of cavalry), about never renouncing a "relentless offensive," and about attacking the "enemy," wherever possible, with the cold steel. we find ourselves wondering how it is that "completely demoralized troops retreating in the open" (by hypothesis the only proper subjects for a steel-charge) can be, nevertheless, conducting a victorious pursuit, and our only escape from the entanglement is that in the case now considered by the author "enemy" means "cavalry," who are, apparently, so far inferior to infantry (though they carry the very[pg 153] weapon which makes infantry formidable) that they can be "relentlessly attacked," even when they are not completely demoralized.

one soon ceases to be surprised at anything in this species of literature, or one would gasp with amazement at the levity with which cavalrymen throw ridicule on their own arm. suddenly and very tardily we come upon an indication of the alternative to that frontal pursuit which gives such meagre results and yet must be continued day and night. "thus, when it becomes no longer possible to show a front to the pursuing cavalry in the open, measures must be taken to block the routes upon which his parallel pursuit is operating," etc. does not the reader feel his brain going when he reads a sentence like this? what antithesis can there be between cavalry "pursuing in the open" and cavalry conducting a "parallel pursuit"? there is no more or less probability of open ground in a parallel than in a frontal pursuit. it is the old story. one half of the writer's brain is back in the days of frederick the great; the other half is in working in the medium of the present.

that is the key to this chapter, from which a cavalry leader could not gain one concrete, definite rule for his guidance in real war. on pursuit, as on many other topics, the author was[pg 154] more clear and instructive in his earlier work, "cavalry in future wars" (chapter iv.), where he was not hampered by having to consider regulations with any pretence to modernity, and where he accordingly spoke with freedom on the absolute necessity of fire-action in pursuit; though he could not even then wholly grasp the corollary, the absolute necessity of fire-action in retreat.

ii.—the british view.

let us now, as in the case of the fight of the independent cavalry, contrast the directions given by our own authorities for the great battle of all arms ("cavalry training," pp. 225-229). one point of difference we may dispose of at once. the divisional cavalry (who are mounted infantry) and the "protective" cavalry (to which there is no german counterpart) behave rationally. they remain with, or drop back to, their respective main bodies, and there make themselves generally useful. the rules for the independent or army cavalry, on the other hand, present a curious study. on the german model, this main mass is, generally speaking, to be posted forward of one of the flanks. (there is no suggestion of a "reserve behind the front.") but we notice at[pg 155] once, with some surprise, that nothing is said about the corresponding hostile cavalry mass, which, according to von bernhardi, should be the primary objective, and whose "absolute and complete overthrow" is, according to sir john french (p. xiv), a "primary necessity."

the explanation is that one of the opposing cavalry masses is assumed to have been already absolutely and completely overthrown—that is, during the pre-battle reconnaissance phase, whose central incident, as described in pp. 192-194 and 200-212 of the manual, and criticized by me in the last chapter, is the great shock-duel of the two independent cavalries—a duel which is to result in the annihilation of one side or the other, and to which i shall have to return once more in the next chapter. the thread is resumed on p. 224 with the words, "once the independent cavalry has defeated its opponent," etc., and from that point onwards nothing is heard of the hostile independent cavalry. the explanation of sir john french's expression is the same. on p. xv he, too, assumes that before the battle the hostile cavalry has been disposed of, and says, somewhat vaguely, that the "true r?le of cavalry on the battle-field is to reconnoitre, to deceive, and finally to support"—functions which he distinctly suggests should be carried[pg 156] out mainly through fire-action by troops "accustomed to act in large bodies dismounted." and we seem to recognize this view in the functions outlined in the manual on p. 225. "reconnoitre," it is true, disappears. we find no echo of von bernhardi's chimerical conception of a double reconnaissance, distant and close, along the whole battle-front; nor, we may add, of his injunction to "occupy defiles and other important places to the flanks and front" of the army.

the r?les suggested for the flank cavalry mass are:

1. to "act against the enemy's flanks."

2. to combine fire concentrically with the main attack.

3. to pursue on parallel lines—a function which it is laid down on p. 229 is to be performed mainly with the rifle.

4. to force the enemy away from his direct line of retreat; which is merely a corollary of no. 3.

so far, good; but the arme blanche, as we might expect, is not going to be suppressed in this summary fashion, and when we pass from pious generalization to the actual "crisis," which "offers the greatest opportunities for cavalry action," we breathe once more the intoxicating atmosphere of the great shock-charge, not against cavalry[pg 157] now (for they are ex hypothesi extinct), but against infantry and artillery. there is a mild caution about the "modern bullet," but it is evidently not intended to be taken very seriously. the relation between the "flank" phase and functions and the "crisis" phase and functions is passed over in silence. von bernhardi's difficulty about deployment and advance under modern fire is surmounted by the simple direction that for what is called the "approach" surprise is essential; yet in the next breath "fire-swept zones" are envisaged which are to be passed over in a "series of rushes from shelter to shelter in the least vulnerable formation"—a process exclusive of surprise; and on the absolutely vital point of the formation for the actual attack one can positively watch the compilers struggling to reconcile cromwellian principles with modern facts, and embodying the result in studiously vague and misleading language. the front of the cavalry is not to be "too narrow," but the imperative necessity insisted on by von bernhardi of wide extension in the whole attacking force is implicitly denied by the direction that "squadrons in extended order may be used to divert the enemy's attention from the real attack." then, there is to be the stereotyped rally, which is to be in "mass," and the[pg 158] resulting mass is apparently to escape from further fire by using "another route."

when will our soldiers base their rules on war facts? as i have said, the facts show that it is still possible, in certain conditions, for men on horses, big target as they are, to penetrate a modern fire-zone, and attack and defeat riflemen and artillery; but it is impossible to do so if they insist on conforming their methods to the assumption that they are to do their killing work by remaining in the saddle and wielding steel weapons. that idea is fatal. it is that idea which promotes these rules about not too narrow fronts, these grotesque mounted rallies in mass, and this pregnant silence about the real point of interest—what actually happens when a line of horsemen, stirrup to stirrup, or in extended order, wielding lances and swords, impinges on an extended line of dismounted riflemen. we know from war experience that such a charge, stirrup to stirrup, is as extinct as the dodo, and is advocated in set terms by no rational being. it has not even been tried or contemplated since 1870. we know that the widely extended type has shared the fate of the other, because, with the loss of physical "shock," the steel weapons have lost their whole historical raison d'être. the only practicable mounted charge known to modern[pg 159] war is that of the mounted riflemen, who fight up to the charge, and use the only weapon which is effective against riflemen—namely, the rifle, fortified, if need be, by the bayonet. this charge is not an essential to victory. heaven knows we lost guns and men and transport enough in south africa without any mounted charging. the very object of a missile weapon is to overcome distance in a way that the lance and sword cannot overcome it. for all we know, even the mounted rifle charge may wholly disappear as science improves the firearm. but that improved firearm will itself rule combat, and banish into still remoter realms of memory the reign of the lance and sword.

i have excepted the case of the "utterly demoralized" enemy—utterly demoralized, of course, by fire. he is, naturally, fair game for any weapon, and experience proves that the firearm once more is incomparably the best weapon. lances and swords are, relatively, slow, cumbrous, and ineffective. a magazine pistol used even from horseback is a better weapon than either.

nothing is said by our authorities as to attack during the battle upon the enemy's reserves and transport, enterprises in which von bernhardi, after dismissing as a rare exception the great shock-charge, concludes that cavalry are to seek[pg 160] their decisive opportunities. we may assume that, like raids on communications, they are ruled out. but no alternative to the shock-charge at the crisis is suggested, for the parallel pursuit is, of course, a subsequent phase. there is only the ominous reservation that, if the ground is not favourable to the shock-charge, the "cavalry commander must look for his chance elsewhere, or wait for a more favourable opportunity" (p. 227).

that is just what we have to fear. that was the old, narrow, ignorant outlook of the continental cavalries, who were always waiting for favourable opportunities, and accounts for the idleness and lack of enterprise which von moltke stigmatized in 1866, and for the paltry character of their performances as a whole, which von bernhardi recognizes and condemns. it accounts for the miserable failure of the cossacks in manchuria, and explains the success of the japanese cavalry, once they realized the worthlessness of their german instruction and textbooks, and discovered for themselves the worth of the rifle as a stimulus to activity and mobility. von bernhardi says (p. 202): "the greatest imaginable error ... is to adopt a waiting attitude ... in order that the possibility of a great charge might not slip by unutilized." that error is[pg 161] precisely what we have to fear. teach cavalry that their lances and swords are their principal weapons, and that the rifle is a defensive weapon; tell them that the "climax of training" is the steel charge, "since upon it depends the final result of the battle"; found their "spirit" on the steel; make it in theory their "proper r?le"; give it a vocabulary of stirring epithets, like "glorious," "relentless," "remorseless," and all the rest, and they are only too likely, eager for battle as they are, to "wait for favourable opportunities" which will never occur, when they ought to be busy and active with their horses and rifles.

the sections on pursuit and retreat are modelled on similar sections in von bernhardi's earlier book, "cavalry in future wars," and escape therefore some of the contradictions of the later work. since they lay predominant stress on fire, we can only hope that their obvious blindness to the true reasons for fire does little harm. pursuits, whether by infantry or cavalry, be they frontal, parallel, or intercepting, will always be governed by fire. the thing that really distinguishes cavalry from infantry is that they have horses, which give them a vast scope for a class of intercepting tactics which infantry cannot undertake so easily. but even infantry will be better at any form of pursuit than a purely shock-[pg 162]trained cavalry. sir john french would have intercepted the boers, not only at paardeberg, but at poplar grove, karee siding, dewetsdorp, and zand river, if his cavalry had understood the rifle as well as they understood the horse. retreat is the counterpart of pursuit, and the same principles apply. cavalry ought to be able to fight a rearguard action better than infantry, because, thanks to their mobility, they can choose defensive points more freely, hold them longer, and fall back to others quicker. but if they are taught that it is beneath them to entrench and to defend a fire-position with stubborn tenacity, and that their proper r?le is to be performing frederician fantasias with the lance and sword, then they are likely, "in real war," to be relegated to a sphere "outside effective rifle-range," and to find their place usurped by infantry and mounted riflemen. there is very little to be known about rearguard actions which the boers have not taught us, and yet they were, in cavalry parlance, "defenceless"—in other words, steelless riflemen.