Humanitarian Law and Just War

The Question of Restraint in Armed Conflict

Humanitarian law begins with a difficult proposition: organized violence remains subject to obligations whose authority survives disagreement over the justice of the conflict. Governments may disagree about sovereignty, historical wrongs, territory, self-defense, or self-determination. The wounded on either side nevertheless require care, prisoners require humane treatment, and civilians retain protection from direct attack. This normative continuity is the distinctive achievement and unresolved philosophical problem of the modern law of armed conflict. A general permission to resist aggression does not answer whether one may strike a particular apartment building, detain a person without review, destroy a water system, or endanger a medical convoy (Geneva Convention I, 1949, common Art. 3; Protocol I, 1977, Arts. 48, 51, 54, 57).

The underlying question is why rules should bind a defender confronted by a lawless aggressor. One answer derives from individual rights: the fact that force is used for protection cannot erase the separate standing of persons exposed to it. A second answer derives from the nature of public law: operational constraints become reliable only when their application is not made contingent on each belligerent's self-assessment of its cause. A third answer is institutional: reciprocal compliance protects soldiers and civilians, but the duties of humanity also apply when reciprocity fails. These accounts reinforce one another without being conceptually identical. If the rule were only a bargain, a violation by one party might dissolve the other's duties. If it were only an abstract moral claim, questions of classification, feasibility, proof, and enforcement would remain unanswered.

The inquiry therefore separates four levels of assessment. The legitimacy of resorting to force concerns the political decision to fight. The lawfulness of military operations concerns the conduct of fighting. The treatment of persons in a party's power concerns obligations that continue after immediate combat has ceased. The termination of war and disposition of its consequences concern the transition to peace. Legal doctrine and philosophical argument meet across these levels, but each question requires its own criteria. The monograph's independent focus is the first three levels' relation to legal restraint; the larger general ethics of war remains the subject of the separate Just War Theory research dossier (Lazar, 2016; Haque, 2017).

This study addresses international humanitarian law as law, just war as moral philosophy, and their interaction as a problem of institutional normativity. It does not resolve particular armed conflicts by assertion. Applying any legal rule to particular conduct requires verified facts concerning conflict classification, treaty obligations, operational knowledge, expected effects, command decisions, and jurisdiction. Those evidentiary conditions are themselves central to the philosophy of wartime judgment.

Definition, Scope, and Sources of Normative Authority

International humanitarian law (IHL), also called the law of armed conflict, is the body of treaty and customary international rules governing the conduct of hostilities and protecting persons who are not, or are no longer, participating in them. Its central purposes include limiting means and methods of warfare and requiring humane treatment under conditions in which organized armed violence exists. The law distinguishes international armed conflicts from non-international armed conflicts, and the precise treaty obligations vary with the classification and the parties bound. Certain fundamental customary prohibitions and treaty minima apply across that divide (Geneva Convention I, 1949, common Art. 3; ICRC, 1977a, Arts. 48–58; ICRC, 1977b, Art. 13; Henckaerts & Doswald-Beck, 2005).

Jus ad bellum designates norms concerning recourse to force; contemporary positive law is framed principally by the United Nations Charter, especially Articles 2(4), 39–42, and 51. Jus in bello designates norms governing the conduct of hostilities, detention, protection, and treatment during armed conflict. The two differ in question, legal source, and institutional consequences. A prohibited initiation of force can be accompanied by conduct that complies with IHL; a defensive war can include unlawful attacks or mistreatment. International criminal law adds a further distinction: not every breach of IHL satisfies the specific elements and jurisdictional conditions of an international crime (United Nations, 1945; International Criminal Court [ICC], 1998, Arts. 5, 8, 8 bis).

Moral just war theory asks about permissible collective defense, legitimate authority, necessity, intention, proportionality, the status of combatants, and obligations after war. Its conceptual language partly overlaps with the language of international law. It would nevertheless be a mistake to treat moral permissibility and treaty validity as synonyms. A philosopher can dispute whether all soldiers possess an equal moral permission to kill while recognizing that public humanitarian constraints apply equally to belligerent parties. A lawyer can identify an act prohibited under a convention without claiming to have settled every moral question about the agents' blameworthiness (Walzer, 1977/2015; McMahan, 2009; Haque, 2017).

The legal sources require careful ordering. Treaty rules derive authority through legally binding agreements and their interpretation. Custom arises through general practice accepted as law, conventionally termed opinio juris. The ICRC's 2005 customary-law study catalogued 161 rules, while expressly presenting a study of evidence rather than an omniscient source of every possible norm. Judicial decisions, military manuals, expert commentaries, and institutional practice are evidence and interpretive materials with different legal status. An ICRC commentary is particularly valuable for historical and doctrinal analysis but is not itself a treaty amendment (ICRC, 2005; ICRC, 2025).

The philosophical problem of authority has two parts. First, why should persons respect protective duties in the extremity of war? Second, through what institutions do duties acquire public, interpretable, and enforceable form? Answering the first question without the second leaves a moral appeal vulnerable to ambiguity. Answering the second without the first reduces protection to whatever institutions happen to command. Humanitarian law is coherent insofar as legal form gives practical effect to reasons arising from the standing of people whose lives are at stake.

Ancient and Medieval Moral Genealogies

The moral regulation of warfare has multiple histories. Classical Mediterranean authors discussed justice, power, civic duty, and the treatment of enemies, while different Asian intellectual traditions developed critiques of aggressive war, rules of conduct, and the political costs of conflict. These traditions did not create a uniform modern system of international humanitarian law. Their significance lies in the older question they share: whether organized power can claim exemption from moral reasons because it commands the means of force. The terms of each discussion must be reconstructed in its own institutional context rather than presented as an embryonic version of a twentieth-century convention.

Augustine's discussion in Contra Faustum, Book XXII, section 74, distinguishes legitimate political or divine authority from motives of cruelty, vengeance, and the desire to injure. His polemic occurs within Christian theological argument, not within modern jurisprudence. The essential conceptual contribution is that the purpose and authorized origin of coercion are independent objects of moral examination. A claimed public end does not settle the morality of the actor's disposition, and political force remains open to ethical assessment (Augustine, ca. 400, XXII.74).

Thomas Aquinas gives a more systematic account in Summa Theologiae, II–II, question 40, article 1. He identifies sovereign authority, a just cause, and right intention as conditions for just war. These are conditions about the authorization and purpose of armed conflict; they are not a complete specification of modern battlefield obligations. Aquinas also discusses deception and ambush (q. 40, a. 3), revealing that the moral analysis of war includes rules about how force may be used. It is historically inaccurate to claim that the full contemporary list of proportionality, distinction, and precautions appears already in his three conditions (Aquinas, ca. 1270, II–II, q. 40).

The continuity between medieval inquiry and contemporary humanitarian law is consequently conceptual rather than legislative. Medieval writers ask who may fight and for what ends. Modern humanitarian law specifies protected persons and forbidden methods in a detailed public regime that binds belligerents regardless of their ultimate claims. The transition involves a transformation of moral questions into more operational rules: what can be attacked, what must be spared, how evidence must be assessed, and who may be held responsible. Its history includes inherited moral vocabularies, political transformations, military practice, and humanitarian mobilization.

Early Modern Natural Law and the Juridification of War

Early modern jurists made war an object of systematic legal and moral argument within an expanding world of states, empires, trade, and colonial conflict. Francisco de Vitoria's lectures on war and the peoples of the Americas challenged some asserted titles to conquest, while remaining embedded in the hierarchies and theological assumptions of his period. His discussion matters because it subjects claimed authority to reasons accessible to opponents, including questions of injury, restitution, and proportionality. An analysis faithful to the texts must also examine the colonial setting: a doctrine invoking universal rights can become an instrument of unequal power when its empirical assumptions about those to whom it applies are distorted (Vitoria, ca. 1539/1991).

Hugo Grotius's De jure belli ac pacis (1625) brings together natural-law reasoning, the rights of belligerents, and rules governing wartime action. Its three-book architecture links the grounds of war with what is permitted or forbidden in its conduct and aftermath. Grotius differentiates strict legal permissions from moral counsel and shows why the existence of war does not erase limits on promises, persons, property, and treatment. This dual perspective is decisive for the later problem of legal and moral separation: a public legal permission or immunity can be broader than what morality ideally approves (Grotius, 1625/2005, Books I–III).

The early modern genealogy contains both expansion and restraint. European powers formalized doctrines of sovereign prerogative at the same time that jurists sought universal norms governing strangers. The contradiction is philosophically productive: claims to universal law must be assessed against the historical exclusions through which they were administered. Vitoria and Grotius should neither be reduced to present-day humanitarian liberals nor treated as merely ideological masks. Their arguments combined serious normative insight with historical limitations. The task of reconstruction is to show where a principle genuinely constrains authority and where the allocation of authority defeats the principle in practice.

Immanuel Kant subsequently places the law of war within the wider problem of peaceful international order. In Toward Perpetual Peace (1795), his preliminary articles oppose forms of war-making that destroy the possibility of future trust, while his definitive articles emphasize republican government, international federation, and cosmopolitan right. His approach shifts attention from the isolated judgment whether one war is justified to the legal conditions under which states can cease treating force as a normal means of settling disputes. Restraint in war becomes one component of a larger problem of public right and the construction of peace (Kant, 1795/1917).

Humanitarian Reform and the Nineteenth-Century Codification Problem

The nineteenth century makes the tension between military necessity and humanity unusually visible. Francis Lieber's Instructions for the Government of Armies of the United States in the Field were promulgated as General Orders No. 100 in April 1863. The Code attempts to discipline armies through public rules, including restraints on cruelty, protection for persons and property, and treatment of prisoners. Article 16 rejects cruelty under the guise of military necessity, while Article 17 accepts starvation as a method of subjection in language incompatible with later prohibitions on starvation of civilians. Its historical importance therefore lies in both its restriction of military conduct and the gaps subsequently narrowed by humanitarian law (Lieber, 1863, Arts. 14–17, 75–77).

Henry Dunant's Un souvenir de Solférino (A Memory of Solferino, 1862) supplies a different foundation. Its account of suffering after the battle at Solferino and its proposal for organized aid were central catalysts for the humanitarian movement and the first Geneva Convention of 1864. Dunant's intervention begins with wounded bodies, the capacity of volunteers to relieve pain, and the absence of assistance at precisely the moment it is needed. It reframes the relation between armed enemies by emphasizing the vulnerability that soldiers share after becoming wounded. Care is due to the injured person because the injuries place that person under a demand for aid, not because one side has proved its political cause just (Dunant, 1862; ICRC, 2009a).

The Hague stream codified restraints on the means and methods of warfare, particularly through the 1899 and 1907 conferences, while the Geneva stream developed protections for victims and the wounded. The distinction is historiographically useful but does not define two sealed legal orders. The later Geneva Conventions and Additional Protocol I integrate aspects of conduct-of-hostilities regulation with protection of persons. The 1907 Hague Regulations state in Article 22 that belligerents' choice of means of injuring the enemy is limited, a proposition that contradicts the premise of unlimited discretion once war begins (Hague Convention IV, 1907, annexed Regulations, Art. 22).

The Martens Clause, originating in the preamble to the 1899 Hague Convention II and appearing in later instruments, addresses cases not covered by specific regulations through references to established custom, the principles of humanity, and dictates of public conscience. Philosophically, it resists an inference from silence to unlimited permission. Legally, its precise interpretive consequences remain debated; the clause cannot be used as a magic formula that independently settles every unsettled weapons question. It is best understood as a reason to interpret the legal regime as incomplete without supposing that incompleteness abolishes normative constraint (Hague Convention IV, 1907, preamble; International Court of Justice [ICJ], 1996).

The Geneva Conventions and the Protection of Persons

The four Geneva Conventions of 12 August 1949 reorganized humanitarian protection after the immense injuries of the Second World War. The First Convention concerns wounded and sick members of armed forces on land; the Second addresses wounded, sick, and shipwrecked members at sea; the Third governs prisoners of war; and the Fourth protects civilians in time of war, including in occupied territory. Their arrangement reflects differences in vulnerability and legal status while maintaining the idea that persons who have fallen into an adversary's power retain enforceable claims. The relevant treaty provisions address states and military organizations, but their human purpose is expressed in obligations owed to individuals (Geneva Conventions I–IV, 1949).

Common Article 1 binds parties to respect and ensure respect for the Conventions in all circumstances. Common Article 2 establishes a central basis for applying the Conventions to declared war and other armed conflicts between contracting parties, even without recognition of war, and to occupation. Common Article 3 sets a minimum for non-international armed conflicts, requiring humane treatment of persons taking no active part in hostilities, prohibiting murder, torture, hostage-taking, degrading treatment, and sentencing without essential judicial guarantees, and requiring care for the wounded and sick. Its final provision clarifies that applying humanitarian minima does not alter the legal status of the parties (Geneva Convention I, 1949, Arts. 1–3).

These provisions have profound philosophical structure. Common Article 3 refuses to make elementary protection depend on political recognition. A state may deny an insurgency's legitimacy, and an insurgent group may deny the state's legitimacy; neither may infer that persons outside hostilities can be treated as having lost their human standing. The rule answers an institutional problem that abstract declarations often leave unresolved: duties must remain intelligible and enforceable even when recognition, political membership, and lawful authority are contested.

The Third and Fourth Conventions make protective status consequential in specific, sometimes complex, ways. Prisoner-of-war status in international armed conflicts normally concerns persons falling within defined categories, including relevant armed forces; the Third Convention's protections concern detention, interrogation, labor, discipline, communication, and release. The Fourth Convention contains protections for civilians who satisfy its definition of protected persons, rules for occupation, and limits on transfers. Not every civilian in every setting meets every category's definition, and customary rules and human-rights law may still apply. A philosophical account that collapses all statuses into a single moral label loses the legal precision necessary to protect them (Geneva Convention III, 1949; Geneva Convention IV, 1949).

The Additional Protocols and the Architecture of Conduct

The 1977 Additional Protocols elaborate rules for two significant types of conflict. Protocol I primarily concerns victims of international armed conflicts; Protocol II develops treaty protection in certain non-international conflicts under a scope narrower than every conflict covered by common Article 3. Protocol I contains an extensive body of conduct-of-hostilities rules, including the limitation of means and methods, the distinction between civilians and combatants, proportionality, and precautions. Protocol II specifies protections including humane treatment, care, protection of the civilian population, and limits on displacement and starvation (ICRC, 1977a; ICRC, 1977b).

Article 48 of Protocol I states the basic rule of distinction. Article 51 prohibits direct attacks on civilians and certain indiscriminate attacks; Article 52 specifies the concept of military objectives; Article 57 sets duties of feasible verification, precautions, cancellation or suspension when facts change, and warning where circumstances permit. Articles 54–56 address starvation, objects indispensable to civilian survival, the natural environment and dangerous installations. These provisions form an interlocking framework, not a collection of independent slogans. A lawful target must be identified; foreseeable incidental harm must be assessed; feasible measures must be selected; the assessment must be revised as information changes (Protocol I, 1977, Arts. 48, 51–57).

The Protocols' precise treaty reach depends on ratification and conditions of application. Many associated rules also have customary status, but the determination of customary law requires practice and opinio juris, including careful attention to contested formulations and state positions. It is therefore methodologically unsound to treat the verbatim text of every Protocol article as universally binding treaty law or to deny customary restraint merely because a party has not ratified a particular instrument. Humanitarian philosophy explains the reasons for the rules; legal method determines which rules bind whom and in what form (Henckaerts & Doswald-Beck, 2005; ICRC, 2005).

The philosophical significance of this system lies in the operational translation of humanity. An injunction to spare the innocent cannot by itself determine how a commander should respond to mixed civilian and military locations, incomplete intelligence, multiple possible targets, or risk to attacking forces. The Protocols connect broad prohibitions with processes of judgment under uncertainty. They make restraint both a limit on conduct and a professional duty to acquire relevant knowledge before acting.

The Charter System, Aggression, and Self-Defense

The United Nations Charter transforms the legal regulation of inter-state force. Article 2(4) prohibits the threat or use of force in international relations against territorial integrity, political independence, or otherwise inconsistently with United Nations purposes. Chapter VII establishes Security Council powers, and Article 51 preserves the inherent right of individual or collective self-defense if an armed attack occurs, subject to the Charter's terms. Philosophical theories about preemptive defense, humanitarian rescue, resistance to aggression, and state legitimacy cannot be substituted for these legal rules merely because the proposed moral purpose is urgent (United Nations, 1945, Arts. 2(4), 39–42, 51).

A distinction is essential between a violation of the prohibition on force, a breach of IHL, and the international crime of aggression. These involve different elements, actors, and jurisdictional arrangements. Legal responsibility may attach to states, individuals, or both through distinct processes. The Rome Statute contains provisions for war crimes and, under its amendments and jurisdictional requirements, the crime of aggression. A claim that a military campaign was unlawful under Article 2(4) does not identify which attacks within it violate the law of targeting; conversely, a lawful resort to self-defense does not immunize commanders from charges concerning particular attacks (ICC, 1998, Arts. 8, 8 bis).

The legal independence of the regimes serves a humanitarian purpose. An aggressor's prisoners do not become legitimate objects of abuse because their government initiated an unlawful war. Civilians belonging to the aggressor state do not forfeit their immunity by citizenship. The opponent's armed forces remain bound by protective rules. Treating the justice of the political cause as a general exception would return the entire law of armed conflict to the parties' mutually incompatible declarations of righteousness.

At the same time, legal independence does not require moral indifference to the cause of war. A soldier who knowingly participates in aggression may bear moral responsibility that cannot be dissolved by compliance with targeting rules. The stronger claim is institutional: the law must be capable of constraining anyone who fights, while moral analysis may assign different responsibilities for participating in the conflict at all. This separation is an architecture of legal restraint rather than a declaration that the causes of wars are morally equivalent (Haque, 2017; McMahan, 2009).

Classification: International and Non-International Armed Conflicts

Conflict classification determines important consequences for treaty application. An international armed conflict ordinarily arises through resort to armed force between states, whereas non-international armed conflict involves organized armed groups and sufficient intensity of hostilities under applicable criteria. Common Article 3 reaches non-international armed conflicts under its own conditions; Additional Protocol II adds requirements that include organized armed groups under responsible command and sufficient territorial control to conduct sustained military operations and implement the Protocol. Domestic disturbances and isolated violence are governed by other legal regimes when they do not meet the armed-conflict threshold (Geneva Convention I, 1949, Arts. 2–3; Protocol II, 1977, Art. 1; ICTY, Tadić, 1995).

The classification is based on facts relevant to the legal criteria, not the words governments choose. That feature has philosophical importance. Were a state able to suspend humanitarian duties by labeling a conflict a police operation, the persons most exposed to force would lack dependable protection. Conversely, treating every episode of unrest as war can weaken the distinct human-rights and law-enforcement standards appropriate outside armed conflict. The authority of law depends on resisting strategic relabeling in either direction.

Multiple conflicts may coexist in one territory, and the classification of relationships between parties may differ. External intervention can introduce international armed-conflict dimensions alongside non-international fighting. Occupation presents a separate basis for application of certain rules. The factual assessment of organization, intensity, state involvement, and territorial control is essential. Philosophical analysis must therefore distinguish the moral concept of resistance from the legal classification that allocates particular treaty rules.

The ethical challenge remains consistent across classifications: persons do not disappear behind organizational labels. But precision protects them. Who counts as a member of an organized armed group, who is a civilian directly participating in hostilities, who is captured, and who is entitled to prisoner-of-war protections are different questions. A coherent humanitarian regime treats these differences as reasons for specific safeguards rather than opportunities for erasing responsibility.

Distinction, Civilian Protection, and the Problem of Status

The principle of distinction constitutes the organizing condition for lawful attacks. Under Additional Protocol I, Article 48, belligerents must distinguish civilians from combatants and civilian objects from military objectives, directing operations only against military objectives. Article 51 develops civilian protection and prohibits attacks intended to spread terror among the civilian population, as well as indiscriminate attacks. For non-international armed conflicts, Article 13 of Additional Protocol II states the prohibition on attacking civilians and provides that civilian protection from attack is lost for such time as civilians take a direct part in hostilities. Similar protections are found in customary law as identified by the ICRC study (Protocol I, 1977, Arts. 48, 51; Protocol II, 1977, Art. 13; Henckaerts & Doswald-Beck, 2005, Rules 1–6).

The distinction involves a classification with consequences, not a moral ranking of human worth. A civilian's protection is not earned by moral virtue or loyalty, and an opposing soldier's exposure to attack is not a finding of criminal guilt. Status rules identify the legal circumstances in which particular persons may be intentionally attacked and the conditions under which that exposure ends. In an international armed conflict, membership in the armed forces of a party ordinarily provides a relevant combatant classification, subject to treaty detail. In a non-international armed conflict, the status of organized armed groups and direct participation requires a different analysis; the concept of continuous combat function appears in the ICRC's 2009 Interpretive Guidance and remains a subject of legal disagreement about its limits (ICRC, 2009b).

The moral importance of these distinctions becomes clear when legal and ethical ideas of innocence diverge. A conscript may be compelled to fight for an unjust cause without possessing personal culpability commensurate with intentional killing. A civilian financier may knowingly advance an unjust campaign without thereby becoming lawfully targetable merely by moral complicity. The law allocates targeting protection through public criteria that an adversary can in principle apply without accessing each person's inner moral history. This allocation can be morally imperfect in individual cases while retaining a compelling justification under conditions of uncertain information and systematic risk (Lazar, 2015; McMahan, 2009).

Direct participation presents a narrow yet difficult departure. A person who performs a hostile act may lose protection from direct attack for the relevant time under the applicable rule; the meaning of direct participation depends on conduct and the relevant legal standard. Political advocacy, sympathy, generalized economic contribution, and ordinary civilian employment cannot simply be equated with an immediately targetable hostile act. Broadening the category by association would dissolve the distinction. Narrowing it so far that organized armed violence could be concealed entirely within civilian status would also endanger the rule's coherence. The proper analysis must attend to the nature of the act, causal contribution, belligerent connection, and temporal boundaries while recognizing disagreements over interpretation (ICRC, 2009b).

A system that makes civilian immunity dependable must also protect uncertainty against self-serving certainty. The obligation to verify does not require infallibility. It requires disciplined inquiry proportionate to the information and capabilities available. The adversary's violation of distinction, including unlawful use of human shields, does not release the attacking party from its own duties toward civilians (Protocol I, 1977, Art. 51(7)–(8)). This proposition rejects guilt by proximity as a basis for targeting.

Military Objectives and the Moral Structure of Things

Military objectives are not identified by the importance an attacker assigns to destruction. Article 52(2) of Additional Protocol I concerns objects that, by their nature, location, purpose, or use, make an effective contribution to military action and whose destruction, capture, or neutralization, under circumstances ruling at the time, offers a definite military advantage. Civilian objects retain general protection. Article 52(3) supplies an explicit presumption in cases of doubt regarding normally civilian objects such as homes and schools under the Protocol's terms (Protocol I, 1977, Art. 52).

This definition carries two analytically distinct tests. Contribution addresses the object's relationship to the adversary's military action. Advantage addresses the attacker's anticipated operational gain from neutralizing it. Both are temporal and contextual. A bridge used by armed forces may have a military function, yet the consequences of its destruction for civilians and alternative routes still matter under other rules. A power station may simultaneously serve military communications and a hospital's life support. Classification does not erase the separate question whether a proposed attack would cause excessive incidental harm, or whether less harmful means are feasible.

The concept of military necessity is often abused through an equivocation. In strategic discourse, necessity can mean whatever a commander strongly desires in pursuit of victory. In IHL, military necessity operates within the legally permitted means and purposes of war, and cannot transform an expressly prohibited act into a permitted one merely by demonstrating its utility. The Lieber Code already distinguished necessity from cruelty, though its historical standards differed from modern rules. Contemporary law converts this distinction into limits on eligible targets, methods, expected effects, and the treatment of persons in custody (Lieber, 1863, Arts. 14–16; Protocol I, 1977, Arts. 35, 48–57).

The philosophical foundation of these rules is that objects are normatively significant through their relations to persons and protected activities. Houses, water networks, cultural monuments, agricultural areas, and hospitals bear values that military advantage cannot simply absorb. Some possess special treaty protection; others receive the general protection of civilian objects. A rule about objects is thus partly a rule about the conditions of human survival, collective memory, and the ability to return to peace. The destruction of a civilian water network may have enduring effects on health and displacement that are not captured by counting deaths at the moment of impact.

A disciplined targeting judgment must accordingly ask: What precisely makes this object a military objective? What definite advantage is anticipated now? What civilian uses remain? What harmful effects are reasonably foreseeable? Which alternative means are available? The question of the target's category precedes, and cannot be replaced by, the question of proportionality. No favorable ratio between military advantage and civilian loss legitimizes an object that is not a lawful military objective.

Proportionality in Attack: A Norm of Comparative Judgment

Proportionality in attack prohibits an attack expected to cause incidental civilian death, injury, or damage to civilian objects excessive in relation to the concrete and direct military advantage anticipated. This is distinct from jus ad bellum proportionality, which evaluates the overall harms and goods of resorting to or continuing a war. It is also distinct from a simple numerical ratio of civilians to combatants. The relevant comparison involves foreseeable incidental harm and the anticipated concrete and direct military advantage of the attack, assessed on the information reasonably available before the decision (Protocol I, 1977, Arts. 51(5)(b), 57(2)(a)(iii)).

A fundamental philosophical difficulty is that the values being compared are not commensurable in a straightforward mathematical unit. Civilian death, permanent disability, deprivation of medical care, and damage to housing cannot be directly converted into a standardized currency of military benefit. The law nevertheless requires judgment. This does not render the principle meaningless; many moral and legal standards involve qualitative assessment under incomplete knowledge. The judgment must be reasoned, contextual, and constrained by the prohibition against treating civilian lives as expendable inputs in a calculus of institutional prestige.

Suppose a lawful military objective lies next to a densely populated building. A commander expects neutralizing the objective to interrupt hostile operations for a limited period but anticipates severe civilian losses. Proportionality asks whether those incidental losses would be excessive in light of the advantage concretely expected. It does not ask whether the attacking state is generally fighting a just war, or whether the opposing government has committed atrocities elsewhere. A cumulative record of enemy wrongs does not increase the acceptable harm of this particular attack. The decision must rest on the target, expected effects, and relevant military advantage as determined under applicable law.

Foreseeable indirect effects complicate the analysis. A destroyed transformer may disable hospital equipment, pumps, refrigeration, or communications. Some effects follow through complex causal chains and may be uncertain. Reasonable foreseeability, available information, temporal proximity, and causal evidence therefore become important to legal and moral analysis. A decision-maker who deliberately avoids readily available information to preserve an appearance of innocence violates the spirit of the duty to assess harm. Conversely, hindsight alone cannot establish that a responsible ex ante assessment was defective simply because an unanticipated consequence occurred (Haque, 2017; Dill, 2015).

Proportionality does not authorize intentional attacks on civilians. The rule concerns incidental harm arising from attacks on military objectives; an attack directed at civilians fails at the threshold of distinction. This logical order matters. If proportionality were treated as a universal permission to trade any civilian harm for sufficiently large military gain, the fundamental prohibition on targeting civilians would collapse. The analysis must therefore proceed through target eligibility, foreseeable consequences, comparison, and feasible precautions, with each stage possessing independent normative force.

Precautions and the Ethics of Responsible Knowledge

Additional Protocol I, Article 57, requires constant care to spare civilians and civilian objects in military operations. With respect to attacks, it requires feasible verification of objectives, feasible precautions in the choice of means and methods, assessment of excessive incidental harm, cancellation or suspension when later information shows the target or anticipated effects fail the governing rules, and effective advance warning unless circumstances do not permit. Where several targets yield a similar military advantage, the rule calls for selection of the one expected to endanger civilians and civilian objects least (Protocol I, 1977, Art. 57).

The obligations establish an epistemic dimension of humanitarian law. Moral responsibility concerns not only what decision-makers know but what they reasonably ought to seek to know. The sequence between information, deliberation, and action becomes legally significant. A commander who can review updated surveillance, distinguish hospital infrastructure from an armed position, or alter a munition's timing has reasons to do so that cannot be negated by the convenience of a preexisting operational plan. The relevant standard is feasibility under circumstances; it is neither omniscience nor unlimited expenditure of resources.

The distinction between positive and negative duties matters. A prohibition on intentionally attacking civilians tells a belligerent what cannot be chosen as an aim. A precautionary duty also governs what must be investigated, altered, warned about, or abandoned before an otherwise eligible attack proceeds. This renders restraint proactive. The law's protective value would be substantially diminished if parties could satisfy it simply by averring that they did not subjectively intend the civilian harm they failed to examine.

Duty to cancel or suspend is especially revealing. It makes legality sensitive to time. An attack that appeared lawful at the planning stage may become unlawful when civilians enter the area or evidence undermines the target's military character. An order is therefore not a moral certificate whose value survives every change of circumstance. Command responsibility includes the willingness and institutional capacity to revise action. Operational doctrine, information flow, and authority to abort an attack can determine whether the legal rule has practical force.

The philosophical conclusion is that responsible wartime agency is informationally structured. A military organization must build systems able to identify protected persons, represent uncertainty, communicate changes, record decisions, and learn from errors. Humanitarian obligation cannot be reduced to the character of one morally conscientious commander. It is an institutional discipline of knowing, deciding, and correcting in conditions where a mistake can irreversibly destroy lives.

Intention, Double Effect, and Foreseeable Harm

The doctrine of double effect is often invoked in the ethics of war to distinguish intending harm from foreseeing harm as a side effect of pursuing a permissible end. In its familiar form, the argument differentiates the object of an action, the agent's intention, the role of the harmful effect in producing the desired good, and the proportionality of the resulting consequences. Just war theorists debate whether these distinctions explain why some foreseen civilian deaths may be morally permissible while direct targeting is prohibited (Walzer, 1977/2015; Lazar, 2016).

A simplistic reading of double effect is dangerous. An actor cannot avoid moral responsibility for foreseeable devastation merely by describing it as regrettable. Nor can the agent's self-reported intention substitute for evidence about the selected object, manner of attack, or actual structure of the causal plan. Choosing a method because it will terrorize civilians remains a choice about civilians even when the ultimate political end is described as defense. Conversely, attacking a genuine military objective after serious precautionary assessment may create foreseeable incidental risks that call for proportionality analysis rather than the assertion that the civilians were intended targets.

The legal structure and philosophical doctrine overlap without coinciding. Additional Protocol I prohibits direct attacks on civilians, indiscriminate attacks, and attacks involving excessive expected incidental harm. It does not simply incorporate one philosopher's account of double effect. In criminal proceedings, mental elements are determined by the applicable offense and general principles of liability; an academic distinction between intended and foreseen consequences cannot override statutory language (Protocol I, 1977, Art. 51; ICC, 1998, Arts. 8, 30).

Critics of intention-centered approaches ask why persons suffering an identical harm should receive radically different moral treatment depending on the perpetrator's internal description of the causal process. Defenders argue that the way an agent incorporates a person into a plan is independently important. Both concerns are valid. Intention helps identify whether persons are treated as means, while foreseeability and proportionality test whether the agent may impose the anticipated side effects. A full theory of restraint must unite rather than substitute these dimensions.

The stronger account regards intention as one element of responsible practical reasoning. Target classification, anticipated military advantage, feasible alternatives, expected suffering, and the agent's reasons for choosing a course of action jointly determine whether an attack is defensible. The burden lies with those exercising force to give reasons that others can assess, including reasons about foreseeable harm they would prefer not to cause.

Combatant Equality, Legal Symmetry, and Moral Inequality

One of the most difficult disputes concerns the treatment of combatants on opposing sides. Traditional just war theory frequently defends a moral equality of soldiers under the jus in bello: fighters on either side may fight within the same constraints regardless of the overall justice of the war. Michael Walzer's Just and Unjust Wars offers an influential defense of this position. Jeff McMahan argues that combatants on an unjust side generally lack the moral permissions attributed to them by traditionalists, since participation in an unjust threat is morally relevant to liability and defense (Walzer, 1977/2015; McMahan, 2009).

The debate can be reconstructed through competing arguments. First, if killing normally requires a justification grounded in preventing wrongful threats, a soldier's cause appears relevant to whether the killing is justified. Second, if combatants in an unjust war contribute to wrongful killing, their exposure to defense cannot be treated as morally identical to the exposure of defenders merely because both wear uniforms. Third, however, the capacity of particular soldiers to recognize the justness of the overall war may be limited by propaganda, coercion, incomplete information, and conflicting claims. Finally, a public legal system must specify workable rules that continue to bind both sides despite the foreseeable disagreement over which side is right.

Legal symmetry can thus be defended without claiming complete moral equality. The law can apply the same conduct restrictions to all parties because unequal rules based on unilateral assessments of justice would undermine the reliability of protection. Adil Ahmad Haque develops a further point: the law's symmetrical prohibitions should not be mistaken for an unconditional legal moral permission to participate in aggression. Equal limits on killing are compatible with independent rules against the wrongful resort to force (Haque, 2017).

This distinction resolves an important conceptual confusion. Equality of restraint means that each side owes protective duties. It does not mean that each cause is equally defensible, or that individuals have no responsibility for knowingly advancing an unjust campaign. The most defensible legal architecture separates the public ability to identify and enforce protective standards from the moral appraisal of participation. The moral assessment can be richer and more discriminating than the law's status categories, while the legal categories protect people precisely because they do not require agreement on every contested moral question.

The difficulty remains acute for conscription, coercion, ignorance, and resistance. A conscript who reasonably but mistakenly believes the war defensive may possess a different degree of culpability from a leader who knowingly initiates aggression. A soldier who obeys an unlawful order to attack civilians faces different questions from one who disobeys and accepts grave personal risk. A serious moral theory must keep permission, liability, excuse, and obligation distinct. The law's independence is a safeguard against indiscriminate punishment, not an exemption from these ethical questions.

Surrender, Wounds, and the Moral End of Threat

A person who is hors de combat has ceased to be a legitimate object of attack under the relevant rules. Additional Protocol I, Article 41, recognizes persons in an adversary's power, those clearly expressing an intention to surrender, and those incapacitated by wounds or sickness and unable to defend themselves, subject to the stated conditions concerning hostile acts and escape. Common Article 3 similarly protects persons who have laid down their arms or are no longer active because of detention, injury, or other causes (Protocol I, 1977, Art. 41; Geneva Convention I, 1949, common Art. 3).

The philosophical basis is straightforward but profound. Even if participation in armed violence could justify force against a person, that justification is conditional upon the relevant threat and legal status. Once a soldier surrenders or is incapacitated, the reasons for intentionally killing that person are transformed. Continued violence against a captive would convert military force into an injury inflicted on someone who can no longer pose the same immediate combat threat. Revenge cannot restore the justification that active fighting once supplied.

The notion of humanity here is expressed through temporal responsiveness. Legal and moral obligations change as the person's situation changes. An opponent can become a prisoner, a wounded patient, or a civilian again; the categories have distinct legal routes, but all prohibit treating mere enmity as permanent forfeiture of protection. This places a demand on command practices: surrender must be recognizable, signals must be communicated, and forces must have procedures for handling detainees and wounded persons under dangerous conditions.

Complex battlefield cases require factual judgment. A feigned surrender can implicate perfidy rules; a purported surrender while continuing hostile acts may fail the legal conditions. Such possibilities do not authorize a policy of refusing all surrender. The law protects the categories through standards that can be applied amid uncertainty rather than through presumptions that every adversary is deceptive. The philosophical achievement lies in making the enemy's changed relation to violence morally decisive.

Detention, Prisoners of War, and the Limits of Captive Power

Detention is a distinctive test of humanitarian authority. Once a party has physical control over a person, the public reasons that may justify temporary deprivation of liberty cannot justify humiliation, torture, starvation, disappearance, or arbitrary punishment. The Third Geneva Convention establishes extensive protections for prisoners of war in international armed conflicts, including humane treatment, respect for persons, conditions of captivity, discipline, contact, and release. Common Article 3 protects persons in non-international conflicts at a minimum, while other applicable rules supply additional safeguards (Geneva Convention III, 1949; Geneva Convention I, 1949, common Art. 3).

Prisoner-of-war status involves legal classification rather than an award for morally just fighting. In the relevant international-conflict settings, combatants who meet the applicable status conditions may be detained while receiving the Convention's protections. Detention serving to prevent return to hostilities differs from punishment for an ordinary crime. A person may be held because of combatant status without having been individually convicted of wrongdoing, and mistreatment remains impermissible regardless of the merits of the military cause. This is a particularly important example of the distinction between preventive authority and punitive desert.

A humanitarian theory of detention should examine four questions. What justifies the initial deprivation of liberty? Which authority is competent to detain? What procedures allow the person's status, health, and rights to be assessed? What obligations survive until release and return? The legal answers vary across international and non-international armed conflicts, treaty regimes, and potentially applicable human-rights law. Moral reasoning nonetheless identifies a constant concern: control over another person makes the captor responsible for foreseeable harms produced by that control.

The transformation from enemy to detainee also creates an asymmetry of capacity. Captors decide access to food, medical care, communication, shelter, and legal process. Because prisoners cannot normally provide these necessities for themselves, rights require positive duties of provision. The ethical relation is therefore more demanding than a simple rule to refrain from killing. Human dignity is expressed institutionally through adequate conditions and accountable procedures, not only through ceremonial respect.

Humanitarian law remains especially important where fear and anger are strongest. A claim that an enemy committed atrocities cannot justify torture of prisoners. The distinction between investigating individual criminal conduct and imposing collective reprisals protects both justice and the factual process needed to identify genuine offenders. The same restraint that prevents abuse also makes credible evidence and lawful proceedings possible.

Medical Protection, Humanitarian Relief, and the Priority of Need

The Geneva tradition began in part with concern for the wounded. Humanitarian law protects medical personnel, units, transports, and emblems under applicable conditions, while imposing duties to collect and care for the wounded and sick without adverse distinction. Common Article 3 requires that the wounded and sick be collected and cared for. Additional Protocol I extends medical protections and sets conditions concerning misuse and the possible loss of special protection. Its Article 13 specifies, for civilian medical units, safeguards including warning in circumstances of harmful use before protection ceases (Geneva Convention I, 1949; Protocol I, 1977, Arts. 12–16).

The philosophical principle is need-responsive duty. A wounded opponent's claim to assistance arises from vulnerability and the basic human interest in survival. Medical judgment allocates care according to clinical urgency and available capability rather than national loyalty. This does not mean wartime medical systems have no resource constraints; it means that political hostility is not an admissible reason to withdraw the basic protections. Medical neutrality and impartiality convert this ethical demand into an institutional expectation that armed actors must recognize.

Relief operations involve additional legal complexities. Under relevant treaty provisions, humanitarian and impartial relief actions may be subject to consent and control arrangements, but those provisions must be interpreted consistently with protections against starvation, unlawful collective punishment, and the requirements imposed by occupation. Article 70 of Additional Protocol I addresses relief for inadequately supplied civilian populations outside occupied territory, while the Fourth Geneva Convention has specific duties concerning occupied territories. The legitimacy of screening relief shipments cannot be converted into a practical veto that makes protective duties meaningless (Protocol I, 1977, Art. 70; Geneva Convention IV, 1949, Arts. 55, 59; ICRC, 2025).

Relief highlights why wartime morality cannot be exhausted by the law of killing. The person threatened by lack of water, medical oxygen, shelter, or insulin is endangered through a chain of decisions about access and infrastructure. A purely kinetic account of warfare overlooks systematic harms that may continue after immediate hostilities. Positive obligations concerning aid and care are therefore integral to the same humanitarian architecture as distinction and proportionality.

The special protections also reveal a structural paradox: medical institutions must remain identifiable to the parties from which they seek protection. Distinctive emblems, notification, coordination, and recordkeeping can reduce risk, yet may be abused by parties who target rather than spare protected services. The possibility of abuse cannot justify abolishing the legal commitment. It strengthens the need for verification, independent monitoring, and accountability for attacks on protected medical activity.

Starvation, Siege, and the Conditions of Civilian Survival

Starvation of civilians as a method of warfare is prohibited under Additional Protocol I, Article 54(1), which further protects objects indispensable to civilian survival under specified conditions. Additional Protocol II, Article 14, prohibits starvation of civilians as a method of combat and protects essential objects such as foodstuffs, crops, livestock, drinking water installations, and irrigation works. Relevant customary rules likewise prohibit starvation of civilians (Protocol I, 1977, Art. 54; Protocol II, 1977, Art. 14; Henckaerts & Doswald-Beck, 2005, Rules 53–56).

These rules demonstrate a historical movement from a narrower conception of siege toward stronger protection of civilian survival. The Lieber Code's Article 17 accepted starvation of an adversary in broad terms, reflecting nineteenth-century assumptions that later humanitarian law significantly constrained. The difference is an instructive historical case: principles described as military necessity change in content when law more fully recognizes the vulnerability and rights of those subject to them (Lieber, 1863, Art. 17).

The ethical wrong of deliberate civilian starvation has several dimensions. It exploits dependency on food and water systems, transforms ordinary life-sustaining needs into instruments of coercion, distributes suffering among children and other vulnerable persons without regard to responsibility, and may force displacement. Its effects can include disease, disability, and mortality well beyond immediate military encounters. The prohibited method therefore reveals how violence can be mediated through control over social reproduction and essential infrastructure.

Not every siege or disruption of supplies has the same legal classification, and specific prohibitions, exceptions, evacuation conditions, and duties of relief must be examined in their treaty and factual context. The law's analytical demand is to identify whether civilians are deliberately deprived of indispensable objects, what the foreseeable consequences are, and what separate obligations govern humanitarian access. Calling an operation militarily valuable cannot settle these inquiries.

This topic also exposes limitations of a morality centered exclusively on the attacker's immediate intention to kill. A policy of systematic deprivation may cause mass suffering without a single projectile directed at a civilian. Humanitarian law protects persons through rules about conditions of life and the means by which an adversary may seek advantage. The object of protection is the continued possibility of human survival in an environment distorted by armed power.

Occupation, Political Authority, and Protection Without Sovereignty

Military occupation raises a question of public power: what obligations arise when a belligerent exercises effective control over territory whose sovereignty it does not acquire through that control? The Hague Regulations of 1907, particularly Article 43, address the occupant's duty to restore and ensure public order and civil life as far as possible while respecting the existing laws unless absolutely prevented. The Fourth Geneva Convention adds extensive protections for persons and resources in occupied territory, including provisions concerning transfers, food, medical supplies, and relief (Hague Convention IV, 1907, annexed Regulations, Art. 43; Geneva Convention IV, 1949, Arts. 47–78).

Occupation is a situation of factual authority under legal restraints. The authority of the occupant to take some security measures does not generate an unrestricted right to annex territory, permanently transform its population, or use its inhabitants as instruments of military advantage. Article 49 of the Fourth Convention prohibits forcible deportations and transfers of protected persons from occupied territory, subject to the narrow terms concerning temporary evacuation, and prohibits the occupying power from transferring parts of its own civilian population into the occupied territory. The 2025 ICRC commentary explains the protective purposes and contemporary interpretations of these rules (Geneva Convention IV, 1949, Art. 49; ICRC, 2025).

The legal structure expresses the difference between control and legitimate ownership. Military power can establish the practical capacity to command, administer, or detain; the legitimacy and scope of each exercise require separate legal justification. This problem connects occupation law to general jurisprudence: legal authority cannot be inferred solely from the fact that an actor is obeyed. A ruler's obligations may arise precisely because it possesses de facto power over persons who have not consented to its political project.

Occupation also shows why humanitarian law and human-rights law can interact. The International Court of Justice in its 2004 Wall advisory opinion addressed the applicability of human-rights instruments alongside humanitarian law to occupied territory. The exact relation depends on the right, treaty, and factual context. A blanket assertion that human rights disappear upon the outbreak of war would be inconsistent with this jurisprudence. Equally, the specific structure of occupation law must be taken seriously rather than replaced by an undifferentiated moral vocabulary (ICJ, 2004).

A philosophical assessment of occupation must keep temporary administration, security needs, protection of local institutions, property rights, political self-determination, and future peace within one framework. The fact that the occupant controls an administrative system increases its duty to preserve civilian life and lawful institutions. The order that follows conquest cannot be justified by the capacity to impose it.

Non-International Conflict and the Authority of Armed Groups

Non-international armed conflict presents a particular challenge because some actors are not states and may lack the kind of public authority assumed in classical just war theory. Common Article 3 is binding on each party to a qualifying non-international conflict, and its operation does not confer recognition or legitimacy upon an armed group. Additional Protocol II applies in the narrower circumstances specified in Article 1. A non-state party's obligation to treat captives humanely is thus conceptually separable from whether its resort to arms is politically or morally justified (Geneva Convention I, 1949, common Art. 3; Protocol II, 1977, Arts. 1, 4–6).

The distinction matters where governments portray opponents as criminals and armed groups portray governments as illegitimate. These classifications may be relevant to domestic criminal law or political philosophy, but they do not dissolve elementary humanitarian protections. A state's duty to protect detainees remains. An insurgent command's duty not to target civilians remains. At the same time, direct participation by civilians in a non-international armed conflict does not automatically grant combatant privilege under domestic law. The ICRC's analysis distinguishes loss of protection from direct attack from the separate question of domestic criminal liability (ICRC, 2009b).

The philosophical authority of humanitarian norms therefore extends beyond recognized sovereign command. Their binding character can be understood through the exposure and control that armed organization creates. A group capable of capturing persons, organizing attacks, and holding territory is also capable of causing predictable harms that the law demands it restrain. Political recognition is a different question from the existence of the obligations generated by organized hostilities.

This is not a theory that all armed groups are institutionally equivalent to states. Organization, capacity, territorial control, and the applicable legal framework vary. Different treaty regimes use different thresholds, and rights of combatants in international conflict are not simply copied into civil war. The point is that the absence of statehood does not entail the absence of law. A legal order committed to protecting persons must operate where violence is organized, including where the political status of the parties is radically disputed.

War Crimes, Criminal Liability, and Command Responsibility

Humanitarian law establishes obligations of conduct, while international criminal law assigns individual liability for specified serious violations through distinct legal elements. The Rome Statute of 1998 establishes the International Criminal Court and, under Article 8, defines categories of war crimes for international and non-international armed conflicts. The Statute must be read with its relevant amendments, Elements of Crimes, jurisdictional provisions, admissibility rules, and standards of culpability. It does not convert every tactical failure, treaty breach, or morally reprehensible policy automatically into a conviction for a war crime (ICC, 1998, Arts. 5, 8, 25, 28, 30).

The philosophical relation between prohibited action and individual guilt is complex. A commander may make a decision upon inaccurate information despite taking reasonable precautions. Another may deliberately target civilians while describing them as military objectives. A third may omit to prevent or repress criminal acts committed by subordinates in circumstances that satisfy applicable command-responsibility standards. Each case involves different mental elements, causal relations, forms of responsibility, and evidentiary burdens. A legal theory committed to the protection of victims must also preserve the fairness of judgment about individual defendants.

Command responsibility is especially important because armed organizations distribute knowledge and control. Senior leaders may establish doctrine, supply weapons, issue orders, create incentives, and control reporting systems without physically carrying out each attack. The law accordingly asks questions about authority, knowledge or required awareness, effective control, failure to take necessary and reasonable measures, and causal or relational conditions set by the applicable provision. The precise elements differ between legal regimes and should never be replaced by a generalized claim that every superior is responsible for every subordinate act (ICC, 1998, Art. 28).

International adjudication also requires distinguishing responsibility for the lawfulness of war from responsibility for its conduct. The international crime of aggression concerns a specified form of leadership responsibility for a state's act of aggression, subject to jurisdictional limitations. War crimes focus on listed conduct within armed conflict. A prosecution for one offense does not logically prove the other. This distinction preserves the separate authority of the jus ad bellum and jus in bello while permitting responsibility on both grounds where legal conditions are met.

The deeper philosophical task is to reconcile the impersonal protection of rules with the personal character of punishment. The duty not to attack civilians binds every party. The culpability of an accused person must be demonstrated through law and evidence. Humanitarian accountability is therefore a structured process rather than a declaration of collective moral condemnation. A credible system requires investigation, preserved evidence, public reasoning, independent adjudication, and remedies for victims; without these, the vocabulary of war crimes risks becoming a rhetorical instrument of the stronger side.

Judicial Interpretation and the Institutional Meaning of Humanity

Judicial practice clarifies how abstract principles operate in hard circumstances. The International Court of Justice's 1996 advisory opinion on the legality of the threat or use of nuclear weapons discusses the fundamental importance of distinction between combatants and civilians and the prohibition of unnecessary suffering. The Court analyzed those principles within the applicable international law, including the United Nations Charter and humanitarian law, while its ultimate conclusions about nuclear weapons included important limitations and divisions. A responsible citation should report both its recognition of cardinal rules and the restricted reach of its answers, rather than attributing an uncomplicated categorical decision the Court did not make (ICJ, 1996).

The 2004 advisory opinion on the construction of a wall in the Occupied Palestinian Territory provides another important example. The Court examined the interaction of the Hague Regulations, the Fourth Geneva Convention, human-rights instruments, and the law of self-determination. Its reasoning demonstrates how military occupation activates obligations that cannot be reduced to the simple legality of one attack. An occupying power's use of coercive authority affects movement, livelihood, property, administration, and a population's ability to exercise political rights (ICJ, 2004).

The Tadić litigation at the International Criminal Tribunal for the former Yugoslavia helped articulate the threshold for armed conflict and the relevance of conflict classification to international humanitarian adjudication. The Appeals Chamber's Decision on the Defence Motion for Interlocutory Appeal on Jurisdiction, issued on October 2, 1995, articulated the now frequently cited conception of armed conflict as involving interstate armed force or sufficiently protracted armed violence between governmental authorities and organized armed groups or among such groups. The doctrinal value of the decision depends on reading its actual procedural and jurisdictional context (ICTY, 1995, para. 70).

Judicial interpretation necessarily occupies a space between general rule and disputed fact. Judges must determine the content of law, assess testimony, evaluate documents, and identify whether legal elements are satisfied. This creates a distinct epistemology of accountability. Proof beyond a reasonable doubt in criminal proceedings, the evaluation of state responsibility in contentious inter-state cases, and advisory opinions concerning general legal obligations have different functions. Their conclusions cannot be transferred across contexts without analyzing the distinct legal and evidentiary questions.

The development of jurisprudence also illustrates why humanitarian law is historically continuous yet open to refinement. Treaty text remains binding in its own right; later interpretation can identify implications for technologies, detention practices, and displacement that the original drafters could not describe in detail. Good-faith interpretation protects continuity with the norm's legal source and responsiveness to new circumstances. The ICRC's updated 2025 commentary on the Fourth Geneva Convention participates in this interpretive conversation as an expert resource, while remaining distinguishable from an authoritative judicial ruling or a new treaty (ICRC, 2025a).

Sexual Violence, Gender, and the Unequal Burdens of Conflict

International humanitarian law prohibits rape and other forms of sexual violence in armed conflict, and the Rome Statute includes specified forms among war crimes where the legal elements are met. Sexual violence may also constitute other international crimes under their respective conditions. The harm extends beyond immediate bodily injury to coercive control, trauma, social exclusion, and the destruction of relationships within families and communities (ICC, 1998, Art. 8; ICRC, 2016).

Historically, accounts centered on battlefield tactics sometimes treated the experiences of persons subjected to sexual violence, displacement, or coerced labor as peripheral to the moral evaluation of war. Such omissions narrow the object of analysis without justification. Humanitarian law recognizes the treatment of persons in a party's power as a central component of wartime obligation. A philosophy of restraint that focuses solely on the lawfulness of attacks leaves extensive spheres of coercion morally invisible.

Gendered patterns of harm require factual study and care in attribution. Women and girls may face particular risks in some conflicts, while men, boys, and persons of diverse identities also experience sexual violence and detention-related abuse. The ethical point is universal protection combined with attention to the forms in which vulnerability is actually distributed. The need for specialized medical assistance, confidential evidence handling, and survivor-centered treatment follows from concrete harms rather than from an assumption that all victims occupy identical circumstances.

This subject also tests the difference between humanitarian language and institutional capability. A legal prohibition is effective only when reporting does not expose victims to retaliation, investigators are competent to preserve evidence, and military and judicial authorities treat allegations with seriousness. The victims' rights to care and dignity cannot wait until criminal culpability has been established. Protective duties and criminal adjudication have different temporal structures. One responds to immediate vulnerability; the other determines responsibility through evidence and procedure.

The general conclusion is that the law of armed conflict regulates domination as well as destruction. It governs what the powerful may do to those rendered vulnerable through captivity, territorial control, social collapse, and physical dependence. This broader conception better captures the moral foundation of humanitarian law than a theory that begins and ends with permissible killing.

Reprisals, Reciprocity, and the Temptation of Retaliation

A striking feature of humanitarian norms is their resistance to the claim that a breach by the enemy excuses further breaches. Additional Protocol I, Article 51(6), prohibits attacks against civilians by way of reprisals. Article 52 addresses reprisals against civilian objects, and other provisions protect additional categories. Precise legal treatment of reprisals varies by context, applicable treaty, and customary rule, but the core philosophical issue is general: can wrongdoing by one actor justify targeting persons who possess no individual responsibility for it? (Protocol I, 1977, Arts. 51(6), 52(1), 54(4); ICRC, 2026a).

Retaliatory logic often conflates political membership with liability. If a government attacked civilians, members of that government's civilian population do not thereby become legitimate targets. Punishment by association would make the protective rule contingent on the adversary's behavior and would undermine the law's commitment to distinction. The stronger party could simply define its violence as a response to the weaker party's previous violation, producing a continuous cycle of ostensibly justified cruelty.

Reciprocity nevertheless matters to the practical history of compliance. Parties may be more willing to protect prisoners when they expect their own soldiers to be protected. These instrumental incentives can reinforce the legal order, but they cannot form its sole foundation. Humanitarian protection is especially necessary when expectations of reciprocity deteriorate. The rule's point is to preserve the status of persons amid failure of political trust, not merely to reward cooperative opponents.

An institutional theory of restraint must therefore distinguish incentives from authority. Reciprocal agreement can help explain why states adopt rules and why they may comply. The validity of an obligation to spare civilians does not disappear because the adversary violated it. This distinction is the public-law analogue of a familiar moral idea: another person's wrongdoing changes what can justifiably be done to that wrongdoer, but does not license arbitrary injury to third persons.

The Environment, Cultural Heritage, and Intergenerational Harm

War damages more than immediate participants. It can contaminate soil, disrupt ecosystems, destroy archives and monuments, and permanently impair the conditions under which later generations can live. Additional Protocol I contains protections relating to environmental damage, including Articles 35(3) and 55, while various treaty and customary provisions protect cultural property and other civilian objects under specific conditions. The 1954 Hague Convention for the Protection of Cultural Property and subsequent instruments add specialized obligations. The ICRC's 2020 guidelines synthesize relevant legal rules and recommendations concerning the natural environment (Protocol I, 1977, Arts. 35(3), 55; UNESCO, 1954; ICRC, 2020).

Environmental destruction is a difficult case for proportionality because harm can unfold across time and across populations. Contamination of water systems can produce disease after military operations have ended. Damage to agricultural land can destroy livelihood and cause displacement. A legal analysis must identify the applicable prohibition, its threshold, causation, and anticipated consequences rather than assuming that every ecological harm meets the requirements of a specialized environmental war crime. Philosophically, however, the enduring effects enlarge the temporal field of responsibility.

The destruction of cultural objects also requires differentiated analysis. Museums, archives, historical sites, places of worship, and scientific collections may have special protection under applicable law, subject to its terms. Their loss can damage communal memory and the possibility of reconstructing political life. The moral wrong is not exhausted by market value or immediate military utility. Cultural heritage can be a condition of collective identity, education, and historical continuity, so its destruction may affect people who were neither present nor alive at the moment of attack.

This leads to a general conceptual distinction between immediate military advantage and the persistence of war's effects. The former is assessed under legal standards applicable to particular attacks; the latter matters to the evaluation of responsibility, relief, compensation, recovery, and peace. A legal framework that took only immediate casualties into account would fail to recognize significant pathways through which warfare harms human life.

The philosophical horizon is intergenerational. Those initiating and conducting war exercise power over conditions that future persons cannot consent to and may be unable to repair. This establishes strong reasons for precaution and for a comprehensive accounting of foreseeable ecological and cultural effects. These reasons complement existing humanitarian law while also showing where further rulemaking and institutional capacity may be necessary.

Autonomous Weapons, Human Judgment, and Accountability

Weapons that select and engage targets without further human intervention create a new version of an old problem: who must determine that a particular use of force satisfies the law? Under Additional Protocol I, Article 36, a contracting party studying, developing, acquiring, or adopting a new weapon, means, or method of warfare must determine whether its employment would be prohibited by the Protocol or other applicable law. Existing conduct-of-hostilities requirements remain relevant to autonomous systems, whatever additional rules states may develop (Protocol I, 1977, Art. 36).

The ICRC's 2021 position and 2026 briefing distinguish the applicability of existing IHL from proposals for additional prohibitions and restrictions. The organization recommends prohibiting unpredictable systems and autonomous systems designed or used to target human beings, alongside strong restrictions on other systems. These are proposals in the international regulatory debate, not descriptions of a treaty already universally adopted. The distinction between an existing legal duty and a proposed additional prohibition is crucial to a scientifically honest analysis (ICRC, 2021; ICRC, 2026b).

The deeper challenge is that an autonomous system may execute the physical act without a person choosing its precise target or time. A human operator can authorize a deployment area, target profile, or operating duration, while software and sensors produce the immediate selection. That arrangement complicates verification, proportionality assessment, foreseeable-effects analysis, and the opportunity to cancel an attack. The legally relevant question is not simply whether a human was involved at some earlier stage; it is whether responsible persons retain sufficient knowledge and control for the applicable obligations to be met in the actual circumstances.

Responsibility should be analyzed through the complete chain of decisions: design, testing, legal review, authorization, operational planning, selection of environment, supervision, and response to malfunction. Each stage can generate duties without making every designer or operator automatically guilty of every resulting injury. Technical complexity is a reason to develop traceability, meaningful testing, and clear command authority. It is not a reason to treat lethal effects as morally or legally orphaned.

The philosophical significance extends beyond artificial intelligence as a tool. Automated use-of-force systems place distributed technical processes within the execution of public coercion. The authority to use force remains a question about law, institution, and human responsibility even when the mechanism of target selection changes. The enduring principle is that the normative demand to justify an act cannot be satisfied merely by replacing the immediate actor with a machine.

Cyber Operations, Data, and Civilian Infrastructure

Cyber operations conducted in connection with armed conflict can disrupt medical facilities, water systems, transportation, communications, and other services upon which civilian populations depend. The ICRC emphasizes that existing IHL governs relevant cyber operations during armed conflict, including rules of distinction, proportionality, and precaution. There remain disagreements among states and specialists on particular questions, including whether all digital data is protected as an object under traditional conduct-of-hostilities provisions and the classification of some operations as attacks (ICRC, 2022; ICRC, 2025b).

The problem is conceptually distinctive because material harm may arise through interference with information. A system can be disabled, data corrupted, or access denied without an explosive strike on a physical structure. A hospital's functioning may depend on devices and records distributed across networks. The lawfulness of an operation consequently requires attention to reasonably foreseeable effects on protected people and services, not solely to the immediate physical surface of the operation.

Civilian involvement complicates the division between combatants and noncombatants. A civilian hacker may provide assistance ranging from political speech to technical support to directly hostile operations. Legal consequences depend on the person's conduct and applicable rules, not on the presence of a keyboard or a political affiliation. The ICRC's 2025 study on civilian cyber participation analyzes the protective risks and distinct legal questions arising from individual hackers, organized groups, and private technology companies (ICRC, 2025b).

The ethical principle is continuous with earlier humanitarian law: protected lives cannot be made invisible by technical mediation. Yet the evidentiary problem intensifies. Remote operations, attribution uncertainty, cascading failures, and opaque network dependencies may make reliable consequence assessment difficult. Precautions can require cyber-specific expertise, testing, limits on propagation, and ongoing monitoring. A weapon or operation that cannot reliably discriminate between military and civilian consequences raises serious questions about lawful deployment.

The philosophical contribution of this case is to show that the concept of force depends on effects, causal agency, and organized intent as well as physical form. New technical means expand the work needed to protect civilians while leaving the duty of restraint intact. What changes is the epistemic and institutional machinery through which that duty must be fulfilled.

Human Rights, Humanitarian Law, and Overlapping Legal Orders

Human-rights law and humanitarian law have distinct histories, legal structures, and fields of application, yet they may apply simultaneously during armed conflict. Humanitarian law addresses the specific situation of armed conflict; human-rights treaties establish broader duties that may remain binding in war, subject to their own conditions and lawful derogation regimes. The International Court of Justice's 2004 Wall opinion expressly considered both fields in assessing obligations concerning occupied territory (ICJ, 2004).

A philosophical error would be to treat humanitarian law as an exception that exhausts all protections, or to treat human-rights concepts as automatically decisive without interpreting the rules specialized for warfare. The relation is an interpretive problem of overlapping normative orders. Relevant principles of legal interpretation, treaty scope, state responsibility, and the concrete right under examination determine how obligations interact. The phrase lex specialis can help identify a specially applicable rule, but it is not a general instruction to discard the rest of international law.

The ethical significance lies in continuity of personhood across institutional settings. The same person can be a civilian living under occupation, a detainee, a patient, a displaced individual, or a resident whose freedom of movement is constrained. Different legal classifications select different duties, but the person's basic claim to lawful treatment does not cease whenever one regime becomes more specialized. The multiplicity of applicable sources may therefore reinforce protection rather than undermine it.

There is nevertheless a practical concern: overlapping obligations can make decisions more complex. Commanders, officials, and courts must know which law applies and how its standards interact. This is a reason for better institutional training and legal advice, not for the blanket replacement of legal specificity by broad moral declarations. A strong philosophy of humanitarian law must defend both the universal grounds of protection and the technical competence through which protection is realized.

The Critique of Formal Equality and the Politics of Humanitarianism

The equal application of humanitarian rules is an essential protection, but it can create a risk of misleading political narratives. A report that two belligerents have committed violations of comparable legal categories does not demonstrate that their resort to force was equally justified, that they possessed comparable resources, or that the magnitude of their conduct was identical. The juridical classification of an individual attack and the political history of a war are different objects of judgment. Legal analysis becomes distorted when one is used to erase the other.

Critics of humanitarianism also ask whether formal neutrality can obscure empire, occupation, economic power, or structural inequality. Historical arguments about civilization sometimes placed unequal populations under supposedly universal rules whose application was controlled by colonial authorities. Contemporary scholarship must confront those histories when explaining the genealogy of the law. The universal significance of civilian protection is compatible with rigorous criticism of the institutions through which claims to universality have been selectively administered (Vitoria, ca. 1539/1991; Dill, 2015).

A defensible response begins by preserving the difference between the moral standing of protected persons and the political claims of competing actors. Refusing to attack civilians does not imply neutrality about aggression or domination. Delivering impartial medical treatment does not imply that the causes of all wounded persons' suffering are morally equivalent. The humanitarian duty follows the person's need and protected status; the judgment of historical responsibility follows evidence about decisions, institutions, and acts.

A further criticism concerns the possibility that legal compliance becomes a legitimating performance. Sophisticated military organizations can deploy legal terminology, targeting documentation, or technical precision to create the appearance of restraint even while their overall strategy devastates civilian life. Formal procedures deserve no automatic presumption of substantive success. Their value depends on the quality of information, integrity of assessments, willingness to cancel attacks, effective external scrutiny, and actual protection of persons. Dill's research on the social construction of legitimate targets illustrates why institutional adoption of legal vocabulary and humanitarian results must be assessed separately (Dill, 2015).

The strongest theory combines universal rules with critical scrutiny of power. Equal restraints protect persons who would otherwise become captives of political disagreement. Critical analysis examines whether those restraints are being interpreted or administered in ways that reproduce domination. Neither function can substitute for the other. They jointly preserve the capacity to demand justification from every actor exercising organized force.

Humanitarian Intervention, Rescue, and the Limits of Moral Permission

Humanitarian intervention is often invoked where massive atrocities create an urgent moral demand for protection. Just war philosophers disagree about when external force can be justified, who possesses authority to use it, how risks to outsiders and local populations should be assessed, and whether apparently benevolent intervention can generate new forms of domination. The international legal framework adds separate questions about the prohibition on the use of force, Security Council authorization, self-defense, and contested unilateral claims of rescue (United Nations, 1945; Fabre, 2012; Walzer, 1977/2015).

The philosophical argument for rescue typically begins with the rights of people facing grave wrongs. If protective force is the only feasible means of preventing mass slaughter, refusing assistance may itself be morally troubling. The counterargument emphasizes that military intervention imposes grave risks on the very people it claims to protect, may be based on incomplete or manipulated information, and may allow powerful states to reinterpret self-interest as humanitarian necessity. Both arguments require attention to authority, necessity, proportionality, probability of success, and conditions after intervention.

The distinction between jus ad bellum and jus in bello remains decisive. A moral argument that intervention is justified cannot legalize attacks on civilians. A determination that intervention is unauthorized under the Charter does not release participating soldiers from humanitarian constraints. The relevant legal basis for using force cannot be inferred from the existence of atrocities alone without analyzing applicable international law. Likewise, the legality of using force must not be confused with the separate moral evaluation of refusing to protect endangered populations.

The synthesis is that rescuing persons through armed force is an especially demanding form of power. The protective end must be tested against the risks created by the means, the political legitimacy of decision-makers, and the prospect of preserving meaningful agency for the affected population. Humanitarian law is a floor of conduct that applies whichever side of the intervention debate prevails; just war theory examines whether initiating the enterprise was justified at all.

Institutions of Compliance: Training, Review, and Public Reasons

A rule's existence in a treaty does not ensure its realization. Military doctrine, training, rules of engagement, legal advisers, intelligence practices, detention procedures, medical coordination, and disciplinary mechanisms determine whether humanitarian obligations enter actual decisions. Public authorities must organize their institutions so that personnel can identify applicable law and implement it under operational pressure. The law's moral purpose becomes institutional through education, documentation, accountability, and the possibility of correcting orders.

The difficulty is especially acute where command systems reward rapid escalation, suppress evidence of civilian harm, or render responsibility diffuse. A prohibition may be formally acknowledged yet practically neutralized by defective incentives. The question for institutional ethics is whether the organization has made compliance possible and sufficiently authoritative at the level where decisions occur. A commander cannot satisfy precautionary obligations through an abstract policy whose practical implementation is systematically impossible.

Transparency also has limits in war. Intelligence sources, operational plans, and the identities of vulnerable persons may require confidentiality. But justified secrecy should not prevent credible post hoc scrutiny or independent investigations into serious violations. The structure of public reasons can accommodate protected evidence, judicial controls, and limited disclosure while maintaining the obligation to explain and account for coercive power. An unreviewable assertion of military necessity would leave humanitarian rules dependent on the very discretionary force they are designed to constrain.

This institutional dimension helps explain why ethical courage matters but cannot carry the entire burden of restraint. Individuals should be able to refuse clearly unlawful orders and report serious violations through legitimate mechanisms. Institutions must also reduce the personal costs of law-abiding action and prevent the concealment of wrongdoing. The philosopher's task is to identify the reasons for compliance; the jurist's and administrator's tasks include building durable procedures that make those reasons effective.

The Epistemology of Wartime Claims and Investigations

Wartime information is routinely fragmented, strategically presented, and emotionally charged. Reports may rely on anonymous sources, incomplete imagery, social-media fragments, contested casualty counts, or interpretations of military purpose that cannot be independently established. Philosophical inquiry must therefore distinguish an allegation, a supported factual finding, a legal classification, and a final judicial determination. Conflating them undermines both the protection of victims and the fairness owed to those accused.

Legal assessment often turns on what was known or reasonably available to decision-makers before an attack. This differs from what investigators learn afterward. An ex ante targeting judgment may have been unreasonable even if the intended target was eventually revealed to be military, because the planner lacked adequate verification. Conversely, a tragic outcome does not by itself prove an unlawful attack if the expected effects and feasible precautions met applicable standards at the relevant time. The distinction is methodological and should never become an excuse to ignore evidence of foreseeable harm (Protocol I, 1977, Art. 57; Haque, 2017).

Credible reconstruction requires a disciplined sequence: identify the applicable legal framework; establish the conflict and parties; reconstruct the relevant timeline; locate orders and available intelligence; determine the status of persons and objects; assess the method and foreseeability of harm; evaluate alternative precautions; and distinguish legal breach from individual criminal responsibility. Each stage admits uncertainty that should be specified rather than concealed by rhetorical conviction.

This procedure respects the different interests at stake. Victims deserve serious inquiry and remedies where violations are established. Accused persons deserve assessment under the governing legal elements and standards of proof. Public institutions require accurate information to reduce future harm. Philosophical analysis contributes by clarifying the concepts that organize an investigation and the inferences that the evidence can actually support.

Philosophical Synthesis: Protection as Publicly Answerable Restraint

The historical and doctrinal analysis supports a conception of humanitarian law as publicly answerable restraint. Its first ground is equal human standing: persons retain claims to humane treatment and protection from unlawful attack irrespective of their citizenship, beliefs, political allegiance, or an adversary's moral appraisal of their cause. Its second ground is feasible institutional judgment: duties must govern identifiable acts, actors, and decisions under conditions of uncertain information. Its third ground is accountability: an obligation becomes practically credible when violations can be investigated, explained, corrected, and where legally appropriate punished.

These grounds explain why humanitarian law should not depend on reciprocal compliance. Reciprocity encourages observance but cannot define the scope of a civilian's rights. They also explain why legality and morality must be distinguished without being isolated. Law translates reasons for restraint into public rules and procedures. Moral reasoning evaluates those rules, criticizes their omissions, and addresses the justice of war itself. Their interaction is productive when the moral argument deepens protection while respecting the institutional reason for common standards.

The conceptual architecture has five connected judgments. First, the resort to organized force requires a separate justification under the relevant moral and Charter standards. Second, the person or object selected for attack must fall within the applicable legal category. Third, expected incidental harm must be assessed against the anticipated definite military advantage under the governing proportionality rule. Fourth, feasible precautions and changing information can alter what may lawfully proceed. Fifth, the treatment of persons who are wounded, detained, displaced, or living under occupation generates obligations whose force continues beyond any immediate attack. Each judgment identifies a different failure that an appeal to military necessity cannot absorb.

The most important theoretical result is the separation of the moral assessment of a war's cause from the legal protection of those exposed to its conduct. It allows a philosophical critique of aggression without making the aggressor population unprotected. It allows criticism of unlawful attacks by a defensive belligerent without implying that defensive and aggressive purposes are equivalent. It allows legal symmetry in restraint while leaving moral responsibility for participation differentiated. This separation makes protective rules capable of working amid profound political conflict.

A second result concerns the relation between persons and institutions. Civilian immunity, medical care, prisoner protections, and rules for occupied territory require organizational capability. The dignity of persons is made effective by intelligence procedures, military training, humanitarian coordination, detention administration, judicial review, and documentary accountability. A culture of humanitarian restraint must therefore include systems that prevent foreseeable injury, not only individual intentions that deplore it.

A third result concerns the changing material form of conflict. Cyber operations and autonomous systems alter the distribution of decision-making and the pathways through which harm reaches civilians. Existing legal duties continue to provide a framework, but their practical realization requires new forms of technical knowledge and, where justified, additional public regulation. Humanitarian principles retain their authority through changing technologies precisely because they regulate responsibility for foreseeable effects rather than the mere appearance of a weapon.

Open Questions in Humanitarian Jurisprudence

Several questions remain open after the foregoing synthesis. The first concerns the relation between legally defined combatant status and individual moral liability. Revisionist arguments expose moral differences between defending against wrongful aggression and participating in it, while common legal rules preserve operationally workable restraints. A complete account must explain how law should respond where moral asymmetries are significant but difficult to determine reliably.

A second question concerns proportionality's epistemic content. What degree of investigation is feasible when information is limited, the adversary deliberately obstructs verification, and a decision must be taken quickly? How should long-term effects on essential services be represented in anticipatory assessment? These are questions of practical reason as well as legal interpretation. No abstract mathematical formula substitutes for a doctrine grounded in evidence, institutional capability, and the seriousness of the harms.

A third question concerns the boundaries of legal authority over new technical systems. When selection and timing are delegated to software, what constitutes adequate human assessment, control, and responsibility? Which design choices are themselves prohibited by existing law, and which call for new treaty regulation? The ICRC's current proposals provide one developed answer in the ongoing debate, not its final resolution (ICRC, 2026b).

A fourth question concerns the conditions of effective accountability. International criminal courts have limited jurisdiction, states may fail to investigate their own forces, and evidence can be inaccessible or manipulated. How should institutions combine national courts, international jurisdictions, fact-finding mechanisms, reparations, and humanitarian monitoring without prejudging guilt or reducing the victim's position to a tool of geopolitical messaging?

A fifth question concerns the temporal bounds of wartime obligation. Environmental degradation, destroyed infrastructure, displacement, and cultural loss often persist after hostilities cease. The relation among IHL, human-rights law, international responsibility, and the morality of reconstruction demands further analysis. A legal system that governs the moment of attack must also remain capable of recognizing the enduring conditions created by that attack.

A sixth question concerns the effects of inequality in political and technical power. Can protective law operate equally when parties have very different surveillance capacities, weapons, institutional support, and access to authoritative interpretation? Equal norms may demand different feasible precautions from actors with different capabilities while remaining equal in their protective rationale. This tension is central to any future philosophy of humane warfare.

Conclusion

Humanitarian law establishes obligations of restraint within armed conflict through a historically developed regime of treaty and customary rules. Its philosophical foundation lies in the protection of persons exposed to organized violence and in the demand that coercive power remain answerable to public norms. The history from theological just-war reasoning through early modern natural law, the Lieber Code, Dunant's humanitarian intervention, the Hague Regulations, and the Geneva Conventions shows a gradual transformation of general ethical commitments into status-specific, operational, and institutional rules.

The analysis establishes the autonomy of four essential questions: whether the resort to force is justified; whether a person or object may be attacked; whether anticipated incidental harm and precautions satisfy the governing requirements; and how persons in the power of a belligerent must be treated. None can be settled solely by another. A state exercising justified self-defense remains bound by every applicable humanitarian rule. An aggressor's violation cannot deprive prisoners or civilians of legal protection. Compliance with battlefield rules does not supply a justification for initiating aggression.

The philosophical dispute between traditional and revisionist just war theory reveals why moral judgments about soldiers and legal rules governing soldiers cannot simply be identified. Legal symmetry supplies a framework of common restraint where political causes are bitterly contested. Moral analysis retains its authority to distinguish culpable aggression from justified defense and culpable individuals from those acting under coercion or error. Together, they provide a stronger account than either an unconditional moral equality of all participation or a legal system whose protection would depend on the victory of one moral narrative.

The final conclusion is that humanitarian law is a continuing discipline of the use of power. It protects people by requiring classification, verification, proportionality, precautions, care, accountable detention, lawful administration, and serious investigation of wrongdoing. As warfare incorporates autonomous targeting, networked infrastructure, and dispersed technical agency, those requirements become more demanding in their implementation. Their enduring purpose remains clear: organized force must be governed by obligations that preserve the humanity and legal standing of those whom war places in danger.