4 Interpretation and modi fi cation
6.3 The Pure Theory’s constitutional theory
6.3.2 The architecture of the constitution of international law
What is the upshot of this discussion? We now know that we can only enquire how the sources of international law could be arranged. We can also ask how scholars in the past have arranged them. This ‘arrangement’ of the sources of inter- national law is different from the question posed in Chapter 5, however. We are not asking about derogability, but about which forms of norm derive their validity from which norms. In other words, we are asking about the hierarchy of legal conditionality (Stufenbau nach der rechtlichen Bedingtheit) in the relationship of the meta- norms of international law-creation to each other. What form can this archi- tecture take? Logic allows us to formulate three options for the normative order(s)
‘international law’.
186 ‘[Die Grundnorm] beschrọnkt sich auf die Einsetzung der Rechtssetzungsautoritọt und ist deshalb begrifflich schon ihrem Inhalt nach immer nur singulọr.’ Jỹrgen Behrend, Untersuchun- gen zur Stufenbaulehre Adolf Merkls und Hans Kelsens (1977) 28.
187 Rub (1995) supra note 20 at 311.
6.3.2 A constitution for international law 235
(1) One source of international law is subordinate to another or one source of international law is the supreme source and origin of all international law. All other sources are derived from the supreme source, because that super-ordinate source contains a norm authorising the creation of subordinate sources. All three sources mentioned in Article 38(1)(a)–(c) ICJ Statute have in the past been proposed as the supreme source of international law.188
(a) Alfred Verdross, in his Die Verfassung der Vửlkerrechtsgemeinschaft (1926), as well as the early Kelsen of the Das Problem der Souverọnitọt (1920),189 base their construct of the constitution on pacta sunt servanda as Grundnorm,190 which is at once the basis for treaties and custom, because one is an express conclusion of a convention and the other is a tacit convention (pactum tacitum).191 Strictly speaking, however, this con- struct does not propose one superior source, but two. The creation of treaties and customary law is authorised by the same Grundnorm.192 This construct is problem- atic not only because of the difficulties encountered in construing customary international law as tacit pact (Section 3.2), but also that the Grundnorm as epi- stemological tool is not able to create unity (Section 6.3.1).193
(b) In Principles of international law (1952) Hans Kelsen postulates that customary international law is the highest source. International law’s Grundnorm is: ‘The states ought to behave as they have customarily behaved.’194 The norm ‘pacta sunt servanda’, as the Grundnorm of the subordinate legal order ‘international treaty law’, is merely a norm of positive (customary) law.195 The problem with customary law as superior source is that it cannot integrate other norms as norms, because they cannot form a factual pattern of usages (Section 3.2.5).
(c) From about 1931 onwards, Verdross claims that general principles of law are the fount of all other sources:
We have to assume that the validity [of the general principles of law] can be derived from the same Grundnorm as that of customary law . . . International law’s Grundnorm would roughly have this content: sovereign and partially sovereign legal communities ought to behave in their inter-relationships according to the general principles of law,
188 Rub (1995) supra note 20 at 337.
189 Hans Kelsen, Das Problem der Souverọnitọt und die Theorie des Vửlkerrechts. Beitrag zu einer reinen Rechtslehre (1920) 217, 262, 284; cf. Rub (1995) supra note 20 at 313.
190 Verdross (1926) supra note 14 at 29.
191 Verdross (1926) supra note 14 at 43–44. Verdross later changed his views on the nature of custom and the foundation of the international legal order.
192 Verdross (1926) supra note 14 at 44.
193 Métall’s arguments against this construct (Métall (1931) supra note 9 at 418–422), namely that since pacta sunt servanda is already a positive norm of international law it cannot at the same time be its Grundnorm cannot succeed, inter alia because customary law cannot conceive of ‘servanda’
as state practice (Chapter 3 and infra).
194 Kelsen, (1952) supra note 418; Kelsen (1960) supra note 7 at 222 (Ch 34 h), 324–325 (Ch 42 c);
Métall (1931) supra note 9 at 425.
195 Kelsen (1952) supra note 11 at 314, 417; Kunz (1934) supra note 8 at 403; Virally (1968) supra note 25 at 128; Rub (1995) supra note 20 at 314.
Uncertainty in International Law 236
unless international relations have created particular valid norms which deviate from these principles.196
Admittedly, for the purposes of classification this statement is not quite as clear as could be hoped for. While general principles are considered originary (they are explicitly referred to in Verdross’ proposed Grundnorm), and while treaty law is subordinate to it as particular norms,197 customary international law has a similar value and is somehow parallel to them198 – perhaps as a sort of ‘manifestation’? As this theory relies upon natural law arguments (general principles of law as abso- lute standard, not posited by humans), which we have problematised above, we will not pursue this strand further.
(2) A constitution of international law or historically first constitution as a meta-meta-level of the kind mentioned in Section 6.3.1, whose task would con- sist in authorising the creation of sources, rather than prescribing subject behaviour (although this is not excluded by theory), could also be imagined.
Thus, while ‘international treaty law’ as a source is not derived from customary international law, and while the two sources are ‘two separate branches of law’199 of equal standing, they would be connected by a superstructure of meta-meta- laws which regulate the relationship between the formal sources of international law. This opens up the question of a closed or open catalogue of sources of international law. If we assume a ‘constitution’ to govern all international law (i.e.
one normative order) then the number of possible sources is closed and only if a new source conforms to the constitutional law for adding sources is it added to the catalogue of international law’s sources. That special supra-law would have to be composed of positive norms – the historically first constitution cannot simply be presupposed. As mentioned above, it is doubtful whether such positive norms exist.
(3) Another version could be called the ‘default theory’. The three main formal sources are not hierarchically ordered200 and the sources are themselves not normatively connected.201 Applied to current international law this would mean
196 ‘Vielmehr mỹssen wir annehmen, daò [die] Geltung [der allgemeinen Rechtsgrundsọtze] aus derselben Grundnorm herzuleiten ist, der auch das Gewohnheitsrecht entspringt . . . Die vửlker- rechtliche Grundnorm wỹrde also etwa folgenden Inhalt haben: Souverọne und teilsouverọne Rechtsgemeinschaften verhaltet Euch in Eueren gegenseitigen Beziehungen nach den allgemeinen Rechtsgrundsọtzen, soweit sich nicht im internationalen Verkehre besondere, von jenen Grund- sọtzen abweichende, gỹltige Normen herausgebildet haben.’ Verdross (1931) supra note 69 at 362.
197 Verdross (1931) supra note 69 at 362; Métall (1931) supra note 9 at 424.
198 Métall (1931) supra note 9 at 423–424.
199 Grigory Tunkin, Is general international law customary law only?, 4 European Journal of International Law (1993) 534–541.
200 There are cases where a subordination is obvious. Not every manifestation of norms is com- pletely autonomous. Security Council Resolutions derive their validity from the United Nations Charter, as do International Court of Justice judgments.
201 In an early book Kelsen shows the theoretical possibility of this solution: Kelsen (1920) supra note 189 at 106–107.
6.3.2 A constitution for international law 237
that both ‘pacta sunt servanda’ and ‘consuetudines sunt servanda’ are basic norms. No constitution which binds these two types of norms in one normative order is cognisable and neither is the subordination of treaty to custom or vice versa. If no connection is proven as part of positive law, it is possible that no connection between the sources exists. Both types of law may be part of ‘international law’, but the connection can only be an empirical classification. Without an overarching constitution regulating what kinds of formal sources international law has, the two, three or more sources currently recognised might be two, three or more different legal orders.202 This might well be the consequence of orthodox international legal doctrine that sees the sources as ‘equals’.203
If, for example, the constitution of a municipal legal system does not reco- gnise customary law (thereby denying its validity), then for the legal system char- acterised by the constitution, customary law does not exist, just like a concurrent and competing legal system (e.g. the People’s Republic of China and the Republic of China on Taiwan). On the other hand the municipal constitution could recog- nise customary law and could purport or claim to subordinate customary law to its legal system. The consequence of this theory regarding the catalogue of sources – if we assume international law to be only an empirical category – is that the catalogue of sources is open and whatever claim by whomever fulfils the empirical classification criteria can be counted as belonging to the ‘family of international law’. While a ‘constitution of international law’ is one normative order, the ‘default theory’ is not more than an empirical communality (more than one – a multitude of – normative orders and one defining empirical com- munity). If this is how the architecture of sources is shaped, we are faced anew with the question of the derogatory power of norms of the same quality, but of a different kind,204 which is as uncertain as all other possibilities of derogability.
Yet another solution to the two problems of the meta-norms would be to incorporate all conditions for the creation of, for example, customary law (material and subjective element, time frame, participation level, repetitions, persistent objector) into the postulated Grundnorm of, for example, customary international law. This, in turn, makes all the conditions for law-creation (Rechtserzeugungsbedin- gungen) part of one postulated norm. However, as mentioned repeatedly, the Grundnorm is not positive, but presupposed. It cannot create what is not already positive. It only gives validity as existence as Ought.
202 This theory raises interesting problems of the succession of treaties by customary law and vice versa.
203 E.g.: Bos (1977) supra note 26 at 73–74; Cassese (2005) supra note 26 at 154; Gihl (1957) supra note 14 at 75; Neuhold (2004) supra note 25 at 31 (RN 174).
204 Wolfram Karl discusses this topic in: Karl (1983) supra note 35 at 86 et seq. The conclusion he draws from the equality and non-connectedness of treaty and custom is that both can derogate from the other; a statement that does not conform to the approach presented in this book.
Uncertainty in International Law 238