íkovo54 to realise that careful expositions of legal positions that do not form part of the decision can develop a life of their own, often forming the basis of subsequent legal argument by a range of international actors. While dissenting and separate opinions provide a platform for individual judges to develop their own theories of international adjudication, it has also been suggested that regular dissent limits the opportunities for a judge to influence the majority in its crafting of the court’s opinion.55
The ICJ has remained largely impervious to the receipt of information or argument from any source other than the parties in contentious cases56 and even in its advisory jurisdictions accepts arguments or information only from an international organisation considered ‘likely to be able to furnish information on the question’.57 As discussed in Chapter 2 other international courts are more receptive to submissions from non-parties, including from NGOs in the form of amicus briefs. Unless explicitly referred to in the judgment, the extent to which such information is influential in the process of decision-making remains unclear.
The greater number of multilateral instruments containing compulsory or optional dispute resolution clauses has ensured that judicial tribunals have had greater opportunity both to amplify (and sometimes confuse) the understanding of how international law derives from the sources listed in Article 38 (1) of the ICJ Statute and have developed substantive rules and principles. Clauses in multilateral treaties providing for the jurisdiction of the ICJ invariably do so in cases of disputes over the interpretation or application of the particular convention. Other clauses may spell out the law to be applied by the ICJ in making its determination. All three of these interlocking tasks—determining the applicable law, application and interpretation of the convention—assume existing law to be applied but in fact contribute to the processes of law-making. Determination of the applicable law may frame an issue as falling within the parameters of general international law or a regional or subject-specific lex specialis. This initial categorisation then shapes the development of law in that area. Judicial reliance upon, and interpretation of, some principle or instrument (whether in hard or soft law form) bestows it with meaning and authoritative weight.58 It is not only the ICJ that carries out these functions. Other courts, including national courts, may look to the ICJ’s interpretation of international law and to its methodology when determining the existence and content of a rule of conventional or customary international law.59 This section looks at some aspects of adjudication in the development and clarification of international law through determination of applicable law, application and interpretation.
It should not be assumed that in deciding cases brought before them international courts and tribunals always apply the whole of international law. They have jurisdiction to decide cases only insofar as the parties have conferred jurisdiction on them to do so. No court exercises universal compulsory jurisdiction over legal disputes, although some specialised tribunals, such as the WTO Dispute Settlement Body, or a tribunal established under the 1982 UNCLOS, have compulsory jurisdiction within their particular field. The political processes involved in the creation of international courts and the negotiation of dispute settlement clauses in treaties significantly affect the discretion left to adjudicators in determining the substantive law they are to apply. The widest discretion is accorded by Article 38 (2) of the Statute of the ICJ which envisages that with agreement of the parties the Court may decide a case ex aequo et bono—in effect a decision not necessarily based on legal rules.60 Although this choice has never been exercised, states have sometimes agreed that a dispute will be adjudicated on the basis of rules that are not yet law. Thus in the Tunisia–Libya Continental Shelf case the compromis provided that the Court would apply international law including ‘the recent trends admitted at the Third Conference on the Law of the Sea’.61
There are many other formulae for determining the applicable law. The simplest is where judges are directed to decide cases in accordance with the ‘applicable rules of international law’.62 In other instances the general rules of international law are incorporated as a yardstick against which claims or defences are to be measured, thereby bringing into a convention legal principles of international law external to it. For example, in the Oil Platforms case, the ICJ interpreted the provisions of a treaty of friendship in such a way that international law on the use of force became applicable for the purpose of applying the treaty.63 Similarly, the European Convention on Human Rights, Protocol 1, Article 1 provides that no one shall be deprived of property ‘except . . . by the general principles of international law’. The applicable law may also include related treaties or soft law instruments which set standards with which the parties to the principal treaty are required to conform. As we saw in Chapter 5, the 1982 UNCLOS makes reference for this purpose to ‘generally accepted international rules and standards established through the competent international organisation or general diplomatic conference’.64
Many treaties do not incorporate other rules of international law quite so liberally or at all. Some impose explicit limits on the application of other rules. One result of limiting the applicable law in this way is that some disputes cannot easily be decided in accordance with all the relevant international law potentially applicable between the parties.65 This is particularly a phenomenon of treaty disputes arising under compulsory jurisdiction clauses. For example, in accordance with Article 293 of the 1982 UNCLOS a court or tribunal must apply the ‘Convention and other rules of international law not incompatible’ with it. It has been argued that Article 293 allows a court to adjudicate an UNCLOS dispute on the basis of customary law or other treaties binding on the parties to the dispute, and that the only limit on doing so is compatibility with the Convention. On that basis in the Mox Plant Arbitration Ireland sought to persuade a tribunal that it could decide the case in accordance with the more detailed rules on environmental impact assessment and the precautionary principle found in European law and European regional treaties.66 The problem with this view is that it would turn a dispute settlement system that confers only limited compulsory jurisdiction for the purposes of a particular treaty into a general jurisdiction clause over non-UNCLOS aspects of a dispute. Using an applicable law provision for this purpose is not an argument that any international tribunal has so far found persuasive.67 The alternative view is that UNCLOS tribunals may adjudicate on questions of general international law only insofar as it is within their jurisdiction and not inconsistent with UNCLOS to do so: that is, only where other rules of law are expressly incorporated by specific articles of the Convention, or where it is necessary to apply other rules in order to decide the UNCLOS dispute.68 For example, if the question is whether freedom of fishing on the high seas can be suspended in order to carry out nuclear tests, the legality of such tests under general international law would have to be decided in order to apply the Convention’s rule on use of the high seas with ‘due regard’ for the rights and interests of other states.69 On that basis, Article 293 does not extend the jurisdiction of tribunals under the Convention and its reference to applicable law should be read accordingly.
The problem of ensuring coherent application of international law in such situations can partially be addressed through interpretation. Article 31 (1) of the Vienna Convention on the Law of Treaties provides for a treaty to be interpreted in its context. Article 31 (2) gives a broad meaning to a treaty’s context by including any instrument made ‘in connection with the conclusion of the treaty’. The WTO adjudicating bodies have brought in under this rubric70 the Harmonised System—a treaty negotiated under the auspices of the World Customs Organisation—recognising it as relevant context for the interpretation of obligations under the WTO.71 Moreover, as we saw in Chapter 5, Article 31 (3) also allows other rules of international law to be taken into account when interpreting a treaty. In all these ways the applicable law of a particular convention can be extended beyond its own terms, allowing up to a point for consistent judicial application across connected subject matters. They maintain the fiction that the law is there simply to be identified and applied by the judges. But in reality, by extending the scope of applicable law, adjudicators may in effect incorporate principles outside those specifically agreed by the parties to the dispute and thereby develop the law beyond the framework of the convention upon which jurisdiction is founded.
In other instances, however, states may be concerned that they will not view such judicial determinations favourably and thus seek in advance to pre-empt any law-making effect. For example the NAFTA Agreement provides that the parties may collectively issue interpretations of their specific agreements under NAFTA and that any such interpretative statement is binding on the Tribunal.72 Similarly, Article 3 (2) of the WTO Dispute Settlement Understanding provides that ‘Recommendations and rulings of the DSB [Dispute Settlement Body] cannot add to or diminish the rights and obligations provided in the covered agreements’. In this way the parties have asserted their own competence to interpret these agreements, thereby prioritising their own political determinations over potential judicial decisions. There is an interesting contrast between the Statutes of the ICTY and ICTR on the one hand and the ICC on the other in this regard. The former provide virtually no guidance to the judges in formulating the applicable rules of international criminal law and procedure. The Report of the Secretary-General that was preparatory to the establishment of the ICTY stated that the Tribunal would apply international humanitarian law that exists in conventional and customary law.73 The Secretary-General emphasised that the Tribunal should ‘apply rules of international humanitarian law which are beyond any doubt part of customary law’. However no definitions of offences were provided (other than genocide) and there had been no international war crimes trials since the adoption of the 1949 Geneva Conventions for the Tribunal to draw upon. The Nuremberg and Tokyo Tribunals did provide a body of case law but these Tribunals were established over 50 years ago, with different jurisdiction and before the evolution of much contemporary international law, for example international human rights law. It was evident that the ICTY (and subsequently the ICTR) would have to determine its own definitions, processes and rules. Indeed the judges were entrusted with drawing up the Rules of Procedure and Evidence,74 again without any directional guidance.
In contrast, in the highly politicised negotiations for the Rome Statute of the ICC, states sought to minimise the Court’s future law-making capacity by determining that the Elements of Crimes and the Rules of Procedure and Evidence would be set out through negotiation by states at the Assembly of States Parties. These important matters were thus not left to the judges to determine through judicial decision-making or adoption of an instrument.
Nevertheless, despite the conclusion of the Elements of Crimes and the Rules of Procedure and Evidence, the ICC judges will have the detailed and comparatively prolific jurisprudence of the ad hoc Tribunals before them as well as that of other courts such as the ICJ. In some instances this jurisprudence was incorporated into the ICC definitions adopted by states parties, in others it was not. But legal history cannot be turned back and it is certain that lawyers arguing before the ICC will draw upon these cases. Assuming the ICC develops a reasonable case-load, it may have to choose whether to seek consistency with the jurisprudence of the ad hoc Tribunals, in effect recognising an applicable international criminal law, and regarding itself as justified by this drafting history in deviating from that law by applying the law of the Statute. If it favours the former approach it will assist in the evolution of a coherent body of international criminal law emanating from a mix of SC resolutions, treaty negotiation and the case law of the ad hoc Tribunals. If it relies heavily on its own Statute to the exclusion of these other sources it could marginalise the work of the ad hoc (and time limited) Tribunals and may retrospectively undermine their authority.
We saw in Chapter 5 how some treaties evolve through interpretation, a point of particular resonance in the field of human rights. For example the European Convention on Human Rights was adopted in 1950. Its provisions are short and claims are made for rights that do not fall within their precise terms. The ECHR has to strike a balance between interpreting the Convention in such a way that it continues to have validity in changing social, economic and political conditions and not asserting new rights that have not been accepted by the parties. New rights can be introduced by adoption of a Protocol and it is not the Court’s role to usurp the ‘legislative power’ of states parties.
The concept of a ‘living instrument’ has been the Court’s answer, that is using existing treaty articles to reach innovative outcomes that reflect society’s changing attitudes. The ECHR has said that the Convention is ‘a living instrument which . . . must be interpreted in the light of present day conditions’75 enabling decisions that would not have been anticipated at the time of its adoption. A frank admission of the consequences of the judicial role was made by Judge Garlicki in Öcalan v. Turkey:
This may result (and, in fact, has on numerous occasions resulted) in judicial modifications of the original meaning of the Convention . . . The Strasbourg Court has demonstrated such a creative approach to the text of the Convention many times, holding that the Convention rights and freedoms are applicable to situations which were not envisaged by the original drafters.76
Thus in Soering v. UK 77 the Court stated that it ‘cannot but be influenced by the developments and commonly accepted standards in the penal policy in the member states of the Council of Europe’.78 In this case the Court noted the ‘virtual consensus’ within such states that the death penalty should not be imposed for crimes committed in times of peace as it is no longer ‘consistent with regional standards of justice’. This was not the case in 1950. Accordingly the Court found it to be a violation of Article 3 (prohibition of torture, cruel, inhuman treatment) to extradite a person to a state (outside the Council of Europe) where he might face the death penalty and thus be subjected to the agonies of detention on death row. In Öcalan itself, after taking account of the current attitude towards the death penalty in the member states of the Council of Europe, the Court went further, holding that ‘capital punishment in peacetime has come to be regarded as an unacceptable, if not inhuman, form of punishment which is no longer acceptable under Article 2’.79 Similarly there are no articles specifically protecting the human rights of lesbians, gay men, transexuals and bisexuals. However, the Court has interpreted the Convention to bring such rights within Article 8 (the right to privacy). Although there is neither a right to health nor to a clean environment within the Convention the ECHR has also relied on Article 8 to develop environmental rights. For example it has found a violation of the right to respect for private and family life through a government’s failure to regulate environmental nuisances, or to enforce the law in such cases, or to warn of the associated dangers and health risks.80 It considered that serious harm to the environment may affect the welfare of persons and deprive them of the enjoyment of their homes in such a way as to damage their private and family life.
The law-making effect of such decisions is especially powerful because of the direct effect of the European Convention in the domestic law of many member states. In the UK (where incorporation of a treaty into national law is required) the Human Rights Act 1998, section 2 requires any court or tribunal making a determination relating to a Convention right to take into account any ‘judgment, decision, declaration or advisory opinion’ of the ECHR. ECHR decisions are argued before UK courts as authoritative precedents and analysed by judges as such. Decisions of national courts are formative of state practice and opinio juris for the formation of at least a regional customary international law of human rights. The cross-fertilisation of human rights principles between the human rights bodies of the Inter-American and African systems and the UN treaty bodies strengthens arguments of customary international law standards emergent from the terms of the conventions and contributes to the development of general principles of law.
The Statute of the ICJ, Article 38 (1) sets out where the ICJ should look for existing law where there is no relevant treaty.81Other courts deciding issues on the basis of international law look to the same sources. The first listed is ‘international custom, as evidence of a general practice accepted as law’. Judicial decision-making does not of course constitute customary international law, which long pre-dates the establishment of the PCIJ.82 But judicial reasoning may clarify its mysteries. In particular the ICJ has had many occasions to consider the processes by which customary international law is formed and identified. Thus although the ‘existence and content of custom is usually determined by states and academics . . . the Court remains the ultimate arbiter in some cases’.83 The choices a court (and individual judges) makes in determining its recourse to, and application of, customary international law (and the other listed sources) are political, value choices. The Court’s approach and structure of legal argument with respect to custom have been subject to significant analysis and critique, most notably in recent years from critical legal scholars.84
It is commonplace that the ICJ has set out what it regards as the appropriate criteria for determination of customary international law: the actual practice of states and opinio juris.85 As we saw in Chapter 5, it has asserted in a number of cases whether there is, or is not, a rule of customary international law, and the content of particular rules. It has refined understandings of custom (for example the existence of regional customary law)86 and procedural requirements for pleading custom (for example that the state relying on custom must prove it). These have in turn become part of the body of customary international law. The Court has recognised its constraints and has refrained from blatant law-making, for example by speculating what future law might be.87 However, applying the criteria for establishing custom is not a scientific process, the accuracy of which can be measured. Rather it requires an evaluation of the facts and arguments selected for presentation to the Court by advocates appearing before it. Indeed it has been observed that ‘custom, if considered from a technical point of view, is not so much the rule; it is the procedure of creating the rule’,88 that is a methodology.
The Court has considered customary international law in a great many cases, only a few of which can be referred to here.89However even a brief examination of the way the Court has addressed the making of customary international law shows both the law-making potential in this process and how the Court has not been consistent in applying its own criteria for the determination of customary international law.
In a number of early cases the PCIJ developed its approach to determining when a rule of customary international law exists, the content of the rule, to whom it is applicable and the limits of its application.90Despite these pronouncements it has been inconsistent in its examination of state practice and opinio juris and in the respective weight it has given to these distinct elements. For example in the 1951 Anglo-Norwegian Fisheries case91 it referred to no practice with respect to the drawing of straight baselines around the skjaregaard from any state other than Norway and indeed noted that the Second Sub-Committee of the Second Committee of the 1930 Conference for the Codification of International Law had formulated the low water mark rule ‘somewhat’ strictly.
In the 1974 Fisheries Jurisdiction case the Court found two concepts to have crystallised as customary international law: the concept of a 12-mile exclusive fishing zone and that of preferential rights for coastal states beyond 12 miles. These concepts were said to have arisen out of the general consensus revealed at the 1960 Geneva Conference on the Law of the Sea—which had failed to reach agreement on the extent of fishery rights. Furthermore, UNCLOS III (which had commenced in 1973)92 had not yet reached any conclusions—as it would not for another eight years. Without citing any concrete instances of state practice the Court noted that it was ‘aware that a number of States has asserted an extension of fishery limits’.93 It left unspoken such questions as: how was it aware of this? By arguments presented before it? If so why not acknowledge the fact? How many states? How great an extension of fishery limits? Have they been widely accepted? By whom? The Court was also ‘aware’ of the manifest desire of states to codify the law through UNCLOS III. While asserting that it could not usurp the legislator by anticipating the law, the Court did precisely that. In the same case at its jurisdiction stage the Court noted that the Vienna Convention on the Law of Treaties, 1969, Article 62 ‘may in many respects be considered as a codification of existing customary law’.94 Again this leaves open many questions: in what respects? On what basis? It did not refer to its own methodology expounded in the North Sea Continental Shelf cases on when a treaty provision may be regarded as customary international law.95 No state practice was offered as evidence. There was no reference to the ILC commentaries on Article 62, nor to the debates at the Vienna Conference. Unlike the Law of the Sea Convention, at least in this instance the treaty had been completed, although it was not yet in force. Such failure to act consistently with its own asserted methodology96 undermines the legitimacy of judicial decision-making, and the content of the espoused customary laws. Indeed the ‘practice of the ICJ to vary at will, and without acknowledging standard rules is not legitimacy enhancing for both the new norms of law created and for law creating mechanism involved’.97
As discussed in Chapter 5, in the 1986 Nicaragua case the Court apparently reversed its traditional approach of seeking state practice supported by opinio juris by finding first opinio juris in the form of UNGA resolutions and then looking for state practice.98 The Court discounted the inconvenient fact that much state practice was inconsistent with the ideals expressed in those resolutions as constituting evidence weighing against the opinio juris but instead viewed such practice as breaches of those rules.99Nor did it inquire whether there was any state practice or opinio juris to support its opinion that Articles 1 and 3 of the Geneva Conventions on the Laws of War had become customary international law.100
The Court’s inconsistency with respect to the weight to be accorded to state practice and opinio juris has elicited different responses. Schachter has argued that there are different types of customary rules.101 The first type is ‘the great body of customary rules’ such as those on jurisdiction, immunity and state responsibility where consistent state practice must still be sought. The second type comprises those rules that prohibit state conduct that offends against core values and sustain the international order: the prohibition of aggression, genocide, torture, slavery and systematic racial discrimination. These norms are ‘brittle’ in their susceptibility to violation and must therefore be reaffirmed even in the face of such violations.102 Kirgis has similarly argued for different processes according to the substantive nature of the proposed rule. Greater consistency in state practice is required where there is little evidence of opinio juris but contradictory practice is tolerated where there is expressed consensus about illegality, for example through UNGA resolutions.103
These analyses fuse the making of customary international law with its content. The use of ethical and human rights values in this way adds to the inherent uncertainties that surround customary international law and dispel any pretence at an objective methodology. They assume that international law is imbued with the core values of an international community which debates about universalism make uncertain.104 But in any case such analyses are further confused by subsequent decisions of the Court. For example, in the 2002 Arrest Warrant case the ICJ had to determine whether a foreign minister has immunity from the criminal jurisdiction of foreign states asserted under principles of universal jurisdiction. The case therefore involved one of Schachter’s ‘great body of customary rules’—immunity. There was at that time no international treaty on sovereign immunity105 and there remains much disagreement about the substantive content of any customary rule on criminal immunities. Further this was the first opportunity for the ICJ to consider immunity of state officials after the ruling of the UK House of Lords in the Pinochet case106 which had held there to be no immunity (for a former Head of State) in the case of alleged torture. According to Schachter’s analysis above, evidence of consistent state practice would be required for the pronouncement of a rule of customary international law. In these circumstances it might have been anticipated that the Court would be especially conscientious in this regard. However the Court made no reference to any methodology for reaching its conclusion on the immunity of a foreign minister. It cited no state practice (although it claimed to have examined some)107 nor any evidence of opinio juris and decided on the basis of its conclusions about the functions of a foreign minister. Similarly with respect to Belgium’s claim that under customary international law immunity is displaced in the case of war crimes and crimes against humanity, the Court stated that it had examined state practice, but did not provide any information about what state practice, from which states and in what context. Nor was there any reference to the need for, or existence of, opinio juris. In this instance its decision upheld state sovereignty and implicitly rejected the moves to develop the law along the lines commenced by the House of Lords.108
Judge ad hoc Van Den Wyngaert regretted the lack of a ‘principled perspective’109 in the majority’s approach and stressed that there is no support for the proposition that foreign ministers enjoy immunity under customary international law. In criticising the majority’s methodology and structure of its argument, the disregard of its own case law, the failure to undertake a ‘rigorous approach’, the flaws in drawing an analogy between Heads of State and Foreign Ministers, and the brevity of the judgment, Judge Van Den Wyngaert provides grounds for questioning the legitimacy of the decision and the rule of customary law pronounced therein.
In contrast in the 1996 advisory opinion on the Legality of the Threat or Use of Nuclear Weapons the Court had considered both the consistency of state practice and the normative effect of UNGA resolutions. On this occasion there was consistent state practice but it was of omission not commission—no state had used a nuclear weapon since the end of World War II. The Court concluded that it could not be sure if states had refrained from the use of such weapons out of a conviction of their illegality or because they favoured the political policy of deterrence. In these circumstances there was no evidence of opinio juris to be derived from state practice. The Court then looked to whether UNGA resolutions could be used as evidence of opinio juris. The relevant factors for making this determination are the ‘content and the conditions’ of the adoption of the resolution and ‘whether an opinio juris exists as to its normative character’. It also accepted that while a single resolution is not determinative ‘a series of resolutions may show the gradual evolution of the opinio juris required for the establishment of a new rule’.110 In light of the fact that several relevant resolutions were adopted with negative votes and abstentions they fell ‘short of establishing the existence of an opinio juris on the illegality of the use of such weapons’.111 The Court accepted that the series of resolutions indicated the concern of the international community about the issue, and even that there was an evolving principle of customary international law, but that the conflicting evidence as to opinio juris denied the existence of an existing law.
In Nicaragua the Court had given effect to UNGA resolutions as constituting opinio juris while in the Legality of Nuclear Weapons it did not. It is evident that it is not the ‘soft law’ nature of UNGA resolutions that determines the issue. In the Legal Consequences of the Construction of a Wall in the Occupied Palestinian Territory case the Court drew broadly on soft law. It accepted the opinions of a ‘soft’ tribunal, the Human Rights Committee, given through a soft adjudicative process (individual complaints procedure) and a non-binding General Comment112 to find the International Covenant on Civil and Political Rights to be applicable in the Occupied Territories.113 The suggestion has been made that greater recourse to adjudication through the proliferation of international courts might lessen the importance of soft law because of a judicial preference for hard law sources, made less vague through their interpretation and application and given effect through hard remedies. This ‘half-truth’ has been exposed by José Alvarez114 and is well illustrated by this case.
The Court may adopt one of three procedures: declaring existing law; crystallising a rule of customary international law, that is by articulating an emergent rule and transforming it into an existent rule; and generating or constituting a rule whereby the Court’s pronouncement becomes the ‘focal point of a subsequent practice of states which, in due course, hardens into a rule of customary international law’.115 An oft-cited example is the Anglo-Norwegian Fisheries case where Norway’s novel assertion of the legality of the use of straight baselines and long usage was accepted by the ICJ, despite strong legal argument from the UK that this was against general international law and lack of supporting practice. The judgment has been described as serving one purpose: sanctioning the Norwegian government’s breach of international law on the acquisition of maritime territory.116 Nevertheless it generated practice both from the UK itself and from other states. Only seven years later the substance of the ruling was incorporated into the Geneva Convention on the Territorial Sea and Contiguous Zone, 1958, Article 4 and in almost identical language into the 1982 Convention on the Law of the Sea, Article 7.
Where there is no previous practice or other sources (hard or soft) the Court must use other techniques to support its reasoning. In Anglo-Norwegian Fisheries the Court paid attention to what it termed ‘the realities’ of the situation: that the coast of the mainland is exceptional in that it ‘does not constitute, as it does in almost all other countries, a clear dividing line between land and sea’; that along the coast are shallow banks that are veritable fishing grounds, known to the Norwegian fishing people since time immemorial; and that in these barren regions the inhabitants derive their livelihoods essentially from fishing.117 The Court also stressed ‘certain basic considerations’ that determine the relevant criteria for the drawing of straight baselines, for example the ‘close dependence of the territorial sea upon the land domain’, and the ‘more or less close relationship existing between certain sea areas and the land formations which divide or surround them’ and the economic factors peculiar to the region, ‘the reality and importance of which are clearly evidenced by a long usage’.118 Finally the Court also found that the ‘general toleration of foreign states with regard to the Norwegian practice is an unchallenged fact’.119 In contrast, Norway’s rejection of the ten-mile rule meant that it was not bound. David Kennedy has concluded that:
It seems that doctrines about the creation of custom seem to be about the conditions of justice or the nature of the system of international law, while doctrines about the limits of custom seem to be about the failure of consent by the state to be bound.120
In selecting one of the three procedures (declaring existing law, crystallising a rule of customary international law or constituting a rule) the Court has fluctuated in the respective weight it has given to the components of state practice and opinio juris and the amount of evidence it has required for either. At times it has placed greater emphasis on state practice (termed the classic or traditional approach121), although varying its requirements as to time, number of states, identity of states and consistency of practice. In other decisions it has favoured expressions of opinio juris over inconsistent state practice (termed the modern approach). Koskenniemi has famously termed this the dilemma of apology and utopia: law that is reliant on state practice defers too much to the realities of power while that which rests on principles unrelated to actual behaviour is Utopian and destined never to be achieved.122 In other instances it has given little attention to the need to ascertain either. While the ICJ has identified a methodology for identifying rules of customary international law it follows it neither consistently nor rigorously. Nor has it clarified the ‘mysteries’ of customary international law expounded by academics123