4 Interpretation and modi fi cation
6.3 The Pure Theory’s constitutional theory
6.3.1 The Stufenbau determines the sources of
The answer of the Pure Theory to the question ‘Where do the sources of law come from?’ is clear. For this theory of normative orders which are hierarchically ordered, a ‘source of law’ is the meta-law on law-creation,169 irrespective of where that norm resides within the hierarchy. In other words, the source is the norm- function ‘authorisation’, even though we might not usually call authorising norms which hold a subordinate position within a normative order a source of law. If we combine the preliminary notion of ‘constitution’ as the highest echelon of author- ising norms in a given normative order (in international law traditionally treaty law and customary law)170 with the notion of the hierarchy of norms (Stufenbau)171 (Sections 5.5.2 and 6.1.1), the correct question to ask in response to ‘Where do the sources come from?’ is, ‘What norm of international law authorises the creation of the norms that authorise the creation of (for example) customary international law?’172 The Pure Theory of Law enjoins us to ask for and find positive norms of international law that create source-law, such as the law on custom-creation (Chapter 3).173
How we are to proceed depends upon the answer we can give to that question.
How uncertain the constitution of international law is depends (a) upon whether we can give an answer to that question with any degree of certainty (Section 6.3.3) and (b) whether definite answers to the question lead to further theoretical prob- lems and make the highest normative ontology of international law uncertain (Sections 6.3.1 and 6.3.2). The answer, if we can give it, can take two forms, for either we find a positive norm authorising source-creation or we do not.
If there is a norm of international law authorising the creation of the sources of international law, i.e. a meta-meta-law of source-creation, it would probably fit Kelsen’s ‘historically first constitution’ (historisch erste Verfassung).174 Despite the name, that specific form of constitution is not only historically but also logically prior in the hierarchy of validity:
If one asks why norms which regulate the creation of general norms are valid, one may
169 Kelsen (1952) supra note 11 at 303; Kelsen (1960) supra note 7 at 238–239 (Ch 35 e); Petev (1984) supra note 15 at 273.
170 Kunz (1934) supra note 8 at 412.
171 Verdross (1923) supra note 16 at 129.
172 Rub (1995) supra note 20 at 312–313.
173 Verdross (1926) supra note 14 at 43.
174 Kelsen (1952) supra note 11 at 411.
6.3.1 A constitution for international law 231
find a yet older constitution, i.e. the validity of the present constitution is based in its being created according to the provisions of a previously valid constitution by way of an amendment of the constitution. Thus at the end one comes to the historically first constitution, which cannot be founded in a positive norm, a constitution which came into validity through a revolutionary process. If one asks why the historically first constitution is valid, the answer can only be that the validity of this constitution, the assumption that it is a binding norm, must be presumed.175
The passage quoted exhibits all the relevant elements of a historically first consti- tution we need to transpose it to the realm of international law. The historically first constitution is the highest positive norm of a positive normative order. Because it is historically first and hierarchically highest, it is not derived from a previous (higher) norm – if that were so, it would not be the highest norm – and thus its creation cannot be based on a previous constitution. Its creation was revolutionary (or at least ‘originary’).
Take the Austrian post-1945 order as an example. In a ‘law’ enacted just after the war ended, Austria’s main political parties declared that the 1920/1929 con- stitution was reinstated.176 Since that act itself did not conform to the provisions of that constitution, they actually created a new constitution identical in content with the old constitution. The validity of the new constitution, however, is based on the norm created by the parties in 1945, not by the original enactment of the constitution in 1920. Despite its claims, Austria’s post-1945 constitutional order is based on a revolutionary act and the declaration of reinstatement by the parties is its highest positive norm. As Kelsen notes, where a positive norm cannot be based in another positive norm, its validity can only be presupposed if one wants to perceive it as a norm. The historically first constitution is thus directly below the Grundnorm of a given normative order (Grundnormunmittelbarkeit).177 In
175 ‘[F]ragt man nach dem Geltungsgrund der Normen, die die Erzeugung der generellen Normen regeln . . . so gerọt man vielleicht auf eine ọltere Staatsverfassung; daò heiòt: man begrỹndet die Geltung der bestehenden Staatsverfassung damit, daò sie gemọò den Bestimmungen einer vorangegangenen Staatsverfassung im Wege einer verfassungsmọòigen Verfassungsọnderung . . . zustande gekommen ist; und so [gerọt man] schlieòlich auf eine historisch erste Staatsverfassung, die nicht mehr auf eine [positive] Norm zurückgeführt werden kann, eine Staatsverfassung, die revolutionọr . . . in Geltung getreten ist . . . [F]ragt man nach dem Grund der Geltung der historisch ersten Staatsverfassung . . . dann kann die Antwort . . . nur sein, daò die Geltung dieser Verfassung, die Annahme, daò sie eine verbindliche Norm sei, vorausgesetzt werden muò . . .’; Kelsen (1960) supra note 7 at 203 (Ch 34 c). Nota bene:
‘Staatsverfassung’ (‘constitution of a state’) has been deliberately mistranslated only as ‘constitu- tion’ in order to show that Kelsen’s theory is not limited to the municipal realm. Due to the lack of succinctness of the German original, the translation is less literal here.
176 The situation is more complicated, but essentially as portrayed here, see Art I, III Unabhọngig- keitserklọrung, in: Proklamation [ỹber die Selbstọndigkeit ệsterreichs], StGBl 1945/1; Art 1 Verfassungsgesetz vom 1. Mai 1945 über das neuerliche Wirksamwerden des Bundes- Verfassungsgesetzes in der Fassung von 1929 (Verfassungs-ĩberleitungsgesetz – V-ĩG), StGBl 1945/4.
177 Métall (1931) supra note 9 at 421.
Uncertainty in International Law 232
our case the post-1945 Austrian legal order is valid only if we presuppose the Grundnorm: ‘the norms of the political parties are valid norms’ or something to that effect.
But why would we need to discuss the notion of a ‘historically first constitution’
in a book dealing with international law? Surely, one might object, this is good and proper in countries with a written constitution, but what relevance does it have for international law which clearly does not have anything approaching a constitution, let alone a ‘first’ constitution? The issue is dealt with here for two reasons. First, any normative order necessarily has a historically first constitution, just as any order has a constitution in the theoretical sense. The question for international law is merely what exactly it looks like (Section 6.3.2). Second, in Universelles Vửlkerrecht (1984) Alfred Verdross and Bruno Simma have used this notion in their section on sources of international law.178 Given Verdross’ know- ledge of and respect for Kelsen, it is likely that he was aware of the similarity; thus it apposite to discuss their idea here.
In a 1973 publication, at the end of a historical argument regarding the development of international law, Verdross claims that history shows that:
their original norms were thus neither created by a formal treaty nor by customary international law, but through informal consensus of the then powerful entities, where they acknowledged certain principles of law as binding. . . . Still, these constitutional norms are not a series of hypothetical norms, but actual norms constituting the basis for customary international law and formal treaty law.179
Verdross and Simma continue in 1984: ‘The originary source of international law – international consensus – is not only the historical basis of the formalised methods of creation [in Article 38], but is still superimposed upon them.’180 They use consent as a historically first constitution in Kelsen’s sense, even if only impli- citly. Consent as foundation for the sources of international law is a relatively widespread and well-established theory,181 but we will not discuss it in the abstract, only in the specific form Verdross and Simma advocate. They consider ‘consensus’
to be originary and that it came into being uno actu with the coming-into-existence
178 Verdross and Simma (1984) supra note 23 at 59–60, 324–327; cf. also: Verdross (1973) supra note 44 at 20–21. Contra: Bleckmann (1978) supra note 42; Fastenrath (1991) supra note 27 at 112–113.
179 ‘ihre ursprỹnglichen Normen sind also weder durch einen fửrmlichen Staatsvertrag noch durch die vửlkerrechtliche ĩbung, sondern durch einen formlosen Konsens zwischen den damaligen Machthabern entstanden, durch den sie bestimmte Rechtsgrundsọtze als rechtsverbindlich anerkannt haben. . . . Gleichwohl bilden jene Verfassungsnormen kein bloò hypothetisches, sondern ein dem VGR und dem fửrmlichen Vertragsrecht tatsọchlich zugrundeliegendes Normengebilde.’ Verdross (1973) supra note 44 at 20–21 (emphasis added); Verdross and Simma (1984) supra note 23 at 59.
180 ‘Die originọre Vửlkerrechtsquelle des zwischenstaatlichen Konsenses liegt diesen formalisierten Erzeugungsarten [in Artikel 38] aber nicht nur historisch zugrunde, sondern überlagert sie nach wie vor.’ Verdross and Simma (1984) supra note 23 at 324.
181 To name only very few: Charney (1997) supra note 56; Corbett (1925) supra note 31; Hoof (1983) supra note 3; contra: Métall (1931) supra note 9 at 424.
6.3.1 A constitution for international law 233
of the modern state.182 If one were to try to express this thought in terms of the Pure Theory one would say that this is the first positive norm of the constitution of international law, not itself based upon another positive norm.
But there are problems with re-reading Verdross and Simma in a Kelsenian light. While they consider the sources of international law to have been derived from consensus, they leave it open whether they mean to construct a hierarchical normative order. It seems they do not conceive of the historically first constitution as a meta-meta-law in the sense employed by the Pure Theory of Law. Also, they warn against misunderstanding ‘consensus’ as the source of validity of international law.183 While this was clearly meant to distinguish their position from crude voluntarism, the other consequence is a delimitation also from the Pure Theory, for which ‘source of law’ and ‘source of validity’ are necessarily identical.
We may not be able to take Verdross’ and Simma’s specific theory on board in a study of the sources of international law basing itself upon the Pure Theory of Law, but what if we were to argue that consensus is the historically first constitution of international law? In this case consensus, as a positive norm of international law, would have authorised the creation of custom and treaty as the formal sources of international law. International law’s Grundnorm would author- ise consensus to create norms of international law and everything would work out – except that we would first have to prove that consensus is the originary constitutional norm of international law. That is where certainty ends and the problems begin: how can we prove this contention? How can we prove that there was a positive act of will creating, say, customary international law, at some point in the past? That is the point of a historically first constitution in the positivist theoretical edifice of the Pure Theory. All positive norms have to be positus (a product of human willing) and cannot be presupposed.184 Our epistemological horizon is too limited to answer this question with more than a presumption. As long as we are presupposing, we could presuppose any norm to found inter- national law, even absurd ones. We need to prove positivity, because mere fictional norms (Section 6.2.2) cannot found the validity of sources of positive law.
If, on the other hand, the answer to the question posed above – whether we are able to find a norm of international law authorising the creation of its sources – is negative, the argument stops at the Grundnorm. If, for example, the creation of customary international law is not authorised by a positive norm of international law, we must presuppose its validity in order to perceive customary international law as law, which means presupposing the Grundnorm. Apart from this authorising function, the Grundnorm also unifies the norms under it into one normative order.185 While the notion of the Grundnorm will be discussed in Chapter 7, we
182 Verdross (1973) supra note 44 at 20.
183 Verdross and Simma (1984) supra note 23 at 327.
184 Fitzmaurice (1958a) supra note 25 at 163–164.
185 Métall (1931) supra note 9 at 416.
Uncertainty in International Law 234
need to point to a problem we encounter when we presuppose it at an early stage in international law.
The basic norm unifies, but it also excludes every norm outside its purport.
Furthermore, the Grundnorm cannot fulfil the same function as a positive meta- meta-norm. Because it is presupposed by the observer, rather than created as a positive norm, its content is determined by ‘its’ positive normative order, rather than vice versa. This means that legal scholars cannot unify two source-norms by simply presupposing a Grundnorm on top – it is too much of a stretch for the concept. In other words, the presupposition Grundnorm cannot create unity where none is in positive law. ‘[The Grundnorm] limits itself to appointing the law-creative authority and its content is thus by its very terms limited to singularity.’186
A far worse problem is the potential this has for fragmenting normative orders.
If the Grundnorm can be presupposed at any stage – if, therefore, positive norm- creation is ‘autonomised’ (Section 5.5.3.3) – we do not have a fixed line where one normative order ends and another begins. It needs to be stressed that this is not the Pure Theory of Law’s fault, but a feature of all positive normative orders which the Pure Theory helps to see clearly. Positive norm-creation has the tendency to be fragmenting, because a superior norm authorising norm-creation as one of the two elements needed for the creation of a positive norm can be ‘supplanted’ by presupposing the Grundnorm. The second element – a positive act of will187 – if existent, could thus be regarded as sufficient for norm-creation. The observer decides where the Grundnorm is to be ‘placed’ and thus becomes the maker of normative ordering. This is uncertainty on the highest level – the divisibility of normative orders due to the ease with which the Grundnorm is presupposed con- spires with the inability of the Grundnorm to combine positive sources to make the sources of international law uncertain.