Abstract

This volume exists to prevent a misreading. SAE's four theorems are most easily read as a new just war theory — determine who is the just party, and go to war and punish on that basis.

The overlap with the just war tradition is real, in three groups: a high threshold for going to war; means not exempted by the justice of the end; and war directed toward the restoration of peace. Three reasons differ, and the differences in range can be named one by one. Last resort as a moral duty can be offset by exigency; as a structural condition it cannot. Proportionality and SAE's bridge clause on means intersect rather than contain one another. Rousseau's principle of discrimination depends on both parties to a war being states, and fails in non-state war; SAE's does not.

Two disagreements: auctoritas principis does not block the self-defence of the powerless — what it blocks is that a private act of defence does not automatically carry the juridical powers of public war. And supreme emergency permits cancelling the other party's carrier of cultivation to preserve one's own; SAE declines that permission and bears the cost of declining. This is registered as a ground conflict, not as internal failure.

One structural divergence: applying these criteria in full requires a position from which to determine who is the just party, and among states no such position exists. Grotius faced that absence squarely, and SAE stands on the same side as he does — both decline to establish the position and turn instead to governing conduct in its absence. The cost of leaving it empty is that SAE issues no certificate of legitimacy usable to persuade a third party.

1. The Overlap, in Three Groups

The overlap is stated in full first; otherwise what follows will read as fault-finding, and the likeliest error in this volume is to state the overlap too thinly. The tradition's jus ad bellum requires just cause, right authority, right intention, last resort, reasonable hope of success and proportionality; SAE's Theorem One requires that the carrier of cultivation be lost, that cultivation in its mild forms can no longer continue, and that four gates be passed. Both hold that beginning a war requires a high threshold, and both place have the other roads been walked among the hardest of the conditions.

Jus in bello has two core principles: discrimination — non-combatants may not be made the direct object of attack — and proportionality. SAE has the bridge clause constraining means: no means may be used that forecloses the other party's possibility of reorganising itself as a subject of cultivation. Both hold that means acquire no exemption from the justice of the end. This is no small matter: it means both systems reject the end justifies the means, and in the domain of war that rejection is not self-evident. And the tradition requires right intention — the aim is the restoration of peace, not conquest, revenge or plunder — while SAE's Theorem Three requires that war be directed toward ending.

And one sentence must be added, not as argument but as fact. From Augustine through Aquinas to Grotius, this line drew war out of the strong may do anything and put it inside a set of criteria that could be publicly argued, cited, and used to indict rulers. Modern law of war grew from it. Every divergence in this volume proceeds on that premise.

2. Last Resort: a Moral Duty and a Structural Condition

The tradition's is a moral duty: you are obliged to try peaceful means first, and a ruler who goes to war without trying, even with just cause, is morally at fault. SAE's is a structural condition: cultivation in its mild forms can no longer continue. That is not a duty but a judgement about a situation — whether those forms are still open.

A last resort of the duty type can be offset by exigency; one of the structural-condition type cannot. In moral reasoning the stringency of a duty is generally modulated by circumstance: I ought to have negotiated first, but they were already massing; there was no time is an available defence, and often an accepted one. Requirements adjusting to feasibility is not a flaw in moral reasoning — it is how it normally works. A structural condition admits no such adjustment. Urgency does not change whether those roads are still open. If the road of negotiation remains open and was not walked, Theorem One's condition is not met, however pressed the agent.

One practical consequence: at the moment when error is likeliest — the other side massing, reports confused, the window for decision narrowing — the binding force of a duty-type criterion is at its weakest, because that is exactly when circumstances did not permit is most persuasive. The criterion is loosest when it is most needed.

But one cost must be acknowledged. A condition wholly unmodulated by circumstance may in practice appear inhuman, and may require an agent to walk through a procedure that has become meaningless when he knows the other side is about to strike. SAE's only answer is that the condition asks not was the procedure completed but is that road still open — and a road the other party has closed does not count as open. But that answer pushes the burden of judgement back onto the agent — and the agent is in the position least fit to make it: angry, exhausted, afraid, and with good reason to think that waiting longer will kill him. These are not his faults; they are his situation. This volume does not solve that; it registers it. (The full treatment of that burden is in volume seven.)

3. Proportionality and the Bridge Clause: Intersection, Not Containment

Proportionality: the harm caused by a means may not exceed the value of the legitimate end pursued. The bridge clause: no means may foreclose the other party's possibility of reorganising itself as a subject of cultivation. The two block partly different things, and the bridge clause is not an upgrade of proportionality.

Direction one — the bridge clause blocks what proportionality does not. Proportionality is a judgement of magnitude, and it requires someone to put both sides on one scale. In war that weigher is usually the belligerent itself: it weighs the value of its own ends and the harm to the other, and every party will hold its ends to be of very high value. The bridge clause is a judgement of direction and needs no scale. It asks not how heavy was this blow but is that road still there afterwards. A belligerent may sincerely hold its end to be of infinite value; it cannot sincerely hold that a means which forecloses the other's possibility of reorganisation has not foreclosed it — that is a fact about the structure of consequences, not an assessment of value.

Direction two — proportionality blocks what the bridge clause does not. A military action that forecloses nothing yet is grossly disproportionate to its aim — heavy civilian casualties in taking a secondary objective, where those casualties do not destroy the other party's capacity to reorganise — draws no verdict from the bridge clause, and does draw one from proportionality. This must be stated plainly: SAE has a gap here. The bridge clause handles does it foreclose; it does not handle is it excessive. And excess is a real moral dimension, not something to be waved off with we do not make judgements of magnitude.

Direction three — the relation is intersection, not containment. So it is inaccurate to say the bridge clause is more fundamental or more thorough. They are two lines cut in different directions, and a system holding the bridge clause and not proportionality lets through an entire class of acts that ought to be blocked. Whether SAE needs an additional criterion of magnitude is not answered here — and the difficulty is that a criterion of magnitude requires a weigher, and admitting a weigher re-establishes the very position this volume declines.

4. Discrimination: the Premise of Rousseau's Version

Rousseau gives the most famous formulation, in The Social Contract I.4: war is not a relation between man and man, but between State and State, and individuals are enemies only accidentally — not as men, nor even as citizens, but as soldiers; not as members of their country, but as its defenders. And then the conclusion: since the object of war is to destroy the enemy State, it is legitimate to kill its defenders as long as they are carrying arms; but as soon as they lay them down and surrender, they cease to be enemies or instruments of the enemy, and become once more merely men, whose life no one has any right to take. The reasoning is clean, and its force comes from that premise.

Rousseau's reason is the social contract: the parties to a war are two juridical persons, and individuals stand in an adversarial relation only while serving as units of war — lay down the arms and that status ends, and with it the adversarial relation. SAE's reason is different: the other party may not be struck out of the position of a subject of cultivation, whether or not he is at this moment serving as a unit of war.

Rousseau's principle of discrimination depends on the premise that both parties to a war are states, and fails in non-state war; SAE's does not depend on that premise. Rousseau states the premise very explicitly himself: in the same passage he holds that war between private persons can exist neither in the state of nature nor in the social state; that duels and private affrays do not constitute a state of war; that the private wars of the Middle Ages were an abuse of feudalism. He expressly excludes non-state violent conflict from war. So in civil war, insurrection, non-state armed resistance and decentralised resistance, individuals are enemies only accidentally does not obtain, because there are not two juridical persons.

The consequence is not hypothetical. The 15DD defence this series is concerned with is very often of exactly that kind: resistance in occupied territory, colonial revolt, self-defence in a society whose legitimate institutions have been destroyed. A slave uprising has no juridical personality — on Rousseau's framework the principle of discrimination does not apply there, not that it is violated, but that it does not apply. SAE's does not depend on juridical statehood, because a subject of cultivation need not be a state. And what Rousseau has that SAE lacks must be acknowledged: his reasoning yields an observable moment — the moment arms are laid down, a line both sides can identify. SAE's still exists as a subject of cultivation has no such moment; it requires judgement. In actual fighting, an observable line is more useful than a more correct line that requires judgement.

5. Legitimate Authority: a Reading Corrected

The tradition requires that only legitimate public authority may make war. An earlier version of this series read that as excluding the self-defence of the powerless. That reading is too broad, and it is corrected here.

Aquinas explicitly permits private self-defence. Summa Theologiae II-II q.64 a.7: an individual may use proportionate force to preserve his own life, even where the act foreseeably results in the assailant's death — provided the defender's intention is his own preservation rather than killing, and the force is not excessive. He says the act may have two effects: the preservation of one's own life, and the death of the assailant. The line he draws is elsewhere: a private person may generally not intend the assailant's death; intentional lethal coercion belongs to public authority acting for the common good. A soldier against an enemy, an officer against a felon, may intend lethal action under public authority — and even they are at fault if acting from private enmity.

So what auctoritas principis blocks is not the self-defence of the powerless. It is that a private act of defence does not automatically carry the juridical powers of public war — to initiate, to punish, to mobilise, to intend to kill. And Aquinas gives more than a slogan: a private person may not make war, because in a normal political condition he may assert his right before a superior's tribunal; nor can a private person summon the multitude that public war requires; conversely, the care of the common good is entrusted to public authority, which must repel enemies without as it uses public coercion against wrongdoers within. The substance of this gate is public jurisdiction, responsibility for the common good, and the prevention of a private quarrel being converted into public war.

6. SAE's Actual Distance from Him, and Where the Threshold Sits

The distance is smaller than it is usually taken to be, and that must be acknowledged. The four gates on the second face of Theorem One — do not open the second face on its own; see without substituting; genuine rising; the intervening party must have interests of its own — block the same class of things. An armed body claiming to represent the oppressed with no one having asked it fails the third gate; an intervener with no connection to the party intervened for fails the fourth; an action packaging a private quarrel as liberation fails the first. Aquinas's reasons are the same class of reason as what the four gates exist to prevent. To say that SAE rejects legitimate authority and is therefore more permissive is inaccurate.

The real difference is where the threshold is set: he sets it on the standing of the subject, SAE on the structure of the situation. Standing is a judgement about who this person is — does he hold public office, whence does his authority derive. Structure is a judgement about what this situation is like — has the carrier been lost, has the other party risen, is the intervener connected.

And the difference in range runs both ways. Direction one: a collective self-defence in a society whose legitimate institutions have been destroyed cannot be raised to the status of war under auctoritas principis — it may be an indefinite number of individual acts of justified defence, but it cannot be a war with standing to be initiated, to mobilise, to punish. And 15DD defence is very often exactly that situation. SAE's four gates do not ask this; they ask after structural conditions, and those have nothing to do with holding public office.

Direction two: standing is verifiable; structure is hard to verify. Does this king have the authority to declare war could be checked at the time; is the rising genuine requires judgement. This tells against SAE, and the cost must be stated: it is another appearance of one and the same problem — a structural condition is verifiable in principle while lacking, in practice, any settled way of being observed. The attribution of that problem divides in two: can this criterion be broken into publicly contestable sub-propositions is a question about the form of a criterion, and falls within what this series must answer; can it be broken down in wartime is an execution judgement, and does not.

7. Supreme Emergency: a Ground Conflict

Walzer holds that when the threat facing a political community is extreme enough to destroy its very existence, the limits of jus in bello may be breached. This is the strongest direct challenge to the bridge clause on means, because it does not claim the clause is wrong. It claims there exists a tier of exception at which rules give way to survival.

What supreme emergency permits is: to preserve our own carrier of cultivation, the other party's may be cancelled. SAE declines that permission and bears the cost of declining. The registration must be precise. To write this does not hold within SAE would not itself be wrong — it indeed cannot be uttered within SAE — but it reads as a determination that the concept is self-contradictory, and Walzer is not inside SAE, so that determination does not apply to him. The accurate statement is: this is SAE's choice of ground, not a determination about the internal consistency of the concept. Walzer's argument is coherent within his own framework, and he advanced it in full awareness of the difficulty.

The argument: the bridge clause prohibits foreclosing the other party's possibility of reorganising itself as a subject of cultivation — and what supreme emergency asks to exempt is precisely that class of means; it exempts what is ordinarily forbidden, which is to say the most thorough. So the content of the exception is: when one party's carrier of cultivation is threatened, the other's may be cancelled. And that cannot be uttered within SAE, for a reason of the same form as volume two's: cultivation cannot authorise its own cancellation. A criterion in the name of preserving cultivation whose content is cancelling the other party's cultivation cancels the condition of its own standing. It has not violated a rule; it does not obtain.

But the cost must be acknowledged, and the acknowledgement cannot be formal. Walzer's exception was not set down carelessly; it answers to a real situation — there are moments at which keeping the rules may mean the destruction of a community. SAE's refusal of the exception is the bearing of that cost: it issues no exception clause available to be invoked at such a moment, and a community in that situation gets no dispensation from SAE. That bearing must be written down; the situation may not be pretended away. And what SAE bears it on must also be written: not that the situation will not arise, but that once the exception is established, any party may claim to stand at that tier, and no one has standing to verify the claim.

8. That Position

The foregoing can be discussed item by item. One thing cannot, because it determines that the two are not of one kind. The full application of these criteria requires a position from which to determine who is the just party, and among states no such position exists. Just cause, right intention, proportionality — each of these judgements must be made, and among states there exists no maker of them whom all parties acknowledge.

The position then has only these destinations: left empty, and the criteria cannot be applied; occupied by the belligerents themselves, and each judges itself just, so the criteria lose their discriminating power; or occupied by some third party claiming standing — and whence that third party's standing? Historically the position has been occupied in turn by the Church, by the sovereign state itself after Westphalia, and in our time by something called international society. All three have had real functions, and all three have had their standing questioned.

9. Grotius Faced It Squarely, and SAE Stands on His Side

An earlier version of this series wrote that the position has only three destinations, and none of them holds. That exhaustive list omits a fourth, supplied here. Grotius knew clearly that among sovereign states there is no common judge of the kind available to litigants within a civil state — that absence is the very problem his theory of war addresses, not something he failed to see. And his answer has three tiers.

First, the tier of objective natural right: contradictory claims cannot both be true in the same respect, so objectively a war cannot ordinarily be just on both sides. Second, the tier of subjective culpability: a party whose cause is objectively defective may nonetheless act from excusable or invincible error — so objectively both sides cannot be just and both sides may have justified belief in their own justice do not conflict. Third, the tier of the law of nations: the law of nations may confer bilateral legal effects on a formal public war — the validity of capture, the standing of acts in war — even where the natural justice of the original cause cannot belong to both. The third tier does the most important work: it lets every soldier and every third party know which legal consequences of the war apply without first settling the dispute over just cause. So the fourth destination of that position is: do not adjudicate; govern conduct in the unadjudicated state.

The criticism that remains live is that the response yields no authoritative third-party determination. This tradition has no answer to the question does not. And here SAE and Grotius are on the same side, which changes the shape of this volume. Both decline to establish the adjudicating position, and both turn instead to governing conduct in the unadjudicated state. The difference lies in what is governed: Grotius governs the symmetrical distribution of legal effects — both sides fall under the same rules on capture, the status of prisoners, the validity of acts in war, without asking whose cause is objectively just. SAE governs whether the standing as a subject is preserved — the bridge clause asks whether, after the fighting, the other party can reorganise itself, and likewise does not ask whose cause is objectively just. Neither regime requires an adjudicator, and both are verifiable from outside. At this point SAE is not opposed to the tradition; it has taken another road in a direction the tradition had already opened.

10. Kant: the Position Cannot Be Established

Theorem One does not ask whether this war is just. It asks: has the carrier of cultivation been lost; can cultivation in its mild forms no longer continue; do the four gates pass. These are judgements about structure, requiring not standing but sight.

And Kant, from within Recht, holds that the position cannot be established. Perpetual Peace, Sixth Preliminary Article: war is only the sad recourse in the state of nature — where there is no tribunal which could judge with the force of law — by which each state asserts its right by violence; and in it neither party can be adjudged unjust, for that would presuppose a juridical decision; in place of that decision the issue of the conflict, as in a so-called judgment of God, decides on which side the right is. But between states no punitive war is conceivable, because there exists between them no relation of superior and inferior.

The position of this passage matters: it speaks from within Recht, not against Recht from outside. Kant does not say the moral concerns of just war theory are wrong. He says the adjudicating position does not exist among states — not ought not to be established but cannot be, because establishing it requires a juridical decision for which there is no judge, and because punishment presupposes a relation of superior and inferior which does not obtain. So SAE need not draw this distinction itself: someone on that road has already drawn it. (In the Doctrine of Right §60 of 1797 Kant employs the concept of an unjust enemy, which does not sit easily with this; that is treated in volume one.)

11. The Cost of Leaving It Empty

The cost of leaving the adjudicating position empty is that SAE issues no certificate of legitimacy usable to persuade a third party. Just war theory issues that certificate. It is one of the tradition's real functions and one reason it has been needed for two thousand years — a community deciding to resist needs to explain to its own members, to possible allies and to posterity why the thing is right. Our war is just is a sentence that can be said, quoted, and written into a declaration.

SAE's criterion produces no such sentence. What it produces is in this structure, entering war constitutes a cannot-not — a sentence meaningful only where the specific structure obtains, and acceptable only to someone willing to go and check that structure. This is not clumsiness of expression but structural necessity: once SAE issues that certificate, the adjudicating position has been re-established, because a certificate requires a signatory.

From this follows one requirement on the reader. The place where the four theorems are most easily conflated with just war theory is exactly where they most resemble it — high threshold, limits on means, peace as the aim. So it must be said flatly: the four theorems do not determine who is the just party. They tell no one you are right and he is wrong. They determine one thing: whether, in this specific structure, entering war constitutes a cannot-not, and which practices, once entered, turn the thing into something else. This determination faces inward — a criterion turned inward is reflection; turned outward it is judgement. A use of the four theorems to argue that one is oneself just, and the other unjust, has read them as a just war theory, and that is the misreading this volume exists to prevent.

12. One Asymmetry That Must Be Acknowledged

Leaving the adjudicating position empty does not exempt SAE from misuse. Anyone who wants a war can say the mild forms are exhausted, just as he can say our war is just. Both require external verification, and both are hard to verify in wartime. The difference is only this: the former can be broken into sub-propositions that can be publicly argued and separately evidenced — which roads were walked, what did the other party answer, which measures remain untried — while the latter points at a holistic judgement.

Three things must be stated precisely, because decomposability is very easily read as verifiability. First, each item after decomposition can still be disputed — whether a road was genuinely open, whether a given response constitutes closure, whether an attempt was sufficient: none of these has an automatic answer. Decomposability is not automatic transparency of fact. Second, what decomposability buys is localisation of dispute, assignable burdens of proof, locatable error, and a handle for revision — not the truth. Third, a holistic judgement cannot be refuted is also too strong: it can be opposed, there is simply nowhere for the opposition to take hold — an opponent can only deny it wholesale, and a wholesale denial is as unadjudicable as the wholesale assertion.

What survives the narrowing: SAE's criterion directs dispute toward places where it can be handled item by item, and just war theory's does not. This is an advantage of degree, not of kind. And the realisation of that advantage depends on someone actually asking — which is Theorem Four's business, and one of SAE's thinnest places.

13. Left Unresolved in This Volume

Proportionality blocks a class the bridge clause does not — foreclosing nothing, yet grossly disproportionate. Whether SAE needs an additional criterion of magnitude is undecided, and the difficulty is that a criterion of magnitude requires a weigher, while admitting a weigher re-establishes the position this volume declines. Without it, an entire class that ought to be blocked is not blocked.

Structural checks are harder to verify than checks of standing, and the attribution of that problem is handled separately. Supreme emergency is declined partly on the ground that any party may claim to stand at that tier and no one has standing to verify the claim — but the same words can be turned against SAE's own threshold, as this volume concedes. The difference is one of degree, and a difference of degree needs its size stated; this volume does not quantify it.

Finally: Rousseau's discrimination supplies an observable moment — arms laid down — while SAE's still exists as a subject of cultivation does not. Whether SAE needs an observable operational marker has not been considered.