How International Law Revises Itself — Treaties, Conventions, and Precedent

The fundamental problem of international law is the absence of a sovereign. In domestic legal systems, the law is backed by a state — an entity with a monopoly on legitimate violence that can enforce compliance. Remove that entity, and you have the condition of international relations: sovereign states that acknowledge no authority above themselves, interacting in what Thomas Hobbes called a state of nature at the interstate level.

In this environment, law cannot operate through the mechanisms of domestic legal systems. There is no world government that can imprison states that violate international norms. Enforcement depends on the voluntary cooperation of states, on the political calculations of great powers, and on the normative pressure of international opinion — all of which are partial, inconsistent, and politically mediated. A realist would say this means international law is not really law. A more defensible position is that international law is law operating under uniquely difficult conditions, and that its effectiveness must be evaluated against those conditions rather than against the standard of well-functioning domestic legal systems.

Viewed this way, the history of international law's revision is a history of progressively sophisticated responses to the core problem: how do you create normative constraints that bind sovereign actors who have agreed to no external authority? The three main mechanisms — treaties, custom, and precedent — each represent a different solution to this problem, with different strengths and failure modes.

The treaty mechanism is international law's most transparent revision tool. States negotiate, sign, and ratify written agreements that create explicit legal obligations. The obligations are binding only on states that have ratified, which means the treaty system is opt-in — states choose which constraints they accept. This is a significant limitation but also a source of the system's durability: states accept obligations they have explicitly chosen, rather than having them imposed.

The historical pattern of treaty development reveals a recurring structure. A catastrophic failure of the existing normative framework occurs. Horrified observers document what happened and what existing law did not prevent. Diplomatic momentum builds for a multilateral response. States negotiate, with each seeking to preserve maximum freedom of action while accepting minimum constraints. The resulting treaty reflects these competing pressures, often being weaker than advocates wanted and stronger than resisters preferred.

The first Geneva Convention (1864) is the archetype. Henri Dunant witnessed approximately 40,000 casualties at Solferino going unattended, dying of treatable wounds because no organized system for treating battlefield wounded existed and no agreement protected medical personnel from attack. He organized the Red Cross movement and lobbied for a multilateral treaty. The 1864 Convention established the first protections for the wounded in land warfare and for medical personnel and facilities. It was a minimal framework — covering only battlefield medicine in land wars — but it established the principle that had never been formally codified: that some constraints on warfare apply even between belligerents.

The subsequent history is a series of expansions and revisions, each triggered by a failure of the prior framework. The 1906 revision added more detail. The 1929 revision added prisoner of war protections, directly responding to the miseries of World War I captivity. The four 1949 Geneva Conventions, negotiated in the immediate aftermath of World War II's atrocities, vastly expanded the framework: civilian protection, protections in occupied territories, and more robust POW rules. The two 1977 Additional Protocols extended coverage to armed conflicts not between states. The 2005 Third Protocol added a new protective emblem. Each round of revision addressed failures that the prior framework had exposed.

This pattern generalizes. The Chemical Weapons Convention (1993) was the revision triggered by the confirmed use of chemical weapons in the Iran-Iraq War. The Ottawa Treaty banning landmines (1997) was the revision triggered by the humanitarian toll of landmine casualties documented across post-conflict regions. The Arms Trade Treaty (2014) was the revision triggered by the role of unregulated arms flows in atrocities across Africa and elsewhere. In each case, documented failure of existing norms created political space for a treaty that would have been blocked in the absence of that failure.

This is revision through catastrophe, which is an inefficient but reliable mechanism. The alternative — proactive international lawmaking that anticipates failures before they occur — exists in embryonic form in environmental law (particularly climate agreements, which attempt to prevent catastrophe rather than respond to it) but is much harder to achieve politically. States are more willing to constrain their behavior after a catastrophe has made the cost of unconstrained behavior visible than before. This limits the revision rate of international law to a pace governed by catastrophe, which is too slow for fast-moving threats.

Customary international law is less well understood than treaty law but may be more durable. A norm becomes customary when it satisfies two criteria: state practice (states actually behave consistently with the norm) and opinio juris (states regard the norm as legally obligatory, not merely habitual or convenient). When both criteria are met, the norm binds all states, including those that have not ratified any treaty codifying it.

The revision mechanism for customary law is the same as the formation mechanism: when practice changes, and when the opinio juris accompanying that change reaches sufficient generality, the custom changes. This is slow. Documenting customary law requires surveying state practice across dozens or hundreds of states over sustained periods. Determining opinio juris — distinguishing between behavior that reflects legal obligation and behavior that reflects policy preference or diplomatic courtesy — is genuinely difficult.

But the slow pace of customary law revision is a feature as well as a limitation. A norm that has become genuinely customary is embedded in the actual behavior of states, not merely in their written commitments. Customary norms are harder to dislodge than treaty norms, because dislodging them requires changing actual state practice across the international community, not simply withdrawing from a treaty. The prohibition on torture, which has both treaty and customary status, is a norm that states cannot simply opt out of — they can violate it, but they cannot publicly repudiate the prohibition without severe reputational consequences, because the prohibition is embedded in the fundamental framework of the international legal order.

The formation of new customary norms is also a revision mechanism. The law of space, the law of cyberspace, and the evolving norms around autonomous weapons systems are all domains in which customary norms are currently forming — through state practice, through diplomatic statements, through bilateral and multilateral negotiations — without the existence of comprehensive treaty frameworks. States are, in effect, writing the first draft of these norms through their behavior, and the opinio juris that accompanies that behavior will determine what constraints future actors face.

The growth of international adjudication over the past century — from the Permanent Court of International Justice in 1920 to the current ecosystem of international courts, tribunals, and arbitral panels — has created a third revision mechanism: the accumulation of judicial decisions that interpret, clarify, and extend international norms.

International judicial precedent is not formally binding in most international legal systems the way common law precedent is in domestic systems. The International Court of Justice explicitly states that its decisions are binding only on the parties to the specific case. But in practice, international courts treat their prior decisions as authoritative guidance, and states, lawyers, and other courts cite ICJ decisions as representing the settled interpretation of international law. The normative weight of precedent is real even without formal stare decisis.

The revision function of international adjudication operates at several levels.

At the level of specific norms, courts interpret treaty provisions in ways that expand or restrict their application. The International Criminal Tribunal for the former Yugoslavia's interpretation of "systematic attack on a civilian population" in its early decisions expanded the definition of crimes against humanity in ways that influenced subsequent ICC prosecutions. The ICJ's interpretation of maritime boundary delimitation principles has created a body of precedent that shapes how states negotiate maritime disputes, even disputes the ICJ itself does not adjudicate.

At the level of customary law, international courts are uniquely positioned to confirm or contest the customary status of particular norms. A court finding that a particular practice constitutes customary international law accelerates the norm's consolidation; a finding that a purportedly customary norm has not achieved that status delays or prevents its consolidation. The ICJ's 1986 Nicaragua decision, which ruled that the US mining of Nicaraguan harbors violated customary international law prohibitions on the use of force, consolidated those prohibitions in ways that extended well beyond the specific facts of the case.

At the level of institutional architecture, adjudication reveals gaps in the existing framework that create pressure for treaty revision. The gaps exposed by the ICTY and ICTR in processing mass atrocity cases — the limitations of ad hoc tribunals created for specific conflicts — contributed directly to the political momentum for the Rome Statute and the creation of the ICC as a permanent institution. The adjudicative experience revealed the inadequacy of the existing revision to atrocity law and helped produce the next revision.

The most significant structural failure of international law as a revision system is its inability to constrain the behavior of powerful states in domains where those states resist constraint. The UN Security Council's veto structure ensures that the five permanent members can prevent enforcement action against themselves or their close allies. The ICC has jurisdiction over states that have ratified the Rome Statute, but the United States, Russia, China, India, and Israel — collectively responsible for a substantial fraction of controversial military conduct — have not ratified. The WTO's dispute settlement mechanism has been significantly impaired by the United States' refusal to approve new Appellate Body members.

This power asymmetry is a structural feature of the system, not an accidental failure. International law emerged from and reflects the distribution of power among states; the states most capable of shaping the international legal order have systematically preserved their freedom of action while using the legal order to constrain weaker rivals. This is not hypocrisy in the simple sense — powerful states do accept constraints, do engage with the legal order in ways that constrain their behavior — but the constraints they accept are systematically less binding than those they help impose on others.

The revision implication is significant: international law's revision mechanism tends to codify the moral learning of the international community in domains where no major power has a strong interest in maintaining the prior norm, and to stall in domains where a major power has such an interest. The codification of prohibitions on anti-personnel landmines advanced rapidly precisely because no major military power was significantly invested in landmine use as a military capability. The codification of prohibitions on nuclear weapons has stalled precisely because the states with the largest nuclear arsenals have a fundamental interest in maintaining them.

This does not mean revision never happens against great power resistance. The long-run trend in international norms constraining the conduct of warfare, the treatment of prisoners, and the protection of civilians has moved consistently toward greater restriction even as great powers have resisted particular provisions. But the pace is slow, the gaps are large, and the accountability for powerful states is substantially lower than for weak ones.

Despite its limitations, international law has produced genuine civilizational revision across several domains.

The law of war. The normative framework governing armed conflict has been substantially revised since the mid-nineteenth century. The laws of war now recognize the protected status of civilians, medical personnel, cultural property, and prisoners of war in ways that were not formalized in prior centuries. States still violate these norms, sometimes systematically. But the norms shape the behavior of military planners, the rhetoric of political leaders justifying military action, and the accountability of individuals who violate them in captured contexts. The existence of the norm matters even when enforcement is imperfect.

The prohibition on aggressive war. The Kellogg-Briand Pact of 1928 and the UN Charter's prohibition on the use of force against the territorial integrity of other states represent a fundamental revision of the international system's normative framework. Prior to this, conquest was a recognized legal procedure — states could and did legally acquire territory through military victory. The revision of this norm has not eliminated military aggression, but it has eliminated the legal legitimacy of conquest, which changes the political economy of aggression in ways that matter. Occupying powers now universally disclaim conquest as their objective, which constrains their behavior in occupied territories and makes prolonged occupation politically more costly.

The international human rights regime. The post-World War II human rights treaty system — beginning with the Universal Declaration of Human Rights and extending through the two International Covenants, the Convention on Torture, the Convention on the Rights of the Child, and dozens of specialized instruments — has created an international normative framework for the relationship between states and their populations that simply did not exist before 1945. Implementation is inconsistent and often nominal. But the framework creates accountability mechanisms, empowers civil society, and provides reference points for domestic political contestation that have measurable effects on state behavior.

International criminal accountability. The Nuremberg and Tokyo tribunals, followed by the ICTY, ICTR, and ICC, have created the principle — still contested, still inconsistently applied — that individuals are accountable under international law for crimes against humanity, war crimes, and genocide. The application of this principle to specific individuals has been limited and politically mediated. But the principle itself represents a revision of the prior framework, in which states bore international legal responsibility but their leaders did not.

These are not small achievements. They are the product of two centuries of iterative revision of an international system that had no formal mechanism for self-correction. The mechanism is slow, it is biased toward powerful states, and it fails catastrophically in high-stakes cases. But the long run trend is toward a normative framework that is more constraining, more universal, and more enforced than what preceded it — and that is what civilizational revision, operating at its characteristic pace, looks like.