A Interdependence of the Parts in a War

According as we have in view the absolute form of war or one of the real forms deviating more or less from it, so likewise different notions of its result will arise.

In the absolute form, where everything is the effect of its natural and necessary cause, one thing follows another in rapid succession; there is, if we may use the expression, no neutral space; there is -- on account of the manifold reactionary effects which war contains in itself, on account of the connection in which, strictly speaking, the whole series of combats follow one after another, on account of the culminating point which every victory has, beyond which losses and defeats commence -- on account of all these natural relations of war there is, I say, only one result, to wit, the final result. Until it takes place nothing is decided, nothing won, nothing lost. Here we may say indeed: the end crowns the work. In this view, therefore, war is an indivisible whole, the parts of which (the subordinate results) have no value except in their relation to this whole. [ ... ]

To this view of the relative connection of results in war, which may be regarded as extreme, stands opposed another extreme, according to which war is composed of single independent results, in which, as in any number of games played, the preceding has no influence on the next following; everything here, therefore, depends only on the sum total of the results, and we can lay up each single one like a counter at play.

Just as the first kind of view derives its truth from the nature of things, so we find that of the second in history. There are cases without number in which a small moderate advantage might have been gained without any very onerous condition being attached to it. The more the element of war is modified the more common these cases become; but as little as the first of the views now imagined was ever completely realised in any war, just as little is there any war in which the last suits in all respects, and the first can be dispensed with.

If we keep to the first of these supposed views, we must perceive the necessity of every war being looked upon as a whole from the very commencement, and that at the very first step forwards, the commander should have in his eye the object to which every line must converge.

If we admit the second view, then subordinate advantages may be pursued on their own account, and the rest left to subsequent events.

As neither of these forms of conception is entirely without result, therefore theory cannot dispense with either. But it makes this difference in the use of them, that it requires the first to be laid as a fundamental idea at the root of everything, and that the latter shall only be used as a modification which is justified by circumstances.
[ ... ]

Theory demands, therefore, that at the commencement of every war its character and main outline shall be defined according to what the political conditions and relations lead us to anticipate as probable. The more that, according to this probability, its character approaches the form of absolute war; the more its outline embraces the mass of the belligerent states and draws them into the vortex, so much the more complete will be the relation of events to one another and the whole, but so much the more necessary will it also be not to take the first step without thinking what may be the last.