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 XXIV

Operating Against a Flank

WE need hardly observe that we speak of the strategic flank, that is, a side of the theatre of war, and that the attack from one side in battle, or the tactical movement against a flank, must not be confounded with it; and even in cases in which the strategic operation against a flank, in its last stage, ends in the tactical operation, they can quite easily be kept separate, because the one never follows necessarily out of the other.

These flanking movements, and the flanking positions connected with them, belong also to the mere useless pageantry of theory, which is seldom met with in actual war. Not that the means itself is either ineffectual or illusory, but because both sides generally seek to guard themselves against its effects; and cases in which this is impossible are rare. Now in these uncommon cases this means has often also proved highly efficacious, and for this reason, as well as on account of the constant watching against it which is required in war, it is important that it should be clearly explained in theory. Although the strategic operation against a flank can naturally be imagined, not only on the part of the defensive, but also on that of the offensive, still it has much more affinity with the first, and therefore finds its place under the head of defensive means.

Before we enter into the subject, we must establish the simple principle, which must never be lost sight of afterwards in the consideration of the subject, that troops which are to act against the rear or flank of the enemy cannot be employed against his front, and that, therefore, whether it be in tactics or strategy, it is a completely false kind of notion to consider that coming on the rear of the enemy is at once an advantage in itself. In itself, it is as yet nothing; but it will become something in connection with other things, and something either advantageous or the reverse, according to the nature of these things, the examination of which now claims our attention.

First, in the action against the strategic flank, we must make a distinction between two objects of that measure—between the action merely against the communications, and that against the line of retreat, with which, at the same time, an effect upon the communications may also be combined.

When Daun, in 1758, sent a detachment to seize the convoys on their way to the siege of Olmutz, he had plainly no intention of impeding the king's retreat into Silesia; he rather wished to bring about that retreat, and would willingly have opened the line to him.

In the campaign of 1812, the object of all the expeditionary corps that were detached from the Russian army in the months of September and October, was only to intercept the communications, not to stop the retreat; but the latter was quite plainly the design of the Moldavian army which, under Tschitschagof, marched against the Beresina, as well as of the attack which General Wittgenstein was commissioned to make on the French corps stationed on the Dwina.

These examples are merely to make the exposition clearer.

The action against the lines of communication is directed against the enemy's convoys, against small detachments following in rear of the army, against couriers and travellers, small depôts, etc.; in fact, against all the means which the enemy requires to keep his army in a vigorous and healthy condition; its object is, therefore, to weaken the condition of the enemy in this respect, and by this means to cause him to retreat.

The action against the enemy's line of retreat is to cut his army off from that line. It cannot effect this object unless the enemy really determines to retreat; but it may certainly cause him to do so by threatening his line of retreat, and, therefore, it may have the same effect as the action against the line of communication, by working as a demonstration. But as already said, none of these effects are to be expected from the mere turning which has been effected, from the mere geometrical form given to the disposition of the troops, they only result from the conditions suitable to the same.

In order to learn more distinctly these conditions, we shall separate completely the two actions against the flank, and first consider that which is directed against the communications.

Here we must first establish two principal conditions, one or other of which must always be forthcoming.

The first is, that the forces used for this action against the flank of the enemy must be so insignificant in numbers that their absence is not observed in front.

The second, that the enemy's army has run its career, and therefore can neither make use of a fresh victory over our army, nor can he pursue us if we evade a combat by moving out of the way.

This last case, which is by no means so uncommon as might be supposed, we shall lay aside for the moment, and occupy ourselves with the accessory conditions of the first.

The first of these is, that the communications have a certain length, and cannot be protected by a few good posts; the second point is, that the situation of the line is such as exposes it to our action.

This weakness of the line may arise in two ways—either by its direction, if it is not perpendicular to the strategic front of the enemy's army, or because his lines of communication pass through our territory; if both these circumstances exist, the line is so much the more exposed. These two relations require a closer examination.

One would think that when it is a question of covering a line of communication forty or fifty miles long, it is of little consequence whether the position occupied by an army standing at one extremity of this line forms an oblique angle or a right angle in reference to it, as the breadth of the position is little more than a mere point in comparison to the line; and yet it is not so unimportant as it may seem. When an army is posted at a right angle with its communications, it is difficult, even with a considerable superiority, to interrupt the communications by any detachments or partisans sent out for the purpose. If we think only of the difficulty of covering absolutely a certain space, we should not believe this, but rather suppose, on the contrary, that it must be very difficult for an army to protect its rear (that is, the country behind it) against all expeditions which an enemy superior in numbers may undertake. Certainly, if we could look at everything in war as it is on a sheet of paper! Then the party covering the line, in his uncertainty as to the point where light troops or partisans may appear, would be in a certain measure blind, and only the partisans would see. But if we think of the uncertainty and insufficiency of intelligence gained in war, and know that both parties are incessantly groping in the dark, then we easily perceive that a detached corps sent round the enemy's flank to gain his rear is in the position of a man engaged in a fray with numbers in a dark room. In the end he must fall; and so must it also be with bands who get round an army occupying a perpendicular position, and who therefore place themselves near to the enemy, but widely separated from their own people. Not only is there danger of losing numbers in this way; there is also a risk of the whole instrument itself being blunted immediately; for the very first misfortune which happens to one such party will make all the others timid, and instead of bold attacks and insolent dodging, the only play will be constant running away.

Through this difficulty, therefore, an army occupying a perpendicular position covers the nearest points on its line of communications for a distance of two or three marches, according to the strength of the army; but those nearest points are just those which are most in danger, as they are the nearest to the enemy.

On the other hand, in the case of a decidedly oblique position, no such part of the line of communication is covered; the smallest pressure, the most insignificant attempt on the part of the enemy, leads at once to a vulnerable point.

But now, what is it which determines the front of a position, if it is not just the direction perpendicular to the line of communication? The front of the enemy; but then, again, this may be equally as well supposed as dependent on our front. Here there is a reciprocal effect, for the origin of which we must search.

If we suppose the lines of communication of the assailant, a b, so situated with respect to those of the enemy, c d, that the two lines form a considerable angle with each other, it is evident that if the defensive wishes to take up a position at e, where the two lines intersect, the assailant from b, by the mere geometrical relation, could compel him to form front opposite to him, and thus to lay bare his communications. The case would be reversed if the defensive took up his position on this side of the point of junction, about d; then the assailant must make front towards him, if so be that his line of operations, which closely depends on geographical conditions, cannot be arbitrarily changed, and moved, for instance, to the direction a d. From this it would seem to follow that the defender has an advantage in this system of reciprocal action, because he only requires to take a position on this side of the intersection of the two lines. But very far from attaching any importance to this geometrical element, we only brought it into consideration to make ourselves the better understood; and we are rather of opinion that local and generally individual relations have much more to do with determining the position of the defender; that, therefore, it is quite impossible to lay down in general which of two belligerents will be obliged soonest to expose his communications.

If the lines of communication of both sides lie in one and the same direction, then whichever of the two parties takes up an oblique position will certainly compel his adversary to do the same. But then there is nothing gained geometrically by this, and both parties attain the same advantages and disadvantages.

In the continuation of our considerations we shall, therefore, confine ourselves to the case of the line of communication of one side only being exposed.

Now as regards the second disadvantageous relation of a line of communication, that is to say, when it runs through an enemy's country, it is clear in itself how much the line is compromised by that circumstance, if the inhabitants of the country have taken up arms; and consequently the case must be looked at as if a body of the enemy was posted all along the line; this body, it is true, is in itself weak without solidity or intensive force; but we must also take into consideration what the close contact and influence of such a hostile force may nevertheless effect through the number of points which offer themselves one after another on long lines of communication. That requires no further explanation. But even if the enemy's subjects have not taken up arms, and even if there is no militia in the country, or other military organisation, indeed if the people are even very unwarlike in spirit, still the mere relation of the people as subjects to a hostile government is a disadvantage for the lines of communication of the other side which is always felt. The assistance which expeditionary forces and partisans derive merely through a better understanding with the people, through a knowledge of the country and its inhabitants, through good information, through the support of official functionaries, is, for them, of decided value; and this support every such body will enjoy without any special effort on its own part. Added to this, within a certain distance there will not be wanting fortresses, rivers, mountains, or other places of refuge, which of ordinary right belong to the enemy, if they have not been formally taken possession of and occupied by our troops.

Now in such a case as is here supposed, especially if attended with other favourable circumstances, it is possible to act against the communications of an army, although their direction is perpendicular to the position of that army; for the detachments employed for the purpose do not then require to fall back always on their own army, because being in their own country they are safe enough if they only make their escape.

We have, therefore, now ascertained that—

1. A considerable length,

2. An oblique direction,

3. An enemy's province,

are the principal circumstances under which the lines of communication of an army may be interrupted by a relatively small proportion of armed forces on the side of the enemy; in order to make this interruption effectual, a fourth condition is still requisite, which is a certain duration of time. Respecting this point, we beg attention to what has been said in the fifteenth chapter of the fifth book.

But these four conditions are only the chief points which relate to the subject; a number of local and special circumstances attach themselves to these, and often attain to an influence more decisive and important than that of the principal ones themselves. Selecting only the most essential, we mention the state of the roads, the nature of the country through which they pass, the means of cover which are afforded by rivers, mountains, and morasses, the seasons and weather, the importance of particular convoys, such as siege trains, the number of light troops, etc., etc.

On all these circumstances, therefore, will depend the effect with which a general can act on his opponent's communications; and by comparing the result of the whole of these circumstances on the one side with the result of the whole on the other, we obtain a just estimate of the relative advantages of both systems of communication, on which will depend which of the two generals can play the highest game.

What here seems so prolix in the explanation is often decided in the concrete case at first sight; but still, the tact of a practised judgment is required for that, and person must have thought over every one of the cases now developed in order to see in its true light the absurdity of those critical writers who think they have settled something by the mere words "turning" and "acting on a flank," without giving their reasons.

We now come to the second chief condition, under which the strategic action against the enemy's flank may take place.

If the enemy is hindered from advancing by any other cause but the resistance which our army opposes, let that cause be what it may, then our army has no reason to be apprehensive about weakening itself by sending out detachments to harass the enemy; for if the enemy should attempt to chastise us by an attack, we have only to yield some ground and decline the combat. This is what was done by the chief Russian army at Moscow in 1812. But it is not at all necessary that everything should be again on the same great scale as in that campaign for such a case to happen again. In the first Silesian war, Frederick the Great was each time in this situation, on the frontiers of Bohemia and Moravia, and in the complex affairs relating to generals and their armies, many causes of different kinds, particularly political ones, may be imagined, which make further advance an impossibility.

As in the case now supposed more forces may be spared to act against the enemy's flank, the other conditions need not be quite so favourable: even the nature of our communications in relation to those of the enemy need not give us the advantage in that respect, as an enemy who is not in a condition to make any particular use of our further retreat is not likely to use his right to retaliate, but will rather be anxious about the direct covering of his own line of retreat.

Such a situation is therefore very well suited to obtain for us, by means less brilliant and complete but less dangerous than a victory, those results which it would be too great a risk to seek to obtain by a battle.

As in such a case we feel little anxiety about exposing our own line of communications, by taking up a position on one or other flank, and as the enemy by that means may always be comspelled to form front obliquely to his line of communications, therefore this one of the conditions above named will seldom fail to occur. The more the rest of the conditions, as well as other circumstances, co-operate, so much the more certain are we of success from the means now in question; but the fewer favourable circumstances exist, the more will all depend on superior skill in combination, and promptitude and precision in the execution.

Here is the proper field for strategic manœuvres, such as are to be found so frequently in the Seven Years' War, in Silesia and Saxony, and in the campaigns of 1760 and 1762. If, in many wars in which only a moderate amount of elementary force is displayed, such strategic manœuvring very often appears, this is not because the commander on each occasion found himself at the end of his career, but because want of resolution and courage, and of an enterprising spirit, and dread of responsibility, have often supplied the place of real impediments; for a case in point, we have only to call to mind Field Marshal Daun.

As a summary of the results of our considerations, we may say, that the action against a flank is most effectual—

1. In the defensive;

2. Towards the end of a campaign;

3. Above all, in a retreat into the heart of the country; and

4. In connection with a general arming of the people.

On the mode of executing this action against the communications, we have only a few words to say.

The enterprises must be conducted by skilful detachment leaders, who, at the head of small bodies, by bold marches and attacks, fall upon the enemy's weak garrisons, convoys, and small detachments on the march here and there, encourage the national levies (landsturm), and sometimes join with them in particular undertakings. These parties must be more numerous than strong individually, and so organised that it may be possible to unite several of them for any greater undertaking without any obstacle from the vanity or caprice of any of the single leaders.

We have now to speak of the action against the enemy's line of retreat.

Here we must keep in view, above all things, the principle with which we commenced, that forces destined to operate in rear cannot be used in front; that, therefore, the action against the rear or flanks is not an increase of force in itself; it is only to be regarded as a more powerful application (or employment) of the same; increasing the degree of success in prospect, but also increasing the degree of risk.

Every opposition offered with the sword which is not of a direct and simple nature, has a tendency to raise the result at the cost of its certainty. An operation against the enemy's flank, whether with one compact force, or with separate bodies converging from several quarters, belongs to this category.

But now, if cutting off the enemy's retreat is not to be a mere demonstration, but is seriously intended, the real solution is a decisive battle, or, at least, the conjunction of all the conditions for the same; and just in this solution we find again the two elements above-mentioned—the greater result and the greater danger. Therefore, if a general is to stand justified in adopting this method of action, his reasons must be favourable conditions.

In this method of resistance we must distinguish the two forms already mentioned. The first is, if a general with his whole force intends to attack the enemy in rear, either from a position taken up on the flank for that purpose, or by a formal turning movement; the second is, if he divides his forces, and, by an enveloping position with one part, threatens the enemy's rear, with the other part his front.

The result is intensified in both cases alike, that is—either there is a real interception of the retreat, and consequently the enemy's army taken prisoners, or the greater part scattered, or there may be a long and hasty retreat of the enemy's force to escape the danger.

But the intensified risk is different in the two cases.

If we turn the enemy with our whole force, the danger lies in the laying open our own rear; and hence the question again depends on the relation of the mutual lines of retreat, just as in the action against the lines of communication, it depended on the relation of those lines.

Now certainly the defender, if he is in his own country, is less restricted than the assailant, both as to his lines of retreat and communication, and in so far is therefore in a better position to turn his adversary strategically; but this general relation is not of a sufficiently decisive character to be used as the foundation of a practical method; therefore, nothing but the whole of the relations in each individual case can decide.

Only so much we may add, that favourable conditions are naturally more common in wide spheres of action than in small; more common, also, on the side of independent states than on that of weak ones, dependent on foreign aid, and whose armies must, therefore, constantly have their attention bent on the point of junction with the auxiliary army; lastly, they become most favorable for the defender towards the close of the campaign, when the impulsive force of the assailant is somewhat spent; very much, again, in the same manner as in the case of the lines of communication.

Such a flank position as the Russians took up with such advantage on the road from Moscow to Kaluga, when Buonaparte's aggressive force was spent, would have brought them into a scrape at the commencement of the campaign at the camp of Drissa, if they had not been wise enough to change their plan in good time.

The other method of turning the enemy, and cutting off his retreat by dividing our force, entails the risk attending a division of our own force, whilst the enemy, having the advantage of interior lines, retains his forces united, and therefore has the power of acting with superior numbers against one of our divisions. This is a disadvantage which nothing can remove, and in exposing ourselves to it, we can only be justified by one of three principal reasons:—

1. The original division of the force which makes such a method of action necessary, unless we incur a great loss of time.

2. A great moral and physical superiority, which justifies the adoption of a decisive method.

3. The want of impulsive force in the enemy as soon as he has arrived at the culminating point of his career.

When Frederick the Great invaded Bohemia, 1757, on converging lines, he had not in view to combine an attack in front with one on the strategic rear, at all events, this was by no means his principal object, as we shall more fully explain elsewhere, but in any case it is evident that there never could have been any question of a concentration of forces in Silesia or Saxony before the invasion, as he would thereby have sacrificed all the advantages of a surprise.

When the allies formed their plan for the second part of the campaign of 1813, looking to their great superiority in numbers, they might very well at that time entertain the idea of attacking Buonaparte's right on the Elbe with their main force, and of thus shifting the theatre of war from the Oder to the Elbe. Their ill-success at Dresden is to be ascribed not to this general plan but to their faulty dispositions both strategic and tactical. They could have concentrated 220,000 men at Dresden against Buonaparte's 130,000, a proportion of numbers eminently favourable (at Leipsic, at least, the proportion was as 285 : 157). It is true that Buonaparte had distributed his forces too evenly for the particular system of a defence upon one line (in Silesia 70,000 against 90,000, in the Mark—Brandenburg—70,000 against 110,000), but at all events it would have been difficult for him, without completely abandoning Silesia, to assemble on the Elbe a force which could have contended with the principal army of the allies in a decisive battle. The allies could also have easily called up the army of Wrede to the Maine, and employed it to try to cut Buonaparte off from the road to Mayence.

Lastly, in 1812, the Russians might have directed their army of Moldavia upon Volhynia and Lithuania in order to move it forward afterwards against the rear of the principal French army, because it was quite certain that Moscow must be the extreme point of the French line of operations. For any part of Russia beyond Moscow there was nothing to fear in that campaign, therefore the Russian main army had no cause to consider itself too weak.

This same scheme formed part of the disposition of the forces laid down in the first defensive plan proposed by General Phul, according to which the army of Barclay was to occupy the camp at Drissa, whilst that under Bragathion was to press forward against the rear of the main French army. But what a difference of circumstances in the two cases! In the first of them the French were three times as strong as the Russians; in the second, the Russians were decidedly superior. In the first, Buonaparte's great army had in it an impulsive force which carried it to Moscow 80 miles beyond Drissa: in the second, it is unfit to make a day's march beyond Moscow; in the first, the line of retreat on the Niemen did not exceed 30 miles: in the second it was 112. The same action against the enemy's retreat therefore, which was so successful in the second case, would, in the first, have been the wildest folly.

As the action against the enemy's line of retreat, if it is more than a demonstration, becomes a formal attack from the rear, there remains therefore still a good deal to be said on the subject, but it will come in more appropriately in the book upon the attack; we shall therefore break off here and content ourselves with having given the conditions under which this kind of reaction may take place.

Very commonly the design of causing the enemy to retreat by menacing his line of retreat, is understood to imply rather a mere demonstration than the actual execution of the threat. If it was necessary that every efficacious demonstration should be founded on the actual practicability of real action, which seems a matter of course at first sight, then it would accord with the same in all respects. But this is not the case: on the contrary, in the chapter on demonstrations we shall see that they are connected with conditions somewhat different, at all events in some respects, we therefore refer our readers to that chapter.