THE RIGHT OF RESISTANCE AND THE ETHICS OF FORCE

Full report

On this page

Authority does not become morally innocent because it wears a uniform.

Rebellion does not become morally innocent because it calls itself liberation.

Those propositions are not rhetorical bookends. They are the beginning of the problem. Political ethics becomes dishonest when it grants one side a moral exemption in advance. States point to legality, sovereignty, procedure, elections, uniforms, courts, and public office. Insurgents point to oppression, history, liberation, popular anger, martyrdom, and necessity. Neither vocabulary is self-validating. Legal authorization can coexist with atrocity. A righteous cause can coexist with atrocious means. Contemporary political philosophy itself distinguishes sharply between having a legal obligation and having a moral duty to obey; the existence of a legal command is not proof of its moral authority.

The central conclusion of this investigation is therefore conditional rather than pacifist or revolutionary:

Force can sometimes be morally legitimate in resistance to oppression, but the justification is exceptional, burdened, discriminating, and revocable. A serious right of resistance arises most plausibly where grave and persistent violations of basic rights are occurring; where government has converted public power into an instrument of predation; where less harmful means cannot reasonably protect those under threat in time; where those claiming to resist can plausibly speak for or protect the oppressed rather than merely substitute themselves for them; where the expected harms are proportionate to the wrong resisted; and, above all, where violence against civilians is rejected rather than rationalized.

That answer is not a license. It is a structure of restraint.

The most important of those restraints is the civilian line:

CIVILIANS ARE NOT SUBSTITUTES FOR POWER.

A government is not its population. An army is not its taxpayers. A ruling party is not everyone who voted for it. A political constituency is not a military formation. A nationality is not a chain of command. A civilian's bad opinions do not automatically convert that civilian into a legitimate object of physical attack. International humanitarian law is categorical at the level of its governing principle: parties to armed conflict must distinguish civilians from combatants, and attacks may be directed against combatants or military objectives, not civilians as such. The rule binds state and non-state parties in armed conflicts to which it applies.

That distinction will carry most of the moral weight of this report.

The moral grammar of resistance

Political arguments about force often begin too late. They begin after the words have already done the moral work. A government calls its conduct “law enforcement.” A resistance organization calls its conduct “armed struggle.” Each expression is capable of concealing what actually happened.

The first discipline is therefore conceptual.

The following are working definitions for moral analysis, not claims that every legal system, historian, or political movement uses the terms identically. “Terrorism,” “revolution,” “civil war,” and even “violence” remain contested concepts. Philosophical accounts commonly understand terrorism as a form of violence using fear or terror as an intermediate political mechanism, while philosophical accounts of revolution emphasize profound transformation of political order outside ordinary consensual or legal processes.

Several distinctions follow.

All force is not morally equivalent. The moral difference between pulling a child away from an attacker, imprisoning a convicted murderer after due process, beating a prisoner during interrogation, breaking an illegal restraint, damaging an empty object, and massacring civilians is not erased by the fact that each involves physical force. Moral analysis asks at least: who is acting; against whom; to avert what wrong; with what authority; by what means; with what intention; under what alternatives; causing what foreseeable harm?

Calling all of these “violence” and stopping there is not moral seriousness. It is taxonomy masquerading as judgment.

The same applies to property.

Property destruction is not ordinarily morally equivalent to violence against persons. Persons possess bodies, agency, consciousness, relationships, plans, and interests that cannot be reduced to their possessions. Killing someone cannot normally be repaired by compensation. A broken window usually can. A destroyed vehicle usually can. This difference is one reason theories of defensive force and just war place extraordinary moral weight on bodily integrity and civilian life.

But the distinction is not absolute. Destroying a family's home, food supply, medical facility, cultural patrimony, or indispensable civilian infrastructure can expose people to grave physical danger, displacement, or destitution. International humanitarian law therefore protects civilian objects and restricts attacks to military objectives under its governing rules.

The correct distinction is therefore not:

persons matter; property does not.

It is:

harm to property and harm to persons belong to different moral categories, but destruction of property becomes increasingly grave as its relationship to human survival, security, livelihood, identity, and foreseeable bodily harm becomes closer.

This matters politically. Governments routinely inflate damage to property into a language suggesting equivalence with homicide. Some radicals make the inverse mistake, treating “property” as though destroying anything classified under that noun were morally trivial. A burned family home, a destroyed hospital, a smashed shop belonging to an uninvolved person, an empty governmental symbol, and a piece of replaceable machinery are not morally interchangeable because all are objects.

The same discipline applies to coercion. Gandhi's satyagraha was nonviolent, but it was not politically inert. Martin Luther King Jr. explicitly described nonviolent direct action as a means of producing a crisis that made continued refusal to negotiate politically untenable. Nonviolence is therefore compatible with pressure, disruption, confrontation, and coercion. The alternative to violence is not necessarily polite request.

This is why civil disobedience occupies a conceptual space between lawful protest and organized forcible resistance. Classical accounts describe it as public, conscientious, normally nonviolent lawbreaking designed to change law or policy.

A final preliminary distinction is indispensable:

Illegality is not identical to immorality, and legality is not identical to morality.

Political philosophy treats that distinction as foundational. A tyrannical regime can create valid legal obligations in the descriptive sense—tax rules, prohibitions, administrative duties—while failing to generate a moral duty of obedience merely because those commands are law.

That is not anarchism. It is the refusal to let the state serve simultaneously as defendant and moral judge in its own case.

Authority, state force, and the historical right of resistance

Max Weber's famous formulation is nearly unavoidable because it clarifies both the power and the danger of the modern state. In Politics as a Vocation, Weber characterizes the state as the human community that successfully claims a monopoly on the legitimate use of physical force within a territory. He immediately notes that force is neither the state's normal nor its only means; rather, it is a means specific to the state in the sense that other actors' authorized force is ordinarily derivative of state permission.

This proposition is often misused.

Weber was not proving that whatever violence the state authorizes is morally justified. The statement identifies an institutional characteristic of modern statehood: the state successfully asserts a special claim to authorize physical coercion. “Legitimate” here belongs to a sociological account of authority and recognition; it cannot simply be substituted for the conclusion of moral philosophy.

Otherwise the reasoning becomes circular:

The state alone may legitimately use violence because it is the state. How do we know it is the state? Because it successfully claims the legitimate use of violence.

That might distinguish institutions. It does not establish justice.

The state's monopoly does, however, matter morally. A reasonably just public authority can supply things private violence ordinarily lacks: published law, institutional continuity, procedural accountability, generality, adjudication, democratic authorization, chains of responsibility, and a public rather than purely factional claim to act. Those properties can create presumptions in favor of state coercion. They cannot create immunity from judgment.

The state also possesses capacities that make its wrongdoing unusually dangerous: prisons, police, armed forces, intelligence agencies, tax systems, administrative databases, borders, and the authority to classify its own opponents as criminals. The fact that state violence is bureaucratically organized can make it more accountable—or more systematic.

State and non-state violence are therefore neither morally identical nor morally separated by an absolute categorical wall.

They differ institutionally.

They do not differ in the elementary fact that a civilian intentionally killed without justification is still dead.

International humanitarian law reflects part of this distinction. In non-international armed conflict, humanitarian rules increasingly bind government forces and organized non-state armed groups alike; the ICRC emphasizes that the relevant rules are equally binding on both. At the same time, domestic law ordinarily does not give rebels the combatant privilege enjoyed by qualifying combatants in international armed conflict: states have historically refused to surrender their legal power to punish rebellion merely because rebels comply with humanitarian rules.

That creates an important asymmetry:

legal status can remain asymmetric even where moral duties regarding civilians are symmetric.

A rebel may be legally punishable for participating in hostilities while still being morally justified in resisting. A government soldier may be legally authorized while participating in an unjust system. Neither legal proposition settles the ethical one.

The history of resistance theory developed precisely because political thinkers understood that public authority could become predatory.

Aquinas. Aquinas does not offer a modern revolutionary charter, and it is careless to enlist him as a simple theorist of private tyrannicide. His treatment of sedition is subtler. Tyranny is government directed toward the ruler's private good against the good of the multitude, and Aquinas can place responsibility for social disorder on the tyrant himself. At the same time, resistance is conditioned by the danger that upheaval may produce harms worse than the tyranny resisted.

The important inheritance is not “kill tyrants.” It is that political order is morally subordinate to the common good. Stability is valuable, but stability in service of systematic predation is not the same good as political peace.

Grotius. Grotius's Rights of War and Peace was written in the shadow of catastrophic European war and attempts to subject political violence to rational legal restraint. His treatment of subjects resisting superiors is more restrictive than later Lockean resistance theory. He fears the dissolution of political order that follows when every subject privately claims authority to wage war, yet his framework does not make sovereignty metaphysically unlimited. The Grotian contribution is therefore two-sided: force must be constrained by public law, but rulers are not exempt from the moral order merely because they rule.

Locke. Locke supplies the canonical liberal theory of resistance. Government exists through delegated power for preservation of the political community. When rulers arbitrarily place themselves over people's lives, liberties, and estates, especially when they seek to reduce subjects to slavery, they violate the trust on which governmental power rests. Locke describes such rulers as putting themselves into a state of war with the people, thereby forfeiting the claim to continued obedience.

Locke's argument is often vulgarized into “government does something wrong, therefore revolt is legitimate.” His actual structure is more demanding. The breach must concern the basic purposes for which political authority was entrusted. He also observes that populations normally tolerate many injuries before resisting and stresses patterns of abuse rather than every isolated injustice.

That distinction remains crucial. A right of resistance cannot plausibly arise every time one loses a policy dispute. If routine political defeat created a right to war, constitutional government would be impossible.

Revolutionary republicanism. The Atlantic republican tradition radicalized the fiduciary view of government. Algernon Sidney attacked theories of unanswerable royal authority and defended government grounded in public good and consent; such currents fed later republican theories in which political authority could be altered when it betrayed its constitutive ends. The American Declaration of Independence states the familiar version: government derives just powers from consent, and a people may alter or abolish a government destructive of the ends for which government exists; yet the same document invokes prudence against overthrow for light and transient causes.

The conjunction matters more than the slogan:

right of resistance plus prudential restraint.

Without the first, government becomes unanswerable. Without the second, every faction can baptize impatience as revolution.

Abolitionism exposes the limit of obedience most starkly. Chattel slavery asks the enslaved person not merely to obey a law but to accept legal transformation into an object of another person's domination. Frederick Douglass's political thought explicitly treated slavery as a profound moral wrong and engaged the legitimacy of self-defense and resistance. David Walker's Appeal likewise stands in the radical abolitionist tradition that rejected the idea that legally sanctioned enslavement generated a duty of passive submission.

Slavery therefore provides one of the strongest arguments against a general moral doctrine of obedience: a legal regime can make obedience itself the mechanism by which an extreme injustice is maintained.

The Haitian Revolution makes that abstraction historical. The uprising in Saint-Domingue beginning in 1791 became the most successful modern slave revolt, ending in emancipation and the establishment of independent Haiti. Whatever one concludes about particular acts committed during that long and brutal revolution, the proposition that enslaved people were morally required to remain enslaved because slaveholders had positive law on their side is extraordinarily difficult to defend.

Anti-colonial resistance expanded the same problem from persons to peoples. United Nations General Assembly Resolution 1514 of December 14, 1960, declared colonial domination contrary to the right of peoples to self-determination and helped formalize the international normative rejection of colonial rule. Additional Protocol I to the Geneva Conventions later included certain conflicts against colonial domination, alien occupation, and racist regimes within its treaty framework for international armed conflict—a provision historically important but legally and politically controversial, and one of the provisions criticized by the United States when it declined to ratify Protocol I.

None of that proves that every anti-colonial war was morally justified in all its conduct. It proves something narrower and more important: international legal development itself eventually ceased treating imperial sovereignty as an automatic answer to the moral claims of colonized peoples.

Contemporary philosophy presses further. Magnus Reitberger challenges the idea that recognized state authority must be a necessary condition of a just resort to force, arguing that rights of self-defense and protection of basic human rights can ground force independently of state status. Christopher Finlay's work on legitimate armed resistance similarly treats rebels as moral agents who can be assessed through a theory of justified resistance rather than categorically excluded from ethics because they lack state recognition. Candice Delmas broadens the concept of resistance beyond the polite boundaries of classical civil disobedience, arguing that grave injustice can sometimes generate duties to resist through a wider repertoire of political conduct.

The authority problem is therefore not resolved by choosing between two absolutes:

state recognition cannot be sufficient for morally legitimate force; complete absence of public authority cannot be irrelevant either.

A serious theory should ask what legitimate authority is supposed to contribute morally.

It contributes at least four things:

representation: the actor plausibly acts for those whose rights are invoked rather than merely claiming them;

accountability: agents can be restrained, judged, replaced, or punished;

publicity of purpose: the violence serves defensible political aims rather than private revenge;

capacity for ordered peace: the movement can plausibly replace wrongful domination with political order rather than permanent predation.

Recognition by foreign ministries is at most an imperfect proxy for these properties.

Just war, rebellion, and the conditional legitimacy of force

Just-war theory is valuable here because it refuses both easy positions.

It does not say that war is always forbidden.

It does not say that a good cause makes war good.

Instead, it divides the problem. There is the justice of resorting to force and the justice of conduct within force. Traditional criteria include just cause, legitimate authority, right intention, necessity or last resort, proportionality, discrimination, and—on many accounts—a reasonable prospect of achieving the just aim. Just-war theory further insists that discrimination and proportionality constrain conduct even when the underlying cause is just.

For resistance to oppression, the framework can be reformulated as follows:

This framework immediately raises the hardest structural objection.

If “legitimate authority” means “recognized sovereign state,” then classical just-war theory appears to exclude rebellion by definition.

The result would be perverse. Imagine a government conducting systematic torture or extermination while retaining diplomatic recognition. The state's forces would pass the authority test because they belong to the institution committing the wrong. Those resisting extermination would fail because they do not.

That cannot be an adequate moral theory.

Reitberger's critique directly challenges the privileged status of a state-only authority requirement. Contemporary theory accordingly tends to ask whether the authority condition can be reconstructed in functional moral terms: representation, collective authorization, accountability, respect for rights, and capacity to establish a defensible peace. Finlay's work on rebel ethics similarly rejects the assumption that lack of sovereign recognition places armed resistance outside moral evaluation altogether.

But eliminating authority entirely is also dangerous.

Suppose fifteen factions all claim to defend “the people.” Suppose none possesses broad support. Suppose they fight one another while each claims revolutionary necessity. Suppose commanders cannot control their members. Suppose no institution can investigate wrongdoing. Suppose anyone can invoke the oppression of a population as authorization to start a war that the population itself will have to endure.

The authority criterion is doing real moral work there.

The better conclusion is therefore:

statehood is neither necessary nor sufficient for legitimate authority; some form of public authorization and accountable political agency is still morally necessary for sustained collective force.

Individual self-defense is different. A person under immediate wrongful attack does not need a parliament to authorize survival. The more force moves from immediate personal defense toward organized political war, however, the stronger the demand for collective authorization becomes because the actor imposes risks not only on aggressors but on whole communities.

Just cause likewise requires severity.

Not every injustice permits force. Corruption, discriminatory policies, censorship, disputed elections, economic inequality, even serious police abuse can create powerful rights of protest, disobedience, refusal, litigation, political organization, and perhaps disruptive resistance without yet generating a right to civil war. The threshold must be high because war externalizes risk onto others.

At the clearest end are threats such as enslavement, genocidal destruction, ethnic expulsion, systematic torture, or an occupying power's violent subjugation. The case for defensive force grows as the wrong becomes more severe, direct, systematic, and resistant to correction.

Yet just cause never migrates from the cause to every act performed in its name.

This is the central error of romantic revolutionary ethics:

We are fighting injustice. Therefore what advances our struggle is just.

It does not follow.

A people may possess an impeccable right not to be enslaved and no right to massacre children.

A population may possess a right to resist ethnic cleansing and no right to ethnically cleanse another population.

A colonial subject may possess a right to independence and no right to torture a captive.

The enemy government's crimes do not enter a moral bank account from which unlimited retaliatory permissions can be withdrawn.

Proportionality is therefore double.

There is macro-proportionality: is resort to force likely to cause evils grossly disproportionate to the wrong resisted?

And there is micro-proportionality: even within a justified resistance, is a particular use of force excessive relative to the immediate defensive purpose?

The Internet Encyclopedia of Philosophy's account of just war emphasizes precisely this separation: discrimination asks against whom force may be directed; proportionality asks how much force may permissibly be used.

Reasonable prospect of success is more difficult than it appears.

As a general rule it prevents political leaders from converting civilians into sacrificial material for doomed gestures. But an absolute probability threshold produces morally grotesque results in exterminatory circumstances.

The Warsaw Ghetto Uprising began on April 19, 1943, when German forces entered to deport the ghetto's remaining inhabitants. Roughly 700 Jewish fighters participated; the uprising was crushed after nearly a month. The wider context was deportation and mass killing, and the resistance had overwhelmingly inferior prospects. USHMM notes that Jewish resistance in ghettos and camps sometimes proceeded despite awareness that defeat was virtually certain.

If “reasonable prospect” meant “probable military victory,” the theory would tell people already marked for extermination that resistance was immoral because the killers were too strong.

That cannot be right.

The criterion should instead ask:

reasonable prospect of what?

Victory is one good. Rescue is another. Delay can be another. Creating an opportunity for escape can be another. Refusing forced deportation can matter even where political victory is impossible. At the same time, symbolic dignity cannot automatically justify exposing uninvolved persons to unlimited danger. The Warsaw case therefore does not abolish reasonable prospects; it demonstrates that the relevant morally legitimate objective cannot always be defined as conventional victory.

Necessity and last resort likewise require reconstruction.

“Last resort” should not mean that an oppressed population must file one more petition indefinitely. A regime capable of manufacturing procedures can manufacture endless prerequisites. If peaceful demonstrators are disappeared, courts are controlled, elections fictitious, appeals ignored, journalists imprisoned, and ongoing atrocities irreversible, saying that another administrative remedy technically exists can become a parody of ethics.

Conversely, opposition movements routinely have incentives to proclaim peaceful avenues exhausted before they are. Violence can increase organizational prestige, intensify group cohesion, eliminate moderate competitors, and create a self-confirming world in which repression proves the movement's diagnosis.

A serious last-resort principle therefore asks not whether every imaginable alternative has been attempted but whether a less harmful means is reasonably available, timely, and sufficiently likely to protect the rights at stake.

The burden of justification rises with the irreversibility of the proposed force.

The civilian line, collective guilt, terrorism, and property

Here the report reaches its governing limit.

CIVILIANS ARE NOT SUBSTITUTES FOR POWER.

International humanitarian law expresses this through distinction. Parties to armed conflict must distinguish between civilians and combatants and between civilian objects and military objectives. Civilians enjoy protection from attack unless and for such time as they directly participate in hostilities. Persons who take no active part—including those who have laid down arms or are otherwise hors de combat—receive fundamental protections under Common Article 3.

These rules are legal. Their philosophical foundation is deeper.

A theory of resistance must be capable of identifying who is responsible for the wrong being resisted. Once that capacity collapses, resistance becomes collective vengeance.

Consider the categories that political violence routinely blurs:

This taxonomy produces an uncomfortable but indispensable proposition:

civilian immunity is not a reward for personal innocence.

Some civilians are morally vicious. Some support atrocious policies. Some soldiers are conscripts with humane private beliefs. Some taxpayers oppose the war their taxes finance. Some politicians bear greater causal responsibility than low-ranking soldiers. Moral blame and vulnerability to attack are not simple mirrors of one another.

Just-war theory has long struggled with this problem. Its discrimination principle refuses to make the enemy population one undifferentiated targetable mass, even while acknowledging that civilians can possess varying degrees of political responsibility.

The correct inference from “some civilians are culpable” is therefore not “civilians are legitimate targets.”

It is that culpability has degrees and forms that must be handled by political judgment, law, prosecution, electoral consequences, truth processes, sanctions, or other forms of accountability appropriate to the conduct.

The alternative is collective guilt.

Collective guilt is philosophically dangerous for five reasons.

First, it substitutes identity for agency. A person becomes liable because of nationality, ethnicity, citizenship, residence, religion, or presumed political sympathy rather than because of what that person actually did.

Second, it destroys limiting principles. If taxpayers are targetable because taxation finances an army, almost every adult civilian in a modern state becomes targetable. If voters are targetable because their votes causally support a government, democracy perversely becomes a mechanism for abolishing civilian immunity. If political supporters are targetable because rhetoric contributes to oppression, speech becomes combatancy.

Third, it is symmetrical in the worst sense. The state can say the village “supports insurgents.” Rebels can say the city “supports the regime.” Each converts a population into a military abstraction. The categories then eat the civilians from both sides.

Fourth, it converts political liberation into ethnic or social war. A resistance movement ostensibly fighting a structure of domination begins treating the people living under that structure as embodiments of the structure.

Fifth, it prefigures the politics after victory. A movement that trains itself to assign inherited collective guilt is unlikely to become magically individualist, legal, and restrained when it acquires state power.

This is where the distinction between government and people is not merely humanitarian. It is constitutional.

A liberation movement is making an implicit claim about what political power ought to be. If that movement denies individual moral standing whenever individuals belong to the wrong population, it contradicts part of its own emancipatory claim.

The same standard must apply to governments. Counterinsurgency language has repeatedly blurred civilian communities into “support networks,” “host populations,” or environments in which insurgents operate. Modern humanitarian law resists precisely that move: non-state armed groups and governments remain bound by rules protecting civilians in non-international armed conflict. Contemporary counterinsurgency doctrine has also often recognized civilian protection and political legitimacy as central, even from a strategic perspective; but strategic usefulness is weaker than the moral point. Civilians should not be protected merely because harming them is counterproductive. They should be protected because they are not disposable instruments.

Terrorism terminology must be handled with the same discipline.

There is no single universally accepted global definition of terrorism. Philosophical definitions usually emphasize political violence in which fear or terror beyond the immediate victim is central; national statutes define the term differently. United States federal law, for example, defines international and domestic terrorism around dangerous or violent acts intended to intimidate or coerce civilian populations or influence governmental conduct through intimidation or specified forms of violence. UN Security Council Resolution 1566 described certain criminal acts, including attacks on civilians intended to cause death or serious injury for purposes such as provoking terror or compelling governments, as within the core problem.

Historically, the terminology is even less clean. “Terror” was initially associated with the governmental Reign of Terror during the French Revolution; later “terrorism” became strongly associated with revolutionary and non-state violence. Colonial authorities also routinely used terrorism or banditry terminology against insurgent movements, some of which later became governments.

Two opposite mistakes follow.

The first is governmental:

They are terrorists, therefore no further moral analysis is necessary.

The label does not answer whether particular grievances are just, whether state violence preceded the insurgency, whether every member of the movement shares responsibility, or whether government countermeasures are lawful and moral.

The opposite mistake is revolutionary:

They are freedom fighters, therefore conduct commonly described as terrorism cannot really be terrorism.

That is equally evasive.

A movement does not cease deliberately terrorizing civilians because its political objective is independence. Anti-colonial justice cannot alter the moral status of a child into that of a military objective. Nor does a state's uniform make deliberate terror against civilian populations morally different in essence. Hannah Arendt herself treats terror as a mode of domination that can become governmental when violence substitutes for genuine political power.

The analytically useful question is therefore not:

terrorist or freedom fighter?

It is:

What was done, to whom, with what intent, under what conditions, and for what coercive purpose?

When political violence intentionally attacks civilians in order to frighten a larger population or compel political authorities, the moral problem remains even if terminology is disputed.

“State terrorism” can thus be a defensible analytic concept where governments deliberately employ terror against populations for political control, though legal systems often classify such conduct under different categories: war crimes, crimes against humanity, unlawful killing, torture, enforced disappearance, or other offenses. The semantic distinction should not become moral immunity.

Property destruction again requires discrimination.

There is a morally significant difference between damaging an unoccupied object and injuring an uninvolved person. This difference explains why some resistance traditions have tried to distinguish attacks on property from attacks on human beings. Nelson Mandela's Rivonia statement defended the ANC's move away from exclusive nonviolence after severe apartheid repression while insisting that the form initially selected was intended to avoid loss of life rather than embrace terrorism.

One need not endorse every later action of any movement to see the ethical structure: if force becomes unavoidable, a movement may seek to narrow harm by prioritizing forms that do not injure persons.

But “only property” can be morally misleading. A civilian house is property. A clinic is property. A water system is property. A small shop supporting a family is property. Burning occupied buildings is not ethically converted into harmless vandalism because flames initially contact objects.

Property destruction should therefore be judged by at least five questions:

what ownership and function the property has; whether the owner is implicated in the wrong; whether destruction risks bodily harm; whether the object is essential to civilian life; and whether a less harmful means can achieve the legitimate political purpose.

The priority remains persons.

Pacifism, nonviolence, the hard cases, and last resort

The strongest case against political violence is considerably stronger than the caricature offered by its opponents.

Pacifism need not rest on the belief that oppressors are kind.

It can begin from the opposite belief: human beings are extraordinarily good at finding noble reasons to kill one another, and extraordinarily bad at recognizing the moment when necessity has become vengeance.

The strongest pacifist argument has several parts.

The first is deontological. Human beings possess a moral status that political objectives cannot simply override. Deliberately killing another person is not converted into a clean act because history is moving in one's preferred direction. Some pacifists take this prohibition to be absolute.

The second is epistemic. Every side of a violent political struggle tends to believe its own cause exceptional. If “sufficient injustice” creates permission to kill, who decides whether the threshold has been crossed? Militants select evidence confirming necessity. Governments select evidence confirming emergency. Human beings are partisan judges of their own violence.

The third is reciprocity. Once one side relaxes civilian immunity, the other can use the same categories. “Their civilians sustain their war.” “Their community shelters extremists.” “Their voters chose this.” “Their taxpayers finance that.” The logic of collective guilt becomes mutually self-authorizing.

The fourth is consequential. Violence can create martyrs, reprisals, militarized hierarchies, emergency politics, displacement, and cycles of retaliatory harm. The organization most capable of winning a violent struggle may be the organization most internally centralized around coercion, not the one best suited to pluralistic government afterward.

The fifth is political in Arendt's sense. Arendt sharply distinguishes violence from power: power arises from people acting in concert; violence can destroy power but cannot manufacture the durable collective authorization on which political order rests. A movement that can only command obedience through violence may have acquired control while failing to acquire political power in the richer sense.

The sixth is strategic. Nonviolent resistance is not merely moral witness. It can impose real costs through withdrawal of cooperation.

Its repertoire includes strikes, boycotts, refusal, noncooperation, civil disobedience, conscientious objection, mass demonstrations, and the construction of parallel civic institutions. King's account of direct action explicitly sought to create political pressure sufficient to force negotiation when normal requests had failed. Gandhi's satyagraha similarly understood nonviolence not as passivity but as disciplined resistance involving self-control and willingness to bear suffering rather than inflict it.

This matters empirically.

Erica Chenoweth's widely cited research on major resistance campaigns reports that, among 565 campaigns in the historical dataset discussed in Journal of Democracy, roughly 51 percent of predominantly nonviolent campaigns achieved full success compared with about 26 percent of violent campaigns. The same research associates successful civil resistance with broader participation and, in many cases, more favorable post-conflict political outcomes.

This finding should be taken seriously.

It should not be converted into scripture.

The campaign data use particular definitions of campaigns and success. Observational comparisons face selection problems: movements do not randomly choose violence or nonviolence; regimes differ; geopolitical environments differ; unsuccessful movements may be underdocumented. Christoph Dworschak's replication and critique argues that prominent results about nonviolent superiority can be highly sensitive to variable selection, undercoverage, bootstrapping, and omitted-variable bias.

The defensible empirical conclusion is therefore narrower:

There is substantial evidence that organized nonviolent resistance can be far more politically effective than common stereotypes suggest, and often more effective than armed campaigns; but the literature does not establish a law that nonviolence always works or that armed resistance never does.

That distinction matters enormously.

A moral preference for nonviolence should not depend on a statistically absolute claim that does not exist.

Gandhi represents the most demanding fusion of moral and strategic nonviolence. Satyagraha sought to place suffering on the resister rather than the opponent and to transform the relationship through disciplined truth-force. Gandhi's position cannot be reduced to safety. It expected courage, loss, imprisonment, and suffering.

Its difficulty is exactly there.

May one choose suffering for others?

A person can nobly refuse to kill even at the cost of their own life. It is harder to infer that they may require a child, captive, or persecuted neighbor to die rather than use force in that person's defense.

Martin Luther King Jr. grounds nonviolent direct action in a political structure rather than mere moral purity. Birmingham's problem, in his account, was not absence of negotiation as an ideal but the refusal of entrenched power to negotiate until pressure altered the political environment. Nonviolence here is coercive in the broad political sense: it changes the costs of continued injustice without intentionally injuring adversaries.

King therefore provides one answer to the false binary between passivity and violence.

Malcolm X articulates the opposite pressure point. His defense of self-defense rejected the idea that Black Americans were morally required to accept physical attack without protecting themselves. In a 1963 speech he distinguished obedience to law and respect for others from refusal to defend oneself when attacked.

The strongest version of Malcolm's argument is not an unrestricted doctrine of retaliatory violence. It is considerably narrower:

nonviolence chosen as a political strategy does not automatically create a moral duty to submit to immediate unlawful violence.

That proposition is difficult to dismiss. It is also easy to enlarge beyond recognition. Self-defense against an attacker is not the same thing as vengeance against the attacker's community. The moral work is being done by immediacy, wrongfulness, and discrimination.

Fanon poses a more radical challenge. Writing from the colonial world and particularly the Algerian struggle, he treats colonialism not as an ordinary government containing regrettable injustices but as an order constituted through coercion, hierarchy, racialization, and violence. The Wretched of the Earth describes decolonization as violent and considers violence capable of reshaping the psychology and agency of the colonized.

Fanon's insight should not be domesticated until it becomes innocuous. Colonial order often did rest on coercive violence, and describing resistance as the point at which “violence enters politics” can erase the force that made the colonial relationship possible in the first place.

But Fanon's rhetoric of violence is also where criticism must be hardest.

The suggestion that violence can cleanse, unify, or restore humanity risks converting an instrument that may sometimes be necessary into a moral therapy. Once violence becomes redemptive rather than regrettable, proportionality becomes harder to sustain. Violence that must constantly justify itself is safer than violence believed to transform its practitioners into authentic political subjects.

Fanon himself was not blind to postcolonial corruption; his critique of the national bourgeoisie warns that formal independence can reproduce domination under new rulers. That warning cuts against romantic readings of revolutionary force. A movement can defeat colonial rule and still fail to create freedom.

Camus supplies the counterweight. The Rebel asks what happens when rebellion against humiliation becomes revolution without limits. Camus sees authentic rebellion as arising from an affirmation of human value and solidarity, but attacks forms of revolutionary nihilism in which a promised future licenses murder in the present.

Camus's principle is not politically comfortable. Rebellion establishes a boundary—“this cannot be done to human beings”—and therefore contradicts itself when it becomes willing to do anything to human beings.

His limitation is that “measure” does not itself solve every emergency. In a death camp, slave plantation, torture chamber, or genocidal roundup, restraint may still require force. But Camus's warning remains devastating: a revolution that declares itself historically innocent has begun reproducing the logic of absolute power.

Arendt adds a different distinction. She does not simply call violence evil. She argues that violence and power are different phenomena. Violence can be instrumentally effective and can destroy existing power, yet cannot by itself create the collective acting-in-concert that constitutes genuine political power. She even acknowledges that nonviolence would not have achieved the same results against every possible regime, explicitly contrasting the British imperial setting in India with totalitarian regimes capable of massacre without comparable restraint.

That concession is fatal to simplistic strategic pacifism.

It does not refute moral pacifism.

The hard cases expose the difference.

Does absolute pacifism survive these cases?

As a coherent moral position, yes. As a universally persuasive political duty imposed on victims, much less clearly.

The absolute pacifist can say: there are acts one must never perform, even to stop terrible consequences. Moral limits exist precisely when violating them would be useful. Otherwise they are not limits. The pacifist can also argue that human beings systematically exaggerate the necessity of killing and underestimate the indirect suffering produced by armed struggle.

That argument deserves respect.

Its hardest problem is defense of third parties.

Suppose an aggressor is actively killing defenseless people and proportionate force is the only available means to stop the attack. The pacifist who refuses personal participation may be exercising moral conscience. But to claim that the victims themselves, or someone capable of protecting them, are morally prohibited from stopping the aggressor requires a much stronger argument. At that point the doctrine risks transferring the moral costs of one's own purity onto the bodies of people who did not volunteer to die.

The strongest conditional resistance theory therefore begins not from a love of violence but from duties of protection.

Its central argument is:

1. Persons possess rights against killing, enslavement, torture, expulsion, and comparable grave wrongs.
2. Those rights do not disappear because the wrongdoer controls the legal system.
3. Ordinary morality recognizes defensive action against wrongful threats.
4. Political institutions exist in part to secure those basic rights.
5. When a state systematically becomes the author of the very threats that justify protective institutions, its claim to exclusive moral authority over force weakens.
6. In extreme circumstances, persons and political communities can therefore possess a right of resistance incorporating proportionate defensive force.
7. Because innocent third parties retain their rights, the resistance right is constrained by necessity, discrimination, proportionality, legitimate collective authorization, and accountability.

The argument is powerful.

Its objections are equally serious.

Who decides that the threshold has been crossed?

How representative must the resistance be?

What if violence worsens the oppression?

What if multiple factions claim authority?

What if “temporary” emergency leadership becomes dictatorship?

What if the movement sincerely but falsely believes peaceful means are futile?

What if protecting some citizens predictably exposes many more to retaliation?

These are not technical details. They are reasons why a right of resistance must remain conditional rather than becoming a revolutionary equivalent of sovereign immunity.

The last-resort problem crystallizes this tension.

A sound test should reject both mechanical extremes.

A peaceful remedy is genuinely “available” only when there is a reasonable possibility that it can operate in time. A court that will consider a case after the population has been deported is not an adequate remedy for imminent ethnic cleansing. An election occurring in five years under free conditions may be relevant to bad policy but irrelevant to tonight's torture.

Availability also requires some institutional reality. A grievance procedure controlled by the perpetrators, an election whose opposition candidates are systematically imprisoned, or a petition process answered by disappearance cannot indefinitely reset the last-resort clock.

But peaceful possibilities are not exhausted merely because they require patience, sacrifice, organizing, electoral defeat, compromise, or negotiation with people one despises. King explicitly defended direct action as a way of forcing negotiation; the point was not to declare negotiation obsolete but to alter the conditions under which it became possible.

A morally serious movement should therefore have to answer:

What peaceful remedies exist?

Why are they inadequate to protect the threatened interest?

What evidence shows that waiting creates serious irreversible harm?

What narrower coercive measures remain possible?

Who besides the movement's own leadership has assessed the claim of necessity?

The more irreversible the force contemplated, the greater the burden of proof.

Historical comparison and accountability after resistance

History does not provide moral algorithms. It provides tests that abstractions must survive.

The cases below are useful precisely because they resist one-dimensional interpretation.

Several lessons survive comparison.

First: extreme oppression strengthens just cause but does not automatically determine means.

Haitian slavery and the Warsaw Ghetto make the right of passive submission almost impossible to defend as a universal political duty. But even the strongest cause does not prove that every person associated with the oppressive system becomes liable to attack.

Second: nonviolence is strongest when it can exploit dependence.

Governments require workers, administrators, transport systems, soldiers, civil servants, commercial cooperation, and public acquiescence. Strikes, boycotts, civil disobedience, and mass noncooperation can attack those dependencies without deliberately attacking human bodies. Civil-resistance scholarship emphasizes precisely the advantages of broad participation and defections from regime pillars.

But Arendt's warning is essential: a regime willing and able to answer nonviolent power with unlimited massacre may alter what strategies can protect people in time.

Third: historical causation is usually mixed.

It is tempting to transform a successful political transition into proof of one's preferred ethic. India becomes proof that nonviolence always works. Algeria becomes proof that armed struggle is necessary. South Africa becomes proof either of armed resistance or of sanctions and mass politics depending on the narrator.

History is rarely that generous.

Political outcomes result from interacting causes: domestic organization, elite fractures, economic conditions, external pressure, geopolitical change, repression, war fatigue, mass participation, leadership, and contingency. The empirical case for nonviolent resistance is important precisely because it corrects underestimation of nonviolence; it should not be converted into monocausal mythology.

Fourth: means affect the political order that follows.

This is where Arendt and Camus converge despite very different philosophies. Arendt argues that violence cannot by itself generate political power; Camus warns that rebellion destroys its own principle when unlimited murder becomes a normal instrument of historical salvation.

Organizations learn habits.

A movement that normalizes disappearance may inherit prisons and continue disappearing people.

A movement that treats criticism as treason during struggle may retain that category in government.

A movement that divides populations into collectively guilty communities may carry that division into citizenship law.

A movement that builds genuine deliberative institutions, disciplines its own members, protects political opponents, distinguishes civilians, and accepts external scrutiny is practicing—imperfectly—the political order it claims to seek.

This is why post-conflict accountability is part of the ethics of resistance, not an afterthought.

International criminal law emphasizes individual rather than mystical collective criminal responsibility. Article 25 of the Rome Statute is structured around individual criminal responsibility, while Article 27 rejects official capacity as an automatic shield from the Statute's application. The principle is philosophically important even outside the ICC's jurisdiction: neither a government office nor revolutionary status should erase personal responsibility for atrocities.

OHCHR's account of transitional justice similarly treats prosecution, truth-seeking, reparations, recognition of victims, and guarantees of non-recurrence as interconnected components of post-conflict justice.

A morally serious resistance movement should therefore accept at least the following post-conflict propositions:

its own crimes remain crimes;

its victims remain victims even if they belonged to the former oppressor's social group;

officials of the old regime deserve individualized judgment rather than collective revenge;

combatants who surrender or are detained cease being objects of vengeance;

civilian populations associated with the defeated side remain citizens or rights-bearing persons, not spoils of victory;

truth about crimes by one's own side is not betrayal;

reparation is owed according to victimization, not ideological allegiance;

and victory does not constitute acquittal.

The ICRC's legal protections for persons not or no longer participating in hostilities embody the same refusal to let enmity erase personhood.

This is one of the sharpest differences between liberation and replacement domination.

Liberation says: the arbitrary power under which human beings were subjected must end.

Replacement domination says: the wrong people previously possessed arbitrary power; now we should possess it.

No ethical theory of resistance should confuse them.

Judgment, contested claims, propositions, unresolved questions, and sources

The evidence supports neither absolute faith in nonviolence nor a romantic right of revolution.

The strongest positions can now be stated without caricature.

The strongest pacifist case says that political killing corrupts precisely the moral principle resistance claims to defend: the irreducible worth of persons. Human beings systematically overestimate their righteousness, and warfare transfers the costs of political conviction onto others. Nonviolent coercion can often mobilize broader participation, weaken authoritarian dependence, encourage defections, preserve civil society, and produce better post-conflict outcomes. Once civilians become permissible instruments, every faction acquires a vocabulary for atrocity. Gandhi supplies the ethics of refusing to inflict suffering; King demonstrates the coercive political sophistication of nonviolent direct action; Arendt demonstrates why violence cannot itself create legitimate political power. Empirical civil-resistance research gives this argument substantial, although not universal, strategic support.

The strongest objection to pacifism is defense of others under catastrophic, immediate, and otherwise unstoppable attack. A doctrine under which an enslaved person may never physically resist enslavement, a victim may never stop a torturer, or witnesses to an ongoing massacre may never use proportionate force to protect those being killed imposes extraordinary moral costs on the victims. Strategic evidence for nonviolence cannot solve the case because effectiveness varies by political environment, and Arendt herself recognized that strategies effective against British imperial power might have met massacre under totalitarian regimes.

The strongest conditional-right-of-resistance case begins with basic rights and defensive morality. Public authority exists partly to protect persons against grave coercive wrongs. When government systematically becomes their author—enslavement, extermination, torture, ethnic expulsion, comparable tyranny—it can forfeit the presumption that its commands generate obedience. A right to resist can then arise, including proportionate force where necessary to protect persons. Locke provides the classical breach-of-trust version; abolitionism exposes its application to slavery; anti-colonial self-determination extends the problem to imperial domination; contemporary philosophers such as Reitberger and Finlay reject state recognition as a necessary condition for all morally justified force.

The strongest objection to that theory is epistemic and institutional. Every violent faction can claim the threshold has been crossed. Armed movements may be unrepresentative, internally coercive, incapable of controlling members, or indifferent to the communities they purport to defend. War can intensify the very atrocities it seeks to stop. The exception can swallow the rule. “Last resort” can become ritual rhetoric. “The people” can become a word used by a small organization speaking over actual people.

The right of resistance therefore needs constraints stronger, not weaker, than the state right to force.

The decisive hierarchy should be:

basic human protection over sovereign prestige;

civilian immunity over collective vengeance;

less harmful means over more harmful means where comparably effective;

defense over punishment;

individual responsibility over collective guilt;

public accountability over clandestine moral self-certification;

political settlement over permanent militarization.

The following register identifies the claims on which scholarship or philosophy remains genuinely contested.

Twenty propositions suitable for serious publication

1. The legality of political force is evidence about institutional authorization, not proof of moral justification. Political obligation and legal obligation remain conceptually distinguishable.

1. The fact that the modern state claims a monopoly over legitimate physical force does not entail that every successful state use of force is legitimate in the moral sense. Weber's definition cannot perform ethical work it was never designed to perform.

1. A plausible right of resistance is derivative of rights against grave domination, not an independent entitlement to revolution. Its strength therefore rises and falls with the severity, immediacy, and systematic character of the wrong resisted.

1. Political authority is fiduciary before it is possessive. When public coercive institutions systematically attack the elementary interests they exist to secure, their claim to obedience can be morally weakened or forfeited; Locke's theory remains the canonical formulation of this logic.

1. Just cause is necessary but radically insufficient. The justice of liberation cannot transfer automatically to the means used in liberation.

1. Statehood is neither sufficient nor strictly necessary for legitimate authority in the ethics of force. The morally relevant properties are public authorization, representativeness, accountability, discipline, and capacity to establish a defensible peace.

1. The moral threshold for individual defensive force and the threshold for organized revolutionary warfare are not identical. Large-scale political force imposes involuntary risks on third parties and therefore requires stronger public justification.

1. Last resort should be interpreted as a requirement of reasonable necessity, not chronological ritual. Populations facing irreversible atrocities are not morally required to exhaust fictitious procedures, while movements remain obligated to use genuinely available less harmful alternatives.

1. Proportionality is not a mathematical ratio but it is not empty. The gravity of the protected interest, probability of achieving it, scale of expected harm, distribution of risk, and availability of alternatives all matter.

1. Reasonable prospect of success must be indexed to the legitimate objective. Conventional military victory cannot be the sole measure in exterminatory situations where rescue, survival, escape, or delay are morally significant.

1. Civilian immunity is strongest when detached from claims of perfect civilian innocence. A person may be politically culpable and still not be liable to violent attack.

1. Voting, taxation, nationality, residence, and generalized political sympathy do not by themselves constitute direct participation in hostilities. Any contrary rule would convert modern populations into universally targetable collectives and destroy distinction.

1. Collective guilt is equally corrupting when invoked by states and rebels. “Insurgent population” and “enemy population” can become mirror-image devices for abolishing individual moral status.

1. Property destruction is generally morally less grave than intentional injury or killing, but its justification depends on its relation to human safety, livelihood, civilian survival, ownership, and political purpose. Civilian objects do not become morally negligible because they are objects.

1. Terrorism should be analyzed primarily through conduct rather than political identity. Deliberately terrorizing civilians does not become morally acceptable when done for liberation, and state terror does not become morally acceptable when done under law.

1. Nonviolence is not non-coercion. Strikes, boycotts, civil disobedience, and mass noncooperation can exert severe political pressure while preserving the prohibition on intentional bodily harm.

1. The empirical superiority of nonviolent campaigns is a serious presumption in favor of nonviolent strategy, not a universal theorem. Both the prominent empirical findings and their methodological criticisms must be kept in view.

1. Violence may destroy an oppressive political arrangement without possessing the capacity to constitute legitimate political power afterward. Arendt's distinction between violence and power remains indispensable here.

1. Post-conflict accountability is part of jus ad resistendum itself. A movement unwilling in advance to accept responsibility for its own crimes gives reason to doubt that its project is liberation rather than a competing claim to impunity. International criminal and transitional-justice frameworks reinforce individualized responsibility, truth, reparations, and non-recurrence.

1. The deepest moral test of a resistance movement is whether it can oppose a structure of power without converting the human beings living under that structure into embodiments of the enemy. Once political association becomes sufficient for violent liability, the normative distinction between resistance and collective punishment begins to collapse.

Ten questions philosophy has not resolved

1. How grave must oppression become before a collective right to force arises? There is no uncontested metric capable of comparing censorship, disenfranchisement, systemic brutality, torture, occupation, enslavement, and extermination along a single threshold.

1. Who may authorize force when representative political institutions have themselves been destroyed? Requiring democratic authorization can be impossible precisely where resistance is most necessary, yet eliminating authorization empowers self-appointed factions.

1. How much risk may resisters impose on unwilling members of the population they claim to defend? Defensive action can protect some while predictably exposing others to retaliation.

1. Does a person who is morally culpable for oppression but remains legally and functionally civilian ever lose some moral immunity from force? Revisionist theories of defensive liability pressure the traditional civilian-combatant distinction, while humanitarian law has powerful reasons to preserve clear status protections.

1. How should reasonable prospects be assessed when the alternatives are probable defeat and near-certain atrocity? The Warsaw Ghetto demonstrates why ordinary victory metrics are insufficient.

1. Can violent resistance remain genuinely defensive once its aim becomes revolutionary transformation rather than immediate protection? The larger the political objective, the less directly self-defense analogies seem to apply.

1. Can the psychological liberation Fanon associates with violent resistance ever count as a legitimate good in proportionality calculations? Or is treating violence as psychologically redemptive already a dangerous departure from necessity?

1. Can absolute pacifism coherently distinguish refusing to kill from refusing to rescue another person when rescue requires proportionate force? The moral status of third-party defense remains one of pacifism's hardest problems.

1. How long must oppressed populations preserve confidence in peaceful reform before declaring it no longer reasonably available? Neither the state nor the resistance can be permitted unilaterally to answer that question without evidence.

1. Can a movement use substantial coercive force during revolution without reproducing coercive hierarchy after victory? Arendt's warning that violence can destroy power but cannot create it suggests that institutional design during resistance may matter as much as intentions after success.

Source bibliography

Foundational political and philosophical texts. Thomas Aquinas, Summa Theologiae, treatment of sedition and tyranny, especially the distinction between common good and tyrannical private rule. Hugo Grotius, The Rights of War and Peace, including the treatment of war, sovereign authority, and resistance by subjects. John Locke, Second Treatise of Government, especially Chapter XIX on dissolution of government, breach of trust, slavery, and the state of war between rulers and people. Max Weber, Politics as a Vocation, for the state as claimant to the monopoly of legitimate physical force. The U.S. Declaration of Independence, for the republican articulation of consent, alteration or abolition of destructive government, and prudential restraint.

Republican and abolitionist traditions. Materials on Algernon Sidney and republican resistance to unanswerable political authority. Stanford Encyclopedia of Philosophy, “Frederick Douglass,” especially slavery, self-defense, and political resistance. David Walker, Appeal to the Coloured Citizens of the World, in primary and historical collections documenting radical abolitionist resistance to slavery. Library of Congress and Smithsonian materials on the Haitian Revolution and its emergence from slave revolt into emancipation and independence.

Anti-colonialism and self-determination. United Nations General Assembly Resolution 1514 (XV), Declaration on the Granting of Independence to Colonial Countries and Peoples, adopted December 14, 1960. Additional Protocol I to the Geneva Conventions, Article 1(4), and historical commentary concerning conflicts against colonial domination, alien occupation, and racist regimes. Materials concerning the controversy surrounding Article 1(4), including U.S. objections to Protocol I. Frantz Fanon, The Wretched of the Earth, read in the Algerian decolonization context and alongside his criticism of postcolonial elite domination.

Nonviolence and civil resistance. Gandhi's satyagraha as examined through Columbia's Uprising materials, with emphasis on self-discipline, self-suffering, and noninfliction of suffering. Martin Luther King Jr., Letter from Birmingham Jail, especially the relation between nonviolent direct action, crisis, negotiation, and unjust law. Stanford Encyclopedia of Philosophy, “Civil Disobedience,” for the conceptual relation between lawful protest, conscientious disobedience, uncivil resistance, and forcible resistance. Erica Chenoweth's synthesis of civil-resistance data in Journal of Democracy, including the comparative success figures for violent and nonviolent campaigns. Christoph Dworschak, “Civil Resistance in the Streetlight,” Comparative Politics, for robustness and selection critiques of the empirical nonviolence literature.

Civil-rights and self-defense traditions. Malcolm X materials preserved by Columbia's Malcolm X Project, including his explicit distinction between initiating aggression and defensive response to attack. Nelson Mandela's Rivonia statement, including the argument that the transition away from exclusive nonviolent resistance followed the closure of peaceful political avenues and initially sought forms of resistance avoiding loss of life. South African historical materials on the shift from mass nonviolent resistance to armed struggle after Sharpeville and intensified apartheid repression.

Limits on revolutionary violence. Albert Camus, The Rebel, particularly the argument that rebellion arises in affirmation of human value and contradicts itself when revolutionary nihilism licenses unlimited oppression. Hannah Arendt, “Reflections on Violence” and the later argument developed in On Violence, especially the distinction between violence and power and the critique of political terror. Fanon should be read against both writers rather than either canonized or reduced to caricature.

Contemporary resistance ethics. Magnus Reitberger, “License to Kill: Is Legitimate Authority a Requirement for Just War?”, arguing against treating traditional legitimate authority as an indispensable state monopoly and deriving justificatory force from self-defense and basic rights. Christopher J. Finlay's work on terrorism, rebel ethics, and the right to resist, which extends just-war reasoning to non-state resistance rather than treating rebels as outside moral theory. Candice Delmas, A Duty to Resist, on duties of resistance extending beyond the classical model of polite civil disobedience. The contemporary debate also includes Virginia Held, Seth Lazar, Jeff McMahan, Cécile Fabre, Jonathan Parry, James Pattison, Michael Walzer, and others concerned with authority, defensive liability, discrimination, proportionality, and non-state force; Reitberger's article maps a substantial part of this literature.

Just-war theory and humanitarian law. Internet Encyclopedia of Philosophy, “Just War Theory,” particularly discrimination, proportionality, and individual responsibility for conduct within war. ICRC Customary IHL Rule 1 on distinction between civilians and combatants. Additional Protocol I, Article 51, on protection of the civilian population. ICRC rules concerning civilian objects and direct participation in hostilities. Common Article 3 and ICRC materials governing persons taking no active part in hostilities. ICRC's treatment of non-international armed conflicts is particularly important because it stresses both the legal asymmetry surrounding rebel participation and the equal binding force of core humanitarian rules upon state forces and organized non-state armed groups.

Terrorism terminology. Stanford Encyclopedia of Philosophy, “Terrorism,” surveying conceptual definitions centered on political violence and terror. 18 U.S.C. § 2331, illustrating one influential statutory approach emphasizing dangerous violent acts intended to intimidate or coerce civilian populations or governments. United Nations materials documenting both the absence of a universally agreed comprehensive definition and Security Council efforts to specify core acts involving violence against civilians for purposes of terror or coercion. Historical research tracing “terror” from the governmental French Revolutionary Terror into later revolutionary and non-state usages.

Historical test cases. U.S. Holocaust Memorial Museum materials on the Warsaw Ghetto Uprising and other Jewish armed and civilian resistance to Nazi deportation and extermination. Historical materials on Gandhi's Salt March and the wider Indian independence struggle, including scholarship cautioning against monocausal accounts of independence. Materials on the Algerian War, its anticolonial setting, violence, and eventual independence. Mandela and South African historical sources concerning apartheid, repression, nonviolent resistance, the turn toward armed struggle, and eventual constitutional politics.

Post-conflict accountability. Rome Statute of the International Criminal Court, particularly Article 25 on individual criminal responsibility and Article 27 on irrelevance of official capacity within the Statute's framework. OHCHR materials on transitional justice, emphasizing truth-seeking, prosecution, reparations, victim recognition, and guarantees of non-recurrence.

The final judgment can now be stated.

Force is not morally legitimate because history is cruel. Cruelty is what makes the question arise, not what answers it.

There are conditions under which refusing all force becomes ethically difficult to defend: a person being enslaved; a captive being tortured; a population being deported toward extermination; civilians being systematically expelled or killed while every effective peaceful protection has been destroyed. To tell such people that the state retains an exclusive moral title to force merely because it possesses uniforms, statutes, and diplomatic recognition is to mistake organized power for justice. Locke, abolitionism, anti-colonial self-determination, defensive morality, and contemporary resistance theory all supply reasons to reject that mistake.

But the opposite mistake is no less destructive.

Oppression does not produce an unlimited moral credit. A liberation movement cannot inherit the victim's innocence and carry it into every act. The child of an oppressor is not the oppressor. The voter is not the soldier. The taxpayer is not the commander. The civilian administrator is not automatically a combatant. The family of a torturer is not a substitute for the torturer. Political support, cultural identity, citizenship, and collective history cannot be transformed into general licenses for bodily attack without annihilating the very principle of individualized moral standing that resistance to oppression ordinarily invokes.

That is why the civilian line is not one restriction among many. It is the point at which a theory of resistance reveals what it actually believes about human beings.

If it believes persons remain persons even inside enemy societies, it can still claim to resist domination.

If it believes whole populations become morally fungible once attached to an oppressive power, it has adopted one of domination's oldest intellectual instruments.

Just-war theory matters because it separates cause from conduct. Gandhi matters because he shows that resistance need not be passive to reject violence. King matters because he shows how nonviolent coercion can force negotiation. Malcolm X matters because voluntary nonviolence is not obviously a duty of undefended submission. Fanon matters because political orders built on coercion should not disguise their own violence and then condemn resistance as the first appearance of force. Camus matters because the revolt against murder can become incoherent when it makes murder historically innocent. Arendt matters because destroying domination and creating political power are not the same achievement.

The thesis has therefore survived its strongest objections, but only after qualification.

The distinction between power and people does not mean every official is innocent, every civilian blameless, or every member of an armed force equally culpable. It does not forbid individualized legal responsibility. It does not prohibit proportionate defense against persons actually carrying out grave wrongful violence. It does not require oppressed people to imagine that political structures float above human agents.

It means something more precise:

responsibility must remain connected to agency.

The moment a resistance movement abandons that connection, its enemy ceases to be oppressive conduct and becomes a category of human beings.

At that point proportionality becomes difficult because there is no bounded wrong to stop.

Distinction becomes impossible because everyone belongs to the suspect population.

Last resort becomes meaningless because destruction of the population can be redescribed as destruction of the regime.

Post-conflict reconciliation becomes treason because collective enemies never become innocent.

And liberation ceases to mean freedom from arbitrary domination. It begins to mean transfer of the authority to dominate.

The ethical right of resistance therefore exists, but under permanent suspicion—not because oppression is morally ambiguous, but because human beings wielding force are.

Its strongest form is defensive rather than redemptive.

Its object is protection rather than purification.

Its authority must be accountable rather than self-anointed.

Its violence, where genuinely necessary, must remain regrettable rather than sacred.

Its scale must remain proportionate.

Its first presumption must favor less harmful means.

Its enemies must be identified through conduct and responsibility rather than ethnicity, nationality, taxation, voting, or collective identity.

Its prisoners remain human.

Its own crimes remain crimes.

Its victory does not rewrite the moral law backward.

And its treatment of civilians is not peripheral evidence about its morality. It is among the clearest evidence available.

A REBELLION THAT DESTROYS THE DISTINCTION BETWEEN POWER AND PEOPLE HAS ALREADY SURRENDERED ONE OF ITS STRONGEST MORAL CLAIMS.