As Prof. Clive Parry asserts, the ascertainment of the law on any given point in domestic legal orders is not usually too difficult a process. The reason for this assertion is probably that, in every developed legal system, there are laws which stipulate in express terms what the sources of law are. For instance, in Ghana, article 11 of the 1992 Constitution spells out in unequivocal terms what the sources of Ghanaian Law are. In the case of the international system, the contrary situation prevails. There is no world constitution to clearly spell out what the sources of International Law are, although attempts have been made by some authors to consider the United Nations Charter as the ‘Constitution of the World’. This argument is not without flaws. However, that is not the focus of this article, and as such not much will be said of it. Malcolm Shaw notes that the lack of a legislature, executive and structure of courts within international law has diverse obvious ramifications. There is no single body able to create laws internationally binding upon everyone, nor a proper system of courts with comprehensive and compulsory jurisdiction to interpret the law as we find in every developed legal system. A similar view is expressed by Mario Prost who posits that international law constitutes a distinctly anarchical order, not so much because it is chaotic and disorderly, but because it lacks a centralized and hierarchically structured law-making and law enforcing authority. This poses a significant challenge in discerning where law is to be found and how one can tell whether a particular proposition amounts to law.

Nonetheless, if we accept that International Law is truly law, then it must have some sources as it is the case with domestic law. Sir Robert Jennings, a former Judge of the International Court of Justice once stated that although lawyers know that the quality of certainty of law is one on which there must be much compromise, not least in the interests of justice, it is a desideratum of every strong law that there is reasonable certainty about where one should look to find it. The fundamental question therefore is, what do we mean by ‘sources of international law’? There are several controversies regarding what the term ‘sources’ as used in international law is. However, as Emmanuel Yaw Benneh puts it, it has generally been used in reference to the methods of law creation in the international front.

In our enquiry to find answers to the question of what the sources of international law are, we come close to a document which has been authoritatively cited as giving a clear exposition on what the sources of international law are. This document is the Statute of the International Court of Justice (ICJ), article 38(1) of which provides as follows:

“The Court, whose function is to decide in accordance with international law such disputes as are submitted to it, shall apply:

(a) international conventions, whether general or particular, establishing rules expressly recognized by the contesting states;

(b) international custom, as evidence of a general practice accepted as law;

(c) the general principles of law recognized by civilized nations;

(d) subject to the provisions of Article 59, judicial decisions and the teachings of the most highly qualified publicists of the various nations, as subsidiary means for the determination of rules of law”.

Malcolm Shaw maintains that since all member states of the United Nations are ipso facto parties to the Statute by virtue of article 93 of the United Nations Charter, there is no serious contention that the provision expresses the universal perception as to the enumeration of sources of international law.

Within the changing structure and character of international law, a very perplexing question arises whether article 38 of the Statute of the International Court of Justice is exhaustive of all the sources of international law. The widely accepted view has always been that it is not, as contemporary international law admits of other sources not contemplated under article 38. These include; jus cogens, acts and decisions of international organizations such as the United Nations, unilateral declarations of States and draft articles of the International Law Commission.

The focus of this paper is to ascertain whether there is any hierarchical categorization between the sources of international law as enunciated in article 38 of the Statute of the ICJ. In fact, the travaux preparatoires (preparatory works) of the Statute reveals that an attempt to include the words ‘in the undermentioned order’ in Article 38 was rejected by the committee so that the Article in its final form included no express reference to hierarchy.

That notwithstanding, the question of hierarchy has resulted in a great legal conflict. As regards judicial decisions and writings of the most highly qualified publicists, a strong argument has been proffered that these sources only play a subordinate role and there is no contention that they are inferior to the other sources. This is because of their description in article 38 as subsidiary means for the determination of rules of law. General principles of law is also viewed by many international law scholars as complementing treaties and customs. The greatest question of priority lies between treaties and custom.

For Christopher Greenwood, there is no strict sense of hierarchy between treaty and customary law. This assertion finds support in the opinion of Phillippe Blancher who posits that an international treaty does not have a priori a superior value over the one of a unilateral document or over a customary rule. Other authors have expressed the view that conventional accounts of international law-making depict an eclectic and uncoordinated system in which can be engaged simultaneously or in competition with one another, no process been intrinsically or normatively superior to the other. Based on this view, James Crawford is quick to conclude that the concept of a formal, a priori hierarchy of sources is alien to the structure of international law.

Contrary to these points of view is the position taken by Mario Prost that the absence of rigid and formal hierarchies in the doctrine of sources should not serve to conceal the fact that States, adjudicators and legal scholars have, historically, expressed clear preferences for particular sources, and have thus established informal hierarchies, if not of validity, at the very least of importance or pre-eminence among law-making processes. This, he asserts, is not entirely alien to what Prof. Clive Parry refers to as the “logical ordering of scales” i.e. the logical ordering of sources according to specific value judgments about their respective merits.

The legal conflict as to which ranks superior, treaty law or customary law, has given rise to two major schools of international jurisprudence: the treaty primacy school of thought and the custom primacy school of legal reasoning. We shall now revert to consider these views seriatim.




The leading view has been that treaties are the “most prominent”, “most important”, “most fundamental”, “dominant”, “major”, “principal” or “primary” source of international law. Some proponents of this view, such as Patrick Dailler, Mathias Forteau and Alain Pellet, have even gone as far as categorizing the sources of international law into mainly two: treaty law and non-treaty law. There are mainly two strands of the treaty-primacy thesis.

The first stream posits that treaties and custom are normatively equivalent but that, as a matter of procedural order, treaties take priority over other sources of international law. When deciding a case, and so the argument continues, courts and tribunals do routinely – and should as a matter of principle – look at treaties first, before considering non-treaty sources. Treaties are generally thought to be superior instruments for resolving disputes owing to their written character, which confers a greater degree of precision and textual determinacy to treaty norms. Treaty norms are easier to locate, ascertain and apply than other norms, particularly customary norms, the precise content of which can be difficult and onerous to establish. As noted by Jason Beckett, “state practice is widely dispersed, often awkward to identify, hard to weigh, and generally not uniform …it is easier to consult a written source”. This is a sound statement of law, considering the fact that what constitutes the opinio juris sive necessitatis, a pre-requisite for the establishment of the existence of customary law, is cumbersome to prove. The resultant pattern of judicial decisions, notably the Lotus case, the Nicaragua case and the North Sea Continental Shelf Cases, reveals that the court has adopted and maintained a high threshold as regards the overt proving of the subjective constituent of customary law formation. For this reason, Hilary Charlesworth considers “ease of identification” as the determinant of the pre-eminent source of law.

Then again, Hersch Lauterpacht expressly supports the view that treaties should enjoy operational priority by opining that the rights and duties of states “are determined, in the first instance, by their agreement as expressed in treaties, just as in the case of individuals their rights are specifically determined by any contract which is binding upon them. When a controversy arises between two or more States with regard to a matter regulated by a treaty, it is natural that the parties should invoke and that the adjudicating agency should apply, in the first instance, the provisions of the treaty in question …. In the above sense, treaties must be considered as ranking first in the hierarchical order of the sources of international law.”

In the North Sea Continental Shelf Cases, the ICJ appeared to have remarked that “rules of [general] international law can, by agreement, be derogated from in particular cases or as between particular parties” and more specifically in the Nicaragua case that “in general, treaty rules being lex specialis, it would not be appropriate that a State should bring a claim based on customary-law rule if it has by treaty already provided means for settlement of such a claim.” It is not uncommon for proponents of the procedural pre-eminence of treaties to cite these decisions in support of their opinions.

The other strand of the treaty primacy thesis posits that treaty law is not just operationally, but normatively superior to other law-making processes. Though legal scholars have expressed a wide range of views in this regard, treaty law is generally thought to possess three essential qualities that set it apart from other sources: ontological determinacy, practical versatility, and process legitimacy as identified by Mario Prost. Ontological determinacy is used in reference to the fact that the nature of treaties as a source of international law is “unambiguous and uncontroversial.” Thus, unlike customary law, the ascertainment of the constituent elements of which poses a significant challenge, the text and context of a treaty is not so difficult to identify. This is because of the definition of a treaty in article 2(1) of the Vienna Convention on the Law of Treaties. It provides: “treaty” means an international agreement concluded between States in written form and governed by international law, whether embodied in a single instrument or in two or more related instruments and whatever its particular designation. In effect, at all material times, treaties must be written. As such, challenging the content of a treaty is unlikely to arise. The practical versatility of treaties is also predicated on the premise that treaties can be created for a variety of purposes and in variety of contexts and can be used to codify or restate pre-existing customary law, or to make a fresh start and create new rules almost instantly. Lastly, process legitimacy is used with the assumption that treaty making is premised on the principle of freedom of contract. States are free to sign up to a treaty or to choose not to become a party thereto. They enjoy full freedom as regards the modalities and forms of agreements. They are free to enter reservations, limiting or modifying the effect of the treaty in its application to them, as guaranteed by section 2 of the Vienna Convention on the Law of Treaties. Thus, treaty making, the argument goes, takes affirmative cognizance of the concept of contractual autonomy of States. More so, treaties are regarded as more democratic in their outlook and process of formation. As such, Thomas Franck posits that treaty making is comparatively more transparent and democratic than other law-making processes, especially the nebulous process of customary law formation. This is said to increase the effectiveness of international law, as norms generated through legitimate processes and are thought to exert greater “compliance pull” and thus harder to disobey.




Almost all contemporary discussions on the hierarchy of sources of international law postulate that treaty law is of more contemporary relevance than custom. That notwithstanding, some writers still maintain that customary law ranks higher than treaty law.

Practitioners in this school of thought state categorically that custom is superior to other sources, including treaty law, insofar as it precedes and pre-determines them. For them, a treaty properly so called cannot exist without a pre-existing framework of customary law governing its formation. For instance, writers like Lassa Oppenheim notoriously opine that custom “is the original source of international law.” As Mario Prost exemplifies, by this Oppenheim did not mean to say that from a chronological standpoint custom comes first and treaties second, but that custom was not dependent on any other source to exist, unlike treaties which draw their lives from customs. This can be gleaned from Oppenheim’s categorical remarks that, “treaties are a source the power of which derives from custom. For the fact that treaties can stipulate rules of international conduct at all is based on the customary rule of the Law of Nations that treaties are binding upon the contracting parties.”

Hans Kelson, in his treatise General Theory of Law and State, observed that “if we ask why a treaty is valid, we are led to the general norm which obligates the States to behave in conformity with the treaties they have concluded, a norm commonly expressed by the phrase pacta sunt servanda. This is a norm of general international law, and general international law is created by custom …. Customary international law … is the first stage within the international legal order.”

Paul Reuter, a leading scholar of the law of treaties and a firm believer in the “central position” of treaty law in international life, conceded that “treaties are binding by virtue not of a treaty but of customary rules. In that sense, international custom is even more central than the law of treaties since it is the very pillar on which treaties rest. If one were to speak of a ‘constitution’ of the international community, it would have to be a customary one.” James Crawford has recently taken the same view, arguing that “international law is a customary law system, despite all the treaties: even the principle pacta sunt servanda, the obligation to comply with treaties, is a customary law obligation.” Common to all these views about the primacy of custom is the notion of Briggite Stern in his seminal work: Custom at the Heart of International Law, that customary law has a privileged, elemental status “at the heart” of the international legal order and represents the source of all sources, the background that determines the condition of validity of all other legal norms and processes.

Then again, there are some extremists in this school of jurisprudence who argue rather that, custom is the only process capable of producing law in the proper sense of the term. Fitzmaurice classifies treaties as mere contracts which can do little more than create specific rules, applicable to specific parties, in specific contexts. Even after Fitzmaurice, many scholars such as Clive Parry toed his line. Clive Parry asserts that treaty is essentially peripheral as a source of international law. It is custom, in his view, that defines the basic constitutional structure and general principles of international law as a system. Ian Brownlie went as far as listing treaties alongside General Assembly resolutions and drafts adopted by the International Law Commission as “material sources” exercising direct influence on the content of the law, rather than “formal sources” of law proper, thereby relegating the essence of treaty law as a source of law properly so called.

Yet still, there are others who argue extensively that custom ranks superior normatively, inasmuch as it has the ability to generate universally applicable norms, unlike treaties which are often met by unnecessary reservations. In fact, the ICJ, stated in the North Sea Continental Shelf Cases that “customary rules and obligations, by their very nature, must have equal force for all members of the international community and cannot therefore be the subject of any right of unilateral exclusion.” Ordinarily, States are not permitted to opt out of customary law unless they have persistently and unambiguously objected to its formation.

This can be gleaned from the role of protests and acquiescence in the formation of customary law. Authors such as James Brierly hold the view that implied consent of States is not a philosophical tenable and sound explanation of customary law, domestic or international. Nonetheless, MacGibbon and others take the view that actual protests are called for to break the legitimizing process of customary law, otherwise States would be deemed to have consented by their silence. The Chamber of the International Court in the Gulf of Maine Case defined acquiescence as ‘equivalent to tacit recognition manifested by unilateral conduct which the other party may interpret as consent and as founded upon the principles of good faith and equity.’

As was stated unambiguously by the ICJ in the North Sea Continental Shelf Cases, “customary rules and obligations, by their very nature, must have equal force for all members of the international community and cannot therefore be the subject of any right of unilateral exclusion.” This seems to suggest that customary laws create obligations capable of universal application.




Having considered at great length the major opposing views as to whether customary or treaty law ranks superior, it is worth noting that each of these views is not without its own flaws. For instance, Lauterpacht famously noted that, if one subscribes to the view that treaties are binding only because there is a customary rule to that effect, as the proponents of the treaty superiority view postulate “there remains the question why custom is binding”, a question, the answer to which is not easy to identify. Then again, Mohammed Bedjaoui argues that custom as a source of International Law is “backward looking, conservative because static, iniquitous in its content, ponderous in its formation, custom as traditionally conceived cannot be of real use in the development of new rules, and could actually be an obstacle to any attempt at change”, thereby calling for lesser attention to be paid to it as a source of law.

Then the obvious question arises. To which view does the present writer subscribe? As a matter of fact, practitioners in both schools proffer very sound and convincing arguments in support of their propositions. Even so, the present writer does not fully subscribe to any of the above propositions. In fact, an attempt to add the words ‘in the undermentioned order’ to the wording of article 38 was rejected by the committee of experts that drafted the statute.  It is relevant, as far as contemporary international law is concerned, that inter-State relations be governed by expressly laid down rules, so that a defaulting party to an agreement can be held liable for any breach occasioned. Be that as it may, it is equally relevant that in the case of a dispute arising, the adjudicating tribunal looks out for any existing custom between the contesting States, placing the onus of proving the existence of any customary rule on the party who so alleges.

A strong view has been held by some scholars that the later in time takes precedence, as treaties are usually formulated to replace or codify existing custom, whereas treaties in turn may themselves fall out of use and be replaced by new customary rules. This, to my mind, is a sounder postulation. In this connection, a question that is likely to be posed is, in the case where the provisions of a treaty and the existing custom between the contesting States are in sharp contrast with each other, which law shall be applied? The answer is simply that, the latest in time to be formed should prevail, insofar as its application will not occasion a grave miscarriage of justice. From the foregoing analysis, it can be gleaned that the present writer takes the view that there is no need for any hierarchical categorization between the sources of International Law.