READING THE COSTANZO OBLIGATION IN THE LIGHT OF THE PURE THEORY OF LAW

Ewa Fabian ^*

https://orcid.org/0000-0003-3061-396X

ANALIZA OBOWIĄZKU COSTANZO W ŚWIETLE CZYSTEJ TEORII PRAWA

1. INTRODUCTION

It can be argued that Hans Kelsen is most commonly known for his views on law and morality. The assumption may be, therefore, that discussing Kelsen or legal normativism inevitably leads to debates on ethics, which, though very important, at times provide little guidance in terms of practical application. The problem reflected in this article is related to the position of administrative authorities in EU member states. On the one hand, some may claim that it is not for the administrative body to decide how to apply general principles of EU law. Public sector bodies function on the basis of, and within the limits of, the law. Administrative authorities are at times discouraged from using sophisticated methods of interpretation. Judicial supervision may be viewed as a sufficient safeguard:

Although it would not be justified to expect that the tax authorities would disregard the binding provisions of national law, which are unambiguous in their content, and interpret them in a pro-EU manner, either using the formula of legislative modification or following the analogia legis legal reasoning, it is still up to the administrative courts controlling the activities of public administration in terms of compliance with the law, including Community (EU) law, to examine the possibility and justification of carrying out such interpretation. Its result is the discovery of the existence of a gap in the law (...) which must be filled in a way that enables the implementation of the norms of Community (EU) law (...). (Judgment of the Polish Supreme Administrative Court – wyr. NSA 2.02.2017, II FSK 506/16, Legalis 1578653, translation my own – Author)

Even though the judgment quoted above was issued, in essence, in accordance with the spirit of EU law and with careful consideration of relevant judgments of the Court of Justice of the EU, there remains an argument that the application of the principles of EU law by the administrative branch is “not to be expected.” On the other hand, the EU Court – previously known as the European Court of Justice, the ECJ, currently known as the Court of Justice of the EU, the CJEU – requires national administrative authorities to be proficient in applying the principles of EU law when dealing with EU law. Under the Costanzo Obligation, which is a rule formulated in EU case law, administrative authorities are required to apply the EU’s direct effect principle. Administrative authorities are expected, at least by EU courts and institutions, to apply certain EU law principles in the same way that domestic courts do. This means that they are deemed authorised to assess if their local laws are compliant with EU law, and, if they are not, whether to decide on the basis of other provisions or invoke general principles of EU law.

This matter of the application or non-application of EU laws may be of secondary importance in matters such as in the judgment quoted above. There, interested parties had access to an administrative court due to the subject matter of the case under domestic law. There are, however, matters where the application or non-application of EU laws by the administrative branch determines whether the matter may be reviewed by an administrative court, or not at all. In such matters, the application of EU law by an administrative official directly reflects on the access to court (the right to an effective remedy under the Charter of Fundamental Rights of the European Union). According to the spirit of EU law, as expressed in the judgments issued by the CJEU, if the domestic law directs matters involving EU law to a domestic procedure where access to courts is never granted, it is for the administrative body to establish that the domestic procedure is contrary to EU law. In order to make EU law fully effective, the administrative body should apply such domestic laws that will make it possible for the matter to be brought to a court at a later stage, if needed. If the official is not expected to apply EU laws, the individual applicant has no clear path to have their case reviewed by a court of law.

It can be argued that the application of the principles of EU law demands knowledge that can be obtained, generally speaking, either through clear instruction, which would pose practical and constitutional challenges, or through an analysis of a great amount of legal texts. For the legal tradition which is said to reject certain elements of Neo-Kantianism adapted by Kelsen in his legal theory, the application of EU law would require comparative interpretation in search of the substance of EU law in national regulations so as to make sense of open textuality of EU law and eliminate the carelessness of meaning (Zirk-Sadowski 2009, 59–60). In the light of that practical obstacle, opposition against the requirements of sophisticated legal reasoning becomes all the more understandable. The extent of the Costanzo Obligation is also debated among legal scholars (Verhoeven 2011). The relationship between the EU and EU member states is of a fundamental nature while also undergoing change and growth, both politically and in the field of law.

In this article, I oppose the view that the application of the principles of EU law by administrative authorities is not justified. I will present the relevant arguments with reference to some of the points contained in the works by Hans Kelsen. The focus is on three aspects of the problem, namely: 1) the matter of the choice between applying domestic law and EU law; 2) whether the law applied by the judicial branch is different from the law applied by the administrative branch; 3) the political nature of any such application of law. With any luck, the application of Kelsen’s theory in an analysis of judgments issued by the CJEU will shed new light on the role of administrative authorities in the EU.

2. THE COSTANZO OBLIGATION

In general terms, the legal order of the EU rests upon a number of principles, relying, when invoked by the CJEU, mainly on the primary law of the EU, i.e. the founding treaties and general principles of EU law (Syrpis 2015, 1). The CJEU is also known to tie its reasoning to the pan-European law of fundamental rights, now set in the framework of the Charter of Fundamental Rights of the European Union – the EU Charter – on the basis of Article 6(1) of the Treaty on European Union (Paunio 2013, 31). The EU is a union of law; the main body of legal principles is assumed to be shared among all EU member states. It is understood that this shared legal tradition predates the existence of the European Communities. The CJEU invokes general principles of law and expects them to be applied when making sense of legal provisions in all kinds of cases, not only when the so-called “hard cases” are brought before the high courts. As far as the principles of law are concerned, references are also made to the European Convention of Human Rights, in particular Article 52(3) of the EU Charter (Paunio 2013, 31). Whatever their origin, the principles of EU law are taken into consideration and further elucidated in judgments issued by the CJEU. When making such references, the CJEU often refers to its “case law” and, within that, to certain general principles of interpretation:

[i]n accordance with settled case-law, it is necessary, in interpreting a provision of Community law, to consider not only its wording, but also the context in which it occurs and the objectives pursued by the rules of which it is part (...) In addition, under a general principle of interpretation, a Community measure must be interpreted, as far as possible, in such a way as not to affect its validity and in conformity with primary law as a whole. (Judgment of the Court (First Chamber) of 16 September 2010, C-149/10, Zoi Chatzi v Ypourgos Oikonomikon, ECLI:EU:C:2010:534, 42–43)

The level of discretion enjoyed by the CJEU is usually determined by a number of factors, which include the lack of clarity and imprecision of interpreted provisions, the level of conflict between “conclusions suggested by the applicable interpretative topoi”, and indetermination in previous CJEU judgments (Beck 2012, 434).

The CJEU often deals with the issue of whether a national court of a EU member state should apply national law even if it is incompatible with EU law, or should it apply EU law and, if so, what conditions apply. Thus, the CJEU plays an important role in the shaping of the EU legal order at every level, not excluding the protection of individual rights in first-instance decisions; the reach of EU law should not be limited to national legislature and the high courts. The question remains whether the EU legal order can be described as hierarchical, heterarchical (Avbelj 2011), or a multi-central system with a quoad usum division (Łętowska 2005). One can observe that some legal issues are resolved only within the national legal system, while other matters, regulated by EU laws, are expected to be resolved with reference to EU laws, including the primary law of the EU, and by laws based on EU laws, as far as reasonable. EU member states retain their procedural autonomy, but this is also limited by a set of principles, namely the principle of effectiveness and the principle of equivalence, often referred to as the Rewe or Rewe-Comet effectiveness formula (see Judgment of the Court of 16 December 1976, C-33/76, Rewe-Zentralfinanz eG and Rewe-Zentral AG v Landwirtschaftskammer für das Saarland, ECLI:EU:C:1976:188; Judgment of the Court of 16 December 1976, C-45/76, Comet BV v Produktschap voor Siergewassen, ECLI:EU:C:1976:191; Widdershoven 2019). In an attempt to make laws of EU member states and their application as compatible with EU laws as possible, this expectation is not understood only as an obligation of national legislatures to enact new laws. Pointing to the duty of sincere cooperation, the CJEU goes beyond national legislation in a variety of ways, relying on the interpretation and application of law in each and every individual case decided by common courts, administrative courts, and – in accordance with the Costanzo Obligation – by national administrative bodies in every EU member state. The issue in question here is whether differences between the judicial power and the administrative branch have enough bearing to justify the disregarding of EU obligations with respect to the latter. For the sake of argument, it is assumed here that hypothetical national laws are irreconcilable with hypothetical EU laws to the point where an administrative authority deciding a case has to choose on its own initiative whether to apply EU laws or national laws. In certain cases, by applying EU law principles and rejecting an express provision of national law, an administrative body would choose a procedure which allows redress to court, thus facilitating access to court in a matter concerning EU laws. The analysis described in this article is, therefore, not limited to the Costanzo Obligation, but also takes into account the principle of judicial protection. When an administrative body decides on matters pertaining to EU law, their decision involves also the form that the act should take. By selecting a form which cannot be brought before a court of law nor an administrative court, they act as gatekeepers barring access to justice.

The main principle of EU law is the principle of primacy, set out in a number of groundbreaking CJEU judgments: Costa v ENEL (Judgment of the Court of 15 July 1964, 6–64, Flaminio Costa v E.N.E.L., ECLI:EU:C:1964:66), Internationale Handelsgesellschaft (Judgment of the Court of 17 December 1970, 11–70, Internationale Handelsgesellschaft mbH v Einfuhr- und Vorratsstelle für Getreide und Futtermittel, ECLI:EU:C:1970:114), Simmenthal (Judgment of the Court of 9 March 1978, 106/77, Amministrazione delle Finanze dello Stato v Simmenthal SpA, ECLI:EU:C:1978:49) and IN.CO.GE (Judgment of the Court of 22 October 1998, Joined cases C-10/97 to C-22/97, Ministero delle Finanze v IN.CO.GE.’90 Srl, Idelgard Srl, Iris’90 Srl, Camed Srl, Pomezia Progetti Appalti Srl (PPA), Edilcam Srl, A. Cecchini & C. Srl, EMO Srl, Emoda Srl, Sappesi Srl, Ing. Luigi Martini Srl, Giacomo Srl and Mafar Srl, ECLI:EU:C:1998:498). It establishes a conflict rule which “imposes an obligation on all national authorities to «set aside conflicting national measures» and leave them inapplicable” (Claes 2015, 182). This rule does not affect the validity of national laws conflicting with EU law.

One can say, in such a situation, that there are two or more valid legal systems contemplated by a national authority in parallel. The application of EU law is often portrayed this way by the judiciary and even in academic works. After all, the entire theory of a multi-centric legal system is based upon this argument as it seeks to point out “which law” is to be applied in a given legal matter (Łętowska 2005). This divided view (i.e. on the one hand, the national law, and on the other, the EU law, or EU and international law) is, for practical reasons, a familiar starting point for the purposes of discussion. However, the problems one faces when required to choose one law over another shows why the monistic theory of international law provides a promising change of perspective. If an entire legal system is to be chosen over another, one can make a false assumption that there is an option to disregard an entire “body of law” as “alien” to national law, without any further distinction between principles and specific provisions. This view of the body of law is then simplified – “our law” as opposed to “international” or “community” law. This divided view also feeds arguments against the democratic legitimacy of EU law. Since there is a division between “our law” and community law coming from “the outside”, democratic legitimacy of the latter is also debated. Scholars introduced the term of “functional legitimacy”, which “arises from the (hoped-for) realisation of certain values” such as the common welfare (Bindreiter 2000, 9). An increase of democratic legitimacy in the EU could justify the process of supra-national decision-making (Bindreiter 2000, 297), thus solving the initial riddle of legal cohesion. However, it is not clear whether it could, in fact, change the way in which EU law is thought and written about in practice. It appears that the EU court attempted to make EU law “our law” of EU member states early on through a series of rules of interpretation and application of law, as can be seen in early judgments of the ECJ.

As a sui generis system, EU law is based upon the direct effect principle. The direct effect principle was established in 1963 in Van Gend & Loos (Judgment of the Court of 5 February 1963, 26–62, NV Algemene Transport- en Expeditie Onderneming van Gend & Loos v Netherlands Inland Revenue Administration, ECLI:EU:C:1963:1). In Van Gend & Loos, Dutch revenue authorities applied expressly worded national rules on tariffs, while Van Gend & Loos relied upon the EEC Treaty. The Tariefcommissie, listed as a judicial branch, established that the matter at hand raised a question concerning the interpretation of the EEC Treaty. It then referred the matter to the CJEU (then the ECJ). The Court ruled that “according to the spirit, the general scheme and the wording of the Treaty, Article 12 must be interpreted as producing direct effects and creating individual rights which national courts must protect” (Van Gend & Loos, 13). Thus, the CJEU required that national authorities disregard national laws which were incompatible with EU law. The principle was further explained and developed in subsequent judgments (Judgment of the Court of 21 June 1974, 2–74, Jean Reyners v Belgian State, ECLI:EU:C:1974:68; Judgment of the Court of 8 April 1976, 43–75, Gabrielle Defrenne v Société anonyme belge de navigation aérienne Sabena, ECLI:EU:C:1976:56 and in other cases). In the Reyners case, the CJEU reaffirmed, in the context of EU law, the rules commonly used by national authorities when applying international law, namely – the requirement that a provision must be sufficiently clear and precise to have direct effect (Reyners 2–74, passim). In this context, when discussing the direct effect from the academic perspective, Verhoeven invokes the term of “justiciability”, which is a situation when “a norm which has direct effect is suitable for application by a court” (Verhoeven 2011, 21).

Then there is the important point of the language used by the CJEU. The notion of a kind of a EU common court was observed first between 1988 and 1990 (Wróbel 2010, 474). The CJEU ruled that “(...) a duty of sincere cooperation with the judicial authorities of the Member States, which are responsible for ensuring that Community law is applied and respected in the national legal system” (Order of the Court of 6 December 1990, 2/88 Imm, J.J. Zwartveld and Others, ECLI:EU:C:1990:440, 10). The notion was followed by phrases such as “ordinary courts of Community law” (Order of the President of the Court of First Instance of 22 December 1995, T-219/95 R, Marie-Thérèse Danielsson, Pierre Largenteau and Edwin Haoa v Commission of the European Communities, ECLI:EU:T:1995:219, 77) and “Community courts of general jurisdiction” (Judgment of the Court of First Instance of 10 July 1990, T-51/89, Tetra Pak Rausing SA v Commission of the European Communities, ECLI:EU:T:1990:41, 42). Since then, national courts have been regarded as the courts of that state and, in EU-related matters, as EU courts. The use of such phrases appears to make a difference, at least in the sense of the legal culture currently taking shape.

Around the same time, the CJEU interpreted the established principle of direct effect and the principle of primacy as requiring “national administrative authorities to set aside provisions of national law which are incompatible with EU law. When necessary, this may also imply the obligation to apply provisions of European law instead of the unapplied provisions of national law, if the disapplication leads to the emergence of a legal gap” (Verhoeven 2011, 10). This rule, known as the Costanzo Obligation, was expressed in the Fratelli Costanzo case (Costanzo 103/88), referred to earlier in the article. The rule was expressly confirmed in the Ciola case: “all administrative bodies, including decentralised authorities, are subject to that obligation as to primacy, and individuals may therefore rely on such a provision of Community law against them” (Judgment of the Court (Second Chamber) of 29 April 1999, C-224/97, Erich Ciola v Land Vorarlberg, ECLI:EU:C:1999:212, 30).

It was later confirmed that the Costanzo Obligation applies to tax authorities, decentralised authorities, constitutionally independent authorities, and even authorities providing public health services (Verhoeven 2011, 9). It also “applies regardless of the question whether the Court of Justice has already established the incompatibility between rules of national law and rules of European law” (Verhoeven 2011, 286). This approach strongly supports EU integration, requiring all authorities of EU member states to read and interpret EU law in line with interpretations provided by the CJEU. Moreover, when analysing incompatibility with EU law, the CJEU invokes not only the context of interpretation as when law is applied by courts, but also legislation: “provisions of national law which conflict with such a provision of Community law may be legislative or administrative (see, to that effect, Case 158/80 Rewe ν Hauptzollamt Kiel [1981] ECR 1805, paragraph 43)” (Ciola C-224/97, 31). Thus, the CJEU circumvents any theoretical arguments related to the double role of administrative authorities in the legal system, bearing in mind that administrative authorities are at times expressly delegated to create laws, if only to a limited extent.

In one of the judgments, the CJEU put forward a definition of a body required to follow the direct effect principle. It is

a body, whatever its legal form, which has been made responsible, pursuant to a measure adopted by the State, for providing a public service under the control of the State and has for that purpose special powers beyond those which result from the normal rules applicable in relations between individuals is included in any event among the bodies against which the provisions of a directive capable of having direct effect may be relied upon. (Judgment of the Court of 12 July 1990, C-188/89, A. Foster and others v British Gas plc, ECLI:EU:C:1990:313, 20)

In this approach, the CJEU expressed a view that any authority which is not a private individual and has “special powers” is under the obligation to respect the rights of individuals regulated by EU law.

The expectation that national administrative authorities apply EU laws has several justifications. One of them, as mentioned in the opening chapter, is linked to the access to court (the right to an effective remedy assessed in the light of the Rewe principles of equivalence and effectiveness as well as the EU principle of effective judicial protection (Widdershowen 2019)). It is possible to imagine a case where a national entity questions its authority to issue administrative decisions, acting on the basis of the provisions of domestic law. It then decides on EU law matters informally, finally, and prohibiting any review, barring the applicant from seeking judicial protection. A domestic court of law, believing itself to be a EU court or not, may refuse to hear the case on formal grounds, pointing out that, since there were no grounds to issue an administrative decision, there is no matter for an administrative review and, thus, no matter for a judicial review either.

A similar problem was described in the Judgment of the Court (Fourth Chamber), 17 September 2014, C-562/12, Liivimaa Lihaveis MTÜ v Eesti-Läti programmi 2007–2013 Seirekomitee, ECLI:EU:C:2014:2229. In that judgment – pts 70, 71, and 75 – the CJEU explained and reminded that:

It is therefore not possible for an applicant whose application for aid has been rejected to contest that rejection decision. (...) In those circumstances, the lack of any remedy against such a rejection decision deprives the applicant of its right to an effective remedy, in breach of Article 47 of the Charter. (...) the requirement for judicial review of any decision of a national authority constitutes a general principle of EU law. Pursuant to that principle, it is for the national courts to rule on the lawfulness of a disputed national measure and to regard an action brought for that purpose as admissible even if the domestic rules of procedure do not provide for this in such a case. (see, to that effect, judgment in Oleificio Borelli v Commission, EU:C:1992:491, paragraphs 13 and 14)

In other words, “individuals are entitled to have access to the national courts if Union law confers rights on them (‘ubi Union jus, ibi national remedium’)” (Widdershowen 2019, 20).

3. A BREAKDOWN OF ARGUMENTS

The problems which arise when EU law is applied by the administrative branch in EU member states invite analysis in the light of some elements of Kelsenian Pure Theory of Law. In this article, I disagree with the view that an administrative body should not be allowed to apply principles of EU law and in so doing refuse to apply express provisions of domestic law. Such principles may, for instance, include the principles of effectiveness and equivalence setting the limits of procedural autonomy of EU member states – also known as the principle of effective judicial protection.

In the judgment quoted in the opening chapter, the administrative branch was said to be “not expected” to act in favour of EU law even if the meaning of the principles of EU law, as they are linked directly to the provisions of primary law of the EU, has already been elucidated by the CJEU and usually leaves little room for legal manoeuvre. That judiciary claim is based, I assume here, on certain preconceived ideas related to the understanding of law.

One of these preconceived premises is that an interpreter of law may choose between applying EU law and the domestic law in EU-law-related cases as if there were two “legal systems” operating in parallel to each other, concurrently binding but offering different norms when applied to the same legal matter. An administrative official is expected to apply the domestic law even knowing that, in doing so, they would act in breach of EU law. The official is to apply an express provision of the domestic law which is contrary to EU law. If there was an express EU law provision, the official would probably be allowed to apply it; the principles of law, however, are different in their nature. The official is not permitted to find an express provision of the domestic law inapplicable by reference to principles, since every act issued by public administration must invoke legal grounds which are deemed to be sufficiently clear. If an express provision of the domestic law is in breach of EU law, the domestic law must be amended in the procedure provided for in the domestic law. Only when the domestic law is amended and the official has a clear provision of law to invoke as legal grounds, the official is allowed to act according to the spirit of EU law, but not before then. The official acts always on behalf of a sovereign state and not on behalf of the EU. The official must obey the domestic law; in other words, they need to choose the domestic law over EU law, despite the fact that, in cases involving EU law, all state officials are supposed to ensure that EU laws are effective and that the citizens are granted access to the courts in all EU-related matters. The reason for these “two legal systems” being concurrently in force and binding does not exactly lie in any lack of hierarchy between them. The difference recognised here is between express provisions of legal rules (such as the domestic rules of procedure or an express, directly applicable provision of EU law) and the (general) principles of law which may be perceived as less clear or less conclusive.

This viewpoint – which I oppose – can be analysed in the light of Kelsen’s works despite the fact that his theory – namely the monistic constructions of international law – concerned international law. This is possible partly because Kelsen contemplated sovereignty in international law, which is a term still in frequent use, in politics and in legal studies. Arguments raised by Kelsen are very general and rather on the macro scale.

Secondly, the claim that an administrative body is not allowed to treat EU law as primary to the domestic law contains an idea that laws should be applied differently by the administrative branch and by the judiciary. There are areas, such as the interpretation of the principles of law, which are sophisticated, require deep understanding, and are, to a point, unpredictable. Relying on principles when refusing to apply provisions of the domestic law is similar to the interpretation of law performed by high courts – in particular, constitutional tribunals. It is generally assumed that only high courts make such judgements, or – at the very least – only the judiciary but not the administrative branch. This second side of the argument may result in a presumption that there are, in fact, “two laws” not only on the plane where EU law and the domestic law appear to compete with each other but also in relation to the branch applying it, distinguishing between the law applied by the administrative branch and the “more sophisticated” law applied by the judiciary. This distinction is often made without reference to any specific rules of procedure but solely on the basis that, if a matter involves the application of a legal principle instead of a specific provision of the domestic law, it is not for an administrative body to see that distinction and make the decision. At the same time, it is assumed that, in the same situation, a court of law would be allowed or even expected to make such a choice.

The third, additional facet of the argument is the question whether the application of EU law in accordance with the principles of EU law, including its primacy and effectiveness, is a political act. This third point is brought on directly by the views of Hans Kelsen. It is obvious, though, that any choice expected to be made between express provisions of the domestic law and the more general provisions of EU law or international law contains a strong political charge. One can try to avoid thinking about it, but it remains, obviously, at the bottom of the argument. Otherwise, discussions about sovereignty in that context would not really resonate.

4. THE FIRST ARGUMENT

When discussing the notion of sovereignty vis-à-vis international law, Hans Kelsen notes that, in reality, international law is addressed to individuals; the state is not a superhuman organism (Kelsen 1952 (1959), 100).

If international law imposes obligations and confers rights on the state to behave in a certain way, this means that it imposes obligations and confers rights on human beings, in their capacity as organ of the state, to behave in this way (Kelsen 1962 (1998), 526).

The main point of Kelsen’s views on sovereignty is, however, that when the notion of sovereignty is used to oppose international law, it only betrays a political agenda. According to Kelsen, sovereignty is not a valid argument against the effectiveness of international law. Since both international law and domestic laws clearly exist, one needs to understand how they relate to each other. When one argues that a state is not a superhuman being that may be subject to rights and obligations, it is unsustainable – for Kelsen – to claim that there are two legal systems – i.e. international law and the domestic law – in effect and in force at the same time

If one recognises that the imposition of obligations and the conferral of rights on the state by international law simply means that international law delegates powers to the state legal systems to specify the human beings whose behaviour makes up the content of these obligations and rights, then the dualistic construction of the relation between international law and state law collapses. (Kelsen 1962 (1998), 527)

Kelsen does not distinguish state from its legal system and from that perspective there is no real subject for discussion about the relationship between the state, its domestic law, and international law. Theoretical analysis may only concern the relationship between the domestic law and international law. In the monistic concept of international law, there are two ways of looking at the way they relate to each other, either from the perspective of the domestic law as enabling international law, or from the perspective of international law as enabling the domestic law. In both propositions, one law is contained within the other. Kelsen explains that it is impossible to determine which view is correct, as, from the theoretical perspective, both of them are (Kelsen 1962 (1998)).

Since there is an “epistemic unity” between international law and the domestic law, it could be ventured at this point that the domestic law should not be chosen in favour of international law as a whole, in any discriminatory matter, with the invocation of the notion of sovereignty. Yet, the problem discussed in this article rests with the application of law and the hierarchy of norms inferred from their level of generality (express provisions of domestic laws confronted with the general principles of EU law). Kelsen discussed the practical possibility of a conflict between “an established norm of international law” and the domestic law, assuming that “the state organs are bound to apply national law, even if it is contrary to international law” (Kelsen 1952 (1959), 419–420). A system where the judiciary is empowered to disregard national law in such situations is possible (Kelsen 1952 (1959), 419; Kelsen 1942, 188). Yet, in the view supported by Kelsen, a lower norm may only be invalidated in a procedure leading to its invalidation (Kelsen 1952 (1959), 532). A solution to the problem of conflict between international law and the domestic law in the application of norms “cannot be deduced from the relation which is assumed to exist between international law and national law” (Kelsen 1952 (1959), 420). For Kelsen, the solution is the same as when a conflict arises between higher and lower norms within the domestic legal systems – in particular, with the constitution (Kelsen 1952 (1959), 421). A legal norm deemed contrary to a higher principle of law (be it international or domestic) has to go through the formal procedure of invalidation. The principles of international law enjoy no special treatment in this respect. Since both systems – international law and the domestic law – enable each other, their hierarchy cannot be established at the theoretical level.

At this point, it would appear that EU law is not compatible with Hans Kelsen’s basic theory of international law. The order brought by EU law at the international law level – i.e. as far as the treaties are concerned – pierced the epistemic unity where international law and the domestic law described by Hans Kelsen enjoy similar treatment. Instead, the EU established a different system, agreed upon by EU member states. In EU-law-related matters, EU law is privileged; state organs are in some cases expected to disregard the provisions of the domestic law to make sure that the principles of EU law are effective. If state organs are usually not entitled to refuse the application of domestic norms on the grounds that they deem them unconstitutional, their powers in EU-law-related matters go further.

In EU member states, state organs need to treat EU law as if it were for them to decide whether their domestic law remains applicable in the light of EU law or not. Such is the legal system shaped by the CJEU in its judgments. Given this expectation, it is clear that EU law goes beyond the basic model described by Hans Kelsen in his theory of international law. However, Kelsen also stated that

[p]ositive international law (...) sets no bounds on limiting state sovereignty, that is, the freedom of action or the authority of the state. An international treaty can create an international organisation so centralised that it has itself the character of a state, with the result that the states entering into the treaty and incorporated into the organisation lose their character as states. (Kelsen 1952 (1959), 533–534)

He also advocated for international tribunals and deemed them more necessary in the process of creating law than an international legislator “there can be no legislator without a judge, even though there can very well be a judge without a legislator” (Kelsen 1943, 401). For these reasons, I assume here that, since international agreements come in all shapes and forms, there is a scale between a fully sovereign statehood and “states losing their character as states.” On that scale, one can imagine an organisation where certain, but not all, legal matters are centralised. Since, for Kelsen, there is no real difference between the state and its laws (Widłak 2018, 60), if specified areas of regulation are transferred to the organisation, the state cedes, by agreement, a specified part of its sovereignty. From then on, it is no longer for the state to decide upon the shape of the legal system concerning the matters which had been ceded. The legal system, including not only the wording of specific provisions, but also the binding rules related to the interpretation and application of law, is transferred from the state to the organisation. Since accession to such an organisation was freely agreed upon, the state organs can indeed be deemed to apply laws pertaining to that organisation within a different legal environment than their domestic one. They were allowed to do so from the moment of accession to that organisation by their state. I would conclude that, under such circumstances, an analysis of the relationship between the state law and an external legal system, such as EU law, cannot be made by reference to any general theory of international law. Instead, it needs to be limited to international treaties that created the organisation in question. All acts issued by EU member state organs with reference to EU law need to be analysed in the light of the treaties and their interpretation, not in the light of the general theory of international law.

It is possible for a state to agree upon obligations leading to the creation of a federation or a similar organisation, placed, like the EU, somewhere halfway towards a federation. Such a state takes on the obligation to loyally cooperate under the treaty. The obligations of state organs should therefore be analysed in the light of the particularities of EU treaties and other acts comprising primary EU laws (in particular the EU Charter). The obligations of state organs to treat EU laws differently are rooted in the treaties, as they were ratified in EU member states. MacCormick refers to it as “a self-referential and independently valid legal order” based on the “pacta sunt servanda” norm (MacCormick 1997, 336, 337). The privileged manner in which EU laws are to be applied was, therefore, agreed upon by each EU member state and should be obeyed not because there is any intrinsic, theoretical supremacy of principles of international or EU law, but because such a shape of the EU legal system was agreed upon by each member state at the fully sovereign, international level. Sovereignty existing at the level of international law enabled EU member states to freely agree to join the EU. After that, each time EU laws are expected to take precedence over domestic laws without recourse to the procedure of invalidation, a reference is made to the principle of sincere cooperation. Since EU member states agreed to such a system of enforcement at the international level, state organs which follow the rules and principles of EU law are not only loyally fulfilling promises made by their respective states but are, in fact, acting on the grounds of their domestic laws, in particular their constitutions and accession treaties.

What this means for the problem contemplated in this article is that it is not for the state courts to decide whether administrative bodies of that state are allowed to disregard domestic laws which are contrary to the principles of EU law. State courts of law are to ensure that EU laws are effective. Even when state courts act as EU courts, in their basic function when applying EU laws, they should avoid reshaping the EU legal system with the patterns they apply in purely domestic matters. As long as state administrative bodies are deemed permitted – and obligated – under EU laws and in judgments issued by the CJEU, to refuse to apply domestic laws contrary to EU law, they should be assumed to act within their rights when they do so.

5. THE SECOND ARGUMENT

The second preconceived idea analysed in this article is the difference – or lack thereof – between the law applied by the judicial branch and by the administrative branch. There may be situations where the same provisions of substantive law apply in cases decided upon by the administrative branch and by the judiciary. Most of all, the same general principles of law may apply when a case is decided by an administrative body and, when a case is adjudicated, by a court of law. No legal system can perfectly separate the two branches. The question here is whether an administrative body official enjoys the same discretion as a judge in the court of law when applying laws in matters related to EU law in the context of the Pure Theory of Law. One must agree that the administrative branch applies the same law but it does not hold the judicial power which is reserved for the judiciary. Public administration is generally linked to the executive branch and, within this sphere, can sometimes be granted powers to create laws. However, in Kelsen’s Pure Theory of Law, creating laws takes place every time the law is applied; in the dynamic process of concretising norms, the application of a higher-ranking norm with the hierarchy of norms translates into the creation of a lower norm (Bernstorff 2015, 37).

In Kelsen’s view, the application of law involves an act of will. Kelsen put strong emphasis on the element of volition: “[a]t the same time, it must be remembered, the activity of the judge is in no way exhausted by the act of recognition: this is only the forerunner of an act of will by which is to be set up the individual norm of the judicial sentence” (Kelsen 1934a, 480). There is a “degree of freedom granted to the judge by the dynamic idea of law-creation”, which “allows the court to adjust the law to the current needs of society” (Bernstorff 2015, 41). Discretionary choice is, however, enjoyed not only by judges, but also by other officials, thus involving administrative bodies (Chiassoni 1995, 41; Paulsson 2019, 209–211). Kelsen advocated for equal treatment of administration and the judiciary because of the judicial oversight of the administrative branch (Kelsen 1928a, 110; Techet 2024, 8). A legal norm functions as a scheme of interpretation (Kelsen 1967, 4); it does not dictate it, also as far as the method of interpretation is concerned. Kelsen refrained from prescribing any specific methods of interpretation (Chiassoni 1995, 47; Paulson 1990, 139; Bernstorff 2015, 39). As I understand it, this means that both branches – the judicial branch and the administrative branch – apply the law within the same theoretical framework. This theoretical framework and the application of law as an act of will describes all cases involving EU law – from the decision of an administrative official, through a judgment of a domestic court of law, to the judgments issued by the CJEU. It is important, however, as far as Kelsenian ideas are concerned, that one should assume judicial oversight of the administrative branch. This, again, shows how important it is to uphold a system where effective judicial protection is the primary rule which should take precedence over specific norms of the domestic law (without recourse to the procedures of invalidation).

6. THE THIRD ARGUMENT

The third argument which needs to be discussed in this article is whether the application of EU law in accordance with the principles of EU law, including its primacy and effectiveness, is a political act. For Hans Kelsen, since the application of law is an act of will, not of intellect, it should be presented as the creation of law (Paulsson 2019, 211). In the framework known as “the law qua politics”, legislators, judges and administrative officials are all lawmakers and political actors. They are empowered to act and, according to Kelsen, they are authorised to act “for political reasons” (Paulsson 2019, 189). Making decisions on political grounds is, for Kelsen, “entirely appropriate” (Paulsson 2019, 189). He claimed that “there is only a quantitative, not a qualitative difference between the political character of legislation and that of the judiciary” (Kelsen 1931, 586; transl. Techet 2024, 11), since “every legal dispute is a political dispute” (Kelsen 1931, 587; transl. Techet 2024, 11).

If put this way, there is no room for separation between law and politics. For that reason, I must conclude that, in the light of the Pure Theory of Law, any decisions made by administrative officials in EU member states are political in nature. Acts of will involved in issuing administrative acts create legal norms and, as such, cannot be perceived merely as the “execution” of law. In reality, every time an administrative official issues any act concerning EU laws (or any other laws, for that matter), the reasoning behind the application or non-application of the domestic legal provisions or principles of law (EU law or any other law) may involve other concerns, such as ethics, morality, economic reasons, or politics. Thus, an official whose tasks involve, for example, the allocation of EU funds or deciding on the petitioner’s future access to a court of law in a EU matter, takes up a position on the membership in the EU of their member state every time they make a decision. I would add that such officials are not exempt from political responsibility by virtue of them not taking part in the legislative process. Maintaining the shape of the EU’s legal system rests with all political actors participating in the application of EU laws, including the judiciary and the administrative branch.

* * *

To conclude, under the Costanzo Obligation, administrative officials are authorised to refuse to apply express provisions of the domestic law and apply other norms on the basis of the principles of EU law. When arguments against such obligations are voiced, they may be made on the premise that there are differences between the administrative branch and the judiciary. In the light of the Pure Theory of Law, however, there are no inherent differences between these branches as far as the application of law is concerned. What is more, the acts of the administrative branch can be perceived as political in nature. There remains the question whether an international organisation may impose upon its members an obligation to treat its laws in a different manner than their domestic laws. In a member state where lower-ranking norms are binding until the procedure of invalidation is carried out, a different approach would constitute a privileged position. I conclude, however, that such obligations are possible if agreed on at the international level. The obligations of the state officials to treat the “external” legal system differently would stem from the agreement entered into by that state based on the “pacta sunt servanda” norm. Acting in accordance with such obligations goes with the principle of sincere cooperation. General models concerning sovereignty and the relationship between international law and the domestic law no longer apply when the treaties of international law altered the relationship between the organisation and its member state by privileging external law in relation to its domestic law.