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 II

Nature of the Strategical Attack

WE have seen that the defensive in war generally—therefore, also, the strategic defensive—is no absolute state of expectancy and warding off, therefore no completely passive state, but that it is a relative state, and consequently impregnated more or less with offensive principles. In the same way the offensive is no homogeneous whole, but incessantly mixed up with the defensive. But there is this difference between the two, that a defensive, without an offensive return blow, cannot be conceived; that this return blow is a necessary constituent part of the defensive, whilst in the attack, the blow or act is in itself one complete idea. The defence in itself is not necessarily a part of the attack; but time and space, to which it is inseparably bound, import into it the defensive as a necessary evil. For in the first place, the attack cannot be continued uninterruptedly up to its conclusion, it must have stages of rest, and in these stages, when its action is neutralised, the state of defence steps in of itself; in the second place, the space which a military force, in its advance, leaves behind it, and which is essential to its existence, cannot always be covered by the attack itself, but must be specially protected.

The act of attack in war, but particularly in that branch which is called strategy, is therefore a perpetual alternating and combining of attack and defence; but the latter is not to be regarded as an effectual preparation for attack, as a means by which its force is heightened, that is to say, not as an active principle, but purely as a necessary evil; as the retarding weight arising from the specific gravity of the mass; it is its original sin, its seed of mortality. We say: a retarding weight, because if the defence does not contribute to strengthen the attack, it must tend to diminish its effect by the very loss of time which it represents. But now, may not this defensive element, which is contained in every attack, have over it a positively disadvantageous influence? If we suppose the attack is the weaker, the defence the stronger form of war, it seems to follow that the latter can not act in a positive sense prejudicially on the former; for as long as we have sufficient force for the weaker form, we should have more than enough for the stronger. In general—that is, as regards the chief part—this is true: in its detail we shall analyse it more precisely in the chapter on the culminating point of victory; but we must not forget that that superiority of the strategic defence is partly founded in this, that the attack itself cannot take place without a mixture of defence, and of a defensive of a very weak kind; what the assailant has to carry about with him of this kind are its worst elements; with respect to these, that which holds good of the whole, in a general sense, cannot be maintained; and therefore it is conceivable that the defensive may act upon the attack positively as a weakening principle. It is just in these moments of weak defensive in the attack, that the positive action of the offensive principle in the defensive should be introduced. During the twelve hours rest which usually succeeds a day's work, what a difference there is between the situation of the defender in his chosen, well-known, and prepared position, and that of the assailant occupying a bivouac, into which—like a blind man—he has groped his way, or during a longer period of rest, required to obtain provisions and to await reinforcements, etc., when the defender is close to his fortresses and supplies, whilst the situation of the assailant, on the other hand, is like that of a bird on a tree. Every attack must lead to a defence; what is to be the result of that defence, depends on circumstances; these circumstances may be very favourable if the enemy's forces are destroyed; but they may be very unfavourable if such is not the case. Although this defensive does not belong to the attack itself, its nature and effects must re-act on the attack, and must take part in determining its value.

The deduction from this view is, that in every attack the defensive, which is necessarily an inherent feature in the same, must come into consideration, in order to see clearly the disadvantages to which it is subject, and to be prepared for them.

On the other hand, in another respect, the attack is always in itself one and the same. But the defensive has its gradations according as the principle of expectancy approaches to an end. This begets forms which differ essentially from each other, as has been developed in the chapter on the forms of defence.

As the principle of the attack is strictly active, and the defensive, which connects itself with it, is only a dead weight; there is, therefore, not the same kind of difference in it. No doubt, in the energy employed in the attack, in the rapidity and force of the blow, there may be a great difference, but only a difference in degree, not in form.—It is quite possible to conceive even that the assailant may choose a defensive form, the better to attain his object; for instance, that he may choose a strong position, that he may be attacked there; but such instances are so rare that we do not think it necessary to dwell upon them in our grouping of ideas and facts, which are always founded on the practical. We may, therefore, say that there are no such gradations in the attack as those which present themselves in the defence.

Lastly, as a rule, the extent of the means of attack consists of the armed force only; of course, we must add to these the fortresses, for if in the vicinity of the theatre of war, they have a decided influence on the attack. But this influence gradually diminishes as the attack advances; and it is conceivable that, in the attack, its own fortresses never can play such an important part as in the defence, in which they often become objects of primary importance. The assistance of the people may be supposed in co-operation with the attack, in those cases in which the inhabitants of the country are better disposed towards the invader of the country than they are to their own army; finally, the assailant may also have allies, but then they are only the result of special or accidental relations, not an assistance proceeding from the nature of the aggressive. Although, therefore, in speaking of the defence we have reckoned fortresses, popular insurrections, and allies as available means of resistance; we cannot do the same in the attack; there they belong to the nature of the thing; here they only appear rarely, and for the most part accidentally.