Standalone Essays · Triad of Stewardship

What Is Owed to a War Criminal

There is a particular kind of man who appears in every war. He is not the soldier who kills across the line of battle, terrible as that is. He is the one who waits until his enemy is disarmed, bound, and helpless — and only then begins. He tortures the man who can no longer resist. Sometimes he photographs it. Sometimes he keeps trophies. And when the war is over and he is caught, the question of what to do with him tends to arrive already answered, in the form of an impulse: do to him exactly what he did.

The impulse is understandable. It may even be righteous in its origin — it is grief with nowhere to go. But it is the wrong answer, and the reasons it is wrong turn out to be the whole of the problem. To reject it cleanly, and to see what should be done instead, is to work out an entire ethic of justice after atrocity: what is owed to such a man, what is forbidden to us in dealing with him, and how to carry out the ugly necessity without becoming a smaller version of the thing we are judging.

The crime is the crime

Begin with the one distinction that organizes everything else. The judgment of the crime is categorical. The machinery of answering it is not.

Whether a man is guilty of an atrocity, and how grave it is, does not depend on his nationality, his uniform, his side, or how near he stands to us. A foreign general and a local conscript who commit the identical act have committed the identical crime and earn the identical verdict. The severed ear does not become less of a crime because the hand that took it wore the enemy’s flag, nor more of one. To let nationality or nearness bend the verdict is not justice — it is tribalism wearing the robes of a court.

What nationality and relationship do bear on is narrower and later: which lawful instrument is available to carry the sentence out. That is a question of machinery, and it comes only after the judgment is fixed. Confusing the two — letting the machinery reach back and soften the verdict — is the characteristic corruption of victor’s justice, and the framework guards against it from the first step.

The rabid dog and the cage

Timing decides which kind of act we are even contemplating.

On the battlefield, against an armed man still fighting, lethal force is permitted to stop the threat. The old image is exact: a rabid animal is put down while it is biting, and the justification is the ongoing danger, not any verdict about its past. But note what this permission is and is not. It licenses neutralizing a threat. It does not license punishing a crime. A soldier may not execute a suspected war criminal on the field as retribution, because the battlefield cannot supply the thing a punishment requires — a verified, deliberate finding of guilt beyond reasonable doubt. If the man can be captured, he must be.

And the moment he is captured, everything changes. He is no longer a threat; he is a prisoner, disarmed and wholly dependent on his captor for survival. The obligation not to harm him does not lapse at that moment — it intensifies, precisely because his vulnerability is now total. The rabid dog is in the cage. Killing it now, quietly, “to be done with it,” is a different act from shooting it mid-bite, and it demands a different and higher justification: the verified diagnosis, produced by a process, not the horror of the photographs.

So the phase sorts the logic. In active combat, the logic is incapacitation. At capture, it is custody. After the war, when institutions however thin can function again, it is justice — and justice, unlike the battlefield, can meet the standard of verification, and therefore must.

Three kinds of prisoner

Not every captured man is a war criminal, and the framework refuses to treat them alike. Judged on the act itself, blind to nationality, captured combatants fall into three bands.

The lawful combatant who fought within the laws of war and committed no atrocity is a prisoner of war, held only to keep him off the battlefield, and repatriated when the war ends. He is owed his freedom back.

The reducible offender — the low-level, coerced, or situational participant who committed no atrocity of his own hand — is owed accountability oriented toward repair. This is the domain of the community court and the graduated sentence, of disclosure and restitution that can earn genuine mitigation. Most people swept into the machinery of atrocity belong here: pressured, junior, afraid. The framework’s restorative instinct exists for them, and it is a mistake — a grave one — to let the worst cases harden our treatment of these.

The irreducible offender is the torturer: the man who committed a verified, willing atrocity, the deliberate infliction of suffering on the helpless. He is owed permanent removal from the community of the living-among-others, and he is owed it through a fair process — but he is not owed reconciliation, and he cannot buy his way back with a confession.

Between the reducible and the irreducible runs a line that must not be blurred, because it is where mercy is either kept honest or turned into enablement. An act evil in itself — torture above all — is never laundered by orders, by pressure, or by a poisoned unit culture. The man who tortured under command remains fully accountable for his own hand. What the coercive culture does is not excuse him; it pulls others toward the line alongside him. The commander who knew and either approved or looked away is culpable by that knowledge — the law has long called this command responsibility, from the Yamashita precedent to Article 28 of the Rome Statute — and so are the fellow soldiers who knew and encouraged, or covered, or simply kept silent. The code of silence is part of the machinery that produced the act. But none of that lightens the load on the hand. Coercion moves guilt up and outward; it never moves it off.

Reconciliation is not immunity

The great achievements of transitional justice — Rwanda’s community courts, South Africa’s Truth and Reconciliation Commission — are held up, rightly, as the endpoint a wounded society should reach for. Their goal is not the defeat of the enemy but the restoration of the community. It is easy to read that as an argument for amnesty even for the worst, and to feel that permanence of punishment betrays the reconciling spirit.

It does not, and the reason is a truth about forgiveness the framework takes seriously elsewhere (see Forgiveness and Its Limits). Forgiveness is a debt, and it is owed to a specific person — the one who was harmed. Not to his family, not to his community, not to the state, and not to history. Only the injured party holds the standing to forgive the injury. And a crime that kills its victim extinguishes the only person who could ever have given that forgiveness. In that exact sense the gravest atrocities are structurally unforgivable: not because we are too hard-hearted to forgive them, but because the one soul with the authority to do so has been placed permanently beyond reach.

So reconciliation and immunity are not the same thing, and the framework separates them. Reconciliation is what a society builds around the removed perpetrator — the peace it makes with itself, the future it refuses to poison. It is real, and it is necessary. But it is not a door opened for the torturer. Mercy that returns a predator to the vulnerable is not mercy; it is the financing of the next harm.

Without becoming the thing

Two prohibitions bind everything above, and they are absolute — not guidelines to be traded against necessity, but the walls that keep justice from decaying into what it judges.

The first: no torture, ever. There is no exception for the severity of what the prisoner himself did. The impulse we began with — do to him what he did — fails here, finally and completely. To mirror the torturer is to adopt his ethic: that a helpless person’s suffering is available to be used. The moment we accept that premise in order to punish him, we have conceded his central claim and become a variation on him. The prohibition is not softness toward the criminal. It is the refusal to let him remake us in his image, which is the one victory we can always deny him.

The second: the same justice for our own side. A process reserved for the enemy is not a court; it is a reprisal in a court’s clothing, and a wounded population sees through it instantly. Selective justice is worse than no justice, because it destroys the very expectation of fairness that justice exists to build. If we will not try our own torturers by the identical standard, we have no standing to try theirs — and worse, we hand the enemy’s torturer the one thing that could redeem him in his followers’ eyes: the status of a victim of a rigged tribunal.

This is also why, for the foreign perpetrator, the forum matters as much as the verdict. A neutral international tribunal is preferable to trial by the victim nation, not because the foreign criminal is owed less than a local one — he is owed exactly the same — but because a victim sitting in judgment of its own brutalizer is the structural definition of the temptation we are trying to resist. Where a neutral forum can be reached, it should be. The victim-state trial is a last resort, and it should know itself as one.

The manner of the thing

Suppose the process is done, the diagnosis verified beyond doubt, the sentence permanent. There remains the question of how, and it is not a small one, because it is where a martyr is either made or refused.

These men are, in the main, disappointingly ordinary. They are not the towering monsters they are made into after the fact. They are cowards who required a dozen armed comrades and a bound victim before they were brave, and many of them feed on the very reputation for monstrousness that horror confers. To make a spectacle of their punishment — to give them a stage, a defiant last word, a dramatic death — is to hand them the one reward the sentence could still deny them.

So the manner should be plain, deliberate, bureaucratic, and witnessed. Quiet in affect, but public in record. Not secret — secrecy breeds the myth — but stripped of theater. And the true weapon against the martyr’s legend is not silence, which merely leaves the story to his admirers to write, but the exhaustive, factual, unglamorous record of exactly what he did, preserved in a register too dull to be sung. Let the accounting be boring and complete. The truth, encoded plainly, is the answer to the myth.

The hardest question

Whether the permanent removal of the irreducible offender should extend to execution is the point on which reasonable frameworks divide, and honesty requires naming where this one stands and where it departs from the current consensus. Much of modern international law — the Rome Statute among it — has moved away from capital punishment entirely, resolving even the gravest cases through life imprisonment. This framework does not follow it there, and it is worth being exact about why, and about the reasons it refuses.

It does not rest the case on cost. The argument that execution is cheaper than lifelong confinement is both false wherever real verification is practiced — the safeguards cost more than the cell — and corrosive in principle, because a cost-justification licenses killing anyone whose confinement is expensive. That is a horror engine, and it is barred.

It does not rest on deterrence as its foundation, either. That the next would-be torturer should know a quiet, un-glorious death awaits him is a real and legitimate good — but a secondary one. Certainty of consequence deters far more reliably than its severity, and a justice grounded primarily in making an example of a man has begun to use him rather than to judge him.

The ground is narrower and harder: the protection of the innocent from an irreducible threat. A verified atrocity-perpetrator does not cease to produce harm when the cell door closes. He carries his culture with him, and among the imprisoned — many of whom are the reducible, the salvageable — he finds fresh material to corrupt. The only confinement that would truly neutralize him is indefinite isolation, which is itself a form of deliberate degradation the framework forbids. Cornered between an execution and a cruelty it has already ruled out, the framework resolves toward the execution — carried out under the strictest verification, by a legitimate and even-handed forum, plainly and without spectacle.

And this is not a license for the state to kill freely. It is the exact opposite. The same tradition suspends ordinary capital punishment until a justice system can prove it deserves the trust such an irreversible act requires. The war criminal’s execution is not an exception to that caution — it is that caution’s condition being met: the guilt witnessed and documented beyond doubt, the forum neutral, the standard applied to our own as readily as to theirs. Strip away any one of those and the sentence is no longer justice. It is only the rabid-dog impulse, dressed for court.

What the discipline is for

The man who tortures the helpless is betting on something. He is betting that his cruelty will provoke cruelty in return, that his enemies will either sink to his level and prove his cynicism right, or recoil into a paralysis that lets him escape. Both of those are victories for him.

The whole of this ethic is the refusal of that bet. It says: we will judge you, exactly and without flinching, by a standard we hold ourselves to as well. We will not torture you, because we decline to be remade in your image. We will not make you famous, because your name is not worth the breath. And if the protection of the living requires your death, we will bring it plainly, quietly, and with a full and boring record of why — and then we will get back to the harder work of rebuilding a world in which men like you are rarer.

That is what is owed to a war criminal: not vengeance, which he has earned but which would cost us too much to pay, and not mercy, which is not ours to give on the dead’s behalf — but justice, verified and even-handed, delivered by people who took the greatest care precisely where it would have been easiest to take none.

Part of the Triad of Stewardship’s Standalone Essays. Companion pieces: Forgiveness and Its Limits (forthcoming) · The Mandates of Conflict · Criminal Justice.