Constitutional Courts Asking Questions: A Deliberative Potential of Preliminary Reference Mechanism

Abstract

This Article interrogates the role of national constitutional courts within the Article 267 TFEU preliminary reference mechanism from both descriptive and normative angles. First, I demonstrate that although a clear majority of the constitutional courts submit references to the ECJ on a more frequent basis, differences in individual approaches remain significant. Subsequently, I argue that the core normative attractivity of the questions submitted in the course of domestic constitutional review lies in their participative and deliberative potential. Compared to ordinary courts, constitutional courts are better suited to amplify the ‘unheard’ voices of immobile EU citizens. By counterbalancing the demands of the EU's functional constitution, which is primarily based on the ideals of market capitalism, constitutional courts’ questions may contribute to the EU's capacity to generate legitimate decisions. Finally, I put my theoretical claims in context and analyse the main ways in which such deliberative potential can translate into practice.

I. Introduction: National Constitutional Courts Lost in The Process of European Integration?

The process of European integration through law has put the Member States’ constitutional courts¹ under a lot of pressure.² Ordinary courts took up their new role of ‘European Union courts’³ rather swiftly as multiple fuzzy doctrines of the European Court of Justice—primacy principle, doctrine of direct effect, and CILFIT, to name just a few⁴—empowered them to effectively disregard any piece of domestic legislation they deemed incompatible with EU norms.⁵ Consequently, under the rule of EU law, all domestic judges could perform a specific kind of diffuse judicial review of national legislative outputs even though such task had traditionally been in the hands of a small number of constitutional justices.⁶ This naturally put the overwhelmingly privileged and entrusted institutions in an odd position.⁷

The marginalising effects of the diffuse judicial review were strengthened in 2009 when the Charter of Fundamental Rights of the European Union (‘Charter’) entered into force.⁸ Since then, constitutional courts (‘CCs’) need to cope with the fact that they have to share their once-mastered agenda of human rights not only with the European Court of Human Rights in Strasbourg but also with the Court in Luxembourg.⁹ As Michal Bobek aptly puts it: constitutional courts have always been rather special creatures, and by the process of accession to the EU, such creatures were caged as some parts of their former habitat were declared out of bounds.¹⁰

How did constitutional courts react to these unfavourable institutional dynamics? Gradually, they developed several strategies in order to claim their sphere of influence over the development of EU law as well as to regain their monopoly in reviewing national legislation. Three tactics stand out in particular.

First, although initially hesitant,¹¹ most of the constitutional courts have by now started to use EU law as a yardstick for constitutional review, be it directly or indirectly.¹² The former approach is well illustrated by the Austrian Verfassungsgerichtshof, which held that the constitutionality of national legal acts might be assessed directly from the perspective of rights guaranteed by the Charter.¹³ We can trace down similar stances in the jurisprudence of Italian,¹⁴ Belgian,¹⁵ and German¹⁶ constitutional courts. As to the latter—indirect—approach, one can point out the case law of the Spanish Tribunal Constitucional. Even though it formally claims that EU law lacks constitutional status,¹⁷ it has also confirmed that a violation of EU law may, under certain circumstances, amount to a violation of the rights guaranteed by the Spanish Constitution. In that vein, a failure of a ‘court of the last instance’ to refer a preliminary reference to the European Court of Justice (‘ECJ’), or to grand direct effect to the provision of an EU directive, can result in a violation of the right to effective judicial protection according to Article 24 of the Spanish Constitution.¹⁸ Such an indirect inclusion of EU norms into the standard for constitutional review can also be found in the jurisprudence of French Conseil constitutionnel,¹⁹ and Czech,²⁰ Slovenian,²¹ Hungarian,²² Bulgarian,²³ or Slovak²⁴ constitutional courts. Putting contextual specifics aside, it is safe to claim that most of the CCs are at least sometimes willing to look at national legal acts through the EU lenses.

Turning to the second kind of anti–marginalisation tactics, most of the constitutional courts have over time also developed several defensive doctrines which pose limits on the primacy principle of EU law within national legal orders. These concepts include ‘higher level of fundamental rights’ review (Solange), ultra vires review, controlimiti doctrine, and national/constitutional identity review.²⁵ Even though they differ in some important aspects, they all share the same theoretical premise: the transfer of powers to the EU is limited, because domestic constitutions do not allow the Member States to surrender their sovereignty to the EU altogether.²⁶ Some have argued that national constitutional courts should cautiously use these concepts as substantive tools to counterweigh the EU's democracy-diminishing mechanisms.²⁷ Others have alerted against them by highlighting that their flawed potential may lead to dangerous outcomes.²⁸

In any case, what once used to be the subject of rather theoretical debates has now become an acute concern for real-life European politics. Indeed, everyone who takes an interest in the EU context has read about cases in which national constitutional courts invoked the ultra vires and constitutional identity cards—Czech Landtova saga,²⁹ German PSPP,³⁰ and Polish K 3/21³¹ need no elaborate introduction here. There is, however, one particular moment in the discussion about these major cases that deserves further attention.

Unlike their German counterparts, Czech and Polish constitutional judges were heavily criticised for the fact that they had not given the ECJ chance to rule on the validity of EU acts before proclaiming them unconstitutional.³² In particular, the unwillingness of the Polish Constitutional Tribunal to engage in the preliminary reference mechanism according to Article 267 of the Treaty on the Functioning of the European Union (‘TFEU’)³³ was depicted as proof of its malign intentions to undermine the EU's authority.³⁴ Even though such insights are based on the rather contextual ‘backlash’ and ‘rule of law crises’ narratives, they do bridge the discussion towards the third larger tactic most of the constitutional courts have employed by now—using preliminary references as a procedural tool to re-centralise their position within the EU judicial system.

Now, what do we do and do not know about the national constitutional courts’ involvement in the preliminary reference procedure? On the one hand, previous research has mainly focused on the approaches of CCs from the doctrinal perspective.³⁵ Even though studies of individual cases provided us with valuable contextual insights, a more comprehensive picture of how often and what type of questions constitutional courts submit as a group is still missing (a descriptive gap). On the other hand, the existing literature has so far depicted the preliminary reference mechanism in the context of constitutional courts predominantly as an appropriate tool to solve constitutional conflicts between the EU and Member States.³⁶ As much as this account is attractive in the current context of the EU's rule of law crisis, it fails to expound the value of constitutional courts’ involvement in the formal dialogue with the ECJ outside those rather extreme instances of constitutional clashes (a normative gap).

This Article aims to address some of these blind spots in the literature on national constitutional courts’ place within the EU and examine the third anti-marginalisation tactic—submission of preliminary references to the ECJ—in more detail. Thus, it interrogates the role of national CCs within the Article 267 TFEU mechanism from both descriptive and normative angles. Drawing on the analysis of 122 preliminary references posed between the years 1958 and 2022, I first demonstrate that a majority of the constitutional courts have started to send questions to the ECJ on a more frequent basis. Nonetheless, the differences in how individual constitutional courts make use of the mechanism remain significant. In the second part, I explore whether such development can be regarded as normatively attractive. Here, I contend that due to their specific deliberative design, constitutional courts have the capacity to represent a unique part of communicative arrangements within the EU. Lastly, I put my theoretical claims in context and analyse the main ways in which such potential can translate into practice.

II. Constitutional Courts Asking Questions

It is a well-known fact that at first, constitutional courts were quite hesitant to ask the European Court of Justice any questions whatsoever. Until the year of 2009, they submitted only 29 references altogether.³⁷ This was traditionally explained by two sets of reasons: legal and behavioural.³⁸ As to the former, some of the CCs first refused to identify themselves as courts or tribunals in the meaning of Article 267 TFEU as well as they refused to use EU law as a standard for constitutional review.³⁹ As a result, those courts had no opportunities to apply EU law in their day-to-day adjudication.⁴⁰ As to the latter reasons, the initial reluctance to engage with the ECJ has been explained predominantly on the basis of ‘judicial ego’ and ‘jealousy’ of national constitutional judges who did not want to become subjects of the ECJ's authority and on the contrary attempted to maintain their position of ‘highest courts’ in the country.⁴¹

However, the initial hesitatnt position has changed considerably over the years. The total number of references submitted by constitutional courts every year (Figure 1⁴²) suggests that constitutional courts reconsidered their negative stance as they referred multiple times more references in the following decade. Indeed, the increasing number of constitutional court referrals is not a mechanical byproduct of the fact that more EU Member States (and hence constitutional courts) joined the EU over time. The increasing trend is evident even when considering the annual referral rate, i.e, the total number of referrals from constitutional courts relative to the number of active constitutional courts across the EU (Figure 2⁴³).

Figure 1. The figure depicts the total number of references submitted by constitutional courts per year in time.

Figure 2. The figure depicts the annual referral rate, i.e, the total number of referrals from constitutional courts relative to the number of active constitutional courts across the EU.

In total, CCs sent 122 preliminary questions between the years 1958 and 2022.⁴⁴ Moreover, only 4 out of 22 analysed constitutional courts have not posed any preliminary reference whatsoever—these include constitutional courts of Bulgaria, the Czech Republic, Croatia, and Hungary (see Table 1⁴⁵).

Table 1.

It is clear that with their 45 and 40 preliminary references, the Irish Supreme Court and the Belgian Cour constitutionelle are without any doubt the most proactive CCs when it comes to both the total number of questions posed as well as the annual referral rate. The Irish court submitted its first question concerning the exemptions from the compulsory acquisition of rural land in 1983.⁴⁶ The Belgian one in 1997.⁴⁷ The latter case concerned the conditions for specific training in general medical practice. The Belgian CC simply asked the ECJ to interpret the provisions of the respective directive and whether they should be read as requiring persons who are about to undertake such practice to first obtain a formal diploma in medical practice.⁴⁸ It is telling that neither the ECJ, nor the AGs seemed to think that these first references should be seen as a significant moment for the EU's constitutional setting.⁴⁹ Since then, both constitutional courts have made great use of the preliminary reference procedure and referred questions concerning both the interpretation and the validity of EU acts.⁵⁰ It was probably the Pringle case in which the ECJ dealt with the question of unconstitutional constitutional amendment of the Treaties and which has attracted the most academic interest.⁵¹

The Latvian Satversmes tiesa, the Italian Corte costituzionale, the Austrian Verfassungsgerichtshof, and the Cyprian Anotato Dikastirio belong—with their six and four preliminary references—to the group of ‘rather active’ constitutional courts.

The Austrian CC referred its first question already in 1999.⁵² Due to its friendly approach to the preliminary reference mechanism and the use of the Charter in general, it has been suggested that in the future, the Austrian CC might act as sort of a ‘gatekeeper’ or a ‘privileged partner’ of the ECJ. Particularly, Austrian constitutional judges were expected to first filter domestic cases relating to fundamental rights, clarify, and prepare them for the ECJ, and then adapt the preliminary rulings for the peculiarities of the domestic legal order in the follow-up cases.⁵³ Nevertheless, the Austrian Verfassungsgerichtshof has not sent any reference to Luxembourg since 2012. Indeed, the fairly low annual referral rate (0, 148) do suggest that the Austrian CC might not be excessively active when compared to other CCs.

Unlike its Austrian counterpart, the Italian Constitutional Court had denied for a long time its status of a ’court’ or a ‘tribunal’ in the meaning of (current) Article 267 TFEU.⁵⁴ Its position has, however, changed significantly over time as it started to accept EU law as a standard for abstract constitutional review. Consequently, it referred its first preliminary reference in 2008.⁵⁵ One can identify the case MAS and MB, which concerned the limitation periods in regard to VAT frauds, as the most significant preliminary reference.⁵⁶ In that case, the Italian CC re-sent the question after the ECJ had already issued one preliminary ruling regarding that matter based on the preliminary reference of an ordinary Italian court (the so-called Taricco saga).⁵⁷ The Italian CC provided the ECJ with the necessary constitutional context (mostly highlighting the fundamental principle which requires that rules of criminal law are precisely determined and are not retroactive) and provided Luxembourg judges with an opportunity to reconsider their initial position. Such approach has been praised as a positive example of constitutional dialogue within the EU.⁵⁸

Finally, a slightly different development can be traced in Latvia. The Latvian CC started to actively supply the Luxembourg Court with the questions only quite recently – it posted its first reference in 2017⁵⁹ and then five other references in 2020. Indeed, when considering the annual referral rate, the Latvian CC is the third most active constitutional court. Such an unparalleled activity has, however, remained without a further response from the side of scholarship so far. Similarly, academia seems not to pay great attention to the questions posed by the Cyprian Supreme Court which submitted its first reference in 2009.⁶⁰

Now, as it is evident from Table 1, the remaining constitutional courts, which have made at least some use of the Article 267 mechanism, have submitted not more than one or two references, and their annual referral rates are fairly low. Yet, academia has paid enormous attention especially to the questions sent by the German Bundesverfassungsgericht. Such interest is probably caused by the fact that in cases Gauweiler and Others ⁶¹ and Weiss and Others ⁶² (which both concerned the competencies of the European Central Bank and its economic and monetary policies), German constitutional judges ‘played hardball’ and raised ultra vires and constitutional identity arguments while questioning the validity of Union legal acts for their incompatibility with the TFEU and German Grundgesetz.⁶³ Similarly, the literature has closely covered the preliminary reference in the case of Melloni, in which the Spanish Tribunal Constitucional questioned the validity of the execution of European Arrest Warrant with regard to judgments issued in absentia.⁶⁴ It has been claimed that the Spanish CC showed great effort by elaborating its interpretations of the right to a fair trial and of Article 53 of the Charter from the perspective of the Spanish Constitution on the one hand, and by expressing respect for the ECJ's authority and the autonomy of EU fundamental rights regime on the other. The ECJ's preliminary ruling, however, has been criticised for an acute lack of responsivity.⁶⁵ The remaining preliminary questions of other CCs have clearly not enjoyed the same level of attention of existing scholarship.

Now, what do these descriptive insights tell us about the constitutional courts’ involvement in the preliminary reference mechanism? One can highlight at least a couple of broader points. The overview of submitted questions seems to confirm the expectations of the proponents of the so-called ‘second’ emancipation theory. According to them, the initial phase of ‘judicial empowerment’ of lower courts needs to be perceived merely as a self-eroding Act I because national high courts are now beginning to engage in the formal dialogue with the ECJ on a more frequent basis in order to reassert their control over national judicial hierarchies and to substantively influence the development EU law.⁶⁶ Indeed, the statistics show that the constitutional courts that have not yet referred any preliminary questions to the ECJ now represent a clear minority. Thus, from an overall perspective, the upward trend is clear—national constitutional courts collectively submit more and more preliminary references each year.

At the same time, however, it must be stressed that the differences in the number of questions raised by the various constitutional courts remain significant. The Belgian Cour Constitutionelle and the Irish Supreme Court are evident outliers in this respect as they clearly stand out from the crowd. Only four other courts can be classified as more active ones while the rest of the courts remain rather passive. Similarly, there are evident differences in the types of questions various constitutional courts submit.

It is important to highlight here that such evident inter-court variation in the number and type of submitted preliminary references cannot be explained by the mere ‘willingness’ of individual constitutional courts to engage in the formal dialogue with the ECJ. Indeed, it is crucial to further explore why a few ‘prominent’ CCs (the Belgian and Irish CCs) have asked many more questions compared to those courts that have referred few (if any) cases to the ECJ. In other words, what matters in not only that CCs as a group submit more references collectively, but also which courts make use of the Article 267 TFEU procedure and how they do so.

Elsewhere, I argued that the constitutional courts’ decision to engage with the ECJ depends on quite complex mix of factors—legal, institutional, personal, and strategic.⁶⁷ Thus, it is for instance understandable from the strategic perspective that the Polish Constitutional Tribunal—following its illegal unconstitutional capture—have not submitted other preliminary references in cases in which it openly pushes against the authority of the ECJ and EU fundamental principles in general.⁶⁸ Moving swiftly along, from the institutional perspective, the low activity of the French Conseil constitutionnel might have something to do with the fact that the constitutional review conducted by means of the question prioritaire de constitutionnailité mechanism adheres to strict time limits which clash with the rather lengthy preliminary reference mechanism.⁶⁹

Thus, although the descriptive statistics provides us with valuable insights regarding the increasing ‘referencing’ activity of the collective of constitutional courts in time, these should be taken in caution when it comes to the context of individual Member States. The differences in approaches are significant and dependent on numerous legal and extra-legal factors. This—as we shall see in the following Part—matters not only for descriptive analysis, but also for the purposes of evaluating the activity of constitutional courts from the normative angle.

III. A Deliberative Potential of Constitutional Courts’ Questions

The previous Part provided us with a more comprehensive descriptive picture of how constitutional courts as a group make use of the Article 267 TFEU mechanism. Let us now inquire whether the phenomena of constitutional courts asking more EU law questions may be in any way regarded as normatively attractive.

Preliminary references submitted by CCs have been mainly valued in the context of constitutional conflicts between the EU and its Member States. The references have been recognized as ‘one of the more powerful procedural connectors among courts serving the cause of constitutional pluralism’.⁷⁰ The mechanism ought to represent ‘the only appropriate forum’ for constitutional dialogue because it provides the national CCs with an opportunity to challenge EU law, alert the ECJ and inform it of arguments why an EU measure may be problematic.⁷¹ Under the right conditions, such dialogue can serve a productive ‘auto-correct function’ and result in a full agreement between the rivalling parties.⁷² From this perspective, constitutional courts should participate in the formal dialogue with the ECJ because diplomacy and mutual engagement are essential when solving constitutional conflicts.⁷³ Correspondingly, the engagement is perceived as valuable because it has the capacity to prevent constitutional crises and preserve the ‘coexistence equilibrium’ between the high judicial authorities within the EU.⁷⁴

While this narrative highlights an important function of preliminary references submitted by CCs, it hardly reveals their full normative potential outside the rather extreme instances of open constitutional conflicts. Against this background, I argue that the core normative attractivity of questions submitted by CCs lies in their participative and deliberative democratic capacity. In order to support my claim, I will first show that preliminary references submitted by ordinary courts do not represent a tool that would effectively supplement the well-identified EU's democratic legitimacy gaps. Secondly, I will contend that compared to ordinary courts, constitutional courts are better designed to use preliminary references in order to channel through the voices of immobile EU citizens who have not benefited from the EU integration as much as those participating in cross-border economic activity.

A. Fighting EU's Inherently Anti-democratic Constitution through Ordinary Courts?

The Treaty on European Union (‘TEU’) formally commits the EU to the ideal of representative democracy. However, many have argued that despite numerous institutional attempts, the EU's complex system of democratic legitimation is ultimately insufficient to approximate such principal traditionally tailored to nation-states.

Indeed, it would be unproductive to lament here again about the lack of the European Parliament's legislative and oversight competencies, low involvement of national parliaments, fragmented elections, and other institutional roots of the EU's representative democracy ‘deficit’.⁷⁵ It suffices to highlight the underlying argument: the EU's decisions with far-reaching consequences for the Member States’ constituencies are primarily made through executive, administrative, judicial, and technocratic avenues (most importantly, the Council, the Commission, the ECJ, and the ECB) which are inherently anti-majoritarian.⁷⁶ In that way, the EU's formal commitment to the ideal of representative government is emasculated in practice and remains unfulfilled.

In trying to rectify this so-called democratic deficit, it has been suggested that what the EU lacks on the side of representation might be substituted by various institutional channels of participation and deliberation. After all, the TEU also formally declares that all EU citizens ought to ‘have the right to participate in the democratic life of the Union’ and ‘receive equal attention from [EU] institutions’.⁷⁷ In line with such account, the legitimacy of the decisions made on the EU level is to be based on the normative principle of deliberative supranational democracy under which mutual reflection about various preferences is facilitated not through traditional tools of representation, but by means of productive institutional discourse.⁷⁸

Following this logic, some have argued that the preliminary reference mechanism might embody one of the effective avenues for citizens’ participation in the EU's decision-making processes. When mixed with the ECJ's doctrines of supremacy, direct effect, state liability, and individual freedoms, the Article 267 TFEU procedure was said to empower not only domestic courts, but EU citizens as well (at least in an indirect way).⁷⁹ As Mancini and Keeling argued, the involvement of the citizens in the decentralised system of judicial review was a ‘dramatically democratising factor’ that took the EU law out of the hands of politicians and bureaucrats, and gave it to the people.⁸⁰

Seen from this perspective, the preliminary reference procedure is democratically significant because it gives EU citizens an opportunity to be listened to, to open or reopen a conversation based on arguments about the EU's policies, so that explicit and reasoned justifications for and against them become available for public deliberation. In this way, citizens are empowered to make effective use of their right to participate in ongoing political struggles for determining the proper scope, content, and limits of the EU's regulative actions.⁸¹ To put it differently, this specific form of judicial review ought to give voice to interests which are largely excluded from the EU's political processes and which might not otherwise be heard.⁸²

But does the ordinary preliminary reference procedure really meet such a robust democratic narrative? To answer this question, it is first important to highlight that the participative and deliberative reading of the Article 267 TFEU mechanism necessarily presumes that the procedure secures effective participation among all citizens on equal terms.⁸³ In other words, all EU citizens ought to have an equal right to legal contestation of EU's policies by means of the preliminary references submitted to the ECJ. Although this might seem unproblematic at first sight, more critical inspection reveals some serious conceptual difficulties of the democratic account of the procedure.

First and foremost, due to the fuzzy CILFIT criteria⁸⁴, the domestic courts enjoy wide discretion in deciding whether to pose the reference or not. This allows them to act as ‘critical gatekeepers’ in the formal dialogue between the citizens and the ECJ.⁸⁵ One could argue that the deliberative purpose of the procedure might be fulfilled even in cases where domestic judges decide not to submit the question, but only deal with citizens’ arguments without the involvement of the ECJ. However, that line of reasoning seems rather naïve. Due to their primary rationale of solving the disputes before them, ordinary courts often employ a wide range of tactics which allow them to avoid thorough engagement with the arguments of the citizens who ask them to initiate the procedure. Consequently, more often than not, parties to the proceedings will be left out with no answers to their EU law questions. More importantly, their arguments might not even be documented in the judgment's reasoning for the future engagement of the public. This necessarily questions the existence of an equal opportunity for the citizens to even engage in the deliberative discourse about the EU's policies.

There is, however, a deeper concern that challenges the democratising effect of the ordinary preliminary reference procedure. Although the mechanism undeniably enables the political agency of some citizens, one cannot overlook that it has predominantly been used by the ECJ to further a very specific purpose: to enable the pro-integration preferences enhancing the material scope of market freedoms and economic union.

Following Scharpf's logic, the use of the mechanism by the ECJ asymmetrically empowers private parties who have a major stake in increased capital or personal mobility as well as the financial and organizational resources to pursue their interests through litigation against laws of the Member States. On the contrary, the ECJ has historically been less responsive to the interests of the less mobile majority of European individuals and firms that benefit from existing national laws and regulations.⁸⁶ Indeed, preliminary rulings of Viking and Laval where the ECJ effectively subordinated the right to strike/collective bargaining to the freedom of establishment,⁸⁷ or of Omega where the Court did not uphold the ban on the video games simulating murder on the basis of the freedoms to sell and buy them, are instructive of such asymmetry.

Indeed, the ‘asymmetry’ narrative is also supported by number of studies on European legal mobilization, which in short suggest that individuals—‘have nots’—are disadvantaged in EU litigation while the corporate ‘haves’ necessarily come out ahead.⁸⁸ Thus, it is claimed that the EU law decentralised enforcement mechanisms increase opportunities for participation of citizens and firms, but only if they possess domestic courts access and sufficient resources to use it.⁸⁹ Similarly, Tommaso Pavone illustrated how group of prominent ‘Euro-lawyers’ managed to lobby some Member States’ judges against their domestic governments.⁹⁰

As a result, by now, it is clear that ‘integration through law’ has been uneven.⁹¹ Accordingly, as with other EU's representative institutions, the preliminary reference mechanism may be criticised for purposefully and systematically favouring economic freedoms that outweigh the personal, communicative, social, and cultural commitments of national policies.⁹²

Now, it has been suggested that the fact that the EU's representative institutions (among which we included the preliminary reference procedure) give cross-border market actors privileged platforms should not be perceived as accidental, but rather as illustrative of the EU's material constitution.⁹³ In her powerful account, Turkuler Isiksel argues that the EU's institutional mechanisms illuminate the distinctive configuration of its system of functional constitutionalism, whose normative claim to authority is founded primarily on the promise of effective government through the free market and economic union rather than on the principle of collective autonomy.⁹⁴ In this vein, the so-called democratic deficits are not bugs, but systematic features of the EU.⁹⁵

Indeed, others have argued that the EU's inherently anti-majoritarian processes are indispensable for both the affirmation and perpetuation of the EU's economic and monetary order and the particular kind of capitalism it serves.⁹⁶ Since at the very heart of the functional constitution of the EU lies a radical conception of private property based on the neo-liberal ideals of ‘sound’ money, economic freedoms, and ‘free’ competition, these ideals represent the key parameters by reference to which the validity and soundness of all national policies are determined.⁹⁷ Thus, it should not come as a surprise that the powers of the Member States are constrained not by the democratic popular will of ‘the people’, but rather by the will of ‘the market’.⁹⁸

As a result, it seems that the suggested participative and deliberative democratic reading of the Article 267 TFEU mechanism is at odds with the specific material and functional constitution of the EU. At the very least, it is reasonable to argue that the democratic potential of the procedure has been historically weakened by the ECJ which asymmetrically favoured the mobile cohort of the EU's citizens while leaving the voices of those who benefit from the national social-policy choices unheard. In what follows, I will contend that the normative attractivity of the questions submitted by national constitutional courts lies precisely in their institutional capacity to amplify these ‘unheard’ voices of immobile EU citizens.

B. Rectifying the Asymmetries: A Deliberative Potential of Constitutional Courts’ Design

Unlike ordinary courts, constitutional courts have the general power to challenge, oversee, and usually override the acts of elected national parliaments in the name of constitutional supremacy.⁹⁹ As we have seen, such capacity alone loses its uniqueness in the matrix of the EU since essentially every court and every administrative agency with or without the help of the ECJ may determine the incompatibility of national parliament's policies with the higher law—the EU law.¹⁰⁰ What has been mostly overlooked, however, is that the ‘uniqueness’ of the constitutional courts is not only mirrored in the enumeration of their formal powers, but also in their institutional design. Indeed, constitutional courts were specifically designed to represent ‘deliberative forums of a distinctive kind’ within the communicative processes of deliberative democracy.¹⁰¹ It is because of this specific design, CCs are said to be better equipped than ordinary courts to perform responsive judicial review—to counter the risks of anti-democratic monopoly power, democratic blind spots, and burdens of legislative inertia.¹⁰²

It is obviously not possible, nor desirable to deeply analyse all the institutional features that ought to secure this deliberative democratic promise of constitutional courts here.¹⁰³ Nevertheless, one can highlight at least a few such institutional traits.

First, unlike ordinary courts, CCs ‘communicate’ through various institutions of participation, intervention, and amici curiae with a larger number of privileged and non-privileged political actors. Higher diversity of litigants is supposed to maximise the range of arguments that are part of the formalised constitutional dialogue.¹⁰⁴ In this sense, the different types of constitutional proceedings as well as the relaxed means of formal involvement are intended to contribute to the creation of an open, inclusive and participatory discourse on the content of the constitution, in which a wide range of actors—state institutions (government, parliament, administrative bodies, ombudsman, etc), private actors (individuals, corporations, NGOs) and supranational institutions (European Court of Human Rights)—can participate. In other words, constitutional review is meant to institutionalise the diverse democratic debate on the concrete meaning of constitutional norms in the life of society.¹⁰⁵ Importantly enough, not resolving disputes between two parties, but deliberation about the legitimacy of democratic policies is the primary goal of the constitutional review.

Secondly, in comparison with ordinary courts, CCs typically enjoy larger financial and personal resources (including a high number of law clerks or analytical departments) that ought to enable them to carefully consider all the parties’ arguments from a rather systemic perspective.¹⁰⁶ Also, CCs usually enjoy greater docket-control discretion so they can filter out the most important cases—this in turn gives them more time to adjudicate. Such benefits ought to help CCs to provide all citizens with explicit and well-reasoned justifications for and against the contested policies that are then made publicly available, so that they can be inspected and challenged with counterarguments potentially leading to a change in public opinion.¹⁰⁷

Thirdly, the decision making of constitutional courts is of greater interest to the public and media, which places greater demands on the decision making of the constitutional judges in terms of both results and reasoning. It should therefore be more difficult for specialised constitutional courts to avoid ‘hard cases’ through classical formalistic judicial techniques which once again serves deliberative purposes.¹⁰⁸

Fourthly, unlike in case of ordinary judges, the appointment procedures for constitutional judges are intended to reflect the higher political importance of their office, and therefore typically involve a wider range of institutions (Senate, judicial council, etc). Moreover, the public and media typically place higher demands on the diversity of justices in terms of both personal characteristics (gender, colour of skin, sexual orientation, etc),¹⁰⁹ as well as professional backgrounds (professional lawyers, academics, politicians, ethicists, philosophers, etc).¹¹⁰ Thus, the deliberative logic is pursued once again—the higher the diversity of the deliberators, the more inclusive the decision-making process is supposed to be.

Now, the exact scope of the deliberative potential of individual constitutional courts obviously very much depends on the particular mixture of deliberative devices they have at their disposal. For instance—recalling the descriptive part of the Article here—the deliberative potential of Italian Constitutional Court and Belgian Constitutional Court might be lower in comparison to their Czech and German counterparts when it comes to the diversity of political actors they might engage into the constitutional review. Since the former CCs lack jurisdiction to hear individual complaints, their deliberative potential might be limited by their ability to hear arguments originating only from the ‘privileged’ political institutions, not individual citizens directly.

Nevertheless, the above-described standard characteristics of most of the European constitutional courts are illustrative of the point I am trying to highlight here: constitutional courts are specifically designed to be—at least abstractly—responsive to the interests of the ‘unvoiced, marginalised, and insular minorities’ within the national democratic process.¹¹¹ By having the institutional capacity to take all citizens’ arguments seriously, constitutional courts aim to generate substantively attractive outcomes that are later subjected to the ongoing process of contestation from the side of other political actors.¹¹² Accordingly, national constitutional courts represent by their design one of the avenues that generate legitimacy of collective decisions made by the state.

How does such specific deliberative potential of constitutional courts translate in the context of the EU? In his insightful account, Jan Komárek argued that national constitutional courts form ‘an important component of communicative arrangements, which generate decisions that remain open to further revision, and are subject to communicatively generated legitimacy’ within the space of European Constitutional Democracy.¹¹³ Against this background, constitutional courts’ primary role is not to further promote the benefits of those who are well served at the EU level, but to amplify and protect the interests that are systematically ignored by the process of European integration—which are primarily those of the cohort of immobile citizens.¹¹⁴

Following this logic (but unlike Komárek¹¹⁵), I contend that the constitutional courts’ involvement in the preliminary reference procedure might be normatively attractive precisely because of their deliberative institutional design. Unlike ordinary courts that have not been able to rectify EU's democratic ‘deficits’ through preliminary references, constitutional courts are better equipped to channel through the interests of the less mobile majority of European citizens and firms who mostly benefit from existing national laws, and who might want to challenge the omnipotent pro-integration market logic of the EU's functional constitution.

Indeed, thanks to the mix of various kinds of constitutional proceedings, high diversity on the side of both participants and judges, and larger resources, constitutional courts have better chances of identifying and amplifying the arguments that might outweigh the predominant perspective of the neo-liberal ideals of ‘sound’ money, economic freedoms, and ‘free’ competition. This can, in turn, generate overall higher legitimacy of the EU's decision making.

To be sure, I do not claim that preliminary references submitted by constitutional courts are necessarily participative and deliberative. As it was suggested in the descriptive part above, it matters which constitutional courts and in what context submit the references. Although some constitutional courts are by design more accessible to a variety of actors and the ‘immobile citizens’ than others and thus, they might have higher deliberative potential as a starting point, having such preconditions by design does not necessarily mean that they will use the actual potential when making the decision to submit the reference. Following this logic, even if the Polish and German Constitutional Courts might enjoy higher deliberative potential by their institutional design, their references do not have to necessarily fulfil it either due to the ‘rule of law crises’ context, or due to their inability to be responsive to the arguments of the less privileged EU citizens in the particular case.

As we shall see in the next Part, one needs to put particular references in concrete perspective in order to assess whether constitutional courts managed to engage with and convey the ‘still overlooked’ arguments to the ECJ in a persuasive manner. I only contend that the above-described trend of constitutional courts asking more and more questions might be perceived as normatively attractive because constitutional courts’ involvement in the Article 267 TFEU procedure has the potential to rectify some of the asymmetries present in the EU's decision-making processes. Thus, since constitutional courts have a better capacity than ordinary courts to maximise the deliberative potential of the formal dialogue with the ECJ, their increasing activity in that regard might intensify the political legitimacy of the EU's policies as well.

IV. Assessing Constitutional Courts’ Deliberative Performance in Context

So far, I hope to have shown that the constitutional courts’ tendency to submit more preliminary references to the ECJ might be perceived as normatively appealing due to their deliberative capacity. In what follows, I try to put this theoretical claim into practical context. Obviously, it is not possible to assess here the deliberative performance of all constitutional courts in all references. Nevertheless, drawing on the contextual analysis of two cases, I will try to argue that the constitutional courts’ deliberative potential might evolve in two main directions in practice. The first story shows how that potential can translate into productive questions channelling diverse institutional voices to the ECJ. The second one then demonstrates how constitutional courts leave their institutional potential unfulfilled while ending up not addressing, but rather intensifying the EU's legitimacy concerns sketched out above.

Let us start with the more promising story—the case of the bargaining power of Lithuanian milk farmers.¹¹⁶ In Lithuania, the raw milk market is quite specific as on the one hand, there were more than 20,000 very small milk producers who usually owned only a handful of cows, while there were only six companies processing and further selling 97% of the raw milk. In Case C-2/18, the Lithuanian Constitutional Court asked the ECJ whether EU law precludes national measures which ought to combat the unfair practices of the dominantly concentrated milk processors (buyers) who would simply inform producers (farmers) of the purchase price of the milk, without prior negotiation, leaving those individual farmers with no option but to accept the conditions imposed.

In the order for preliminary ruling, the Lithuanian constitutional judges communicated the arguments that were presented before them prior to the submission of the question to the ECJ. In particular, the reference included not only the arguments of the group of MPs who challenged the national measure, but also of the rest of the Lithuanian Parliament, the Ministry of Justice, and the Competition Council as well.¹¹⁷ It is precisely the inclusion of a number of national institutions which makes the involvement of the Lithuanian Constitutional Court in the preliminary reference mechanism normatively attractive. From our deliberative perspective, only by bringing more institutional views to the table, the Constitutional Court increased the ECJ's likelihood of generating a ‘good’ outcome in a way which would not be possible in the course of the judicial review conducted by an ordinary court.

Nevertheless, what is more important is that the Lithuanian Constitutional Court not only amplified a higher number of institutional voices in general, but that it amplified specific kind of those voices in particular. One cannot overlook that in comparison with the arguments of the other institutions, the Lithuanian Constitutional Court predominantly highlighted the Parliament majority's position that maintained that the freedom of negotiation is not absolute. The Parliament specifically argued that the restrictive measure in question ought to ensure equity and justice because the bargaining power of raw milk producers, on the one hand, and processors or buyers, on the other hand, is unbalanced in favour of the latter.¹¹⁸ In that regard, constitutional judges described the Lithuanian milk market wider context in detail (main actors, history etc.). Moreover, they also referred to the explanatory notes accompanying the draft law on the prohibition of unfair practices and emphasized that one of the purposes of that law was to limit the exercise of disproportionate market power by milk processors and the unfair advantage that operators trading in milk products obtained from a reduction in wholesale prices of such products.¹¹⁹

Indeed, this way of communicating the Parliament's majority arguments in the order for preliminary ruling illustrates well how constitutional courts might use the participatory and deliberative potential of the national constitutional review in order to highlight the interests of the less mobile part of the EU's citizens who mostly benefit from national legislation. Even though Lithuanian Constitutional Court was not explicit about its view on the outcome of the question it submitted,¹²⁰ it made sure that the voice of the historically marginalised group of EU citizens was heard.

Nevertheless, there are also clear instances where the deliberative potential of the constitutional courts’ preliminary references submitted is not fulfilled. The well-known case of Coman and Others ¹²¹ serves as a good example in that regard. To put it shortly, in that case, the Romanian Constitutional Court asked whether a Member State has an obligation to recognise, for the purpose of granting a right of residence to a national of a non-EU state, the marriage of that national to an EU citizen of the same sex lawfully concluded during the period of genuine residence in another Member State, in accordance with the law of that Member State.¹²² Importantly enough, in the order for preliminary ruling, the Romanian constitutional judges mainly argued that if the right of residence were not granted, interference with the freedom of movement of EU's citizens might occur—the possibility of moving in one or another Member State would vary depending on whether provisions of national law allow marriage between persons of the same sex.

Now, it should be first highlighted that unlike the Lithuanian Constitutional Court in the previous example, the Romanian Constitutional Court did not include in its order any observations of other institutions besides the arguments of the parties of the original dispute. Although the constitutional review in that case was not initiated by MPs, but rather by an ordinary court by means of question constitutionality referral (ie a form of ‘concrete constitutional review’), the Romanian Constitutional Court could have also obtained submissions from the presidents of the two chambers of the Romanian Parliament, the Romanian government, and the Romanian Advocate of the People in that kind of proceedings as well.¹²³ The fact that it did not do so—or it did not communicate it in the order for preliminary ruling—suggests a lower deliberative value of that question.

Nevertheless, there is a deeper aspect of that Romanian referral that ultimately diminishes its potential. The normative ideal of participative and deliberative decision making requires that even if some external actors are not officially included in the process of persuasion and thus cannot formally present their arguments, the court nevertheless forms its stance inclusively, empathically, and responsively.¹²⁴ Consequently, the referring judges are expected not only to engage with the arguments they heard, but also to imagine hypothetical and unheard arguments that could appear in the public sphere including those of marginalised groups. One cannot but notice that the Romanian constitutional judges presented only one specific paradigm of the question—that of free movement.

Indeed, in its order, the Romanian Constitutional Court emphasised that the issue of non-recognition of same-sex marriages had come about in a situation relating to free movement within the European Union and cannot be addressed separately. In that way, the Constitutional Court framed the issue as a question of market freedom, not as a question of human rights more broadly.

By highlighting that the case only concerns the EU citizens who have already made use of the freedom of movement and entered the same-sex marriage in the Member States where they ‘genuinely’ resided, the Romanian judges necessarily left out the concerns of the remaining EU LGBTQ+ citizens who have not decided to live the truly ‘cosmopolitan’ way of life and have not travelled, worked, and lived in other Member States. In other words, by framing the issue in this very specific way, the Romanian Constitutional Court ignored the interests of the ‘bad EU citizens’, who unlike the ‘good EU citizens’, fail to boost the internal market and live by the ideology of the cross-border movement.¹²⁵

As a result, the preliminary reference in Coman and Others is illustrative of the way in which constitutional courts may not make use of their participative and deliberative potential. More significantly, it also shows that by means of preliminary references, constitutional courts can further contribute to the reproduction and exacerbation of the dominating EU's paradigm of economic union and market freedoms. Thus, the example uncovers how the lack of constitutional courts’ responsivity to the ‘unheard voices’ can diminish their ability to generate overall higher legitimacy of the EU's decisions.

V. Conclusions: A Yet Unfulfilled Promise?

Even though constitutional courts have always been rather ‘special creatures’ of the jungle of the European Union, their institutional potential within the EU's legitimating mechanisms has been mostly overlooked both from descriptive and normative perspectives.

In this Article, I showed that national constitutional courts have developed a third anti-marginalisation tactic as they started to submit more preliminary references to the ECJ. Although constitutional courts, which have asked no question whatsoever, represent a clear minority by now, the differences in the involvement of individual courts in the Article 267 TFEU mechanism remain significant. I argued that the most normatively attractive way to assess such development is to appreciate the unique participative and deliberative institutional design of constitutional review. Indeed, due to their institutional capacity allowing them to be more responsive, constitutional courts can use preliminary references to channel the voices of the less mobile part of EU citizens to the ECJ and thus balance the inherently anti-democratic material constitution of the EU that hinges on the neo-liberal ideals of ‘sound’ money, economic freedoms, and ‘free’ competition. Nevertheless, as the contextual analysis of the two cases demonstrated, the legitimacy-enhancing potential of preliminary references translates in contradictory directions in practice. On the one hand, the more promising story of Lithuanian farmers showed that in comparison with their ordinary counterparts, constitutional courts have used the mechanism to amplify specific arguments of various institutional actors – including national parliament – which have in turn provided the ECJ with an opportunity to consider the interests of the less vocal members of the EU's community. On the other hand, the story of Coman and Others demonstrated a riskier side of the potential. Instead of addressing the existing asymmetries at the EU level, constitutional courts may by means of preliminary references contribute to the reproduction and exacerbation of the dominating EU's paradigm of the economic union and market freedoms.

Now, it is crucial to emphasise that the specific deliberative reading of the constitutional courts’ role in the preliminary reference mechanism presented here does not claim to be something more than it could be. Even if all constitutional courts used their deliberative potential to the fullest in every reference, they could not distort the EU's material constitution based on the specific version of market capitalism on their own. Constitutional courts’ questions do represent only one part of an iterative sequence of unceasing political communication in the long run.¹²⁶ Consequently, even deliberative preliminary questions should be perceived cautiously as mere catalysers of a wider democratic discourse.

Nonetheless, I hope to have shown that it is important to be receptive to the deliberative performance of individual preliminary references in order to be able to assess whether constitutional courts in fact help to generate more legitimate decisions on the EU level, or whether their unique institutional design still remains in the sphere of unfulfilled promise. A detailed ‘deliberative’ analysis of the (future) individual questions seems to be a rather productive research agenda in that regard.