Ascertaining Customary International Law: An Inquiry into the Methods Used by Domestic Courts

Based on analysis of a large number of recent domestic court cases on matters of customary international law (2000–2014), this article demonstrates that, rather similar to the International Court of Justice, domestic courts do not normally identify customary norms of customary international law on the basis of the textbook method of ascertaining a general practice accepted as law. Rather, they tend to outsource the determination of custom to treaties, non-binding documents, doctrine or international judicial practice. Sometimes, it appears that domestic courts simply assert, without citing persuasive practice authority, the existence of a customary norm. In rare cases, however, domestic courts do engage in extensive analysis.


Introduction
In the formalist understanding of the sources of international law, 1 a norm becomes a norm of international law insofar as it has been generated by accepted processes of international law formation. The existence of a norm of customary international law, in line with the dominant understanding of custom as a source of international law, is only acknowledged when sufficient evidence of state practice and opinio juris is adduced. 2 There is no agreement, however, on the precise method to be used to determine specific norms of custom, or on how to make sense of disparate evidentiary materials. Reflecting on the method used by the International Court of Justice (ICJ), Stefan Talmon has submitted that 'when determining the rules of customary international law, the ICJ does not use one single methodology but, instead, uses a mixture of induction, deduction and assertion'. 3 In most cases, the ICJ simply asserts or posits the customary norm, without strong evidentiary backing. 4 This concern was also raised by the Asian-African Legal Consultative Organisation (AALCO) Informal Expert Group on Customary International Law. 5 The ICJ's lack of rigorous customary international law-identification methodology has even been admitted (although at the same justified on pragmatic grounds) by a former President of the Court. 6 What this means is that a determination of the law may not depend on the community of states' opinio juris, but on the opinio juris of the ICJ itself. This suggests that the ICJ has considerable agency, if not simple discretion in determining whether a claimed norm enjoys customary status, thus making the process of content-determination of the sources of international law a rather unpredictable endeavour. 7 A similar process can be witnessed in the practice of domestic courts determining norms of customary law. Inevitably, a somewhat messy picture emerges from this practice, but it is striking that in the large majority of decisions, courts offer little in 1 See on formalism: d'Aspremont (2011). 2 Art. 38(1)(b) Statute of the International Court of Justice. See ILC Special Rapporteur (2014), p. 72 ('To determine the existence of a rule of customary international law and its content, it is necessary to ascertain whether there is a general practice accepted as law'). 3  , p. 443. Induction is being understood as a method of inferring the customary norm from an iterative process of state practice and opinio juris. Deduction infers a specific customary norm (e.g., immunity from jurisdiction) from a more general principle (e.g., the principle of sovereign equality). Assertion means that the ICJ neither uses inductive nor deductive reasoning, but simply asserts customary international law. Ibid.,pp. 420,434. 4 Ibid., AALCO IELG, Report of 24 March 2015, available at http://www.aalco.int/54thsession/AALCOIEG% 20Chairman%27s%20Statement%20and%20Special%20Rapporteur%27s%20Report%2020150324.pdf (accessed 20 January 2018). 6 Tomka (2016), p. 5. 7 Compare also Prosecutor v. Kupreškić, 14 January 2000, IT-95-16-T, para. 527, finding a rule of customary international law, citing 'the demands of humanity or the dictates of public conscience', in the face of scarce state practice. Arajärvi (2017), pp. 189-190. the way of evidence supporting the existence of the posited customary norm. 8 In so doing, they largely mimic the ICJ's approach, 9 although offering even less evidence than the ICJ does. 10 In particular, while they duly pay lip-service to the two constitutive elements of custom (state practice and opinio juris), 11 the analysis of state practice-supposedly the backbone of the process of customary norm determination process-is generally shallow, 12 and findings as to the (non-) existence of a customary norm may appear to be of an oracular nature. 13 This article shows, on the basis of a large number of cases, that domestic courts, when determining custom, use other materials as proxies for a thorough analysis of state practice and opinio juris. The article is a comparative categorisation study of the methods used by domestic courts to identify customary international law. The research is methodologically based on grounded theory, a social science-based qualitative research approach, which constructs theory inductively on the basis of data analysis. Grounded theory works on the basis of coding, conceptualizing, and categorizing data with a view to generating dense theory. 14 No specific legal systems or explanatory theories were selected beforehand, as such would have limited the scope of the research. The aim was instead to show the prevailing divergence of identification methods that are globally in use. 15 While it is theoretically possible to examine cases without a specific demarcation in time, for a proper understanding of contemporary identification practices, the authors have examined domestic court decisions rendered between 2000 and 2016. For reasons of practical feasibility as well as to understand contemporary rather than historic court practices, it was decided to limit the sample to cases decided from 2000 onwards.
In social-scientific research, sample size should be large enough to reach theoretical saturation, i.e., a situation where '(a) no new or relevant data seem to emerge regarding 8 See also Stirling-Zanda (2004), p. 3 (submitting that domestic courts tend to offer little evidence of state practice and opinio juris). 9 ILC Special Rapporteur (2013a), para. 84;Stirn (2013) Corbin and Strauss (2015). 15 See for the method of selecting legal systems in light of the aim of the research: Oderkerk (2015), p. 608.
Ascertaining Customary International Law: An Inquiry into… 3 a category, (b) the category is well developed in terms of its properties and dimensions demonstrating variation, and (c) the relationships among categories are well established and validated' 16 or large enough to have sufficient 'information power'. 17 Methodologists do not set a minimum sample size, as this depends on the context and the research question. To answer the research question posed in this article and to increase theoretical validity, it was considered important to identify as many relevant decisions (data) as possible, from as many different jurisdictions as possible. Any domestic case in which the court ascertained customary international law, using one of the methods mentioned below, was considered relevant. Cases have been largely drawn from the databases International Law in Domestic Courts (ILDC) of Oxford University Press, and International Law Reports (ILR) of Cambridge University Press. Basic keywords, such as 'customary international law' or any equivalent or translation thereof were used to identify the cases. The cases were not selected based on their geographical location, but based on their relevance in relation to the keywords. 18 We initially identified 300 cases, of which 60 turned out to be relevant. We cannot exclude the existence of other relevant cases that have not been reported in ILDC or ILR. National reporters for these databases may be inactive, or even non-existent, as a result of which domestic cases relevant to customary international law may not have been reported. Whether ILDC or ILR suffer from reporting bias is not part of our inquiry, however.
The focus of the examination has been placed on the methods or techniques used by domestic courts to identify norms of customary international law, with a view to identifying methodological categories. To this effect, all data, i.e., all domestic court decisions, which were based on customary international law, have been coded by both authors. This led to the categorisation of the data in a number of provisional subcategories, 19 and, after conceptual saturation was reached in the data analysis, in five final categories, i.e., five customary law identification methods used by domestic courts. Obviously, reasonable observers may take issue with this categorisation and take the view that other categories may be more apt to systemize the data. 20 Such alternative views may be entirely legitimate. The authors 16 Corbin and Strauss (1998), p. 212. 17 Malterud et al. (2016Malterud et al. ( ), pp. 1752Malterud et al. ( -1760 Cases were drawn from Australia, Austria, Belgium, Canada, France, Germany, Ghana, Hong Kong, Italy, Kosovo, the Netherlands, Philippines, Poland, South Africa, Singapore, Sweden, the United Kingdom, the United States of America and Zimbabwe. 19 E.g., in an initial phase of the research 'executive statements regarding CIL' and 'simple assertion of CIL' were identified as relevant (sub-)categories of techniques used by domestic courts. These were eventually integrated into broader, final categories. 20 E.g., for slightly different, although overlapping categories capturing the ICJ's method of customary international law-identification: Tomka (2016), p. 24 ('The Court has taken a pragmatic approach to determining the existence and content of international custom, considering whether a rule has been convincingly identified before moving on to consider the primary evidence of State practice and opinio juris. In this regard, four particular methods for the assessment by the Court of evidence of customary law have played an important role: (1) referring to multilateral treaties and their travaux preparatoires, (2) referring to UN resolutions and other non-binding documents which are drafted in normative language, (3) considering whether an established rule applies to current circumstances as a matter of deduction and (4) resorting to an analogy'). considered the chosen categories as being most adequate to capture the practice in domestic courts, and being most useful for pedagogical purposes. As they are the outcome of empirical research, they should by no means be used in a prescriptive manner: this contribution does not, and does not aspire to offer precise guidelines regarding how domestic courts should identify customary international law.
Most of the identified methods draw on other (legal) materials. Indeed, only rarely does a domestic court thoroughly examine relevant state practice and opinio juris at length. 21 Instead, domestic courts tend to draw on multilateral treaties (Sect. 2), authoritative non-binding instruments (Sect. 3), the doctrine (Sect. 4), and international judicial practice (Sect. 5). Exceptionally, domestic courts conduct an extensive analysis of state practice and opinio juris (Sect. 6).
The focus of the article lies on categorising domestic court practices, which contribute to a more systematic understanding of such practices. However, the article also signals, from an evaluative perspective, problematic aspects of techniques applied by domestic courts. The judgmental yardstick used for this evaluative analysis is the requirement, drawn from the classic theory of customary norm formation, that a customary norm evidences 'a general practice accepted as law', 22 which in principle requires a thorough analysis of state practice and opinio juris. 23 The normative perspective in this article is however subordinate to its primary systematisation objective.
This article only addresses the customary international law-identification methods used by domestic courts. It does not as such examine the normative impact which domestic court decisions have had on the formation of customary international law, or on the decisions of domestic courts in other jurisdictions confronted with similar questions. 24 Nor does it carry out a comparative analysis of whether there are any differences in how particular states or groups of states (such as civil law versus common law states) identify customary international law. 21 E.g. Reinisch and Bachmayer (2012), p. 10 (demonstrating, on the basis of an empirical study of post 1945 Austrian judgments, that Austrian courts rarely conduct an extensive analysis of state practice and opinio juris). In fact, only in one case did a court engage in such an analysis. See Dralle v. Republic of Czechoslovakia, OGH 1 Ob 171/50, SZ 1950No. 23/143, 17 ILR 155, 10 May 1950Supreme Court, pp. 157-163. 22 Art. 38(1)(b) Statute of the International Court of Justice. 23 A rigorous application of the two-element approach is also the yardstick suggested by the AALCO Informal Expert Group on Customary International Law. See Yee (2016), p. 37 ('The AALCOIEG agreed that the two-element approach was the proper one but was very concerned about the uneven rigor with which international courts and tribunals applied it in the identification of customary international law rules in their decisions, in identification operations which were sometimes described as ''pragmatic'', ''truncated'', ''incomplete'', ''untidy'' or ''assertive'''). 24 See on this e.g., Roberts (2011), p. 62;.
Ascertaining Customary International Law: An Inquiry into… 5

Multilateral Treaties
Very much like the ICJ, 25 domestic courts may consider widely, or at times less widely 26 ratified multilateral treaties as evidence of the existence of customary international norms (or general principles). Thereby, they dispense with a detailed analysis of state practice or opinio juris. They may do so because they feel more comfortable with written sources of law, 27 or because state parties to treaties already accepted that certain provisions are declaratory of customary international law. 28 Referring to treaties may be problematic, however, insofar as there is no existing law, no 'general practice accepted as law by states' on the subject. This is arguably the case for most treaties. 29 Accordingly, domestic courts will tend to consider treaty provisions as reflecting customary norms in case of widespread ratification of the treaty. 30 The practice of states which ratified the treaty and act in accordance with the latter's provisions, is then considered as evidencing state practice and opinio juris for customary law purposes. 31 A US court has for instance held that 'a treaty will only constitute sufficient proof of a norm of customary international law if an overwhelming majority of States have ratified the treaty, and those States uniformly and consistently act in accordance with its principles'. 32 In another case, a court ruled that, while a number of specially affected states had ratified relevant treaties, as those states did not act in accordance with the provisions of those treaties, they could not be considered as reflecting customary international law. 33 Insofar as treaties have been very widely ratified, and pertain to human rights or international humanitarian law, domestic courts tend to conceive of such treaties as reflecting custom, without submitting elaborate evidence in terms of state practice and opinio juris. Thus, relying on treaty law, domestic courts have held that torture is a violation of the law of nations, 34 and that genocide is a specific intent crime. 35 Courts have satisfied themselves with relying on the Geneva Conventions to the effect that there is a customary international law norm stipulating that opposing sides in a time of armed conflict are to ensure the well-being of the civilian population and to respect its dignity and basic rights, 36 and that a distinction should be made between combatants and civilians, 37 as no one would contest that these are basic principles of humanitarian law. Likewise, the prohibition of slavery is also considered as 'one of the most well-established customary rules'. 38 Given the widespread ratification of the aforementioned relevant treaties, and the fundamental nature of the relevant norms, which may even amount to jus cogens (peremptory) norms, 39 such a law-ascertainment method may appear legitimate. Republic of Equatorial Guinea, 153 ILR 697, 23 January 2008, Zimbabwe; High Court, paras. 703, 711 (referring to international and regional human rights instruments). 35  However, doubts can be raised over the legitimacy of relying on multilateral treaties as evidence of custom when these treaties are not widely ratified, and when state practice is not widespread or even at loggerheads with the text of the treaty. Thus, domestic courts should be cautious to consider the Rome Statute of the International Criminal Court 40 or treaties on state succession as reflective of customary international law. Not all courts have demonstrated such caution. 41 Even more caution is needed when a treaty has not even entered into force, e.g., the UN Convention on Jurisdictional Immunities of States, 42 in which case additional evidentiary materials may be required. 43 To their credit, some domestic courts realise the trap of deriving custom from multilateral treaties, and may make an effort at referring to concordant 40  practice. 44 Sometimes they may do so without further identifying this practice, 45 but they might just as well be willing to engage in a rather detailed analysis of state practice, sometimes also with a view to proving that an invoked treaty norm does not reflect custom. 46 As far as the latter method is concerned, the practice of US courts stands out in this respect. Through analysing state practice, the status of ratifications and state party reservations, US courts have held that customary international law did not prohibit military use of herbicides in spite of a 1925 Geneva Convention, 47 that the International Convention for the Suppression of the Financing of Terrorism did not constitute customary international law, 48 and that provisions of the Convention on the Rights of the Child did not necessarily reflect custom. 49 Finally, some questionable practices may be flagged that replace rigorous scrutiny of state practice with an analysis of whether the convention containing the alleged customary norm is judicially enforceable (an approach which conflates 44  validity with enforceability), 50 whether the convention is not just aspirational and vague (which betrays a concern over judicial law-making to the detriment of the foreign policy prerogatives of the political branches), 51 or whether it concerns a 'matter of universal concern' (which conflates formal validity with the substantive content and scope of obligations of the norm). 52 Also of note is the practice of admitting that a multilateral treaty not ratified by the forum state does not constitute customary international law, but nevertheless going on to apply its provisions on the ground that customary law is too difficult to determine. 53 The latter methodology, which is arguably inspired by the wish to prevent a non liquet, obviously makes a mockery of the theory of custom-formation.

Authoritative Non-Binding Instruments
The ICJ has the habit of regarding non-binding instruments such as UN General Assembly resolutions and Draft Articles of the International Law Commission (ILC) as, under some circumstances, reflective of customary international law, again without much analysis of its own. 54 Domestic courts are not much different: while sometimes conceding that these instruments are as such not binding, they may readily apply their provisions to the case at bar. 55  ILC Articles are of course imbued with a rather high degree of legitimacy, emanating from a UN body tasked with codifying international law. 60 However, domestic courts may also rely on far more contestable non-binding instruments, a scientific consensus, or just proper policy, to buttress a predetermined outcome, or at least one reached on other grounds. 61 For example, the English Court of Appeal cited a non-binding resolution adopted by the Institut de Droit International concerning the customary immunity of heads of state. 62 The Hong Kong Court of First Instance referred to a 'roundtable' of 35 experts who concluded that the principle of nonrefoulement of refugees is a principle of customary international law, as well as to the non-binding San Remo Declaration on the Principle of Non-Refoulement. 63 The Federal Constitutional Court of Serbia asserted that a document on the independence of the judiciary adopted by just one expert meeting contained generally accepted principles of international law, 64 without citing any further evidence. Finally, the 57 Tomka (2016), p. 11. But see , p. 64 (submitting that 'there is no evidence that any common law court has ever referred to the work of the ILC'). 58 66 and partially on that basis ordered the State to reduce the joint volume of Dutch annual greenhouse gas emissions by at least 25 per cent at the end of 2020 compared to the level of the year 1990. 67 None of the just-cited instruments would normally produce direct legal effects in municipal legal orders. A special class of non-binding authoritative instruments on which courts rely to establish the customary nature of treaty provisions are the travaux préparatoires or preamble of a treaty. The Dutch Supreme Court, for instance, held that Article 19 of the UN Convention of Jurisdictional Immunities (a convention not yet in force for that matter) reflects customary international law, because the preamble states that jurisdictional immunities are generally accepted as customary international law. 68 Surprisingly, a Belgian Court of Appeal considered the same article not to be reflective of customary law, by referring to the travaux préparatoires. 69 Another class of non-binding instruments, which domestic courts have considered as persuasive authority regarding the identification of customary international law, are statements by the executive branch of government. This is a common practice in the United States, where courts may sometimes accept at face value executive determinations regarding the existence, or not, of customary norms, without an analysis of their own. 70 In so doing, they effectively outsource the customary law identification process to the Executive. In other cases, courts may rely only in part on executive determinations regarding the (non-)existence of customary norms, while still treating them as authoritative enunciations. 71 65  That domestic courts sometimes accept non-binding authoritative instruments as statements of positive law does not mean that they always do so. 72 Courts may well put a high premium on the consent of states in the formation of international law, and on that ground refuse to accept soft law instruments as evidence of customary international law. 73 There are a considerable number of domestic cases in which courts did not accept the customary status of provisions in ILC Draft Articles in the absence of sufficient state practice. As previously discussed with regard to the preamble and travaux préparatoires, a Belgian court considered that Article 19 of the ILC State Immunity Articles, which deals with the immunity from execution, was not a codification of customary international law. 74 A year later, the ICJwhich should have had access to exactly the same contemporary state practiceconsidered the essence of the article to be reflective of customary law, or at least of a well-established practice. Regardless of the law-ascertainment methodology used, the ICJ's view is obviously more authoritative, but this goes to show that there appear to be discretionary elements in the customary international law identification, or at least that law-identifiers have considerable agency. 75 A refusal to accept ILC Draft Articles as restatements of lex lata, could also be discerned from the English High Court's ruling that Article 8 of the ILC Draft Articles on Diplomatic Protection, which allowed diplomatic protection of refugees, was lex ferenda rather than lex lata, 76 and from a judgment of a German regional court with respect to Article 6 of the ILC Draft Articles on State Responsibility, which deals with attribution in cases of lending of state organs to another state. 77 When rendering such decisions, domestic courts are not necessarily at loggerheads with the ILC, however: after all, the ILC not only codifies international law, but also progressively develops it. The ILC may even openly admit that a particular rule which it lays down, amounts to progressive development of international law. Thus, in the commentary to the aforementioned Article 8 of the ILC Draft Articles on Diplomatic Protection, the ILC characterised the article as 'an exercise in progressive development of the law, [which] departs from the traditional rule that only nationals may benefit from the exercise of diplomatic protection', while citing, apparently approvingly, the English court's ruling that the article was 'not yet part of international law'. 78 It is of note that where courts do accept non-binding instruments as positive law, possibly alongside other practice, individual judges may sometimes take issue with the majority's 'uncritical' evidentiary analysis, and its creation rather than 'determination' of a norm. 79 Some international judges have criticised the majority in similar terms. 80 When domestic courts do give legal effect to non-binding instruments, it bears emphasis that they may not just act as simple law-appliers (in case the customary character of the relevant provision had already been acquired and its scope of application been delimited), but also as veritable law-creators. As domestic court decisions count as instances of state practice, they may contribute to the hardening into positive law of norms that were initially meant only as progressive development of international law. This goes to show that domestic courts, on par with international courts and tribunals, might be active developers of international law, a point that recent doctrine has repeatedly emphasized.

Doctrine
Unlike the ICJ, a considerable number of domestic courts seem to rely quite heavily on what the doctrine believes is custom. 82 This is arguably so because domestic courts consider that they lack the required expertise to properly determine international law. 83 Domestic courts may thus outsource their custom-determination to scholars, assuming-possibly mistakenly-that scholars have painstakingly and objectively analysed relevant practice. Examples of strong reliance on doctrine are legion. 84 In the seminal case of United States v. Smith (1823), the US Supreme Court held that the definition of an offence under 'the law of nations' (i.e., customary international law) was determined (apparently primarily) on the basis of 'the works of jurists writing professedly on public law', apart from 'the general usage and practice of nations', and 'judicial decisions recognizing and enforcing that law'. 85 The Smith methodology was more recently confirmed in the Filartiga case (1980), which rediscovered the US Alien Tort Statute that gives aliens a cause of action in US federal courts for violations of the law of nations. 86 The US Supreme Court also relied rather heavily on a study conducted by legal scholars when considering that the imposition of life sentences without parole on juveniles was cruel and unusual. 87 It may even happen that courts refer to the views of scholars, without mentioning whose views they have consulted. 88 Courts rarely specifically reject any role for the doctrine in the determination of customary international law. 89 That being said, while domestic courts may consider the doctrine important as a subsidiary source of international law for purposes of customary law determination, they will nonetheless often cite other evidence of the purported norm, or absence thereof. 90 In this respect, domestic courts at times begin their process of identifying customary law by examining international judicial practice or treaties, but eventually base their conclusions on doctrinal analyses. 91 Domestic courts' reliance on doctrinal sources again evinces that they may feel more comfortable referring to pre-existing 'written' evidence, identified by other agencies, in the process of customary law identification.
When relying on doctrine to determine customary international law, domestic courts may sometimes conveniently overlook the contestation within the doctrine regarding the existence or not of a customary rule. For instance, the Court of Appeals of Milan based the principle that there is no customary rule which recognised functional immunity from criminal proceedings for all state officials for acts performed in their official functions on 'authoritative doctrine' and 'univocal views of scholars', 92 although in reality there is no consensus on this principle. 93 The Canadian Supreme Court, for its part, drew heavily on academic works when positing that the precautionary principle in environmental law may be a customary international law norm, 94 although there are other works that contest the customary character of the precautionary principle. 95 Sometimes, courts are explicitly aware of a discussion in the doctrine on the existence of a customary norm, but nevertheless go on to apply (or 'create') the norm. The High Court of Hong Kong, for instance, while noting the absence of a universal consensus among academic writers regarding the customary status of the principle of nonrefoulement, eventually took the view that the principle did reflect customary international law on the ground that universal adherence to the principle was not required. 96

International Judicial Practice
Most domestic courts embrace an attitude of deference to international courts when it comes to the determination of customary international law. They may do so out of respect for international courts' stronger expertise and legitimacy in matters of international law, 97 possibly because they perceive international courts to be higher up in the judicial hierarchy, as well as out of concern about the uniform application of international law across jurisdictions. 98 Again, the examples are rife. 99 Domestic courts may not consider international case law as necessarily binding, but clearly they believe that it has persuasive authority, 100 much like the ICJ has treated relevant ICTY case law in the Bosnia Genocide Case. 101 However that may be, domestic courts appear to be outsourcing the determination of customary international law norms to international courts, much like they outsource it to the doctrine. When an international court has addressed the matter, they do not normally carry out an extensive law-determination process of their own. 102 Domestic courts may have little concern that the method of lawdetermination applied by international courts may sometimes be backed by only flimsy evidence. 103 After all, domestic courts can be expected to defer to international courts on grounds of the latter's superior, expertise-based authority.
It may happen that domestic courts consciously refrain from relying on pertinent international judicial practice regarding the determination of a customary international law norm, without necessarily replacing the international courts' law-determination process with their own, however. Three scenarios can be distinguished. First, very exceptionally, domestic courts may take issue with the fact, alluded to above, that international courts often just assert rather than find the law. 104 A US court of appeals, for instance, refused to heed relevant decisions of the ICJ and the European Court of Human Rights on the ground that these tribunals 'are not empowered to create binding norms of customary international law' and that 'their decisions could not be considered primary sources of international law'. 105 Such an attitude may at first sight be understandable, but it obviously threatens the uniform application of customary norms and begs the question whether domestic courts are better placed than international courts to determine international law. Second, domestic courts may refrain from applying a customary norm affirmed by an international court when it clashes with domestic constitutional law protections that are considered to prevail over any other legal norms. This attitude appears legitimate insofar as the constitutional law norm is 'consubstantial' with another international norm. 106 Finally, domestic courts may hold that legal concepts recognised by international tribunals, while entirely valid in inter-state relations, have no relevance for relations under domestic law. 107 In this vein, the German Constitutional Court held that the concept of 'necessity' could not be invoked as an international law defence to justify the non-performance of contractual obligations owed to private debtors, as these were governed by domestic private law, whereas decisions of international tribunals concerned legal relations under international law. 108 Somewhat along the same lines, a French appeals court held that it could be inferred from the case law of the ICJ that customary international norms are 104 See Talmon (2015). 105 See Tzanakopoulos (2014) on the term. Accordingly, from a substantive perspective, the domestic court in fact balances different international law norms instead of giving precedence to domestic law. This has notable relevance for fundamental human rights, which feature in both (domestic?) constitutional and international law instruments, in particular the right to a remedy. Claimants in domestic law proceedings may invoke the right to a remedy so as to restrict the scope of application of international immunities and sanctions. 107 See more generally on domestic courts' views that customary norms only govern inter-state relations: Wouters (2004), p. 34 (referring in particular to the practice of not granting direct effect to international law). See on the law of responsibility Wittich (2013), p. 646 ('While the relevant reference to adjudging the approach of domestic courts towards questions of state responsibility is the 2001 ILC Articles, it is not clear whether these are applicable at all to the relations between states and individuals or non-state entities', although ultimately writing that 'with regard to the content of state responsibility (reparation, that is) the ILC Articles may indeed be viewed as providing the general rules on the matter and can legitimately be employed as a reference for analysing the rules on compensation also in relation between states and individuals'). 108 Argentine Necessity Case, K and ors v. Argentina (represented by President Néstor Kirchner), Decision of the Federal Constitutional Court, Order of the Second Senate, 2 BvM 1-5/03, 1, 2/06, vol. 118, 124, 60 NJW (2007), 2610, 138 ILR 1 (2010, ILDC 952 (DE 2007), 8 May 2007, paras. 51-66 (adding that an extension of the necessity doctrine was not supported by state practice and opinio juris, and that, given the differentiated treatment given to the concept of necessity in domestic legal systems, it was not a general principle of law). applicable to states only. 109 Since in domestic settings international law is often invoked by or against private actors, such an approach has the potential to seriously restrict the domestic applicability of international law. 110 Finally, domestic courts may not only refer to international courts, but also to other courts, both from within and outside their own jurisdiction. Rather selfevidently, they tend to defer to higher courts within their national legal system, on the basis that the latter courts have more authority, expertise, and legitimacy. 111 In some jurisdictions lower courts may even refrain from applying international law, in particular identifying customary international law, when a higher court has not yet addressed the matter. 112 Such a reserved attitude obviously limits the opportunities for domestic courts' determination of international norms, and their contribution to the content of the sources of international law. Apart from referring to court decisions from within the jurisdiction, domestic courts may also cite relevant decisions of domestic courts from other jurisdictions as evidence of the existence of customary international law norms. 113 This process has been usefully characterised as 'comparative international law', coming into being on the basis of a transnational judicial dialogue. 114 While possibly increasing the quality of a court's customary law identification methodology, this however comes with its own risks, such as cherry-picking and undue reliance on misguided decisions rendered by courts lacking international law expertise. 115

Extensive Analysis
For all the legitimate criticism that can be levelled at courts' international lawdetermination method, one has to concede that exceptionally they do carry out a relatively extensive analysis of relevant state practice for purposes of customary law determination. A few higher court decisions, from the Supreme Courts of Spain, Germany, and Poland, can be singled out in this respect by way of illustration. 116 Such decisions may be commended for aligning with the two-element approach to customary law identification. As this article is primarily concerned with systematization, however, it is not our ambition to develop a best practices-based normative blueprint of how to properly identify customary international law in concreto. Nor have we reviewed whether the extensive analysis and the conclusions reached in the specific cases were sound. The examples given here are only meant to show that domestic courts do sometimes engage in extensive analysis. It can obviously not be excluded that even an extensive analysis may lack rigor.
In the Guatemala Genocide case, the Spanish Constitutional Court analysed in detail abundant foreign legislation before reaching the conclusion that the principle of universal jurisdiction was not subject to the existence of a 'connection' with national interests. 117 In the Yemeni Citizens Extradition case, the German Constitutional Court clearly set out its methodology of ascertaining customary international law-the two-elements approach-and subsequently engaged in detail with foreign courts' case law in order to find an answer to the question whether the fact that a prosecuted person had been lured out of his or her state of origin is an obstacle precluding extradition. 118 In the Natoniewski case, the Polish Supreme Court carried out a thorough analysis of national legislation and case law (i.e., actual state practice), treaties, case law of the ICJ and the European Court of Human Rights, and doctrine, before concluding that there is no customary law-based exception to state immunity for grave violations of human rights. 119

Conclusion
The analysis in this article has demonstrated that, similar to the ICJ, domestic courts do not normally identify norms of customary international law on the basis of the textbook method of ascertaining a general practice accepted as law. 120 Rather, they tend to outsource the determination of custom to treaties, non-binding documents, doctrine or international judicial practice. Sometimes, it appears that domestic courts simply assert, without citing persuasive practice authority, the existence of a customary norm. Arguably, domestic courts feel more comfortable referring to (more or less) authoritative international written sources. They may be forgiven for doing so, as they are not experts in international law. Indeed, for domestic courts it may be particularly challenging to gather and assess relevant state practice, and identify opinio juris. 121 At the same time, by using 'shortcuts', domestic courts may want to signal their impartiality and legitimacy in determining customary international law, for much the same reason as the ICJ uses such shortcuts. As Niels Petersen has observed, how exactly a law-applying agency-in his study the ICJ-analyses state practice is often difficult to observe and will always be selective, or at least carry a whiff of selectivity. 122 Thus, they may cite international treaties with a view to more strongly anchor an identified customary norm in the consent of states, and on that basis, preempt accusations that they favour one party over another. 123 From this perspective, a painstaking analysis of state practice and opinio juris may backfire politically. That being said, such political dynamics may be rather particular to the ICJ. As the Court's jurisdiction is not compulsory, states may withdraw their acceptance of jurisdiction in case the Court is seen as taking decisions that are insufficiently rooted in state consent. This risk of state withdrawal is obviously not present before domestic courts. Still, by taking decisions that are insufficiently attuned to the interests of the community of states, domestic courts risk legislative intervention by the forum state, which may clip courts' customary law-finding wings. 124 It remains that from a doctrinal perspective, a serious legal analysis of all available materials is called for. Ultimately, courts, and especially domestic courts, which function in a state governed by the rule of law, are supposed to identify and apply the law, regardless of the political ramifications of their decisions. Therefore, a rigorous application of the two-elements approach to customary law identification by domestic courts is called for. How such rigor is to be achieved has not been the subject of this article. In fact, it is the subject of a large-scale, on-going ILC study, which has triggered intensive academic activity on customary law-identification methods. 125 In any event, in order to preserve the unity of customary law formation methodology, it is posited that no distinction should be drawn between law-applying agencies: thus, both international and domestic courts should use the same methodology.
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.