Chapter 7: the formation of troops for combat,1 and the separate or combined employment of the three arms.Two essential articles of the tactics of battles remain for us to examine: the one is the manner of disposing the troops in order to conduct them to combat, the other is the employment of the different arms. Although these objects belong to logistics and to secondary tactics, it must be owned meanwhile that they form one of the principal combinations of a general-in-chief when it is the question to deliver battle; hence it necessarily eaters into the plan that we have proposed to ourselves. Here doctrines become less fixed, and one falls back of compulsion into the field of systems: it is not therefore without astonishment that we have seen quite recently one of the most celebrated modern writers, pretend that tactics is fixed, but that strategy is not, whereas it is the contrary. Strategy is composed of invarible geographic lines, the relative importance of which is calculated upon the situation of the hostile forces, a situation  which can never lead but to a small number of variations, since the hostile forces are found divided or collected either upon the centre, or upon one of the extremities. Nothing is more possible than to subject elements so simple to rules derived from the fundamental principle of war, in spite of the efforts of fastidious writers to perplex the science in endeavoring to render it too abstract and exact. It is the same with the combinations of orders of battles, which can be subjected to maxims equally referable to the general principle. But the means of execution that is to say the tactics properly so called, depend upon so many circumstances, that it is impossible to give rules of conduct for the innumerable cases which may present themselves. To be assured of this, it is sufficient to read the works which succeed each other every day upon these portions of the military art without any being able to agree; and if we bring together two distinguished generals of cavalry or of infantry, it is very rare that they succeed in having a perfect understanding as to the most suitable method of executing an attack. Add to this the enormous difference which exists in respect to the talents of chiefs in their energy in the moral of the troops, and we shall be convinced that the tactics of execution will forever be reduced to contrary systems, and that it will be a great deal if one succeeds in laying down a few regulating maxims, which prevent the introduction of false doctrines into the systems that shall be adopted.
Article XLIII: the posting of troops in line of battle.After having defined in Article 31, what should be understood by the line of battle, it is proper to say in what manner they are formed, and how the different troops should be distributed in them. Before the French revolution, all the infantry, formed by regiments and brigades, were found united into a single battle corps, subdivided into first and second lines which had each their right and left wings. The cavalry was ordinarily placed on the two wings, and the artillery, yet very heavy at this epoch, was distributed upon the front of each line (they  dragged sixteen pounder guns, and there was no horse artillery). Then the army always encamping united, put itself in march by lines or by wings, and as there were two wings of cavalry and two of infantry, if they marched by wings they formed thus four columns. When they marched by lines, which was especially suitable in flank marches, then they formed but two columns, unless, through local circumstances, the cavalry or a part of the infantry had encamped in a third line, which was rare. This method simplified logistics, since the whole disposition consisted in saying: “You will march in such a direction, by lines or by wings, by the right or by the left.” They seldom deviated from this monotonous, but simple formation, and in the spirit of the system of war they followed it was the best they could do. The French determined at Minden, to try a different logistical disposition, by forming as many columns as brigades, and opening roads for conducting them abreast upon a given line, which they could never form.2 If the labors of the staff were facilitated by this mode of encamping and marching by lines, it must be owned that, applied to an army of a hundred or a hundred and fifty thousand men, this system would produce columns without end, and that routs would often occur like that of Rosbach.3 The French Revolution brought about the system of divisions, which broke the too great unity of the old formation, and gave fractions capable of moving on their own account upon all kinds of ground, which was a real benefit, although they fell perhaps from one extreme into another, by returning almost to the legionary organization of the Romans. Those divisions, composed ordinarily of infantry, artillery and cavalry, manoeuvred and fought separately; whether they were extended beyond measure for causing them to live without magazines, or whether they had the mania for prolonging their line, with the hope of outflanking that of the enemy, we often see seven or eight divisions of which an army is composed, march abreast upon as many routes at four or five leagues from each other; the Headquarters was placed at the centre without other reserve than five or six slender regiments of cavalry of three or four thousand horses; so that if the enemy chanced to unite the bulk of his forces upon one of those divisions and defeat it, the line was found pierced, and the general-in-chief, having no infantry reserve in hand, saw no other resource than to put himself in retreat to rally his scattered forces.  Bonaparte, in his first Italian war, remedied this inconvenience as much by the mobility and rapidity of his manoeuvres, as in uniting always the bulk of his divisions upon the point where the decisive blow was to be directed. When he was placed at the head of the State, and saw each day increase the sphere of his means and that of his projects, Napoleon comprehended that a stronger organization was necessary; he took then a mean term between the ancient and the new system, at the same time preserving the advantage of the division organization. He formed inithe campaign of 1800, corps of two or three divisions, which he placed under lieutenant generals for forming the wings, the centre or the reserve of the army.4 This system was definitively consolidated at the camp of Boulogne, where were organized permanent army corps, under marshals who commanded three divisions of infantry, one of light cavalry, and from thirty-six to forty pieces of artillery, with sappers. They were as many little armies proper to form, at need, any enterprise by themselves. The heavy cavalry was united into a strong reserve, composed of two divisions of cuirassiers, four of dragoons, and one of light cavalry. The grenadiers united and the guard formed a fine reserve of infantry; later, in 1812, the cavalry was organized into corps of three divisions, in order to give more unity to the ever increasing masses of this arm. It must be owned, this organization left little to be desired, and that grand army, which effected such great things, was soon the type upon which all Europe was modeled. Some military men, dreaming of the perfectibility of the art, would have desired that the infantry division, called sometimes to fight by itself, were increased from two brigades to three, because his number, gives a centre and two wings, which is of a manifest advantage, since without it the number two gives for centre an opening an interval, and that the fractions forming the wings, deprived of central support, could not operate separately with the same security. Besides that, the number three permits two brigades to be engaged, and have one in reserve which evidently augments the disposable forces for the decisive shock. But if thirty brigades, formed in ten divisions of three brigades each, are better than distributed into fifteen divisions of two brigades, it would be necessary,  to obtain this division organization, par excellence, to augment the infantry by a third, or to reduce the divisions of the corps d'armee to two instead of three, which would be a more real evil, since the corps d'armee being oftener called to fight alone than a division, it is to it especially that the number three is the most suitable.5 As for the rest, the best organization to give to an army entering the field, will be for a long time a logistical problem to resolve, because of the difficulty that is experienced in maintaining it in the midst of the events of the war, and the incessant detachments which they more or less necessitate. The grand army at Boulogne, which we have just cited, is the most evident proof of it. It seemed that its perfect organization should have secured it from every possible vicissitude. The centre under Marshal Soult, the right under Davoust, the left under Ney, the reserve under Lannes, presented a regular and formidable battle corps of thirteen divisions of infantry, without counting those of the guard and of the united grenadiers. Besides that, the corps of Bernadotte and Marmont, detached to the right, and that of Augereau detached to the left, were disposable for acting upon the flanks. But from the passage of the Danube at Donauwert, all was disordered; Ney, at first reinforced to five divisions, was reduced to two; the main body was dislocated, part to the right, a part to the left, so that this fine order of battle became useless. It will ever be difficult to give an organization at all stable; meanwhile events are not always as complicated as those of 1805, and the campaign of Moreau in 1800, proves that the primitive organization can, to a certain point, be maintained, at least for the bulk of the army. To this end it seems that the organization of the army into four fractions, viz: two wings, a centre, and a reserve, is the only rational one; the composition of those fractions may vary according to the strength of the army, but in order to be able to maintain it, it will be indispensable to have a certain number of divisions out of line, to furnish the necessary detachments. Those divisions whilst they are detached, could reinforce the one or the other of those fractions which should be the most exposed to receive or to strike great blows; or else they would be employed either upon the flanks of the main body, or to double the reserve. Each of the  four grand fractions of the main body may only form a single corps of three or four divisions, or else be divided into two corps of two divisions In this last case we should have seven corps, by counting but one for the reserve; but it would be necessary that the latter should always have three divisions, in order that the centre and the wings have each their reserve. In forming thus seven corps, if we had not always some out of line for detachments, it would often happen that the corps of the two extremities would be found detached, so that there would remain for each wing but two divisions, from which it would be necessary even at times to detach still a brigade to flank the march of the army, in such a manner that there would remain no more than three brigades, which does not constitute a very strong order of battle. These truths lead to the belief that an organization of the line of battle into four corps of three divisions of infantry and one of light cavalry, besides three or four divisions destined for detachments, would be less subject to variations than one of seven corps of two divisions. For the rest, as all depends in these kinds of arrangements, on the strength of the army and the units which compose it, as much as on the nature of its enterprises, there result many variations which it would take too much space to detail here, and I will confine myself to tracing on the accompanying plate, the principal combinations which a formation would present, according as the divisions should be of two or three brigades, and the corps of two or three divisions. I have traced there the formation for two corps of infantry upon two lines, either one behind the other, or one by the side of the other. The latter leads us to examine if it can ever be suitable to place thus two corps the one behind the other, as Napoleon has often done, especially at Wagram. I believe that with the exception of the reserves, this system could only be applied to a position of expectation, and by no means to an order of combat; for it is much preferable that each corps have in itself its second line and its reserve, than to accumulate several corps under different chiefs. However well disposed a general may be to sustain one of his colleagues, it will ever be repugnant to him to divide his forces to that effect, and when, instead of a colleague, he shall see in the commander of the first line but an envied rival, as happens only too often, it is probable that he will not furnish with haste the succors of which it might be in need. Besides that, a chief whose command is spread upon a long extent, is much less sure of his operations, than if he had only embraced half of this front, and when he would find in exchange in greater depth, the support which might be necessary to him.  Finally, in order to complete this sketch, it will be seen by the table hereafter,6 how much this question of the best formation is subordinate to the strength of the army, and how complicated it is. We can scarcely be regulated now a days, by the enormous masses put in action from 1812 to 1815, where we have seen one army form fourteen corps which had from two to five divisions. With such forces, it is incontestable that nothing can be imagined better than an organization by army corps of three divisions; eight of these corps would be destined for the line of battle, and there would remain six as well for detachments as for reinforcing such points of this line as should be judged suitable. But to apply this system to armies already very respectable of one hundred and fifty thousand men only, we can scarcely employ divisions of two brigades, where Napoleon and the Allies employed entire army corps. In effect, if we destine nine divisions to form the main body, that is to say, the two wings and the centre, and design six others for the reserve and the eventual detachments, there would be necessary fifteen divisions or thirty brigades, which would number one hundred and eighty battalions, if the regiments are of three battalions each. Now this supposes already a mass of a hundred and forty-five thousand foot, and an army of two hundred thousand combattants. With regiments of two battalions it would require, it is true, but a hundred and twenty battalions, or ninety-six thousand foot, but if the regiments have only two battalions, then the force of the latter ought to be increased to a thousand men, which would always give an hundred and twenty thousand foot, and an army of a hundred and sixty thousand men. These calculations alone prove how much the system of formation of inferior fractions influence that of the grand fractions.  If an army does not exceed a hundred thousand men, the formation in divisions, as in 1800, would be better perhaps than that by corps. After having sought the best mode for giving a somewhat stable organization to battle corps, it will not be out of place to examine whether this stability is desirable, and whether we do not better deceive the enemy by frequently changing the composition of corps and their position. I do not deny this last advantage, but it is possible to harmonize it with that which procures approximate stability in the order of battle. If we unite the divisions destined for detachments with the wings and the centre, that is to say, if we compose those fractions of four divisions, instead of three, and if at times we add one or two divisions to that one of the wings which should be the most probably destined to the principal shock, we shall have at the wings corps which will be nominally of four divisions, but which by detachments will ordinarily have but three, and at times might be reduced to two, whilst that the opposite wing, reinforced by a part of the reserve until the concurrence of five divisions, would present a sufficiently marked difference, in order that the enemy should never know exactly the real force of the fractions of the main body which he would have before him. There would be by this means more unity in the orders of movements of the staff, more facility for daily expeditions, and in the mean time not enough regularity to allow the enemy to know always precisely with whom he would have to do. I perceive, however, that I am engaged too far in an arena into which I ought not even to enter. It is for governments to decide those questions which merit a mature examination, and ought to make the object of an instruction for the staff; instruction, nevertheless, which should not impose absolute chains on the generalissimo, who ought always to have the power to regulate his forces according to his particular views, and the extent of the enterprises which he should form. Definitively whatever may be the force and the number of the subdivisions or fractions of the army, the organization by corps d'armee, will remain probaby a long time the normal type with all the great continental powers, and it is on this truth that the line of battle should be calculated. If the distribution of the troops in them is different from what it formerly was, the line of battle itself has also undergone some changes which result from the reserves, and the light cavalry attached to the various corps of infantry. Formerly it was composed ordinarily of two lines, now it is composed of two lines, with one or several reserves. But in latter times the European masses which encountered each other became, go considerable, that the corps d'armee, themselves formed upon two lines,  being found often placed the one behind the other, formed thus four lines; and the corps of reserve being formed also in the same manner, there resulted frequently, even six lines of infantry, and several of cavalry, a formation good perhaps for a preparatory position, but which is too deep for battle. However that may be, the classic formation, if this name can be given it, is still, for the infantry, that upon two lines; the more or less confined extent of the field of battle, and the forces of the armies could well give rise sometimes to a deeper formation, but this will always be an exception, or used for a last effort, for the order upon two lines besides the reserves, appearing to suffice for solidity, and giving more forces fighting at a time, seems also the most suitable. When the army possesses a permanent corps as an advanced guard, this corps could also be formed in advance of the line of battle, or withdrawn to the rear for augmenting the reserve,7 but as has already been said elsewhere, that rarely happens after the manner of the present formations, and the mode of combining the marches they require; each wing of the army has its own advanced guard, and that of the main body finds itself quite naturally furnished by the troops of the army corps which should march in front; when the army arrives in presence, those divisions reenter into their respective battle positions. Often even the reserves of cavalry are found almost entirely in the advanced guard, which does not prevent their taking the post assigned them, at the moment of delivering battle, either from the nature of the ground, or from the views of the general-in-chief. After what we have just explained, our readers will be assured that the methods followed since the revival of the art of war and the invention of gunpowder until the French Revolution, have undergone great changes through the present organization, and that in order to appreciate well the wars of Louis XIV, of Peter the Great, and of Frederick II, it is necessary to refer them to the system adopted in their time. However, a part of the ancient methods can still be employed, and if, for example, the position of the cavalry on the wings is no longer a fundamental rule, it might be good for an army of fifty or sixty thousand men, especially when the centre is found upon a ground less suitable to this arm than the one or the other of the extremities. It is generally the custom  to attach one or two brigades of light cavalry to each of the infantry corps; those in the centre place it in preference behind the line, those of the wings may place it upon their flanks. With regard to the reserves of this arm, if it be sufficiently strong for organizing three corps, to the end that the centre and each of the wings have its reserve, it would be an order as perfect as could be desired. In default of that, we could dispose this reserve in two columns, the one at the point where the centre is connected with the right, the other between the centre and the left; these columns could thus arrive with the same facility upon every point of the line which should be menaced.8 The artillery, now more movable, is indeed as formerly distributed upon the whole front, since each division has its own. Meanwhile it is well to observe that, its organization being perfected, we can better distribute it according to need, and it is ever a great fault to scatter it too much. There exists, for the rest, few positive rules upon this distribution of artillery, for who would dare to counsel, for example, to block up a gap in a line of battle, by placing a hundred pieces in a single battery, far from the whole line, as Napoleon did with so much success at Wagram? Not being able here to enter into all the details of this arm, we will limit ourselves to saying: 1. That horse artillery ought to be placed upon the ground where it can be moved in every direction. 2. That foot artillery, especially that of position, would be better posted, on the contrary, upon a point where it would he covered by ditches, or by hedges which would secure it against a sudden charge of cavalry. I need not say that, in order to preserve to it its greatest effect, we should be careful not to post it upon too elevated eminences, but rather upon flat grounds or slopes like a glacis; this is what every sous-lieutenant ought to know. 3. If the horse artillery be principally joined to the cavalry, it is well, however, that each army corps have its own, for gaining rapidly a point essential to occupy. Besides that, it is proper that there be some of it in the artillery reserve, in order to be able to direct it with more promptitude to the succor of a menaced point. General Benningsen had cause to congratulate himself at Eylau for having united fifty-eight pieces in reserve, for they contributed powerfully to re-establishing affairs between the centre and the left where his line chanced to be broken.  4. If one be on the defensive, it is proper to place a part of the batteries of heavy calibre upon the front, instead of holding them in reserve, since it is the object to batter the enemy at the greatest possible distance, in order to arrest the impulsion of his attack and to scatter confusion in his columns. 5. In the same condition it seems suitable, that apart from the reserve, the artillery be equally distributed upon the whole line, since one has an equal interest in repelling the enemy upon every point; this, meanwhile, is not rigorously true, for the nature of the ground, and the evident projects of the enemy, might necessitate the carrying of the bulk of the artillery upon a wing or upon the centre. 6. In the offensive it may be equally advantageous to concentrate a very strong artillery mass upon a point where we should wish to direct a decisive effort, to the end of making a breach in the hostile line, which would facilitate the grand attack upon which might depend the success of the battle. Having to treat here only of the distribution of the artillery, we shall speak farther on of its employment in combats.
Article XLIV: the formation and employment of infantry.The infantry is, without contradiction, the most important arm, since it forms the four-fifths of an army, and it is it which carries positions or defends them. But if it must be admitted that next to the talent of the general it is the first instrument of victory, it must be owned also that it finds a powerful support in the cavalry and artillery, and that without their assistance it would often find itself much exposed, and able only to gain half successes. We shall not evoke here the old disputes upon the shallow and the deep order, although the question, which was thought to be decided, is far from being exhausted, and placed in a point of view which permits the resolving it by examples and probabilities, at least. The war with Spain and the battle of Waterloo have renewed the controversies relative to the advantages  of fire, or the shallow order, over the impulsion of columns of attack, or the deep order; we shall express our opinion farther on. In the meantime we must not be misunderstood; it is no longer the question now to dispute whether Lloyd was right in wishing to give to the infantry a fourth rank armed with pikes, to the end of offering a greater shock in moving upon the enemy, or more resistance in receiving his attack; every experienced military man acknowledges in our day, that there is already sufficient difficulty in moving with order, battalions deployed in three closed ranks, and that a fourth rank would add to this embarrassment, without adding the least thing to their strength. It is astonishing that Lloyd, who had made war, should have insisted so much upon this material force; for the contact is very rarely sufficiently close, in order that this mechanical superiority be put to the test; and if these ranks turn their backs, it is not the fourth that will restrain them. This augmentation of a rank diminishes, in the defensive, the front and the fire, whilst that in the offensive, it is far from offering the mobility and the impulsion which are the advantages of columns of attack. We might affirm even that it will diminish that impulsion, for it is more difficult to cause eight hundred men to march in line of battle with four full ranks than in three, although there be a quarter less extent of front: the difficulty of the jointing of the two middle ranks, amply makes up for this slight difference. Lloyd has not been much more happy in the choice of the means which he proposes for diminishing the inconvenience of narrowing the front; it is so absurd that we cannot conceive how a man of genius could have imagined it. He would deploy twenty battalions, leaving between each of them a hundred and fifty yards, that is to say, an interval equal to their front; we may imagine what would become of those battalions all disunited and isolated at such a distance, leaving between them twenty gaps where cavalry could penetrate in strong columns, to take them in flank, and sweep them like dust before the wind. The question, we have said, no longer consists in discussing upon the augmentation of the number of ranks of a line, but merely to decide whether it ought to be composed of deployed battalions, acting only by their fire, or rather of columns of attack formed, each of a battalion ployed upon the two platoons of the centre, and acting only by their impulsion and their impetuosity. Several modern writers have treated these matters with sagacity, without any one of them succeeding to present any thing conclusive, because in tactics all is much more subjected to unexpected events, to sudden inspirations, to the moral, and to individualities. Guibert was the most  eloquent advocate of the shallow order and of fires, and a hundred victories of the late wars has given it a hundred denials. The Marquises of Chambray and Ternay have approached the same questions, and have given birth to doubts without resolving them. The course of tactics of the latter presents nevertheless, for orders of battle especially, valuable developments, not for prescribing absolute rules, but for familiarizing us with the different combinations which may result from them; this is all the advantage that can be promised from a tactical work.9 General Okounief, in his argued disquisition upon the three arms, has not shown less penetration, nor obtained less success. Perhaps he has not been sufficiently conclusive and has allowed yet some uncertainty to hover over the solution of the problem. Like his predecessors, he has not enquired whether the French columns, repulsed by the fire of the deployed English, were not masses much too deep, instead of being merely columns of a single battalion, like those of which we have just made mention, which would constitute a capital difference. I shall resume the points of view which the question presents. There exist, in fact, but five modes in forming troops for encountering the enemy. 1. As skirmishers; 2. Into deployed lines, either continuous or checker-wise, (en échiquier;） 3. In lines of battalions ployed upon the centre of each battalion; 4. In deep masses; 5. In small squares The skirmishers are an accessorary, for they ought only to cover the line properly so called by favor of the ground, to protect the march of the columns, to fill up the intervals, or defend the approaches of a post. These divers modes of formation are thus reduced to four systems: the shallow order, or the one deployed into three ranks: the half deep order, formed of a line of battalions in columns of attack upon the centre, or of squares by battalions; the mixed order where the regiments should be in part deployed, and partly in columns; finally, the deep order, composed of heavy columns of battalions, deployed the one behind the; other The order deployed upon two lines, with a reserve, was formerly generally used, it is especially useful in the defensive. Those deployed lines may be continuous, formed checker-wise, (en échiquier,) or in echelons.  The order by which each battalion of a line is found formed in column of attack by divisions upon the centre, is more concentrated; it is in some sort a line of small columns, (like the figure 5 of the opposite plate.) In the present regulation of three ranks, the battalion having four divisions,10 this column would present twelve ranks in depth, which gives perhaps too many non-combattants, and too much exposure of artillery. To diminish these inconveniences, it has been proposed, whenever it should be desired to employ infantry in columns of attack, to form it into two ranks, to place but three divisions of each battalion behind each other, and to deploy the fourth as skirmishers in the intervals of the battalions and upon the flanks, but to rally them behind the three divisions, if the enemy's cavalry chanced to charge. (See figure 6.) Each battalion would have by this means two hundred more shots, besides those which the increase, by a third, of the front would give by putting the third rank in the first two. Thus there would be in fact but a depth of six men, and we should obtain one hundred files front, and four hundred shots for each column of attack of a battalion. There would thus be strength and mobility united.11 A battalion of eight hundred men, formed, after the method in use, into column of four divisions, presents about sixty files to each division, and the first alone firing by two ranks, there would be but one hundred and twenty shots to furnish for each of the battalions thus placed in line, whereas, according to the mode proposed, there would be four hundred delivered. But whilst seeking the means of obtaining more fire at need, it is important to recollect also that the column of attack is not destined to fire, and that it ought to reserve this means for a desperate case; for, if it commences to fire in marching upon the enemy, its impulsion will become null, and the attack will fail. Besides that, this reduced order would be advantageous only against infantry, for the column of four sections of three ranks, forming a kind of solid square, is better against cavalry. The Arch-Duke Charles was fortunate at Essling, and especially at Wagram,  in having adopted this last order, which I proposed in my chapter upon the general principles of war published in 1807; the brave cavalry of Bessieres could do nothing against those little masses.12 In order to give more solidity to the column proposed, we could in truth call in the skirmishers and reform the fourth section; but there would always be but two ranks, which would present much less resistance against a charge principally upon the flanks. If for diminishing this inconvenience, we wish to form square, many military men think that in two ranks it would offer less consistency still than the column. Meanwhile the English squares were of only two ranks at Waterloo, and in spite of the heroic efforts of the French cavalry, there was but a single battalion broken. I have explained all the parts of the process; it remains for me to observe that if it were desired to adopt the formation in two ranks for the column of attack, it would be difficult to preserve that in three ranks for deployed lines, an army being scarcely able to have two modes of formation, or at least to employ them alternately on the field of battle. Hence what European army (if we except the English) could be risked to deploy in lines of two ranks? It would be necessary in this case never to move but in column of attack. I conclude from thence that thesystem employed by the Russians and the Prussians, that of forming the column of four divisions in three ranks, of which one could at need be employed as skirmishers, is that which is generally applicable to all situations, whilst that the other of which we have spoken is suitable only in certain cases, and would require a double mode of formation. Independently of the two orders above mentioned, there exists a mixed, which Napoleon employed at the Tagliamento, and the Russians at Eylau; their regiments of three battalions deployed one in first line, and formed the other two behind this one, upon the platoons of its extremities, (fig. 2, same plate.) This regulation, which belongs also to the semi-profound order, is suitable, in fact, for the offensive-defensive, because the troops deployed in first line resist a long time by a murderous fire, the effect of which always somewhat shakes the enemy; then the troops, formed in column,  can debouch through the intervals and throw themselves upon him with success. perhaps we could augment the advantage of this formation by placing the two battalions of the wings upon the same line as that of the centre, which would be deployed in such a manner that the first divisions of those battalions would be in line. There would thus be a half battalion more for each regiment in the first line, which for fire would not be inconsiderable; but it might be feared that those divisions putting themselves in condition for firing, the two battalions kept in column to be launched upon the enemy would be less easily disposable. However, there are many cases where such an order would be advantageous, it is sufficiently so for rendering it a duty to indicate it. The order in very deep masses is certainly the least suitable, (fig. 3.) We have seen in the late wars, divisions of twelve battalions deployed and compressed behind each other, forming thirty-six crowded and accumulated ranks. Such masses are exposed to the ravages of artillery, diminish mobility and impulsion, without adding any strength. This was one of the causes of the small success of the French at Waterloo. If the column of Macdonald succeeded better at Wagram, it paid dearly for it, and but for the success of the attacks of Davoust and of Oudinot upon the left of the Arch-Duke, it is not probable that it would have came out victorious from the position in which, for a moment, it saw itself placed. When it is decided to risk such a mass, it is necessary, at least, to take care to establish upon each flank a battalion marching by files, in order that if the enemy chanced to charge in force upon those flanks, it would not oblige the column to halt, (see fig. 3;) protected by those battalions which will face to the enemy, it will be able at least to continue its march to the object assigned it, otherwise this great mass, battered by converging fires to which it has no means of opposing even a proper impulsion, will be put in disorder like the column at Fontenoi, or broken as the Macedonian phalanx was by Paulus Aemilius. Squares are good in plains and against an enemy superior in cavalry. They were made formerly very large, but it is acknowledged that the square by regiment is the best for the defensive, and the square by battalion for the offensive. We can, according to circumstances, form them into perfect squares or into long squares, in order to present a greater front, and obtain more fire on the side from whence the enemy is expected to come, (see fig. 8 and 9.) A regiment of three battalions would easily form a long square by breaking the middle battalion and causing each half battalion to move, the one to the right, and the other to the left. In the wars with Turkey, squares were almost exclusively employed,  because hostilities took place in the vast plains of Bassarabia, of Moldavia and of Wallachia, and the Turks had an immense cavalry. But, if operations have place in the Balkan or beyond, and if their feudal cavalry give place to an arm organized in the European proportions, the importance of squares will diminish, and the Russian infantry will show all its superiority in Romelia. Be that as it may, the order in squares by regiments of battalions appears suitable to every kind of attack, whenever there is a superiority in cavalry, and we manoeuvre on even ground, favorable to the charges of the enemy. The long square, applied especially to a battalion of eight platoons, of which three should march abreast, and one upon each of the sides, would be better for moving to the attack than a deployed battalion; it would be less suitable than the column proposed farther back, but there would be less wavering and more impulsion than if it marched in a deployed line; it would have, moreover, the advantage for being in condition against cavalry. It would be difficult to affirm that each of those formations are always good, or always bad; but it will be admitted, at least, that it is an incontestable rule that, for the offensive, there is necessary a mode which should unite mobility, solidity, and impulsion, whilst for the defensive there is wanted solidity united to the greatest possible fire. This truth admitted, it will remain to decide whether the bravest offensive troops, formed in columns and deprived of fire, will hold out long against deployed troops having twenty thousand musket shots to send it, and able to deliver it two or three hundred thousand in five minutes. In the late wars, we have often seen Russian, French and Prussian columns, carry positions at the support arms, without firing a shot; it is the triumph of impulsion and of the moral effect which it produces, but against the murderous fire and the sang froid of the English infantry, columns have not had the same success at Talavera, at Busaco, at Fuente di Onor, at Albuera, and still less at Waterloo. Meanwhile, it would be imprudent to conclude from thence that this result should cause the balance to incline decidedly in favor of the shallow order and of fire; for, if the French were accumulated in all these affairs into masses too profound, as I have more than once seen with my own eyes, it is not astonishing that enormous columns, formed into deployed and wavering battalions, battered in front and flank by a murderous fire, and assailed on all sides, have experienced the fate which we have pointed out above. But would the same result have taken place with columns  of attack formed each of a single battalion ployed upon the centre according to rule? I do not think so, and in order to judge of the decided superiority of the shallow or firing order, over the half deep order, or that of offensive impulsion, it would be necessary to witness repeatedly what would happen to a deployed line which should be boldly attacked by an enemy thus formed, (fig. 6 of plate 2.) As for myself, I can affirm that, in all the actions in which I have been, I have seen these little columns succeed. Moreover, is it easy to adopt another order for marching to the attack of a position? Is it possible for this purpose to conduct an immense line in deployed order and firing? I believe that every one will pronounce for the negative: to throw twenty and thirty battalions in line, executing a fire by file or by platoon, with the object of crowning a position well defended, is to arrive there in disorder like a flock of sheep, or rather it is never to succeed. What ought we to conclude from all that we have just said? 1st, That if the deep order is dangerous, the semi-profound order is excellent for the offensive. 2d, That the column of attack by battalions is the best order for carrying positions, but that is is necessary to diminish as much as possible its depth, to give more fire at need, and to diminish the effect of the enemy's fire; it is proper, moreover, to cover it by many skirmishers, and to sustain it by cavalry. 3d, That the deployed order as first line, with the second in column, is that which is the best suited to the defensive. 4th, That the one and the other may triumph according to the talent a general shall have for employing seasonably his disposable forces, as we have said in treating of the initiative, in Article 16 and Article 31. In truth, since this chapter was written, the numerous inventions which have had place in the art of destroying men would he able to militate in favor of the deployed order, even for moving to the attack. However, it would be difficult to anticipate the lessons which it is necessary to look for from experience alone, for despite all that rocket batteries, the howitzers of Schrapnel or of Bourman, and even the guns of Perkins, could offer redoubtable; I own that I should have difficulty in conceiving a better system for leading infantry to the assault of a position, than that of the column by battalions. Perhaps it will even be necessary to give back to the infantry the casques and cuirasses that it wore in the fifteenth century, before throwing it upon the enemy in deployed lines. But if we return decidedly to this deployed system, it would be necessary, at least, in marching to the attack to find a more favorable means than that of long continuous lines, and to adopt either columns at distances for deploying on arriving at the enemy's position, or lines broken en échiquier, or  finally the march in battle by the flank of platoons, operations all more or less dangerous in front of an adversary who knows how to profit from them. Meanwhile, as we have said, a skillful general can, according to circumstances and localities, combine the employment of the two systems. If experience has proved to me long since that one of the most difficult problems of the tactics of war was the best mode of forming troops for going to combat, I have found out also that to resolve this great problem in an absolute manner, and by an exclusive system, is a thing impossible. In the first place, the nature of countries differ essentially. There are those where we can manoeuvre two hundred thousand men deployed, as in Champagne; there are others, like Italy, Switzerland, the valley of the Rhine, the half of Hungary, where we could scarcely deploy a division of ten battalions. The degree of instruction of the troops in all kinds of manoeuvres, their armament and their national character, could also have an influence upon formations. By favor of the great discipline of the Russian infantry, and of its instruction in manoeuvres of every species, it is possible that they may succeed in moving it in great lines with sufficient order and harmony for causing it to adopt a system which would, I think, be impracticable with the French or the Prussians at this day. My experience of this kind has taught me to believe everything possible, and I am not of the number of orthodox persons who admit but one same type and one same system for all armies, as for all countries. In order to approach the nearest possible to the solution of the problem, it seems to me then that we ought to seek-- (a) The best mode of moving in sight of the enemy, but still out of reach of his shot; (b) The best mode of advancing to the attack; (c) The best order of defensive battle. Whatever solution may be given to these questions, it appears to me suitable, in every case, to exercise the troops:-- 1. In the march in columns of battalions upon the centre, for deploying, if desired, within reach of the musket, or for advancing on the enemy, even with the columns, if it be necessary; 2. In the march in deployed and continuous lines, by eight or ten battalions at a time; 3. In the march en échiquier of battalions deployed, which> offer broken lines more easy to move than long continuous lines;  4. In the march in advance by the flanks of platoons; 5. In the march in advance by small squares, either in line or en échiquier; 6. In the changes of front, by means of these various methods of marching; 7. In the changes of front executed by columns of platoons at full distances, in order to reform without deployment; a means which is more expeditious than the other modes of changing front, and which is better adapted to all kinds of ground. Of all the modes of moving in advance, the march by flanks of platoons would be the easiest if it did not offer some danger; on level ground it answers marvellously, on rough ground it is the most convenient. It has the inconvenience of much fracturing the line; but by habituating the chiefs and the soldiers to it, by dressing well the guides of platoons, and the directing colors, all confusion could be avoided. The only objection which could be offered to it would be the fear of exposing the disjointed platoons to the danger of a rush of cavalry. I do not deny the danger, but it can be avoided either by being well watched by the cavalry, or by not employing this order too near the enemy, but only for crossing the first part of a great space which should separate the two armies. At the least sign of the approach of the enemy, the line could be reformed in a second, since there would only be necessary the time required for a platoon to place itself by file in line at the marching step. However, whatever precautions we may take, it must nevertheless be confessed that this manoeuvre could only be employed with troops well disciplined and well exercised, but never with militia or young soldiers. I have never seen it used before the enemy, but only in manoeuvres; and for the changes of front especially, it was employed with success. We could always try it in the great annual manoeuvres. I have also seen tried, marches in lines of battalions deployed en échiquier; these marches did very well, whilst those in full or continuous lines were always horribly bad. The French, especially, have never known how to march well in deployed lines. Perhaps those marches en échiquier would be found also dangerous in case of an unexpected charge of cavalry; we could, however, employ them for the first moment of the march, to the end of rendering it more easy, then the second échiquiers could enter in line with the first before assailing the enemy. Besides, by placing but a small distance between the échiquiers it would be always easy to form the line at the instant of a charge, for it must not be forgotten  that the échiquiers do not constitute two lines, but a single one, which has been divided in order to avoid the wavering and the disorder of a march in continuous line. The best formation for charging seriously the enemy is not less diffcult to point out. Of all the trials which I have seen made, that which appeared to me to succeed the best was the march of twenty-four battalions upon two lines of columns by battalions formed upon the centre for deploying; the first line went at the charge step upon the enemy's line, and arrived within twice the range of musketry, it deployed in the march. The company of voltigeurs of each battalion was deployed as skirmishers, the others were formed, then commenced a sustained fire by file; the second line of columns followed the first, and the battalions which composed it threw themselves at the charge step through the intervals of the companies which were firing. This was done, in truth, without an enemy, and it seemed that nothing could have resisted this double effect of the fire and of the column. Besides those lines of columns, there are yet three other means of moving to the attack in semi-deep order. The first is that of lines mingled with deployed battalions and battalions in column upon the wings of those deployed, of which we have spoken at page 297. The deployed battalions and the first divisions of those in column would fire at half musket range, and afterwards throw themselves upon the enemy. The second is to advance with the deployed line, and firing, to within half musket range, then to throw the columns of the second line through the intervals of the first. The third is the echelon order, mentioned on page 213, and in figure 11 of plate 1. Finally, the last mode is to advance entirely in deployed order, by the sole ascendant of the fire until one of the two parties retreat, which appears almost impracticable. I cannot affirm which of those modes would be the most suitable, for I have seen nothing of the like in the field. In fact, in war, I have never seen anything in the combats of infantry, but battalions deployed beforehand, which commence firing by platoon, then engaging by degrees a fire by file; or else by columns marching fiercely upon the enemy, which fled without awaiting the shot, or which repulsed those columns before the actual meeting, either by its firm continuance or by its fire; or, finally,  by taking, itself, the offensive by advancing to the rencounter.13 It is scarcely but in villages and defiles that I have seen real melees of infantry in column, the heads of which encountered with the bayonet; in battle position I have never seen the like. However it may be with regard to these controversies, we could not too often repeat, it would be absurd to reject the fire of musketry, as well as to renounce semi-profound columns, and the imposing an absolute system of tactics for all countries and against all nations indiscriminately, would be to ruin an army. It is less the mode of formation than the well combined employment of the different arms which will give the victory; I except from it, nevertheless, columns too deep, which should be proscribed by all theories. We will terminate this dissertation by recalling, that one of the most essential points for conducting infantry to the combat, is to secure our troops from the fire of the enemy's artillery as much as possible; not in withdrawing them unseasonably, but by profiting by the inequalities of the ground, or other accidents which are found before them, in order to shelter them from the batteries. When we have arrived under the fire of musketry, then shelters are not to be calculated upon; if we be in condition to assail, we must do so; shelters are suitable only, in this case, for skirmishers and for defensive troops. It is sufficiently important, generally, to defend villages which are upon the front, or to seek to carry them if we be the assailant; but it is equally necessary not to attach an undue importance thereto, forgetting the famous battle of Hochstaedt: Marlborough and Eugene seeing the bulk of the French infantry buried in the villages, forced the centre and took twenty-four battalions, sacrificed to guard those posts. For the same reason it is useful to occupy clumps of trees or copses, which may give a support to that one of the two parties which is the master of them. They shelter the troops, conceal their movements, protect those of the cavalry, and hinder that of the enemy from acting in their proximity. The skeptic Clausewitz was not afraid to sustain the contrary maxim, and under the singular pretext that he who occupies a wood acts blindly, and discovers nothing of what the enemy is doing, he presents their defense as a fault of tactics. Blinded himself, probably, by the results of  the battle of Hohenlinden, the author is too prone to confound here the occupation of a wood in the line of battle with the fault of throwing a whole army in a vast forest without being master of the issues, either of the front or of the flanks; but he must never have seen a combat who denies the incontestable importance of the possession of a wood situated in proximity with a line that he wishes to defend or attack. The part which the park Hougeumont played in the battle of Waterloo is a great example of the influence that a post well chosen and well defended can have in a combat; in advancing his paradox, M. Clausewitz had forgotten the importance which woods had in the battles of Hochkirch and of Kollin. But we have already dwelt too long upon this chapter of the infantry, it is time to speak of other arms.
Article XLV: the cavalry.The formation of the cavalry, subjected to nearly the same controversies as that of the infantry, has been subjected also to the same uncertainty, and the too much vaunted treatise of the Count de Bismark, has not done much to clear them up. As we have been scarcely better settled upon its employment, I shall be permitted to submit what I think of it to the decision of generals habituated to conducting it. The employment which a general should make of cavalry, naturally depends a little on the relative strength of that of the enemy, either in number or in quality. Nevertheless, whatever modification those variations may induce, a cavalry inferior, but well conducted, may always find occasions to do great things, so decisive is the proper moment in the employment of this arm. The numerical proportion of the cavalry to the infantry has much varied. It depends upon the natural disposition of nations, whose inhabitants are more or less fit to make good horsemen; the abundance and the quality of the horses also exercise a certain influence. In the wars of the revolution, the French cavalry, though disorganized, and very inferior  to that of the Austrians, served marvellously. I saw, in 1796, in the army of the Rhine, what they pompously called the reserve of cavalry, and which formed scarcely a feeble brigade, (fifteen hundred horses.) Ten years afterwards I saw those same reserves fifteen or twenty thousand horses strong, so much had ideas and means changed. As a general thing, we may admit that an army in the field ought to have a sixth of its force in cavalry; in mountainous countries, a tenth is sufficient. The principal merit of the cavalry lies in its rapidity and its mobility; we might add even in its impetuosity, if it were not feared to see a false application made of the last quality. However important it may be in the ensemble of the operations of a war, the cavalry could not defend a position without the assistance of infantry. Its principal object is to prepare or to finish the victory, to render it complete by taking prisoners and trophies, by pursuing the enemy, by rapidly carrying succor to a menaced point, by breaking the shaken infantry, finally by covering the retreats of the infantry and the artillery. This is why an army, wanting in cavalry, rarely obtains great successes, and why its retreats are so difficult. The mode and the moment most suitable for engaging the cavalry, belongs to the coup d'oeil of the chief, to the plan of battle, to what the enemy is doing; and to a thousand combinations too numerous to mention here. We shall indicate then their principal features. It is acknowledged that a general attack of cavalry against a line in good order, could not be attempted with success unless sustained by infantry and much artillery, at least at a certain distance. It was seen at Waterloo how much it cost the French cavalry for having acted against this rule, and the cavalry of Frederick experienced the same fate at Kunersdorf. We may, nevertheless, find ourselves called upon to engage the cavalry alone; but, in general, a charge upon a line of infantry which should already be found engaged with the adverse infantry, is that from which we could expect the most advantages; the battles of Marengo, of Eylau, of Borodino, and ten others, have proved this. Meanwhile there is a case in which the cavalry has a decided superiority over infantry; it is when there falls a beating rain or snow, which wets the arms and deprives the infantry of its fire; the corps of Augereau had a cruel proof of it at Eylau, and the left of the Austrians experienced the same fate at Dresden. Great charges are also executed with success against infantry, when  we should have already succeeded in shaking it by a fearful fire of artillery, or in any other manner. One of the most remarkable charges of this kind was that of the Prussian cavalry at Hohenfriedberg, in 1745, (see Treatise of Operations.) But every charge against squares of good infantry not broken, must fail. Great charges are made for carrying the batteries of the enemy, and facilitating for the masses of infantry the means of crowning his position; then it is necessary that the infantry be in condition to sustain them without delay, for a charge of this nature has but an instantaneous effect, of which it is necessary to profit briskly before the enemy drive back your cavalry disunited. The fine charge of the French upon Gosa, at the battle of Leipsic, 16th of October, is a great example of this kind. Those which they executed at Waterloo with the same object, were admirable, but without results, for want of support. In the same manner the audacious charge of the feeble cavalry of Ney upon the artillery of the Prince Hohenloe, at the battle of Jena, is an example of what may be done in such a case. Finally, general charges are made against the enemy's cavalry for driving it from the field of battle and returning afterwards against his battalions with more liberty. The cavalry could be launched with success for taking the hostile line in flank or in reverse, at the moment of a serious attack, which the infantry should execute in front. If it be repulsed, it can return at a gallop, and be rallied upon the army; if it succeed, it may cause the ruin of the hostile army. It is rare that it is given this destination, and I do not see, nevertheless, what obstacle there could be to it, for cavalry well conducted could not be cut off, even when it should find itself in rear of the enemy. For the rest, this is the part which belongs especially to irregular cavalry. In the defensive, the cavalry can equally obtain immense results, by engaging at the proper moment a hostile body of troops, which, having approached the line should be ready to penetrate it, or which should already have pierced it; it could in this case re-establish affairs, and cause the destruction of an adversary shaken and disunited even by its first successes; a fine charge of the Russians at Eylau, and the English cavalry at Waterloo proved this. Finally, the especial cavalry of the army corps make timely charges, either for favoring an attack, or for profiting from a false movement of the enemy, or in order to finish his defeat in a retrograde movement. It is not so easy to determine the best mode of attack, it depends upon the object that is proposed, and other circumstances which have an influence upon the choice of the moment. There are four modes of charging  in columns at distance, in lines at the trot,14 in lines at the gallop, finally at a helter-skelter, (a la debandade:) all may be employed with success. In the charge en muraille or in line, the lance offers incontestable advantages; in melees, the sabre is better, perhaps: hence comes the idea of giving the lance to the first rank which is to break, and the sabre to the second, which is to finish by partial struggles. The firing with the pistol is suited only to advanced posts in a charge as foragers, or when the light cavalry wishes to harrass the infantry and draw its fire, in order to favor a more serious charge. As for carbine firing, we scarcely know what it is good for, since it requires the whole troop to halt in order to fire deliberately, which will expose it to a certain defeat if it be attacked boldly. It is skirmishers only who are able to fire running. We have just said that all the modes of charging could be equally good. Meanwhile it is necessary to guard against believing that impetuosity is always decisive in a shock of cavalry against cavalry. The fast trot on the contrary appears to me the best gait for charges in line, because here everything depends upon harmony, steadiness and order, conditions which we do not find in charges at a gallop. Those are suitable especially in charges against artillery, because it is more important to arrive quickly than to arrive in order. For the same reason, with cavalry armed with sabres, we may throw ourselves at a gallop at two hundred paces against a hostile line which awaits us steadily. But if we have a cavalry armed with lances, the fast trot is the true gait, for the advantage of this arm depends above all upon the preservation of order; as soon as there is a melee, the lance loses all its value. When the enemy comes upon you at a fast trot, it does not seem prudent to run upon him at a gallop, for you arrive all disunited against a compact and close mass, which will pass through your disjointed squadrons. There would only be the moral effect produced by the apparent audacity of your charge which would be favorable to you; but if the enemy appreciate it at its just value you will be lost, for in the physical and natural order, success ought to be for the compact mass against horsemen galloping without harmony. In charges against infantry, the Mamelukes and the Turks have sufficiently proved the importance of impetuosity; where the lancers or the  cuirassiers at the trot will not penetrate, no cavalry will pierce. It is only against infantry much shaken, or whose fire could not be kept up, that the impetuous charge can have any advantange over the trot.15 In order to force good squares, cannon and lancers are necessary, better still cuirassiers armed with lances. For charges as foragers or helter-skelter, so frequent in the daily recounters, it is necessary to imitate the Turks or the Cossacks: these are the best examples that can be taken: we shall return to this subject. Whatever system is employed for going to the shock, a recognized truth for all possible charges is, that one of the best means of succeeding is to know how to throw at the proper time some squadrons on the flanks of the enemy's line which is to be assailed in front. But in order that this manoeuvre should obtain a full success, in charges of cavalry against cavalry especially, it is necessary that it be executed only at the instant when the lines come to be engaged, for a minute too soon or too late the effect would probably be nothing: thus it is that the greatest merit of an officer of cavalry consists in this exact and rapid coup d'oeil. The armament and the organization of cavalry have been the subject of many controversies, which it would be easy to reduce to a few truths. The lance is, as has just been said the offensive arm for a troop of horsemen charging in line, for it attains an enemy that could not approach them; but it may be well to have a second rank or a reserve armed with sabres, more easy to handle when in a meee, and when the ranks cease to be united. Perhaps it would even be better still to cause a charge of lancers to be sustained by an echelon of hussars, who penetrating the hostile line after them, would better finish the victory. The cuirass is the defensive arm par excellence. The lance, and the cuirass of strong leather doubled, or a buffalo hide, seems to me the best armament for the light cavalry; the sabre and the iron cuirass for that the heavy cavalry. Some experienced military men incline even to arming the cuirassiers with lances, persuaded that such a cavalry, very similar to  the ancient men at arms, would overturn all before it. It is certain that a lance would suit them better than the musketoon, and I do not see what should prevent giving them weapons similar to those of the light cavalry. With regard to the amphibious troop, dragoons, opinion will ever be divided; it is certain that it would be useful to have some battalions of mounted infantry, which could anticipate the enemy at a defile, to defend it in retreat, or to scour a wood; but to make cavalry of infantry, or a soldier who would be equally proper for either arm, appears a difficult thing: the fate of French foot dragoons would seem to have sufficiently proved it, if on the other side the Turkish cavalry had not fought with the same success on foot as on horseback. It has been said that the greatest inconvenience of dragoons arises from the circumstance that you are obliged to preach to them in the morning that a square cannot resist their charges, and to teach them in the evening that a footman armed with a gun ought to overcome all possible horsemen: this argument is more specious than true, for instead of preaching to them maxims so contradictory, it would be more natural to tell them, that if brave horsemen can break a square, brave infantry can also repulse that charge; that the victory does not depend always upon the superiority of the arm, but rather upon a thousand circumstances; that the courage of the troops, the presence of mind of the chiefs, a seasonably made manoeuvre, the effect of the artillery and the fire of musketry, the rain, the mud even, have contributed to checks or successes; but that in general, a brave man on foot or on horseback ought to beat a poltroon. By inculcating these truths to dragoons, they will be able to believe themselves superior to their adversary, either when employed as infantry or when charging as horsemen. It is thus that the Turks and Circassians act, whose cavalry often dismount to right in the woods or behind a shelter gun in hand. Meanwhile, it cannot be concealed, good chiefs and good soldiers are necessary to carry the education of a troop to that degree of perfection. However that may be, a regiment of dragoons attached to each corps d'armee of infantry or cavalry, as well as to an advanced or rear guard, could be very useful; whilst that forming whole divisions of dragoons is reducing them to the impossibility of being employed as infantry in the small number of unexpected cases where that would become necessary. It would then be better to make lancers of them. All that has been said in respect to the formation of infantry may be applied to the cavalry, saving the following modifications. 1. Lines deployed checkerwise, or in echelon are much more suitable to cavalry than full lines; whilst that in the infantry the order deployed en  échiquier would be too broken, and dangerous if cavalry changed to penetrate and take the battalions in flank; the disposition en échiquier is sure only for preparatory movements previous to the contact with the enemy, or for lines in columns of attack able to defend themselves alone in every direction against cavalry. Whether we form the echiquier, or prefer full lines, the distance between the lines ought to be sufficiently great in order that they should not reciprocally drag each other on in case of a check, in view of the rapidity with which they are rallied if the charge is unfortunate. It is merely well to observe that, in the echiquier, the distance may be less than in the full line. In no case, could the second line be full. It ought to be formed in columns by divisions, or at least to leave in it openings for two squadrons which we ploy in columns on the flanks of each regiment, in order to facilitate the passage of the troops rallied. 2. In the order in columns of attack upon the centre, the cavalry should be by regiments, and the infantry only by battalions. To comply well with this order, regiments of six squadrons are then necessary, in order that in ploying upon the centre by divisions they may be able to form three. If they have only four squadrons, they would then only form two lines. 3. The column of attack of cavalry should never be compact like that of infantry, but at full or half squadron distance, with a view to having ground for separating and charging. This distance is only intended for troops thrown out to combat; when they are in repose behind the line they can be closed together in order to cover less ground and to diminish the space which they would have to pass over in order to engage, provided, nevertheless, that those masses shall be under shelter or out of reach of cannon. 4. An attack on the flank being more to be feared in cavalry than in a combat of infantry against infantry, it is necessary to establish, upon the extremities of a line of cavalry, some squadrons in echelon by platoons, in order that they be able to form, by a right wheel or a left wheel against the enemy who should come to disturb the flank. 5. For the same motive it is essential, as has already been said, to know how to throw seasonably some squadrons upon the flanks of a line or cavalry which we are about to attack; if there be irregular cavalry present, it is especially for that we ought to use it in the combat, because for this use it is worth as much and perhaps more than the regular. An important observation also is that, in the cavalry especially, it is  well that the commander-in-chief extend in depth rather than in length. For example, in a division of two brigades which should deploy, it would not be expedient that each brigade should form a single line behind the other, but rather that each brigade should have a regiment in first line, and one in the second: thus each unit of the line will have its own reserve behind it, an advantage which cannot be misunderstood, for events pass so quickly in charges, that it is impossible for a general officer to be master of two deployed regiments. It is true that in adopting this mode each general of brigade will have the disposition of his reserve, and that it would be well, nevertheless, to have one for the whole division; for this reason it is believed that five regiments for a division is very suitable for cavalry. If it is wished to engage in line by brigades of two regiments, the fifth serves as general reserve behind the centre. If it is wished, we may also have three regiments in line, and two in column behind each wing. Is it preferred, on the contrary, to take a mixed order by deploying but two regiments at a time, keeping the remainder in column, in this case, we have also a suitable order, since three regiments formed by divisions behind the line cover the flanks and the centre, at the same time leaving intervals for passing the first line if it is beaten. (See fig. 10, plate 3.) Two essential maxims are generally admitted for combats of cavalry against cavalry : the one is, that every first line ought to be sooner or later led back, for, in the supposition even that it should have made the most fortunate charge, it is probable that the enemy, by opposing to it fresh squadrons, will force it to rally behind the second line. The other maxim is, that with equal merit in the troops and chiefs, the victory will remain to him who shall have the last squadrons in reserve, and who shall know how to launch them at the proper moment upon the flanks of the hostile line, already engaged with his. It is upon these two truths we shall be able to form a just idea of the system of formation most suitable for conducting a heavy body of cavalry to the combat. Whatever order may be adopted, it is necessary to guard against deploying large bodies of cavalry in full lines; for they are masses difficult to handle, and if the first is driven back, the second will be dragged along with it without being able to draw a sabre. To the number of a thousand proofs that the late war has give us of this, we will cite the attack executed by Nansouty in columns by regiments, upon the Prussian cavalry deployed in advance of Chateau-Thierry.  In the first edition of this treatise I opposed the formation of cavalry upon more that two lines; but I have never intended to exclude several lines en echiquier or in echelon, nor reserves formed in columns; I wished to speak only of cavalry deployed for charging en muraille, and the lines of which uselessly accumulated the one behind the other, would be swept away as soon as the first should chance to retreat.16 For the rest, in cavalry more still than in the infantry, the moral ascendancy does a great deal; the coup d'oeil and the sang froid of the chief, the intelligence and bravery of the soldier, whether in the melee, or for rallying, procure victory oftener than such or such another formation. Meanwhile, when we can unite these two advantages, we are only the more sure of conquering, and nothing could legitimize the adoption of a mode recognized as vicious. The history of the late wars (1812 to 1815) has renewed also ancient controversies for deciding if cavalry fighting in line can triumph in the long run over irregular cavalry, which avoiding all serious engagement flies with the speed of the Parthian, and returns to the combat with the same vivacity, limiting itself to harrassing the enemy by individual attacks. Lloyd has pronounced for the negative, and several exploits of the Cossacks against the excellent French cavalry seems to confirm his judgment;17 but we must not be deceived, and think that it would be possible to execute the same things with disciplined light cavalry, which we should launch as foragers against squadrons well united. It is the constant habit of moving in disorder which causes irregular troops to know how to direct all their individual efforts towards a common end; the best exercised hussars will never approach to the natural instinct of the Cossack, of the Tscherkes or of the Turk. If experience has proved that irregular charges may bring about the defeat of the best cavalry in partial combats, it is necessary to acknowledge also the impossibility of counting upon helter-skelter charges in  pitched battles, upon which depends often the fate of a whole war. Such a charge could without doubt aid an attack in lines) but alone it would produce nothing important. We ought then to consider those irregular charges as a powerful auxiliary in the daily rencounters of cavalry, and as a useful accessory in decisive shocks. From all that which precedes, we ought to conclude, in my opinion, that for battles, a regular cavalry, furnished with long arms, and for petty warfare, an irregular cavalry armed with excellent pistols, with lances and with sabres, will ever be the best organization for this important branch of the service of war. For the rest, whatever system we adopt it appears not less incontestable that a numerous cavalry, whatever be its nature, ought to have a great influence upon the results of a war; it can carry to a distance terror into the ranks of the enemy; it captures convoys, blockades the army, thus to speak, in its positions; renders its communications difficult, if not impossible; disturbs all harmony in its enterprises and in its movements. In a word it procures almost the same advantages as a rising in mass of the people, by carrying confusion upon the flanks and upon the rear of an army, and by making it impossible for its general to calculate anything with certainty. Every organization, then, which should tend to double the strength of the cavalry, in case of war, by incorporating militia into it, would be a good system; for those militia, aided by a few good squadrons, will be able at the end of some months' campaign, to make good partizans. Without doubt those militia will not have all the qualities which the warlike and wandering populations possess who pass, thus to speak, their lives on horseback, and whose first instincts are those of petty warfare; but they will supply them in part. In this respect Russia has a great advantage over all her neighbors, as much by the number and quality of her horses of the Don, as by the nature of the irregular militia which she can raise at a moments' warning. The following is what I wrote twenty years ago in Chapter 35 of the Treatise of Grand Military Operations, upon this same subject:--
The immense advantages which the Cossacks have given to the Russian armies are incalculable. Those light troops, insignificant in the shock of a great battle, (unless it be for falling upon the flanks,) are terrible in the pursuit and in a war of posts; this is the most redoubtable enemy for all the combinations of a general, because he is never sure of the arrival and execution of his orders, his convoys are always exposed,  and his operations uncertain. So long as an army has a few half regular regiments of them, their whole utility is not recognized; but when the number of them is increased to fifteen or twenty thousand, their importance is felt, especially in countries where the people are not hostile to them. When a convoy is carried away by them, it is necessary to escort all such, and that the escort be numerous and well conducted. We are never certain of making a tranquil march, because we know not where our enemies are; these labors require considerable forces, and the regular cavalry is soon rendered unserviceable by fatigues which it is not able to sustain. For the rest, I believe that hussars or volunteer lancers, raised or organized at the moment of the war, well conducted, and moving where bold chiefs conduct them at their will, would accomplish nearly the same object; but it is necessary to regard them as independent, for if they were to receive orders from the headquarters, they would no longer be partizans. They would not, perhaps, have all the qualities of good Cossacks, but they might approach them.Austria has also in the Hungarians, the Transyivanians and the Croats, resources which other States have not; however, the services rendered by the mounted landwehr prove that we can draw also upon this species of cavalry, were it only for relieving the regular cavalry in the accessory services which abound in all armies, as escorts, despatches, detachments for conducting convoys, flankers, &c. Mixed corps of regular and irregular cavalry can often render more real services than if they were composed only of cavalry of the line, for the fear of compromising and ruining the latter, often prevents launching it into audacious movements which may produce immense results. I would not terminate this article without noticing the by far too passionate attacks of which it has been the object on the part of General Bismark, and with which, unfortunately, I have become acquainted too late for replying to them as I ought to do. The passage which seems to have especially excited his wrath, is that in which I have advanced, after many others, that cavalry could not defend a position by itself. The General, who doubtless pretends that cavalry can make war of itself alone, and that it could hold a position quite as well as infantry, thinks to justify such sophisms in going for examples even to the war of Hannibal upon the Ticino, as if musketry, shells and grape shot had brought about no change in the employment of this arm! Proud of his equestrian erudition, he treats as ignorant all who do not think like him.  Without being a Seydlitz or a Lagueriniere, one may very well reason upon the employment of cavalry in war, and although I have no pretension to being a trooper, I can say that the most experienced of generals in our day have partaken of my ideas upon the cavalry, and that in many battles I have often judged of it better than those who have commanded large masses of it. The only one of my maxims which has excited some controversies, is that relative to the gait of the trot for charges against cavalry. Whatever may have been said of it, I believe still, at the moment at which I am writing, that success depends much upon the maintenance of order until the instant of the shock; and that for lancers especially, the shock of a mass well in hand and at the trot, would triumph over a troop scattered by the gallop. As for the rest, to maintain order as much as possible in the shock, to endeavor to have it seconded at the opportune moment by a flank attack; to be able to give moral impulsion to one's troop, and to have an echelon ready for support, are the only elements of success which I have ever recognized as practicable in the charges of cavalry against cavalry, for all the fine maxims in the world vanish in a struggle rapid as the lightning, where the most skillful professors would only have time to parry sabre cuts, without even being in condition to give an order which could be heard and executed. With regard to the good employment of the cavalry, in the whole of a battle as in that of the whole of a war, I believe that no experienced general would repudiate the ideas which I have advanced upon this subject. I have never denied that cavalry would not concur in the defense of a position; but that it would defend it by itself, I shall ever deny. Posted on a position, behind a hundred pieces of artillery, it will be able to maintain itself there if one be contented with cannonading it, as the French cavalry so bravely defended itself at Eylau; but let infantry and artillery march upon it after having paralyzed its batteries, and you will see if the position will be defended. For the rest, the true cause of the great wrath of General B**** is easy to divine. I have had the imprudence to say that his Treatise upon the Cavalry, albeit very erudite, had not caused much progress to be made in this arm. This judgment has doubtless appeared to him severe, and in spite of the wrongs of the author in regard to myself, I agree that it was pronounced in too absolute a manner. Meanwhile, after the teachings we have been able to receive from the cavalry of Seydlitz and  of Napoleon, I do not know whether that which M. B**** would organize and conduct according to his doctrines, would do much better; here lies the question. For having dared to resolve it negatively, I am but an ignoramus; there is good criticism for you! If opinions be free, cannot one discuss them without injuries? As for myself, I recognize in M. B**** much mind and erudition; perhaps he has even too much for the subject he treats. When wit sparkles and the passions speak, reason and judgment sleep. As for the rest, I have already observed in the notice which precedes this work, that it was not in serious books that a military man ought to reply to personalities especially after having been ignorant of them for six years.
Article XLVI: the employment of the artillery.The artillery is at the same time an offensive and defensive arm, equally formidable. As an offensive means, a great battery, well employed, crushes a hostile line, shakes it, and facilitates to the troops which attack it the means of breaking it. As a defensive arm, it must be acknowledged that it doubles the strength of a position, not only by the harm it does an enemy from afar, and by the moral effect which it produces at a long distance upon troops which march to the attack, but yet by the local defense which it will make of the position itself, and within grape shot range. It is not less important in the attack and defense of places, or of intrenched camps, for it is the soul of modern fortification. We have said a few words upon its distribution in the line of battle, but we are more embarrassed in speaking of the mode in which it should be made to act in combat. Here the chances multiply in such a manmer, by reason of the particular circumstances of the affair, of the ground and of the movements of the enemy, that we cannot say that the artillery has any action independent of that of the other arms. In the meanwhile we have seen Napoleon at Wagram throw a battery of a  hundred pieces in the gap occasioned in his line by the departure of the corps of Massena, and thus to hold in check all the efforts of the Austrian centre; but it would be very dangerous to set up as a maxim such an employment of the artillery. We shall limit ourselves then to presenting here a few fundamental data, observing that they are based upon the condition of this arm, such as it existed in the late wars; the employment of the new discoveries not being yet well determined could not find place here. 1. In the offensive, we ought to unite a certain mass of artillery upon the point where we are preparing to direct our heaviest blows; we will employ it at first for shaking by its fire the hostile line, in order to second the attack of the infantry and cavalry. 2. There are necessary, besides, a few batteries of horse artillery, for following the offensive movement of the columns, independently of the light foot batteries which have the same object. We must not, however, throw too much foot artillery in an offensive movement; it can be placed in such a manner as to attain the object without following the columns. 3. We have already said that the half, at least, of the horse artillery ought to be united in reserve, in order to be directed rapidly wherever its services shall be most required.18 To this effect it is necessary to place it upon the most open ground, where it can be moved in every direction. We have also mentioned the best post for the artillery of position. 4. Batteries, although spread in general over the whole of a defensive line, ought to know how to direct their attention upon the point where the enemy would find more advantages and facilities to penetrate; it is necessary then that the general commanding the artillery should know the strategic and the tactical point of a field of battle, as well as the ground itself, and that every distribution of the reserves be calculated upon this double data. 5. Every one knows that artillery posted on level ground, or in the midst of declivities gently inclined en glacis, is that whose effect, in  direct or ricochet firing, will be the most murderous. No person is ignorant, either, that the concentric fire is the most suitable. 6. Artillery of every kind employed in battles ought never to forget that its principal destination is to batter the troops of the enemy, and not to reply to his batteries. Meanwhile, as it is well not to leave the field free to the action of the hostile artillery, it is useful to combat it for drawing its fire; a third of the disposable pieces may be destined to that object, but two thirds at least ought to be directed upon the cavalry and the infantry. 7. If the enemy advance in deployed lines, the batteries should seek to cross their fires in order to take those lines obliquely; those which could place themselves upon the flanks, and batter the lines in their prolongation, would create a decisive effect. 8. When the enemy advances in columns, they can be battered in front; that is to say, in their depth. However, it is not less advantageous to batter them obliquely, and especially in flank or in reverse. The moral effect produced upon troops by artillery taking them in reverse, it incalculable. It is rare that the most valiant soldiers are not astonished and shaken. The fine movement of Ney upon Preitiz (battle of Bautzen) was neutralized by a few pieces of Kleist, which took his columns in flank, arrested them, and decided the Marshal to change his good direction. A few pieces of light artillery, thrown at every risk upon the flanks for obtaining a like result, would never be ventured without utility. 9. It is acknowledged that batteries should be constantly sustained by infantry or cavalry, and that it is advantageous to support them properly upon the flanks. Meanwhile many cases present themselves when it is necessary to deviate from this maxim, and the example of Wagram, of which we have spoken, is one of the most remarkable of them. 10. It is very important that, in the attacks of cavalry, the artillery do not allow itself to be frightened, and that it fire with ball, but especially with grape shot, as long as possible.19In this case, the infantry charged with protecting batteries ought to be formed in squares in proximity, in order to give refuge to the horses, and afterwards to the cannoniers; long squares, proportioned to the extent of the front of the battery, seem the most proper for accomplishing this object, when the infantry is in rear of the pieces. If it be found at the side, perfect  squares will be preferable. We are assured that rocket batteries can be employed against cavalry, the horses of which they frighten; but I repeat, this is still an experiment to make, and we could base no maxim upon data so uncertain. 11. In the attacks of infantry against artillery, the maxim to fire as long as possible, without, nevertheless, commencing at too great a distance, is yet more rigorous than in the case above mentioned. The cannoniers will always have the means of securing themselves from infantry, if they are properly sustained. Here is one of the cases for engaging the three arms at the same time, for if the hostile infantry be shaken by the artillery, a combined attack of the infantry and cavalry will cause its destruction. 12. The proportions of the artillery have considerably varied in the late wars. Napoleon went to the conquest of Italy in 1800, with forty or fifty pieces, and succeeded completely; whilst in 1812 he invaded Russia with twelve hundred pieces, and did not succeed. This sufficiently proves that no absolute rule could fix those proportions. It is generally admitted that three pieces to a thousand combattants are sufficient, and even in Turkey, as well as in the mountains, this is a great deal too much. The proportions of heavy artillery, the reserve, so called, with those of lighter artillery, equally vary. It is a great fault to have too much heavy artillery, for in battles six or eight pounder guns produce nearly the same effect as twelve pounders, and there is meanwhile a great difference in the mobility and the accessory embarrassments of these calibres, For the rest, one of the most notable proofs which can be cited for appreciating the influence of the proportions of the armament upon the success of armies, was given by Napoleon after the battle of Eylau; the cruel losses which his troops sustained by the fire of the numerous artil lery of the Russians, made him feel the necessity of increasing his own. With an activity difficult to conceive, he set all the arsenals at work in Prussia, on the line of the Rhine, and even at Metz, to increase the number of his pieces, and to cast new ones, for turning to account the munitions which he had captured in the campaign. In three months he doubled, at four hundred leagues from his frontiers, the personel and the materiel of his artillery, a thing unheard — of in the annals of war. 13. One of the most suitable means for obtaining the best possible employment of the artillery, would be always to give the superior command of this arm to a general of artillery who is at the same time a good tactician and strategician; this chief would have the faculty of  disposing not only of the artillery reserve, but even of half of the pieces attached to the different corps or divisions. He could thus concert with the generalissimo as to the moment and the place where considerable masses of artillery could best contribute to the victory; but he will never make such a union of masses without having taken previously the orders of the commander-in-chief. At the moment when I was about to publish this article for the second time, I received a pamphlet from General Okounieff upon the importance of the artillery. However interesting it may be, it could not decide me to change what I have said upon this arm. The author avows, with a laudable frankness, that he had not sufficiently appreciated that importance in his work upon the employment of the three arms; and as if to make reparation to the artillery, he sustains now that it is henceforth to decide battles, and to become for that purpose even the principal arm of European armies. As I have recognized at all times the part that a well employed artillery may have in victories, I am very much disposed to admit with the author, that its influence would be greater if it were known always how to realize from it the part of which it is susceptible. I acknowledge, also, that several quite recent inventions, which will augment its effect whether for ricochet firing, or for grape at long range, are of a nature to call the attention of generals who shall be at liberty to make use of them, and who have at command the means of trying their effects, as also finding the means of securing themselves against them. The pamphlet of General Okounieff would then have already attained an important end in in opening this vast quarry; but after having rendered him justice, I shall be permitted to say that the author has rather overstepped the mark, for if it were necessary to believe all he advances, there would no longer, be required in an army anything but cuirassiers, artillerists, and the infantry necessary for holding enclosed posts, for the rest would be but food for projectiles. Setting out with this dominant idea, M. Okounieff concludes from it by a very natural consequence, that the means of gaining battles will be reduced to breaking the centre of an army by dint of cannon shots, and in having masses prepared to fall upon this breach; a means which he finds very preferable to those he calls movements of conversion, and which to this day, according to his own confession, have gained very many battles. Here, I own, I am obliged to contend that there is something too absolute in these assertions. In the first place, I do not perfectly comprehend  those movements of conversion; they are doubtless attacks for outflanking a wing at the same time that a part of the front is assailed. If I am not deceived, these kinds of manoeuvres are not always movements of conversion; at best it is but a quarrel of definition, which is really of little importance; that which I do not consider well founded, is the idea that an exclusive manoeuvre can be adopted as an universal panacea, and that it is necessary to renounce all other tactics than that of immense batteries and heavy masses piercing centres. For my part, if I had to combat an enemy professing such exclusive ideas, I should be no wise embarrassed in opposing to him means which would defeat his favorite attacks. At first I should employ that which M. Okounieff himself cites on page 35, as having been adopted with success by the Prince de Lichtenstein at the battle of Wagram, against the famous column of Macdonald; the system employed at Cannae by Hannibal, could all the better find here its application, as such a mass battered by the concentric fires of an artillery equal in number, and disposed in a concave line, like that of the Arch-Duke Charles at Essling, would be much compromised. Finally, in order to avoid cutting the army in two parts, who knows if one of those movements of conversion which the author would repudiate, would not be an excellent means to oppose to his system, since it would transport the decisive effort of the combat on quite another point than the centre? Far be the thought from me of contesting all merit in a strong attack upon the centre; I have often recommended it, but especially when it should be combined with an attack upon the extremity of the line (agreeably to figure 12 of plate 1, page 210,) or where it should be made on a rather too extended line. Be that as it may, it appears to me that the author has rather lost sight of the fact that the moral of the troops, the character and genius of the chiefs have also a great influence upon the issue of battles. These are batteries less murderous, but not less efficacious. It must not be forgotten either that all fields of battle and all countries do not offer the same advantages to artillery; in Italy, in Switzerland, in Vendee, in many parts of Germany, in every very broken country, in a word, we do not find fields of battle like Wagram and Leipsic. As for the rest, there are useful lessons in his pamphlet, to which no other reproach could be made than that of having drawn him from one extreme to the other. The author has without doubt wished to imitate those advocates who, after a fine defense, draw exaggerated conclusions certain that the judges will always abate the half of them; wise men  will be able to take what they find in them true and useful, and give him credit for them. The first result of this treatise should be to awaken the attention of men who have the mission of influencing the destinies of. armies, that is to say, of governments and generals. The second will be, perhaps, the doubling of the materiel and personal of the artillery, and the adoption of all improvements capable of augmenting its destructive effect. And as artillerists will be in the number of the first victims, it will be very necessary to engage in instructing in the infantry, men chosen to serve the pieces at need, and to fill even the vacancies which battles would leave in the ranks of the artillery. Finally, it will be necessary to endeavor to find the means of neutralizing the effects of this carnage, and the first which occur seem to be the modification in the armament and the equipment of troops, then the adoption of a new tactics which will render results as prompt as possible. This task will be for the rising generation, when we shall have tested by experience all the inventions with which we are occupied in the schools of artillery, whilst awaiting better. Happy will be those who, in the first rencounters, shall have a plenty of schrapnel howitzers, many guns charged at the breech, and firing thirty shots a minute; many pieces richocheting at the height of a man, and never failing their mark upon, one or another part of the field of combat; finally, the most improved rockets — without counting even the famous steam guns of Perkins, reserved to the defense of ramparts, but which, if the written statement of Lord Wellington is to be believed, will yet be able here to make cruel ravages. * * * What a beautiful text for preaching universal peace and the exclusive reign of railroads! I shall be pardoned if I terminate a discussion so grave, by a phrase bordering upon pleasantry. But we must take a less sombre view of the future with which so many brave men menace us, who by a cruel foresight combine the means of rendering war still more bloody than it is, and that, too, in the hope of assuring the triumph of their banners. A terrible but indispensable emulation, if we would remain on an equality with our neighbors so long as the law of nations shall not have placed limits to those inventions.