Custom
Chapter Ten
Syllabus topic 1.5, "Sources of International Law: Custom"
Pages 67 to 74 of 612
In one line
A customary rule is one that States generally follow because they believe the law requires it, and both halves of that sentence have to be proved.
In the wording a student can write in an exam: international custom, under article 38(1)(b) of the Statute of the International Court of Justice, is a general practice accepted as law. It has two elements: a material element, the general and consistent practice of States, and a psychological element, opinio juris sive necessitatis, the belief that the practice is required by law. Practice without that belief is usage or comity and creates no obligation.
Why custom matters more than it looks
A student meeting a subject full of long conventions may wonder why an unwritten source still matters. Three reasons.
It binds States that have signed nothing. A treaty binds its parties. A customary rule binds every State, including those that were not consulted and those that came into existence afterwards.
It is what fills the gaps. No treaty covers every question, and custom is what a tribunal reaches for when the treaty runs out.
It is what makes codification possible. Every great convention in this book, on treaties, on the sea, on diplomatic relations, was largely a writing down of custom, and the rules in them bind non-parties to the extent that the custom survives alongside the treaty.
The two elements
The material element: general practice
What counts as practice. State practice is not only what States do physically. It includes diplomatic correspondence, statements by government officials, national legislation, decisions of national courts, votes and explanations of vote in international organisations, military manuals, and the conclusion of treaties. Anything a State does or says in its official capacity can be practice.
Whose practice. All States count, but not equally. The practice of States particularly affected by a rule carries more weight, which is why maritime States matter most for the law of the sea and space-faring States for space law.
How general. Not universal. The requirement is generality, not unanimity. What is required is that the practice be extensive and virtually uniform, including the practice of States whose interests are specially affected.
How consistent. Substantially consistent. Minor departures do not destroy a rule; they are treated as breaches of it. That is a point students get wrong, and the reasoning is important: if a State that departs from a rule defends itself by appealing to exceptions or justifications recognised within the rule, its conduct confirms rather than weakens the rule.
How long. There is no fixed period. The North Sea judgment says the passage of only a short period is not necessarily a bar, provided that within that period the practice has been extensive and virtually uniform. The idea of instant custom is associated with space law, where the practice of a handful of States over a few years produced the rule that outer space is not subject to appropriation.
Custom
The psychological element: opinio juris
Opinio juris sive necessitatis means the belief that the conduct is required by law or by a sense of legal obligation.
It is what separates a customary rule from a courtesy. States have for centuries given a departing ambassador a ceremonial send-off; nobody thinks a legal obligation exists, so no custom has formed. States have for centuries refrained from arresting foreign diplomats; that is a rule, because States believe they are bound.
How it is proved. Opinio juris is a state of mind, and States do not have minds. In practice it is inferred from what States say when they act: from protests, from claims, from the terms of national legislation, from the reasons given for votes, and from the arguments States make when accused of a breach.
The paradox. For a new custom to form, some State must act first, and at the moment it acts there is no rule, so its belief that the law requires the act is false. The usual answer is that the belief may be prospective: the State asserts a rule and other States acquiesce, and the rule crystallises around the assertion. That is exactly what happened with the continental shelf, and the Truman Proclamation of 28 September 1945 is the standard example.
The leading case on how a treaty rule becomes custom
Facts. North Sea Continental Shelf Cases (Federal Republic of Germany/Denmark; Federal Republic of Germany/Netherlands), ICJ Reports 1969, page 3, decided 20 February 1969. Germany, Denmark and the Netherlands had agreed partial continental shelf boundaries near the coast in 1964 and 1965 but could not agree on their prolongation. Denmark and the Netherlands wanted the equidistance method of article 6 of the 1958 Geneva Convention on the Continental Shelf. Germany's North Sea coast is concave, and equidistance lines drawn from a recessing coast meet a short way out and cut the State behind them off from the shelf beyond. Germany had signed the 1958 Convention but never ratified it. By two special agreements of 20 February 1967 the parties asked the Court to state the applicable principles.
Held. By eleven votes to six, equidistance was not obligatory. Germany was not bound by article 6, having never ratified, and no estoppel arose against it. Equidistance was not inherent in the concept of the continental shelf, whose basis is the natural prolongation of the land territory, giving rights that exist ipso facto and ab initio. And article 6 had not passed into customary law. For that to happen the provision would have to be, at all events potentially, of a norm-creating character, and article 6 was not: it puts the obligation to use equidistance after a primary obligation to delimit by agreement, it is qualified by the notion of special circumstances whose meaning was controverted, and reservations to it were permitted, which the Court took as a legitimate inference that it was not thought to reflect emergent customary law. Further, a very widespread and representative participation in a convention might show that a conventional rule had become general law, but the number of ratifications and accessions was hardly sufficient. As to time, the passage of only a short period was not necessarily a bar, but within the period, however short, State practice, including that of States whose interests are specially affected, should have been both extensive and virtually uniform, and should have occurred in such a way as to show a general recognition that a rule of law was involved.
Custom
Why it matters. This is the leading statement of how a treaty rule becomes custom, and the three conditions it sets, norm-creating character, widespread and representative participation including specially affected States, and extensive and virtually uniform practice accompanied by opinio juris, are the answer to the standard essay question.
How a court finds a rule of custom: The Paquete Habana
The Paquete Habana; The Lola, 175 U.S. 677 (1900), Supreme Court of the United States, 5 January 1900.
Facts. Two fishing smacks, the sloop Paquete Habana and the schooner Lola, were regularly engaged in fishing on the coast of Cuba, sailing under the Spanish flag, each owned by a Spanish subject residing in Havana, with crews who also resided there and took shares of the catch. Each left Havana on a coast fishing voyage and was returning with a cargo of live fish when she was captured near Havana by a vessel of the United States blockading squadron during the war with Spain. Neither carried arms or ammunition; neither knew of the blockade or even of the war until stopped; neither ran the blockade or resisted; and there was no evidence that either vessel or crew was likely to aid the enemy. The District Court condemned both vessels and their cargoes as prize of war and they were sold.
Held, reversing the decrees, that the captures were unlawful and without probable cause, the proceeds of the sales to be restored to the claimants with damages and costs. Fuller CJ, with Harlan and McKenna JJ, dissented.
The method, which is why the case is in this chapter. International law is part of our law, and must be ascertained and administered by the courts of justice of appropriate jurisdiction, as often as questions of right depending upon it are duly presented for their determination. For that purpose, where there is no treaty, and no controlling executive or legislative act or judicial decision, resort must be had to the customs and usages of civilized nations; and, as evidence of these, to the works of jurists and commentators, not for the speculations of their authors concerning what the law ought to be, but for trustworthy evidence of what the law really is.
Custom
The rule found. Applying that method to several centuries of orders, treaties and practice, the Court held that at the present day, by the general consent of the civilized nations of the world, and independently of any express treaty or other public act, it is an established rule of international law that coast fishing vessels, with their implements and supplies, cargoes and crews, unarmed, and honestly pursuing their peaceful calling of catching and bringing in fresh fish, are exempt from capture as prize of war, and that prize courts administering the law of nations are bound to take judicial notice of that rule and to give effect to it in the absence of any treaty or other public act of their own government.
Three things to take from it.
It shows the two elements being proved. Centuries of practice, and orders and treaties showing that the practice was followed as a matter of obligation rather than of grace, which is opinio juris in the material a court can actually examine.
It fixes the place of the writings of publicists, and it is a better statement of article 38(1)(d) than the article itself: they are evidence of what the law is, not authority for what it should be. [Judicial Decisions and the Writings of Jurists] works that point.
It is the standard illustration of a municipal court applying custom directly, and it belongs with [International Law and Municipal Law] and with the Indian position in [International Law in Indian Courts], where the corresponding proposition comes from Gramophone Company of India Ltd v. Birendra Bahadur Pandey.
Local and regional custom
Custom need not be universal. A rule can bind a group of States, or even two.
Facts. Asylum Case (Colombia v. Peru), ICJ Reports 1950, page 266, decided 20 November 1950. On 3 October 1948 a military rebellion broke out in Peru and was suppressed the same day. Victor Raul Haya de la Torre, head of the American People's Revolutionary Alliance, was prosecuted for military rebellion. On 3 January 1949 the Colombian Ambassador in Lima granted him asylum, qualified him as a political offender and asked Peru for a safe-conduct. Peru refused, saying he had committed common crimes. The two States submitted the dispute to the Court.
Custom
Held. By fourteen votes to two, Colombia was not entitled to qualify unilaterally, and in a manner binding on Peru, the nature of the offence. The Court rejected each treaty Colombia relied on: the Bolivarian Agreement of 1911 recognised asylum in accordance with the principles of international law, and those principles do not entail a right of unilateral qualification; the Havana Convention of 1928 conferred no such right expressly or by implication; and the Montevideo Convention of 1933 had not been ratified by Peru. As to a regional American custom, Colombia had not proved the existence, regionally or locally, of a constant and uniform practice of unilateral qualification as a right of the State of refuge and an obligation on the territorial State, because the facts disclosed too much contradiction and fluctuation to disclose a usage peculiar to Latin America and accepted as law. By fifteen votes to one the Court also held Peru was not bound to grant a safe-conduct, and by fifteen to one rejected Peru's contention that Haya de la Torre was accused of common crimes, military rebellion not being in itself a common crime.
Why it matters. Two things. It establishes that a regional or local custom can exist but must be proved by the party asserting it, and to a strict standard: a constant and uniform practice accepted as law. And it is the leading case on diplomatic asylum, worked for that purpose in [Asylum].
The persistent objector
A State that objects to a rule from the beginning, and keeps objecting, is not bound by it when it crystallises.
The authority is the Fisheries case, worked in [Baselines and the Territorial Sea], where the Court held that the ten-mile rule for closing lines of bays had not acquired the authority of a general rule of international law and, in any event, was inapplicable as against Norway inasmuch as she had always opposed its application to the Norwegian coast.
Three limits on the doctrine are worth writing down.
The objection must be from the start. A State that acquiesces while a rule forms and objects afterwards is bound. Silence is treated as acquiescence.
It must be maintained. An objection abandoned is an objection lost.
It cannot work against jus cogens. No State can persistently object its way out of the prohibition of genocide.
Distinctions that carry marks
| Custom | Usage or comity | Treaty | |
|---|---|---|---|
| Practice | Yes | Yes | Not required |
| Opinio juris | Yes | No | Not required; consent instead |
| Binds | All States, subject to persistent objection | Nobody | The parties |
| Breach gives | A claim of illegality | Complaint only | A claim of illegality |
| Proof | The party asserting it must prove practice and opinio juris | Not applicable | Produce the treaty |
Custom
A worked example
For forty years the States bordering the Gulf of Aral have permitted each other's fishing vessels to shelter in their ports during storms without formality or charge. State E now begins charging State F's vessels a fee. State F protests that a customary rule has been broken.
Is there practice? Yes: forty years, consistent, and by all the States concerned. That is the easy half.
Is there opinio juris? That is the question, and State F must prove it. Evidence would include any statement by a government that the shelter was granted as of right, any legislation framing it as an obligation, any earlier occasion on which a State tried to charge and was told it could not, and any diplomatic protest. Evidence against would include statements describing the practice as a courtesy, or occasions on which shelter was refused without any complaint being made.
Whose custom would it be? Regional or local, binding only the Gulf States. On the Asylum reasoning State F must prove a constant and uniform practice accepted as law, and contradiction and fluctuation in the record will defeat it.
Could State E escape as a persistent objector? Only if it had objected as the rule formed and kept objecting. Forty years of granting shelter without protest is acquiescence, not objection.
What if there were a treaty? Then this whole analysis would be unnecessary, which is why a tribunal looks for a treaty first.
What it does NOT mean
Custom is not old practice. Age helps, but the North Sea judgment says a short period is not necessarily a bar. What matters is that the practice was extensive and virtually uniform and accompanied by the belief that law required it.
Custom is not unanimity. Generality is required, not universal participation. A State that says nothing is treated as acquiescing.
Breach of a customary rule does not destroy it. If the State in breach justifies itself by exceptions recognised within the rule, the rule is confirmed. It is only if States generally claim a right to act inconsistently that the rule erodes.
A treaty codifying custom does not abolish the custom. The two run in parallel, which is why non-parties remain bound by the customary rule and why a party that withdraws from the treaty remains bound by it.
Limits and criticism
Proving opinio juris is close to impossible in a hard case. Tribunals often infer it from the practice itself, which makes the two elements collapse into one and is circular. This is the most common criticism of the doctrine and it is a fair one.
Custom
It advantages the powerful. The practice of specially affected States counts for more, and States with navies, satellites and global reach are specially affected more often than others. Newly independent States made exactly this objection in the 1960s and 1970s, and it is why they preferred codification by treaty, where every State has one vote.
The relationship with treaties is now tangled. Conventions codify custom, parties then develop the custom through their practice under the convention, and non-parties are affected by rules they never joined. The North Sea judgment tried to keep the categories apart and the practice has run ahead of it.
Quick revision
Article 38(1)(b): international custom, as evidence of a general practice accepted as law. Two elements: general practice, and opinio juris sive necessitatis.
Practice includes acts, statements, legislation, national judgments, votes and treaties; it must be extensive and virtually uniform, including the practice of specially affected States; generality, not unanimity; no fixed duration.
Opinio juris is inferred from what States say when they act, and it is what separates custom from usage and comity.
North Sea Continental Shelf (ICJ Reports 1969, page 3): a treaty rule becomes custom only if the provision is of a norm-creating character, there is widespread and representative participation including specially affected States, and practice is extensive and virtually uniform with opinio juris. Article 6 of the 1958 Convention failed all three, partly because reservations to it were permitted.
Asylum (ICJ Reports 1950, page 266): local and regional custom exists but must be proved by the asserting party as a constant and uniform practice accepted as law; contradiction and fluctuation defeat it.
Persistent objector: Fisheries (ICJ Reports 1951, page 116). Norway had always opposed the ten-mile rule, so it was inapplicable to her. The objection must be from the start, maintained, and it cannot defeat jus cogens.
Test yourself
1. State the two elements of custom and explain why the second is needed.
The material element is a general and consistent practice of States; the psychological element is opinio juris sive necessitatis, the belief that the practice is required by law. The second is needed because practice alone cannot distinguish a legal rule from a habit. States have always shown courtesies to one another without regarding themselves as bound, and if practice alone made law every such courtesy would become an obligation and no State could ever stop.
2. What did the North Sea Continental Shelf cases decide about the passage of a treaty rule into custom?
That it can happen but requires three things. The provision must be, at all events potentially, of a norm-creating character, which article 6 was not, since it subordinated equidistance to prior agreement, was qualified by a controverted notion of special circumstances, and was open to reservation. There must be very widespread and representative participation in the convention, and the ratifications and accessions there were hardly sufficient. And State practice, including that of specially affected States, must have been extensive and virtually uniform, and must have occurred in a way showing general recognition that a rule of law was involved. A short passage of time is not by itself a bar.
Custom
3. Can a custom bind only two States? Give authority.
Yes. The Asylum case accepted in principle that a custom could be regional or even local, but held that Colombia had failed to prove one: she had not established a constant and uniform practice of unilateral qualification as a right of the State of refuge and an obligation upon the territorial State, because the facts disclosed too much contradiction and fluctuation to reveal a usage peculiar to Latin America and accepted as law. The burden lies on the State asserting the special custom and the standard is strict.
4. Explain the persistent objector rule and its three limits.
A State that has objected to a rule consistently from the time of its formation is not bound by it when it crystallises; the Fisheries case is the authority, the Court holding the ten-mile bay rule inapplicable to Norway because she had always opposed its application to her coast. The limits are that the objection must begin before the rule crystallises, since silence during formation counts as acquiescence; it must be maintained, since an abandoned objection is lost; and it cannot be raised against a peremptory norm, from which no derogation is permitted.
5. "Breach of a customary rule proves the rule does not exist." Is that right?
No, and the reasoning matters. If a State that departs from a rule seeks to justify itself by appealing to exceptions or justifications contained within the rule, its conduct confirms the rule rather than weakening it, because the State is not claiming a right to act inconsistently but arguing that its act falls within the rule. The rule erodes only where States generally claim a right to depart from it, and inconsistent conduct is then met without protest.
6. What is instant custom, and where is it said to have occurred?
The proposition that a customary rule can form very quickly where practice is extensive and virtually uniform and is accompanied by clear opinio juris, notwithstanding that little time has passed. The North Sea judgment leaves it open by saying that a short period is not necessarily a bar. It is usually illustrated by space law: within a few years of 1957 the practice of a very small number of space-faring States, with the acquiescence of everyone else, produced the rules that outer space is free for exploration by all and is not subject to national appropriation, which were then written into the Outer Space Treaty and are dealt with in [International Space Law].
The rest of this subject
These notes are cut from the University's printed syllabus. Open the syllabus itself, or the past papers, for the same subject.