BLC LLB (Pretoria) LLM (Connecticut) PhD (Rotterdam)
Professor of International Law, University of Pretoria, South Africa; Member of the UN International Law Commission; Special Adviser to the Minister of International Relations and Co-operation of South Africa
Edition: AHRLJ Volume 16 No 2 2016
Pages: 310 - 338
Citation: D Tladi ‘Interpretation and international law in South African courts: The Supreme Court of Appeal and the Al Bashir saga’ (2016) 16 African Human Rights Law Journal 310-338
Download article in PDF
The Constitutional Court was to hear an appeal in this matter on 22 November 2016. At the time of the finalisation of the article, it was unclear whether this hearing would proceed in light of the subsequent decision of the government of South Africa to withdraw from the Rome Statute and, consequently, withdraw its appeal. The author wishes to thank the anonymous reviewers for their helpful comments. Any flaws in the article, of course, remain his alone.
The South African Constitution is regarded as an international-law friendly constitution. Much has been written about the willingness of South African courts to refer to international law instruments when interpreting and applying South African law. Yet, the extent to which South African courts have applied recognised tools and methods for the identification and interpretation of international law has not similarly been considered. The recent case concerning South Africa’s decision not to arrest the President of Sudan, Al Bashir, highlights the importance of a proper approach to the interpretation and identification of international law by South African courts. In this case, the Supreme Court of Appeal had to consider the complex interrelationships between two treaties, namely, the AU-South Africa host country agreement and the Rome Statute of the International Criminal Court, customary international law and a UN Security Council resolution. The objective of the article is not to determine the correctness or not of the decision. Rather, the article is aimed at assessing the Court’s approach to the methodological questions of interpretation and identification of international law. The article, therefore, evaluates whether the rules of interpretation as contained in the Vienna Convention on the Law of Treaties have been applied by the Court in searching for the meaning of the instruments under consideration. It also assesses whether the relationship between the various sources of international law at play in the Al Bashir matter is adequately considered.
The South African Constitution is reputed to be one of the most international law-friendly constitutions in the world. It provides, for example, that the interpretation of the Bill of Rights must take into consideration international law;1 that when interpreting any legislation, any reasonable interpretation consistent with international law must be preferred over any other interpretation that is inconsistent with international law;2 and that customary international law is law in the Republic except where it is in conflict with the Constitution or an Act of Parliament.3
These provisions would seem to require the interpretation and identification of international law by the judiciary. Yet, while much has been written about the role of international law in this international law-friendly constitutional framework, 4 very little has been written about how South African courts approach the actual identification and interpretation of international law. After all, to ‘consider’ international law, to prefer an interpretation of legislation that is consistent with international law, and to apply customary international law as law in South Africa, all presuppose that the rules of international law can be identified and interpreted. The recent decisions of the High Court and Supreme Court of Appeal (SCA), respectively, in relation to the non-arrest of Al Bashir illustrate the importance of careful attention to the methodology of identifying and interpreting international law.5
Over the last few months, much has been written about the continuing Al Bashir saga and, in particular, South Africa’s failure to arrest him. 6 Of course, while the events of June 2015 caused a stir, it was not the first time that this drama had played itself out - there had been similar cases of non-arrest in the Democratic Republic of Congo (DRC), Malawi, Chad (twice), Kenya, Djibouti and Nigeria.7 In fact, since the events of June 2015, this drama has been repeated twice: in Djibouti (for the second time) and in Uganda.8
I wish to focus not on the substantive question of whether there was a duty on South Africa to arrest Al Bashir under international and South African law, but rather on the methodological question concerning the way in which South African courts, in particular the SCA in the Al Bashir case, have approached the question of interpretation of international law or, in the case of customary international law, its identification. The Al Bashir cases have demonstrated the immense importance of a proper understanding of international law, given South Africa’s international law-friendly constitutional framework. At issue in both the High Court and the SCA cases were the interpretation and identification of rules of international law in a complex network involving various inconsistent treaty, customary and other obligations. Further complicating the application of this network of international law rules is the similarly complex network of domestic law rules giving effect to the international law rules. How these would be applied affects (and is affected by) the methodology used to identify the relevant rules of international law.
In light of this, the purpose of the article is to assess the use of recognised methods and doctrines for the interpretation of international law by the SCA in the matter involving the non-arrest of Al Bashir. The article is not concerned with whether there was, or was not, a duty to arrest. It concerns only the methodology applied by the Court when considering the international law materials before it. It is, therefore, not a general comment on the judgment, but rather an evaluation of the Court’s methodological approach to the identification and interpretation of international law.
In the next section, a few comments are made about some of the methodological tools available for interpreting and identifying rules of international law. Section three provides a brief overview of the aspects of the judgment concerning the methodology of interpreting and identifying rules of international law. Section four provides an evaluation of the Court’s methodological approach to the identification and interpretation of international law in the Al Bashir case. Finally, some concluding remarks are offered in section five.
Subsequent to the finalisation of the article, a number of developments occurred which, while beyond the scope of the article, are worth mentioning. First, on 19 October 2016, the government of South Africa decided to withdraw from the Rome Statute. The instrument of withdrawal, signed by the Minister of International Relations and Co-operation, was deposited with the Secretary-General of the United Nations (UN) on the same day. As a consequence, South Africa decided to withdraw its appeal on the Supreme Court of Appeal decision. This set off a strange course of events in which the government, having disagreed with the findings of the Supreme Court of Appeal, now presumably accepted the finding, while the respondent, having opposed the appeal, now sought the appeal to continue. To understand this strange turn of events requires an understanding of the nuances of the Supreme Court of Appeal decision which is very often missed in the popular media.
While, as a general matter, the Supreme Court of Appeal found in favour of the respondent, the Supreme Court of Appeal found for the government in one important respect. As will be shown below, the Court found that, under international law, there was a duty on South Africa to respect the immunities of Al Bashir and that there were no exceptions to this duty in customary international law. The implications of this hardly-noticed fact were that the government, in the conduct of foreign affairs, would, if necessary, be entitled to take action to avoid a conflict of obligations. This resulted in the counter-intuitive situation where the successful party sought to have the appeal proceed and the unsuccessful party sought to prevent the appeal from proceeding.
The decision to withdraw from the Rome Statute has resulted in a challenge by the Democratic Alliance (DA) to the constitutionality of the right of the executive to withdraw from the Rome Statute. While these issues fall beyond the scope of the article, it is worth describing the issues raised in the challenge and to provide preliminary, albeit unsubstantiated, observations. Given their preliminary nature, these may be taken with a grain of salt.
First, the challenge is based on the role of parliament. In its founding affidavit, the DA alleged that, whatever the merits or demerits of the decision to withdraw from the Rome Statute, the executive ought to have sought parliamentary approval before submitting the instrument of withdrawal. This argument is based on the fact that parliamentary approval was required prior to ratification. The argument postulates that the same procedure is required for withdrawal. In the author’s view, because parliament approves and does not oblige the ratification, parliamentary approval is not required. Nonetheless, given the fact that the Constitution is not explicit on the matter, the prudent approach would have been to secure parliamentary approval.
The challenge also raises substantive and complicated issues about the right of the executive to withdraw. These issues depend on matters that fall beyond the scope of the article, such as whether the decision negatively impacts on the fight against impunity and South Africa’s values. For the purposes of the article, it suffices to state that the determination of these complex legal questions would require an assessment of whether, beyond the Rome Statute, mechanisms existed for advancing the values of the South African Constitution.
These and other interesting questions are not addressed in this methodologically-inclined article. Instead, the article purely focuses on the methodological question whether, in the Al Bashir case, the courts have paid attention to the rules of international law for the identification and interpretation of these rules.
It is obviously not possible to comprehensively address the question of how international law is to be identified and interpreted. For this reason, only broad conceptual issues are addressed. Moreover, the analysis will be limited to the main sources of international law, namely, customary international law and treaty law. I have also opted to restrict the description to only those issues that were directly relevant for the judgment of the SCA. On account of these parameters, otherwise important concepts are excluded, such as general principles of law and jus cogens; jus cogens is referred to only in passing.
Any student textbook on international law will tell one that customary international law is both formed and identified by two elements, namely, state practice, also called usus, and acceptance as law, or opinio juris.9 This two-element approach, based on settled jurisprudence of the International Court of Justice (ICJ),10 has been endorsed by the International Law Commission (Commission or ILC) in its consideration of the topic, entitled ‘Identification of Customary International Law’.11 The Commission’s consideration of customary international law was occasioned precisely to offer guidance to, amongst others, domestic courts, which may not always be well versed in the intricacies of international law on how to identify and interpret customary international law.12
During its session in 2016, the ILC adopted a set of draft conclusions on first reading which sets out to provide a normative description of the methodological rules for the identification of customary international law.13 A number of provisions are of particular relevance to the decision by the SCA in the Al Bashir case. The first relevant provision is Draft Conclusion 4, which provides that ‘[g]eneral practice means that it is primarily the practice of states that contributes to the formation, or expression of, rules of customary international law’.14 The word ‘primarily’ is used because the Draft Conclusions recognise that ‘in certain cases, the practice of international organisations’, such as the UN, the African Union (AU) and the European Union (EU) may, as such, contribute ‘to formation, or expression, of rules of customary international law’.15 However, the views or ‘conduct of other actors’ are ‘not practice’ for the purposes of ‘formation, expression, of the rules of customary international law’. Another provision that is of some significance in the assessment of the methodological approach to the sources of international law in the Al Bashir case is Draft Conclusion 7, listing the forms of state practice. According to Draft Conclusion 7, state practice may include diplomatic acts and correspondence; conduct in connection with resolutions; conduct in connection with treaties; and decisions of national courts.
It is important, however, that conduct in connection with a treaty should be approached with some caution. The Commission has determined, first, that the fact that a rule is found in a number of treaties (and other instruments) does not necessarily indicate that the treaty rule reflects a rule of customary international law.16 More to the point, Draft Conclusion 11 sets out the possible relationships between customary international law as determined by the ICJ in the North Sea Continental Shelf cases, namely, that a treaty rule may codify an existing rule of customary international law; may lead to the crystallisation of a rule of customary international law; or may give rise to a general practice accepted as law.17 In each case, however, the rule in question must be ‘confirmed by practice’.18 Moreover, it ‘is important that states can be shown to engage in the practice not (solely) by virtue of the treaty obligation, but out of a conviction that the rule embodied in the treaty is or has become customary international law’.19
Finally, on customary international law, it is worth noting the significance of court decisions. The Statute of the International Court of Justice already determines that decisions of courts are ‘a subsidiary means for the determination of rules of law’. The ILC’s Draft Conclusion identifies two roles for decisions of domestic courts. First, as noted above, decisions of national courts are a form of practice.20 The value of domestic court decisions as a form of practice is not dependent on the quality of the reasoning or correctness of such decisions. In other words, even a court decision that is wholly inconsistent with customary international law is a form of practice and may, thus, contribute to the formation of a (new) rule of customary international law. Decisions of domestic courts, however, may also be relevant in the identification of rules of customary as ‘a subsidiary means for the determination’ of rules of customary international law.21 In this context, the value of the decision is significantly affected by ‘the quality of the reasoning’ and the ‘reception of the decision’. 22
Central to the determination of the rules of international law relevant to the adjudication of the Al Bashir case is the interpretation of the treaty rules. Unlike customary international law, the rules relating to treaties are the subject of a comprehensive treaty regime, namely, the Vienna Convention on the Law of Treaties. There are two aspects of this regime, both of which are accepted as reflecting customary international law, that are significant for the Al Bashir case. The first is that under international law, a treaty rule cannot affect the right of third states, that is, states that are not a party to the said treaty cannot be subject to obligations flowing from that treaty.23 The second aspect, in a sense related to the first, is that international law contains a set of rules for addressing conflict between different treaties. These rules include, amongst others, the lex posterior rule (that a later treaty trumps an earlier treaty); and the lex specialis rule (that the more specific rule trumps the more general).24 It is, however, important to understand the limits of these rules to resolve conflicts between treaties. The most important limit is that these rules can only be applicable if the respective treaties have the same parties. Therefore, the lex posterior rule cannot be applied where some parties to the earlier treaty are not parties to the later treaty, since to do that may have the result that the rights and obligations of states that are not parties to the later treaty are affected contrary to article 34 of the Vienna Convention.
The third, and perhaps the most important aspect of the regime, are the Vienna rules of interpretation. The Vienna rules of interpretation can be found in articles 31 and 32 of the Vienna Convention on the Law of Treaties. The primary rule, contained in article 31(1),25 requires that a treaty ‘be interpreted in good faith in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in the light of its object and purpose’. The general rule set forth in article 31(1) requires that these three elements, namely, ordinary meaning, context and object and purpose, be assessed in good faith. Unlike in domestic interpretation, all three elements are important and contribute equally to the interpretation of a text. In addition to these three elements, international law also requires that subsequent agreements of the parties relating to the interpretation of the treaty and subsequent practice which establish the agreement of the parties as to the interpretation ‘shall be taken into account’ in the interpretation of a treaty.26 Subsequent agreement and subsequent practice are not supplementary means of interpretation. Rather, they are primary means and it is obligatory to take them into account in the interpretation of treaties. Where subsequent practice does not establish the agreement of the parties, it may nonetheless be considered as a supplementary means of interpretation under article 32 of the Vienna Convention.27 The Vienna rules additionally provide that in the interpretation of treaty rules, other ‘relevant rules of [applicable] international law’ must be considered.28
It is worth referring to article 32 of the Vienna Convention which makes it plain that one of the purposes of treaty interpretation is to avoid ‘manifestly absurd or unreasonable results’.29 Article 32 similarly applies in cases where a treaty provision is ambiguous. In these cases - where the normal rules result in absurd or unreasonable interpretation or where the treaty provision is ambiguous - the Vienna Convention allows recourse to supplementary means of interpretation, including the preparatory works to the treaty and other subsequent practice that do not meet the requirements of article 31(3)(b) of the Vienna Convention.
Finally, a few words concerning the interpretation of United Nations Security Council (UNSC) resolutions are appropriate. The first point to make is that resolutions of the Security Council are not treaties. However, Security Council resolutions are international law texts that make (or at the very least contribute to the making of) law. Thus, while Security Council resolutions are not treaties, there is no a priori reason why the rules of interpretation applicable to treaties as contained in the Vienna Convention cannot be used. The ICJ has, albeit cautiously, endorsed the principle of applying the Vienna rules to Security Council resolutions.30 Although the Court proceeded to apply the means of interpretation in the Vienna rules,31 it cautioned that the peculiarities of Security Council resolutions ought to be considered and that, because of these peculiarities, other factors have to be considered in the interpretation of Security Council resolutions.32 The Court noted, in particular, that Security Council resolutions were ‘issued by a single, collective body’, were drafted through a very different process than that for the drafting of treaties, the product of vote under procedures under the UN Charter and may be binding on all states.33 In truth, these differences, to the extent that they reveal anything peculiar about Security Council resolutions,34 do not necessitate a different approach to interpretation. These peculiarities may be true of how a text comes into being, but that in itself does not make the elucidation of the text of a resolution any different from the elucidation of the text of a treaty. After all, it is not inconceivable that a treaty is drafted in similar fashion. This is not to say that there are no differences between resolutions and treaties that affect the interpretation process. These differences, however, to the extent that they exist, should be identified on the basis of an assessment of the Vienna rules.
Having set out the basic rules applicable to the identification and interpretation of international law, it is important to say a word or two about whether South African courts are obliged to apply these rules. As a starting point, one ought to distinguish the case of the interpretation of international law rules, on the one hand, from the case of the interpretation of domestic legislation implementing an international law rule, on the other. Domestic legislation, even when implementing an international rule, remains domestic in nature and the rules applicable to their interpretation, save where the legislation itself provides otherwise, are domestic law rules of interpretation.
Therefore, the question of whether the Implementation of the Rome Statute of the International Criminal Court Act (Implementation Act)35 and the Diplomatic Immunities and Privileges Act (Immunities Act),36 both of which incorporate different treaty regimes, are to be interpreted in accordance with international law rules of interpretation, is to be determined with reference to domestic law. To the extent that the two legislative Acts are silent on their interpretation, domestic rules of interpretation are applicable. Any problem that may arise due to inconsistencies between the implementation and the international law rule ought to be avoided by the application of section 233 of the Constitution, which requires an interpreter to prefer reasonable interpretations that are consistent with international law over other interpretations. Different considerations, however, apply to the interpretation and identification of international law rules as such.
As a doctrinal matter, whether the rules of international law applicable to identification and interpretation are applicable or not is dependent on the domestic system. In South Africa, these rules can be found in the Constitution. According to section 232 of the Constitution, customary international law is law in South Africa. This means that, to the extent that the rules of identification and interpretation described above form part of customary international law, South African courts are obliged to apply these rules when identifying or interpreting international law. This is certainly the case with the Vienna rules of interpretation and the rules relevant to the identification of customary international law.
A related question is whether, as a normative question, South African courts should apply the international rules of interpretation and identification of international law when dealing with international law. The answer to this question should be a resounding ‘yes’. To take treaty interpretation as an example, if the objective is to find the true objective meaning of the treaty, the application of a common set of agreed rules for treaty interpretation should facilitate this search. The idea that there is an objectively correct interpretation, even if arriving at it is difficult, in part explains the importance of having and applying common rules of interpretation. By the same token, it is meaningless to proclaim that customary international law is law in South Africa if the method of identification and, therefore, the likely content of those rules are different.
The facts of the Al Bashir case are by now well known and will not be repeated here.37 As in the High Court case, the SCA had to address a number of issues requiring the interpretation and identification of international law. The first and most obvious question pertained to the interpretation of the Rome Statute and, in particular, the provisions relating to immunities and co-operation. As is now well known, there is an inherent tension between articles 27 and 98 of the Rome Statute, and this tension is at the heart of the Al Bashir debate. While article 27 provides that no person, including sitting heads of state, enjoys immunity before the International Criminal Court (ICC), article 98 provides an exception to the duty to co-operate in the arrest and surrender of persons possessing immunity.38 The second important question relating to the interpretation of international law relates to the host country agreement between South Africa and the AU for the hosting of the AU Summit (host country agreement).39 The SCA also had to determine the content of the rules of customary international law relating to immunities. Finally, the Al Bashir cases required the interpretation of UN Security Council Resolution 1593 which had referred the situation in Darfur to the ICC.40 For the convenience of the reader, I describe the Court’s approach to each of these in the order used by the Court. Thus, I will first describe the Court’s approach to the host country agreement. I shall then proceed to consider the Court’s approach to the Rome Statute, in particular article 98, before addressing the Court’s approach to customary international law. In fairness, the customary international law findings are intertwined with the Rome Statute findings since, in essence, customary international law is raised as a treaty exception under article 98.
As a matter of international practice, international conferences organised under the auspices of international organisations are governed by host country agreements. Where the organisation is based in the territory of the state hosting the conference, such as, for example, AU meetings in Addis Ababa or UN meetings in New York or Geneva, a permanent host country agreement governs the conference. Accordingly, a host country agreement was concluded between South Africa and the AU. One of the arguments of the South African government in the Al Bashir case was that the host country agreement granted Al Bashir immunity from arrest and surrender. The SCA began its assessment of the government’s argument as follows:41
The High Court gave the argument short thrift. It said that on its own terms the hosting agreement conferred immunity on members or staff of the AU Commission and on delegates and other representatives of international organisations. This did not include member states or their representatives or delegates.
Therefore, in the view of the High Court, the relevant provisions of the host country agreement ‘did not cover heads of state or representatives of states’.42 The relevant sections only covered the AU itself and other international organisations.43 The SCA, for its part, stated that there ‘is little that can be added to that reasoning’.44 Before delving into the reasoning of the SCA, it is useful to restate the contents of the relevant section of the host country agreement. Article VIII of the host country agreement provides, in part, that the45
government shall accord the members of the Commission and staff members, the delegates and other representatives of inter-governmental organisations attending the meetings the privileges and immunities set forth in sections C and D, articles V and VI of the General Convention on the Privileges and Immunities of the OAU.
The SCA interpreted this provision as excluding Al Bashir from the scope of article VIII on several grounds. First, the Court found that a head of state was not a delegate but rather ‘an embodiment of the state itself’.46 Apparently this conclusion was based on the description of the Assembly on the website of the AU as the AU’s ‘supreme organ’ and comprising of ‘Heads of State and Government from all member states’.47 According to the Court, because the AU is composed of members states, and the Assembly is its governing body, and heads of state constitute the Assembly, heads of state are, therefore, ‘the embodiment of members states [and] not delegates from them’.48 Moreover, the Court stated that ‘there [was] no basis’ for concluding that heads of state are included in the reference to ‘delegates’ in article VIII. Second, according to the Court, there was ‘nothing to indicate that the AU was representing the heads of member states or their delegations in concluding’ the host country agreement. The key phrase, according to the judgment, was ‘delegates and other representatives of inter-governmental organisations’.49 This, according to the Court, ‘relates only to persons who are there because of their entitlements to be there on behalf of one or other inter-governmental organisation’.50 Heads of state were, accordingly, not covered by the provision.
While the Court spent time describing the architecture of the Rome Statute, such as the jurisdictional provisions and the duty to co-operate under the Rome Statute, the main interpretative issue for the Court concerned the much-debated tension between articles 27 and 98.51 The Court recognised that article 98 (and its relationship to article 27) had ‘occasioned much debate’.52 It then proceeded to give the two opposing views, namely, the view that article 98 ‘operates to protect party states from the obligation to co-operate’ in cases involving, inter alia, heads of non-state parties.53 The provision, therefore, ‘provides a justification’ for non-co-operation with respect to the arrest and surrender of Al Bashir.54 The other view, the Court stated, was that the fact that the ICC’s jurisdiction in the situation in Darfur was pursuant to a UN Security Council resolution that has the effect that ‘article 27 is made applicable to the non-party state and, therefore, it is not open to it to rely on article 98’.55 Having set out these two opposing views, the Court proceeded to conclude that the tension between articles 27 and 98 ‘has not as yet been authoritatively resolved’.56 Specifically regarding the Security Council-related argument, the Court simply noted that the relevance of the Security Council referral was ‘hotly contested by the commentators’.57 The Court thus offered no resolution to the conflict. Finally, the Court determined that ‘South Africa is bound by its obligations under the Rome Statute to co-operate with the ICC’ to arrest persons under arrest warrants, such as Al Bashir.58
Having referred to the tension between articles 27 and 98, the Court considered the rule of customary international law relating to the immunity of sitting heads of state.59 In particular, the Court considered ‘the suggestion’ that there was an exception under customary international law to immunity where Rome Statute crimes were concerned. In considering the question of whether there was an exception to the rule on sitting heads of state immunity, the Court referred to the Arrest Warrant case, 60 and decisions of national courts as gleaned from the literature.61 The Court then drew a distinction between immunity before international courts - which has generally not applied - and immunity before foreign domestic authorities for the purposes of co-operation with the ICC.62 Based on the views of ‘a number of commentators’, the Court noted that the fact that an individual does not have immunity before the ICC ‘does not necessarily mean that a state is entitled to ignore head of state immunity when requested to co-operate with the ICC to bring such person before it’.63 The Court, having considered decisions of the European Court of Human Rights, 64 concluded that there was no exception to the customary rule on immunity, both with respect to civil claims and criminal prosecution, even in connection with jus cogens crimes.65
At various places, the Court made it clear that it would rather have found that there was an exception under customary international law.66 However, it found that its role did not permit it to go beyond existing customary international law, because its role was ‘one of discerning the existing state of law’ and that the ‘[d]evelopment of customary international law occurs in international courts and tribunals, in the content of international agreements and treaties and by its general acceptance’ by states.67 In the Court’s view, while it may be tempting, it was not permissible for domestic courts to seek to expand the boundaries for customary international law - this, it stated, was the role of international courts.68
Having considered these international law issues, the Court made the following concluding remarks:69
Ordinarily, that would mean that President Al Bashir was entitled to inviolability while in South Africa last June. But SALC argued that that the position was different as a result of the enactment of the Implementation Act.
Essentially, it seems to me, the Court did the analysis of international law merely for the sake of completeness.70 The rules of international law, in the view of the Court, did not affect the outcome. The Court then proceeded to resolve the matter by resolving an apparent conflict between the Implementation Act and Immunities Act. The Court considered that71
when South Africa decided to implement its obligations under the Rome Statute by passing the Implementation Act, it did so on the basis that all forms of immunity, including head of state immunity, would not constitute a bar to the prosecution of international crimes in this country or to South Africa co-operating with the ICC ...
This conclusion, it appears, is irrespective of the position under international law.72
As mentioned above, the purpose of the article is not to determine the correctness or not of the judgment. The purpose is to assess the Court’s methodological approach to the identification and interpretation of international law. While in the current case, the Court suggested that international law issues are not material to the resolution of the primary question before it, the analysis below aims to show that the question before the Court cannot fully be answered without a rigorous approach to international law. It is for this reason that a methodological approach to the identification and interpretation of international law is so important.
Before addressing the Court’s approach to the interpretation of the host country agreement, a few preliminary points about the arguments presented before the Court should be noted. According to the Court, counsel for the government argued that article VIII of the host country agreement had been ‘promulgated’ because ‘Sudan had requested that [Al Bashir] be afforded immunities of a delegate attending the AU’.73 If this is indeed what counsel for the government argued, then counsel was mistaken. As stated above, the immunities provision is a standard provision in any host country agreement, and the AU insists (as do other international organisations) on it as a precondition for hosting any conference in the territory of any state. It was not inserted into the agreement for the individual benefit of Al Bashir, nor would it be correct to assume that had Al Bashir not been coming, the provision would not have been inserted. Second, it appears that counsel for the government argued that the host country agreement, together with the Minister’s notice, gave Al Bashir immunities in domestic law. In the view of the author, the better argument would be that the Immunities Act, in particular section 6,74 enabled the Minister of International Relations and Co-operation to incorporate the host country agreement into domestic law by means of a notice. As Dugard points out, an ‘enabling Act of Parliament may give the executive the power to bring a treaty into effect in municipal law by means of a proclamation or notice in the Government Gazette’.75 Thus, what grants the participants of the Summit immunity in domestic law is the incorporation of the host country by the Immunities Act.76 At any rate, whatever effect this approach would have would depend ultimately on the interpretation given to the host country agreement, to which I now turn.
The Court’s interpretation of article VIII rests on two pillars, both of which are based on an interpretation of the words of article VIII. Under the first pillar, heads of state are the ‘embodiment of the member state itself’ and, therefore, cannot be ‘delegates’.77 It appears that the Court arrived at this conclusion based on the definitions in the AU Constitutive Act, in which the phrase ‘member state’ is defined as ‘member state of the AU’. Because the AU, so the argument goes, ‘is composed of the heads of state and government or their duly accredited representatives’, heads of state are an embodiment of member states and are not its delegates.78 With respect, this argument is ill-conceived. The AU is not composed of the Summit. The Summit is an organ of the AU, the highest policy-making organ, but certainly not the only organ of the AU. It, being the AU, therefore, is not composed of heads of state. It is the Summit that is composed of heads of state, not the AU. The AU is composed of several organs, in addition to the Summit, including the Executive Council (the organ comprising the Minister for Foreign Affairs), the Permanent Representative Committee (the organ comprising ambassadors accredited to the AU), the Commission (the Secretariat of the AU) and the Specialised Technical Committee (various organs of Ministers in specialised areas such as agriculture and justice).79 However, at any rate, even if the Summit were the AU, it would not follow that its heads of state were not delegates because the Summit is not merely a collection of its members, but an entity separate from its individual members.
The second pillar of the literal interpretation advanced by the Court does, however, have some basis in the text. The second pillar, which in fact is the interpretation advanced by the High Court,80 is based on the meaning of the words in article VIII, namely, that immunities are to be granted to ‘members of the Commission and staff members, the delegates and other representatives of inter-governmental organisations attending the meetings’. In the view of the Court, this did not include delegates from AU member states. The word ‘delegate’ in article VIII, in the view of the Court, applied only to representatives of other inter-governmental organisations. Although the SCA did not explain why this was the case, purely based on the language, it was a plausible interpretation.81 In short, the word ‘other’ before ‘representatives of inter-governmental organisations’ must mean that the word ‘delegates’ also qualifies inter-governmental organisations, so that ‘delegates’ are a species of ‘representatives’. Thus, according to this interpretation, it is delegates of the inter-governmental organisations and representatives of inter-governmental organisations that are covered by this provision.
This, of course, is a plausible literal interpretation, although a contrary interpretation is equally plausible as the text is ambiguous. The word ‘delegate’ could qualify inter-governmental organisations, in which case the Court’s interpretation would have been correct, or it could be self-standing, in which the case the Court’s interpretation would have been incorrect. At any rate, the Court’s interpretation is open to at least two criticisms. The first criticism, also based on the meaning of the words, is that, in practice, while ‘representatives’ is a species of ‘delegates’, the reverse is not true. All members of a delegation are ‘delegates’, but only the more senior members acquire the title of ‘representative’. Thus, if the text read ‘representatives and other delegates’, then, under a purely literal interpretation, one could argue that both representatives and delegates qualify inter-governmental organisations. However, since not all delegates are representatives, ‘delegates and other representatives of other inter-governmental organisations’ cannot mean delegates of inter-governmental organisations and their representatives. However, this flaw in the literal interpretation of article VIII is not the most serious methodological flaw.
From a methodological perspective, the Court’s interpretation relied almost exclusively on a textual approach and did not account for the other, obligatory, means of interpretation. In particular, the interpretation did not consider, as part of the context, the reference to the General Convention on the Privileges and Immunity of the OAU (General Convention), which makes it clear that the immunities it seeks to give are principally for ‘representatives of member states’. It is perhaps true, as the Court noted, that this could not be a form of incorporation of the General Convention into the host country agreement, since it is the type of immunities in the General Convention that is incorporated and not the beneficiaries of the immunities. Nonetheless, the General Convention does provide a context, which methodologically the Court ought to have taken into account, in shedding light on the persons who normally enjoy immunities at AU conferences. Moreover, since the text in article VIII is a standard text, the Court could have considered how it has been understood in the past, again with a view to determining who is ordinarily entitled at AU meetings.82 Subsequent practice, including the ministerial notice, reflecting how South Africa, a party to the agreement, understood the host country agreement, can also not simply be ignored. The Court noted that the Minister of International Relations and Co-operation and the Cabinet interpreted the host country agreement as conferring immunities on Al Bashir and other delegates. Yet, the Court concluded that this ‘erroneous belief [on the part of South Africa concerning what the content of the host country agreement is] is neither here nor there’.83 But this dismissive statement reflects a lack of understanding of basic principles of interpretation of international law. It ignores the fact that, under international law, subsequent practice, as defined above, is an element of interpretation that must be taken into account. The Court ought to have considered whether it established the agreement of the parties, that is, whether both South Africa and the AU shared this interpretation. If it appeared that the AU adopted a similar interpretation, the Court ought to have considered this as a subsequent practice establishing the agreement of the parties, and thus given it considerable weight in the interpretation. At any rate, even if the practice did not establish the agreement of the parties, the Court ought to have considered it as a supplementary means of interpretation under article 32 of the Vienna Convention.84
The purely literal interpretation advanced by the Court also ignores the rules of interpretation because it accepts, without pause, an interpretation leading to a ‘manifestly absurd and unreasonable’ result. According to the Court’s interpretation, representatives of observers, such as the EU, UN, ECOWAS and other inter-governmental organisations which have no official role in the Summit are given immunity, but the main participants of the conference are left out of the immunity regime. How can this be? It may be argued that covering heads of State is unnecessary since they would be covered by customary international law, but this is an unconvincing argument. First, other practice indicates that host country agreements cover persons ordinarily covered by other rules of international law. The UN Immunities and Privileges Agreement, for example, covers persons who would enjoy immunities under customary international law, since diplomats attending UN conferences have immunities under customary international law. Again, the application of context and object and purpose ought to have led the Court to consider how previous host country agreements have been understood. The fact is that, as a rule, without exception, host country agreements bestow immunities on the participants of the conference and where, for some or other reason, there is no host country agreement, the host state unilaterally ensures such immunity in terms of its own domestic law. The statement of Switzerland before the General Assembly in 2013 is instructive in this regard: 85
Although ... many international conferences are held under the auspices of an international organisation with which a host state has concluded an agreement, not all of them are: When a conference is not connected with an international organisation, the host state often has to confer privileges and immunities on the conference and its participants unilaterally, based on national law. In this regard, we would like to underline that the privileges and immunities ... are based on international law, even though they may be formalised by a unilateral decision.
Several points, relevant as context to the interpretation of the host country agreement, arise from this Swiss statement. First, it is clear from the statement that immunities are, in the view of Switzerland, as a matter of practice primarily conferred on the conference and the participants of the conference. Second, it is clear that these immunities are, in its view, in accordance with international law. The view expressed by the Swiss delegation is uncontroversial and reflects general practice and law. There is simply no host country agreement that confers immunities on non-participants to the exclusion of the main participants. As such, under article 31(3)(c) of the Vienna Convention, this ought to have been taken into account as an applicable principle of international law. These considerations, at the very least, ought to have played some role in the Court’s interpretation of article VIII of the host country agreement. At any rate, the Court could have made an effort to source other examples of practice to contribute to the interpretation of the agreement to determine whether, in practice, host country agreements bestow immunities on the participants of the conference or on observers to the conference.
As stated above, the arguments concerning customary international law and the duty under the Rome Statute to co-operate are, in fact, intertwined. Simply put, the government’s argument is that there is no duty under the Rome Statute to co-operate with the ICC because article 98 of the Rome Statute creates an exception, based on customary international law, to the duty to co-operate. This argument, thus, required both an interpretation of article 98 and the search for the content of customary international law.
I begin with the Court’s methodological approach to the identification of customary international law. I am, in general, in agreement with the substantive conclusions of the Court, namely, that there is no exception under customary international law, as it currently stands, to the duty to respect the immunity of a sitting head of state.86 I am also in general agreement with the methodological approach of the Court. The Court came to its conclusion having considered state practice in the form of national judicial decisions,87 and decisions of international courts in the form of the International Court of Justice and the European Court of Human Rights.88 There is, of course, more direct practice that the Court could have referred to, including statements by states before the General Assembly, for example, in response to the International Law Commission’s work on immunities.89 A consideration of the debate of the General Assembly would have revealed that states generally agree that heads of state enjoy immunity for all acts for the duration of their terms in office.90 A study of the responses of states may, however, also have shown that there are some states that hold the view that international law should - rather than does - recognise some exceptions even for immunity ratione personae.91
The potential for evolution of the law on immunities, as evidenced by the call by some states for development, puts the spotlight on a comment by the Court concerning the role of courts generally in the development of customary international law. The Court stated that it was not for it to develop customary international law because the ‘[d]evelopment of customary international law occurs in international courts and tribunals, in the content of international agreements and treaties and by its general acceptance’ by states.92 First, the role of international courts and tribunals is not to develop customary international law. In fact, national courts, whose decisions constitute state practice, are more empowered to develop customary international law than international courts. While international courts, in particular the International Court of Justice, are in a better position to judge the state of existing customary international law, national courts contribute directly to its development.93 There are two implications flowing from this distinction. First, the Court understated its own importance for the development of customary international law. Second, the contribution that domestic courts can make to the development of international law underscores the importance of ensuring that decisions of national courts, including South African courts, are premised on the correct assessment of the law. Where the Court deviates from existing international law, it should be done because domestic law requires it to - in other words, it should be a conscious decision and not based on an incorrect assessment of customary international law. A careful and rigorous application of the methodology of the identification of customary international law, as in the current case, will contribute to this end.
While the Court’s approach to customary international law was satisfactory, the Court’s approach to the interpretation of article 98 was less so. Article 98 of the Rome Statute is critical to resolving the tension between immunities under customary international law and the duty to co-operate under the Rome Statute. There are various interpretative models that have been used by the Court to resolve this tension. The Court identified two. These are, first, that the UNSC makes Sudan akin to a state party to the Rome Statute, such that article 98 is inapplicable to Sudan and, second, that by virtue of article 98, the customary international law rule on the immunity of sitting heads of state nullifies the duty to co-operate under the Rome Statute. There is a third textual interpretation which has been advanced by the author, namely, that on its terms, article 98 is limited to very specific types of immunity.94 The Court, however, declined to engage in any sort of interpretation, either of article 98 or of the Security Council resolution referring the situation in Darfur to Sudan. The Court only stated that ‘there is a tension between article 27 and 98 that has not as yet been authoritatively resolved’.95 The function of the Court, however, was to resolve all legal issues necessary to address the controversy before it. The tools of interpretation, as described above, could have assisted the Court to address the tension by interpreting the provisions of the Statute.
These tools of interpretation include the text of articles 98 and 27, the context and the object and purpose of the Statute. For example, the Court ought to have considered, as part of the text and context, that articles 27 and 98 address different things. While article 27, which removes immunity, clearly is concerned with the relationship between an accused and the ICC itself, article 98 is concerned with the relationship between state parties and non-parties, on the one hand, and the relationship between the Court and state parties, on the other. Thus, in addressing the question whether Al Bashir was owed immunity under the Rome Statute, the Court ought to have considered the forum. This might suggest that article 98, rather than article 27, is applicable to the dispute. That article 98, and not article 27, applies to the situation of a person under an ICC arrest warrant before South African courts does not necessarily mean that there is no duty to arrest Al Bashir under the Rome Statute. Whether this is the case would ultimately depend on the interpretation of article 98 itself. Elsewhere it has been suggested that the text of article 98, in its context, together with the object and purpose of the Rome Statute, appears to exclude the scope of article 98, the immunities of heads of state, and is rather limited to immunities of diplomats and the immunities of the state itself.96 Time and space do not permit the full reproduction of the arguments here, but it suffices to say that they are based on the following considerations:97
(ii) International law recognises a distinction between different types of immunities, namely, state immunity, diplomatic immunity and heads of state immunity.98
While there is little in the way of preparatory works to the Rome Statute,99 the little that there is suggests that article 98 ought to be interpreted restrictively.100 The various texts provided to the Rome Conferences (and discussed there) suggest that the two main options were to have unlimited exceptions or restricted exceptions to the duty to co-operate. It appears that the Conference decided on a restricted interpretation and, in particular, that any exceptions should be specifically enumerated.101 Again, this does not suggest that article 98 does not, as a matter of law, protect Al Bashir. This simply points to considerations that ought to have been taken into account in the process of interpreting article 98.
There is yet another argument, advanced by the Southern African Litigation Centre (SALC), the effect of which would be that article 98 would be inapplicable to Sudan. Sudan is a party to the Genocide Convention. SALC had, in this context, argued that, since the Genocide Convention removes immunity, and since both South Africa and Sudan are parties to the Genocide Convention, the customary international law rule of immunity does not apply between South Africa and Sudan in respect of genocide.102 The Court gave this argument short thrift, stating that it had not received sufficient submissions to give it due consideration. While I do not share the interpretation by SALC that the Genocide Convention removes immunity, the Court ought to have considered this argument as an applicable principle of international law under article 31(3)(c) of the Vienna Convention on the Law of Treaties. Whether, as argued by SALC, the Genocide Convention removes immunity can only be determined by the application of the rules of interpretation, which surely the Court must do in the exercise of its judicial functions.
All the tools of interpretation discussed above could also be usefully applied to the interpretation of the relevant UN Security Council resolution.103 The application of the normal rules of treaty interpretation could have led the Court to the ordinary meaning of the resolution, the object and purpose not of the Rome Statute, but of the resolution, as well as subsequent practice in the form of the Security Council debate on the report of the prosecutor.
The Court justified its decision not to address the tension by stating that ‘in view of [its] conclusion on the effect of the Implementation Act it is unnecessary to address these submissions’.104 The Court essentially concluded that the Implementation Act trumped all, such that the resolution of international law questions was not critical for the Court to come it its conclusion. Presumably this reasoning could have applied equally to the question of customary international law and possibly the interpretation of the article VIII of the host country agreement. Moreover, for reasons discussed below, it is not correct that because of the conclusions on the Implementation Act, the resolution of the tension between articles 27 and 98 can be dispensed with. The failure of the Court to resolve, at least for itself, the article 98 conundrum results in an internal incoherence in the judgment.
There are at least two reasons why the resolution of the international law issues was important, if not critical, to the interpretation and application of the domestic legislation in question. First, section 233 of the South African Constitution provides that in the interpretation of any legislation, including the Implementation Act and Immunities Act, any reasonable interpretation that is consistent with international law must be preferred over any other interpretation. This provision can only be applied if ‘international law’ on the matter is identified and interpreted. The provision does not call for an interpretation that is consistent with ‘a source’ of international law, but rather an interpretation that is consistent with international law. Section 233 cannot be applied unless the position under international law relating to immunities has been determined, and this cannot be done without a resolution of the tension between articles 98 and 27. In this regard, the author has, prior to the Al Bashir situation, made the following observation, commenting on the importance of interpreting (and not just referring to) international law:105
However, if international law is to be relied upon for the interpretation of the Constitution and legislation, it is reasonable to expect that international law itself would be interpreted, for how can international law be relied upon without first finding its meaning? Finding the meaning of international law requires its interpretation through the application of rules of interpretation. Thus, while these interpretive provisions do not directly call for the interpretation of international law, there is an indirect requirement, or at the very least an expectation, that international law will be interpreted.
However, even more important than the significance of international law for the purposes of section 233, it is not clear how the domestic law questions can be addressed without a resolution of the international law questions. The Implementation Act implements the Rome Statute. The obligation to co-operate under the Implementation Act, by definition, is dependent on whether a duty exists to co-operate under the Rome Statute. The SCA, however, without addressing the article 98 issue, stated that ‘South Africa is bound by its obligations under the Rome Statute’ and that it ‘is obliged to co-operate with the ICC and to arrest and surrender to the Court persons in respect of whom the ICC has issued an arrest warrant and a request for assistance’.106
The conclusion that there is a duty under the Rome Statute to arrest Al Bashir is important even for the finding that there is a duty under the Implementation Act. Presumably, if the ICC is not entitled to make a request for arrest and surrender because of the exception in article 98, then South African courts, being courts based on the rule of law, cannot order the arrest and surrender of Al Bashir. How then did the Court come to the conclusion that there was a duty under the Rome Statute without resolving the tension between articles 27 and 98? After all, the argument of the government was that there was no duty to arrest and surrender because of article 98 of the Rome Statute. In other words, the customary international law argument relates to the interpretation of article 98. The Court appeared to agree with the government that there was a duty under customary international law not to arrest Al Bashir. Surely this means that article 98 applied. If that were so, then there was no duty to arrest Al Bashir under the Rome Statute. The Court could not determine whether there was a duty under the Rome Statute to arrest Al Bashir unless it engaged with the various issues relating to article 98, including the possible effects of the UN Security Council and the textual limitations of article 98.
This internal incoherence - finding that there is a duty to arrest and surrender without resolving the article 98 debate - cannot be remedied by simply stating that the Implementation Act trumps all. This is because for arrest and surrender under the Implementation Act, there has to be a duty under the Rome Statute. However, this internal inconsistency may have another far more insidious effect. The effect of the Court’s decision is that South Africa is obliged to arrest and surrender a sitting head of a non-state party even in the absence of a UN Security Council resolution as the basis for the referral - a decision that would be inconsistent with the ICJ’s decision in the Arrest Warrant case and even the ICC Pre-Trial Chamber in the DRC non-co-operation case, since in those decisions the ICJ and the ICC respectively found that sitting heads of state enjoyed absolute immunity from the jurisdiction of foreign courts.107 Moreover, given the Constitutional Court’s judgment that the Implementation Act obliges South Africa to investigate Rome Statute crimes even where the perpetrator is not in South Africa, subject to certain limitation principles,108 the current decision effectively means that South Africa can investigate sitting heads of non-state parties contrary to international law - and potentially prosecute them should they ever be present on South African soil. Akande, on whom the Court relied heavily, commenting on the judgment, makes the following observations about the judgment:109
[T]hese conclusions regarding the lack of immunity ratione personae in South Africa ... are odd [and] it is not clear why section 4(2) [of the Implementation Act] is not given a meaning which aligns more closely with the words used and with customary international law.
The oddity that Akande identifies is directly linked to the Court’s decision not to address the article 98 argument. Akande, who supports the view that the UN Security Council referral turns Sudan into an analogous state party,110 although he agrees with the conclusion of the Court that there are no exceptions under customary international law to incumbent head of state immunity, states that ‘it was wrong for the Court’ not to address the article 98 tension.111
I do not mean to suggest that the Court ought not to have reached the conclusion it reached - that is a substantive point which falls beyond the scope of the article. The pathways for reaching the same outcome while applying a sound methodological approach to the issues are numerous. I do, however, wish to point to the importance of resolving the international law issues in the case. This in turn, requires attention to the methodological issues raised above which, save in connection with questions of customary international law, the Court did not pay attention to.
The South African legal community prides itself on having an international law-friendly framework. Our Constitution has several provisions enjoining the consideration (and sometimes application) of international law. Our jurisprudence is similarly replete with references to international law materials. 112 Yet, South African courts have, generally, struggled with the methodological questions of the interpretation and identification of international law.113 The result has been superficial references to international law.114 For the most part, this has not harmed the judicial reasoning accompanying these cases as, in most instances, the Court was concerned with the interpretation of domestic law, such that ensuring a rigorous approach to international law might be seen as secondary.
However, South African courts are more often faced with cases in which the central issues revolve around international law. In these cases, international law cannot be treated as secondary, and attention to the details and methodology of international law is crucial. In the Al Bashir case, the approach of the SCA, first of all, was inconsistent. While it adopted a rich and somewhat rigorous approach to the identification of customary international law, on the one hand, on the other it adopted a superficial, a contextual interpretation of the host country agreement and decided not to resolve the key international law question at issue in the case, namely, the interpretation of article 98 of the Rome Statute. The Court also seemed to adopt the approach that international law was secondary. It is for this reason that, having gone through a careful analysis of customary international law, it proceeded to ignore international law in its assessment of domestic law. This is also the reason that the Court decided that it was unnecessary for it to resolve the tension between articles 27 and 98 of the Rome Statute. Much will be written about the substantive aspects of the judgment, but there is a need to pause and think about the methodology of international law in this and other South African cases.
4. See, eg, H Strydom & K Hopkins ‘International law’ in S Woolman et al (eds) Constitutional law of South Africa: Vol 2 (2013). See also N Botha ‘Justice Sachs and the interpretation of international law by the Constitutional Court: Equity or expediency?’ (2010) 25 Southern African Public Law 253; N Botha ‘Rewriting the Constitution: The “strange alchemy” of Sachs indeed!’ (2009) 34 South African Yearbook of International Law 253. A notable exception is N Botha ‘Interpreting the International Child Abduction Act 72 of 1996’ (2003) 28 South African Yearbook of International Law 330. See also D Tladi ‘Interpretation of treaties in an international law-friendly framework: The case of South Africa’ in HP Aust & G Nolte (eds) The interpretation of international law by domestic courts: Uniformity, diversity, convergence (2016).
5. Southern African Litigation Centre v Minister of Justice and Constitutional Development & Others 2015 (5) SA 1 (GP) (SALC v Minister of Justice); Minister of Justice and Constitutional Development & Others v Southern African Litigation Centre 2016 (4) BCLR 487 (SCA) (Minister of Justice v SALC).
6. MJ Ventura ‘Escape from Johannesburg? Sudanese President Al Bashir visits South Africa, and the implicit removal of head of state immunity by the UN Security Council in light of Al Jedda’ (2015) 13 Journal of International Criminal Justice 995; D Tladi ‘The duty on South Africa to arrest and surrender President Al Bashir under South African and international law: A perspective from international law’ (2015) 13 Journal of International Criminal Justice 1027; E de Wet ‘The implications of President Al Bashir’s visit to South Africa for international and domestic law’ (2015) 13 Journal of International Criminal Justice 1049. See also L Kohn ‘The Bashir judgment raises the red flag about the rule of law and the judiciary’ (2016) 133 South African Law Journal 246. See further JD van der Vyver ‘The Al Bashir debacle’ (2015) 15 African Human Rights Law Journal 559. See, particularly on Minister of Justice v SALC (n 5 above), D Akande ‘The Bashir case: Has the South African Supreme Court abolished immunity for all heads of states?’ EJIL:TALK! 29 March 2016.
7. These cases of non-arrest of Al Bashir have resulted in several decisions, including Decision Pursuant to Article 87(7) on the Failure of Republic of Malawi to Comply with the Co-operation Request Issued by the Court With Respect to the Arrest and Surrender Omar Hassan Ahmed Al Bashir: The Prosecutor v Al Bashir (ICC-02/05-01/09), Pre-Trial Chamber I, 12 December 2011; Décision Rendue en Application de l’article 87(7) de la Statut de Rome concernant le refus de la Républic du Tchad d’accédur aux demandes de coopération délivrées par la Cour Concernant l’arrestation et la Remise d’Omar Hassan Ahmad Al Bashir: L’Procureur v Al Bashir (ICC-02/05-01/09), Pre-Trial Chamber I, 13 December 2001; Decision on the Co-operation of the Democratic Republic of the Congo Regarding Omar Al Bashir’s Arrest and Surrender to the Court: The Prosecutor v Omar Hassan Ahmad Al Bashir (ICC-02/05-01/09), Pre-Trial Chamber II, 9 April 2014.
9. See, eg, A Aust Handbook of international law (2005) 6-7; H Thirlway ‘The sources of international law’ in MD Evans International law (2014) 98; I Browlie Principles of international law (2003) 6; P Malanczuk Akerhurst’s modern introduction to international law (1997) 39; J Dugard International law: A South African perspective (2005) 29.
11. See Report of the International Law Commission, 68th Session (2 May-10 June and 4 July-12 August 2016), Supplement 10 (A/71/10) ch V, Identification of Customary International Law, Text of the Draft Conclusions on the Identification of Customary International Law adopted by the Commission on First Reading. Draft Conclusion 2, contained in Part Two titled ‘Basic Approach’, provides as follows: ‘To determine the existence and content of a rule of customary international law, it is necessary to ascertain whether there is a general practice that is accepted as law (opinio juris).’
12. See M Wood First Report of the Special Rapporteur on Formation and Evidence of Customary (A/CN.4/663) 17 May 2013 para 13: ‘In the view of the Special Rapporteur ... the aim of the topic is to offer some guidance to those called upon to apply rules of customary international law on how to identify such rules in concrete cases. This includes, but is not limited to, judges in domestic courts, and judges and arbitrators in specialised international courts and tribunals.’
23. See art 34 of the 1969 Vienna Convention on the Law of Treaties, which provides that ‘[a] treaty does not create either rights or obligations for a third state’. Similarly, art 26 of the Vienna Convention, which establishes the basis for the binding nature of a treaty, provides that ‘[e]very treaty in force is binding upon the parties to it ...’
24. See generally Report of the International Law Commission at its 58th Session, 2006, Supplement 10 (A/61/10), Conclusions of the Work of the Study Group on Fragmentation of International Law: Difficulties Arising from the Diversification and Expansion of International Law. See, in particular, Conclusion 2, paras 5-10 for the lex specialis rule and Conclusion 5, paras 24-30, for the lex posterior rule.
25. Compare Tladi (n 4 above) 145, fn 48, where the view is expressed that art 31(1) constitutes the primary rule and the other paragraphs of art 31 serve to contribute to the process of art 31(1). A contrary view, namely, that the other elements of art 31 have the same weight as the rule in art 31(1), is also presented but not supported.
30. In Accordance with International Law of the Unilateral Declaration of Independence in Respect of Kosovo, Advisory Opinion of 22 July 2010, 2010 ICJ Reports 403 para 94, the Court stated that the Vienna rules ‘may provide guidance’ in the interpretation of Security Council resolutions.
31. In the course of the interpretation of UN Security Council Resolution 1244 (1999), the Court in the Kosovo Opinion (n 30 above) relies on the means of interpretation in Vienna rules such as ‘text’ and ‘object and purpose’. See eg para 98, where the Court refers to ‘the text of Resolution 1244 (1999)’, and para 100, where the Court refers to the ‘object and purpose of the resolution’.
34. The ‘peculiarities’ referred to by the Court could just as easily apply to treaties. Eg, treaties themselves may be adopted by a vote; treaties may be adopted by a body such as the General Assembly; and treaties are the product of intense negotiations just as is the case with resolutions.
38. The precise language of art 27(2) of the Rome Statute is: ‘Immunities or special procedural rules which may attach to the official capacity of a person ... shall not bar the Court from exercising its jurisdiction over such a person.’ Art 98(2) provides that the ‘Court may not proceed with a request for surrender or assistance which would require the requested state to act inconsistently with its obligations under international law with respect to state or diplomatic immunity of a person or property of a third state unless the Court can first obtain the co-operation of that third state for the waiver of the immunity’.
39. Agreement between the Republic of South Africa and the Commission of the African Union on the Material and Technical Organisation of the Meeting of the 30th ordinary session of the Permanent Representative Committee from 7 to 9 June 2015, the 27th ordinary session of the Executive Council from 10 to 12 June and the 25th ordinary session of the Assembly from 14 to 15 June (on file with author).
51. See further on the debate, in addition to the sources cited in n 6, D Akande ‘The legal nature of the Security Council referrals to the ICC and its impact on Bashir’s immunities’ (2009) 7 Journal of International Criminal Justice 333; P Gaeta ‘Does President Al Bashir enjoy immunity from arrest?’ (2009) 7 Journal of International Criminal Justice 315; W Schabas ‘Introductory remarks: Annual Ben Ferencz Session’ (2012) 106 American Society of International Law Proceedings 305; D Tladi ‘The ICC decisions in Malawi and Chad: On co-operation, immunities and article 98’ (2013) 11 Journal of International Criminal Justice 199; D Tladi ‘Immunity in the era of “criminalisation”: The African Union, the ICC and international law’ (2015) 58 Japanese Yearbook of International Law 17.
61. In particular, the Court states that the principle that there is no exception as determined in the Arrest Warrant case ‘has been widely accepted by national courts which have rejected attempts to implead sitting heads of state’. For this proposition, the Court refers to MA Tunks ‘Diplomats or defendants? Defining the future of head of state and sovereign immunity’ (2002) 52 Duke Law Journal 651; and T Weatherall ‘Jus cogens and sovereign immunity: Reconciling divergence in contemporary practice’ (2015) 46 Georgetown Journal of International Law 1151.
70. See Separate Opinion of Ponnan JA (Lewis JA concurring) in Minister of Justice v SALC (n 5 above), para 115 (‘with due deference to my learned colleague, that conclusion, I daresay, renders his discussion on customary international law unnecessary’).
72. Eg, having considered that customary international law requires, without exception, the respect of immunity of heads of state, the Court holds that ordinarily this would mean that South Africa may not arrest him (Minister of Justice v SALC (n 5 above) para 85), and proceeds to find that there is a duty to arrest him. Similarly, having found that there is a tension between involving the rule of customary international law on immunities, art 27 of the Rome Statute, the duty to co-operate in the Rome Statute and art 98, the Court states that ‘in view of my conclusion on the effect of the Implementation Act it is unnecessary to’ resolve the conflict (Minister of Justice v SALC (n 5 above) para 106).
74. Sec 6(1) of the Immunities Act provides as follows: ‘The officials of ... any organisation, and representatives of any state, participating in an international conference or meeting convened in the Republic enjoy, for the duration of the conference or meeting such privileges and immunities as ... are specifically provided for in any [international] agreement for this purpose.’ Sec 6(2), in its part, provides that the ‘Minister [of International Relations and Co-operation] must by notice in the Gazette recognise a specific conference or meeting for the purposes of subsection (1)’.
76. An interesting legal question falling beyond the scope of this article is whether the lex posterior rule of interpretation would have the effect that the incorporation of the host country agreement takes priority over the Implementation Act since the former would be the later legislative act.
82. See, as examples of similar texts, art 13 of the Agreement between the Government of the Republic of South Africa and the Commission of the African Union (AU) on the Material and Technical Organisation of the Ministerial Conference on the African Diaspora, 16-18 November 2007, Pretoria, South Africa; and art 11 of the Agreement between the African Union Commission (AUC) and the Government of the Republic of South Africa on the Material and Technical Organisation of the 3rd Africa-India Trade Ministers Conference and the 2nd Africa-India Business Council Meetings, 30 September-1 October 2013, Sandton Sun, South Africa. Incidentally, what appears to be the 2013 negotiation comments is mistakenly left in the final text. The comment, presumably from the AU in response to a suggestion for an amendment from the South African side, reads: ‘In keeping with the African Union legal format, we would prefer to keep using this phrase in order to maintain consistency as in previous agreements.’
87. Minister of Justice v SALC (n 5 above) para 67, where the Court refers to various cases in which national courts have ‘rejected attempts to implead sitting heads of state including Prime Minister Sharon of Israel, President Gaddafi of Libya, President Mugabe of Zimbabwe, Prime Minister Thatcher of the United Kingdom, President Castro of Cuba, President Zemin of China, President Kagame of Rwanda and President Arastide of Haiti’.
88. In addition to the Arrest Warrant case (n 60 above), the Court also considers the Case Concerning the Jurisdictional Immunities of the State (Germany v Italy: Greece Intervening), Judgment of 3 February 2012, ICJ Reports (2012) 99; Al-Adsani (n 64 above).
89. See Draft Article 4 of the ILC Draft Articles on the Immunity of State Officials from Foreign Criminal Jurisdiction, ch V of the Report of the International Law Commission on the Work of its 65th Session (A/68/10).
90. See, eg, Statement by the Delegation of the Republic of India on Agenda Item 81 (Report of the International Law on the Work of its Sixty-Fifth Session), at the 6th Committee of the General Assembly, October 2013; statement by Mrs Edwige Belliard, Director of Legal Affairs of the French Republic, on Agenda Item 81 (Report of the International Law Commission on the Work of its 65th Session) at the 6th Committee of the General Assembly, October 2013; statement by the United States of America on Agenda Item 81 (Report of the International Law on the Work of its 65th Session) at the 6th Committee of the General Assembly, October 2013.
91. See statement by Rita Farden, Director of Legal Affairs of the Ministry of Affairs of Portugal, Item 81 (Report of the International Law Commission on the Work of its 63rd and 65th Sessions), 68th Session of the United Nations General Assembly.
98. See for discussion Tladi 231-214. See also dissenting opinion of Judge Oda in the Arrest Warrant case (n 60 above) para 14 and dissenting opinion of Judge Van den Wyngaert para 15. See further Case Concerning Certain Questions of Mutual Assistance in Criminal Matters (Djibouti v France), Judgment of 4 June 2008, ICJ Report 2008 177 para 193. See also judgment of the Court in the Arrest Warrant case where the Court, while recognising that the Vienna Convention on Diplomatic Relations does not contain provisions relevant to immunities of foreign ministers and heads of state. See also I Brownlie Principles of public international law (2003) 322 and S Knuchel ‘State immunity and the promise of jus cogens’ (2011) 9 Northwestern Journal of International Human Rights Law 149.
99. See Tladi ‘Malawi and Chad’ (n 51 above) 217. See also R Cryer et al An introduction to international criminal law and procedure (2010) 148. There are no formal travaux préparatoires for the Rome Statute. Summary UN records (available on the Official Document System of the UN) cover plenary debates only and main committee discussions.
101. As above. See also statement of the Peru Republic stressing that the importance of co-operation necessitated that from the duty to co-operate ‘should be allowed’ (A/Conf.183/SR.3); See also statement of Australia (A/Conf.183/SR.4).
103. D Tladi ‘Interpretation of protection of civilians in United Nations Security Council resolutions’ in D Kuwali & F Viljoen (eds) ‘All means necessary’: Protecting civilians and preventing atrocities in Africa (forthcoming). See, for the application of these principles to UNSC Resolution 1593, Tladi ‘Immunity in the “era of criminalisation”’ (n 51 above) 36 et seq.