1.Introduction

Although the exercise of public power typically reaches the external world in the form of legal instruments, there are times when public authorities regulate human affairs by performing some physical conduct.The terms physical act and factual conduct are used here interchangeably. Many of these acts appear to have no significant legal relevance; for instance, driving a car, translating a document and guarding a building are activities that do not seem to trigger any specific consequences in law. In other instances, however, the profound detrimental effects of factual conduct for individuals and corpo-rations can hardly be overlooked. Restricting an individual’s liberty, passing over sensitive or false information to competitors, searching private premises and confiscating documents are all acts which can cause severe adverse effects for individuals and substantially impact their legal rights.

The idea that the European Union is a polity based on the rule of lawArticle 2 TEU. Also, Case C-294/83 Les Verts v. Parliament, EU:C:1986:166, para 23. means that all types of Union acts must adhere to EU norms, with physical acts performed by public officials being no exception to this rule. If this is correct, then a question arises as to what kind of mechanisms must be in place to assess the compliance of Union authorities with such a duty and provide effective legal remedies in individual cases.See, Article 47 EU Charter. This article aims to analyse what avenues for judicial protection are recognised under the law of the European Union for persons affected by physical acts of Union officials. To this effect, it begins by contextualising this inquiry within a broader threefold uncertainty that characterises EU law (section 2) and illustrating the significance of physical acts in contemporary EU administrative practice (section 3). Thereafter, it presents a novel conceptual framework for analysing acts and their effects in Union law (section 4). It uses this analytical framework to assess the justiciability of physical acts of Union authorities with reference to the case law of Union courts (section 5). Finally, it provides a systematic classification of the different functions of EU administrative factual conduct and identifies the respective avenues for judicial control, making also reference to the complexities arising in composite administrative procedures (sections 6-8).

This article has certain limitations. First, although it traces the conceptual underpinnings of certain EU administrative law notions to the national legal traditions, it does not aim to offer a comparative overview of domestic approaches to the legal effects and justiciability of administrative factual conduct.Recent general comparative works include A.V. Bogdandy et al. (eds.), The Max Planck Handbooks in European Public Law – Volume 1 The Administrative State (OUP 2017); and C. Backes & M. Eliantonio (eds.), Cases, Materials and Text on Judicial Review of Administrative Action (Hart 2019). Moreover, it does not explore the influence of national legal doctrines in the development of EU law in this area, which is nevertheless an interesting and largely unexplored subject matter that will hopefully attract future research. Instead, this article presents a critique of the state of the art under the law and administrative practice of the European Union, as a unique case study of transnational administrative governance.

2.A threefold uncertainty in EU law

Academic discussion has long been trying to decipher the conditions of justiciability of Union acts.See e.g. A. Türk, Judicial Review in EU Law (Elgar 2009) 12-39; K. Lenaerts, I. Maselis & K. Gutman, EU Procedural Law (OUP 2014) 7.08-7.30, 9.05-9.09; 10.03-10.12; and A. Ward, Judicial Review and the Rights of Private Parties in EU Law (OUP 2007). Most analyses, however, focus almost exhaustively on the of various types of legal instruments, paying little or no attention to physical acts.Profound exceptions in the English literature include H. Hofmann, G. Rowe & A. Türk, Administrative Law and Policy of the European Union (OUP 2011) Chapter 20; and T. Rademacher, ‘Factual Administrative Conduct and Judicial Review in EU Law’ [2017/29(2)] European Review of Public Law 399-435. For a recent comprehensive work, see T. Rademacher, Realakte im Rechtsschutzsystem der Europäischen Union (Mohr Siebeck 2014). This is not surprising, given that the case law of Union courts offers limited jurisprudential fragments dealing expressly with the reviewability of factual conduct.See infra section 5. Since the subject matter of this inquiry has been somewhat neglected, it is appropriate to place it within its broader constitutional frame.

Understanding the conditions for judicial control over EU factual conduct has both theoretical value and practical significance. It mitigates part of the inherent uncertainty that surrounds the exercise of public power in the Union.T. Tridimas, ‘Indeterminacy and legal uncertainty in EU law’ in J. Mendes (ed.), EU Executive Discretion and the Limits of the Law (OUP 2019). In particular, EU law seems to give rise to a three-dimensional uncertainty. First, there is uncertainty pertaining to the attribution of action. An external observer commonly faces difficulties in identifying the decision-maker, as well as from where exactly its authority emanates.Ibid. On the one hand, this fusion of authority may occur at the horizontal level or within the Union’s inter-institutional environment,Ibid. and is fed by the agencification of the Union’s executive power. On the other hand, the difficulty in identifying the source of authority can appear vertically, between Union authorities and Member States.See e.g. Joined Cases C-105/15 P to C-109/15 P Mallis and Malli v. Commission and ECB, EU:C:2016:702; Case T-327/13 Mallis and Malli v. Commission and ECB, EU:T:2014:909. The growing use of mixed or composite administrative procedures, involving multiple institutional actors representing distinct functions and jurisdictional levels, adds another layer of obscurity.See infra section 7. Finally, there are the circumstances where it is unclear whether an act is imputable to an EU body or an international forum.See e.g. Case C-370/12 Pringle v. Government of Ireland, EU:C:2012:756. In the field of external affairs see, Case T-257/16 NM v. European Council, EU:T:2017:130. A characteristic example of the symbiosis of all abovementioned forms of uncertainty in locating and attributing the source of authority are the normative and institutional arrangements governing the Economic and Monetary Union, particularly the provision of financial assistance to Member States;N. Xanthoulis, ‘ESM, Union Institutions and Union Treaties: A Symbiotic Relationship’ [2017/1] Revue internationale des services financiers 21-33. although, the phenomenon is traceable in other policy areas.

Second, there is uncertainty about the nature and legal effects of EU norms. Guidelines,Case C-189/02 Dansk Rørindustri v. Commission, EU:C:2005:408, para 209. communications,Case C-526/14 Kotnik v. Državni zbor Republike Slovenije, EU:C:2016:570. press releases,Case C-62/14 Gauweiler and Others, EU:C:2015:400. Also, T. Tridimas & N. Xanthoulis, ‘A Legal Analysis of the OMT Case: Between monetary policy and constitutional conflict’ [2016/23(1)] Maastricht Journal of European Law 17-39. recommendations,Case C-16/16 P Belgium v. Commission, EU:C:2018:7; Case C-322/88 Grimaldi v. Fonds des maladies professionnelle, EU:C:1989:646. and policy frameworks,Case T-496/11 United Kingdom v. ECB, EU:T:2015:133. and Memoranda of UnderstandingCase C-8/15 P Ledra Advertising v. Commission and ECB, EU:C:2016:701. See also, A. Karatzia & T. Konstadinides, ‘The Legal Nature and Character of Memoranda of Understanding as Instruments used by the European Central Bank’ [2019] European Law Review (forthcoming). are recent examples of instruments which were found by the Court to have effects that surprised not only the affected parties but also their authors. Such measures are not novel in the Union’s regulatory toolbox, but they growingly multiply and contaminate new policy areas, such as banking supervisionA. Türk & N. Xanthoulis, ‘Legal accountability of European Central Bank in bank supervision: A case study in conceptualising the legal effects of Union acts’ [2019] Maastricht Journal of European and Comparative Law. and border control.In certain instances, the normative force of Union officials’ conduct is profoundly obscure. For instance, FRONTEX officers are authorised to communicate their ‘views’ to Member States on the instructions given to European Border and Coast Guard teams and the Member State must ‘take those views into consideration and follow them to the extent possible’. See,

Article 21(2), EU Regulation 2016/1624 on the European Border and Coast Guard and amending EU Regulation 2016/399 and repealing EU Regulation 863/2007, Council Regulation (EC) No 2007/2004 and Council Decision 2005/267/EC, OJ L 251 of 16.9.2016. Rademacher (2017), supra n. 6 at 399. EU scholars have put forward various models to classify the effects of EU norms; many suggest distinguishing between ‘hard law’ or legally binding rules and other non-binding ‘soft law’ instruments – sometimes informal or atypical – that do not appear in the eyes of an external observer to entail binding legal effects.Among the most recent works, see F. Terpan, ‘Soft Law in the European Union ‒ The Changing Nature of EU Law’ [2015/21] European Law Journal 68-96; O. Stefan, Soft Law in Court: Competition Law, State Aid and the Court of Justice of the European Union (Kluwer 2012). Notwithstanding the theoretical value of these inquiries, so far, they have not presented a framework that would ease the practical uncertainty pertaining to the justiciability of the various forms of Union conduct. As it will be shown, developing such a conceptual framework requires providing clarity not only about the types of effects of Union acts but also the manner via which these are produced. Equally, as we shall explain below, one must not overlook the fact that legal effects can be produced not only by legal instruments but also the factual conduct of Union authorities.

The combined effects of the former two types of uncertainty – namely regarding the source of authority and legal effects of EU norms – has generated a third one, i.e. the practical uncertainty about the justiciability of EU acts. It is difficult to determine whether Union courts can assess the legality of a certain act if its author and consequences remain unclear. The Court has developed a substantive approach to assess the reviewability of EU acts in annulment actions under Article 263 TFEU,Case 60/81 IBM v. Commission, EU:C:1981:264. yet, the case law has been far from consistent until now.Türk & Xanthoulis, supra n. 21. At surface level, this may be for multiple reasons,See Opinion of AG Bobek in Case C-16/16 Belgium v. Commission, EU:C:2017:959, paras 67 et sec. including, inter alia, the Court’s non-uniform use of the terms legal effects and binding legal effects, the use of a plethora of atypical instruments, and the different functions attributed to them. As it will be shown, however, there appears to be a more fundamental underlying cause – the absence, in the case law and literature, of a comprehensive conceptual framework for the notion of legal effects in EU law, which is the decisive criterion for determining whether an act can be subject to judicial review (see infra section 4.1). As a result, persons and Union institutions face an inevitable difficulty in predicting whether a Union act would be subject to judicial control. More importantly, the lack of clarity pertaining to legal effects of a Union act poses an obstacle for any affected parties in deter-mining what rights and respective duties may arise as result of its occurrence. This ambiguity is of constitutional value; it reduces the quality of EU norms and judicial protection, hence posing tensions for the rule of law and human rights, two fundamental Union values.

3.Factual conduct in EU administrative practice

The Union’s everlasting integration has had transformative effects for the mandate, operation and interaction of national and EU administrative authorities. Agencies are increasingly assigned quasi-regulatory powers.E. Chiti, ‘European Agencies’ Rulemaking: Powers, Procedures and Assessment’ [2013/19(1)] European Law Journal 93-110. At an operational level, EU and national administrations perform their tasks in an institutional environment of complex, multi-level structures, processes and human resources. In parallel, the traditional means of cooperation, such as formal exchange of letters and meetings with set agendas, are widely replaced, or at least complemented, by new informal channels of communications.

Such an evolution of means of exercising public authority enhances the effectiveness of Union policies but also poses important challenges for judicial accountability.See e.g. M. Scholten & M. Luchtman (eds.), Law Enforcement by EU Authorities, Implications for Political and Judicial Accountability (Elgar 2017). Within a continuum of administrative events, it is sometimes hard to identify the individual administrative acts, as well as their consequences and justiciability. At least two factors appear to enlarge this problem. First, the Union’s executive power is increasingly conducted via informal means, both in terms of processes and output. For instance, telephone conversations, brief exchange of emails and simple text messages cannot easily be translated into legal categories, such as instructions, orders or opinions.For instance, in the European Banking Union’s administrative network, officials at the Single Supervisory Mechanism constantly communicate with national supervisors, National Central Banks and the management of credit institutions in performing their tasks pertaining to safeguarding the stability of financial systems across the Eurozone.

Second, the centralisation of administrative power and expansion of the Union’s policy areas multiply the types of circumstances where Union officials are required to perform their tasks by some factual conduct rather than merely adopting legal instruments. In the field of competition law enforcement and banking supervision, Union and national officials cooperate in conducting physical investigations into the premises of corporations. In that context, they seize documents, confiscate computer systems and interrogate witnesses. Such activities do not involve the mere exchange of statements, but require that Union officials perform physical acts, some of coercive nature, which may have important adverse impact on the rights of affected persons. Similarly, in the field of police cooperation and FRONTEX, Union agents sometimes encounter situations which can only be addressed by carrying out coercive acts, including the detention and transfer of individuals from one location to another and conducting interrogations.

At face value, distinguishing between legal acts and physical acts might seem straightforward. A Union official adopting a formal decision to deny a person’s request to enter the territory of the Union is, in all appearances, a legal act. By contrast, the actual communication of that decision by post to the said person involves primarily some physical movement, so it may justify classifying it among physical acts. A closer look, however, reveals that the legal/physical dichotomy is inherently blurred. This becomes more apparent in instances where it is difficult to identify an immediate prior legal act prescribing the public authority’s specific physical conduct. Consider, for example, the instance where an official takes a traveller’s passport away by force. In such a circumstance, differentiating between an authority’s decision to do something (normative act), the subsequent physical performance of that decision by the same or another authority and the legal and factual consequences that are produced is far from a straightforward exercise.

4.Conceptual orientations

4.1.The puzzle

EU law offers two basic remedial pathways for persons seeking judicial protection from unlawful acts adopted by Union institutions.The rights of privileged applicants and the ad hoc or exceptional judicial protection avenues provided for in EU primary and secondary law fall outside the scope of this article. First, they can seek to review the legality of a Union act using two distinct avenues; namely, directly, via an action for annulment (Articles 263 and 265 TFEU) or indirectly, via the preliminary ruling procedure (Article 267 TFEU) and inci-dental review (Article 277 TFEU). Second, a person who has suffered harm as a result of an unlawful Union act can seek compensation on the basis of the Union’s non-contractual liability by bringing a direct action for damages (Articles 268(2) and 340 TFEU).

In determining whether a Union act is reviewable in actions for annulment, the Court applies the test introduced in the IBM case.Case 60/81 IBM v. Commission, EU:C:1981:264. A reviewable act, stated the Court in IBM, is any ‘measure the legal effects of which are binding on, and capable of affecting the interests of the applicant by bringing about a change in his legal position…’.Ibid, para 9. This statement is problematic in many aspects,Türk & Xanthoulis, supra n. 21. but it is clear that the IBM test places strong emphasis on the substance of the act, rather than its form or label. The problem is that, in developing its substantive approach, the Court has not applied this test consistently and with clarity; as a result, there is much uncertainty as to when Union acts may fall under the scope of its reviewing powers. The source of this jurisprudential disparity appears to be the absence of a clear understanding of the notion legal effects.The notion of legaleffects was firstly recognised as the decisive criterion for determining the reviewability of Union acts in the ERTA case (Case 22/70 Commission v. Council, EU:C:1971:32, para 42).

To illustrate this point, we note that the Court’s definition of reviewable act in IBM left open three basic and interrelated questions. First, what does the Court mean when it refers to the notion of act? Second, what distinguishes acts capable of affecting the interestsof a person by bringing about a distinct change in that person’s legal position from acts producing other kinds of effects? In other words, what is the meaning of binding legal effect and how does it differ from other non-binding legaleffects or non-legal effects (the latter is sometimes referred to as factual effects or consequences of fact)?Case 60/81 IBM v. Commission, supra n. 31, para 19; Also, Case T-377/00 Philip Morris International v. Commission, EU:T:2003:6, paras 114-115; and Opinion of AG Mengozzi in Case C-476/14 Citroën Commerce, EU:C:2015:814, para 77. In his Opinion in the IBM case, AG Slynn distinguished legal effects from effects in fact (Opinion of AG Slynn in Case 60/81 IBM v. Commission, EU:C:1981:213, p. 2664). The use of the term effects in fact, in this context, should not be confused with references to the same term that have different meanings in the Union case law, such as: (a) instances where the change in the legal position of a person is not ‘expressly’ laid down in the legal provision (see e.g. Joined Cases C-267/91 and C-268/91 Keck and Mithouard, EU:C:1993:905; para 16; Case T-17/93 Matra Hachette v. Commission, EU:T:1994:89, para 47); and (b) the objective ascertainment of the effects of certain conduct (e.g. Case T-321/05

AstraZeneca v. Commission, EU:T:2010:266, para 309). Finally, if only acts with bindinglegal effects are reviewable under Article 263 TFEU, what justifies the Court’s readiness to review non-binding acts when the question of illegality arises in the context of indirectreview proceedings (Articles 267 and 277 TFEU)? In other words, if a direct review is available only for acts producing binding legal effects, what legal effects, if any, must an act produce to qualify for indirect review? We shall discuss each question in turn.

4.2.Acts and effects

The logic underlying the test in IBM seems to be rooted in the contemporary perceptions of the nature and function of law. This becomes apparent by analysing the notions of reviewableact and legal effects in their basic conceptual components.

Acts can be broadly regarded as a category of events that occur in the factual world.The world of facts can be said to cover the existence of all physical phenomena perceived by human senses and their relations in the physical world, as well as any changes made therein. They normally entail some voluntary change in the physical world caused by humans or some forbearance, the latter understood as ‘a consciously willed absence of physical movement’.A. Corbin, ‘Legal analysis and terminology’ [1919/2] Yale Law Journal 163-173, 163. The public law discourse typically analyses the exercise of public power in separate acts,See e.g. H. Kelsen, General Theory of Law & State (Transaction Publishers, 2006) 205. The notion of administrative act has its foundations in the continental legal tradition. For a comparative perspective see, M. Eliantonio & F. Grashof, ‘Types of Administrative Action and Corresponding Review’ in C. Backes & M. Eliantonio (eds.), Cases, Materials and Text on Judicial Review of Administrative Action (Hart 2019). Originally conceptualised in the French legal tradition as acte administratif, it was borrowed by German Jurists in the 19th century. Prof Otto Mayer, one of the founding fathers of German Administrative Law, offered one of the early authoritative definitions by understanding administrative act (Verwaltungsakt) as an authoritative pronouncement of the administration which in an individual case determined the rights of a subject

(O. Mayer, Deutsches Verwaltungsrecht (Duncker und Humblot 1895/96)). For a historical approach to German public law see M. Stolleis, Geschichte des öffentlichen rechts in Deutschland, Band II: 1800 to 1914 (C.H. Beck, 1992). Mayer’s definition, as subsequently developed by W. Jellinek in the 20th c. (W. Jellinek, Verwaltungsrecht (3rd edition, Springer 1931)), was enriched by the German case law and the currently applied definition is provided in Section 35 of the Administrative Procedure Act 1976 (Verwaltungsverfahrensgesetz, VwVfG) as amended, which states the following: ‘administrative act is every order, decision or other measure taken by an authority for the regulation of a particular case in the sphere of public law and directed at immediate external legal consequences’. In the German tradition, the existence of an administrative act is organically linked with any systematic approach to judicial protection of individuals in administrative law. By governing the rights and duties of citizens in their relationship with the state, the notion of administrative act informs the development of a rights-based perspective in the German legal doctrine. See, M. Eifert, ‘Conceptualizing Administrative Law – Legal Protection versus Regulatory Approach’ in H. Pünder & C. Waldhoff (eds.), Debates in German Public Law (Hart 2014) 203-218, 205. Although the German school has been influenced by the French administrative doctrine, the latter currently differs substantially from the former as French administrative law is not codified and widely depends for its development on the rulings of the Conseil d’État (F. Becker, ‘The Development of German Administrative Law’ [2017/24] George Mason Law Review 453-476, 464). For a classic definition of acte administrative see P. Delvolvé, L'acte Administratif (Sirey 1983) 11. For an overview of the scope of influence of the French doctrine of acte administrative in other European legal orders see M. Fromont, ‘A typology of Administrative Law in Europe’ in A.V. Bogdandy et al. (eds.), The Max Planck Handbooks in European Public Law – Volume 1 The Administrative State (OUP 2017) 579-600, 589-590. By contrast, the common law tradition has not developed any systematic analytical framework for administrative acts, similar to the one found in the continental scholarship. Although the foundations for judicial review date from 17th century, the contemporary power of courts to review executive action only appeared in 1960s, which is when the United Kingdom started developing a system of administrative law. See, P. Craig, UK, EU and Global Administrative Law, Foundations and Challenges (CUP 2015) 25-26. understood as sub-categories of facts.

There are two basic types of effects that can be produced or caused by an act,The conceptual borders of acts and their consequences are profoundly obscure and pose important methodological problems for any attempt to analyse human conduct. See inter alia, G.H. von Wright, Explanation and Understanding (Routledge & Kegan Paul 1971). namely legal effects and factual effects. An act is said to bring about legal effects when it creates some change in a legal relation. A legal relation comprises of rights and counterpart duties between two natural or legal persons.It is possible for A to have separate (bilateral) legal relations with multiple persons (A - B),

(A - C) etc. but not a single legal relation with two or more other persons (A - B, C …). A change in a legal relationIn IBM, supra n. 31, the Court refers instead to a change in the ‘legal position’ of a person. It is possible to consider the bundle of rights held by and duties owed to A as A’s legal position. Yet, given that these rights and duties do not exist independently but only in connection to their respective counterparts (which form e.g. B’s distinct legal position), it is more appropriate to use the notion of legal relation for analytical purposes. can be said to occur if any of its constituting rights and obligations is created, modified or extinguishedH.L.A. Hart, The Concept of Law (2nd edition, OUP 1994) 81. (e.g. where a person’s right is denied). Notably, not all human interests are considered as important so as to justify the law’s protection (by recognising them as rights) and constitute part of a person’s legal position (see section 4.3 below). Any interference with a person’s interests that do not enjoy the status of rights would not be a legal effect. Instead, such consequence can be said to comprise a consequence of fact,Also, sometimes referred to as practical, material, physical or factual effects. in the sense that it causes some change in a person’s factualposition. In general, an act is considered to generate some factual effect if it causes a change in the physical world,Hart, supra n. 42. typically in relations between physical beings, perceivable by human senses.

The subjects or authors of an act can be either private persons or public officials, the latter typically acting on behalf of a public authority to which the act is imputed. Acts of public authorities can be distinguished in terms of the shape or manifestation through which they come into being in legal acts and physical acts.

Legal actsThe literature sometimes refers to the terms juridical act or act-in-law; in French, they are widely known as actes juridiques; note, however that these terms do not always signify the same thing. See B. Seiller, Acte administratif, I – identification; II – Répertoire Dalloz de Contentieux administratif (Dalloz 2010) no. 25 et seq. can be defined here as acts dressed by some institutional cloth that reflect the exercise of public power and normally, but not always, entail one or more normative statements. They typically appear as written legal instruments, but they can also occur in an informal or non-written manner. A legal act can be said to exist or be valid, if it satisfies the rules of recognition of such acts, as applicable in a given legal order.

Physical acts (sometimes referred to as material or natural acts,J. Finnis, Natural Law & Natural Rights (2nd edition, OUP 2011) 200. In the French legal tradition, some scholars refer to them as ‘fait materiel’. See, Seiller, ibid. In German, ‘Realakt’ and ‘schlichtes’ or ‘informales Verwaltungshandeln’, but they do not always signify the same thing. See Rademacher (2014), supra n. 6. as well as factual conduct) are acts whose prime feature can be said to be the exercise of some physical power or movement by a person.A distinction can be made between physical power and legal power. One may have the legal power to do something but lack the physical power required to perform the same. Conversely, one may have the physical power to perform certain conduct but no legal power to do so. The notion of legal power is a mental concept which is not linked to the actual physical capacity to carry out certain conduct in the physical world. The existence of a physical act and its effects in the factual world is determined, to some extent, by human senses.Case C-414/10 Véleclair, EU:C:2012:183, para 33. Physical acts of public officials are normally carried out in the exercise of some duty or function provided in law.

Frequently, legal acts would produce legal effects. At the same time, the occurrence of a legal act would also entail, by nature, the performance of minimum physical acts.J. Austin, How to Do Things with Words (OUP 1962) 11-114, 106. Examples of such acts include the utterance of certain words or the writing and publication of a legal instrument. Besides legal effects, it is clear that legal acts can bring about important factual effects; for instance, a decision to detain a person has profound adverse practical or factual consequences for the life of that individual.

The substantial effect of physical acts often appears to be some change in the physical world (rather than the world of legal relations). Sometimes, the law decides to attach certain legal effects to the performance of physical acts. In these circumstances, physical acts can be said to also produce legal effects, in the sense that they additionally cause a change in the legal position of a person, whether by determining the latter’s rights or (less frequently) imposing duties.Consequently, the meaning and effects of physical acts are not revealed solely by empirical observation or their natural existence but also via interpreting and applying legal norms. For instance, the physical act of processing personal data can interfere with the data protection rights or the right to private life of a person. Similarly, the forceful entrance on private premises and the confiscation of business documents in the context of administrative investigations can restrict the right to property and economic freedoms of persons. Furthermore, it is possible to conceive circumstances where the physical act imputed to one Union authority may interfere with the right (power) of another Union authority, i.e. by the former carrying out an ultra vires act or preventing or obstructing the latter from exercising its conferred powers. It follows that the legal effects of physical acts are not confined to fundamental rights but extend to all rights recognised under primary and secondary EU law, including the powers assigned to Union bodies. Where this happens, the physical act sometimes appears to be contaminated by some normative component. More accurately, physical acts aimed at pro-ducing certain legal effects can be said to be two-fold; they entail both a norm and its implementation.

To summarise, the preceding analysis has shown that normative legal acts often have legal effects and some minimum factual effects. Physical acts necessarily bring about some factual effect and sometimes produce legal effects. If these premises are correct, then the main criterion underpinning the legal/physical dichotomy appears to be the measure or significance that one attributes to the physical component of a given act. In other words, if an act is found to create both legal and factual effects, it would be classified as legal or physical depending on the observer’s viewpoint, specifically whether the focus of the change is placed in the legal world or physical world respectively. Where the conduct of public authority consists of formal legal instruments and implementing physical acts, the legal/physical dichotomy may provide some assistance in analysing the sequence of events and locating the normative source of any legal effects produced. However, in other circumstances where the administrative conduct does not enjoy substantial institutional formalities (e.g. there is no written instrument or record of an oral decision), the legal/physical dichotomy can become blurred and thereby less useful for identifying the act that may have caused a change in a legal relation. If this is correct, then the sub-categories of legal and physical acts can be argued to describe the different expressions of public authorities’ conduct, as well as the means through which such authorities can bring about legal effects.

4.3.Rights and obligations

In order to regulate human relations the lawrecognises the existence of legal relations, comprised of rights and obligations between persons.H. Kelsen, The Pure Theory of Law (University of California Press 1970) 163. We talk about legal rights and legal obligations when these are provided for in a certain norm which is denominated as such in a formally valid legal source – in our case, a valid source of Union law. Neither case law nor the Union’s legislature has so far provided clear guidance on the basic features of rights or how to recognise and distinguish them from other norms, values and objectives in EU law.For a recent comprehensive work on EU individual rights see C. Warin, Individual Rights under European Union Law (Nomos 2019). Also, S. Beljin, ‘Rights in EU Law’ in S. Prechal & B. van Roermund, The Coherence of EU Law: The Search for Unity in Divergent Concepts (OUP 2008) 91-122, 93. For a conceptual approach, see J. Bengoetxea, ‘Rights (And Obligations) in EU Law’ in E. Jones, A. Menon & S. Weatherill (eds.), The Oxford Handbook of the European Union (OUP 2012) 734-746.

Under the lens of the so-called ‘interest’ theories,See, J. Raz, The Morality of Freedom (1986 OUP) Chapter III. Contemporary theorists include J. Griffin, ‘Discrepancies Between the Best Philosophical Account of Human Rights and the International Law of Human Rights’ [2001/101(1)] Proceedings of the Aristotelian Society 1-28; and M. Nussbaum, ‘Capabilities and Human Rights’ [1997/66] Fordham Law Review 273-300. Also, M. Nussbaum & A.K. Sen (eds.), The Quality of Life (OUP 1993). rights can only be explained by reference to the notion of obligation, because rights are sources of (or grounds for) the obligations that are correlative to them.J. Tasioulas, ‘On the Nature of Human Rights’ in G. Ernst & J.Ch. Heilinger (eds.), The Philosophy of Human Rights (De Gruyter 2012) 31; J. Tasioulas, ‘The Moral Reality of Human Rights’ in T. Pogge (ed.), Freedom from Poverty as a Human Rights (OUP 2007) 99. See, also, Opinion of AG Tesauro in Joined Cases C-46/93 and C-48/93 Brasserie du Pecheur v. Germany and R. v. Secretary of State for Transport, ex parte Factortame, EU:C:1995:407, para 39, where the following is stated: ‘the obligations of the Member States and of the Community institutions are directed above all, in the system which the Community system has sought and sets out to be, to the creation of rights of individuals’. A legal obligation can be broadly defined as ‘a categorical reason with a certain kind of exclusionary force’ recognised in law.Tasioulas, supra n. 54 at 99. By categorical, it is meant that the existence and weight of a duty is independent of the desires of the duty-bearer.John Tasioulas, ‘H.L.A. Hart on Justice and Morality’ in L.D. d’Almeida, J. Edwards & A. Dolcetti (eds.), Reading HLA Hart’s ‘The Concept of Law’ (Hart 2013) at 155-175, fn. 60. By exclusionary in its force, it is meant that a duty deprives ‘some otherwise countervailing considerations of their normative force’.Ibid. Consequently, a right exists if a person’s interest, taken by itself, has the requisite kind of importance to justify the imposition of duties on others to respect, protect and promote that interest.Tasioulas, supra n. 54 at 77, with reference to Raz, supra n. 53 at Chapter 7. In other words, the law normally recognises a right if the value of having it or a person’s need for it, ‘is of a kind sufficient to impose duties’ on others.J. Raz, ‘Human Rights in the Emerging World Order’, in: R. Cruft, S.M. Liao & M. Renzo, Philosophical Foundations of Human Rights (2015 OUP) 217-231, 221. That value justifies imposing a duty on others to secure or at least not interfere with the right-holder’s enjoyment of that right. To summarise, a right exists only when such corresponding duties exist;See also, Joined Cases C-6 and C-9/90 Francovich and Bonifaci v. Italy, EU:C:1991:428,

para 12. more accurately, ‘[i]t exists because it gives rise to such duties’.Raz, supra n. 59.

EU rights are not confined to rights widely described as human, basic or fundamental but encapsulate all kinds of rights provided in primary and secondary EU law. Importantly, the sources of EU rights may extend beyond the strict borders of the Union’s legal order. Within the EU, there appears to be a complex interplay between rights that stem from formal Union law, ordinary international law (e.g. UN Charter), the ECHR, as well as common constitutional traditions of Member States.P. Craig, EU Administrative Law (OUP 2006) 483. Rights are provided in EU Treaties, as well as Treaties entered into between the Union and third parties. Furthermore, secondary legislation, such as Directives, Regulations and Decisions, may also create rights. Rights can also be created by legal acts of Union executive authorities or in the case law of the Court. The status of the general principles of EU law is more complicated. Rather than declaring clear rights and obligations, their normative content is provided in an abstract and unwritten form; yet, via legal reasoning, it is possible to derive specific rights which are applicable in concreto.For a comprehensive analysis, see T. Tridimas, The General Principles of EU Law (OUP 2006). Although rights are commonly discussed with reference to individuals, it is clear that public authorities and other legal persons can also be right-holders (and of course duty-bearers).

The HohfeldianW. Hohfeld, Fundamental legal conceptions as applied in judicial reasoning, and other legal essays (Yale University Press 1919). analysis of rights, which although influenced by the common law tradition is not isolated from the continental scholarship,See A. Somek, ‘The indelible science of law’ [2009/7] International Journal of Constitutional Law 424-441, 433, where it is argued that Jellinek’s work (G. Jellinek, System Der Subjectiven Öffentlichen Rechte (2nd edition, Mohr 1905)) resembles, to an extent the Hohfeldian analysis. is widely perceived to be one the most comprehensive attempts to accommodate a wide range of uses of the term right in legal scholarship and practice. It is therefore not surprising that some commentators have used Hohfeld’s classification to analyse EU rights.See inter alia C. Hilson & T.A. Downes, ‘Making sense of rights: Community rights in

E.C. law’ [1999/24(2)] European Law Review 121-138; P. Eleftheriadis, ‘The Direct Effect of Community Law: Conceptual Issues’ [1996/16(1)] Yearbook of European Law 205-221; and Bengoetxea,

supra n. 52. Following Finnis’s interpretation,Finnis, supra n. 46 at 198-230. the Hohfeldian theory entails two fundamental assumptions. First, rights can be classified into four types, namely: (a) ‘claim right’ (‘right stricto sensu’ in Hohfeld’s vocabulary); (b) ‘liberty’ (‘privilege’ in Hohfeld’s vocabulary); (c) ‘power’; and (d) ‘immunity’. Second, the existence of a right implies a tripartite relationship between two persons (natural or legal) and a subject-matter, be it an act, thing or state of affairs. Based on these general premises, Finnis reconstructs four types of logical relations (where A and B signify the right-holder and duty-bearer respectively; and φ some act):

‘(1) A has a claim right that B should [be compelled to] φ, if and only if B has a duty to A to φ.

(2) B has a libertyLiberty entails circumstances where B may do something; in other words, some permissible act or a situation where B is free to carry out a certain conduct without this threatening to interfere with a claim right of A. (relative to A) to φ, if and only if A has no-claim right (‘a no-right’) that B should not φ.

(2’) B has a liberty (relative to A) not to φ, if and only if A has no-claim right (‘a no-right’) that B should φ.

(3) A has a powerPower entails circumstances where B can do something; it is correlative of liability and the opposite of disability (i.e. no-power). It concerns a relation where A can determine a legal relation either between A and B or between B and a third person. (relative to B) to φ, if and only if B has a liabilityLiability is the correlative of power and the opposite of immunity (i.e. no-liability). It refers to a relation where B may have his legal position determined by the voluntary act (power) of A. to have his or her legal position changed by A’s φ-ing.

(4) B has an immunityImmunity is the correlative of disability (no-power) and the opposite of liability (no-immunity). It refers to a relation where A has no legal power to determine the legal position of B. (relative to A’s φ-ing), if and only if A has no power (i.e. a disabilityDisability is the correlative of immunity and the opposite of power. It refers to a relation where A has no power to determine the legal position of B.) to change B’s legal position by φ-ing.’Finnis, supra n. 46 at 199.

In relations (3) and (4), the act φ would typically be an act of public authority. In other words, the conferral (or withdrawal) of some power on a public body brings about a change in a legal position of that body. By contrast, in relations (1), (2) and (2’), the act may denote either acts of public authorities or acts of private persons.Ibid. Perhaps the most important contribution of Hohfeld’s conceptual framework is the distinction between claim right and liberty. A’s claim right would always have B’s duty as its correlative. On the other hand, liberty refers to A’s freedom from duty; hence the correlative of A’s liberty would be the absence or negation of a claim right that B would otherwise have. When the subject matter of a claim of a right is the right holder’s own act or omission, then this is necessarily either a liberty or a power (in the case of public acts).

It is worth further elaborating on the nature and function of claim rights, the paradigmatic manifestation of rights. It has already been mentioned that, by definition, a claim right involves some counterpart duty or obligation.The two terms are used here interchangeably. The existence of counterpart duties is precisely what distinguishes rights from other non-legally protected interests, which may or may not be linked to that right.Tasioulas, supra n. 54 at 33. It separates rights from mere political or social aspirations. Moreover, a statement determining a right can be made only by reference to a counterpart duty.Ibid. A person’s right entails a duty of another person as its counterpart, which may be negative or positive.The distinction between negative and positive duties as counterparts of rights can be blurred and should be treated with some caution; all rights will typically have both positive and negative duties as their counterparts. See, H. Shue, Basic Rights: Subsistence, Affluence, and U.S. Foreign Policy, (2nd edition, Princeton University Press 1996). A positive duty would stand for a duty to undertake a course of conduct – to provide the right-holder with something or to assist the right-holder in a certain way. A negative duty signifies a duty of forbearance – an obligation not to interfere or not to treat the right-holder in a certain way.Tasioulas, supra n. 54 at 27. The rights discourse seems to approach the legal relation between persons from a specific viewpoint, that is of the person who benefits from that relationship.Finnis, supra n. 46 at 205. It is a discourse that focuses on norms generating ‘duties that guide the conduct of others in relation to the right-holder.’Tasioulas, supra n. 54 at 34.

What components must be identifiable for establishing the existence of a right is unsettled in the literature. Some scholars suggest that analysing a claim right in specific instances requires identifying the following elements:Finnis, supra n. 46 at 218-219. (a) the duty-beater(s) who must give effect to a person’s right; (b) the content of the duty, which includes the required act, as well as the circumstances and relevant conditions that are attached to it; (c) the right-holder(s); (d) the conditions under which the right-holder loses the claim right, including conditions (if any) for waiving the relevant duties; (e) the claim rights, powers, and liberties of the right-holder in the event that the duty is not performed; and (f) the liberties of the right-holder and their limits (e.g. non-interference with other persons’ liberties or rights). This latter element (f), which refers to the duties of a right-holder, can only be analysed fully if a comprehensive account is made of the correlative rights of any other persons who may be linked to that legal relation.

In EU law, there is evidence that the degree of specificity of the personal and material scope of a right can be decisive for its successful invocation or claimability, yet not necessarily for the existence of a right. For instance, the Court distinguishes the question pertaining to a right’s existence from an assessment as to whether it can have direct effect. For a right to be recognised as having direct effect and, thus, enable individuals to invoke it before national courts, it needs not only the fact of existence but also to satisfy certain conditions; it must generally be sufficiently clear, precise and unconditional.P. Craig & G. de Búrca, EU Law: Test, Cases and Materials (6th edition, OUP 2015) 202. On this basis, it can be argued that the existence of EU rights does not depend on their claimability. This is not to say that the question of claimability of right is insignificant; by contrast, it is clearly vital for its enforcement in concrete cases.

This leads to another important question, that is whether a right can exist independently of any legal remedies being in place that would enable the right-holder to enforce it. In other words, is the enforceability of a right a sine qua non condition in order to count as a right? EU scholars have been debating this issue for quite some time expressing opposing ideas.For the claim that rights are conceptually dependent on their judicial enforceability see inter alia W. Van Gerven, ‘Of Rights, Remedies and Procedures’ [2000/37] Common Market Law Review 501-536. For the opposite view, see inter alia Beljin, supra n. 52 at 91-122; D. Edward, ‘Direct Effect: Myth, Mess or Mystery?’ [2002/2] Diritto dell ’Unione Europea 215-227. From one point of view, rights are there to be enforced.Raz, supra n. 59 at 230. They are ‘conceptually oriented toward positive enactment by legislative bodies’J. Habermas, The Postnational Constellation: Political Essays (Polity Press 2001). and call for public institutions to intervene and adopt measures that would ensure their implementation. So, it is possible to see rights as necessarily enforceable legal claims which a person can bring against those who shoulder the correlative obligations. The enforceability of such a claim may take one of the following forms: either the duty-bearer would be forced to carry out the obligation or, at a minimum, there must be an alternative remedy, typically some compensation, for their failure to perform that obligation.Tasioulas, supra n. 54 at 79. Quickly, a question arises, namely what kind of mechanisms must be available to satisfy the enforceability requirement of rights? This shifts the focus of an inquiry pertaining to the existence of a right towards the specificity and effectiveness of the applicable institutional arrangements for its enforcement. Yet, one should not confuse the two clearly distinct issues. For it is one thing to ask whether the enforceability of a right is a condition for its existence, and another, what are the conditions for guaranteeing a right’s effective enforcement.

Some evidence indicates that, today, civil law and common law traditionsBeljin, supra n. 52 at 95 et sec. and the works cited therein. conceptually separate the existence of a right from the available legal remedies. The Union’s legal order somewhat also reflects this position;See e.g. Case C-263/02 P Jégo-Quéré, EU:C:2004:210, para 29; Case C-50/00 P UPA, EU:C:2002:462, para 39. Also, H. Hofmann & C. Warin, ‘Identifying Individual Rights in EU Law’, University of Luxembourg Law Working Paper No. 004-2017, available at: https://ssrn.com/abstract=300380. Note, however, that AG Sharpston appears to endorse the opposite view in her Opinion in Case C-413/15 Elaine Farrell v. Alan Whitty and Others, EU:C:2017:492, para 32: ‘Because rights under EU law must be effective, no right can exist without a corresponding remedy (“ubi jus, ibi remedium”).’ sometimes, EU law either does not provide for any remedy or poses restrictions on the kinds of persons able to enforce a right. On this basis, rather than being a condition for an EU right’s existence, a legal remedy is seen as giving effect to the normative implications of a right; in other words, it provides the process or the tool for enforcing something that already exists in EU law.Tasioulas, supra n. 54 at 84. The legal remedy per se does not generate substantive duties; it only facilitates the compliance of duty-bearers with an existing duty that corresponds to an existing right. Consequently, it appears that the claimability and enforceability of EU rights are conditions for their activation but not decisive factors for their existence.

4.4.Types of legal effects and avenues for judicial review

In the light of the preceding analysis, it is possible to reconstruct the IBMCase C-60/81 IBM v. Commission, supra n. 31.test as follows. A reviewable act is a Union act that produces legal effects capable of causing some change in a legal relation between two persons, consisting of rights or duties. However, not all kinds of legal effects would satisfy the IBM test. It appears that the justiciability of a Union act is dependent not only on changing rights or duties in a legal relation but also on the manner via which such a change is created. To illustrate this point, Alexander Türk and I have proposed to classify legal effects in two categories, namely primary legal effects and secondary legal effects.For a more elaborate presentation of this conceptual framework see, Türk & Xanthoulis, supra n. 21.

Primary legal effects are what is broadly understood as binding legal effects. Such legal effects have an impact in a legal relation between two persons by determining rights and counterpart duties therein. Primary legal effects can arise either directly or indirectly, depending on whether the act per se comprises the source of a change in a legal relation (arising directly) or whether the act merely triggers or enables another legal provision to produce legal effects in a given case (arising indirectly). All positive acts and omissions that produce primary legal effects can be challenged using a direct action for annulment under Articles 263 and 265 TFEU respectively. If a Union act can be reviewed directly, then by logic the Court can also consider its legality when the matter is brought indirectly in the context of preliminary ruling proceedings

(Article 267 TFEU) or incidental review (Article 277 TFEU).

An act may also produce what we call secondary legal effects, which are distinct from primary legal effects and emerge in the following manner. An act entailing secondary legal effectsdetermines a subsequent act which produces binding legal effects (in our terminology primary legal effects).The legal connection between the two acts may take various forms. Typically, the prior act becomes the legal basis for the adoption of another act, e.g. by requiring, authorising or permitting its adoption. More broadly, the normative content of an act must be determinative, in one way or another, for the adoption or the normative content of another act. For an account of the case law on this matter see, Lenaerts et al., supra n. 5 at 445-447. From the point of view of the affected legal relation, the prior act can be said to produce secondary legal effects vis-à-vis the parties to that legal relation, whose individual legal positions were changed by the latter (binding) act. Acts (and omissions) producing secondary legal effects vis-à-vis a person would not be reviewable via direct actions filed by that person. Their validity can be contested only indirectly underArticles 267 and 277 TFEU. To this effect, the person whose legal position was affected must first file an action against a final act with primary legal effects at national or Union level, and, thereafter, contest the legality of a prior Union act that has secondary legal effects.

It is possible for a single act to bring about more than one kind of legal effect; in other words, the occurrence of an act can result in different kinds of changes in legal relations. Where an act produces both secondary legal effects and primary legal effects, the available judicial avenue(s) for contesting its validity would be dependent on whether the applicant’s legal position was changed as a result of primary or secondary legal effects.

5.Physical acts in the case law of the Court

The language of acts is dominant in EU legal discourse; courts and scholars commonly refer to specific legislative, executive and judicial acts to analyse the means through which the Union exercises its mandate under EU law. Nevertheless, EU law has yet to offer a definition of acts entailing some physical conduct,The power of courts to review the legality of factual conduct only recently became a relatively settled matter in administrative law. For instance, up until 1950s, German courts did not exercise judicial review over the factual conduct of administrative authorities and persons could bring actions for damages under restrictive conditions. Today, under Article 19(4) of the Basic Law and section 43 of the Code of Administrative Practice 1960 (Verwaltungsgerichtsordnung), a judicial remedy is available against all forms of administrative conduct. See Rademacher (2017), supra n. 6 at 426 and the works cited therein. By contrast, common law never distinguished between legal acts and factual conduct. With certain exceptions, common law courts accept the admissibility of judicial review claims irrespective of the nature of the challenged act, provided that the act is issued by a body acting in public function. See, P. Cane, Administrative Law (OUP 2011) 266-278. explain how they differ from legal acts or provide clarity about their legal effects and justiciability. Instead, the case lawThis section does not seek to review comprehensively all cases in which factual conduct was at stake before EU courts. From a methodological perspective, we applied a selective approach that involved a two-stage process: First, we identified cases of EU courts that contained any of the following search terms: ‘physical conduct’, ‘factual conduct’, ‘physical act’ and ‘factual act’. Thereafter, by applying a qualitative analysis of these cases, we identified a second group of cases, where EU courts followed an equivalent approach or due to the similarity of the type of Union conduct contested. presents examples of conduct sometimes described as physical acts, which include, among others, the processing of personal data,Rademacher (2017), supra n. 6 at 399. the public releaseCase T-193/04 Tillack v. Commission, EU:T:2006:292. and communication of information to certain persons,Case C-53/85 AKZO Chemie BV and AKZO Chemie UK Ltd v. Commission, EU:C:1986:256. the physical transfer of goods or persons,Case C-263/16 P Schenker v. Commission, EU:C:2018:58, para 23. and driving a vehicle.Opinion of AG Gand in Case C-9/69 Sayag and Others v. Leduc and Others, EU:C:1969:31,

paras 340 and 342.

As regards their capacity to produce legal effects, in certain circumstances, the Court has considered that the physical or factual conduct of Union institutions can give rise to a reviewable act on the basis that it underlines an implied or tacit decision. In this scenario, a physical act brings about a change to the external world by implementing some legislative act or executing an administrative decision of the same or another public authority.R. Parker, ‘The Execution of Administrative Acts’ [1957/24(2)] The University of Chicago Law Review 292-313, 292. Such a physical act is not independent, but rather parasitic in nature; it is integrally linked to a prior normative act, which becomes, effectively, realised through the performance of the physical act in question.The executive or implementing function of such physical acts can be twofold: Physical acts may execute either positive or negative prior decisions. By adopting a positive decision, the Union authority would prescribe the carrying out of some physical conduct. A negative decision, on the other hand, would require the exercise of physical forbearance. Here, only the prior normative act would be regarded as capable of producing legal effects vis-à-vis third parties and reviewable under Article 263 TFEU. Where the prior normative act is issued in the form of some (preferably formal) written instrument, the task of locating the source of the legal effects produced is relatively clear.

Less straightforward are circumstances where it is not possible to identify any distinct written instrument containing a prior normative act. In these cases, the absence of an apparent legal act poses questions about the nature, function and effects of physical acts. In AKZOChemie BVCase C-53/85 AKZO Chemie BV and AKZO Chemie UK Ltd v. Commission, supra n. 98. (hereinafter ‘AKZO I’), the Court was asked to rule on the validity of a Commission decision to communi-cate confidential information – allegedly containing the business secrets of a company – to a competitor company, in the context of competition proceedings. In assessing the issue of reviewability, the Court applied a distinction between, on the one hand, the ‘physical act’ of communication of documents to a third party and, on the other hand, the previous ‘tacit’ Commission decision, which enabled the disclosure of these documents. The former was regarded as ‘merely’ implementing the latter; as such, the Court concluded that the Commission decision was solely capable of changing the legal position of the applicant company.Ibid, para 17.

The Court’s finding in AKZO I raises two important questions. First, in circumstances where a Union authority has carried out a physical act, under what conditions would an implied prior decision be deemed to exist? Second, what is the process or the means through which one can determine this matter? The Court has provided little guidance in this regard. Its brief reference to a Commission’s letter of 18 December 1984 to the applicant notifying them about the disclosure does not seem to settle the matter. Despite being the contested act in that case, the said letter had been issued after the actual disclosure of information; hence, logically, the letter per se could not have been the legal act that caused its execution through the performance of the Commission’s physical act of disclosure. Consequently, in the absence of any prior decision, be it written or oral, it seems that the Court had no option other than to conclude that the Commission’s decision had been ‘implied’. In other words, the Court’s conclusion in AKZO I can be explained by accepting that the Commission’s factual conduct was preceded by a fictitious or implied decision, the content of which was communicated to the applicant (via the abovementioned letter) after the disclosure of information had already occurred.

In any event, the fact of existence of an implied Commission decision, could not, on its own, justify its reviewability under Article 263 TFEU. According to the Court, the said decision produced legal effects (and hence was reviewable) because ‘it withheld from the […applicant] the [claimed] protection provided by Community law’.Ibid, para 18. In his Opinion in AKZO I, AG Lenz also found that the circumstances resulted to a change in the legal position of the applicant. He traced the origin of these legal effects in the Commission’s legal assessment, which, in his view, entailed a decision as to which information should be disclosed. By contrast, the physical act of handover of the confidential information was considered to be a mere factual step. Having said that, in a somewhat puzzling statement, which seems to contradict his own and the Court’s conclusion, AG Lenz stated that the ‘infringement of a right protected by law must be distinguished from a change in legal position’ (Opinion of AG Lenz in Case C-53/85 AKZO Chemie v. Commission, EU:C:1986:256, p. 1972). RademacherRademacher (2017), supra n. 6 at 405-406. understands that the Court’s statement is inconsistent with the test introduced in the IBM case. He argues that ‘the only implicit or tacit decision that can be construed as having been inherent in the “physical act” of communication was the affirmation of its own lawfulness’,Ibid. which, on its own, could not change the legal position of the applicant. In this case, he continues, the Commission’s decision did nothing more than reflecting its belief that the applicant had no right to confidential treatment. Rademacher acknowledges that the situation would be different if the Commission had ‘remove[d]’ the applicant’s right.Ibid. Yet, it seems that this is precisely what happened in AKZO I. By adopting its (implied) decision, the Commission pursued to unilaterally give rise to a new legal relation, whereby the Commission had the power to carry out certain conduct – that is, the disclosure of information – and the applicant was liable to accept the disclosure of such information by the Commission. In doing so, the Commission effectively denied the previous legal relation, which the applicant thereafter tried to restore through a direct annulment action before the Court. This view justifies the outcome in AKZO I and enables a broader conclusion to be reached. An act of Union institution A that interferes directly with an EU right of person B – and, hence, fails to comply with A’s counterpart duty owed to B – can be regarded as capable of changing B’s legal position (primary legal effects) and, thus, a reviewable act in the meaning of Article 263 TFEU. This interpretation brings the Court’s reasoning in AKZO I in line with the IBM formula and confirms that the scope of this test extends to Union acts not only imposing duties but also determining the rights of persons.

The Court’s approach in AKZO I has been confirmed in subsequent cases with similar factual backgrounds. In AKZO II,Joined Cases T‑125/03 and T‑253/03, Akzo Nobel Chemicals and Akcros Chemicals v. Commission, EU:T:2007:287, upheld on appeal (Case C-550/07 P Akzo Nobel Chemicals and Akcros Chemicals v. Commission, EU:C:2010:512). the applicant challenged the Commission’s seizure of a company’s document and its placement on the competition investigation file without putting it in a sealed envelope. The Court held that, in the absence of a formal rejection decision, such a physical act ‘necessarily entails a tacit [or implied] decision by the Commission to reject the protection claimed by the undertaking’,Joined Cases T‑125/03 and T‑253/03, Akzo Nobel Chemicals and Akcros Chemicals v. Commission, ibid, para 49. i.e., that this document should be protected on grounds of legal professional privilege. As in AKZO I, the physical act per se was not regarded as producing any legal effects; these were thus traced back to an implied prior legal act. The physical act of seizing the document was regarded as the means through which the prior Commission decision had been ‘expressed’.Ibid, para 52. The rationale underlying the capacity of the Commission’s decision to change the legal position of the applicant follows the line of reasoning in AKZOI. The Commission is consequently assumed here to have adopted a decision to reject the protection of a person’s rights;Ibid. in doing so, the decision altered the legal relation between the Commission and the person.

The significance of the AKZO cases lies further in that they present characteristic examples of instances where the Court has exceptionally accepted to review the validity of preparatory acts of Union authorities. In broad terms, preparatory acts form part of a larger administrative process, which may involve various institutional actors and procedural stages both at Union and national level.Hofmann et al., supra n. 6 at Chapter 11. As a general rule, preparatory Union acts are regarded by the Court as not capable of bringing about a final or conclusive change in the legal position of a person not participating in the decision-making process.Türk, supra n. 5 at 17-23. Such an effect (primary legal effect) is typically accepted to be caused by the final normative output of an administrative process, which, sometimes, is a legally binding measure. The rationale behind this is that, were judicial review of preparatory measures were to be the general rule, this would empower Union courts to decide on questions on which an EU authority may have ‘not yet had an opportunity to state its position definitively’.See Case T-41/16 Cyprus Turkish Chamber of Industry, EU:T:2016:613, para 42. According to the Court, such an outcome can pose risks for the division of powers between judicial and administrative authorities.Case C-60/81 IBM v. Commission, supra n. 31, para 20; Case T-64/89 Automec, EU:T:1990:42, para 46. Nevertheless, in certain circumstances, the Union courts have allowed intermediate steps to be challenged but the justification is not always clear. Intermediate acts have been found reviewable where an action against the final decision was ‘not capable of ensuring sufficient legal protection’Joined Cases C-463/10 P and C-475/10 P Deutsche Post and Germany v. Commission, para 56. or where the effects of the intermediary measure were ‘not capable of being rectified in proceedings against the final decision’.Opinion of AG Bot in Joined Cases C-463/10 P and C-475/10 P Deutsche Post and Germany v. Commission, para 75. These conditions would likely be satisfied where an act causes significant adverse interference with the function and day-to-day conduct of natural and legal persons or is capable of bringing about irreparable damage to their reputation, which cannot be rectified by challenging the final measure. For example, a decision to conduct on-site inspections on the business premises, land and means of transport of persons would likely be reviewable under Article 263 TFEU, where their refusal would trigger the power to impose sanctions.By analogy to what applies with respect to competition law proceedings see, e.g. Case T-289/11 Deutsche Bahn and Others v. Commission, EU:T:2013:404, and EU Regulation 1/2003, Arts 20(4) and 21(1). Union authorities hold powers to undertake such on-site inspections in the field of banking supervision,Article 12, EU Council Regulation No 1024/2013 conferring specific tasks on the European Central Bank concerning policies relating to the prudential supervision of credit institutions, OJ L 287 of 29.10.2013 (SSM Regulation). Article 13(2) SSM Regulation expressly provides that the ECB decision to carry out on-sight inspections are subject to review by the CJEU. police cooperation,See, Article 5(5) and 6(1), EU Regulation 2016/794 on the European Union Agency for Law Enforcement Cooperation (Europol) and replacing and repealing Council Decisions 2009/371/JHA, 2009/934/JHA, 2009/935/JHA, 2009/936/JHA and 2009/968/JHA,

OJ L 135 of 24.5.2016. and anti-corruption,EU Regulation 883/2013 concerning investigations conducted by the European Anti-Fraud Office (OLAF) and repealing EU Regulation 1073/1999 and Council Regulation (Euratom) 1074/1999, OJ L 248/1 of 18.9.2013. among other policy areas.

By applying these principles to the AKZO cases, it is possible to deduct the following conclusions: First, the physical conduct of a Union institution entailing a prior decision would be treated as any other legal act, and hence may function as a preparatory measure within a wider investigation process for an alleged breach of EU law. Second, the material facts in the AKZO cases are amongst the exceptional circumstances where a preparatory measure can be reviewable on the grounds that, if it were otherwise, no action against the final measure would grant sufficient judicial protection to the applicants. For instance, in AKZO I, had the Court decided in an action against the final measure that the Commission’s decision to show (and its physical act of disclosing) the documents to AKZO’s competitor were unlawful, the harm suffered by AKZO could not be undone.See also, Joined Cases T-10/92 R etc. Cimenteries CBR and others v. Commission, EU:T:1992:123.

The legal effects of the contested Commission decisions in AKZO I and AKZO II can be distinguished from instances where a Union institution communicates certain information about a private person to another public authority and the latter, thereafter, uses that information to adopt acts with binding legal effects vis-à-vis that person. In Tillack,Case T-193/04 Tillack v. Commission, supra n. 91. OLAF had forwarded certain information to Belgian authorities implicating an individual to some unlawful conduct. Subsequently, these authorities relied on this information to initiate an investigation against the said individual, in the course of which they searched his home and seized documents that were in his possession. The General Court concluded that the transfer of the information was not a reviewable act under Article 263 TFEU; however, the applicant had the right to ask the national court to make a preliminary reference to the Court of Justice on the question of validity of that EU act.Ibid, para 80.

At first sight, the outcome in Tillack may seem to contradict the line of reasoning developed in the AKZO saga discussed above. Arguably, the risk of a person suffering damage or having its rights interfered is real, irrespective of whether a Union body communicates certain information to a private person or another public authority.Rademacher (2017), supra n. 6 at 409; Hofmann et al., supra n. 6 at 671-672. Having to wait, without any judicial remedies, until a public authority uses certain circulating information to adopt a legally binding decision against a person or until that information is communicated to other private parties may seem unreasonable or excessive.Hofmann et al., ibid at 688, 671.

Yet, in certain circumstances, it is not entirely unjustified to allow the flow of information between public authorities to remain on unchallenged until a final decision has been adopted. It strikes an imperfect balance between the need to guarantee an effective legal protection under Article 47 EU Charter and the value of safeguarding the institutional balance in the Union’s architecture, which includes allowing administrative authorities, both at Union and national level, to effectively perform their mandate. Consequently, in cases like Tillack, where a series of administrative acts are involved, the contested prior Union act can be regarded as capable of producing only secondary legal effects vis-à-vis the person affected by the final decision; as such, it would not be directly reviewable. Its validity can be challenged indirectly in the course of judicial proceedings against the final act under either Article 267 TFEU or Article 277 TFEU. This is in line with our proposed hypothesis (see section 4.4) that the available judicial avenue(s) for challenging the legality of a Union acts is depen-dent not only on the capacity of that act to have an impact on a legal relation but also on the manner through which this consequence occurs.

Limiting the avenues of judicial control of physical conduct to indirect review poses familiar – yet, nonetheless important – restrictions to judicial protection. First, at the time of review of the final act, it may sometimes be too late for the affected person to receive meaningful protection for detrimental effects caused by a preparatory factual act in the course of the administrative process.Ibid, at 688 and 672. Second, in composite administrative procedures where a final act is adopted at the national level, the affected person must satisfy the applicable national procedural requirements guarding any request to the Court for a validity ruling under

Article 267 TFEU. The lack of uniform normative and institutional arrangements combined with the diversity of judicial practices between Member States causes a degree of unequal access to legal remedies. Finally, indirect review is always dependent on the existence of a final act producing legally binding effects; in the absence of such a final act, a person would not be allowed to seek the review of any preparatory factual conduct.  

6.Legal effects of administrative physical acts andavenues for judicial control

The preceding analysis has shown that the justiciability of factual conduct and the available avenues for judicial control depend on the specific function of physical acts in individual circumstances and the manner through which their occurrence results in changes to legal relations. On this basis, it is possible to identify the following categories of physical acts, in terms of the possible effects they produce:

6.1.Discretionary physical acts entailing decisions

First, there are AKZO-type cases, where physical acts of public authorities can become the means for translating the mandate of a prior legal act, specifically an actual or implied decision, into reality.These instances can be distinguished from other cases where the substantial interference with a person’s rights takes place primarily via an actual legal act rather than the performance of substantial physical conduct. Examples of such instances include a decision of the European Parliament to reject a petition by a Union official (Case C-261/13 P Schönberger v. Parliament||CEC||, EU:C:2014:2423, para 22) and a Commission’s letter rejecting a company’s request to access the investigation file (Opinion of AG Mischo in Case C-170/89 BEUC v. Commission, EU:C:1991:112, pp. 5726-5727). The same applies with respect to informal legal acts, such as letters, issued towards implementing a decision (Joined Cases T-369/94 and T-85/95 DIR International Film and others v. Commission, EU:T:1998:39, paras 52-55). The physical act executing an implied decision can be some positive conduct, such as ‘naming and shaming’ of Member States and private persons.For instance, ESMA publishes on its website ‘guidelines compliance tables’ outlining which national authorities comply or intend to comply with its guidelines by indicating a ‘Yes’ in green or a ‘No’ in red. See, Article 24(1)(e), EU Regulation 513/2011 amending Regulation 1060/2009 on credit rating agencies, OJ L 145/30 of 31.05.2011. Furthermore, it can concern certain situations of institutional ‘silence’, where a public authority simply does nothing; for instance, if a person receives no response from a Union authority regarding a written request, following the expiry of the applicable time-limit, the ‘silence’ of the institution, which can be perceived as some form of not only legal but also physical inaction, can be regarded as an implied rejection decision, under certain circumstances.Opinion of AG Darmon in Case C-126/87 Del Plato v. Commission, EU:C:1989:98, paras 15-16, where the Commission’s conduct was described as a ‘material act’. An analysis of the different forms and consequences arising from institutional ‘silence’ exceeds the limits of this work. In general terms, institutional silence can be subject to judicial review when it resembles an omission or failure to act. In the Draft Budget case, AG Mischo considered that a reviewable omission under Article 265 TFEU is the ‘non-adoption…of an act or measure, of whatever nature, form or description, which is capable of producing legal effects vis-à-vis third parties’ (Opinion of AG Mischo in Case C-377/87 Parliament v. Council, EU:C:1988:264, para 28). However, not all kinds of legal effects would enable a person to successfully file a direct action against an omission. For instance, the Court’s case law has prevented a direct action on the basis of a claim that an EU institution failed to adopt a recommendation or an opinion (Case C-15/70 Chevalley v. Commission, EU:C:1970:95; Case T-103/99 ACSV v. European Ombudsman and European Parliament, EU:T:2000:135, para 51-52). On the contrary, the Commission’s failure to provide a reasoned response to a complaint was held to be a failure to exercise its competence to decide whether or not a concentration which has not been notified to it falls within the scope of the merger regulation; hence it gave rise to an action under Article 265 TFEU (Case C-170/02 P Schlüsselverlag and others v. Commission, EU:C:2003:501, para 28-29). Similarly, in another case, the Court held that the complainant has a right to bring an action for failure to act if the Commission fails to comply with its obligation to either initiate a proceeding against a third party which is the subject of a complaint or to adopt a definitive decision within a reasonable time (Case C-282/95 P Guérin automobiles v. Commission, EU:C:1997:159, para 38). The existence of an obligation to act in this kind of cases seems to be decisive for determining the reviewability of an omission. Where a prior decision is classified as a preparatory act within a wider administrative procedure, its reviewability under Article 263 TFEU would be dependent on whether it can be placed among the exceptional circumstances recognised in the case law. In all other circumstances, it would be regarded as producing mere secondary legal effects vis-à-vis the person affected by the final decision; hence, Union courts can indirectly review its validity.

AKZO-type cases pose a persistent ontological question. Where it is not possible to identify an actual decision of a public authority, it remains somewhat unclear what circumstances enable or justify the Court’s recognition of an implied prior decision. As already mentioned, one way of solving this is to suggest that all physical acts that are intended to be carried out by a public authority necessarily give rise to an assumption that a prior decision was made prescribing the performance of such a physical act. For example, under ordinary circumstances, it can be presumed that a Commission official would not have carried out a physical act if he or she had not decided to do so. Whether there is an actual record of such a decision in a written instrument or not would arguably be immaterial. There is an important shortcoming in this argument. To the extent that such a prior decision has not been communicated to the external world, e.g. through an oral statement or written instrument, it may be an exaggeration to equal it to an independent act capable of producing legal effects. This is because, for an act to produce legal effects, it must be externalised rather than merely confined to the internal mental sphere of public officials.It is possible for such a ‘naked’ physical act to be subsequently dressed by institutional cloths e.g. an authority may adopt a legal act by which it approves and provides reasons for the performance of such physical conduct. From one point of view, this might strengthen the idea that the source of any legal effects produced in those circumstances lies in the physical act per se.Opinion of AG Lenz in Case C-53/85 AKZO Chemie v. Commission, supra n. 105 at 1972. Yet, perhaps a more convincing idea would be to accept the duality of the nature and function of physical acts in circumstances where it is not possible to identify any separate prior decision. The physical act in these instances embodies both the decision and its physical performance; it exists partly in the legal world and partly in the physical world. It simultaneously externalises and materialises the decision, which comes into being in the shape or form of a physical act. In this sense, it is not some prior (imagined) decision but the physical conduct to which the law attributes legal effects; a view that is in line with our proposion that any conduct can have legal effects depending on its normative evaluation.I owe this point to Alexander Türk. This is neither a contradiction, nor a novelty. It is in line with the Court’s consistent case law that, in determining its reviewability, the form of a Union act is, in principle, immaterial.Case C-60/81 IBM v. Commission, supra n. 31, para 9. Furthermore, the law recognises that rights and duties in contractual relations can arise merely through the physical conduct of participants, even where no single word has been uttered.

6.2.Physical acts performing non-discretionary duties

So far, we have considered scenarios where a Union institution’s physical act entails the exercise of discretionary powers. A second type of circumstances is where the Union authority enjoys no discretion in undertaking (or abstaining from) certain physical conduct. Here, the performance of a physical act occurs for the purpose of complying with an authority’s duty that does not allow for the exercise of meaningful discretion. When this happens, the physical act can be seen as nothing more than a factual step towards implementing a duty under EU law; hence, it is unable to produce any legal effects per se. In these circumstances, an applicant would normally seek to contest the legality of the EU provision providing for such a strict duty of Union authorities to act. For instance, in the hypothetical scenario where a decision is adopted ordering EU officials responsible for border controls to stamp all non-EU passports, the physical act of stamping per se can be regarded as not entailing meaningful discretion and, therefore, a factual step that ‘mechanically’ enforces a prior (revieable) decision.

6.3.Physical acts with no legal effects

In addition, there are instances where it is clear that no legal effects are produced by the performance of a physical act; for example, walking, driving a car and cooking, among others, are acts which do not in any way affect specific legal positions of others. Similarly, physical acts that function as proof that a legal requirement is met can be regarded as having no legal effects. For instance, the physical act of living in a certain space can be proof of residence that may support a pension claim in a certain jurisdiction (Case T‑416/04 Anna Kontouli v. Council, EU:T:2006:281, para 71).

6.4.Physical acts extinguishing legal relations

The primary function of judicial review, as previously mentioned, is the restoration of a valid legal relation of rights and counterpart duties between two parties, which was disturbed by some event. There appears to be certain circumstances, however, where the disturbance to a legal relation caused by a Union act is such that it cannot be restored back to its status quo ante; in a way, the prior legal relation is extinguished. In these cases, a physical act appears to cause a change in the legal position of a person which cannot be meaningfully remedied by a direct or indirect judicial review of its legality. Consider, for example, the case of a Union official inflicting unlawful physical damage to A at a single event that is now terminated. As it interferes with the rights of a person under Union law, specifically the right to the integrity of a person under Article 3 EU Charter, the physical act of inflicting unlawful damage made a change in A’s legal position. Here, the Court’s declaration of the unlawfulness of the act appears somewhat unfit to provide sufficient protection to the affected person per se. Thus, it becomes necessary to examine whether there are any grounds that justify allowing a person to seek a declaration by the Court that some Union act (physical or legal in nature) which changed its legal position is illegal, even if such a declaration would not effectively restore its legal position.

According to settled case law, non-privileged applicants must show an interest in the annulment of a contested act for a direct action under Article 263 TFEU to be admissible.See e.g. Lenaerts et al., supra n. 5 at 354-356, where it is argued that the Court’s analysis is sometimes blurred with the assessment on whether an act is reviewable under Article 263 TFEU. More specifically, the annulment of an act must bring some benefit or advantage to the applicants. Typically, this would be the elimination of the adverse repercussions on their legal position,Ibid, 355-356 and the case law cited therein. or the prevention of future repetition of the unlawful conduct.Case T-133/03 Schering-Plough v. Commission and EMEA, EU:T:2007:365, para 31. Such an interest must be personally linked to the applicant,Lenaerts et al., supra n. 5 at 359. in the sense that an applicant would not be allowed to challenge an act merely on the grounds of an interest to restore a breach of the law to which he or she is not connected to at all or only remotely connected.Ibid, 364. Moreover, the applicant’s interest must be specific and present. It follows, thus, that a hypothetical interest would not suffice for bringing a direct action for annulment. Where the interest invoked by applicants pertains to avoiding some future change in their legal positions, the occurrence of such an event must be certain and not merely possible. The Court may accept the admissibility of a direct action, notwithstanding that the legal effects of the contested act have, strictly speaking, expired; in other words, the measure has become obsolete. This is justified, according the Court, to prevent the Union institution from repeating the same or similar unlawful conduct in the future,Case C-53/85 AKZO Chemie v. Commission, supra n. 98, para 21; Case C-207/86 Apesco v. Commission, EU:C:1988:200, para 16; Case C-362/05 Wunenburger v. Commission, EU:C:2007:322, para 50; Case C-92/78 Simmenthal v. Commission, EU:C:1979:53, para 32; Joined Cases T-191/96 and T-106/97 CAS Succhi di Frutta v. Commission, EU:T:1999:256,

paras 62-63; Case C-239/12 P Abdulrahim v. Council and Commission, EU:C:2013:331, para 25;

Case T-540/15 De Capitani v. Parliament, EU:T:2018:167, para 32. See also, Case T-299/05 Shanghai Excell M&E Enterprise and Shanghai Adeptech Precision v. Council, EU:T:2009:72, paras 56-57, where the General Court stated that preventing the review of measures whose temporal effects are limited and which will expire after an action for annulment has been brought but before the Court is incompatible with the spirit of Article 263 TFEU and the fact that the Union is based on the rule of law, with reference to Case C-294/83 Les Verts v. Parliament, EU:C:1986:166, para 23. or where the annulment action can form the basis of a possible future action for damages, among other things.Case C-76/79 Könecke v. Commission, EU:C:1980:68, para 9; Case C-239/12 P Abdulrahim v. Council and Commission, EU:C:2013:331, para 64; C-183/12 P Ayadi v. Commission, EU:C:2013:369, para 62; Joined Cases C-68/94 and C‑30/95 France and Others v. Commission, EU:C:1998:148, para 74; Case T-299/05 Shanghai Excell M&E Enterprise and Shanghai Adeptech Precision v. Council, EU:T:2009:72, paras 53-55; Case C-149/12 P Xeda International and Pace International v. Commission, EU:C:2013:433, para 32; Case T-42/06 Gollnisch v. Parliament, EU:T:2010:102, para 71; Opinion of AG Mengozzi in Case C-430/16 P Bank Mellat v. Council, EU:C:2018:345, para 35.

From one perspective, by applying the above principles to physical acts, there may be a case against recognising the existence of a person’s interest to contest Union physical conduct in certain circumstances. Where the legal effects are caused primarily by the performance of some physical act, it seems that Union authorities are unable to do much to restore the prior legal relation. By nature, physical acts cannot be replaced or repealed. Even if they entail an actual or implied prior decision, the latter’s withdrawal or repeal would not substantially reverse the harmful effects. At most, where the factual conduct of a Union institution is performed repeatedly or continues to affect the legal position of a person, a Union institution can end the repetition of the said conduct or take action to prevent it from producing legal effects ex nunc. Yet, none of these steps correspond to a recognition of the unlawfulness of the relevant conduct and, in any event, always have prospective rather than retroactive effects.See by analogy Case T-228/02 Organisation des Modjahedines du peuple d'Iran v. Council, EU:T:2006:384, para 35. On the contrary, the withdrawal of legal acts sometimes takes effect ex tunc and has the same consequences as a declaration of their nullity. See, Lenaerts et al., supra n. 5 at 362, fn. 557. At the same time, by declaring the unlawfulness of a physical act in these circumstances, the Court can neither remove its legal effects ex tunc, nor compensate the affected person for any material harm suffered; the latter is typicallyNote, however, that, under Article 266 TFEU, a Union institution whose act or ommission has been declared unlawful is ‘required to take the necessary measures to comply’, which includes granting compensation to the affected party. made possible through a successful action for damages (see infra section 8).

Yet, the mere fact that a Union institution may be unable to restore a legal relation does not seem to provide a convincing justification for denying a person the right to ask the Court to declare its unlawfulness of the disruptive conduct.Case C-76/79 Könecke v. Commission, EU:C:1980:68, paras 8-9; Case C-239/12 P Abdulrahim v. Council and Commission, EU:C:2013:331, paras 64 and 80; Case C-183/12 P Ayadi v. Commission, EU:C:2013:369, para 77. Despite its weakened effects, an annulment action may still be meaningful in these instances. First, by declaring its unlawfulness, the Court communicates a clear signal to the Union authority that it should avoid carrying out certain conduct in the future,Case C-207/86 Apesco v. Commission, EU:C:1988:200, para 16; Case T-121/08 PC-Ware Information Technologies v. Commission, EU:T:2010:183, para 39; Case T-133/03 Schering-Plough v. Commission and EMEA, EU:T:2007:365, para 31. thus substantially limiting the risk of a person being subjected to the same or similar unlawful treatment. Second, where the unlawful physical act has significant negative consequences and considerable impact on the rights and freedoms of a person, the recognition of the illegality of the contested measure delivers justice, which, on its own, comprises some reparation for the non-material harm that a person has suffered.See by analogy Case C-239/12 P Abdulrahim v. Council and Commission, EU:C:2013:331,

para 72; C-183/12 P Ayadi v. Commission, EU:C:2013:369, para 70, where the Court accepted that individuals had an interest to have their names removed from a list of persons whose funds had been frozen by virtue of restrictive measures adopted with a view to combating terrorism, despite that the contested measures were repealed or withdrawn thereafter. Finally, where a person has suffered damage, a successful annulment action can pave the way for bringing a compensation claim. It is possible, therefore, for a person to retain an interest in challenging the legality of a Union physical conduct, even where it seems practically impossible for a Union institution to fulfil its obligation to comply with a judgment of the Court declaring it void.

Furthermore, it is clear that, where the unlawful conduct remains present or active and continuously produces binding legal effects detrimental to a person, the remedy of judicial review can be used to seek the annulment of that conduct and, by extension, the termination of the infringement of the person’s rights. The decisive criterion for determining the justiciability of an act would not be the duration of a person’s subjective suffering from the detrimental effects of a Union body’s conduct, but rather the capacity of that conduct to sustain the production of such legal effects.

7.Physical acts in composite administrativeprocedures

A more complex picture emerges where a Union physical act belongs to an administrative procedure that extends beyond the jurisdiction of a sole legal order and involves both national and Union authorities. Depending on its starting and ending point, it possible to categorise these so-called mixed or composite administrative proceduresF.B. Bastos, ‘Derivative illegality in European composite administrative procedures’ [2018/55(1)] Common Market Law Review 101-134; M. Eliantonio, ‘Judicial Review in an Integrated Administration: the Case of “Composite Procedures’’’ [2015/7(2)] Review of European Administrative Law 65-102. into two main types.

On the one hand, the administrative procedure may be initiated at the Union level and lead to the adoption of a final act by a national authority. In this instance, national courts may consider the validity of a preceding Union act, be it a legal instrument or some factual conduct; however, they have no jurisdiction to declare it invalid.Case C-314/85 Foto-Frost v. Hauptzollamt Lübeck-Ost, EU:C:1987:452, para 14. The latter belongs to the exclusive jurisdiction of Union courts.Ibid at 15. Whether the matter can be brought before Union courts through direct actions or only indirectly would depend on whether the relevant Union act produces primary or secondary legal effects, in line with the proposed analytical framework (see infra section 4.4).

On the other hand, a composite procedure can start at the national level and end at Union level. This scenario does not seem to pose challenges for the judicial review of the Union acts involved (as these clearly fall within the jurisdiction of Union courts). By contrast, it gives rise to important questions regarding the allocation of competences to assess the legality of national acts, which could also theoretically entail some physical conduct. Although the judicial review of national acts falls outside the scope of this article, it is useful to briefly engage with this matter, to the extent that it touches upon the scope of Union courts’ reviewing powers.

To begin with, we can distinguish between two alternative streams of cases. The first, which can be referred to as the Borelli-typeCase C-97/91 Borelli, EU:C:1992:491. For an annotation of this case see F.B. Bastos, ‘The Borelli Doctrine Revisited: Three Issues of Coherence in a Landmark Ruling for EU Administrative Justice’, [2015/8(2)] Review of European Administrative Law 269-298. cases, concerns instances where the EU administration enjoys limited or no discretion to diverge from a national preparatory act. The national preparatory measure, here, ‘is binding on the Community decision-taking authority and therefore determines the terms of the Community decision to be adopted’.Case C-97/91 Borelli, ibid at 10. Three important principles govern the judicial control of such national measures. The duty to review their legality rests on national courts.Case C-343/07 Bavaria NV, EU:C:2009:415, paras 55-57 and 64-67. Union courts have no jurisdiction to rule on the lawfulness of such national measures, notwithstanding that they form part of an EU decision-making procedure.Ibid. Finally, the possible illegality of a national preparatory measure cannot give rise to derivative illegality;F.B. Bastos, supra n. 149. in other words, the unlawfulness of the national measure cannot contaminate the validity of the final Union act, subsequently adopted within the same administrative process.For a more elaborate discussion on the underlying rationale see F.B. Bastos, ibid at 111 and 117.

The second type of cases, which follows the Court’s approach in Sweden v Commission,Case C‑64/05 P Sweden v. Commission, EU:C:2007:802. concerns situations where, unlike Borelli-type cases, a Union authority has meaningful discretion to depart from national preparatory measures and, hence, performs the defining decisional power in the composite procedure. In other words, the Union authority is not bound to adopt a specific final act with any particular content. In this latter type of circumstances, the Grand Chamber of the Court of Justice held in the recent Berlusconi judgmentCase C-219/17 Silvio Berlusconi and Finanziaria d'investimento Fininvest SpA (Fininvest) v.

Banca d'Italia and Istituto per la Vigilanza Sulle Assicurazioni (IVASS), EU:C:2018:1023. that national courts are prohibited from reviewing the national preparatory measure; instead, such a task is conferred exclusively on the Union courts. The rationale offered by the Court to justify this division of judicial competences focuses on the principles of loyal cooperation between the Union and the Member States (Article 4(3) TEU)Ibid at 47. and effective judicial protection,Ibid at 44. the effectiveness of administrative decision-making process,Ibid at 49. and the exclusivity of the Union court’s jurisdiction to review acts of Union authorities.Ibid at 50. One important question that emerges from the Berlusconi case concerning our analysis pertains to the type of judicial avenues that are available for Union courts to review the legality of national preparatory measures. The need to maintain consistency and coherence in the triggers of judicial review of Union courts seems to justify applying here, by analogy, the principles governing the justiciability of Union acts.This also seems to be the position of the AG Campos Sánchez-Bordona in Case C-219/17

Berlusconi, EU:C:2018:502, paras 108-109. On this basis, whether the question on the legality of national preparatory acts can arise through direct actions or only indirectly (under Article 267 and 277 TFEU) would depend on the kind of legal effects produced (primary or secondary) and the manner through which these eventually change legal relations in the context of the relevant composite procedure.

8.The Union’s non-contractual liability for physical acts

A person who has suffered damage as a result of an unlawful Union conduct has the right to seek compensation by bringing a direct action for damages on the basis of the Union’s non-contractual liability under Articles 268 and 340(2) TFEU. Sometimes, this remedial option would follow the success of an annulment action under Article 263 TFEU or indirect review under Articles 267 and 277 TFEU. At other times, however, where no judicial review is available, the action for damages can be the last resort for a person seeking judicial protection.

Rather than aiming to review the legality of Union acts producing legal effects, the action for damages is designed to provide compensation for harm suffered by a person as a result of unlawful conduct by a Union bodies, offices or agencies.Case C-131/03 Reynolds Tobacco and Others v. Commission, para 83. Note, however, that, in principle, Union courts do not have jurisdiction to consider action for damages pertaining to the conduct of Union authorities in the field of Common Foreign and Security Policy. This is not to say that the subject matter of an action for damages would not involve Union conduct producing legal effects. For instance, a public official unintentionally causing a car accidentCase C-9/69 Sayag and Others v. Leduc and Others, EU:C:1969:37. resulting in another person suffering damage can be regarded as having made a change in the legal position of that person (primary legal effects arising indirectly) to the extent that there was some interference with the latter’s rights to property and physical integrity, as guaranteed under the EU Charter. At the same time, the same act may have indirect primary legal effects vis-à-vis other parties, for instance, by triggering an obligation of an authority under EU law to investigate the circumstances surrounding the accident.

An analysis of the procedural and substantive requirements for a successful compensation claim falls outside the scope of this work. Suffice it to state, the form of Union authorities’ conduct is immaterial for the purposes of establishing the Union’s non-contractual liability; hence, physical acts are clearly capable of becoming the subject matter of an action for damages. The Union’s liability can arise not only from positive physical acts of its officials, but also from the failure to carry out certain factual conduct required under Union law; to paraphrase, where the Union infringes its legal obligation to act.See Case T-203/11 Transports Schiocchet ‒ Excursions v. Council and Commission, EU:T:2012:308, para 37. Only acts of Union officials carried out ‘in the performance of their duties’, in the sense of acts that are ‘necessary extensions’ of the tasks conferred to the institution to which the relevant official belongs, may give rise to Union liability.See Case T-259/03 Nikolaou v. Commission, EU:T:2007:254, paras 93 et sec.

The case law is not entirely clear about the kind of liability that needs to be established.Lenaerts et al., supra n. 5 at 512. Whilst Union courts generally put emphasis on establishing the existence of an objective sufficiently serious breach of a Union provision intended to confer rights,In the past, there was some confusion in the case law as to whether a rule that ‘entailed’ the grant of a right would suffice to satisfy the conditions for liability or something more was required. Possibly influenced by the German and Austrian Schutznorm doctrine, some cases seemed to suggest that the legal provision that was breached must have explicitly intended to confer an individual right on the damaged person so as to give rise to liability (see e.g.

Case C-222/02 Paul and Others, EU:C:2004:606). Hofmann & Warin, supra n. 89 at 5-8 convincingly argue that the Court’s approach, seen as a whole and in the light of more recent cases, does not endorse the Schutznorm doctrine or a ‘strict’ reading of ‘intention to grant right’

(Joined Cases C-5/66, C-13/66, C-16/66, C-21/66 Kampffmeyer and Others v. Commission of the EEC, EU:C:1967:31, p. 262; Joined Cases C-178/94, C-179/94, C-188 to C-190/94 Dillenkofer and Others v. Bundesrepublik Deutschland, EU:C:1996:375, paras 40-41; Case C-445/06 Danske Slagterier, EU:C:2009:178; and Case T-217/1 Staelen v. Ombudsman, EU:T:2015:238, para 73). the case law sometimes also considers whether the damage was caused by some ‘fault’ on behalf of the Union institution.Ibid. Where a Union authority enjoys some discretion in carrying out its tasks, the illegality of any physical act or inaction would typically require establishing some manifest and grave disregard of the outer limits of the conferred discretion.See Case T-285/03 Agraz and Others v. Commission, EU:T:2005:109, paras 39-40.

9.Conclusion

Although the factual conduct of public administrations has for long now been an organic part of traditional administrative law textbooks, it has only recently attracted the interest of EU law scholars. This is no coinci-dence, as it goes hand-in-hand with a rapidly growing transnational administrative architecture which is based on a complex, multi-level cooperation between national and Union authorities. Moreover, it reflects the expansion of EU competences in policy areas requiring ‘(EU) boots on the ground’, such as migration policy, border security and on-site inspections of private corporations in banking supervision. Against this background, analysing the effects and avenues for judicial control over the factual conduct of EU administrative authorities is of high relevance. Legal and physical acts share common conceptual foundations, since they are both means for exercising public power. To the extent that all manifestations of public power must adhere to the applicable legal requirements, the EU administration remains judicially accountable both for its legal acts and the performance of physical conduct, where these have an impact on the rights and obligations of others. In other words, where the performance of some factual conduct changes a legal relation between two persons, it triggers the right of an affected person to seek legal remedies.

In this article, we have shown that the uncertainty that characterises the justiciability of Union acts, including physical acts, conceals an underlying conceptual obscurity surrounding acts and their effects in EU law. In addressing this obstacle, we offered an analytical framework for basic notions of EU law, with primary focus on legal effects. Our main hypothesis is that – in addition to the kind of legal effects produced – the manner through which these have been brought about is decisive for identifying the available judicial avenue(s) for contesting the legality of Union conduct. Union acts with primary legal effects, that is, all acts or omissions that, directly or indirectly, are capable of deetermining rights or obligations of parties in legal relation, can be challenged by way of direct actions under Articles 263 and 265 TFEU. Where the change in the legal relation is caused by at least two distinct acts, the former determining the latter binding act without having itself primary legal effects, then the Union courts may review the former act only indirectly, such as through Articles 267 and 277 TFEU. The significance of the proposed analytical framework is twofold: first, it enables the systematic analysis of the factual conduct of EU authorities in its various forms. Second, it unveils the conceptual relations between legal and physical acts, on the one hand, and rights and obligations, as constitutive parts of legal relations and legalpositions, on the other hand. By unifying these concepts, it becomes possible to translate the language of legal effects of Union acts to a language of EU rights and obligations and vice versa.

The objective difficulty in classifying acts and their effects is mirrored in the Union case law. The Court provides an imperfect but practical equilibrium between effective judicial protection of persons adversely affected by unlawful Union conduct and the need of administrative authorities to effectively perform their mandate. A person can retain the right to seek a declaration of illegality of certain Union conduct, even where its legal position has been extinguished and cannot be restored. The role of Courts as ultimate guarantors of the rule of law in the Union requires that justice must be given in all instances, irrespective of whether a Union authority is practically able to reverse the harmful effect that may have caused to a person. Furthermore, in certain circumstances, the Union’s factual conduct can give rise to non-contractual liability under Article 340 TFEU. Where a Union official’s physical act has resulted to some unlawful damage to a person, this can be regarded as having brought about a change in the person’s legal position, should it interfere with that person’s right protected under EU law.