NOTE: This version of Clausewitz's On War is the long-obsolete J.J. Graham translation published in London in 1873. The 1976/84 Howard/Paret version is the  standard translation today.


Chapter V

Order of Battle of an Army

THE order of battle is that division and formation of the different arms into separate parts or sections of the whole Army, and that form of general position or disposition of those parts which is to be the norm throughout the whole campaign or war.

It consists, therefore, in a certain measure, of an arithmetical and a geometrical element, the division and the form of disposition. The first proceeds from the permanent peace organisation of the army; adopts as units certain parts, such as battalions, squadrons, and batteries, and with them forms units of a higher order up to the highest of all, the whole army, according to the requirements of predominating circumstances. In like manner, the form of disposition comes from the elementary tactics, in which the army is instructed and exercised in time of peace, which must be looked upon as a property in the troops that cannot be essentially modified at the moment war breaks out, the disposition connects these tactics with the conditions which the use of the troops in war and in large masses demands, and thus it settles in a general way the rule or norm in conformity with which the troops are to be drawn up for battle.

This has been invariably the case when great armies have taken the field, and there have been times when this form was considered as the most essential part of the battle.

In the seventeenth and eighteenth centuries, when the improvements in the firearms of infantry occasioned a great increase of that arm, and allowed of its being deployed in such long thin lines, the order of battle was thereby simplified, but, at the same time it became more difficult and more artificial in the carrying out, and as no other way of disposing of cavalry at the commencement of a battle was known but that of posting them on the wings, where they were out of the fire and had room to move, therefore in the order of battle the army always became a closed inseparable whole. If such an army was divided in the middle, it was like an earthworm cut in two: the wings had still life and the power of motion, but they had lost their natural functions. The army lay, therefore, in a manner under a spell of unity, and whenever any parts of it had to be placed in a separate position, a small organisation and disorganisation became necessary. The marches which the whole army had to make were a condition in which, to a certain extent, it found itself out of rule. If the enemy was at hand, the march had to be arranged in the most artificial manner, and in order that one line or one wing might be always at the prescribed distance from the other, the troops had to scramble over everything: marches had also constantly to be stolen from the enemy, and this perpetual theft only escaped severe punishment through one circumstance, which was, that the enemy lay under the same ban.

Hence, when, in the latter half of the eighteenth century, it was discovered that cavalry would serve just as well to protect a wing if it stood in rear of the army as if it were placed on the prolongation of the line, and that, besides this, it might be applied to other purposes than merely fighting a duel with the enemy's cavalry, a great step in advance was made, because now the army in its principal extension or front, which is always the breadth of its order of battle (position), consisted entirely of homogeneous members, so that it could be formed of any number of parts at pleasure, each part like another and like the whole. In this way it ceased to be one single piece and became an articulated whole, consequently pliable and manageable: the parts might be separated from the whole and then joined on again without difficulty, the order of battle always remained the same.—Thus arose the corps consisting of all arms, that is, thus such an organisation became possible, for the want of it had been felt long before.

That all this relates to the combat is very natural. The battle was formerly the whole war, and will always continue to be the principal part of it; but, the order of battle belongs generally more to tactics than strategy, and it is only introduced here to show how tactics in organising the whole into smaller wholes made preparations for strategy.

The greater armies become, the more they are distributed over wide spaces and the more diversified the action and reaction of the different parts amongst themselves, the wider becomes the field of strategy, and, therefore, then the order of battle, in the sense of our definition, must also come into a kind of reciprocal action with strategy, which manifests itself chiefly at the extreme points where tactics and strategy meet, that is, at those moments when the general distribution of the combatant forces passes into the special dispositions for the combat.

We now turn to those three points, the division, combination of arms, and order of battle (disposition) in a strategic point of view.


In strategy we must never ask what is to be the strength of a division or a corps, but how many corps or division an army should have. There is nothing more unmanageable than an army divided into three parts, except it be one divided into only two, in which case the chief command must be almost neutralised.

To fix the strength of great and small corps, either on the grounds of elementary tactics or on higher grounds, leaves an incredibly wide field for arbitrary judgment, and heaven knows what strange modes of reasoning have sported in this wide field. On the other hand, the necessity of forming an independent whole (army) into a certain number of parts is a thing as obvious as it is positive, and this idea furnishes real strategic motives for determining the number of the greater divisions of an army, consequently their strength, whilst the strength of the smaller divisions, such as companies, battalions, etc., is left to be determined by tactics.

We can hardly imagine the smallest independent body in which there are not at least three parts to be distinguished, that one part may be thrown out in advance, and another part be left in rear: that four is still more convenient follows of itself, if we keep in view that the middle part, being the principal division, ought to be stronger than either of the others; in this way, we may proceed to make out eight, which appears to us to be the most suitable number for an army if we take one part for an advanced guard as a constant necessity, three for the main body, that is a right wing, centre and left wing, two divisions for reserve, and one to detach to the right, one to the left. Without pedantically ascribing a great importance to these numbers and figures, we certainly believe that they represent the most usual and frequently recurring strategic disposition, and on that account one that is convenient.

Certainly it seems that the supreme direction of an army (and the direction of every whole) must be greatly facilitated if there are only three or four subordinates to command, but the commander-in-chief must pay dearly for this convenience in a twofold manner. In the first place, an order loses in rapidity, force, and exactness if the gradation ladder down which it has to descend is long, and this must be the case if there are corps-commanders between the division leaders and the chief; secondly, the chief loses generally in his own proper power and efficiency the wider the spheres of action of his immediate subordinates become. A general commanding 100,000 men in eight divisions exercises a power which is greater in intensity than if the 100,000 men were divided into only three corps. There are many reasons for this, but the most important is that each commander looks upon himself as having a kind of proprietary right in his own corps, and always opposes the withdrawal from him of any portion of it for a longer or shorter time. A little experience of war will make this evident to any one.

But on the other hand the number of divisions must not be too great, otherwise disorder will ensue. It is difficult enough to manage eight divisions from one head quarter, and the number should never be allowed to exceed ten. But in a division in which the means of circulating orders are much less, the smaller normal number four, or at most five, may be regarded as the more suitable.

If these factors, five and ten, will not answer, that is, if the brigades are too strong, then corps d'armée must be introduced; but we must remember that by so doing, a new power is created, which at once very much lowers all other factors.

But now, what is too strong a brigade? The custom is to make them from 2,000 to 5,000 men strong, and there appear to be two reasons for making the latter number the limit; the first is that a brigade is supposed to be a subdivision which can be commanded by one man directly, that is, through the compass of his voice: the second is that any larger body of infantry should not be left without artillery, and through this first combination of arms a special division of itself is formed.

We do not wish to involve ourselves in these tactical subtilties, neither shall we enter upon the disputed point, where and in what proportions the combination of all three arms should take place, whether with divisions of 8,000 to 12,000 men, or with corps which are 20,000 to 30,000 men strong. The most decided opponent of these combinations will scarcely take exception at the mere assertion, that nothing but this combination of the three arms can make a division independent, and that therefore, for such as are intended to be frequently detached separately, it is at least very desirable.

An army of 200,000 men in ten divisions, the divisions composed of five brigades each, would give brigades 4,000 strong. We see here no disproportion. Certainly this army might also be divided into five corps, the corps into four divisions, and the division into four brigades, which makes the brigade 2,500 men strong; but the first distribution, looked at in the abstract, appears to us preferable, for besides that, in the other, there is one more gradation of rank, five parts are too few to make an army manageable; four divisions, in like manner, are too few for a corps, and 2,500 men is a weak brigade, of which, in this manner, there are eighty, whereas the first formation has only fifty, and is therefore simpler. All these advantages are given up merely for the sake of having only to send orders to half as many generals. Of course the distribution into corps is still more unsuitable for smaller armies.

This is the abstract view of the case. The particular case may present good reasons for deciding otherwise. Likewise, we must admit that, although eight or ten divisions may be directed when united in a level country, in widely extended mountain positions the thing might perhaps be impossible. A great river which divides an army into halves, makes a commander for each half indispensable; in short, there are a hundred local and particular objects of the most decisive character, before which all rules must give way.

But still, experience teaches us, that these abstract grounds come most frequently into use and are seldomer overruled by others than we should perhaps suppose.

We wish further to explain clearly the scope of the foregoing considerations by a simple outline, for which purpose we now place the different points of most importance next to each other.

As we mean by the term numbers, or parts of a whole, only those which are made by the primary, therefore the immediate division, we say.

1. If a whole has too few members it is unwieldy.

2. If the parts of a whole body are too large, the power of the superior will is thereby weakened.

3. With every additional step through which an order has to pass, it is weakened in two ways: in one way by the loss of force, which it suffers in its passage through an additional step; in another way by the longer time in its transmission.

The tendency of all this is to show that the number of co-ordinate divisions should be as great, and the gradational steps as few as possible; and the only limitation to this conclusion is, that in armies no more than from eight to ten, and in subordinate corps no more than from four or at most six, subdivisions can be conveniently directed.

2.—Combination of Arms.

For strategy the combination of the three arms in the order of battle is only important in regard to those parts of the army which, according to the usual order of things, are likely to be frequently employed in a detached position, where they may be obliged to engage in an independent combat. Now it is in the nature of things, that the members of the first class, and for the most part only these, are destined for detached positions, because, as we shall see elsewhere, detached positions are most generally adopted upon the supposition and the necessity of a body independent in itself.

In a strict sense strategy would therefore only require a permanent combination of arms in army corps, or where these do not exist, in divisions, leaving it to circumstances to determine when a provisional combination of the three arms shall be made in subdivisions of an inferior order.

But it is easy to see that, when corps are of considerable size, such as 30,000 or 40,000 men, they can seldom find themselves in a situation to take up a completely connected position in mass. With corps of such strength, a combination of the arms in the divisions is therefore necessary. No one who has had any experience in war, will treat lightly the delay which occurs when pressing messages have to be sent to some other perhaps distant point before cavalry can be brought to the support of infantry—to say nothing of the confusion which takes place.

The details of the combination of the three arms, how far it should extend, how low down it should be carried, what proportions should be observed, the strength of the reserves of each to be set apart—these are all purely tactical considerations.

3.—The Disposition.

The determination as to the relations in space, according to which the parts of an army amongst themselves are to be drawn up in order of battle, is likewise completely a tactical subject, referring solely to the battle. No doubt there is also a strategic disposition of the parts; but it depends almost entirely on determinations and requirements of the moment, and what there is in it of the rational, does not come within the meaning of the term "order of battle." We shall therefore treat of it in the following chapter under the head of Disposition of an Army.

The order of battle of an army is therefore the organisation and disposition of it in mass ready prepared for battle. Its parts are united in such a manner that both the tactical and strategical requirements of the moment can be easily satisfied by the employment of single parts drawn from the general mass. When such momentary exigency has passed over, these parts resume their original place, and thus the order of battle becomes the first step to, and principal foundation of, that wholesome methodicism which, like the beat of a pendulum, regulates the work in war, and of which we have already spoken in the fourth chapter of the Second Book.