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 IX

Defensive Battle

WE have said, in the preceding chapter, that the defender, in his defensive, would make use of a battle, technically speaking, of a purely offensive character, if, at the moment the enemy invades his theatre of war, he marches against him and attacks him; but that he might also wait for the appearance of the enemy in his front, and then pass over to the attack; in which case also the battle tactically would be again an offensive battle, although in a modified form; and lastly, that he might wait till the enemy attacked his position, and then oppose him both by holding a particular spot, and by offensive action with portions of his force. In all this we may imagine several different gradations and shades, deviating always more from the principle of a positive counterstroke, and passing into that of the defence of a spot of ground. We cannot here enter on the subject of how far this should be carried, and which is the most advantageous proportion of the two elements of offensive and defensive, as regards the winning a decisive victory. But we maintain that when such a result is desired, the offensive part of the battle should never be completely omitted, and we are convinced that all the effects of a decisive victory may and must be produced by this offensive part, just as well as in a purely tactical offensive battle.

In the same manner as the field of battle is only a point in strategy, the duration of a battle is only, strategically, an instant of time, and the end and result, not the course of a battle, constitutes a strategic quantity.

Now, if it is true that a complete victory may result from the offensive elements which lie in every defensive battle, then there would be no fundamental difference between an offensive and a defensive battle, as far as regards strategic combinations; we are indeed convinced that this is so, but the thing wears a different appearance. In order to fix the subject more distinctly in the eye, to make our view clear and thereby remove the appearance now referred to, we shall sketch, hastily, the picture of a defensive battle, such as we imagine it.

The defensive waits the attack in a position; for this he has selected proper ground, and turned it to the best account, that is, he has made himself well acquainted with the locality, thrown up strong entrenchments at some of the most important points, opened and levelled communications, constructed batteries, fortified villages, and looked out places where he can draw up his masses under cover, etc., etc., etc. Whilst the forces on both sides are consuming each other at the different points where they come into contact, the advantage of a front more or less strong, the approach to which is made difficult by one or more parallel trenches or other obstacles, or also by the influence of some strong commanding points, enables him with a small part of his force to destroy great numbers of the enemy at every stage of the defence up to the heart of the position. The points of support which he has given his wings secure him from any sudden attack from several quarters; the covered ground which he has chosen for his masses makes the enemy cautious, indeed timid, and affords the defensive the means of diminishing by partial and successful attacks the general backward movement which goes on as the combat becomes gradually concentrated within narrower limits. The defender therefore casts a contented look at the battle as it burns in a moderate blaze before him;—but he does not reckon that his resistance in front can last for ever;—he does not think his flanks impregnable;—he does not expect that the whole course of the battle will be changed by the successful charge of a few battalions or squadrons. His position is deep, for each part in the scale of gradation of the order of battle, from the division down to the battalion, has its reserve for unforeseen events, and for a renewal of the fight; and at the same time an important mass, one fifth to a quarter of the whole, is kept quite in the rear out of the battle, so far back as to be quite out of fire, and if possible so far as to be beyond the circuitous line by which the enemy might attempt to turn either flank. With this corps he intends to cover his flanks from wider and greater turning movements, secure himself against unforeseen events, and in the latter stage of the battle, when the assailant's plan is fully developed, when the most of his troops have been brought into action, he will throw this mass on a part of the enemy's army, and open at that part of the field a smaller offensive battle on his own part, using all the elements of attack, such as charges, surprise, turning movements, and by means of this pressure against the centre of gravity of the battle, now only resting on a point, make the whole recoil.

This is the normal idea which we have formed of a defensive battle, based on the tactics of the present day. In this battle the general turning movement made by the assailant in order to assist his attack, and at the same time with a view to make the results of victory more complete, is replied to by a partial turning movement on the part of the defensive, that is, by the turning of that part of the assailant's force used by him in the attempt to turn. This partial movement may be supposed sufficient to destroy the effect of the enemy's attempt, but it cannot lead to a like general enveloping of the assailant's army; and there will always be a distinction in the features of a victory on this account, that the side fighting an offensive battle encircles the enemy's army, and acts towards the centre of the same, while the side fighting on the defensive acts more or less from the centre to the circumference, in the direction of the radii.

On the field of battle itself, and in the first stages of the pursuit, the enveloping form must always be considered the most effectual; we do not mean on account of its form generally, we only mean in the event of its being carried out to such an extreme as to limit very much the enemy's means of retreat during the battle. But it is just against this extreme point that the enemy's positive counter-effort is directed, and in many cases where this effort is not sufficient to obtain a victory, it will at least suffice to protect him from such an extreme as we allude to. But we must always admit that this danger, namely, of having the line of retreat seriously contracted, is particularly great in defensive battles, and if it cannot be guarded against, the results in the battle itself, and in the first stage of the retreat are thereby very much enhanced in favour of the enemy.

But as a rule this danger does not extend beyond the first stage of the retreat, that is, until night-fall; on the following day enveloping is at an end, and both parties are again on an equality in this respect.

Certainly the defender may have lost his principal line of retreat, and therefore be placed in a disadvantageous strategic situation for the future; but in most cases the turning movement itself will be at an end, because it was only planned to suit the field of battle, and therefore cannot apply much further. But what will take place, on the other hand, if the defender is victorious? A division of the defeated force. This may facilitate the retreat at the first moment, but next day a concentration of all parts is the one thing most needful. Now if the victory is a most decisive one, if the defender pursues with great energy, this concentration will often become impossible, and from this separation of the beaten force the worst consequences may follow, which may go on step by step to a complete rout. If Buonaparte had conquered at Leipsic, the allied army would have been completely cut in two, which would have considerably lowered their relative strategic position. At Dresden, although Buonaparte certainly did not fight a regular defensive battle, the attack had the geometrical form of which we have been speaking, that is, from the centre to the circumference; the embarrassment of the Allies in consequence of their separation, is well known, an embarrassment from which they were only relieved by the victory on the Katzbach, the tidings of which caused Buonaparte to return to Dresden with the Guard.

This battle on the Katzbach itself is a similar example. In it the defender, at the last moment passes over to the offensive, and consequently operates on diverging lines; the French corps were thus wedged asunder, and several days after, as the fruits of the victory, Puthod's division fell into the hands of the Allies.

The conclusion we draw from this is, that if the assailant, by the concentric form which is homogeneous to him, has the means of giving expansion to his victory, on the other hand the defender also, by the divergent form which is homogeneous to the defence, acquires a a means of giving greater results to his victory than would be the case by a merely parallel position and perpendicular attack, and we think that one means is at least as good as the other.

If in military history we rarely find such great victories resulting from the defensive battle as from the offensive, that proves nothing against our assertion that the one is as well suited to produce victory as the other; the real cause is in the very different relations of the defender. The army acting on the defensive is generally the weaker of the two, not only in the amount of his forces, but also in every other respect; he either is, or thinks he is, not in a condition to follow up his victory with great results, and contents himself with merely fending off the danger and saving the honour of his arms. That the defender by inferiority of force and other circumstances may be tied down to that degree we do not dispute, but there is no doubt that this, which is only the consequence of a contingent necessity, has often been assumed to be the consequence of that part which every defender has to play: and thus in an absurd manner it has become a prevalent view of the defensive that its battles should really be confined to warding off the attacks of the enemy, and not directed to the destruction of the enemy. We hold this to be a prejudicial error, a regular substitution of the form for the thing itself; and we maintain unreservedly that in the form of war which we call defence, the victory may not only be more probable, but may also attain the same magnitude and efficacy as in the attack, and that this may be the case not only in the total result of all the combats which constitute campaign, but also in any particular battle, if the necessary degree of force and energy is not wanting.