• A Introduction

  In his book on unconstitutional constitutional amendments, Yaniv Roznai discusses possible limits of amendment powers, thus touching on various issues that have played (and somehow still play) a part in the Austrian constitutional debate. At first glance, he finds the idea of an unconstitutional constitutional amendment puzzling and paradoxical.¹x Y. Roznai, Unconstitutional Constitutional Amendments, the Limits of Amendment Powers, Oxford, Oxford University Press, 2017, p. 7. It will only add to this puzzlement that the Austrian constitutional doctrine traditionally rejects the idea of a limited constitutional amendment power, but is quite familiar with the notion of unconstitutional constitutional law. This short contribution aims to solve this puzzle (Sections B and C) and will further demonstrate that the traditional view that constitutional amendment powers are per se unlimited has been jeopardized by some scholars in more recent times (Section D).

• B Amendment Procedures

  The Austrian Constitution, originally dating from 1920, provides for various procedures to amend and/or enact laws on all kinds of levels, which constitute the ‘hierarchy of norms’. Two of these procedures are of interest here. Constitutional laws have to be passed with a two-thirds majority in the National Council, with half of its members present. Additionally, they have to be denominated ‘constitutional laws’ or ‘amending the constitution’. Only a ‘total revision’ of the constitution, however, additionally needs to be accepted by the electorate in what is called a mandatory referendum. This procedure was initially envisaged to be applied to all constitutional amendments, but as the framers of the constitution – representatives of the political parties – did neither believe in the viability of the Republic nor the future of democracy, they decided otherwise. The democratic constitution should be subject to a rather easy amendment procedure, which paved the way to very frequent amendments after World War II, as the two big political parties (conservatives and social democrats) cooperated and therefore were able to change the constitution whenever they found it to be necessary or desirable.²x M. Stelzer, The Constitution of the Republic of Austria, Oxford, Hart Publishing, 2011, p. 16. Only in the case of a ‘total revision’, an additional referendum would have been required, thus limiting the power of parliament to amend the constitution.
  This regulatory framework, laid down in Art. 44, para. 1 and (nowadays) para. 3 of the Federal Constitution, tasked Austrian scholars with the interpretation of the term ‘total revision’. Surprising as it may be, they gave this term a substantive reading. According to the Austrian doctrine and the jurisprudence of the Constitutional Court, a ‘total revision’ has to be construed as a severe alteration or amendment of one of the principles underlying and/or setting up the Austrian constitution.³x Stelzer, 2011, pp. 32ff.
  With this interpretation given, one may say that the Austrian Constitution contains (at least) two tiers of constitutional law: the easier amendable sphere of what might be called ‘ordinary’ constitutional law and the deeper entrenched sphere of constitutional principles, which can only be changed by means of an additional referendum. In the concept of the hierarchy of norms, constitutional principles would sit above all other constitutional provisions. As the whole concept of constitutional review is based on the idea that lower ranking law is scrutinized against higher ranking law,⁴x H. Kelsen, ‘Judicial Review of Legislation. A Comparative Study of the Austrian and the American Constitution’, Journal of Politics, Vol. 4, No. 2, May 1942, p. 184. the Constitutional Court, therefore, may scrutinize lower ranking constitutional law against constitutional principles. Should it find that lower ranking (‘ordinary’) constitutional law violates one or more constitutional principles, Austrian lawyers would be very familiar in addressing these constitutional laws as unconstitutional.
  However, with regard to a possible overlap of procedural and substantive questions, the court’s possible review of constitutional amendments is according to the Austrian understanding and, contrary to what Roznai⁵x Roznai, 2017, p. 211. suggests, by no means different from any other form of constitutional review. The Austrian doctrine would emphasize that any finding of the court that a law was unconstitutional would basically suggest that the wrong procedure had been chosen to enact it. Any simple law rescinded by the court might therefore have been re-enacted as a constitutional amendment. Only the principle of Rechtsstaat as construed by the Constitutional Court may nowadays set limits to such an undertaking (see Section 3).

• C The Principles of the Constitution

  It has to be emphasized that the number and the content of the principles of the constitution that rank above all other constitutional laws have been determined by the Constitutional Court and the constitutional doctrine. While the Constitutional Court has accepted three principles so far – the democratic principle, the federal principle and the principle of Rechtsstaat – constitutional lawyers have been more creative, inventing various other principles, some listing as many as thirteen different principles.⁶x H.P. Rill & H. Schäffer, ‘Artikel 44’, in B. Kneihs & G. Lienbacher (Eds.), Rill-Schäffer-Kommentar Bundesverfassungsrecht, Vol. 14., Vienna, Verlag Österreich, 2014, pp. 45ff; see also F. Ermacora, Österreichische Bundesverfassungsgesetze, 2nd ed., Vienna, Böhlau, 1994, pp. 12f; H. Mayer, G. Kucsko-Stadlmayer & K. Stöger, Bundesverfassungsrecht, 11th ed., Vienna, Manz, 2015, p. 77, 146. The disguised aim of inventing such a principle is rather clear: once accepted by the Constitutional Court, it would further restrain the power of parliament to amend the constitution.
  But also with regard to the content of those principles, which have been accepted by the Constitutional Court, many issues are still unclear and highly contentious. Nowadays, the jurisprudence of the court is settled in a way that allows at least for giving some hints on what may constitute a violation of one of the three principles cited above.
  The democratic principle protects as its core the principle of equality⁷x VfSlg 15.373/1998. and, above all, the institution of parliament, comprising periodical elections of its members and its core competence to legislate.⁸x VfSlg 16.241/2001. Therefore, it is widely held that Austria’s accession to the European Union that consequently transferred a whole set of legislative powers from the Austrian parliament to European bodies amended the democratic principle and therefore required a referendum.⁹x VfSlg 13.839/1994. Interestingly, should parliament pass a constitutional amendment providing for further legislative powers of the people (by introducing a Swiss-styled referendum that allows laws to be passed against the will of parliament or without its participation), this would also constitute a violation of the democratic principle and therefore require a referendum.¹⁰x VfSlg 16.241/2001.
  This is a very interesting twist in the Austrian constitutional doctrine that might also touch on the underlying concept of Roznai’s book. Many Austrian scholars would agree that Art 1 of the Federal Constitution, according to which all laws of the Republic emanate from the people, anchors the constitution in the (political) concept of popular sovereignty, a concept that also serves as the explanatory framework for Roznai’s book.¹¹x Roznai, 2017, pp. 105ff. Within that concept, it would have to be argued that the agent (parliament) cannot provide for further power for the principal (the people) without the explicit consent of the latter in a referendum. Although such an argument may be construed, it remains puzzling that it forms part of the core element of a democratic principle, especially when scholars define ‘democracy’ as a form of government that is based on or aims at the identity of rules and subjects.
  Following Kelsen’s theory on democracy, it may be questioned whether the Austrian constitution is based on the concept of popular sovereignty at all. To Kelsen, ‘the people’ was merely a fiction created by the legal order.¹²x H. Kelsen, ‘On the Essence and Value of Democracy’, in A. Jacobson & B. Schlink (Eds.), Weimar, A Jurisprudence of Crisis, Berkeley, University of California Press, 2000, p. 90. For a concept of democracy without reference to ‘popular sovereignty’ see further K. Popper, The Open Society and its Enemies, London, Routledge Classics, 2011, p. 118. Without reference to popular sovereignty, ‘the people’ is not a body that precedes the constitution, but, as ‘the electorate’, it is an organ established by the constitution. The introduction of further power for the people would therefore not be between a principal and its agent, but between two constitutional bodies and the allocation of their power, and it would entirely be up to the constitution to provide for the relevant procedures that have to be followed in case this allocation should be altered.
  Without deciding which theoretical concept underpins the Austrian constitution, Kelsen’s observations are closer to the historical circumstances under which the Austrian constitution was drafted and enacted. There was never a popular vote held on the enactment of the constitution, and it was basically the political parties who drafted it (with the help of experts) and compromised on its content. However, as the Austrian constitution may be read without reference to a concept of popular sovereignty, this may also question as to what extent Roznai’s theoretical framework may be applied to it.
  The federal principle protects the allocation of power between the federation and the (nine) states, as well as the participation of the states in the legislation of the federation and thus the basic institutional setup of the Federal Council.¹³x VfSlg 11.403/1987. This principle did not, of course, prevent constitutional amendments that shifted power from the states to the federation. As such amendments were passed frequently in the 1950s and 1960s, scholars emphasized again and again that at one point the federal principle could be eroded. But as powers were not shifted in large numbers at a time, this point was difficult to be assessed. Scholars therefore coined the term ‘creeping total revision’ of the constitution. It never had an effect on the jurisprudence of the Constitutional Court.
  The democratic principle and the federal principle were already present in the Austrian doctrine as early as the 1920s. Both – democracy and federalism – are explicitly addressed in Art 1 and 2 of the Federal Constitution, although without denominating them ‘principles’; both are the characteristic features that mark the differences to the Habsburg empire, which was neither a democratic republic nor a federal state. Otherwise, the constitution reflected the institutions and the concepts already developed under the Monarchy.
  After World War II, however, the Constitutional Court¹⁴x VfSlg 2455/1952. suggested that the Austrian Constitution was underpinned by a third principle: the principle of Rechtsstaat. Although not mentioned explicitly in the constitution, such a principle seemed desirable in the light of the atrocities committed by the Nazi regime. It was also introduced in textbooks,¹⁵x L. Adamovich & H. Spanner, Handbuch des österreichischen Verfassungsrechts, 5th ed., Vienna, Springer, 1957, pp. 114 ff. with scholars arguing that the Austrian constitution conspicuously implemented the ‘hierarchy of norms’, thus anchoring all state action in the law and, ultimately, in the constitution. A system of courts – ordinary courts, the Administrative Court and the Constitutional Court – offered sufficient protection to safeguard the compliance of lower ranking with higher ranking law. (From a more critical perspective, it may nevertheless be questioned if the protection offered by the courts system, basically stemming from the Monarchy, especially with regard to offering redress against all administrative action was and, despite some reforms, still is ‘sufficient’.) The power of the Constitutional Court to review laws was seen as the ultimate achievement of the Austrian Rechtsstaat. Consequently, this principle safeguards mainly the principle of legality as well as the power of the courts, with the reviewing power of the Constitutional Court sitting at its core.
  In the 1980s, the Constitutional Court started to adopt the proportionality principle when scrutinizing laws against fundamental rights. Consequently, it rescinded a couple of laws, many of which infringed on the setup of businesses by providing for a test that had to prove that an additional business of the type envisaged was needed by the market’s demand. This form of regulating markets was held to be disproportionate in many cases. Reacting to this jurisprudence, parliament passed constitutional amendments in some cases, re-introducing these tests rescinded by the court on a constitutional level. Notwithstanding whether this was correct from a moral or a political point of view, it merely reflected core elements of the Austrian constitutional doctrine: a law that violates a higher ranking law must simply be enacted on the higher rank, following the distinctive procedure.
  In answering this practice, the Constitutional Court suggested that should parliament pass constitutional laws in great numbers to bypass a proportionality test performed by the Court, this would erode the power of the Constitutional Court to scrutinize laws and therefore violate the principle of Rechtsstaat.¹⁶x VfSlg 16.327/2001, VfSlg 15.215/1998. But as parliament did not pass such laws in great numbers at a time, the court found itself in a similar position it had already adopted vis-à-vis the federal principle: as these laws were passed by and by, one could only speak of a ‘creeping total revision’ and it would have been completely arbitrary to assess which amendment was the one that constituted the tipping point.
  In the 1990s, as part of European integration, Austria was forced to finally legislate on public procurement – a realm of strong party influence and arguably of corruption. Initially, this was done on different levels of administration – therefore national parliament passed a law regulating public procurement on the federal level, and each of the nine state parliaments did likewise addressing the administration of each state. These laws were all quite similar and parliaments had to find out that they all contained the same two constitutional problems.
  In its case law, the Constitutional Court started to rescind the relevant provisions of these ten laws by and by. Which laws were scrutinized depended on the cases brought before the court. In this situation, it became quite clear to the national parliament that a general review of these laws was needed, possibly entailing some constitutional amendments. To stop the Constitutional Court from further eroding the state laws on public procurement, the national parliament passed a constitutional law declaring all state laws on public procurement (and not only the incriminated provisions) constitutional.
  Under these specific circumstances, the Constitutional Court could argue that in a whole field of law, the field of public procurement in the states, its power to scrutinize laws against the Constitution was completely abolished, thus violating the principle of Rechtsstaat.¹⁷x VfSlg 16.327/2001. To pass such a law, a referendum would have been required.
  However, it has to be emphasized that this court ruling was an exceptional case without any similar decision to follow up. On the other hand, it shows that the Constitutional Court has ‘reconstitutionalized’ the term ‘total revision’, should it ever have been meant as an avenue to introduce a completely different ‘constitution’.¹⁸x Roznai, 2017, p. 155. The reading of the Constitutional Court given, the term ‘total revision’ serves as a benchmark for scrutiny under the framework of the current constitution.

• D And Beyond?

  So far, the analysis has revealed that the Austrian constitutional doctrine is very familiar with the notion of ‘unconstitutional constitutional law’, but that is due to the concept of a two-tier constitution that distinguishes ‘ordinary constitutional law’ and constitutional principles that are entrenched deeper. Austrian scholars would express this phenomenon with reference to the concept of ‘hierarchy of norms’ and insist that constitutional principles sit above ordinary constitutional law. Therefore, ordinary constitutional law may be scrutinized against those principles and in case of violating them may be coined ‘unconstitutional’. But as constitutional principles may be amended by means of an additional (mandatory) referendum, again, it seems that the legislator simply has to choose the matching procedure to change whatever he intends to change. Regarding Roznai’s discussion of limits to amendment power, it might therefore be held that there are no (substantive) limits at all. ‘Total revision’ may therefore indeed open an avenue to constitutional change, no matter to what extent.
  This was most certainly the ubiquitous view of Austrian constitutional scholars, and nowadays it is still accepted in many quarters. However, things have been slightly changing over the last two decades.
  First, it has to be emphasized that the jurisprudence of the Constitutional Court sets effective limits for constitutional amendments. It is almost unthinkable that a government would call an electorate to the polling stations because it wants to enact a law declaring state laws on public procurement constitutional, arguing in the build-up to the referendum that it would otherwise infringe on the principle of Rechtsstaat. But as this is not a legal argument, Austrian public law scholars would see it as irrelevant. Under the efficient constitution, however, it has to be stressed that the jurisprudence of the Constitutional Court has the effect of limiting amendment power (although a more diligent drafting of the law might have passed the constitutional test).
  Second, some Austrian scholars have started to doubt that a democratic decision in a referendum may legitimately aim at abolishing democracy.¹⁹x P. Oberndorfer, ‘Artikel 1’, in K. Korinek & M. Holoubek (Eds.), Österreichisches Bundesverfassungsrecht, Vol. 3, Vienna, Springer, 2000, pp. 1, 9f; P. Pernthaler, Der Verfassungskern, Vienna, Manz, 1998. These scholars explicitly question as to whether the term ‘total revision’ may open the avenue to any change of the constitution, regardless of its future content.
  The ways to advocate in favour of this position are manifold and depend on the theoretical position one adopts. For those, for instance, who would argue that the electorate is a body set up by the constitution rather than ‘the people’, it might be convincing that the electorate has to operate within the (broader) framework of the constitution and therefore cannot legitimately vote for its own dissolution (although psychologically it might be possible).
  This argument may be underpinned by the observation that the principle of equality is part of the core element of the democratic principle, meaning that each member of the electorate has the equal right to vote only with regard to the same right of all its other members. A vote to deprive others of the power to vote may therefore be deemed to be beyond the power of voting.
  A similar observation may be made by those scholars who suggest that the Austrian constitution is built on the idea of popular sovereignty. If the people are not construed as an amorphous mass but a group of individuals (a collective), then each power to vote again exists only as an equal power to vote. The principle of equality is therefore an inherent part of the concept of popular sovereignty.²⁰x See J. Habermas, Between Fact and Norms, Cambridge, Massachusetts., MIT Press, 1996, p. 101. A constitution based on this concept can therefore only install a democratic government and cannot be misused to introduce an authoritarian regime. Additionally, with regard to Roznai’s understanding of ‘constitution’,²¹x Roznai, 2017, p. 1. an authoritarian constitution without limiting the power of government, accepting the rule of law and fundamental rights may be a ‘sham constitution’ as he puts it and/or no constitution at all.
  This all sounds highly philosophical and, indeed, it is. Roznai’s observation that “the issue of unamendability is a question both of fact and norm” and that the ability of a physical power to force any sort of change remains untouched, is correct.²²x Ibid., p. 132. His conclusion, concurring with Landau, that “the unconstitutional constitutional amendment doctrine may achieve less than one would hope”²³x Ibid., p. 230 citing D. Landau, ‘Should the Unconstitutional Constitutional Amendments Doctrine Be Part of the Canon?’, International Journal of Constitutional Law Blog, 10 June 2013, available at: www.iconnectblog.com/2013/06/should-the-unconstitutional-constitutional-amendments-doctrine-be-part-of-the-canon/. is probably realistic.
  However, the Austrian example may demonstrate that this doctrine is not entirely useless. In 1934, the 1920 democratic constitution was finally abolished and an authoritarian, arguably Austrofascist constitution was enacted. Interestingly, this was not done by means of a ‘total revision’ of the 1920 Constitution but by a cabinet decree. However, the government, shying away from a popular referendum, sought legitimacy by involving a 1917 law allowing for governmental measures in emerging situations caused by the war.²⁴x Stelzer, 2011, p. 11. Although this law was deliberately kept in 1920, the enactment of a totalitarian constitution did most certainly not fall under its scope. Although the authoritarian regime was introduced by means of a ‘coup d’état’, the government nevertheless tried to communicate its action as ‘lawful’. Assuming a force establishing an authoritarian regime would nevertheless seek legitimacy, the unconstitutional constitutional amendment doctrine may at least prevent democratic constitutions from being used as tools to provide legitimacy to an undemocratic regime, even if it contains a ‘total revision’ clause.

• E Conclusion

  The 1920 Austrian constitution provides for two tiers of constitutional law with regard to the amending process. The principles of the constitution, which can only be amended in the way of a ‘total revision’ and as such, requiring a referendum, sit above ‘ordinary constitutional law’ that can be amended by parliament although with a two-thirds majority only. Thus, the Constitutional Court may scrutinize constitutional amendment against the deeper entrenched constitutional principles, and, in case of a proven violation, the constitutional amendment can be deemed unconstitutional.
  For many decades, the Austrian constitutional doctrine held that any amendment infringing on the constitutional principles could be passed following the procedures of a ‘total revision’, thus by a mandatory referendum. Only for the last twenty years, scholars have stepped forward arguing that there were even limits to a ‘total revision’, primarily emphasizing that a democratic vote could never be misused to abolish democracy. Even if such a doctrine might not prevent physical powers from establishing an authoritarian regime, it ensures that a democratic constitution cannot supply legitimacy to such a political development even if the channel of amending the democratic constitution is used.

Noten

• 1 Y. Roznai, Unconstitutional Constitutional Amendments, the Limits of Amendment Powers, Oxford, Oxford University Press, 2017, p. 7.

• 2 M. Stelzer, The Constitution of the Republic of Austria, Oxford, Hart Publishing, 2011, p. 16.

• 3 Stelzer, 2011, pp. 32ff.

• 4 H. Kelsen, ‘Judicial Review of Legislation. A Comparative Study of the Austrian and the American Constitution’, Journal of Politics, Vol. 4, No. 2, May 1942, p. 184.

• 5 Roznai, 2017, p. 211.

• 6 H.P. Rill & H. Schäffer, ‘Artikel 44’, in B. Kneihs & G. Lienbacher (Eds.), Rill-Schäffer-Kommentar Bundesverfassungsrecht, Vol. 14., Vienna, Verlag Österreich, 2014, pp. 45ff; see also F. Ermacora, Österreichische Bundesverfassungsgesetze, 2nd ed., Vienna, Böhlau, 1994, pp. 12f; H. Mayer, G. Kucsko-Stadlmayer & K. Stöger, Bundesverfassungsrecht, 11th ed., Vienna, Manz, 2015, p. 77, 146.

• 7 VfSlg 15.373/1998.

• 8 VfSlg 16.241/2001.

• 9 VfSlg 13.839/1994.

• 10 VfSlg 16.241/2001.

• 11 Roznai, 2017, pp. 105ff.

• 12 H. Kelsen, ‘On the Essence and Value of Democracy’, in A. Jacobson & B. Schlink (Eds.), Weimar, A Jurisprudence of Crisis, Berkeley, University of California Press, 2000, p. 90. For a concept of democracy without reference to ‘popular sovereignty’ see further K. Popper, The Open Society and its Enemies, London, Routledge Classics, 2011, p. 118.

• 13 VfSlg 11.403/1987.

• 14 VfSlg 2455/1952.

• 15 L. Adamovich & H. Spanner, Handbuch des österreichischen Verfassungsrechts, 5th ed., Vienna, Springer, 1957, pp. 114 ff.

• 16 VfSlg 16.327/2001, VfSlg 15.215/1998.

• 17 VfSlg 16.327/2001.

• 18 Roznai, 2017, p. 155.

• 19 P. Oberndorfer, ‘Artikel 1’, in K. Korinek & M. Holoubek (Eds.), Österreichisches Bundesverfassungsrecht, Vol. 3, Vienna, Springer, 2000, pp. 1, 9f; P. Pernthaler, Der Verfassungskern, Vienna, Manz, 1998.

• 20 See J. Habermas, Between Fact and Norms, Cambridge, Massachusetts., MIT Press, 1996, p. 101.

• 21 Roznai, 2017, p. 1.

• 22 Ibid., p. 132.

• 23 Ibid., p. 230 citing D. Landau, ‘Should the Unconstitutional Constitutional Amendments Doctrine Be Part of the Canon?’, International Journal of Constitutional Law Blog, 10 June 2013, available at: www.iconnectblog.com/2013/06/should-the-unconstitutional-constitutional-amendments-doctrine-be-part-of-the-canon/.

• 24 Stelzer, 2011, p. 11.