Abstract

This volume runs on two layers, and the point of it is that they cannot be run together.

Layer one: authorisation. Clausewitz and On Protracted War give constraints on war that share one form — war is authorised by an external political purpose and is therefore bound by it. SAE agrees with the content of both. But the form carries a premise neither states on its own: that the authoriser is itself stable, bounded, and does not expand. Neither authoriser satisfies it. So the form has a common limit — it cannot hold back the authoriser itself.

That limit is registered as a limit of the form, not as an internal failure. Neither party ever claimed its theory could bind the political purpose itself, so your constraint cannot hold back the authoriser is no internal criticism of either.

Layer two: the carrier. Machiavelli does not supply a third specimen of the same kind. He supplies a second-order stress test on layer one: even where the authoriser is itself constrained, the carrier that executes may form a war-interest of its own. Layer one asks whether the authoriser will move. Layer two asks whether, even with the authoriser still, the men who execute it will move on their own. Each layer has its own solution and its own insoluble part — and one layer's solution does not solve the other's problem.

1. War Is Sent Out, So It Must Report Back

Clausewitz and On Protracted War both hold that war should not swell on its own, and the constraint each gives takes the same form: war was sent out, so it has to go back and account to the thing that sent it.

Book Eight, chapter six B of On War puts the form most completely. War is nothing but a continuation of political intercourse with the admixture of other means; the phrase with other means is there to show at once that this political intercourse is not stopped by the war itself, does not change into something wholly different, but continues in its essence. And then: war has its own grammar, but not its own logic. Followed by the sentence that completes the form — war can never be separated from political intercourse; separate them in any way and the threads of the relations are to some extent broken, and what stands before us is a thing without object, meaningless.

Grammar of its own, logic from outside. That is the form.

On Protracted War is the one worth the closest look on this layer, because it writes the stopping condition down in plain words. War is the continuation of politics, put symmetrically: politics is bloodless war, war is bloody politics. And the stopping rule is indexed to the political purpose — the political obstacle removed and the political purpose attained, the war stops; the obstacle not fully removed and the purpose not fully attained, the war goes on. The stopping condition is not military opportunity exhausted and not the enemy destroyed. It is whether that political purpose has been completed.

So the two share one form: war is authorised by an external political purpose and is therefore bound by it. SAE agrees with both constraints. War really should not swell on its own. This is a real overlap, not a courtesy.

2. Machiavelli Is Not a Third Specimen

Machiavelli does not stand alongside the first two, and the reason has to be set out first. Clausewitz and On Protracted War give the authorisation and constraint of war by an external political purpose. Book One of The Art of War gives the incentive structure of the executing carrier. The two are not of one type — the first asks what authorises war, the second asks what the men who carry out war live on. Fold the second into the first, and two different levels get pressed into one.

His correct position here is the second-order stress test: even where the authoriser is itself constrained, the carrier that executes may form a war-interest of its own.

The material for registering it: in Book One, Machiavelli has Fabrizio Colonna argue that a well-ordered republic never allowed any of its citizens or subjects to practise war as a trade. The reason is not efficiency. It is a chain of incentives whose starting point is that the trade has no ordinary livelihood in peacetime, so those who depend on it must wish peace not to exist.

Two limits go with it. The speaker is Fabrizio, not Machiavelli in his own voice — the book is a dialogue. And what is forbidden is private livelihood, not knowledge of war: chapter fourteen of The Prince, in the author's own voice, says a prince should make the art of war his principal study. The two do not conflict.

3. Why the Two Layers Cannot Be Run Together

Layer one asks: will the authoriser move? Answer: yes. Layer two asks: even if the authoriser does not move, will the men who execute it move on their own? Answer: yes — as soon as their livelihood is tied to war.

The two have different solutions, and that is exactly what this volume needs from them. Layer one's problem has no solution here — the legitimacy of the authoriser is supplied from elsewhere. Layer two's problem does have one, and Machiavelli gives it: let no one practise war as a trade.

So this volume does not count him as a failed case. What he gives on layer two is a workable disposition. And the converse holds too: adopt his restriction, and layer one's problem is still there — the survival of a republic can equally demand conquest abroad.

Each layer has its own solution and its own insoluble part; and one layer's solution does not solve the other's problem. That sentence is all this volume takes from Machiavelli. The full treatment of layer two belongs to the question of how a means reworks the party using it, and that is not this volume's business.

4. The Other End of the Rope Is Moving

The form war is constrained by an external purpose carries a premise: that the authoriser is itself stable, bounded, and does not expand. If the authoriser can enlarge its goals in the course of things, then war must not exceed the purpose constrains only the distance between war and purpose — not how far the purpose itself has moved. The shape of the constraint is a rope, and the other end of the rope is moving.

Neither party states the premise on its own, because in each of their contexts it was too natural to state. And neither authoriser satisfies it.

Reason of state can be expansionist. Clausewitz's political purpose is the state's political purpose, and a state's political purpose can perfectly well be acquire more. War must not depart from the political purpose still holds — it merely constrains war not to exceed what the expansion requires. It cannot hold back the expansion.

A political purpose can upgrade itself. A political line redefines its goals in execution; that is not a betrayal of the line, it is the line's normal evolution. And after the redefinition, war is the continuation of politics still holds. What continues is the new politics.

5. The Clearer the Stopping Line, the Plainer What It Hangs On

And On Protracted War, precisely because it writes the stopping condition in plain words, shows the problem most clearly. Its rule is: political purpose attained, war stops. That is the only stopping condition on this layer that can be quoted, questioned, and checked afterwards — in that sense it is the best of the two.

And that plain stopping condition moves as the political purpose moves. Redefine the purpose and the stopping line shifts with it, while the rule political purpose attained, war stops needs not one word changed. So this is not a counter-example. It is the strongest positive example: the more clearly the stopping condition is written, the more plainly one sees what it hangs on. A plain sentence hung on something movable is still movable.

So the form has a common limit: it cannot hold back the authoriser itself. And the registration matters. This is uncovered / a limit of the form, not an internal failure. To call it a failure point would require first showing that the party had claimed some goal or criterion of its own — and neither ever claimed its theory could bind the political purpose itself. Clausewitz's theory sets out to explain what war is and how it stands to policy; On Protracted War answers how one particular war is to be fought. Your constraint cannot hold back the authoriser is an internal criticism of neither, because neither took that job. The object of the proposition is the form, not the men.

And this is not two men's separate oversight. Two different authorisers give the same limit, which says the limit does not lie in the choice of authoriser but in the form of taking authorisation. It takes both side by side to see it: read Clausewitz alone and you will think his reason of state was insufficiently pure; read On Protracted War alone and you will think it a problem with one particular line. Two systems with unrelated sources landing at the same place says the place is structurally determined.

6. One Objection, and How It Is Rightly Handled

The objection: Clausewitz was not naive. He knew policy could be bad policy; he says in the same chapter that if the policy itself is wrong, no correctness of military means makes up for it. So he foresaw that the authoriser is unconstrained, and simply held that this belongs to politics and not to the theory of war.

The objection holds, and this volume accepts it. To say ruling it out is exactly how the failure happens is a quibble: assigning a problem elsewhere is a legitimate declaration of range. This series makes heavy use of range declarations itself — silence on everything at the execution layer — so applying the same standard to a predecessor is the minimum.

So what this volume says narrows to one independently checkable sentence: after that declaration of range, what can the remaining constraint do, and what can it not. It can keep war from exceeding a given political purpose. It cannot make that political purpose itself constrained. Neither sentence is a criticism of Clausewitz. The first is his achievement; the second is what he expressly put outside. What is pointed out is only this: for someone who wants war constrained, this form alone is not enough — and that has nothing to do with whether his theory succeeds.

The correction makes the proposition steadier rather than weaker: it now needs no author's oversight, and denies no author's declaration of range.

7. The End of the Path

Ludendorff walks this path to its end: he swaps the authoriser from politics to the national will, and the criterion of the national will is itself. He expressly inverts Clausewitz's primacy of politics — politics obeys the commander, and the whole life of the nation obeys the war.

The consequences follow straight from the limit above. An authoriser that is its own criterion cannot say when to stop — the object of the war becomes the survival of the nation itself, which can terminate only in one side being destroyed in the sense of survival. It cannot tolerate any mechanism that tests it. And it cannot be falsified by any outcome.

The nature of the evidence has to be written down. The inference is drawn from the structure of the form, not from any particular formulation in Total War; the first-hand text was not obtained for this series, and the passages that would have turned on its specific wording did not pass verification and are not used. That material has been moved out of the release package as an isolated research item.

8. Cultivation Cannot Authorise Its Own Abolition

SAE's Theorem Two also constrains the widening of war: every widening has to pass the criterion again, and cannot carry on automatically because the last one passed.

Theorem Two does not take its authorisation from an external purpose. Its criterion comes from cultivation, and cultivation cannot authorise its own abolition.

A political purpose can demand that everything be sacrificed for it, including the very people it was meant to protect. A republic's survival can demand that citizens be turned into a supply of soldiers. A line can demand that individuals submit all the way down. Each of these demands is self-consistent in its own logic: an authoriser is entitled to dispose of what it authorises. Cultivation has no such entitlement, and the reason is categorial. Cultivation is not a goal. It is the very thing of a subject unfolding itself. A criterion that demands the abolition of that unfolding abolishes its own condition of holding. It is not that it violates cultivation. It cannot be said.

From which one operating discipline follows, for SAE itself: the criterion may not be hung on any particular good thing. The state's security, the republic's freedom, the people's liberation — hang it on one of those and you have an authoriser; and once you have an authoriser you have something it cannot hold back.

The discipline applies to SAE, and applies uncomfortably. It means SAE cannot say our criterion is there to protect the oppressed — that would make the interest of the oppressed the authoriser, and it can expand like any other. A criterion can only be a criterion.

9. A Criterion and a Reason to Mobilise Are Not One Thing

One thing is easily written down as a cost of that discipline, and it is not a cost. The discipline seems to make SAE's criterion hard to mobilise with: for the homeland, for liberation can be shouted, can send people to die; cultivation cannot authorise its own abolition cannot.

The difficulty's premise does not hold. The false premise is that a criterion and a reason to mobilise must be the same thing. Separate four layers and the difficulty dissolves.

A particular good can mobilise. What it cannot do is acquire, on its own, the standing to begin a war, to widen one, or to be exempt from review. What the discipline blocks was never people fight for their homes. It is those four words being taken to replace the threshold. A reason to mobilise enters the first layer and does not enter the second; it does not impersonate authorisation.

So the genuinely open question is a different one: how to let particular reasons for mobilising enter public action without impersonating the standing to authorise — a checkable division of labour between the language of mobilisation and the procedure of authorisation. That is a doable problem, not a dead end.

10. Where the People Stand in the Trinity

End of Book One, chapter one of On War: war is a wonderful trinity, composed of primordial violence, hatred and enmity — to be regarded as a blind natural force; of chance and probability; and of its subordinate instrumental character. The first belongs mainly to the people, the second to the commander and his army, the third to the government.

The arrangement leaves the people, in principle, without standing to question. The trouble is not that he put them in. It is where he put them. The people are the source of the war's energy, not its judge — and the consequence is structural: a corner constituted by blind impulse cannot be asked anything. You cannot demand reasons of a natural force. SAE's Theorem Four requires that a warring party cannot but be questioned, and that the questioning come from inside. Clausewitz's model has no such position.

This is not a belittling, and that has to be granted. Part of why he puts the people in that corner is that he had seen the mobilisation of the French Revolution — what he saw was a real, enormous force that had not existed before. His model describes that force accurately. What it does not describe is whether that force can ask a question.

11. Step Out of the Man, Into the Role

Book Three, on the military virtue of an army: to step out of this man and into the role that war assigns us to play — this is the form military virtue takes in the individual. And he adds that this virtue is a definite moral force, like any instrument whose effect can be calculated.

This sets the executor's no longer existing as a person, but as a role, up as a positive quality — and that touches a person is an end on one's own side.

It has to be put precisely. He is not praising coldness; he is describing how a good army actually works, and the description has an empirical basis — a unit that holds formation under fire really is acting by role rather than by individual reaction. The point of contact is the step of setting it up as a quality. Describing a phenomenon and setting it up as something to be pursued are two things, and he does the second: military virtue is a moral force an army ought to have, one that can be cultivated and can be calculated. The own-side half of the second formula is abolished there.

12. Friction: Two Faces of One Coin

Friction is usually taken to be Clausewitz's gift to SAE. It is — and it is at the same time something else, which has to be said first.

In war everything is very simple, but the simplest thing is difficult. These difficulties accumulate and produce a friction no one who has not seen war can accurately imagine. He gives the example of a traveller changing horses: at dusk you reckon on the last two stages, four or five leagues, nothing much; and at the second-to-last post there are no horses, or bad ones; then hills, bad roads, darkness — and after great trouble you reach the next stage to find wretched accommodation. Friction is the one concept that broadly corresponds to what distinguishes real war from war on paper.

But he says where friction comes from, very specifically, and that sentence changes everything. The military machine — the army and all belonging to it — is in fact simple, and appears on that account easy to manage. But we should bear in mind that no part of it is in one piece, that it is composed entirely of individuals, each of which keeps up its own friction in all directions. And then: a battalion is always composed of a number of men, the least important of whom, if chance so ordains, may occasion delay or even disorder.

Friction comes from this: an army is composed of individuals, and each individual has a direction of his own. And immediately after the description of friction, he writes the remedy. A powerful iron will overcomes this friction; it crushes the obstacles, but it also crushes the machine along with them.

The remedy he gives for friction and the definition he gives of military virtue point at one variable: how much of his own direction is left in the executor.

A stronger claim is not available, and has to be ruled out first. Someone will say: if friction comes from individuals each having their own direction, then reducing friction is reducing those directions — so an executor who has successfully stepped out of himself into the role no longer produces friction. That last sentence cannot be had, for two reasons.

First, friction is not one thing. The sources he lists fall into at least three classes — physical conditions; conditions of information and stamina; and individuals each having their own direction. Only the third bears on the executor's position as a subject. An executor wholly inside the role still meets bad roads, wrong intelligence and fatigue. He does not stop producing friction. He stops producing the third class. Second, no longer produces is a universal claim, and there is no material for it. Whether a man in the role never again hesitates, never adjusts by his own understanding, never pauses when orders and what he sees conflict — that is an empirical question about people, and neither this volume nor Clausewitz offers evidence.

So the only thing claimed is what can be read off the text: on the third class of friction, his remedy and his definition of virtue point the same way — reduce how much of himself is left in the executor. That is coherent in his own text, not stitched together here: friction is diagnosed in Book One, virtue prescribed in Book Three, and on the third class their directions coincide.

13. Three Classes of Friction

SAE accepts the concept, and it fills a gap of SAE's own. The four theorems are all at the layer of principle, and they assume by default that a criterion once set can be applied. §3 concedes the lightness of philosophy and the weight of the executor, but that is a concession of posture — it grants that the executor's task is harder without saying where the difficulty lies. Friction supplies that: at the front the criterion is not violated, it is worn away. Delayed information, distorted orders, fatigue, fear, weather, terrain — each degrades the input conditions for judging is widening now unavoidable. The criterion is still there. What is needed to judge it is not.

But SAE refuses his remedy, and not on moral grounds. Part of friction is not a defect to be eliminated, and that part is exactly the mark of the executor still existing as a subject. The narrowing has to be stated, because friction is very easily equated wholesale with subjecthood — and it is not one thing.

Only by folding the three into one does one arrive at friction at zero means the parts have been reduced to parts. That inference holds for the third class alone. So: the third class cannot be eliminated without abolishing the executor's position as a subject. The first two can be reduced, and reducing them abolishes no one.

And his own remedy still has to be criticized, because what he gives is a remedy that does not classify: a powerful iron will overcomes this friction — it does not distinguish whether what is to be crushed is bad weather or a subordinate raising an objection. And the half-sentence he himself wrote states the price of not classifying: it crushes the obstacles, but it also crushes the machine along with them. So what is criticized is not that he wanted friction reduced. It is that the remedy does not classify. He saw the price and took it for a price that must be paid. SAE says: for the third class, that is not a price. That is the abolition of the thing to be protected.

14. The Criterion Must Be Internalized Before the Friction

The third class cannot be eliminated without abolishing the executor's position as a subject, and all three classes wear away the conditions for applying the criterion. So that road cannot be walked by eliminating friction. What remains is: the criterion has to be internalized before the friction, not computed on the battlefield.

Deepening cultivation in peacetime is not only the accumulation of national strength. It is making the criterion into something that can be drawn on without on-site calculation.

What the inference does not claim has to be written down: it does not claim internalization guarantees correct judgement, and it does not claim internalized people will err in the same direction. What it claims is that some class of question can still be perceived, raised, and enter adjudication under pressure — and that is the only form in which a criterion still works outside its range.

A community that only starts wondering is this right once it is on the battlefield will have its criterion worn off, because that is precisely the moment when information is worst, time shortest, fear greatest. A community that has grown the criterion in during peacetime cannot have it worn off, because there is nothing there that needs computing. And this is taken from Clausewitz: he described the resistance; SAE turned it into a requirement on peacetime.

15. Left Unresolved in This Volume

SAE's criterion is hard to mobilise with is not an unresolved item — the difficulty's premise does not hold. What is genuinely open is a different question: how to let particular reasons for mobilising enter public action without impersonating the standing to authorise. This volume does not answer it, and it is a doable problem.

And one lesson on method goes with it: this series had previously taken this one cannot be filtered out as evidence that a boundary was working, when the actual reason it could not be filtered out was that the question itself had been put wrongly. A question that cannot be filtered out may be evidence that the boundary works — or evidence that the question was badly written.

Internalization handles only cases the criterion already covers. Meet a new situation it does not cover and friction's wear applies in full, with the judge in the worst conditions. That family of problems has been settled by division rather than centrally: the moment the criterion is most needed is the moment judging conditions are worst is ruled out entirely — it belongs to executive judgement, outside this series' range, and is acknowledged rather than booked as unresolved. The criterion must be internalizable stays inside — it is a question about how criteria are written, and internalization there has an exact meaning: move the criterion from the subtracting end to the adding end in advance, because on a battlefield there is no time to subtract.

And the trinity leaves the people without standing to question — while SAE has given no replacement model. In wartime, where does questioning come from, who protects it, how is it kept from being treated as sapping morale? This volume does not answer, and the question bears directly on whether Theorem Four can be operated at all.