If, therefore, such exercises allow us to leave many things out of consideration because we believe they neutralize each other, we still cannot ignore those conditions that have brought about the war and that determine its political purpose. The political purpose and the means available to achieve it give rise to the military objective. This ultimate goal of the entire belligerent act, or of the particular campaign if the two are identical, is therefore the first and most important issue that the strategist must address, for the main lines of the strategic plan run toward this, goal, or at least are guided by it. It is one thing to intend to crush my opponent if I have the means to do so, to make him defenseless and force him to accept my peace terms. It is obviously something different to be content with gaining some advantage by conquering a strip of land, occupying a fortress, etc., which I can retain or use in negotiations when, the fighting stops. The exceptional circumstances in which Bonaparte and France found themselves since the Wars of the Revolution, allowed him to achieve major victories on almost every occasion, and people began to assume that the plans and actions created by those circumstances were universal norms. But such a view would summarily reject all of the earlier history of war, which is absurd. If we wish to derive an art of war from the history of war -and that is undoubtedly the only possible way-we must not minimize the testimony of history. Suppose we find that out of fifty wars forty-nine have been of the second kind- that is, wars with limited objectives, not directed at the total defeat of the enemy -then we would have to believe that these limitations reside in the nature of war itself, instead of being in every case brought about by wrong ideas, lack of energy, or whatever. We must not allow ourselves to be misled into regarding war as a pure act of force and of destruction, and from this simplistic concept logically deduce a string of conclusions that no longer have anything to do with the real world. Instead we must recognize that war is a political act that is not wholly autonomous; a true political instrument that does not function on its own but is controlled by something else, by the hand of policy.

The greater the extent to which policy is motivated by comprehensive interests, affecting the very existence of the state, and the greater the extent to which the issue is cast in terms of survival or extinction, the more policy and hostile feelings coincide. As policy dissolves into hostility, war becomes simpler; it proceeds according to the pure concept of force and destruction, and satisfies whatever demands can be logically developed from this concept, until all its component parts come to possess the coherence of a simple necessity. Such a war may seem entirely apolitical, and on that account has been considered the norm. But obviously the political element exists here no less than it does in other kinds of war. It merely coincides so completely with the concept of force and destruction that it vanishes from sight.⁶

In light of this discussion, I have no need to prove that wars exist in which the objective is even more circumscribed - a bare threat, armed negotiations, or in the case of alliances, the mere pretext of action [by one of the allies]. It would be unreasonable to maintain that such wars are beneath the art of war. As soon as we concede that logically some wars may not call for extreme goals, the utter destruction of the enemy, we must expand the art of war to include all gradations of military means by which policy can be advanced. War in its relation to policy has above all the obligation and the right to prevent policy from making demands that are contrary to the nature of war, to save it from misusing the military instrument from a failure to understand what it can and cannot do.

Consequently I must insist that the military goals of both sides are stated whenever a strategic plan is drawn up. For the most part these goals arise out of the political relations of the two antagonists to each other, and to other states that may be involved. Unless these relations are outlined, a plan can be nothing more than a combination of temporal and spatial relationships, directed toward some arbitrary goal-a battle, siege, etc. To the extent that this goal cannot be shown as necessary or superior to others, it can be challenged and contradicted by other projects, without these coming any nearer the absolute truth than did the first plan. That is, indeed, the history of all strategic discussions until today. Everyone rotates within some arbitrary circle. No one tries to push his argument back to the origins of the war that is to be fought, to its true motive, to the one and only point where the logical development and conclusion of the military operations can alone originate. Whatever correct and effective strategic decisions are made, result from the instinctive tact of talented commanders, who with a glance penetrate and assess a mass of circumstances. This instinct suffices for action, but obviously not for analysis, even though action is something far greater than the laborious unfolding and laying bare of facts.

You see from this, dear friend, how little I can make of your assignment.⁷ Above all, I must ask: do the Austrians intend, or could they intend, to defeat and disarm Prussia with 150,000 men, or would they be satisfied with a limited objective? In the Seven Years War they were unquestionably in a position to achieve the former; their primary means would, have been an advance through Saxony and Lusatia on Berlin. They failed to do so, and this failure was rightly counted as a great error on their part, just as it was considered an error that they directed their operations more toward Silesia than Lusatia. But if circumstances are such that the total defeat of the Prussian military monarchy is not possible, then an advance through Lusatia toward Berlin would no longer be the most suitable operation and consequently no longer present the greatest danger to Prussia.⁸

You will agree, of course, that in circumstances in which Austria has only enough power to seize some territory with one or a few fortresses, and in which only a modest plan promises success, while a more ambitious one promises none because it disregards the means available, the modest Austrian plan would carry the greatest threat to Prussia. You can see that the question as to which operation is the most dangerous assumes an aura of universality, in this problem as well as in both solutions, that it does not really possess, an error that recurs constantly in a strategic discussion, with the result that general strategic arguments nearly always prove inapplicable to reality. The greatest threat for Prussia can only be determined if we know what the Austrian objective can be or will be. In the present instance this objective must either result from a definition of the overall political situation, or it must be arbitrarily specified as a given of the exercise. Both the problem as well as the solutions ascribe an obscure (that is, unarticulated) importance to the capital city, which in reality Berlin cannot possess. The Seven Years War demonstrated this clearly enough.⁹ Equally obscure is the notion, involved in the first assumption, that an Austrian thrust through Lusatia would, so to speak, split the Prussian monarchy into two parts, because if our army were defeated and relentlessly pursued, it would have to decide whether to withdraw behind the Oder or escape across the lower Elbe. This would be a serious crisis -but only in the face of an enemy able to carry out such a great project. It can hardly be doubted that for the Austrians such an operation, flanked from Bohemia by Silesia on the right, and on the left by the Elbe fortresses, would be very difficult. It could only succeed if the difficulties were offset by a great preponderance of force. Of this preponderance the problem has nothing to say; it appears on the contrary to assume a kind of equilibrium of strength.

Furthermore it seems to me that this problem assigns undue significance to the question as to which Austrian operation poses the greatest threat. Attack and defense determine their measures reciprocally. But surely in theory-that is, for the purposes of general analysis-the sequence of ideas arising from this interaction must start somewhere? Quite so. The sequence begins, I am convinced, with the defensive, in part because peacetime military arrangements are directed primarily toward defense, which thus precedes the offensive; in part because offensive plans depend on the disposition of the defense, without which they would have no factual basis. The defensive, on the other hand, does not lack the necessary data in the absence of an attack, because these reside in the overall character and circumstances of the country. At the general level, then, the main characteristics of defensive operations develop directly out of existing conditions. In specific cases, however, this is by no means always necessary, but only if one cannot discover the enemy's intentions early enough to take the required countermeasures. This certainly does not seem to be the case here. From the direction of their advance, the location of their magazines, the mobilization of their transport, and other information, one would soon learn whether the Austrians were directing their main force toward Silesia or Saxony. Whether they intend to advance along one bank of the Elbe or the other, they might perhaps conceal until the last moment. But this would make very little difference to us since the slightest movement right or left on our part would compensate for it. If, as is most natural with respect, to our peacetime deployment, we assemble our three Silesian divisions at Neisse, Liegnitz, and Sprottau respectively, the Second and Third Corps and the Guards in Lower Lusatia, and the Fourth Corps on the right bank of the Elbe at Torgau, we can await information that we would certainly receive on the, direction of the enemy's main force, and then lead our main force against it. For this reason it seems to me that this particular exercise ranges quite unnecessarily into the realm of speculative reflection. Wherever concrete conditions are decisive, such speculations, which all too often degenerate into hair-splitting, are no longer of interest. If the question of Prussia's greatest military weakness is to be raised, the problem would have to be totally recast.

Having now shown that the exercise is too incomplete to permit a solution that is not totally arbitrary and furthermore that the particular issue that is supposed to be the key to a solution -namely the question as to the most dangerous line of advance -would in practice never form a basis for our actions. I shall proceed to the individual points of the second [M's] solution. I shall analyze these points historically, since the solution is too illogical to permit a strictly logical rebuttal.¹⁰

1. Why would an attack on Silesia reveal Austrian intentions sooner than attacks elsewhere? Because the border is somewhat farther from Berlin? That is too petty and insignificant a consideration to influence the choice of operational lines. If the Austrians set out from Vienna and the Danube toward our border, we will certainly grasp that we are their target, and prepare ourselves to be in the right place at the right time. We would not delay our measures until the Austrians had crossed the frontier.

2. The advantage of initiative here means the advantage of surprise. Only when surprise is present does the initiative confer an advantage; otherwise, in war as in card games, it is a disadvantage.¹¹ This inheres in the nature of things, but for want of space cannot be discussed here. If Berlin is to be surprised, this can certainly be achieved more easily by way of Saxony; if it should be Silesia, that would not be the case. Obviously the Austrians can attack us much earlier in southern Silesia than in the Mark Brandenburg.

3. The reason for the Austrians to prefer an advance on Berlin is that if Prussia places a high value on protecting her capital she will be forced into a most disadvantageous defensive. On the mere possibility that Prussia would make this mistake, the Austrians are supposed to base their line of advance. Again, this consideration is far too insignificant. And why should Prussia's defensive be most disadvantageous? What an empty phrase! A strong position behind the Notte and Nuthe rivers, two fortresses on the enemy's left flank [Torgau and Wittenberg], an entire province [Silesia] on his right -how can anyone call that disadvantageous? I do not say that this is where we should make a stand; I merely want to show that the reason given here itself needs justification.

4. What is the purpose of indicating the three distances from Vienna to Berlin and to two points on the Prussian frontier? At most the figures might be used to determine how far forward we could assemble our forces. But the author does not use them for this purpose, to which, in any case, the distance from Vienna to Berlin would be irrelevant. It seems as though this distance is meant to demonstrate the danger for Prussia of the second option. But the author does not draw this conclusion; rather he believes the danger to derive from something else -the Elbe and Torgau.¹² In the end, however, that maneuver [the siege of Torgau] leaves the importance of Berlin quite out of the account. If Berlin really were the keypoint of Prussia's entire, defensive system, its distance from Vienna would not matter very much, since this situation -unlike some other subordinate strategic options - demands more than a mere race.¹³ That Prussia would not learn the direction of the Austrian offensive until her frontier had been crossed, is an arbitrary, improbable assumption, totally unsuited to provide a premise for strategic analysis. I confess, then, that I find these calculations of time and space unnecessary on the one hand, and without practical value on the other.

5. The fact that Austria can move a siege train by water to Torgau is relevant only to a siege of Torgau. Why it should pose the greatest danger to the Prussian state would first have to be demonstrated. Again this is merely an unsupported assertion. There may be cases when a siege of Torgau would be more dangerous to the Prussian state than a siege of Glatz, but the opposite could also be true. If Torgau is invested without great superiority of force, success would hardly be conceivable. The then becomes the best means for Prussia to gain a general siege strategic victory. So long as the relationship between the two sides is not more fully specified, we cannot really think about these issues in a practical way. My analysis merely tries to show that the proposition on which the author's reasoning rests is, as so very often in strategic theorizing, an entirely unproven thesis, a mere phrase.

6. The Austrians are supposed to be able to proceed safely on both banks of the Elbe until they reach the vicinity of Dresden.¹⁴ This is inferred from the defensive system of Prussia. If by this term the author means the Prussian intention to fight the campaign on the strategic defensive, then his inference is false. If circumstances were as described, nothing would be more compatible with this intention than to attack one of the Austrian columns before the two could unite. If the author means a specific defensive system, which might perhaps consist in a deliberate withdrawal, we would have to assume that this system, which incidentally is not mentioned in the assignment, had been betrayed to the Austrians. Otherwise they would always have to accept the possibility of having to pay a price for separating their forces. If we consider that while the Austrians are assembling their army on the Eger, the Prussians assemble theirs on the Elster, it is hard to predict with certainty who would arrive first in the vicinity of Dresden. And I doubt that the Austrian commander would risk an advance in two columns on both banks of the Elbe. Instead of envisioning the situation of both armies shortly before the campaign begins, and asking what the most advisable course of action for each would be, a general concept, namely the concept of the defensive, once again is taken as the basis for a conclusion that is not only without practical value, but also inherently wrong.

7. The conclusion the author now draws, that the Prussians should place themselves near Torgau, again seems entirely illogical to me. It is not clear why the Prussians could not occupy more advanced positions and withdraw from them in time to concentrate on Torgau. Besides, the conclusion assumes an intention that must at least be discussed, since it is by no means inevitable -I mean the intention of the Prussian commander to cover Torgau and prevent a siege. Far from being a universal necessity, this intention is fundamentally contrary to the nature of things. Fortresses exist to be besieged. A siege weakens the enemy, and hastens the moment when we can defeat him more easily. This is a natural sequence in theory, and in all wars fought over major issues it is also the natural course of events. On the other hand, in limited wars, the play of balanced forces frequently causes armies to cover fortresses. The history of war is full of examples. It would take me too far afield to disentangle this seeming anomaly here, and to show that it is entirely, natural and justified. I only assert that in the present case this intention [of preventing a siege of Torgau] is not justified, and cannot be conceded as a general, self-evident necessity.

8. What does it mean to say that the Prussian army should retain its freedom of action? Obviously the farther forward its position, the greater its freedom of action, since by advancing it will increase the number of possible lines of retreat in its rear. The phrase "freedom of action" is among the most pernicious of all strategic cliches, because it is used more often than any other, and no one feels obliged to define its actual meaning.

In general, then, the author concludes that the best position for the Prussian army lies between Herzberg and Torgau, where, in one way or another, it can offer the most effective resistance. I have no wish to criticize this conclusion in and of itself, only to point out that it bears little or no relation to the author's construct of strategic theorizing and calculations of time and space. The Prussian army is strong in this position because of the Elbe, the three fortresses, the sheltered character of the whole deployment, and because the Silesian border runs parallel to the enemy's lines of communication. These things can be said in a few words, sound common sense will accept them, and require no ingenious, long drawn-out strategic deduction.

9. Since so few facts are given [in the problem], it is impossible to determine whether a flanking position with our back to the Elbe would be superior to any other. The Prussian army appears to find itself in such an excellent strategic position here that it can evade an attack in three directions toward the Elbe, toward Berlin, and toward Silesia -and still retain the advantage of posing a strategic threat to the enemy, either on his two flanks or in his rear. Nevertheless, the army's main line of retreat should certainly be determined, that is, the line to be used in case of extreme misfortune, with which all other operational, decisions ought to be correlated. The fact that at Vilna [in 1812] the Russians were still undecided whether to withdraw their main force toward St. Petersburg or toward Moscow nearly led to the disaster of the army's surrender in the open field. If it were considered absolutely necessary to cover Berlin, a flanking position so near the city would not be appropriate. I believe, however, that protecting Berlin is not an essential element of the Prussian defense; consequently in a great many cases, which need of course to be studied in detail, a flanking position would prove highly advantageous.

10. A position behind the Elbe, just as behind any large river, is tactically unassailable. The strength of the position does not depend on the two fortresses. But we do not gain a favorable battlefield by deploying behind the Elbe, because we force the enemy to by-pass us. It is only from this compulsion that our advantages must come.

11. In how many positions do both wings rest on major strategic strong points? To have one such point is already worth a great deal. Once again, the reason why we should not deploy at Luckau, the lack of support for the right wing, is one of those strategic terms that cannot bear close inspection. Whether we might not find some tactical support for the right wing [e.g., some farm buildings, a small rise in the ground, a swamp. Eds.] would have to be discovered on the spot. How many battles are fought in which only one wing of the defender's position is secure!

12. The enemy could no more determine the movement of the Prussian army at Luckau than anywhere else. This is one of the most abominable strategic clichés. Basically it says absolutely nothing, since, after all, the actions and movements of one commander do always strongly influence the conduct and movement of the other. If the statement means anything at all, it can only be that one general can compel the other to follow him everywhere, to submit to him wholly, to deprive him even of the possibility of indirect resistance by means of retaliatory countermovements. This does not in any sense seem to me to hold true for a Prussian army at Luckau, and we may assume that had the author developed his ideas in greater detail, their inadequacy and one-sidedness would have become obvious.

13. Battle is here presented as something evil, at least that battle that is resorted to without hesitation. Here we enter the morass of muddled concepts that made up the general-staff science of the Old Regime. Against a determined enemy, who does not shy away from fighting, battle is the only effective means of resistance. We may fight him under the most advantageous circumstances possible, but we must be resolved to fight. In such cases there is no substitute for battle. If the defender has occupied an exceptionally strong position, he will force the attacker to bypass him. If this position is well-placed strategically, to be bypassed offers an advantage to the defense. But the advantage becomes reality only if it catches the attacker in the process, en flagrant delit, as Bonaparte would say. In short, a battle is inevitable: either a battle in the tactical defensive, if the attacker finds bypassing too dangerous, and therefore proceeds to attack our position; or an offensive battle, if the attacker pursues his [original] objective and risks bypassing our position. If the tactical features and strategic location of the defender's position are so strong that the enemy dares to do neither and gives up his advance entirely, then this success without battle arises only from the strictly necessary presupposition that the defender was in fact prepared to fight.

If the attacking army does not advance resolutely, if it does not plan an energetic offensive, if it has taken the field merely to await some favorable opportunity, and will only attempt something should this opportunity arise, then the defense can certainly do likewise. Against such an opponent, the defense may make it its business to avoid battle entirely, to regard it as an evil, and to direct all actions and movements toward ensuring that the advantageous conditions for battle that the enemy seeks will not occur.

It is important not to confuse the two cases. To expect favorable results from passive resistance against a determined opponent goes against the nature of things. On the contrary, nothing is so certain as that such an approach would lead to half-measures, wasted time, confusion, and-following this splendid purgatory -to the most complete defeat.

All this being said, the advance on Luckau and Herzberg surely suggests that the Austrians are seeking a decisive battle.

14. If the Prussian commander retreats north over Torgau simply to bring this fortress into play, obliging the Austrians to besiege it, and then, after they have been weakened by the siege, to attack them, I would find this most sensible. In that case, however, the Prussian commander has no need to withdraw behind the Nuthe River. Rather he will try to halt behind the Elster. Should the Austrians want to drive him off and force him either to withdraw behind the Nuthe or to give battle, the Prussian commander may prefer to accept battle, even if the enemy had not weakened himself significantly by laying siege to Torgau. In light of the general circumstances there seems to be no reason to shrink from this battle, while a retreat to the vicinity of Berlin would be undertaken only on the most pressing grounds.

But even supposing such grounds were present, I would not find the retreat as disadvantageous as the author does. He says that the Prussian commander would be thrown back on the most disadvadvantageous defensive, a well known, ready-made figure of speech from the strategist's workshop.¹⁵ If the expression disadvantageous refers to the defense itself, the statement is clearly wrong, for a defensive position behind such a line of marshes as exists along the Nuthe cannot possibly be considered a disadvantage. Should the expression refer to the fact that if the enemy lays siege to Torgau, defensive operations would no longer be appropriate, and would therefore be detrimental, then I cannot persuade myself that our army would find it all that difficult to advance from this position by one route or another to relieve Torgau.

15. Here we must first ask, what is the meaning of leaving the Elbe? Does the author mean, evacuate the entire area, and let the Austrians take Torgau at leisure; or is he merely referring to a situation in which one wing of the army is not at all times covered by the river? Again, unfortunately, we are dealing with jargon, which, as usual, bears only a faint resemblance to well defined, specific concepts. Even a retreat behind the Nuthe would make the subsequent relief of Torgau neither impossible nor very difficult, as anyone will agree who thinks of the numerous ways for doing this that are open to the Prussian army. Only if the Prussians were to take no steps to relieve the fortress, could the Austrians occupy Torgau without a battle and so acquire an enduring advantage. But this is an empty supposition. If, on the other hand, the author means that as long as we retreat at least one day's march from the Elbe, the Austrians would succeed in taking Torgau, he lacks all basis in fact. After all, we can advance from any point to attack them.

In general, it is not clear how an attacking army can achieve a permanent advantage over a defending force that is its equal physically and in morale, without defeating it in battle. If such a victory is necessary to the success of the Austrian campaign, this necessity is not a consequence of any particular strategic combination on the part of the Prussians, but follows quite simply from the nature of things. That the author again regards battle as an evil, an incongruity, reflects the same confusion that we have discussed under point 13, above.

With this I conclude my remarks on Solution M. I feel only too keenly how cursory they are, and how much they may leave unclear. But to make good these failings, I would need a great deal more time.

You see, my dear friend, I have scarcely left any propositions unchallenged that are meant to yield a solution to the problem. It is not so much the solution as this type of arguing that I should feel obliged to attack, were I competent to do so.¹⁶ I hate the sort of technical language that leads us to believe we can reduce the individual case to a universal, to the inevitable. Strategists manipulate these terminologies as if they were algebraic formulae, whose accuracy has long been established, brief formulae that may be used as substitutes for the original reality. But these phrases do not even represent clear and definite principles. Rather they are nebulous, ambiguous expressions, whose true meaning remains open to question. This is no accident. Their vagueness is intended, because they did not derive from what is essential and could be presented as universal truth. Consequently the inventors of these terms found it natural to allow a certain latitude in their meaning.

I recognize that my finding fault with more or less everything in Solution M conveys a strong impression of a mind already made up. But I am convinced that I do not hold preconceived opinions, nor could I, because I do not follow a particular [strategic] system, and demand nothing but the plain, straightforward truth, the simple linking of cause and effect. I hope you will not suspect that mere contrariness or, worse, personal antipathy are involved. Pointless contradiction is utterly repugnant to me; and with respect to the author, I am truly sorry not to find him farther along the road toward a natural view of strategic issues, a road I have pursued for many years with the greatest enthusiasm, because despite everything I agree with his practical judgments more often perhaps than with anyone else's.¹⁷

After all this, my analysis of your solution will require far less time, I fully approve of its simplicity and realism. You set up no scaffolding of superfluous circumstances; your reasoning is not based on mere clichés. But let me be more specific about particular issues.

1. You are right to characterize the operation through Lusatia as the most decisive. This is the way to put it. The most decisive operation, however, is not always the one that best suits the enemy's circumstances. Of course, the exercise demands that in planning the preliminary concentration of the [Prussian] army, you determine [in the abstract] what attack would be the most decisive. But, as I have already noted, this would not be necessary in practice.¹⁸

2. Your description of this option is accurate. It proves the soundness of your point of view that you characterize the option as a possibility and not simply as an outright error, as does Solution M.

3. This also holds true for the third operational option. But your assertion that for the Austrians a victorious battle would be more advantageous in Lower Lusatia than at the foot of the Silesian mountains is based on the one-sided assumption that the Austrians are in the position to fight a war aiming at major decisions.

4. I have already indicated my objections and qualifications to this assertion.

5. Your calculations of times are simple and sound, without the false implications of the other solution. Such calculations are, of course, always necessary; but in this case they lack the great significance that the problem seems to attribute to them. Nothing is less likely than that the Austrians would march from the Danube to our border in one fell swoop. Therefore we should not be concerned that the Landwehr units of the 2nd Corps would not be available. But that you have emphasized - or rather, mentioned - this possibility makes your calculations superior to those of the other solution, since it is only on account of such possibilities that these calculations are necessary at all.

6. I have nothing to say against the details of our disposition given here, except that such specifics cannot reasonably be required or given in advance. With the information that the enemy has crossed our border will naturally come much other news about the number and strength of his columns. Only then will we decide the details of our disposition.

7. I consider this eccentric deployment the one true fault of your answer.¹⁹ In such a confined space, at a moment when a great battle may be expected, a divided deployment is always an error, unless it is justified by a preponderance of force. Nothing can protect such a deployment from the danger that the enemy, with one and the same army, may defeat the separate parts one after another, as the famous campaign of 1796 [Napoleon's, in Italy] with its five distinct phases has shown. Over longer distances such an enveloping form of attack or defense becomes less dangerous, and if distances are really great, as in Russia in 1812, a division of forces may cease to be dangerous at all, and its special characteristic will then naturally become advantageous. In 1813 the distances were fairly large, and yet there was always the greatest danger that Blucher would be overwhelmed by the main enemy army, which, in fact, did defeat Schwarzenberg at Dresden. The enveloping form of attack is always the more decisive, the one that leads to greatest success, but for that reason also the riskier, in which success is less certain. Success and danger always stand side by side, and form the dynamic law of war. If we wish to increase the first, the second rises as well, and it then becomes, a question of whether or not this is in accord with the needs and particular characteristics of our situation. Thus, if our circumstances do not allow us to take great risks, we can increase our success only when danger itself is not great, that is, when we possess a preponderance of physical and moral force. This utterly simple principle, grounded directly in the concepts [of success and danger] themselves, allows for clear and definite solutions in a great variety of strategic questions, over which, in the usual way people argue fruitlessly.

It is very tempting, if one of our territories lies to one side or behind the enemy advance, to base a considerable force there. This idea has seduced you, but one must resist. That is not to say that we must give up the advantage of this circumstance entirely. Rather we ought to detach a few small units of rangers, whose combined strength is not so essential to the whole that it could not be spared in a decisive battle.²⁰ In an army of 120,000 men, for instance, 5,000 men would not be missed. But these 5,000, in combination with the forces in the area that they could bring into action-Landwehr units, garrison troops, reservists, etc. -can act very effectively against the enemy's lines of operation. This type of operation against an attacker's lines of communication, which arises, in a sense, automatically as he advances, leaving territory to his left and right that he cannot occupy, is the only kind that offers an absolute advantage. It is a unique advantage of the strategic defensive, but it cannot be achieved by detaching significant forces to act against the enemy's flanks, since those who fight on his flank cannot fight on his front.

8. The division's war of maneuver is one of those miserable catchphrases. You will forgive me if I say that you did not form a clear conception of what this would mean.

9. I believe I have already said that if the Austrians were to advance on Silesia rather than Saxony, we would discover it early enough to assemble our main army in Silesia itself, rather than first in Lower Lusatia. The problem is at fault here.

10. Your conclusion is very sensible, and shows that you regard the major battle from the correct point of view. And now, enough. Perhaps I have already worn you out. If I have now and again failed to make myself understood, or if I have not been able to convince you on every point, we can certainly continue our discussion in person.

If, Roeder argued, Austrians were faced with the Prussian deployment given in the first case (four corps between Senftenberg and Spremberg, the fifth at Görlitz), they would have to leave a significant covering force at Görlitz, perhaps as much as one-and-a-half corps, since the Prussians there threaten the Austrian line of retreat. If the Austrians were to lose a battle, this threat could become particularly dangerous. It would probably be impossible to defeat the Prussians at Görlitz, who would avoid engaging a superior force. (1)²¹ After detaching the covering force, the Austrians would retain three-and-a-half corps for their offensive. How should they be employed? To bypass the left wing of the main Prussian force with an advance between the Spree and Neisse rivers is out of the question. (2) If they bypass the right wing, the Austrians could cross the Black Elster between Elsterwerda and Senftenberg. But the terrain would be difficult for them, and we could attack them under favorable conditions. If they cross below Elsterwerda, they would have the Elbe and Torgau to their rear, and a battle would probably be fought between Senftenberg and Spremberg.

Should the Prussian army lose this battle, it would probably have to retreat to Berlin. A good position can be taken up on the heights near the city; if we did not feel strong enough to accept battle there, we would have to withdraw behind the Havel River. (3) If in the meantime the enemy lays siege to Torgau with half a corps, and we bring up reinforcements, we would outnumber him and could take the offensive.

If the entire Prussian army deploys on the Elbe (the second case), an Austrian advance on the left bank of the river is unlikely. If, on the other hand, the Austrians cross the Black Elster and advance on its right bank, they would lose direct communications with their country, and it isn't clear what they would gain. Consequently this move is also unlikely. (4) An Austrian advance between the Elbe and the Black Elster is their most likely move. How should we respond? We can deploy either between the Elbe and the Black Elster, (5) or before the bridgehead at Torgau. (6) We might also cross to the left bank of the Elbe, and take up a position near Torgau. In this case the enemy could bridge the Elbe at Mühlenberg, cross to the left bank, and advance. The Prussians, however, would still have the option to avoid battle. If they accept battle and lose, they might cross to the right bank at Torgau, and retreat to Wittenberg. (7)

These two defensive plans are comparable in several respects. In either case the enemy must win a battle before he can besiege Torgau. In both cases a Prussian victory would seem to lead to similar results. A Prussian defeat between Senftenberg and Spremberg would take the army further from Torgau, because it would have to withdraw to Berlin. But the retreat would leave the army in touch with the main part of its homeland, and thus with its reinforcements. Stationing one corps at Gör1itz seems advisable, because it would threaten the Austrian flank. The force should, however, be commanded by a skillful, energetic general. If such a man is not available, it would be preferable not to detach the corps.

It would help us to choose between the two plans, if we knew the exact lay of the land where battles are likely to occur. One would also like to know more about the environment of the fortresses at Torgau and Wittenberg, before deciding whether our forces should be deployed there. (8)

In conclusion, if Görlitz were fortified it would offer a position that combines the advantages of both plans. We would threaten the enemy flank, whatever he might do; we would retain contact with the main mass of Prussian territory; and we would remain free to turn toward Silesia. (9) Of course, we would not benefit to the same extent from the advantages of the Elbe with its two fortresses. The choice of our deployment would also depend on whether we could expect to be reinforced by the 1st Corps from East Prussia, and the 7th and 8th corps from the Rhine.

Believing that I had discharged my debt to our friendship by the lengthiness of my remarks if by nothing else, I now see that you call on me for a further exercise. I shall take up pen again, not without the fear that your patience has already been exhausted by my first letter.

In that letter I pointed out the extent to which the first problem lacks the specific information that alone would permit a solution that is not wholly arbitrary.²² The second problem is an extension of the first. Admittedly, it is more detailed, not as generalized; nevertheless its terms are such that the specifics I missed in the first exercise would also be very significant here. On the other hand, we see that as the conditions of the exercise are presented in greater detail, the specific arrangements of the enemy assume increasing importance. Since it is impossible to generate all essential facts in a hypothetical exercise, it is evident that the more detailed our account, the more illusory it becomes. We make any number of tacit assumptions, and develop an analysis that in the end might not be relevant to one case in a hundred. Reasoning of this kind may still be useful to train our judgment, of course. But it is clear that such an analysis can never be satisfactorily refuted by equally arbitrary arguments. If we are contentious we won't be able to reach agreement; if we honestly seek a solution we will eventually fall into a disagreeable state of perplexity, in which we might almost despair of the validity of any theory whatever.

By positing two different means of execution, and asking you to choose between them, the second exercise expects you to criticize each by means of the other. This seems to me particularly inappropriate since the two approaches differ only in one unimportant point. In any strategic problem, but most especially in those that pose alternatives and ask us to choose between them, I feel the need to reduce the issue to general principles, that is to reveal the relationship between one or the other option and the facts that inevitably result from the nature of the situation. In this way, at least, we can recognize the nature of each measure, and its unique characteristics. In the event that we must execute one scheme or the other in real life, we can then decide for ourselves whether the characteristics of one or the other are better suited to our requirements and circumstances. In short, when we evaluate hypothetical plans, we must suspend final judgment on many points, while on others we can be conclusive, because they violate conditions that are set down clearly enough in the problem. I shall now offer such comments on the various aspects of your analysis.

You say that a superior enemy force could not defeat the Prussian corps at Görlitz, or rather, you assume it, and indeed it is no more than an assumption. Both experience and the nature [or logic] of the situation teach us that it is very difficult to avoid a developing battle if we are to maintain contact with the enemy and miss no favorable opportunities. An enveloping attack (what Jomini calls operating on exterior lines) is therefore always very dangerous, and is warranted only by a preponderance of force or the knowledge that the enemy is not seeking a decisive action. Nor can you be certain that the enemy will always keep one-and-a-half corps against your corps at Görlitz. When he is about to attack your main force, he may leave only half a corps there. In any case, the force at Görlitz does not present all that much of a danger for him, since in an emergency he can withdraw to the Elbe and by means of a pontoon bridae cross the river below Dresden as well. The calculations pertaining to the corps at Görlitz are thus extremely uncertain; it would be a mistake to trust them rather than the far safer method of maintaining a unified force.

In strategy we must distinguish between outflanking and bypassing a position. It is one thing to envelop a position with individual corps or even with the whole army, and attack it from the flank or rear. It is obviously something quite different to bypass the position in order to pursue the object of the attack, without regard to the enemy being left behind. In the first case, the position retains its strategic effectiveness; indeed, the attack demonstrates that it cannot be ignored, and it only remains for the tactical features of the position to prove their strength. In the other case, the position has lost its strategic effectiveness. It is extremely rare that an attacker can bypass a position that isn't very badly situated. But it is equally rare that a position cannot be outflanked. In most cases the defender must be prepared for this eventuality, and make his arrangements accordingly. A position that is given up because it is being enveloped, is hardly worth taking in the first place. I do not think the Austrians could bypass your position either on the right or the left. They could certainly envelop it more easily on the right than on the left; but you cannot assert that enveloping it on the left would be out of the question.

3. A retreat behind the Havel River would indicate that we intend to base ourselves on the western part of the monarchy. If that were not the case, the move would be entirely inappropriate.

4. Presumably the real question is whether the Austrians if they want to carry out their offensive, can do anything other than fight a battle. My answer is that they can't. First, because they are not stronger than we are; secondly, because they are entirely surrounded by our territories, and thus clearly at a disadvantage. Their ability to force us to withdraw and then to besiege and capture Torgau is reduced by these two circumstances to such an extent that it would make little sense for them to base their plans on this possibility. If they want to fight, they must cross to the right bank of the Elbe where we are, and naturally they would prefer to cross somewhere out of our reach. But it is obviously wrong to hold that they could not seek battle by advancing on the right bank with part or all of their army. Crossing the river would complicate their communications; but, as I have just indicated, they can hardly bypass us or force us back [on the left bank].²³

5. If we deploy between the Elbe and the Elster, it would have to be south of Torgau. Hemmed in between two rivers that present real obstacles to movement, the one because of its size, the other by the nature of its banks, places us in a situation in which our lines of retreat are exceptionally limited, a fact that exerts a very undesirable influence on the conduct of battle. If, to take advantage of a [tactically] strong position, we choose to fight in this area, we should at least keep open a line of retreat in Torgau itself. That demands that we deploy south of the town, and not too near to it.

6. Deploying before Torgau, with our backs against the town, like a bridgehead of flesh and blood, would mean needlessly resorting to a measure of despair. Such a position, benefitting from two protected flanks, but with a front that - being convex - is very weak, and which in general restricts the defender to extreme passivity, is suitable only to a force that because of its weakness can no longer stay in field. It has been driven into a corner; its back is against the wall because it is on the verge of giving way before the superior strength of the enemy.

7. A large river, flowing in such a direction that the defense can maintain its lines of communication for several days' march on both banks, certainly offers the defense some favorable opportunities. This is even more true when a fortress with a bridgehead protects his movement from one bank to the other, as Torgau does, and when another fortress, as Wittenberg, some distance to his rear, multiplies the number of options. It seems, indeed, that, at the moment the enemy advances, the defender can avoid battle by crossing to the other side of the river, without ceasing to cover the area through the overall strategic effectiveness of his position. At first glance it is not obvious why this game could not be repeated indefinitely. But, first of all, it must be said that in most cases the strategic value of a position declines markedly when a large river separates the defense from the attacker, since this allows the attacker to divide and maneuver his forces in ways that would not be possible without this barrier. Secondly, it is rare that the lines of communication on both banks of the river are of equal value to the defense, and that either one could be given up at any moment. Thirdly, the retreat of a large army over a bridge across a river is not a trivial act; on the contrary, in the presence of the enemy it is barely possible. Fourth, to break out on the enemy's side of the river, should this become necessary, is no simple matter, even in the vicinity of a friendly fortress. Fifth, and finally, it would be very difficult repeatedly to shift all those elements that constitute the rear of the army into their appropriate positions. For all these reasons, a defender could scarcely cross the river more than, a couple of times, even under the most favorable conditions, before, in his effort to counter the movements and actions of the enemy, a part or the whole of his army would suddenly find itself on the same bank as the attacker.

If we apply these considerations to the Prussian army at Torgau, we must agree that advantages could indeed be obtained by skillful use of the Elbe in combination with the two fortresses, whether to avoid battle for a time, or to give it under favorable circumstances. But we cannot say in advance exactly how this might be done, because no hypothetical case can specify all the unique, momentary conditions that give rise to opportunity. It must also be added that we ought not exaggerate the absolute value of these favorable conditions; they merely provide the opportunity for auspicious courses of action. If the Prussian commander does not seize this opportunity with great skill, the advantages of his position would soon disappear.

8. As you have correctly remarked, the position on the Elbe presupposes that the army will base itself on the western part of the monarchy, and give up its connection with the east. Should this not be in accord with the general [political and strategic] circumstances, the position would be unnatural and of doubtful value.

It seems to me that overall the difference between deploying behind the Black Elster and at Torgau is not great. If we want to remain realistic, we would have to say that in the event of a major battle, this particular difference would be only a minute factor in the final result. The degree to which both commanders have united their forces for combat, the good planning and skillful conduct of the battle, the perseverance of the commanders, the courage of the troops, their confidence in their leader, the obedience of the subordinate generals-are these not all factors that have greater significance and that affect the outcome more directly?

Whether the Russians in 1812 withdrew from Moscow to Vladimir or turned south to Kaluga -this one basic alternative, while not primarily determining the direction of the French retreat, as is sometimes claimed, was nevertheless a matter of extraordinary importance.²⁴ There can be no doubt that the change of a single strategic line may have decisive influence. But we ought not to suppose that it is therefore important whether the Prussians fight at Senftenberg or at Torgau, unless we wish to impute strategic value to things that have nothing to do with strategy.

9. Deploying the main army at Görlitz (assuming the place is fortified), which is your ultimate conclusion, would put the army in a flanking position. Naturally the territory that flanking positions are supposed to cover does not lie directly behind them. Such positions can therefore be adopted only if the army has a very broad base of operations, so that the flanking force retains a line of retreat. That is the case here. But this line of retreat will always be rather constricted, sometimes more so, sometimes less. The danger that the army will be forced into an eccentric line of retreat, away from the main part of its home territory, from the center of gravity of the entire military base, constitutes a grave disadvantage of positions of this kind. The position must make up for this flaw by its very great tactical strength, so that the enemy either cannot attack it at all, being immobilized as it were by its strategic power, or is unlikely to succeed if he does attack. This justifies the flanking position, but we still lack the real reason for it: its strategic effectiveness. If you consider the present case with these points in mind, you will agree that if the army at Görlitz were defeated, it would have to retreat on the highway to Breslau, and would therefore be in danger of being driven south to Silesia. A strategist of the old school would say: in that case the army is lost. I won't, because it would be an entirely unjustified assertion. But you will concede that a thousand serious disadvantages would result from such a retreat. Finally, a strong tactical position has not yet been located in the area around Görlitz, and its strategic effectiveness would not be very great in any case, since the Austrians could base themselves on the Elbe.

Your third study meets my entire approval [Roeder had also sent Clausewitz a brief memorandum, which treated the following: a) two armies of equal strength meet, each with a vertical line of retreat; b) the same armies meet, one with a vertical line of retreat, the other with a line of retreat that is an extension of one of its wings. What are their respective advantages and disadvantages?] I would only add that it is difficult to provide a strategic cover for oblique lines of communication.²⁵

1. A Prussian corps, at war-strength consisted of two divisions, one artillery brigade, one Jäger detachment of two companies, and one engineer detachment. A reserve infantry regiment, a reserve Landwehr [second line] infantry battalion, and a reserve Landwehr squadron might also be added.

The division consisted of a line infantry brigade and a Landwehr infantry brigade of two regiments each, and of a cavalry brigade. Besides the two regular cavalry regiments, the cavalry brigade probably also included a Landwehr cavalry regiment; at any rate, a corps had a total of four line and two Landwehr cavalry regiments.

The artillery consisted of nine heavy artillery batteries, and of three light- or horse-artillery batteries. Ed. of the 1937 German text.

2. Numerals in parenthesis in Ms text were added by Clausewitz. His comments are similarly identified, and refer to the numerals in the text. Eds.

3. Numerals in parenthesis in Roeder's text were added by Clausewitz. His respective comments are similarly identified. Eds.

4. In peacetime, the 2nd Corps was stationed on the Baltic, several hundred kilometers farther away from the theater of operations than the Guards and the 3rd, 4th, 5th, and 6th corps. Eds.

5. Clausewitz writes "wie die sogennannte Strategie," which, taken literally, would make little sense. The rest of the sentence makes it apparent that he is referring to the "science" or "discipline" of strategy. Eds.

6. Compare On War, Book I, ch. 1, sections 25 and 26 (pp. 87-88). Eds.

7. "Zwei Briefe des Generals von Clausewitz: Gedanken zur Abwehr," special issue of the Militärwissenschaftliche Rundschau, March 1937, 5-9.

8. The editor of the German text points out that it is Roeder who calls an Austrian advance through Lusatia on Berlin the most dangerous threat to Prussia. See p. 22. Eds.

9. During the Seven Years War Berlin was briefly occupied, once by the Austrians, once by the Russians. Eds.

10. As will be seen, Clausewitz does not mean he will test M's solution against historical examples, though his discussion does occasionally refer to incidents in the past. He is here using "historical" in opposition to "a strictly logical," i.e., theoretical, refutation, to signify a realistic, concrete analysis of M's "individual points." For Clausewitz's belief that history, as the reflection of reality, is the only valid basis for a theory of war, see the statement in his letter to Roeder above, p. 8, and many passages in On War, for example Book II, ch. 6 (pp. 170-74). Eds.

11. The comparison with games points to the belief that the cardplayer tends not to play his best cards at once, but gradually in the course of the game. Ed. of the German text. See also the discussion of Clausewitzs essay "On Progression and Pause in Military Activity," which compares gambling and war, in Peter Paret, Clausewitz and the State, pp. 361-62. Eds.

12. The reference is to the danger that according to M a siege of Torgau constituted for Prussia. Ed. of the German text.

13. This seems to refer to those cases in which it was important to occupy a significant area and establish defensive positions there before the enemy could do so. Ed. of the German text.

14. M wrote "at the Prussian border." Major von Roeder's solution (p. 15) counted on the possibility that the enemy would cross the Elbe in the vicinity of Dresden. Ed. of the German text.

15. In this passage M did not refer to "the most disadvantageous defensive," but to Prussia being "thrown ... completely on the defensive" (p. 18). That difference does not, however, affect the sense of Clausewitz's argument. Ed. of the German text.

16. Clausewitz writes "venn ich dazu berufen ware" -which means either "were I competent to do so," or "were I called upon to do so." Both versions are possible, though the latter seems close to a tautology: "I should feel obliged to attack [the argument] were I called upon to do so." It seems preferable to interpret the statement as an expression of Clausewitz's habitual courteous self-depreciation. Eds.

17. That Clausewitz, having demolished M's solution, can claim he agrees with him perhaps more than with anyone else, may seem strange. In part, the statement is a further expression of the polite, unassertive manner that he habitually assumed toward his subordinates. But his overt tolerance covers a good measure of sarcasm, and probably also a rejection of the conventional military judgment of the day, which Clausewitz considers so inadequate that even an M might not be the most competent strategic analyst. Eds.

18. Apparently Clausewitz is again referring to the absence of specific information on the political situation, which alone would determine whether an advance through Lusatia would be appropriate. Ed. of the German text.

19. Roeder makes no provision for uniting the Görlitz corps with the main Prussian Army concentrated northeast of Hoyerswerda, rather this corps would be forced northeast into Silesia. Eccentric deployment is the same as divergent deployment, i.e., away from the center. See On War, pages 367-369. Eds.

20. Clausewitz writes "Parteigänger," who might be either elite regulars -Jäger (rangers) or Schützen (sharpshooters) - trained in detached, small unit operations, or regular line infantry and light cavalry used for this purpose, or irregulars. Eds.

21. Numerals in parenthesis in Roeder's text were added by Clausewitz. His respective comments are similarly identified. Eds.

22. This sentence introduces a characteristic example of Clausewitz's dialectic method of analysis. Eds.

23. To convey Clausewitz's meaning in this paragraph, a very free translation of the final sentence seems necessary. The German reads: "Der Grund, dass sie dann ihre Kommunikationen aufgeben wurden, bezieht sich auf das Vorbeigehen oder Herausmanovrieren, welches ich eben als untuhlich gezeight habe." Eds.

24. Vladimir is about 150 miles east of Moscow on the road to Gorki while Kaluga is about 100 miles southwest of Moscow. By placing his army at Kaluga, Kutusov was in a position to threaten Napoleon's line of retreat from Moscow. This maneuver also tended to concentrate Kutusov's army with Russian forces operating south of the Konigsberg-Vilna-Smolensk-Moscow line, a move to Vladimir would have continued to drive major Russian armies apart. Eds.

25. See On War Book Six, Ch. Four, page 462 for Clausewitz's discussion of lines of communication.

See also Erich von Manstein's Lost Victories, Henry Regnery Co., 1958, for WW II examples of divergent lines of operation; p. 291, pp. 368-369.

pp. 369-370 for an example of oblique lines of communication for German forces in southern Russia. Eds.