NOTE: This version of Carl von Clausewitz's On War is the long-obsolete J.J. Graham translation of Clausewitz's Vom Kriege (1832) published in London in 1873. The 1976/84 Howard/Paret version is the standard translation today; for the most accurate text one should always consult the 1943 Jolles translation. Consider the more modern versions and other relevant books shown below.
On Waterloo: Clausewitz, Wellington, and the Campaign of 1815. Ed./trans. Christopher Bassford, Daniel Moran, and Gregory W. Pedlow (ClausewitzStudies.org, 2010). ISBN: 1453701508. This book is built around a new and complete translation of Clausewitz's study of the Waterloo campaign [Berlin: 1835], which is a strategic analysis of the entire campaign (not just the Battle of Waterloo), and the Duke of Wellington's detailed 1842 response to it.
Buy the best translation—recommended for serious readers. The Book of War (The Modern Library, February 2000). ISBN: 0375754776. Clausewitz's On War and Sun Tzu's Art of War in one volume. The translation of Clausewitz's On War is the 1943 version done by German literary scholar O.J. Matthijs Jolles at the University of Chicago during World War II—not today's standard translation, but certainly the most accurate.
Buy the standard English translation of Clausewitz's On War, by Michael Howard and Peter Paret (Princeton University Press, 1976/84). ISBN: 0691018545 (paperback). Kindle edition. This quite readable translation appeared at the close of the Vietnam War and—principally for marketing and copyright reasons—has become the modern standard.
Vanya Eftimova Bellinger, Marie von Clausewitz: The Woman Behind the Making of On War (Oxford University Press, 2015), ISBN: 0190225432. A rich biography of Countess Marie von Clausewitz that also sheds enormous light on the life, ideas, influences upon, and character of the great military thinker himself.
BOOK 6 • CHAPTER 25
Retreat into the Interior of the Country
WE have considered the voluntary retreat into the heart of the country as a particular indirect form of defence through which it is expected the enemy will be destroyed, not so much by the sword as by exhaustion from his own efforts. In this case, therefore, a great battle is either not supposed, or it is assumed to take place when the enemy's forces are considerably reduced.
Every assailant in advancing diminishes his military strength by the advance; we shall consider this more in detail in the seventh book; here we must assume that result which we may the more readily do as it is clearly shown by military history in every campaign in which there has been a considerable advance.
This loss in the advance is increased if the enemy has not been beaten, but withdraws of his own accord with his forces intact, and offering a steady continuous resistance, sells every step of ground at a bloody price, so that the advance is a continuous combat for ground and not a mere pursuit.
On the other hand, the losses which a party on the defensive suffers on a retreat, are much greater if his retreat has been preceded by a defeat in battle than if his retreat is voluntary. For if he is able to offer the pursuer the daily resistance which we expect on a voluntary retreat, his losses would be at least the same in that way, over and above which those sustained in the battle have still to be added. But how contrary to the nature of the thing such a supposition as this would be! The best army in the world if obliged to retire far into the country after the loss of a battle, will suffer losses on the retreat, beyond measure out of proportion; and if the enemy is considerably superior, as we suppose him, in the case of which we are now speaking, if he pursues with great energy as has almost always been done in modern wars, then there is the highest probability that a regular flight takes place by which the army is usually completely ruined.
A regularly measured daily resistance, that is, one which each time only lasts as long as the balance of success in the combat can be kept wavering, and in which we secure ourselves from defeat by giving up the ground which has been contested at the right moment, will cost the assailant at least as many men as the defender in these combats, for the loss which the latter by retiring now and again must unavoidably suffer in prisoners, will be balanced by the losses of the other under fire, as the assailant must always fight against the advantages of the ground. It is true that the retreating side loses entirely all those men who are badly wounded, but the assailant likewise loses all his in the same case for the present, as they usually remain several months in the hospitals.
The result will be that the two armies will wear each other away in nearly equal proportions in these perpetual collisions.
It is quite different in the pursuit of a beaten army. Here the troops lost in battle, the general disorganisation, the broken courage, the anxiety about the retreat, make such a resistance on the part of the retreating army very difficult, in many cases impossible; and the pursuer who, in the former case, advances extremely cautiously, even hesitatingly, like a blind man, always groping about, presses forward in the latter case with the firm tread of the conqueror, with the overweening spirit which good fortune imparts, with the confidence of a demigod, and the more daringly he urges the pursuit so much the more he hastens on things in the direction which they have already taken, because here is the true field for the moral forces which intensify and multiply themselves without being restricted to the rigid numbers and measures of the physical world.
It is therefore very plain how different will be the relations of two armies according as it is by the first or the second of the above ways, that they arrive at that point which may be regarded as the end of the assailant's course.
This is merely the result of the mutual destruction; to this must now be added the reductions which the advancing party suffers otherwise in addition, and respecting which, as already said, we refer to the seventh book; further, on the other hand, we have to take into account reinforcements which the retreating party receives in the great majority of cases, by forces subsequently joining him either in the form of help from abroad or through persistent efforts at home.
Lastly, there is, in the means of subsistence, such a disproportion between the retreating side and the advancing, that the first not uncommonly lives in superfluity when the other is reduced to want.
The army in retreat has the means of collecting provisions everywhere, and he marches towards them, whilst the pursuer must have everything brought after him, which, as long as he is in motion, even with the shortest lines of communication, is difficult, and on that account begets scarcity from the very first.
All that the country yields will be taken for the benefit of the retreating army first, and will be mostly consumed. Nothing remains but wasted villages and towns, fields from which the crops have been gathered, or which are trampled down, empty wells, and muddy brooks.
The pursuing army, therefore, from the very first day, has frequently to contend with the most pressing wants. On taking the enemy's supplies he cannot reckon; it is only through accident, or some unpardonable blunder on the part of the enemy, that here and there some little falls into his hands.
Thus there can be no doubt that in countries of vast dimensions, and when there is no extraordinary disproportion between the belligerent powers, a relation may be produced in this way between the military forces, which holds out to the defensive an immeasurably greater chance of a final result in his favour than he would have had if there had been a great battle on the frontier. Not only does the probability of gaining a victory become greater through this alteration in the proportions of the contending armies, but the prospects of great results from the victory are increased as well, through the change of position. What a difference between a battle lost close to the frontier of our country and one in the middle of the enemy's country! Indeed, the situation of the assailant is often such at the end of his first start, that even a battle gained may force him to retreat, because he has neither enough impulsive power left to complete and make use of a victory, nor is he in a condition to replace the forces he has lost.
There is, therefore, an immense difference between a decisive blow at the commencement and at the end of the attack.
To the great advantage of this mode of defence are opposed two drawbacks. The first is the loss which the country suffers through the presence of the enemy in his advance, the other is the moral impression.
To protect the country from loss can certainly never be looked upon as the object of the whole defence. That object is an advantageous peace. To obtain that as surely as possible is the endeavour, and for it no momentary sacrifice must he considered too great. At the same time, the above loss, although it may not be decisive, must still be laid in the balance, for it always affects our interests.
This loss does not affect our army directly; it only acts upon it in a more or less roundabout way, whilst the retreat itself directly reinforces our army. It is, therefore, difficult to draw a comparison between the advantage and disadvantage in this case; they are things of a different kind, the action of which is not directed towards any common point. We must, therefore, content ourselves with saying that the loss is greater when we have to sacrifice fruitful provinces well populated, and large commercial towns; but it arrives at a maximum when at the same time we lose war-means either ready for use or in course of preparation.
The second counterpoise is the moral impression. There are cases in which the commander must be above regarding such a thing, in which he must quietly follow out his plans, and run the risk of the objections which short-sighted despondency may offer; but nevertheless, this impression is no phantom which should be despised. It is not like a force which acts upon one point: but like a force which, with the speed of lightning, penetrates every fibre, and paralyses all the powers which should be in full activity, both in a nation and in its army. There are indeed cases in which the cause of the retreat into the interior of the country is quickly understood by both nation and army, and trust, as well as hope, are elevated by the step; but such cases are rare. More usually, the people and the army cannot distinguish whether it is a voluntary movement or a precipitate retreat, and still less whether the plan is one wisely adopted, with a view to ensure ulterior advantages, or the result of fear of the enemy's sword. The people have a mingled feeling of sympathy and dissatisfaction at seeing the fate of the provinces sacrificed; the army easily loses confidence in its leaders, or even in itself, and the constant combats of the rear-guard during the retreat, tend always to give new strength to its fears. These are consequences of the retreat about which we must never deceive ourselves. And it certainly is—considered in itself—more natural, simpler, nobler, and more in accordance with the moral existence of a nation, to enter the lists at once, that the enemy may out cross the frontiers of its people without being opposed by its genius, and being called to a bloody account.
These are the advantages and disadvantages of this kind of defence; now a few words on its conditions and the circumstances which are in its favour.
A country of great extent, or at all events, a long line of retreat, is the first and fundamental condition; for an advance of a few marches will naturally not weaken the enemy seriously. Buonaparte's centre, in the year 1812, at Witepsk, was 250,000 strong, at Smolensk, 182,000, at Borodino it had only diminished to 130,000, that is to say, had fallen to about an equality with the Russian centre. Borodino is ninety miles from the frontier; but it was not until they came near Moscow that the Russians reached that decided superiority in numbers, which of itself reversed the situation of the combatants so assuredly, that the French victory at Malo Jaroslewetz could not essentially alter it again.
No other European state has the dimensions of Russia, and in very few can a line of retreat 100 miles long be imagined. But neither will a power such as that of the French in 1812, easily appear under different circumstances, still less such a superiority in numbers as existed at the commencement of the campaign, when the French army had more than double the numbers of its adversary, besides its undoubted moral superiority. Therefore, what was here only effected at the end of 100 miles, may perhaps, in other cases, be attained at the end of 50 or 30 miles.
The circumstances which favour this mode of defence are—
1. A country only little cultivated.
2. A loyal and warlike people.
3. An inclement season.
All these things increase the difficulty of maintaining an army, render great convoys necessary, many detachments, harassing duties, cause the spread of sickness, and make operations against the flanks easier for the defender.
Lastly, we have yet to speak of the absolute mass alone of the armed force, as influencing the result.
It lies in the nature of the thing itself that, irrespective of the mutual relation of the forces opposed to each other, a small force is sooner exhausted than a larger, and, therefore, that its career cannot be so long, nor its theatre of war so wide. There is, therefore, to a certain extent, a constant relation between the absolute size of an army and the space which that army can occupy. It is out of the question to try to express this relation by any figures, and besides, it will always be modified by other circumstances; it is sufficient for our purpose to say that these things necessarily have this relation from their very nature. We may be able to march upon Moscow with 500,000 but not with 50,000, even if the relation of the invader's army to that of the defender in point of numbers were much more favourable in the latter case.
Now if we assume that there is this relation of absolute power to space in two different cases, then it is certain that the effect of our retreat into the interior in weakening the enemy will increase with the masses.
1. Subsistence and lodging of the troops become more difficult—for, supposing the space which an army covers to increase in proportion to the size of the army, still the subsistence for the army will never be obtainable from this space alone, and everything which has to be brought after an army is subject to greater loss also; the whole space occupied is never used for covering for the troops, only a small part of it is required, and this does not increase in the same proportion as the masses.
2. The advance is in the same manner more tedious in proportion as the masses increase, consequently, the time is longer before the career of aggression is run out, and the sum total of the daily losses is greater.
Three thousand men driving two thousand before them in an ordinary country, will not allow them to march at the rate of 1, 2, or at most 3 miles a day, and from time to time to make a few days' halt. To come up with them, to attack them, and force them to make a further retreat is the work of a few hours; but if we multiply these masses by 100, the case is altered. Operations for which a few hours sufficed in the first case, require now a whole day, perhaps two. The contending forces cannot remain together near one point; thereby, therefore, the diversity of movements and combinations increases, and, consequently, also the time required. But this places the assailant at a disadvantage, because his difficulty with subsistence being greater, he is obliged to extend his force more than the pursued, and, therefore, is always in danger of being overpowered by the latter at some particular point, as the Russians tried to do at Witepsk.
3. The greater the masses are, the more severe are the exertions demanded from each individual for the daily duties required strategically and tactically. A hundred thousand men who have to march to and from the point of assembly every day, halted at one time, and then set in movement again, now called to arms, then cooking or receiving their rations—a hundred thousand who must not go into their bivouac until the necessary reports are delivered in from all quarters—these men, as a rule, require for all these exertions connected with the actual march, twice as much time as 50,000 would require, but there are only twenty-four hours in the day for both. How much the time and fatigue of the march itself differs according to the size of the body of troops to be moved, has been shown in the ninth chapter of the preceding book. Now, the retreating army, it is true, partakes of these fatigues as well as the advancing, but they are much greater for the latter:—
1. because the mass of his troops is greater on account of the superiority which we supposed,
2. because the defender, by being always the party to yield ground, purchases by this sacrifice the right of the initiative, and, therefore, the right always to give the law to the other. He forms his plan beforehand, which, in most cases, he can carry out unaltered, but the aggressor, on the other hand, can only make his plans conformably to those of his adversary, which he must in the first instance find out.
We must, however, remind our readers that we are speaking of the pursuit of an enemy who has not suffered a defeat, who has not even lost a battle. It is necessary to mention this, in order that we may not be supposed to contradict what was said in the twelfth chapter of our fourth book.
But this privilege of giving the law to the enemy makes a difference in saving of time, expenditure of force, as well as in respect of other minor advantages which, in the long run, becomes very important.
3. because the retreating force on the one hand does all he can to make his own retreat easy, repairs roads, and bridges, chooses the most convenient places for encampment, etc., and, on the other hand again, does all he can to throw impediments in the way of the pursuer, as he destroys bridges, by the mere act of marching makes bad roads worse, deprives the enemy of good places for encampment by occupying them himself, etc.
Lastly, we must add still, as a specially favourable circumstance, the war made by the people. This does not require further examination here, as we shall allot a chapter to the subject itself.
Hitherto, we have been engaged upon the advantages which such a retreat ensures, the sacrifices which it requires, and the conditions which must exist; we shall now say something of the mode of executing it.
The first question which we have to propose to ourselves is with reference to the direction of the retreat.
It should be made into the interior of the country, therefore, if possible, towards a point where the enemy will be surrounded on all sides by our provinces; there he will be exposed to their influence, and we shall not be in danger of being separated from the principal mass of our territory, which might happen if we chose a line too near the frontier, as would have happened to the Russians in 1812 if they had retreated to the south instead of the east.
This is the condition which lies in the object of the measure itself. Which point in the country is the best, how far the choice of that point will accord with the design of covering the capital or any other important point directly, or drawing the enemy away from the direction of such important places depends on circumstances.
If the Russians had well considered their retreat in 1812 beforehand, and, therefore, made it completely in conformity with a regular plan, they might easily, from Smolensk, have taken the road to Kaluga, which they only took on leaving Moscow; it is very possible that under these circumstances Moscow would have been entirely saved.
That is to say, the French were about 130,000 strong at Borodino, and there is no ground for assuming that they would have been any stronger if this battle had been fought by the Russians half way to Kaluga instead; now, how many of these men could they have spared to detach to Moscow? Plainly, very few; but it is not with a few troops that an expedition can be sent a distance of fifty miles (the distance from Smolensk to Moscow) against such a place as Moscow.
Supposing Buonaparte when at Smolensk, where he was 160,000 strong, had thought he could venture to detach against Moscow before engaging in a great battle, and had used 40,000 men for that purpose, leaving 120,000 opposite the principal Russian army, in that case, these 120,000 men would not have been more than 90,000 in the battle, that is 40,000 less than the number which fought at Borodino; the Russians, therefore, would have had a superiority in numbers of 30,000 men. Taking the course of the battle of Borodino as a standard, we may very well assume that with such a superiority they would have been victorious. At all events, the relative situation of the parties would have been more favourable for the Russians than it was at Borodino. But the retreat of the Russians was not the result of a well-matured plan; they retreated as far as they did because each time that they were on the point of giving battle they did not consider themselves strong enough yet for a great action; all their supplies and reinforcements were on the road from Moscow to Smolensk, and it could not enter the head of anyone at Smolensk to leave that road. But, besides, a victory between Smolensk and Kaluga would never have excused, in the eyes of the Russians, the offence of having left Moscow uncovered, and exposed it to the possibility of being captured.
Buonaparte, in 1813, would have secured Paris with more certainty from an attack if he had taken up a position at some distance in a lateral direction, somewhere behind the canal of Burgundy, leaving only with the large force of National Guard in Paris a few thousand regular troops. The allies would never have had the courage to march a corps of 50,000 or 60,000 against Paris whilst Buonaparte was in the field at Auxerre with 100,000 men. If the case is supposed reversed, and the allies in Buonaparte's place, then no one, indeed, would have advised them to leave the road open to their own capital with Buonaparte for their opponent. With such a preponderance he would not have hesitated a moment about marching on the capital. So different is the effect under the same circumstances but under different moral relations.
As we shall have hereafter to return to this subject when treating of the plan of a war, we shall only at present add that, when such a lateral position is taken, the capital or place which it is the object to protect, must, in every case, be capable of making some resistance that it may not be occupied and laid under contribution by every flying column or irregular band.
But we have still to consider another peculiarity in the direction of such a line of retreat, that is, a sudden change of direction. After the Russians had kept the same direction as far as Moscow they left that direction which would have taken them to Wladimir, and after first taking the road to Riazan for some distance, they then transferred their army to the Kaluga road. If they had been obliged to continue their retreat they could easily have done so in this new direction which would have led them to Kiew, therefore much nearer again to the enemy's frontier. That the French, even if they had still preserved a large numerical superiority over the Russians, could not have maintained their line of communication by Moscow under such circumstances is clear in itself; they must have given up not only Moscow but, in all probability, Smolensk also, therefore have again abandoned the conquests obtained with so much toil, and contented themselves with a theatre of war on this side the Beresina.
Now, certainly, the Russian army would thus have got into the same difficulty to which it would have exposed itself by taking the direction of Kiew at first, namely, that of being separated from the mass of its own territory; but this disadvantage would now have become almost insignificant, for how different would have been the condition of the French army if it had marched straight upon Kiew without making the detour by Moscow.
It is evident that such a sudden change of direction of a line of retreat, which is very practicable in a spacious country, ensures remarkable advantages.
1. It makes it impossible for the enemy (the advancing force) to maintain his old line of communication: but the organisation of a new one is always a difficult matter, in addition to which the change is made gradually, therefore, probably, he has to try more than one new line.
2. If both parties in this manner approach the frontier again; the position of the aggressor no longer covers his conquests, and he must in all probability give them up.
Russia with its enormous dimensions, is a country in which two armies might in this manner regularly play at prisoners' base (Zeck jagen).
But such a change of the line of retreat is also possible in smaller countries, when other circumstances are favourable, which can only be judged of in each individual case, according to its different relations.
When the direction in which the enemy is to be drawn into the country is once fixed upon, then it follows of itself that our principal army should take that direction, for otherwise the enemy would not advance in that direction, and if he even did we should not then be able to impose upon him all the conditions above supposed. The question then only remains whether we shall take this direction with our forces undivided, or whether considerable portions should spread out laterally and therefore give the retreat a divergent (eccentric) form.
To this we answer that this latter form in itself is to be rejected.
1. Because it divides our forces, whilst their concentration on one point is just one of the chief difficulties for the enemy.
2. Because the enemy gets the advantage of operating on interior lines, can remain more concentrated than we are, consequently can appear in so much the greater force at any one point. Now certainly this superiority is less to be dreaded when we are following a system of constantly giving way; but the very condition of this constantly yielding, is always to continue formidable to the enemy and not to allow him to beat us in detail, which might easily happen. A further object of such a retreat, is to bring our principal force by degrees to a superiority of numbers, and with this superiority to give a decisive blow, but that by a partition of forces would become an uncertainty.
3. Because as a general rule the concentric (convergent) action against the enemy is not adapted to the weaker forces.
4. Because many disadvantages of the weak points of the aggression disappear when the defender's army is divided into separate parts.
The weakest features in a long advance on the part of the aggressor are for instance;—the length of the lines of communication, and the exposure of the strategic flanks. By the divergent form of retreat, the aggressor is compelled to cause a portion of his force to show a front to the flank, and this portion properly destined only to neutralise our force immediately in his front, now effects to a certain extent something else in addition, by covering a portion of the lines of communication.
For the mere strategic effect of the retreat, the divergent form is therefore not favourable; but if it is to prepare an action hereafter against the enemy's line of retreat, then we must refer to what has been said about that in the last chapter.
There is only one object which can give occasion to a divergent retreat, that is when we can by that means protect provinces which otherwise the enemy would occupy.
What sections of territory the advancing foe will occupy right and left of his course, can with tolerable accuracy be discerned by the point of assembly of, and directions given to, his forces, by the situation of his own provinces, fortresses, etc., in respect to our own. To place troops in those districts of territory which he will in all probability leave unoccupied, would be dangerous waste of our forces. But now whether by any disposition of our forces we shall be able to hinder him from occupying those districts which in all probability he will desire to occupy, is more difficult to decide, and it is therefore a point, the solution of which depends much on tact of judgment.
When the Russians retreated in 1812, they left 30,000 men under Tormassow in Volhynia, to oppose the Austrian force which was expected to invade that province. The size of the province, the numerous obstacles of ground which the country presents, the near proportion between the forces likely to come into conflict justified the Russians in their expectations, that they would be able to keep the upper hand in that quarter, or at least to maintain themselves near to their frontier. By this, very important advantages might have resulted in the sequel, which we shall not stop here to discuss; besides this, it was almost impossible for these troops to have joined the main army in time if they had wished. For these reasons, the determination to leave these troops in Volhynia to carry on there a distinct war of their own, was right. Now on the other hand, if according to the proposed plan of campaign submitted by General Phul, only the army of Barclay (80,000 men), was to retire to Drissa, and Bragathion's army (40,000 men) was to remain on the right flank of the French, with a view to subsequently falling on their rear, it is evident at once that this corps could not possibly maintain itself in South Lithuania so near to the rear of the main body of the French army, and would soon have been destroyed by their overwhelming masses.
That the defender's interest in itself is to give up as few provinces as possible to the assailant is intelligible enough, but this is always a secondary consideration; that the attack is also made more difficult the smaller or rather narrower the theatre of war is to which we can confine the enemy, is likewise clear in itself; but all this is subordinate to the condition that in so doing we have the probability of a result in our favour, and that the main body of the force on the defensive will not be too much weakened; for upon that force we must chiefly depend for the final solution, because the difficulties and distress suffered by the main body of the enemy, first call forth his determination to retreat, and increase in the greatest degree the loss of physical and moral power therewith connected.
The retreat into the interior of the country should therefore as a rule be made directly before the enemy, and as slowly as possible, with an army which has not suffered defeat and is undivided; and by its incessant resistance it should force the enemy to a constant state of readiness for battle, and to a ruinous expenditure of forces in tactical and strategical measures of precaution.
When both sides have in this manner reached the end of the aggressor's first start, the defender should then dispose his army in a position, if such can be found, forming an oblique angle with the route of his opponent, and operate against the enemy's rear with all the means at his command.
The campaign of 1812 in Russia shows all these measures on a great scale, and their effects, as it were, in a magnifying glass. Although it was not a voluntary retreat, we may easily consider it from that point of view. If the Russians with the experience they now have of the results to be thus produced, had to undertake the defence of their country over again, exactly under the same circumstances, they would do voluntarily and systematically what in great part was done without a definite plan in 1812; but it would be a great mistake to suppose that there neither is nor can be any instance elsewhere of the same mode of action where the dimensions of the Russian empire are wanting.
Whenever a strategic attack, without coming to the issue of a battle, is wrecked merely on the difficulties encountered, and the aggressor is compelled to make a more or less disastrous retreat, there the chief conditions and principal effects of this mode of defence will be found to have taken place, whatever may be the modifying circumstances otherwise with which it is accompanied. Frederick the Great's campaign of 1742 in Moravia, of 1744 in Bohemia, the French campaign of 1743 in Austria and Bohemia, the Duke of Brunswick's campaign of 1792 in France, Massena's winter campaign of 1810—11 in Portugal, are all cases in which this is exemplified, although in smaller proportions and relations; there are besides innumerable fragmentary operations of this kind, the results of which, although not wholly, are still partly to be ascribed to the principle which we here uphold; these we do not bring forward, because it would necessitate a development of circumstances which would lead us into too wide a field.
In Russia, and in the other cases cited, the crisis or turn of affairs took place without any successful battle, having given the decision at the culminating point; but even when such an effect is not to be expected, it is always a matter of immense importance in this mode of defence to bring about such a relation of forces as makes victory possible, and through that victory, as through a first blow, to cause a movement which usually goes on increasing in its disastrous effects according to the laws applicable to falling bodies.
Return to the top of this page
Visit ClausewitzStudies.org
Visit the
Clausewitz Bookstores!
US • UK • France • Germany