A. Introduction
The autonomy of the legal order of the European Union (EU or Union) has long been invoked by the Court of Justice of the EU (CJEU or Court) when reviewing the Union’s external action and, in particular, international agreements. Recently, and controversially, the Court’s autonomy analysis has triggered the foreclosure of investment arbitration in an intra-EU context; that is, between an investor from one member state against another member state, on the basis of both Bilateral Investment TreatiesFootnote 1 and the multilateral Energy Charter Treaty.Footnote 2
Still, much of the EU case law and subsequent academic focus suggests that autonomy is a European idiosyncrasy rather than a feature of all international organizations (IOs). This skews the legal discourse regarding the role of the autonomy principle in regulating the Union’s relationship with entities established under international law. To the extent it has been examined, the relevance of autonomy in IOs has mainly been considered from the perspective of international relations, often in relation to the discretion afforded to IO bureaucrats by the organizations’ member states.Footnote 3 However, there have only been limited discussions on the autonomy of IOs from a legal perspective; the relationship between autonomy and international law is thus relatively unexplored. Such an exploration will be attempted here, with a view to determining whether autonomy merely reflects the historical and institutional peculiarities of the European integration project or whether it is more broadly relevant as a characteristic of all IOs established under international law.
First, the EU conception of autonomy will be considered in light of the notion’s understanding under international law. Relevant CJEU case law give expression to known aspects of the autonomy of IOs, namely their political and institutional dimensions. This will provide an opportunity to explore the “multifaceted”Footnote 4 notion of IOs’ autonomy. The above suggests that, far from being an EU invention, it substantially draws from its international law conceptual origins. Second, the different normative status of autonomy under EU and international law will be analyzed. Autonomy has risen to the status of a legal principle within the Union legal order, whereas this is not the case under international law. Lastly, this difference will be sought to be justified. In particular, it will be argued that there is a positive correlation between the degree of an IO’s autonomy and the robustness of the dispute settlement mechanisms for which it provides. This arguably betrays the underlying relationship between autonomy and the rule of law and, more fundamentally, reflects the objectives that each IO has been created to pursue. Such a view casts EU autonomy as different from the autonomy of ordinary IOs, but only in so far as the EU differs from typical, non-constitutionalized IOs. The qualitative uniqueness of the EU principle of autonomy is, in this sense, justified under international law, as would the principle’s further intertwining with the Union’s commitment to judicial protection and the rule of law under Article 2 TEU.
B. The EU Conception of Autonomy in Light of Its International Law Understanding
This section will consider the conceptual similarities between the EU and international law understandings of autonomy. This will not entail a comprehensive exploration of the case law in which the autonomy of the EU legal order appears.Footnote 5 Illustrative references to CJEU case law will be made to facilitate a comparison with how the notion is understood in the international legal order. In general terms, the Court has long characterized the EU legal order as autonomous in order to safeguard the structural integrity of the Union’s legal and judicial system. This includes the allocation of competences between the EU institutions and member states and, crucially, its own jurisdiction under Article 19(1) of the Treaty on European Union (TEU)Footnote 6 and Articles 267 and 344 of the Treaty on the Functioning of the European Union (TFEU).Footnote 7
Considering the extent to which the EU’s understanding of autonomy is sui generis seems important in light of how it is often perceived by legal discourse. In particular, the autonomy principle has been painted as idiosyncratic to EU law—as a tool developed by the CJEU out of “fear”Footnote 8 if not “selfishness.”Footnote 9 The Court’s conceptualization of the autonomous nature of the Union legal order has been argued to “border on autarky.”Footnote 10 The above criticisms are grounded in the assumption that the EU is asserting, in a self-interested manner, a systemic independence that is unjustified under international law. Indeed, as the relevant case law concerns the relationship between the EU and its Court with international dispute settlement, the EU conception of autonomy is often seen as being directly invoked at the expense of international law.Footnote 11
The exploration below centers around two identified dimensions of the autonomy of IOs under international law: political and institutional.
By way of disclaimer, it should be noted that IOs constitute an international community of “an almost infinite diversity,”Footnote 12 which cannot be fully captured by the scope of this article. First, in general terms, this community includes organizations operating as a platform for state-to-state engagement; some have a global reach, such as the United Nations (UN) and the World Trade Organization (WTO), and others maintain a more regional focus, such as the Council of Europe (CoE). Second, some IOs promote regional (economic) integration. The European Union is the most developed organization of the latter category. However, the CJEU has long resisted accepting that the EU is merely an international—that is, intergovernmental—organization, emphasizing instead its supranational characteristics.Footnote 13 Other regional integration organizations include the African Economic Community, the Association of Southeast Asian Nations (ASEAN), the Caribbean Community (Caricom), and the Southern Common Market (Mercosur).
I. Political Independence, Volonté Distincte and International Legal Personality
Considerations of autonomy are relevant for all IOs, despite the morphological diversity outlined above, in the sense that the autonomy of an IO reflects its relationship with its member states. On the most basic level, autonomy may be conceived in political terms as it characterizes the scope of an IO’s powers or, conversely, the extent of sovereign powers member states have granted to or share with it. Jean d’Aspremont framed this as “autonomy as political independence.”Footnote 14 This could be seen to follow international law’s traditional understanding of autonomy on a national, regional, or local level, “determined primarily by the degree of actual as well as formal independence enjoyed by the autonomous entity in its political decision[-]making process.”Footnote 15 In the context of IOs in particular, autonomy in the political sense overlaps with an organization’s volonté distincte (separate will).Footnote 16 An IO’s ability to express and act upon a separate will is an essential by-product of a state’s decision to “structure their cooperation, to institutionalise it …, to become a part of a larger whole because of a shared interest.”Footnote 17 By virtue of such a decision, every IO must “be made capable of acting independently … in the exercise of its tasks”; such a separate will, distinct from those of the IO’s members, needs to “emanate not from any ad hoc grouping of states,” but by an institution or body to which the requisite competences have been allocated by states acting qua members.Footnote 18
It is generally accepted that European integration initially followed a largely intergovernmental paradigm,Footnote 19 which increasingly assumed supranational characteristics.Footnote 20 Since its inception, the European UnionFootnote 21 has had to address the organizational and administrative issues typically experienced by all IOs. These include defending its bureaucratic and operational autonomy; that is, its capacity to develop administrative procedures and to pursue its objectives in accordance with the powers granted by its member states. Indeed, such organizational matters still resurface occasionally.Footnote 22
However, the CJEU has also been central in ensuring that the EU is and remains politically independent, as described by d’Aspremont. First, in its early case law, the Court emphasized that EU member states have “limited their sovereign rights” for their “benefit” to justify the doctrine of direct effect.Footnote 23 In Costa v ENEL, the Union’s political independence was strikingly asserted by the finding that EU law, enacted within the framework established by member states, may not be “overridden by domestic legal provisions, … without the legal basis of the [Union] itself being called into question.”Footnote 24 This was confirmed in Internationale Handelsgesellschaft.Footnote 25 It is in the spirit of maintaining this framework that the CJEU further distinguished the Union legal order from its member states by noting that “[Union] law uses terminology which is peculiar to it” and that “legal concepts do not necessarily have the same meaning [as in] the law of the various Member States.”Footnote 26 This early development of the EU’s autonomy vis-à-vis national law seems to have been motivated by the Court’s desire “to prevent significant distortions as regards the application of EU law in the Member States.”Footnote 27 Such a risk of interpretative anarchy or fragmentation would undermine the EU’s ability to express a separate will effectively, which is a fundamental consideration for all IOs.
Second, the Court similarly safeguarded the effective expression of the Union’s separate will by invoking autonomy in order to ensure that the “allocation of responsibilities defined in the Treaties” between the Union institutions and its member states is not “adversely … affect[ed].”Footnote 28 This aims to safeguard the Court’s own jurisdictional monopoly.Footnote 29 At the same time, the reference to the Treaties and the use of the term “allocation” serves as a reminder of the politically sensitive degree of independence that member states have granted the Union and its institutions. The Court thus used autonomy as a principle that structurally reinforces the Union framework.
Autonomy as political independence typically presupposes an independent legal personality under international law. This may be inferred in light of the objectives the IO was established to pursue and the powers it has been given by its members for this purpose, as was done by the International Court of Justice (ICJ) in the case of the United Nations.Footnote 30 In the EU context, this was similarly clarified by the CJEU in the ERTA judgment,Footnote 31 even though the relevant Treaty provision was textually ambiguous as to whether the scope of the Community’s legal personality extends to international law as well as the national laws of its member states.Footnote 32 Article 47 TEU subsequently confirmed this. The international legal personality of the EU has previously been acknowledged as a constitutive part of the autonomy of its legal order.Footnote 33 This illustrates the connection between the objectives of an IO and the question of its legal personality and, importantly, the role of judicial bodies in the determination of the latter. Indeed, this role more broadly extends to the determination of the organization’s autonomy, which seems to be a more complicated exercise. As the ICJ suggested in its Advisory Opinion on the Legality of the Use by a State of Nuclear Weapons in Armed Conflict, autonomy is situated on a spectrum (“a certain autonomy”)Footnote 34 — unlike legal personality, the recognition of which resembles a “yes or no question.”Footnote 35 This suggests a degree of inherent conceptual flexibility, as identifying the autonomy of an IO entails not just the recognition of its existence but also a delineation of its scope. This exercise is typically undertaken by a judicial body, as with the ICJ Advisory Opinion and relevant CJEU case law. It requires a closer inspection of the organization’s particular institutional features and administrative-political practices. It thus allows a greater degree of interpretative discretion. In this sense, its findings may be somewhat open to interpretation and may be subject to change over time as the organization’s working realities evolve. Arguably, this casts a different light on criticism of the “self-aggrandizing” invocation of the autonomy of the EU legal order by the CJEU:Footnote 36 a court-centric approach to the determination of an organization’s autonomy is required by virtue of the concept itself, as understood by international law.
In light of the above, the international legal personality of an IO (such as the EU) serves as the basis on which it may operate in a manner that is politically independent of its member states, in so far as this is envisaged by the treaties from which it was created. The autonomy of an IO is thus necessarily circumscribed by the powers “bestowed” upon them by its member states “with a view to the fulfillment of [particular] purpose[s].”Footnote 37 This is echoed by the CJEU’s early case law, which speaks of the EU having “real powers stemming from … a transfer of powers from the States to the Community.”Footnote 38 Within this framework, the political independence of an IO can be expressed in terms of policy autonomy through an organization-level agenda and the joint pursuit of goals.Footnote 39
II. Institutional Independence and the Interaction of the EU with the International Legal Order
The understanding of autonomy outlined above may be “too narrow,”Footnote 40 as it largely focuses on the dynamic between the international organization and its members. After all, in the EU context, this matter seems to have long been settled by the CJEU in the case law considered above. Subsequent case law, particularly Opinions rendered by the Court under Article 218(11) TFEU starting with Opinion 1/91,Footnote 41 instead focused on the relationship of the EU and its legal order with international law and dispute settlement mechanisms established thereupon. This dimension of the autonomy of the EU legal order reflects the “institutional independence” possessed by IOs. Though the term will be adopted, “institutional” is meant as encompassing “legal,” as it characterizes the “impermeability of … organization[s] to external institutional interferences”; that is, the “extent to which [IOs] constitute a legal order distinct from the general international legal order.”Footnote 42 This is more sensitive to the autonomous global role of IOs, qualifying the traditional Westphalian outlook of international relations that emphasizes the dominant role of sovereign states.Footnote 43 This suggests that understanding autonomy as institutional independence would be more important for IOs that place an emphasis on having a global role as a single actor when interacting with other international entities. This appears particularly resonant with the EU, as it has increasingly, not least post-Lisbon, developed and pursued external policies qua Union.
Institutional independence implies a degree of systemic distinctiveness, as it would denote the ability of an IO to defend its “political project,” which may be threatened by external “interferences” stemming from the broader international community.Footnote 44 It is well established that the EU legal order has a heightened sense of its own systemic distinctiveness. Indeed, the CJEU made such a claim on behalf of the EU early on and gradually refined it. First, it stated that “the [then European Economic] Community constitutes a new legal order of international law”;Footnote 45 the Commission argued that EU law should not be “consider[ed] … in the light only of the general principles of the law of nations.”Footnote 46 Then, the Court ceased to refer to international law and found that the EU simply constitutes “a new legal order.”Footnote 47 In any event, Union law is of a “special and original nature.”Footnote 48 More recently, the Court emphasized that the EU “has a new kind of legal order, the nature of which is peculiar to the EU, [and] its own constitutional framework.”Footnote 49 In Opinion 1/17, the Court similarly noted that “the Union possesses a constitutional framework that is unique to it.”Footnote 50 While such a persistent assertion of uniqueness by the CJEU may appear grating for its perceived defiance over general international law, it arguably follows the international law understanding of the autonomy of regional IOs. In this regard, d’Aspremont’s reference to the “impermeability” of organizations to “external … interferences”Footnote 51 under international law is echoed by Christina Eckes, who observes, in the EU context, that the CJEU has sought to “shut external interference[s] out” of the Union legal order.Footnote 52 Indeed, the Court further developed its understanding of the Union’s systemic uniqueness by safeguarding the jurisdictional integrity of the EU legal order. In particular, in Opinion 1/00Footnote 53 and, more controversially, Opinion 2/13Footnote 54 and Kadi,Footnote 55 it established that “foreign [that is, external] norms” may be incorporated into the EU legal order only if they comply with “the fundamental values on which the [EU] is founded.”Footnote 56
As understood by international law, institutional independence is an essential feature of organizations with a regional focus, for their “raison d’être lies in the creation of a separate regime from general international law.”Footnote 57 This can be distinguished from organizations such as the World Health Organization and the International Labor Organization. EU discourse seems to conceive of this aspect of autonomy, focusing on IOs’ institutional independence from other international entities, as its “external dimension.”Footnote 58 By contrast, equivalence regarding IOs’ political independence and the “internal dimension”Footnote 59 of autonomy in the EU sense seems less direct—they both broadly concern internal arrangements in accordance with which member states envisage the operation of their respective organizations.
As international, and particularly regional, organizations typically claim their ability to determine and pursue their own political priorities (within the framework determined by their members), it follows that “[t]he defence of [the institutional] dimension of autonomy has always been closely associated with the defence of the political project of the [IO] concerned.”Footnote 60 The CJEU’s defense thereof has been expressed in terms of guaranteeing coherence and self-sufficiency: the Union’s institutional, structural independence grants it “the capacity to operate as a self-sufficient system of [substantive] norms.”Footnote 61 This is only strengthened by the Court’s assertion of the conceptual autonomy of EU law in CILFIT, discussed in the previous section. This amounts to a robust safeguarding of the institutional independence of the EU, so much so that the latter may even be elevated to the level of “jurisdictional sovereignty.”Footnote 62 Still, it may be explained by reference to the EU’s “political project”—or, in more constitutional terms, the Union’s objectives under the Treaties.Footnote 63
In reviewing the assessment of the EU conception of autonomy in light of its understanding under international law, the two seem significantly equivalent in conceptual terms. Importantly, the distinction between the two dimensions of the autonomy of IOs should not be overstated. Institutional independence remains “inevitably intertwined” with political independence: institutional independence means that an IO is able to “defend” its values, principles, and interests from external “interferences,”Footnote 64 but within the limits of the political independence afforded by its members. Furthermore, these principles and interests have been politically determined in accordance with the arrangement and processes established by the organization’s members. The relationship between the two dimensions of autonomy seems to be one of complementarity, particularly in IOs such as the EU, which are more institutionally developed.
C. Principle or Notion? The Normative Status of Autonomy under EU Law and International Law
Notwithstanding the conceptual similarities, the normative status of autonomy seems to differ in EU and international law contexts. First, this section will examine whether autonomy may constitute a principle of international law. Though this does not seem to be the case, the importance of autonomy as an inherent characteristic of IOs is not diminished. Second, the important function of autonomy as a constitutional principle within the EU legal order will be demonstrated. In this light, though it may not (yet) be classified as a general principle of international law, autonomy can justify not only the creation of politically balanced and structurally distinct legal orders by IOs but also of self-contained systems of norms.Footnote 65 In the latter instance, an IO (in accordance with its attributed objectives, powers, and institutions) may well treat it as a general principle within its own internal legal order. The international law understanding of autonomy would then emanate as a principle of the IO’s internal-domestic law and be treated as such by its institutions and members. It is argued that the European integration process, culminating in the formation of the European Union in its current form, is an illustration of this. Such an account would demystify the CJEU’s autonomy-related case law, as the Court would be seen to largely follow trends that already existed within the international legal order. Reference to such case law will again be selective in order to illustrate particular points.
I. Autonomy as a Principle of International Law?
The analysis carried out in the previous section gives rise to a complicated understanding of international organizations’ autonomy conceived as political and institutional independence, whereby both aspects, in principle, co-exist in varying degrees. On this basis, it shall be examined whether autonomy may be a fully-fledged, if underutilized, principle of law rather than merely a feature of IOs.
Jan Klabbers previously acknowledged the question regarding the normative status of autonomy, if only tangentially. However, he did so only with regard to international organizations’ law in particular, though it was said that “there is no such thing as a separate corpus of rules and principle to be referred to as international organisations law” in the first place.Footnote 66 This is important, as an examination of the status of autonomy as a legal concept requires a clear delineation of the area of law within which it may operate as a principle. Notwithstanding the existence or otherwise of a distinct discipline of international organizations law, interactions between IOs are governed by international law as broadly understood.Footnote 67 Hence, autonomy could arguably only be a principle of general international law, as it would accord with the fundamental conclusion of the ICJ that IOs are subjects of an overarching system of international law.Footnote 68
A finding that autonomy constitutes a principle of international law is not unavoidable, even if, as stated earlier, the ICJ has indicated that it is an attribute with which all international organizations are, to an extent, endowed by their member states.Footnote 69 Even so, this does not suffice to justify the existence of a legal principle. Takis Tridimas’s broadly accepted formulation states that “a principle is a general proposition of law of some importance from which concrete rules derive.”Footnote 70 It must be “general”; it must “underlie the legal system as a whole” and be widely recognized by key judicial-institutional, political, and social constituents.Footnote 71 A principle must also “carry added weight”Footnote 72 and bear “wider independent normative meaning.”Footnote 73 Finally, on a conceptual, doctrinal, and functional level, principles “justify and underpin the specificity and detail of rules, both procedural and substantive.”Footnote 74
Violeta Moreno-Lax and Katja S. Ziegler do not believe that autonomy has passed the threshold of being such a general principle of international law, as they consider it to lack normative substance: autonomy is merely “an allusion to an empirical fact or a collective reference to the more defined specific rules” that underlie it.Footnote 75 Moreover, as autonomy is not relevant in purely national contexts, it may not be said to derive from domestic legal orders. The International Law Commission Draft Conclusions on General Principles suggested that general principles “formed within the international legal system” must be “widely acknowledged in treaties and other international instruments,” “underlie general rules” of customary international law, or be “inherent in the basic features and fundamental requirements of the international legal system.”Footnote 76 However, autonomy is not explicitly stated in international agreements, nor does it yet have a legal function that is generally recognized by the international community. Moreover, in the absence of a well-established international consensus, it is difficult to argue that autonomy is a general principle of international law due to it being inherent in the basic features of the international legal system; this has typically referred to principles such as consent to jurisdiction and uti possidetis juris, quasi-customary principles with a history of being invoked before and relied upon by international courts.Footnote 77 Indeed, autonomy has not been central to international judicial reasoning, with the exception of the CJEU.
Lastly, one seems inclined to agree with Moreno-Lax and Ziegler’s assessment that no single normative claim can underpin a principle of autonomy under international law. The rich diversity of IOs, which differ dramatically in terms of organization, foundational objectives, modes of operation, and membership, is a main factor hindering the doctrinal maturity of the concept. In this sense, one could be tempted to merely view autonomy as an “idea,”Footnote 78 as an interpretative scheme that helps to conceptualize the operation of IOs. However, even if disregarded as “a kind of pseudo-sovereignty of international organisations,” autonomy retains considerable importance precisely because it may be conceived as a “structural basis” of public international law.Footnote 79 This would be particularly valuable in explaining (or justifying) the international capacity and limits of the IOs’ actions. Paradoxically, the notion of IOs’ autonomy may be useful in illustrating their dependence on the international legal order. By virtue of their sovereignty, states do not need the legal-conceptual support that autonomy provides in order to (inter)act within the international community on such a basic, structural level.
Therefore, autonomy can be recognized as a legal notion under international law that concretely characterizes all international organizations. It is not a mere “idea” but an inherent feature of IOs as subjects and products of international law. As Klabbers rightly suggested, it may not be necessary to classify it as a principle of international law in order to recognize its importance: autonomy “need not be invoked as principle … in order to materialize.”Footnote 80 Autonomy maintains an essential role in international law as an inherent feature in the establishment and operation of IOs. Though it may not correspond to a single normative claim per se and is rarely relied on by international courts, it has a profound structural importance in explaining IOs’ distinctness from its member states. However, importantly, it also has a relevance in facilitating substantive policies pursued by IOs, which may develop organically in accordance with the particular characteristics of each organization.
This flexibility, encompassing the structural and the substantive, accounts for the ability of different organizations to rely, in varying degrees, on autonomy’s political or institutional dimensions. Though all international organizations are concerned with the exercise of their attributed powers with respect to the framework established by their member states, some organizations seek to build on this externally and even claim a systemic distinctiveness for their own legal order vis-à-vis general international law. In other words, autonomy has a clear structural dimension, as it supports a political balance between an IO and its members and facilitates IOs’ development of their own legal order. At the same time, autonomy can also have a substantive normative dimension, depending on the IO it characterizes: it can cultivate the substantive, rather than structural, normative features underpinning the organization’s formally distinct legal order. This seems, in principle, more prevalent in regional integration organizations.
II. The CJEU as Pygmalion and Autonomy as a (General) Principle of EU Law
Notwithstanding the close conceptual proximity of the international law understanding of the autonomy of international organizations and the CJEU’s understanding of the autonomy of the EU legal order, one must note that the latter seems to operate as a principle of law, thus having a more robust role in an EU rather than in an international context. By contrast, there does not seem to be a formal principle of autonomy under international law, given that it is not recognized as such by international courts, notwithstanding the notion’s considerable importance as an inherent attribute of all international organizations.
The Court has evolved the way in which it invokes autonomy. As seen, for instance, in Van Gend,Footnote 81 Costa,Footnote 82 Commission v Belgium,Footnote 83 and even the ERTA judgment,Footnote 84 the Court avoided explicit references to autonomy, which was treated more as a notion underpinning the development of the principles of direct effect and the primacy of EU law, and the implied external powers doctrine.Footnote 85 This matched international courts’ typical treatment of autonomy as an underlying rationale for the development of the principle of international legal personalityFootnote 86 rather than as a formal principle in itself. Following Opinion 1/91,Footnote 87 however, the CJEU revisited autonomy and began its judicial invocation qua principle. It is interesting to see that Theodor Schilling’s 1996 observation that autonomy constitutes “[t]he single most far-reaching, and probably most disputed, principle” of EU law still seems apt today, more than twenty-five years later.Footnote 88
This raises the question of the normative classification of autonomy. Marise Cremona classified autonomy as a structural principle alongside, inter alia, effectiveness, proportionality, sincere cooperation, and institutional balance; structural principles are argued to be “a type of general principle.”Footnote 89 Autonomy is structural “in the sense of defining and being inherent to the deep structure of the EU.”Footnote 90 Further, it is a “systemic” structural principle, together with effectiveness and coherence, as they “characterise the type of international actor the EU is, and the norms it produces in its external policymaking.”Footnote 91 Indeed, contrary to autonomy’s status under international law, the notion has been elevated to a constitutional principleFootnote 92 in the EU context:
[A]utonomy, which exists with respect both to the law of the Member States and to international law, stems from the essential characteristics of the European Union and its law [which refers to the principles of direct effect and primacy of EU law]. Those characteristics have given rise to a structured network of principles, rules and mutually interdependent legal relations binding the European Union and its Member States reciprocally as well as binding its Member States to each other.
That autonomy accordingly resides in the fact that the Union possesses a constitutional framework that is unique to it [which] encompasses the founding values set out in Article 2 TEU [including respect for the rule of law and human rights], … the general principles of EU law, the provisions of the Charter, and the provisions of the EU and FEU Treaties, which include, inter alia, rules on the conferral and division of powers, rules governing how the EU institutions and its judicial system are to operate, and fundamental rules in specific areas, structured in such a way as to contribute to the implementation of the process of integration described in the second paragraph of Article 1 TEU.Footnote 93
While criticism of the CJEU over the perceived tension between the robust interpretation of autonomy in Opinion 2/13Footnote 94 and the EU’s commitment to human rights under Articles 2, 6, and 21 TEU is understandable,Footnote 95 one should not be too quick to dismiss autonomy as a non-principle within the EU legal order. As a matter of EU law, autonomy does not merely represent a value-less “ontological reality.”Footnote 96 Instead, its relevance in case lawFootnote 97 regarding the rule of law and fundamental rights indicates that it has both an axiological and an ontological dimension.Footnote 98 The Court’s above elucidation of autonomy in Opinion 1/17, as referenced in Komstroy, suggests that it may “entail a claim of a wide or general normative content.”Footnote 99 Namely, the Court has firmly intertwined the traditional role of the autonomy principle in safeguarding the jurisdictional integrity of the EU legal order with the Union’s commitment to judicial protection and Article 2 TEU values, which include respect for the rule of law and human rights. The relationship between autonomy, judicial protection, and the rule of law will be revisited in the next section.
The way in which the CJEU understands autonomy suggests that it satisfies the definition of a legal principle:Footnote 100 it is a legal norm that provides “justification for concrete rules”Footnote 101 and is typically used to review the legality or constitutional compatibility of a measure or agreement. All general principles, including autonomy, are of a “fundamental nature,” meaning that they “inform and guide the interpretation of rules even where those rules have not been adopted to give them specific expression.”Footnote 102 Indeed, the significant evolution that autonomy has undergone since Van Gend and Opinion 1/91 is to be expected: principles “are open to adjustment and re-interpretation over time and in changing circumstances.”Footnote 103
However, the CJEU has intertwined autonomy with other EU law principles and doctrines in a way that almost makes one contingent upon the other. Autonomy “stems” from the principles of direct effect and primacy,Footnote 104 though, at the same time, it served as the underlying rationale for their development. It “resides in” the fundamental normative commitments of the Union – that is, Article 2 TEU values, general principles of EU law, and fundamental rightsFootnote 105 – but it is also used to safeguard the judicial-institutional framework that gives effect to them. Further, Achmea showed that autonomy also relates to principles such as mutual trust.Footnote 106 Lastly, it is connected to the very objective of European integration under Article 1 TEU, the creation of an “ever closer union.” In this sense, autonomy constitutes a general general principle, a foundational principle of EU law that both stems from and justifies the constitutional and institutional uniqueness of the EU as a systemically distinct form of international organization.
Nevertheless, the way in which autonomy has evolved into a legal principle and become integrated within the EU framework cannot eclipse the notion’s conceptual origins in international law. Autonomy’s connection to the EU’s normative foundations still appears as a judicial elaboration based on the political and institutional independence that inherently characterizes international organizations, adapted to the European context. The CJEU’s inspiration from international law cannot be obscured by its bold development of the concept, unlike other international courts, into a benchmark of constitutionality. As Amedeo Arena remarked in a different context, the Court “acted more as a Pygmalion than as a Demiurge: it did not create the principle … ex nihilo, but brought to life a [notion] already established in international law.”Footnote 107 The EU principle of autonomy is not a Union invention per se but a Pygmalian judicial creation, clearly drawing from and building on an international law notion. It will be left to readers to decide whether the myth of Pygmalion,Footnote 108 who eventually fell in love with his own creation, is fully analogous to the CJEU’s development of autonomy.
D. Autonomy Mutatis Mutandis and the (International) Rule of Law
The CJEU developed an understanding of the autonomy of the EU legal order which seems to encapsulate both the “political” and “institutional” dimensions of the notion as conceived by international law, and has transposed it into a fully-fledged constitutional principle. Arguably, this is owed to the process of constitutionalizing, whereby the EU’s institutional structure developed concurrently with its emphasis on judicial protection, the rule of law, and individual rights; the one informed, shaped, and enhanced the other. Though constitutionalization is not, in principle, limited to the EU context,Footnote 109 no other IO has embarked on the path inaugurated by Van Gend. European constitutionalization has imbued the notion of autonomy with a broader normative aspiration going beyond a merely structural – that is, substantively agnostic – purpose relating to the systemic functionality of the EU legal order. In doing so, the CJEU made full use of the notion’s conceptual scope, which already existed under international law; it employed autonomy’s capacity to facilitate the creation of a distinct system of norms by asserting the EU’s political and institutional independence.
More specifically, the adaptation by the Court of Justice of the notion of autonomy into a constitutional principle of EU law can be explained by reference to the “broader normative commitment [of the international community] to the realization of the rule of law.”Footnote 110 International organizations were hoped to be able to constrain the arbitrary exercise of their members’ sovereign will and represented a new way of taming international law’s blunt positivism.Footnote 111 The autonomy of IOs, therefore, can be seen as an essential element of the international rule of law, the pursuit of which may be “the single most important goal of the international system, upon which all other goals – peace, prosperity, and effective international cooperation – depend.”Footnote 112 This points to an underlying relationship between autonomy and the rule of law and suggests a positive correlation between the autonomy of an IO and the rule of law arrangements they establish, including their judicial-institutional framework and dispute settlement mechanisms.
As will be seen, the degree of an organization’s autonomy typically corresponds to the robustness of its dispute settlement structures, as these introduce and regulate limits to the exercise of members’ separate will. More fundamentally, an organization’s dispute settlement mechanisms seem to reflect its intended objectives. An IO’s objectives (“common goals”) have also been closely related to its endowed degree of autonomy by the ICJ.Footnote 113 IOs generally tend to operate either as an agora that hosts political debate and “perhaps” fosters decisionsFootnote 114 or as an independent, institutionalized entity acting collectively across policy areas. However, they may well partly draw from both “competing images.”Footnote 115 By way of illustration, the United Nations and the Council of Europe will be briefly considered in these terms to explain why the conceptual borrowing of the EU conception of autonomy from its international law origins was coupled with its unique development into a concrete legal principle.
Closer to the agora concept of IOs, the United Nations was founded after World War II, inter alia, “[t]o maintain international peace and security, and to that end … to bring about … settlement of international disputes or situations which might lead to a breach of the peace”; “[t]o develop friendly relations among nations”; and “[t]o achieve international co-operation.”Footnote 116 The ICJ, the judicial organ of the UN, is tasked with a significant role in adjudicating disputes between members and issuing advisory opinions, but its jurisdiction is based on the parties’ consent.Footnote 117 If a party fails to comply with a judgment, the other party “may have recourse to the Security Council,” which may “make recommendations or decide upon measures to be taken to give effect” to it.Footnote 118 This raises the possibility of a de facto veto to ICJ’s decisions by permanent Security Council membersFootnote 119 which would fall within a broader debate regarding the reviewability of Security Council decisions.Footnote 120
The Council of Europe (CoE), another international organization, albeit with a regional reach, was itself founded with an aim “to achieve a greater unity between its members … by discussion of questions of common concern and by agreements and common action … and in the maintenance and further realisation of human rights and fundamental freedoms.”Footnote 121 It was envisaged to be complementary to the UN.Footnote 122 The European Court of Human Rights (ECtHR) was established within the CoE framework in order to “ensure the observance of the engagements undertaken by” the parties to the European Convention on Human Rights (ECHR).Footnote 123 Though members’ national judiciaries have an important role in the application of the ECHR,Footnote 124 and the ECHR has had a profound influence on the development of human rights protection within the EU,Footnote 125 the ECtHR is a typical international court. Its jurisdiction only extends to the interpretation and application of the ECHRFootnote 126 and is subject to the parties’ exhaustion of domestic remedies.Footnote 127 Lastly, there are no hard mechanisms for enforcing ECtHR judgments,Footnote 128 which limits the breadth and depth of the CoE’s autonomy.
Important differences between the UN, the CoE, and the EU can be identified in this light. Though the EU also serves an agora function, as a whole, it is situated near the other end of the spectrum as a deeply institutionalized (indeed, constitutionalized) entity. The EU legal order is systemically distinct from but closely connected with the national legal orders of member states, not least by virtue of the principle of primacy.Footnote 129 Moreover, the system of preliminary reference procedure, the “keystone” of the EU judicial system, “set[s] up a dialogue” between the CJEU and national courts, by which the latter are tasked with applying EU law on the basis of the former’s interpretation.Footnote 130 This transforms national courts into EU courts.Footnote 131 Article 19 TEU “established that the EU is bound by the rule of law and recognizes the separation of powers”;Footnote 132 it also introduced an obligation for member states to ensure effective judicial protection in areas of EU law. The Union judicial system is further shaped by the Court’s emphasis on individual rightsFootnote 133 and by the operation of significant enforcement mechanisms.Footnote 134 The EU system thus emulates characteristics usually found in domestic rather than international legal orders, which it accommodates within a supranational framework. This is not surprising, given that the EU was conceived as merely a “stage in the process of European integration,”Footnote 135 and the “process of creating an ever closer union”Footnote 136 whose objectives, set out in Article 3 TEU, cover a wide range of areas, including, crucially, common external action.
Therefore, notwithstanding the CJEU’s own role in the assimilation and development of the notion of autonomy in the EU context, the autonomy of the EU legal order is a natural by-product of the Union’s nature as a highly institutionalized entity, quite removed from other agora-type organizations. After all, the EU’s development of a robust judicial system, premised upon a limitation of member states’ sovereignty within specified fields,Footnote 137 was a matter of practicality as well as values. It was imperative, if its objectives were to be achieved, for its institutional-judicial arrangements to ensure the uniformity and effectiveness of EU law.Footnote 138 If “[i]nternational law’s … institutional failings [stemming from a lack of autonomy], particularly when compared to national legal systems, are … a source of constant apologia in international legal scholarship,”Footnote 139 it seems appropriate that the international organization with the most developed sense of autonomy is the one that is most integrated, supranationally, within the legal systems of its member states.
Such an understanding casts the differing normative status of autonomy under EU and international law in less dramatic terms. The autonomy of the EU legal order can be said to be mutatis mutandis equivalent to the international law understanding of autonomy. Their conceptual proximity and differing status – as a principle and a notion, respectively – may simply reflect the objectives of the organizations they characterize. Though not necessarily denoting a causal relationship, the presence of elaborate judicial-institutional arrangements and effective dispute settlement mechanisms suggests that an organization’s objectives justify, and even require, a robust understanding of autonomy, as derived from the notion’s international law origins.
E. Conclusion
Tracing the conceptual origins of autonomy by reference to international law perhaps qualifies its reputation as the “probably most disputed … principle” of EU law,Footnote 140 notwithstanding the controversial policy outcomes it has facilitated lately.Footnote 141 Autonomy seems to be an inherent attribute of all international organizations rather than a European peculiarity. All IOs seek, to some extent, to safeguard their political independence in order to exercise their will separately from the will of their member states, in accordance with the powers the latter have attributed to it. Thus understood, autonomy is the consequence of an organization’s international legal personality. Organizations that aspire to have a systemically distinct role within the international community are also characterized by a degree of institutional independence, which differentiates their own internal legal order from general international law. On a conceptual level, autonomy under international law can thus facilitate the creation of independent systems of norms.
Still, autonomy does not yet seem to constitute a principle of international law, as it lacks explicit recognition by the international community as a whole and by international courts in particular. However, the CJEU does seem to have developed a principle of autonomy of the EU legal order. Rather than viewing this as an EU construct per se, it should be seen as an instance of the Court drawing Pygmalian inspiration from notions of international law and giving them new life within the Union framework. This should help to demystify autonomy, as it would indicate that this process may not have been motivated by “fear”Footnote 142 or “selfishness.”Footnote 143 A robust understanding of autonomy, as a legal principle firmly integrated into the institutional and normative foundations of the EU, appears as a natural by-product of the member states’ own commitment to an “ever closer union” with global aspirations. The Court’s use of autonomy both as an independent benchmark of constitutionality and as a normative rationale underpinning the broader constitutional edifice of the EU (which includes primacy, the preliminary reference procedure and mutual trust) would suggest that autonomy has become a general principle of EU law.
This seems to be in congruence with the notion’s international law origins. Organizations like the United Nations and the Council of Europe are characterized by less developed senses of autonomy. This is appropriate as they primarily seek to operate as agorae-platforms rather than deeply institutionalized, much less constitutionalized, entities. Though contributing to the international rule of law in important ways, their respective institutional and judicial mechanisms confirm that their autonomy is inherently limited; this corresponds to a limited ability to develop and assert their own volonté distincte over the sovereign will of their members. On the other hand, the way in which the EU legal order has matured institutionally and judicially demonstrates greater commitment to the original objective of the international community that IOs may constrain state arbitrariness. For the EU to do that while also having a global presence, a well-developed network of judicial and institutional arrangements safeguarding its autonomy was required. This suggests a positive correlation between the robustness of the dispute settlement mechanisms established by an IO and its autonomy.
As autonomy under international law characterizes organizations that vary significantly in form, membership, and purpose, the notion demonstrates remarkable conceptual flexibility. The CJEU has adapted it into a constitutional principle in keeping with its nature as a supranational entity and has evolved with it in keeping with the various stages of European integration. The more recent integration of substantive considerations relating to judicial protection and Article 2 TEU values in the EU principle of autonomy, seen since Achmea and Opinion 1/17, is thus more easily understood. In this sense, it seems not only constitutionally acceptable in domestic terms but also justified by international law—in so far as this would address challenges posed by increasing EU global action in the process of creating an ever-closer union according to its attributed competences and objectives.
Competing Interests
The author declares no competing interests exist.
Funding Statement
The author received no specific funding for the study.