4 Interpretation and modi fi cation
6.1 What is a ‘source’ of international law?
6.1.1 The concept of ‘source of law’ in legal theory
For Kelsen, sources are themselves norms authorising humans to create norms11 – the norm-function of Ermọchtigung (empowerment)12 – rather than an absolute concept. The term ‘source’ is not the core of the matter:
The term ‘source of law’ is used only figuratively and has more than one meaning. It may denote any higher norm in its relationship to a lower norm, whose creation the former regulates. Thus, the term ‘source of law’ may denote also the basis of validity and especially the ultimate basis of validity of a legal order: the Grundnorm. In fact,
8 Robert Alexy, Hans Kelsens Begriff der Verfassung, in: Stanley L. Paulson, Michael Stolleis (eds), Hans Kelsen. Staatsrechtslehrer und Rechtstheoretiker des 20. Jahrhunderts (2005) 333–352 at 333: ‘Verfassung als . . . letzten positivrechtlichen Grund für die Geltung von positivem . . . Recht’;
Josef L. Kunz, The ‘Vienna School’ and international law, 11 New York University Law Quarterly Review (1934) 370–421 at 412.
9 Kelsen (1960) supra note 7 at 228 (Ch 35 a); Rudolf Aladár Métall, Skizzen zu einer Systematik der vửlkerrechtlichen Quellenlehre, 11 Zeitschrift fỹr ửffentliches Recht (1931) 416–428 at 421.
10 Lazare Kopelmanas, Essai d’une théorie des sources formelles du Droit International, 22 Revue de Droit International (1938) 101–150 at 119–120; Joost Pauwelyn, Conflict of norms in international law. How WTO law relates to other rules of international law (2003) 92.
11 Hans Kelsen, Principles of international law (1952) 303.
12 Hans Kelsen, Allgemeine Theorie der Normen (1979) 82–84 (Ch 26).
6.1 A constitution for international law 197
however, only the positive basis of validity of a legal norm, i.e. the higher positive legal norm regulating its creation, is called its ‘source’.13
The function of empowerment is to identify the norms created under it and for this reason belonging to it (Section 5.5.2). ‘A norm belongs to a legal order only because it is created under the terms of another norm of the same order.’14 In a specific sense we can say with Petev that the Pure Theory is nothing but a theory of sources,15 because its construction of normative orders largely depends upon the authorisation to create norms. Yet the notion of ‘sources’ as law regulating law-creation (Rechtserzeugungsregel) is doubly relative. Not only is a hierarchy thus established between any authorising norm and the norms created under that norm, even if it is not what international legal scholarship calls an ‘original source’ (custom and treaty). Also, the totality of the conditions for norm-creation may on occasion not be confined to one Rechtsform and a constitutional statute may have to fulfil not only constitutional provisions for its creation, but provisions of ordinary statutes as well.
In line with the Pure Theory of Law, the concept of sources in this book is that of sources as meta-norms on norm-creation. They are on a meta-level, because the validity-relationship it establishes between the source and law created by it creates a hierarchy between the two norms, because the validity (existence) of one norm depends upon another. Thus a necessary Stufenbau exists even in international law.16 The source is a norm, because the source establishes mem- bership in the normative order; the source-norm itself has to be a norm of that selfsame order.17 The process is norm-creation, because source-norms make other norms of that normative order18 and thus may very well be called their
‘source’.
What the term ‘source of law’ means for scholars naturally depends on their
13 ‘Rechtsquelle ist ein bildlicher Ausdruck, der mehr als eine Bedeutung hat. Man kann damit jede hửhere Norm im Verhọltnis zu der niederen Norm bezeichnen, deren Erzeugung sie regelt.
Daher kann unter Rechtsquelle auch der Geltungsgrund und insbesondere der letzte Geltungs- grund, die Grundnorm, einer Rechtsordnung verstanden werden. Doch wird tatsọchlich nur der positivrechtliche Geltungsgrund einer Rechtsnorm, das heiòt die hửhere, ihre Erzeugung regelnde, positive Rechtsnorm als “Quelle” bezeichnet.’ Kelsen (1960) supra note 7 at 238–239 (Ch 35 e).
14 ‘Eine Norm gehửrt zu einer Rechtsordnung nur, weil sie gemọò der Bestimmung einer anderen Norm dieser Ordnung gesetzt ist.’ Kelsen (1960) supra note 7 at 239 (Ch 35 f); Kelsen (1979) supra note 12 at 247 (N 45); Alfred Verdross, Die Verfassung der Vửlkerrechtsgemeinschaft (1926) 21.
Perhaps also: Torsten Gihl, The legal character and sources of international law, 1 Scandinavian Studies in Law (1957) 51–92 at 72.
15 Valentin Petev, Rechtsquellenlehre und Reine Rechtslehre, in: Werner Kawietz, Helmut Schelski (eds), Rechtssystem und gesellschaftliche Basis bei Hans Kelsen (1984) 273–287 at 273.
16 Alfred Verdross, Die Einheit des rechtlichen Weltbildes auf Grundlage der Vửlkerrechtsverfassung (1923) 129.
17 Alexy (2005) supra note 8 at 341.
18 Though, as also pointed out in Section 5.5, the autonomisation of norms through their positive nature may be a hiccup of sorts in this matter.
Uncertainty in International Law 198
theoretical approach.19 Yet in international legal doctrine there seems to be a remarkable co-incidence of views on the question, which Alfred Rub argues is surprisingly close to the position of the Pure Theory: ‘The . . . recently received Kelsenian understanding of the term source of international law as method of law-creation.’20
It is doubtful whether this notion fully reflects the ideas behind Kelsen’s theory on the sources of law. Rub focuses on Kelsen’s definition of a source as method of law-creation, as evidenced by citations of works by Maarten Bos, and emphasises the rejection ‘of the view that they are only declarative manifestations of law that is already in the process of creation without involvement [of the sources]’.21 The problem is that Rub’s connection of the word ‘method’ to the one-sided rejection is liable to cause a misunderstanding of Kelsen’s theory. On his portrayal, Kelsen could be seen as arguing that at some point (at the mystical point of ‘sources’) norms are no longer based on norms, but on facts alone and of thus disavowing the Is–Ought dichotomy – which is not in accord with the Pure Theory. Alfred Verdross, whose natural law approach to sources will be discussed later (Section 6.2.2), held as late as 1955 that sources are norms on the creation of law.22 Verdross himself was influenced by the Pure Theory, but his Universelles Vửlkerrecht (co-authored with Bruno Simma) utilises a different argument. In this work, sources are defined as ‘formalised methods of creation’23 and the introduction of an original and originary ‘consensus’ (Section 6.3.1) shows that he moved away from Kelsen’s position to a more pragmatic argument.24
Generally speaking, however, orthodox international legal scholarship’s theory of sources has several traits which are problematic from our theoretical point of view. (1) It is common among scholars of international law not to recognise that sources themselves are norms, but to speak of them as ‘methods’ or ‘procedures’25
19 Pauwelyn (2003) supra note 10 at 90.
20 ‘Der . . . in neuester Zeit rezipierte Kelsensche Hauptbegriff der Vửlkerrechtsquelle als Rechts- erzeugungsart’ Alfred Rub, Hans Kelsens Vửlkerrechtslehre. Versuch einer Wỹrdigung (1995) 335.
21 ‘der Vorstellung, sie seien nur deklarative Manifestationen ohne ihr Hinzutreten bereits im Rechtsbewusstsein entstehenden Rechts’; Rub (1995) supra note 20 at 310.
22 Alfred Verdross, Vửlkerrecht (3rd ed. 1955) 118.
23 ‘formalisierte Erzeugungsarten’; Afred Verdross, Bruno Simma, Universelles Vửlkerrecht. Theorie und Praxis (3rd ed. 1984) 321.
24 Rub disagrees: ‘Kelsen’s term “source of international law” is . . . the same as Verdross’ and Simma’s term “formal source of international law” ’ ‘Kelsens Begriff der Vửlkerrechtsquelle ist . . . der gleiche wie der von Verdross/Simma vertretene Begriff der formellen Rechtsquelle’; Rub (1995) supra note 20 at 312.
25 Peter Fischer, Heribert Franz Kửck, Allgemeines Vửlkerrecht (6th ed. 2004) 69; Gerald G.
Fitzmaurice, Some problems regarding the formal sources of international law, in: F.M. van Asbeck et al. (eds), Symbolae Verzijl. Présentées au Prof. J.H.W. Verzijl á l’occasion de son LXX- ième anniversaire (1958) 153–176 at 154; Georg Schwarzenberger, International law (3rd ed. 1957) Volume 1, 25–27; Robert Y. Jennings, Arthur Watts (eds), Oppenheim’s international law (9th ed.
1992) Volume 1, 23 (paras 8–9); Hanspeter Neuhold, Waldemar Hummer, Christoph Schreuer (eds), ệsterreichisches Handbuch des Vửlkerrechts (4th ed. 2004) 31 (RN 166); Clive Parry, The sources and evidences of international law (1965) 4; Malcolm N. Shaw, International law
6.1.1 A constitution for international law 199
or to explicitly deny that sources are norms.26 This is based on the idea that sources are not meta-law, but facts or evidences. Ulrich Fastenrath argues that
‘empirical positivism’ (including Herbert Hart and his followers) sees sources ‘only as empirical description of procedures which usually create norms, which are regularly obeyed . . . or are seen as being obligatory’.27 Often international legal scholars do not elaborate on what is the basis of the designation of sources as mere ‘methods’ and might even be unaware of what such a statement entails.
When we stop worrying about international law once we get to the level of sources, international law is not adequately theoretically based. If we define sources as that which usually creates norms, how would we know a source when sources themselves determine when a norm is created? It would be begging the question, because it is a creation according to rules (according to meta-rules of rule-creation) that a norm can be the basis of validity of another norm. Defining sorces as the usual processes invariably means transcending the Is–Ought dichotomy, which, in turn, makes cognition of norms possible in the first place.
(2) In scholarly writings ‘the sources of international law’ are sometimes con- sidered as somehow residing on an absolute level.28 Only on that absolute level is international law created (which explains the propensity to deny that the sources are norms themselves) and above them we will find nothing but a doctrine of the basis of obligations of international law. It is indicative in this respect that subordinate sources, such as certain decisions of international organisations, are sometimes held to be co-equals to treaty norms as source of international law,29 even though the creation of the decisions is authorised by a treaty. However, if we adopt Merkl’s Stufenbau (Sections 5.5.2 and 5.5.3), we cannot see ‘sources’
as an absolute level. Law-creation within a normative order is done on the basis of norms empowering norm-creation and thus law-creation is always also
(6th ed. 2008) 69–71; Helmut Strebel, Quellen des Vửlkerrechts als Rechtsordnung, 36 Zeitschrift fỹr auslọndisches ửffentliches Recht und Vửlkerrecht (1976) 301–346 at 302–303; Michel Virally, The sources of international law, in: Max Sứrensen (ed.), Manual of public international law (1968) 116–174 at 120.
26 Maarten Bos, The recognised manifestations of international law. A new theory of ‘sources’, 20 German Yearbook of International Law (1977) 9–76 at 10–11; Antonio Cassese, International law (2nd ed. 2005) 154–155; Gihl (1957) supra note 14 at 83; Peter Hulsroj, Three sources – no river:
A hard look at the sources of public international law with particular emphasis on custom and
‘general principles of law’, 54 Zeitschrift fỹr ửffentliches Recht (1999) 219–259 at 234; Alf Ross, A textbook of international law (1947) 80, 83; A.J.P. Tammes, Inter-action of the sources of inter- national law, 10 Netherlands International Law Review (1963) 225–238 at 225–227; Verdross and Simma (1984) supra note 23 at 323–324.
27 ‘lediglich als empirische Beschreibung der Verfahren, in denen üblicherweise Normen entstehen, die regelmọòig befolgt werden . . . bzw. als verpflichtend erlebt werden’; Ulrich Fastenrath, Lỹcken im Vửlkerrecht. Zu Rechtscharakter, Quellen, Systemzusammenhang, Methodenlehre und Funktionen des Vửlkerrechts (1991) 86.
28 Hints of this may be found in: Jennings and Watts (1992) supra note 25 at 15 (para 5), 23 (para 9);
Fitzmaurice (1958a) supra note 25 at 154.
29 Bos (1977) supra note 26.
Uncertainty in International Law 200
law-application (with the exception of the Grundnorm). Merkl’s ‘Janus-face of law’
(doppeltes Rechtsantlitz)30 is an apposite metaphor in this respect.
(3) Another argument, mentioned above, is to distinguish the sources of inter- national law and the basis of obligation of international law. ‘Commentators . . . do not seem to have a clear idea of the signification of the word “sources”. Most of them have confused it with the foundation of international law,’31 writes Menon, for example. Apart from wishing to keep apart chapters one and two of international law textbooks on the subject, what would be the reason for such a distinction? This contention can be made when we deny that the source of a norm is also its source of ‘bindingness’:
In the domestic field there is a fairly close identity, or apparent identity, between the source of the obligation (the obligation to obey a given rule) and the source of the law (i.e. of the rule itself) – between what the rule is, and what makes it law.32
The implication here is that this does not obtain in international law. As argued throughout this book, on the basis of a normativist-positivist legal theory the existence, validity, ‘bindingness’ or ‘obligatoriness’ of norms is one and the same thing and cannot be distinguished. On the approach adopted here, the concept of a ‘non-valid’ or ‘non-binding’ norm is a contradiction in terms. Therefore, the source of the validity of a norm is that norm’s source properly speaking.33 The origin of the content of one or the other norm may be percieved as ‘external’ to the normative order in question. The prohibition of operating a motor vehicle while intoxicated in many domestic legal systems may ‘originate’ from the wish of a large part of the population to reduce the number of road traffic accidents caused by intoxicated drivers, and as a historical explanation this may be accurate.
But in a legal explanation, legal scholarship’s view can only be on the influence of norms on other norms. Legal scholarship would explain the origin of that law by pointing to the provision of the constitution empowering parliament to create statutes, and to the act of will whose sense is the enactment of the prohib- ition. A mixture of incompatible methods, e.g. of the legal-normative with the sociological-empirical, is a syncretism of method (Methodensynkretismus) which hinders scholarly cognition.
30 Adolf Julius Merkl, Das doppelte Rechtsantlitz. Eine Betrachtung aus der Erkenntnistheorie des Rechtes, 47 Juristische Blọtter (1918) 425–427, 444–447, 463–465, reprinted in: Hans Klecatsky, Renộ Marcic, Herbert Schambeck (eds), Die Wiener rechtstheoretische Schule. Ausgewọhlte Schriften von Hans Kelsen, Adolf Julius Merkl, Alfred Verdross (1968) 1091–1113.
31 P.K. Menon, An enquiry into the sources of modern international law, 64 Revue de Droit Inter- national, de Sciences Diplomatiques et Politiques (1986) 181–214 at 181. See also: Percy E.
Corbett, The consent of states and the sources of the law of nations, 6 British Year Book of International Law 1925 (1925) 20–30; Hoof (1983) supra note 3; Parry (1965) supra note 25 at 4–5.
32 Fitzmaurice (1958a) supra note 25 at 155 (emphasis added).
33 Robert Y. Jennings, What is international law and how do we tell it when we see it?, 37 Schweizerisches Jahrbuch für internationales Recht 1981 (1982) 59–88 at 60; Petev (1984) supra note 15 at 273; Virally (1968) supra note 25 at 118.
6.1.1 A constitution for international law 201