I am grateful to Anne van Aaken, Michael Anderheiden, Jürgen Bast, Mátyás Bódig, Péter Cserne, Stefan Häußler, Cristina Hoss, Jórg Kammerhofer, András Karácsony, Stephan Kirste, Jana Lachmund, Nele Matz, Silja Vóneky, Robert Walter, and CJLJ's anonymous reviewer for their thoughtful criticism and valuable comments. See Editor's note at footnote 181.

1. The words ‘Pure Theory of Law’ refer here not to Kelsen’s book with the same title, but to the whole Vienna School and its doctrine. For the same terminology, see Walter, Robert, ‘Der gegenwärtige Stand der Reinen Rechtslehre’ (1970) Rechtstheorie 69 Google Scholar; and the founding document of the Hans Kelsen-Institut (Vienna), see Hans Kelsen zum Gedenken (1974) at 77. Kelsen first called his doctrine ‘reine Rechtslehre’ (with a small ‘r’) in the subtitle of his Das Problem der Souveränität und die Theorie des Völkerrechts. Beitrag zu einer reinen Rechtslehre (1920).

2. Kucsko-Stadlmayer, Gabriele, ‘Der Beitrag Adolf Merkls zur Reinen Rechtslehre’ in Walter, Robert, ed, Schwerpunkte der Reinen Rechtslehre (1992) at 107 Google Scholar et seq.; Grussmann, Wolf-Dietrich, Adolf Julius Merkl. Leben und Werk (1989) at 14 Google Scholar and 18 et seq. with further references.

3. Merkl, Adolf, ‘Das doppelte Rechtsantlitz’ (1918) Blätter, Juristische at 425 Google Scholar et seq., 444 et seq. and 463 et seq.

4. Adolf Merkl, Die Lehre von der Rechtskraft entwickelt aus dem Rechtsbegriff (1923).

5. Merkl, Adolf, ‘Prolegomena einer Theorie des rechtlichen Stufenbaues’ in Verdross, Alfred, ed., Gesellschaft, Staat und Recht. Untersuchungen zur reinen Rechtslehre. Festschrift Hans Kelsen zum 50. Geburtstag gewidmet (1931) at 252 Google Scholar et seq. It was, however, as indicated in the title (Prolegomena = Preface), only an introductory article, and Merkl planned, as he himself mentioned (see ibid. at 294), to write a whole monograph on this topic. This announced major work was never written.

6. It has to be mentioned that the Stufenbaulehre of Merkl (particularly in its methodological premises) draws strongly on Hans Kelsen’s Hauptprobleme der Staatsrechtslehre entwickelt aus der Lehre vom Rechtssatze (1911). Hence one speaks best of a ‘cross-fertilisation’. Similarly Jürgen Behrend, Untersuchungen zur Stufenbaulehre Adolf Merkls und Hans Kelsens (1977) at 49.

7. Kelsen, Hans, Hauptprobleme der Staatsrechtslehre entwickelt aus der Lehre vom Rechtssatze 2. Aufl. (1923) at XV Google Scholar et seq. acknowledges the contribution of Merkl to the development of the Pure Theory of Law explicitly.

8. Kelsen, Hans, Introduction to the Problems of Legal Theory (1992), trans. by Litschewski Paulson, Bonnie & Paulson, Stanley L. (Oxford: Clarendon Press, 1992)Google Scholar. [Reine Rechtslehre (1934)] at 55 et seq. [hereinafter Introduction].

9. As Kelsen writes in a letter to Merkl, ‘…if only there is a ‘Vienna School of legal theory’, it is, in a very great part, due you’, cited by Grussmann, Wolf-Dietrich in Walter, Robert, ed., Adolf J. Merkl. Werk und Wirksamkeit (1990) at 142 Google Scholar; see also Kelsen, Hans, ‘Adolf Merkl zu seinem 70. Geburtstag’ (1960) Zeitschrift für öffentliches Recht 313 Google Scholar. See also Kelsen’s letter to Merkl, where he acknowledges ‘most gratefully the great importance of his scholarly work for the Pure Theory of Law’ in Imboden, Max, et al., eds., Festschrift für Adolf J. Merkl zum 80. Geburtstag (1970) at 11 Google Scholar.

10. As Merkl, supra note 4 at 210 states: ‘If one wants to be able to present the chaos of legal forms as a sum of coherent phenomena, in a word as a legal system, as a cosmos of law, […] it has to be […] acknowledged as the outcome of a common origin’. Thus, unity of the legal order is to be found in the chain of delegations (Delegationszusammenhang). For this, cf. Merkl, Adolf, ‘Prolegomena einer Theorie des rechtlichen Stufenbaues [1931]’ in Klecatsky, Hans CrossRefGoogle Scholar et al., eds., Hans Kelsen—Adolf Merkl—Alfred Verdross: Die Wiener Rechtstheoretische Schule (1968) [hereinafter WRS] 1336; Kelsen, Hans, ‘Der Begriff der Rechtsordnung [1958]’ in WRS 1395 Google Scholar et seq.

11. Cf. Dahlmann, Christian, ‘The Trinity in Kelsen’s Basic Norm Unravelled’ (2004) ARSP 149 Google Scholar.

12. Behrend supra note 6 at 68. On the double case of the Pure Theory of Law against the role of, first, causal sciences, e.g., sociology and second of natural law in legal science, see Dreier, Horst, Rechtslehre, Staatssoziologie und Demokratietheorie bei Hans Kelsen (1986) at 42 Google Scholar.

13. Merkl, ‘Prolegomena’, supra note 10 at 1311. Every form may have an arbitrary content, see Merkl, ibid. at 1313; Kelsen, Hans, The Pure Theory of Law, trans. by Knight, Max (Berkeley: University of California Press, 1967)Google Scholar [Reine|Rechtslehre, 2. Aufl., 1960] [hereinafter PTL2] at 201. Thus, every topic may be theoretically (i.e., if the concrete legal system does not impose something else, e.g., in the constitution) regulated by any kind of sources of law, see Behrend, supra note 6 at 54, n. 191.

14. In the terminology of Merkl: original norm (Ursprungsnorm); see e.g., Merkl, supra note 4 at 209, n. 1. Kelsen himself uses the expression original norm sometimes; e.g., in his Allgemeine Staatslehre (1925) at 99 [hereinafter AStL]. The idea of the basic norm (Grundnorm) (but not the expression) appears in Kelsen’s works before the Stufenbaulehre; first in his article ‘Reichsgesetz und Landesgesetz nach österreichischer Verfassung’ (1914) Archiv des öffentlichen Rechts (1914) 216 et seq., but the article of Verdross, Alfred, ‘Zum Problem der Rechtsunterworfenheit des Gesetzgebers’ (1916) Juristische Blätter at 471 Google Scholar et seq. contributed to its final elaboration (the distinction between the constitution of positive law and the basic norm as constitution in a legal-logical sense), as it was explicitly acknowledged by Kelsen, supra note 7 at XV et seq. Perhaps this is why Kelsen dedicated the second edition of the Hauptprobleme to Merkl and Verdross (ibid. at III). On the role of Verdross, see also Walter, Robert, ‘Entstehung und Entwicklung des Gedankens der Grundnorm’ in Walter, Robert, ed., Schwerpunkte der Reinen Rechtslehre (1992) at 51 Google Scholar.

The expression ‘basic norm’ comes from Husserl, Edmund, Logische Untersuchungen. 1. Prolegomena zur reinen Logik (1900) at 45 Google Scholar, see Holzhey, Helmut, ‘Kelsens Rechts- und Staatslehre in ihrem Verhältnis zum Neukantianismus’ in Paulson, Stanley L. & Walter, Robert, eds., Untersuchungen zur Reinen Rechtslehre. Ergebnisse eines Wiener Rechtstheoretischen Seminars 1985/86 (1986) at 177 Google Scholar et seq. Cf. Krawietz, Werner, ‘Grundnorm’ in Ritter, Joachim, ed., Historisches Wörterbuch der Philosophie, Bd. 3, 921Google Scholar et seq.

Just to be sure: the basic norm of the valid U.S. legal system is not the Constitution (1787) (which is the constitution of positive law), but the norm that says: ‘The Constitution (1787) ought to be obeyed’.

15. Kelsen, Introduction, supra note 8 at 57.

16. Ibid. at 61: ‘while the law cannot exist without power, neither is it identical to power’.

17. Kelsen, AStL, supra note 14 at 248 et seq. Another synonym is ‘the constitution in the transcendental-logical sense’, see Kelsen, Hans, ‘Naturrechtslehre und Rechtspositivismus [1961]’ in WRS note 10 at 827 Google Scholar; Kelsen, Hans, ‘Die Funktion der Verfassung’ [1964] in WRS note 10 at 1976 Google Scholar. On synonymity, see Bindreiter, Uta, Why Grundnorm? A Treatise on the Implications of Kelsen’s Doctrine (2003) at 18 Google Scholar.

18. Single levels can also be skipped (an individual act may also be, for example, based directly on the statute and does not have to gain its validity by the ordinance as an intermediate) as Hans Nawiasky rightly point out in his ‘Kritische Bemerkungen zur Lehre vom Stufenbau des Rechtes’ (1927) Zeitschrift für öffentliches Recht at 495. In the same way Raz, Joseph, The Concept of a Legal System. An Introduction to the Theory of Legal System, 2nd ed. (Oxford: Clarendon Press, 1980) at 99, n. 1CrossRefGoogle Scholar, who therefore prefers the metaphor of a tree to the usual image of a pyramid. This is, however, only a slight addition to the Stufenbaulehre, and cannot refute it at all.

19. Merkl, supra note 4 at 209 et seq.; idem, ‘Gesetzesrecht und Richterrecht [1922]’ in WRS note 10 at 1618; on relevant modifications in the work of Merkl, see Behrend, supra note 6 at 19 et seq. Moreover, according to Behrend one needs at least three levels to make a legal order: the two mentioned by Merkl and the norm of competence for the creation of the coercive norm; see Behrend supra note 6 at 26. The argument of Behrend assumes, however, that one wants to maintain the possibility of a later modification of coercive norms. It is—although rational—not conceptually necessary; or, in the usual expression of the Pure Theory of Law, not essential to law (rechtswesentlich).

20. Merkl, supra note 4 at 216; Merkl, Adolf, ‘Das doppelte Rechtsantlitz [1918]’ in WRS“ note 10 at 1097 Google Scholar.

21. The reason for this is not the explicitly approved discretion, but the fact that linguistic expressions are themselves indefinite to a certain extent; see Merkl, ‘Rechtsantlitz’, ibid. at 1111: ‘Every word has, along with a definite core of meaning, more or less wavering peripheral meanings’ (emphasis added)—in the year 1918, a long time before Hart.

22. Merkl, Adolf, Allgemeines Verwaltungsrecht (1927) at 142, 146CrossRefGoogle Scholar. Achterberg, Norbert, ‘Hans Kelsens Bedeutung in der gegenwärtigen deutschen Staatslehre’ (1974) Die öffentliche Verwaltung at 454 Google Scholar explains that the pyramid of norms corresponds to an opposite pyramid showing the role of extra-legal factors: it is the broadest in the constitution, and the narrowest in the coercive act.

23. Merkl, supra note 22 at 142.

24. Behrend, supra note 6 at 51. In this aspect, Ernst Rudolf Bierling, Juristische Prinzipienlehre, Bd. II, 1898, 117 et seq., esp. 119 and 127 may be mentioned as one of the predecessors.

25. Öhlinger, Theo, Der Stufenbau der Rechtsordnung. Rechtstheoretische und ideologische Aspekte (1975) at 10 Google Scholar.

26. Particularly Binding, Bierling and Thon. Cf. Müller-Freienfels, Wolfram, ‘Zur Rangstufung rechtlicher Normen’ in Law in East and West. Recht in Ost und West (1988) at 19 (with further references) and 21Google Scholar.

27. Öhlinger, supra note 25 at 11.

28. Paulson, Stanley L., ‘Zur Stufenbaulehre Merkls in ihrer Bedeutung für die Allgemeine Rechtslehre’ in Merkl., Adolf J. Werk und Wirksamkeit, ed. by Walter, Robert (1990) at 93 Google Scholar et seq.

29. Merkl, supra note 10 at 1314. Kelsen himself emphasises this advantage, see Kelsen, supra note 7 at XII.

30. Kelsen’s inconsequence occasionally gives another ground for criticism, especially the fact that sometimes he refers to the legal order as a condition of the basic norm, but sometimes he does its inverse. See Dreier, supra note 12 at 45, n. 117 with further references.

31. Such criticism, e.g., Dreier, Ralf, ‘Bemerkungen zur Theorie der Grundnorm’ in Die Reine Rechtslehre in wissenschaftlicher Diskussion (1982) at 38 Google Scholar et seq., esp. 45.

32. We do not have to discuss here the criticism of Hart, H.L.A., The Concept of Law (Oxford: Clarendon Press, 1961) at 245 Google Scholar et seq. (ch. VI, n. 1), who challenges Kelsen’s basic methodical approach and therefore states—among other points—that the basic norm is simply ‘redundant’. The criticism of Werner Krawietz, ‘Die Lehre vom Stufenbau des Rechts—eine säkularisierte politische Theologie?’ in Krawietz, Werner & Schelsky, Helmut, eds., Rechtssystem und gesellschaftliche Basis bei Hans Kelsen Rechtstheorie (1984) 5 Beiheft at 264Google Scholar et seq. remains also out of analysis, who claims that the Stufenbaulehre cannot grasp ‘social reality’, since it pretends that decisions are determined from the top to the bottom—although there is another, inverse way of influence, also omitted because this criticism points to a problem of sociology (of law), while the Stufenbaulehre (as a purely normative theory) says nothing about it.

33. E.g., Larenz, Karl, Methodenlehre der Rechtswissenschaft, 4. Aufl., 1979, 79CrossRefGoogle Scholar et seq.; Hippel, Ernst v., Allgemeine Staatslehre (1963) at 146 Google Scholar; Schreiber, Hans-Ludwig, Der Begriff der Rechtspflicht (1966) at 144 Google Scholar; Peczenik, Aleksander, ‘Two Sides of Grundnorm’ in Die Reine Rechtslehre in wissenschaftlicher Diskussion (1982) at 61 Google Scholar.

34. See Hammer, Stefan, ‘Kelsens Grundnormkonzeption als neukantische Erkenntnistheorie des Rechts?’ in Paulson, Stanley L. & Walter, Robert, eds., Untersuchungen zur Reinen Rechtslehre. Ergebnisse eines Wiener Rechtstheoretischen Seminars 1985/86 (1986) at 225 Google Scholar; Dreier, supra note 12 at 51, n. 138 with further references. For similar reasons (namely because of the supposition of a non-empirical, transcendental basic norm), see Hoerster, Norbert, ‘Kritischer Vergleich der Theorien der Rechtsgeltung von Hans Kelsen und H. L. A. Hart’ in Untersuchungen zur Reinen Rechtslehre. Ergebnisse eines Wiener Rechtstheoretischen Seminars 1985/86, ed. by Paulson, Stanley L. & Walter, Robert (1986) at 1 Google Scholar et seq., esp. 18 finds Kelsen’s theory less plausible than that of Hart. I think that the argument of Hoerster is a strong one, however, I am not going to discuss it here, since (as is emphasised by Hoerster himself) it cannot finally refute the Stufenbaulehre, but only weaken it.

35. Kelsen, Hans, ‘Die philosophischen Grundlagen der Naturrechtslehre und des Rechtspositivismus [1928]’ in WRS note 10 at 294 Google Scholar; Kelsen, Hans, General Theory of Law and State (1949) at 437 Google Scholar [hereinafter GTL]. The idea of the basic norm does, so to speak, offer itself to be rewritten in the spirit of Natural Law. For such an attempt see, e.g., Marcic, René, ‘Das Naturrecht als Grundnorm der Verfassung’ (1964) Zeitschrift für öffentliches Recht at 69 Google Scholar et seq.

36. Kelsen, Hans, ‘Die Funktion der Verfassung, [1964]’ in WRS note 10Google Scholar (1975) et seq. He puts it in the same way in his posthumously published work: Kelsen, Hans, General Theory of Norms (Oxford: Oxford University Press, 1991), trans. by Hartney, Michael [Allgemeine Theorie der Normen, ed. by Ringhofer, Kurt & Walter, Robert (1979)] at 258 CrossRefGoogle Scholar et seq. [hereinafter GTN].

37. Raz, Joseph, ‘Kelsen’s Theory of the Basic Norm’ (1974) 19 Am. J. Juris. 99 CrossRefGoogle Scholar.

38. Herzog, Roman, Allgemeine Staatslehre (1971) at 90 Google Scholar.

39. Cf. Lippold, Rainer, Recht und Ordnung. Statik und Dynamik der Rechtsordnung (2000) at 499 et seq., esp. 505Google Scholar.

40. Dreier, supra note 12 at 45, n. 117. Because of this contradictory nature, Priester calls the basic norm a chimera, i.e., something that has many features at the same time, which are not compatible with each other. Cf. Priester, Jens-Michael, ‘Die Grundnorm als Chimäre’ in Krawietz, Werner & Schelsky, Helmut, eds., Rechtssystem und gesellschaftliche Basis bei Hans Kelsen Rechtstheorie (1984) 5 Beiheft, 238Google Scholar. The chimera has the head of a lion, the back of a dragon, and the trunk of a goat. According to Priester, Kelsen’s arguments about the basic norm are similarly ‘manifold’.

41. Here I follow principally Dreier, supra note 12 at 42 et seq.

42. According to Kant (and the neo-Kantian Kelsen), ‘transcendental’ refers to the epistemological method (i.e., a priori) and ‘transcendent’ to the knowledge that is beyond experience. See Immanuel Kant, Kritik der reinen Vernunft.

43. Kelsen, PTL2, supra note 13 at 218; Behrend, supra note 6 at 65.

44. Dreier, supra note 12 at 49.

45. Hoerster, supra note 34 at 2.

46. Kelsen, Hans, ‘Recht, Rechtswissenschaft und Logik’ (1966) ARSP at 547 Google Scholar (emphasis in the original).

47. Dreier, supra note 12 at 47, n. 119.

48. Kelsen, GTLS, supra note 35 at 413.

49. Kelsen, GTN, supra note 36 at 255: ‘Only a norm can be the reason for the validity of another norm.’

50. On the identification of existence, validity and binding force in Kelsen, see infra note 125.

51. I.e., a non-proved supposition becomes sine qua non of scientific knowledge. It seems at first paradoxical, but a great deal of epistemological considerations work this way. See e.g., Quine, Willard V. O., ‘On Empirically Equivalent Systems of the World’ (1975) Erkenntnis at 313 Google Scholar et seq.

52. Raz, supra note 37 at 104 et seq. esp. 107 and 109; Harris, J.W., ‘When and Why does the Grundnorm Change?’ (1971) 29 Cambridge L. J. 103 at 117, n. 57aCrossRefGoogle Scholar: ‘The grundnorm is postulated by Kelsen as something logically essential to explain the practice of legal scientific discourse.’

53. Walter, Robert, ‘Wirksamkeit und Geltung’ (1961) Zeitschrift für öffentliches Recht at 539 Google Scholar et seq.

54. Cf. Walter, supra note 1 at 73.

55. Walter, ibid. at 81; Behrend, supra note 6 at 71.

56. Walter, ibid. at 80 (emphasis added).

57. See also infra note 125.

58. Walter, supra note 14 at 56 et seq.

59. Walter, supra note 1 at 80 et seq.

60. Behrend, supra note 6 at 70.

61. This means that for, e.g., the scientific analysis of the valid German legal order, one does not choose a basic norm that refers to the legal order of the Weimar republic. Thus, if one wants to work scientifically with the valid German legal order, one should not suppose the basic norm to be ‘The Weimar constitution is valid.’ See also supra note 14.

62. Cf. Pitamic, Leonidas, ‘Denkökonomische Voraussetzungen der Rechtswissenschaft’ (1917) Zeitschrift für öffentliches Recht at 347 Google Scholar. On the role of Pitamic in the development of the idea of the basic norm, see Kelsen, supra note 7 at XV; Walter, supra note 14 at 52.

63. Kelsen, Hans, Die philosophischen Grundlagen der Naturrechtslehre und des Rechtspositivismus [1928] in WRS note 10 at 294 Google Scholar.

64. This means that it is not necessary for the existence of the legal order to suppose the basic norm (to put it somewhat sharply: there also have been legal orders before the Pure Theory of Law), only for its scientific analysis. A scientific analysis does not give an answer to the question about the existence of the legal order anyway, but only presupposes it. In fact, the methodical starting points (e.g., moral relativism) are in general of metaphysical nature—but this is a general characteristic of all epistemological starting points. Cf. supra note 51.

65. Dreier, supra note 12 at 54, n. 160. This juristic self-obligation, however, does not mean that a given legal order is somehow morally justified, and lawyers are morally obliged to abide by it, since this is a moral question, and thus definitely eliminated by Kelsen from his theory of law

66. Objective validity means here a validity that is independent from the subjective will of the norm-poser. Cf. also supra note 15.

67. Dreier, supra note 12 at 55. Cf. Kelsen, Introduction, supra note 8 at 58; Kelsen, PTL2, supra note 13 at 204 et seq., that the basic norm expresses only what positivists have always thought.

68. But only if the methodological starting points are accepted (so especially the elimination of sociological and natural-law elements, and the objectivity of science). We cannot examine the question of rejection here.

The defence is a conditional one also in the sense that not all (possible) objections have been answered. (1) Especially, the sophisticated anti-Kelsenian example of Raz, supra note 37 at 98, was not discussed. Raz mentions a colony that becomes independent in peace, its constitution of independence being given by its mother country. In this case, the new constitution can be derived from the legal order of the mother country, i.e., from the basic norm of the mother country—although the former colony considers itself (as every other country, including its mother-country consider it) as an independent country that should thus have an independent basic norm. I.e., it is not an autonomous legal order after all. The objection of Raz can be overcome if one adopts the monism of international law—as does Kelsen himself—and hence considers all legal norms to be parts of a single huge legal order: in this way, arguments based on the ‘independence of a legal order’ are neutralised. This solution may, however, be problematic from other aspects, see below 1.2.3 The Indefensibility of Monism. (On the other anti-Kelsenian example of Raz, see infra note 105.) (2) The criticism of Tammelo, Ilmar, ‘Von der reinen zu einer reineren Rechtslehre’ in Krawietz, Werner & Schelsky, Helmut, eds., Rechtssystem und gesellschaftliche Basis bei Hans Kelsen Rechtstheorie (1984) 5 Beiheft, 252Google Scholar, is also worth mentioning: a plurality of basic norms, could also be logically imagined, as several logical systems exist with more axioms. Tammelo is undoubtedly right on this point. But if it is not about previously given basic norm(s), but constructions assumed by legal scholars themselves, then it is much more elegant to minimize the number of (epistemological) assumptions. Therefore, the objection of Tammelo is not very persuasive.

69. Kelsen, AStL, supra note 13 at 235.

70. On the logical difference in Hungarian legal order see András Jakab, A jogszabálytan fó´bb kérdéseiró´l [On the Main Questions of the Theory of Normative Legal Acts] (2003) at 21 et seq.

71. Walter, Robert, Der Aufbau der Rechtsordnung (1964) at 40 Google Scholar et seq.

72. ‘Concrete’ means specialis, therefore its sphere of validity is a narrower one, see Jakab, supra note 70 at 43.

73. On the debates about the question cf. Kunz, Josef L., Völkerrechtswissenschaft und reine Rechtslehre (1923) at 69 Google Scholar et seq. On Kelsen’s theory of international law in general, see Rub, Alfred, Hans Kelsens Völkerrechtslehre. Versuch einer Würdigung (1995)Google Scholar.

74. Kelsen, supra note 1 at 123. Cf. also Jochen v. Bernstorff, Der Glaube an das universale Recht. Zur Völkerrechtstheorie Hans Kelsens und seiner Schüler (2001) at 70 et seq.

75. See also Mittelstraß, Jürgen, Enzyklopädie Philosophie und Wissenschaftstheorie, Bd. 2. (1984) at 989 Google Scholar: the typically neo-Kantian claim of the ‘unity of the object of knowledge’.

76. Kelsen, AStL, supra note 14 at 121: ‘…the necessary unity of the normative system. Two orders, as legal orders, cannot be declared as valid unless their validity is somehow derived from a unitary reason of validity.’

77. See e.g., Kelsen, Hans, Principles of International Law (1952) at 446 Google Scholar et seq. Cf. v. Bernstorff, supra note 74 at 93. For a detailed presentation of this argument, see Verdross, Alfred, ‘Völkerrecht und einheitliches Rechtssystem. Kritische Studie zu den Völkerrechtstheorien von Max Wenzel, Hans Kelsen und Fritz Sander’ (1923) Zeitschrift für Völkerrecht at 415 Google Scholar et seq.

78. On the changes of formulation of the basic norm of international law in Kelsen’s oeuvre, see Rigaux, François, ‘Hans Kelsen on international law’ (1998) 9 Eur. J. Int’l Law 325 CrossRefGoogle Scholar.

79. As shown by Bleckmann, Albert, ‘Monismus mit Primat des Völkerrechts. Zur Kelsenschen Konstruktion des Verhältnisses von Völkerrecht und Landesrecht’ in Krawietz, Werner & Schelsky, Helmut, eds., Rechtssystem und gesellschaftliche Basis bei Hans Kelsen Rechtstheorie (1984) 5 Beiheft at 340 et seq.Google Scholar

80. Bleckmann, ibid. at 339 with further references.

81. A plurality of basic norms, is used by Schilling, Theodor, ‘Zum Verhältnis von Gemeinschaftsund nationalem Recht’ (1998) Zeitschrift für Rechtsvergleichung at 149 Google Scholar et seq.

82. Öhlinger, supra note 25 at 17.

83. Act Nr. XI of 1987 on Legislation.

84. Federal Act on the Federal Official Journal of 1996 (BGBl 1996/660).

85. Öhlinger, supra note 25 at 17.

86. It is also acknowledged by Walter, supra note 71 at 63. A radical claim of Öhlinger, supra note 25 at 17 et seq. is based on this. According to Öhlinger, the hierarchy of legal order according to the legal conditioning (i.e., the original form of the Stufenbaulehre) is useless for legal theory. After this, Walter, Robert, ‘Die Lehre vom Stufenbau der Rechtsordnung’ (1980) Archivum Iuridicum Cracoviense at 9 Google Scholar refined his view: ‘What I have told is that there are regularly only two different levels of conditions in law-making; it does not preclude the possibility of further levels. Would there be any of them, then Öhlinger too should consequently acknowledge a “hierarchy” according to the conditions of law-making.’ To the argument of Walter, the following answer could be given, according to Öhlinger’s logic: Even if one assumes that validity of a norm stems from one single norm in some cases, no hierarchy emerges there, since the next level is probably (in Walter’s words: ‘regularly’) again an infinitely complicated network. This means that we have only one moment of sanity (a level of sanity) before going back to the jungle again.

87. Stoitzner, Bettina, ‘Die Lehre vom Stufenbau der Rechtsordnung’ in Paulson, Stanley L. & Walter, Robert, eds., Untersuchungen zur Reinen Rechtslehre. Ergebnisse eines Wiener Rechts-theoretischen Seminars 1985/86 (1986) at 58 Google Scholar; Mayer, Heinz, ‘Die Theorie des rechtlichen Stufenbaues’ in Walter, Robert, ed., Schwerpunkte der Reinen Rechtslehre (1992) at 40 Google Scholar et seq.

88. Walter, Robert, Hans Kelsens Rechtslehre (1999) at 8 Google Scholar.

89. ‘Temporally differentiated’ refers to the distinction between the states of the norm in different moments.

90. Therefore, there are not only a few lines of derivation (as one could imagine after Öhlinger), but hundreds (!).

91. Raz, supra note 37 at 97.

92. Cf. Walter, supra note 71 at 62, who sees rightly that the hierarchy of the legal order according to the conditions of law-making is independent from the traditional levels of the hierarchy of norms (e.g., constitution, statute), as the latter is the hierarchy of the legal order according to the derogatory power. See section 1.3 for a more detailed account.

93. See section 1.2.1.

94. Therefore, one can always consider international affairs as an anarchy that is only subject to the principle of power, see Kelsen, Hans, Law and Peace in International Relations (Cambridge: MA: Harvard University Press, 1942) at 48, 54Google Scholar.

95. One may well ask whether these interpretive schemes (i.e., legal orders) are the same at all (at least in their content). There are good reasons for claiming that this is not exactly the case, since interpretation of law, according to Kelsen, always includes a subjective element, see Kelsen, PTL2, supra note 13 at 348 et seq. It means, then, that every legal scholar has a partially different interpretive scheme in his mind. These general problems of the Pure Theory of Law cannot, however, be discussed here at length.

96. Kelsen has adopted this from Pitamic, supra note 62 at 339 et seq., as he confirms it in his work: Kelsen, supra note 7 at XV

97. Cf. supra note 61.

98. Cf. Kelsen, GTLS, supra note 35 at 111, adopted from Achterberg, supra note 22 at 453; Behrend supra note 6 at 77.

99. Kelsen, PTL2, supra note 13 at 211 et seq.

100. As opposed to the first edition, Kelsen, Introduction, supra note 8 at 60 et seq., where only the validity of the legal order as a whole was dependent on the efficacy of the legal order as a whole, but not validity of every single norm from the efficacy of that single norm. But see on this problem Walter, Robert, ‘Bemerkungen zu Kelsen, Geltung und Wirksamkeit des Rechts’ in Walter, , Jabloner, & Zeleny, , eds., Hans Kelsens stete Aktualität. Zum 30. Todestag Kelsens (2003) at 35 Google Scholar.

101. Thus, it is not about the desuetude of the norm empowering the setting of norms.

102. Any other theoretical construction of explanation would contradict the Denkökonomie.

103. Disagreeing (as a strict Kelsenian, rejecting such implicit rules) see Walter, Robert, ‘Die Gewohnheit als rechtserzeugender Tatbestand’ (1963) Österreichische Juristen-Zeitung at 225 Google Scholar et seq.

104. It is underlined by Bacqué, Jorge A., ‘Stufenbau der Rechtsordnung oder Einebnung der Normenpyramide’ in Bulygin, Eugenio & Garzón Valdés, Ernesto, eds., Argentinische Rechtstheorie und Rechtsphilosophie heute (1987) at 115 Google Scholar.

105. This sophisticated anti-Kelsenian example can be found in Raz, supra note 37 at 99. (On another anti-Kelsenian example by Raz—challenging its plausibility—see supra note 68.)

106. One can also not protect the Pure Theory of Law with the argument that the rule regulating the creation of norms in the new customary law came to existence slowly by evolution, since this slow evolution would mean that the rule regulating the creation of customary law was posed by way of customary law. This is, however, impossible, since customary law was not acknowledged (i.e., did not exist) before its rule of creation came into existence in the country of our example.

It would also be implausible to consider the legal order, which acknowledges the customary law, as a new legal order, because the great majority of norms (i.e., all but the rules of customary law) are to be derived from the historically first constitution of the ‘old’ legal order. To assume the existence of a new legal order, would be then, to a certain extent, a duplication of legal reality, and hence be contradictory to the principle of Denkökonomie.

107. It is persuasively proven by Klug, Ulrich, ‘Die Reine Rechtslehre von Hans Kelsen und die formallogische Rechtfertigung der Kritik an dem Pseudoschluss vom Sein auf Sollen’ in Engel, Salo, ed., Law, State, and International Legal Order. Essays in Honor of Hans Kelsen (Knoxville: University Press, 1964) at 153 Google Scholar et seq., esp. 156. He shows, without any doubt, that if one would like to have an ‘ought’ in the conclusion (of a syllogism), one also has to have an ‘ought’ in one of the premises.

108. Kelsen, GTN, supra note 36 at 255: ‘Only a norm can be the reason for the validity of another norm.’

109. This is an important difference between the legal order and the norms of a moral system, see Kelsen, Introduction, supra note 8 at 55 et seq.

110. Kelsen, supra note 7 at 411: ‘This is the great mystery of law and state, that takes place in the act of legislation….’

111. Thus, it is not only a simple descriptive definition that defines the notion of ‘valid norm’. In fact, it is a real ‘ought’ that prescribes which norm ought to be valid. In law, one often uses descriptive language, even if it is obviously a prescription. E.g., Art. 46, para. 3 of the B-VG: ‘The Federal President announces the national referendum.’ It is only a simplification in style.

112. It is very important to see that the rule of creation itself is a norm too. It is not merely a description like ‘If (performance of law-making actions), then (the law exists)’ or like ‘If parliament enacts the law (B ought_l to be done), then (B ought_l to be done)’. The latter seems to be a prescription because it contains an ‘ought’ (precisely: ‘ought_l’), but its actual form would be ‘If parliament enacts the law (B ought_l to be done), then the law (B ought_l to be done) is valid’ which would be merely a description and not a norm.

113. To be sure: in a Kelsenian way, this ‘abnormal exception’ may not be explained by saying that the legal order is only ‘by and large’ effective anyway, and that this is also true for the derivation of validity. Efficacy ‘by and large’ only refers to the accomplishment of the ‘is’ and not to the accomplishment in the ‘ought’. But as it was just shown, even the accomplishment in the Sollen (i.e., ‘ought’) is not an automatic (logical) necessity, only a high level of probability. This means that validity, in fact, cannot be derived. Such a moment of probability that refers to the world of ‘ought’ is very far from the ideas of the Stufenbaulehre: derivation (deduction) of validity excludes this kind of validity based on probability. But validity is actually based on probability. And: this accomplishment of norms does not refer to the physical ‘is’, but to the existence of norms in the ‘ought’ (thus, this is not the usual sociological objection).

114. This was an exclusively logical error and not an ontological one, since the law as well as the disposition ‘Law oughtc to exist’ are in the sphere of ‘ought’.

115. One could clearly refute the previous arguments by showing, by means of symbolic logic, how validity can be derived in legislation in a way that content is not derived at the same time. There was, until now, no such attempt, which is, in itself, not a proof but rather a call for discussion.

It is worth mentioning that, at the end of his life, Kelsen also rejected the derivation of validity of norms from norms (for other reasons than those considered in our argument). Cf. his posthumously published theory of norms, according to which the validity of the individual norm cannot be derived from the validity of the general norm, as an act of will is also needed; see Kelsen, GTN, supra note 36 at 237 et seq. More generally: validity of a norm cannot be derived from another norm; see Kelsen, ibid. at 423 et seq., n. 179; Priester, supra note 40 at 238. Kelsen, however, did not revise the Stufenbaulehre (what would follow from this); see esp. Kelsen, GTN, supra note 36 at 258 for his usual old content text.

On how logic became less and less important in the oeuvre of Kelsen, see van Eikema Hommes, Hendrik J., ‘The development of Hans Kelsen’s concept of legal norms’ in Krawietz, Werner & Schelsky, Helmut, eds., Rechtssystem und gesellschaftliche Basis bei Hans Kelsen Rechtstheorie (1984) 5 Beiheft at 159Google Scholar et seq.

116. Merkl, ‘Prolegomena’, supra note 10 at 1350; Koja, Friedrich, Allgemeine Staatslehre (1993) at 20 Google Scholar.

117. Merkl, Adolf, Die Unveränderlicheit von Gesetzen—ein normlogisches Prinzip [1917] in WRS note 10 at 1088 Google Scholar.

118. Merkl, supra note 4 at 259.

119. Öhlinger, supra note 25 at18.

120. Behrend supra note 6 at 42.

121. Walter, supra note 71 at 55 et seq. See Walter, Robert, ‘Die Reine Rechtslehre—eine Theorie in steter Entwicklung. Einige Klarstellungen’ (1994) Juristische Blätter at 494 Google Scholar. Cf. in terms of Jürgen Bast, EU-law, ‘Handlungsformen’ in von Bogdandy, Armin, ed., Europäisches Verfassungsrecht (2003) at 505 Google Scholar et seq.

122. Öhlinger, supra note 25 at 22; in the same way Walter, supra note 71 at 57 et seq., but with different words (‘complete annulment’ (vollständige Vernichtung), ‘to annul completely’ (gänzlich vernichten) and ‘to repeal irreversibly’ (unwiederbringlich beseitigen) on the one hand and ‘limited repeal’ (beschränkte Beseitigung) on the other hand; Mayer (n. 87), 43 with another different terminology (‘peremptory’ [endgültige] vs. ‘temporal [vorläufige] derogation’).

123. Walter, supra note 71 at 59.

124. Walter, ibid. at 55. But not necessarily vice versa: they may well stand on the same level, in spite of the different legislative processes. E.g., this may be the case with a statutory decree of former socialist countries or statutory instruments according to Henry VIII clauses in England, both of which are able to amend statutes though issued by the executive (thus not in a statute-making process).

125. See ‘validity’ as the mere existence of a norm: Kelsen, Introduction, supra note 8 at 12 and 71 et seq. on the validity of unconstitutional statutes. On the other hand as binding force: Kelsen, Introduction, supra note 8 at 56 et seq. Normally, Kelsen uses the words ‘Geltung’ and ‘Gültigkeit’ as synonyms, see Hoerster, supra note 34 at 8.

126. For a clear conceptual distinction, see also Wiederin, Ewald, Bundesrecht und Landesrecht (1995) at 51 CrossRefGoogle Scholar et seq.

127. This is indicated by the question mark in the title of this section after ‘an’, see 1.3 An(?)other Hierarchy.

128. A particular form of abolishment is the abolishment of existence, which is a kind of abolishment that has a retroactive effect on the beginning of the temporal sphere of validity.

129. Jakab, supra note 70 at 71 et seq. According to that, one can go further into details. Abolishment is in some legal orders (e.g., in Hungary, according to Art. 13 of the Act Nr. XI. of 1987 on legislation) only possible explicitly (formally), while in others also implicitly (materially) (e.g., in England implied repeal doctrine), i.e., in the case of a contradiction, the former norm of the same level is to be considered automatically as abolished.

130. See Marbury vs. Madison 1 Cranch 137 (U.S. 1803).

131. Annulment is actually a kind of abolishment, i.e., because of some violation of the law. It may happen for substantive as well as procedural reasons. It is the former that is of importance here, i.e., annulment because of a contradiction with the content of another norm, since—as it was shown above—the hierarchy based on the process (i.e., the Stufenbaulehre) offers no clear and tenable structure for the legal order.

132. In my view, this is the best way to give an account of the hierarchy of norms in the Austrian legal order, since the hierarchy according to abolishment cannot handle the fact that the federal constitution (Bundesverfassung) cannot entirely abolish the constitution of member states (Landesverfassung)—although it is described by the accounts of the hierarchy of Austrian legal order as superior to them, see e.g., Öhlinger, Theo, Verfassungsrecht, 5. Aufl., 2003, 26Google Scholar. This derogation would be contrary to the fundamental principles of the federal constitution (which are, according to the accounts of the hierarchy, superior to the federal constitution), see Öhlinger, supra note 25 at 25 with further references. Another special feature of the Austrian hierarchy of norms is that single dispositions may also have a hierarchic relation to each other (e.g., in certain statutes there are dispositions that are explicitly defined as having a constitutional status—see Art. 1 of the Datenschutzgesetz). In Hungary, on the contrary, only legal acts (e.g., statutes, ordinances) can have a hierarchic relation (and the criterion for the hierarchy is the derogatory power, but not the annulment by the constitutional court).

133. Jakab, supra note 70 at 74, n. 159.

134. There are two possible solutions: (1) Either one always expresses unequivocally, whichever kind of derogation is spoken of (e.g., derogation as abolishment or derogation as priority in application), (2) or one uses the notion of ‘derogation’ only for abolishment.

135. One of the advocates of this approach is Theo Öhlinger, a sharp critic of the Stufenbaulehre (the hierarchy of legal order according to conditions of law-making) (see above, 1.2.4 The Validity of a Norm Conditioned by One Single Other Norm), Öhlinger, Theo, Verfassungsrecht, 5. Aufl., 2003, 90Google Scholar; Öhlinger, Theo, ‘Unity of the Legal System or Legal Pluralism: The Stufenbau Doctrine in Present-Day Europe’ in Jyränki, Antero, ed., National Constitutions in the Era of Integration (1999) at 163 Google Scholar et seq. For further discussion of this approach, see Schramm, Alfred, ‘Zweistufige Rechtsakte—oder: Über Richtlinien und Grundsatzgesetze’ (2001) Zeitschrift für öffentliches Recht (2001) at 69, n. 11Google Scholar. These objections are surprising particularly because also (a part of) Austrian literature sees clearly that the word ‘derogation’ nowadays refers to more concepts; see Stoitzner, supra note 87 at 64, 67 et seq.; Öhlinger, supra note 25 at 22. I would like to add that the problem here is not that the use of notions is discussed on another level. In that case, Austrian literature should speak of two different ‘hierarchies of the legal order according to derogatory power’. One speaks, however, of a single hierarchy of legal order according to the derogatory power, although two completely different phenomena are meant by ‘derogation’.

136. See my more detailed criticism: András Jakab, ‘Az osztrák EU-csatlakozás alkotmányjogi szempontból’ [The Austrian EU-accession from the Perspective of Constitutional Law] (2002/1) Jogelméleti Szemle (http://jesz.ajk.elte.hu/) II./1.3.2.

137. But see a different view: Mayer, supra note 87 at 41 et seq.

138. This hierarchy (these hierarchies) is (are) ‘weaker’ also in the sense that individual norms (Einzelakte) are not included; see Stoitzner, supra note 87 at 71. A particular survey is needed in the case of every legal order to elaborate on this question.

139. For a view on these points which Walter developed in the light of Kelsen (and differed from his master), see Schild, Wolfgang, Die Reinen Rechtslehren. Gedanken zu Hans Kelsen und Robert Walter (1975) at 33 Google Scholar et seq.

140. Contemporary Pure Theory of Law often claims that it not only reconstructs and applies the doctrines of Kelsen, but also criticizes and develops them. See Walter, Robert, ‘Die Reine Rechtslehre—eine Theorie in steter Entwicklung. Einige Klarstellungen’ (1994) Juristische Blätter at 493 Google Scholar et seq.

141. Walter, supra note 71 at 16 et seq.

142. Raz, supra note 18 at 109 actually points out the same thing. According to him, the following Kelsenian theses contradict each other: 1. every norm is supported by a sanction, 2. some norms are only empowered for the setting of norms, but are not directly supported by sanctions. The reason for this contradiction is that Kelsen makes use of two different concepts of norm, but not explicitly. Raz shows the difference as one between static and dynamic points of view, see Raz, ibid. at 110 et seq.

143. Walter, supra note 71 at 19.

144. Walter, ibid. at 46 et seq.

145. Walter is interpreted in this way by Stoitzner, supra note 87 at 56.

146. Walter, supra note 71 at 51.

147. Walter, ibid. at 18.

148. In the same way Raz, supra note 18 at 115: the dynamic norm is counter-intuitive (it does not correspond to a lawyerly ‘common sense’) and over-complicated.

149. Walter, supra note 71 at 18.

150. Consequently, the basic norm is not part of the positive legal order, and therefore he does not deal with it here. Hence the fact that Walter does not explicitly discuss the basic norm in this work, is—in my opinion—exclusively due to the fact that he only describes the structure of positive law. On Walter’s view about the basic norm in detail, see Walter, Robert, ‘Die Grundnorm im System der Reinen Rechtslehre’ in Aarnio, Aulis et al., eds., Rechtsnorm und Rechtswirklichkeit. Festschrift für Werner Krawietz zum 60. Geburtstag (1993) at 85 Google Scholar et seq.

151. Walter, supra note 71at 26.

152. See above: 1.2.6 Derivation of Validity (Existence) of a Norm in the Case of Simple Legislation.

153. Kelsen, Introduction, supra note 8 at 52 et seq.

154. Walter, supra note 71 at 26 et seq. This is the reason why Raz criticizes Kelsen, cf. Raz, supra note 18 at 117.

155. Walter, supra note 71 at 28.

156. Walter, ibid. at 31. To what extent Walter is right in this particular question is not discussed here.

157. Walter, ibid. at 30 et seq. and 35 et seq.

158. Walter, ibid. at 48.

159. Walter, ibid. Cf. Jakab, supra note 70 at 60.

160. Walter, ibid. at 51 et seq.

161. Walter, ibid. at 39.

162. Moreover, Walter’s theory also includes a striking criticism on Hart: the distinction between primary rules and secondary rules (and their ‘sub-classes’) is actually not as easy to make as one might think at first sight after Hart.

163. To unveil the underlying ideology of a legal theory would refute it only if a legal theory were considered as pure ideology, i.e., if one would deny it any autonomy whatsoever. This approach is characteristic of Marxism. Without discussing this question here at length, it has to be stated that the above analysis of the underlying ideologies of the Stufenbaulehre is not intended to do that. Examining the ideological background may only be an interesting excursus in a legal theoretical analysis, but it belongs to another dimension. To show this background, in my opinion, does not mean supporting (or rejecting) a theory of law, only explaining it (from the perspective of science sociology).

164. Öhlinger, supra note 25 at 10. The autonomy of law in Kelsen’s system is emphasised in the same way by Raz, supra note 37 at 96.

165. Walter, supra note 88 at 19; Aladár Métall, Rudolf, Hans Kelsen. Leben und Werk (1969) at 22 Google Scholar. I do not mean by this that, beside the abovementioned factors, only law can play an integrative role. However, the lack of other factors has shaped the autonomous idea of law in the Pure Theory of Law.

166. Öhlinger, supra note 25 at 28.

167. Kelsen, Hans, ‘Wesen und Entwicklung der Staatsgerichtsbarkeit’ (1929) Veröffentlichungen der Vereinigung der Deutschen Staatsrechtslehrer at 30 Google Scholar et seq.; Kelsen, Introduction, supra note 8 at 71 et seq.; Verdross, Alfred, ‘Zum Problem der Rechtsunterworfenheit des Gesetzgebers’ (1916) Juristische Blätter at 471 Google Scholar et seq.

168. Behrend, supra note 6 at 18. On the legal nature of general internal orders of the administration see also Heinrich Rupp, Hans, Grundfragen der heutigen Verwaltungsrechtslehre (1965) at 11 Google Scholar et seq.; Achterberg, Norbert, ‘Kriterien des Gesetzesbegriffs unter dem Grundgesetz’ (1973) Die öffentliche Verwaltung at 298 Google Scholar; Möllers, Christoph, Staat als Argument (2000) at 154 Google Scholar et seq.; Jakab, supra note 70 at 83, esp. n. 172.

169. Öhlinger, supra note 25 at 31; Kelsen, Hans, Vom Wesen und Wert der Demokratie, 2. Aufl., (1929) at 69 Google Scholar et seq.

170. Merkl, supra note 22 at 339; Kelsen, supra note 169 at 70 et seq.

171. Öhlinger, supra note 25 at 29.

172. Öhlinger, ibid. at 32.

173. Öhlinger, ibid. at 30. Although some—e.g., Stoitzner, supra note 87 at 73 et seq.; Öhlinger, supra note 25 at 30—regard this as a special Stufenbaulehre, it is, in my opinion, something different, since it gives (similarly to the hierarchy according to derogatory power) no answer to the starting questions (1. origin of validity, 2. unity of legal order). It is, therefore, only an underlying ideology and the political consequence of the Stufenbaulehre.

174. Öhlinger, ibid. at 32.

175. Krawietz, supra note 32 at 255 et seq.

176. Öhlinger, supra note 25 at 9, n. 1; according to Krawietz, Werner, ‘Reinheit der Rechtslehre als Ideologie?’ in Krawietz, , Topitsch, & Koller, , eds., Ideologiekritik und Demokratietheorie bei Hans Kelsen Rechtstheorie (1982) 4 Beiheft, 413CrossRefGoogle Scholar, this is why its influence is so strong. It may certainly have contributed to the success of the Stufenbaulehre, but it was perhaps even more important that it could give an answer to a number of complicated questions of legal theory with the help of a surprisingly simple construction.

177. Krawietz, supra note 32 at 258.

178. Behrend, supra note 6 at 54; Dreier, supra note 12 at 43.

179. Trupp, Andreas, ‘Zur Kritik der Stufenbautheorie und wissenschaftstheoretischen Konzeption der Reinen Rechtslehre’ in Krawietz, Werner & Schelsky, Helmut, eds., Rechtssystem und gesellschaftliche Basis bei Hans Kelsen, Rechtstheorie (1984) 5 Beiheft, 300Google Scholar. Cf. Kelsen, GTN, supra note 36 at 27 et seq.

180. The only way remaining is the one that leads back to the state of 1911, see Kelsen, supra note 6 at 10: ‘… Here it is shown unequivocally that any question about the formation and destruction of the “is” is beyond the perception of the “is” (Seinsbetrachtung) and its particular (causal) epistemological method of explication, in the same way as any question about the formation and destruction of the “ought” is not at the level of observation focused on the “ought”, nor in the scope of a normative epistemological method’; or even 411: ‘This is the great mystery of law and state, that takes place in the act of legislation, and this might be the reason why its essence is represented only in images which are far from being satisfactory.’ I.e., one cannot explain law-making in the Pure Theory of Law anymore.

181. For linguistic help I am thankful to Miklós Könczöl. [This article is the author’s translation (with modifications) of the German language article: ‘Probleme der Stufenbaulehre. Das Scheitern des Ableitungsgedankens und die Aussichten der Reinen Rechtslehre’ which appeared in (2005) Archiv für Rechts- und Sozialphilosophie 333 (Franz Steiner Verlag) (Ed. note).]