International law has no Parliament and nothing that can really be described as legislation. While there is an International Court of Justice and a range of specialised international courts and tribunals, their jurisdiction is critically dependent upon the consent of States and they lack what can properly be described as a compulsory jurisdiction of the kind possessed by national courts.
The result is that international law is made largely on a decentralised basis by the actions of the 192 States which make up the international community. The Statute of the ICJ, Art. 38 identifies five sources:-
. (a) Treaties between States;
. (b) Customary international law derived from the practice of States;
. (c) General principles of law recognized by civilised nations; and, as subsidiary means for the determination of rules of international law:
. (d) Judicial decisions and the writings of “the most highly qualified publicists”.
It is convenient to start with customary law as this is both the oldest source and the one which generates rules binding on all States.
Customary law is not a written source. A rule of customary law, e.g., requiring States to grant immunity to a visiting Head of State, is said to have two elements. First, there must be widespread and consistent State practice – ie States must, in general, have a practice of according immunity to a visiting Head of State. Secondly, there has to be what is called “opinio juris”, usually translated as “a belief in legal obligation; ie States must accord immunity because they believe they have a legal duty to do so. As the ICJ has put it:-
‘Not only must the acts concerned be a settled practice, but they must also be such, or be carried out in such a way, as to be evidence of a belief that this practice is rendered obligatory by the existence of a rule requiring it. … The States concerned must feel that they are conforming to what amounts to a legal obligation.’ (North Sea Continental Shelf cases, ICJ Reps, 1969, p. 3 at 44)
A new rule of customary international law cannot be created unless both of these elements are present. Practice alone is not enough – see, e.g., the Case of the SS Lotus (1927). Nor can a rule be created by opinio juris without actual practice – see, e.g., the Advisory Opinion on Nuclear Weapons (1996).
But these elements require closer examination. So far as practice is concerned, this includes not just the practice of the government of a State but also of its courts and parliament. It includes what States say as well as what they do. Also practice needs to be carefully examined for what it actually says about law. The fact that some (perhaps many) States practise torture does not mean that there is not a sufficient practice outlawing it. To quote from the ICJ’s decision in the Nicaragua case:
‘In order to deduce the existence of customary rules, the Court deems it sufficient that the conduct of States should in general be consistent with such a rule; and that instances of State conduct inconsistent with a given rule should generally have been treated as breaches of that rule, not as indications of the recognition of a new rule.’ (ICJ in Nicaragua ICJ Reps, 1986, p. 3 at 98.)
Regarding opinio juris, the normal definition of a belief in obligation (see, e.g., the North Sea Continental Shelf cases (1969) above) is not entirely satisfactory.
First, it ignores the fact that many rules are permissive (eg regarding sovereignty over the continental shelf), for which the real opinio juris is a belief not in obligation but in right. Secondly, and more fundamentally, there is something artificial in talking of the beliefs of a State. It might be better to consider opinio juris as the assertion of a legal right or the acknowledgment of a legal obligation.
Once there is sufficient practice together with opinio juris, a new rule of custom will emerge. Subject only to what is known as the “persistent objector” principle the new rule binds all States. The persistent objector principle allows a State which has persistently rejected a new rule even before it emerged as such to avoid its application.
Treaties (sometimes called agreements, conventions, exchanges of notes or protocols) between States – or sometimes between States and international organizations – are the other main source of law.
Strictly speaking a treaty is not a source of law so much as a source of obligation under law. Treaties are binding only on States which become parties to them and the choice of whether or not to become party to a treaty is entirely one for the State – there is no requirement to sign up to a treaty. Why is a treaty binding on those States which have become parties to it ? The answer is that there is a rule of customary international law – pacta sunt servanda – which requires all States to honour their treaties. That is why treaties are more accurately described as sources of obligation under law.
But many treaties are also important as authoritative statements of customary law. A treaty which is freely negotiated between a large number of States is often regarded as writing down what were previously unwritten rules of customary law. That is obviously the case where a treaty provision is intended to be codificatory of the existing law. A good example is the Vienna Convention on the Law of Treaties, 1969. Less than half the States in the world are parties to it but every court which has considered the matter has treated its main provisions as codifying customary law and has therefore treated them as applying to all States whether they are parties to the Convention or not.
In theory, where a treaty provision codifies a rule of customary law the source of law is the original practice and opinio juris – the treaty provision is merely evidence. But that overlooks the fact that writing down a rule which was previously unwritten changes that rule. From that time on, it is the written provision to which everyone will look and debates about the extent of the rule will largely revolve around the interpretation of the text rather than an analysis of the underlying practice.
Moreover, even where a treaty provision is not intended to be codificatory but rather is an innovation designed to change the rule, it can become part of customary law if it is accepted in practice. See, e.g., the North Sea Continental Shelf cases (1969):
‘Although the passage of only a short period of time is not necessarily, or of itself, a bar to the formation of a new rule of customary international law on the basis of what was originally a purely conventional rule, an indispensable requirement would be that within the period in question, short though it might be, State practice, including that of States whose interests are specially affected, should have been both extensive and virtually uniform in the sense of the provision invoked; – and should moreover have occurred in such a way as to show a general recognition that a rule of law or legal obligation is involved.’ (ICJ Reps, 1969, p. 43)
In reality the fact of a large number of States agreeing upon a treaty provision is itself an important piece of State practice. If those and other States subsequently apply the treaty provision – especially where they are not parties to the treaty – then it can quickly become part of customary international law.
This consideration has led some writers to distinguish between “traités contrats” (contractual treaties) which are only agreements between the parties and traités lois (law-making treaties). In my view this confuses rather than assists. All treaties are contractual as between their parties. But some also have an effect on the general law.
In practice, it has been through the adoption of numerous treaties on different areas of international law (war, terrorism, diplomacy, treaty-making) that international law has undergone its most important changes in the years since 1945.
3. General Principles
While treaties and custom are the most important sources of international law, the others mentioned in Article 38 of the ICJ Statute of the ICJ should not be ignored. General principles of law recognized by civilised nations – the third source – are seldom mentioned in judgments. They are most often employed where the ICJ or another international tribunal wants to adopt a concept such as the legal personality of corporations (eg in the Barcelona Traction Co. case (1970)) which is widely accepted in national legal systems. But international law seldom adopts in its entirety a legal concept from a particular national legal system; instead the search is for a principle which in one form or another is recognized in a wide range of national legal systems.
4. Judicial Decisions
Article 38(1)(d) refers to judicial decisions as a subsidiary means for the determination of rules of law. In contrast to the position in common law countries, there is no doctrine of binding precedent in international law. Indeed, the Statute of the ICJ expressly provides that a decision of the Court is not binding on anyone except the partiers to the case in which that decision is given and even then only in respect of that particular case (Article 59). Nevertheless, the ICJ refers frequently to its own past decisions and most international tribunals make use of past cases as a guide to the content of international law, so it would be a mistake to assume that “subsidiary” indicated a lack of importance.
Article 38(1)(d) does not distinguish between decisions of international and national courts. The former are generally considered the more authoritative evidence of international law on most topics (though not those which are more commonly handled by national courts, such as the law on sovereign immunity). But decisions of a State’s courts are a part of the practice of that State and can therefore contribute directly to the formation of customary international law.
The writings of international lawyers may also be a persuasive guide to the content of international law but they are not themselves creative of law and there is a danger in taking an isolated passage from a book or article and assuming without more that it accurately reflects the content of international law.
6. Other Sources
The list of sources in Article 38 of the Statute is frequently criticised for being incomplete. In particular, it makes no mention of the acts of the different organs of the United Nations. Today there can be no doubting the importance of those acts in shaping international law, although they perhaps fit within the system of Article 38 better than is sometimes imagined.
The United Nations General Assembly has no power to legislate for the international community; its resolutions are not legally binding. However, many of those resolutions have an important effect on the law-making process. Some resolutions are part of the treaty- making process, attaching a treaty text negotiated in the framework of the United Nations and recommended to the Member States by the Assembly (this was the case with the Convention against Torture). While it is the treaty which creates the legal obligation – and then only for the States which choose to become party to it – the importance of the United Nations in the process of creating that treaty should not be underestimated.
In addition, as I have already mentioned, the positions which States take in the United Nations is part of their practice and a resolution (or sequence of resolutions) which commands a sufficiently widespread acceptance and which is regarded by the States as embodying a rule of international law can have an important effect on the development of customary international law, so long as it is not contradicted by what States actually do elsewhere (see, e.g., the discussion of the resolutions on nuclear weapons in the Advisory Opinion on Nuclear Weapons (1996)).
The studies of international law produced by the International Law Commission for the General Assembly, especially if adopted by the Assembly, may also have an important effect on customary international law, even if they are not turned into treaties (the ILC Articles on State Responsibility adopted in 2001 are a good example).
The position of the Security Council is somewhat different. Decisions taken by the Council under Chapter VII of the Charter and framed in mandatory terms are legally binding on all States (Article 25 of the Charter). Moreover, under Article 103 of the Charter the duty to carry out a decision of the Council prevails over obligations under all other international agreements (see the Lockerbie cases (1992)). However, the Council does not create new laws but rather obligations in relation to specific issues and it is not a legislature (see the decision of the ICTY in Tadic (1995)).