Transnational partisan networks and constituent power in the EU

The constitutional politics of the EU have been intensely debated in recent times. Onewidespread perception animating thedebate is that the constitutionalizationof theEUpolity has takenplace in a largely apolitical fashion, in a process driven by decisions of courts, notably the Court of Justice of the EU, which are made in isolation from political contestation and popular participation (e.g. Glencross, 2014; Grimm, 2015; Scharpf, 2017). This has led to increased calls for the politicization of the EU constitutional order (e.g. Fabbrini, 2016; Glencross, 2014; Niesen, 2017). According to this view, citizens should be endowedwith the ability to shape, revise, and legitimize evolving constitutional settlements. This view is voiced not only in the academy. Analogous concerns have also been expressed among political activists, perhaps most prominently in the pan-European Democracy in Europe Movement 2025 (DiEM25). DiEM’s express ambition is to establish a European constituent assembly in which the citizens of Europe may deliberate on how to transform the EU into a “fully fledged democracy” (DiEM25, 2017; see also Marsili & Milanese, 2018). The goal is to transform the citizens of Europe into a constituent power capable of revising the constitutional features of the EU at large. The normative logic underpinning these scholarly and activist perspectives is as old as modern democracy itself: if theEU is to be ademocratic constitutional order then its citizensmust be the source of that order, and the subjectswith whom sovereignty ultimately rests (Cohen, 2012; Habermas, 2011). But how exactly should we think of constituent power in the EU? It is one thing to stress the importance of enhanced constitutional agency, and quite another to provide amodel of how constitutional agency can and should be exercised. As a response to this question, a small number of political and legal theorists have recently elaborated sophisticated accounts of constituent power in the EU that are intended to resolve these complex normative and conceptual issues (Cheneval, 2011; Crum, 2012; Habermas, 2011; Niesen, 2017; Patberg, 2017). In this article I begin by outlining these accounts. I suggest that there are good reasons to favor accounts of mixed constituent power, as defended by Jürgen Habermas (2011) and others, over those that see constituent power as resting solely with the peoples of EU member


INTRODUCTION
The constitutional politics of the EU have been intensely debated in recent times. One widespread perception animating the debate is that the constitutionalization of the EU polity has taken place in a largely apolitical fashion, in a process driven by decisions of courts, notably the Court of Justice of the EU, which are made in isolation from political contestation and popular participation (e.g. Glencross, 2014;Grimm, 2015;Scharpf, 2017). This has led to increased calls for the politicization of the EU constitutional order (e.g. Fabbrini, 2016;Glencross, 2014;Niesen, 2017). According to this view, citizens should be endowed with the ability to shape, revise, and legitimize evolving constitutional settlements.
This view is voiced not only in the academy. Analogous concerns have also been expressed among political activists, perhaps most prominently in the pan-European Democracy in Europe Movement 2025 (DiEM25). DiEM's express ambition is to establish a European constituent assembly in which the citizens of Europe may deliberate on how to transform the EU into a "fully fledged democracy" (DiEM25, 2017; see also Marsili & Milanese, 2018). The goal is to transform the citizens of Europe into a constituent power capable of revising the constitutional features of the EU at large.
The normative logic underpinning these scholarly and activist perspectives is as old as modern democracy itself: if the EU is to be a democratic constitutional order then its citizens must be the source of that order, and the subjects with whom sovereignty ultimately rests (Cohen, 2012;Habermas, 2011). But how exactly should we think of constituent power in the EU? It is one thing to stress the importance of enhanced constitutional agency, and quite another to provide a model of how constitutional agency can and should be exercised.
As a response to this question, a small number of political and legal theorists have recently elaborated sophisticated accounts of constituent power in the EU that are intended to resolve these complex normative and conceptual issues (Cheneval, 2011;Crum, 2012;Habermas, 2011;Niesen, 2017;Patberg, 2017). In this article I begin by outlining these accounts. I suggest that there are good reasons to favor accounts of mixed constituent power, as defended by Jürgen states. However, I also show that existing accounts of that kind are inadequate, in that they under-theorize how real-world citizens can come to act as a constituent power in the first place.
The remainder of the article then aims to explore how the abstract principles of pouvoir constituant mixte may be connected to political practice. I look for inspiration to a much-neglected political practice-transnational partisanship.
Drawing on the case of Christian Democratic transnationalism, I specifically discuss how partisan networks honed by elected representatives from different European states have provided the backbone for the Christian Democrats' successful exercise of supra-state constituent power. This distinctive mode of cross-border organization and coordination helps us to imagine how today's citizens can engage as agents of EU constitutional politics.
In sum, the article's argument is that there is much to learn from Christian Democratic transnationalism when it comes to the exercise of constituent power in the EU. Not only does it offer a promising general model for organizing and coordinating constitutional agency; as the final sections of the article will show, reflecting on its less attractive features also permits us to see what mistakes are to be avoided in future efforts to politicize the EU constitutional order.

CONSTITUENT POWER IN THE EU
A polity can be considered democratic if power vests ultimately in the people. This is the essence of the modern doctrine of popular sovereignty that underpins most democratic constitutional orders. The people of a democratic polity is accordingly ascribed the status of constituent power-the agent to whose will the original constitution can be traced and who retains the authority to decide all issues of constitutional design. This means that the people are endowed with two powers: the legitimate power to bring a democratic constitutional order into existence, and the power to reform its foundational norms and rule as they see fit (see Arato, 2016;Böckenförde, 2006;Kalyvas, 2005;Loughlin, 2014).
On the face of it, it is difficult to make sense of this concept in the context of the EU (Walker, 2007). For the EU is not a democratic federal state with a central constitutional document that can in some relevant sense be seen as the product of the people, but a supranational polity that has developed from a relatively loose association of nation-states into a constitutional order consisting of many member states (on the nation-state/member state distinction, see Bickerton, 2012; on constitutionalization, see Grimm, 2015). Who in this polity might be the agent to whom the ultimate authority to decide constitutional issues can be attributed, and who is thus responsible for enacting constitutional reform and revision?
Two different responses to that question have emerged in the recent political theory literature. The first assumes that the EU is "not constituted by a single people or 'demos'" (Crum, 2012, p. 45) but that it has several pouvoirs constituents; namely, the demoi of the member states (Cheneval, 2011;Cheneval, Lavenex, & Schimmelfennig, 2015, p. 4).
Authors defending this sort of position follow broadly Rawlsian argumentative strategies, invoking such devices as an original position and public reason, and suggest that decisions about EU constitutional design ought to be determined by the peoples of member states. As Cheneval et al. (2015, p. 4) write, sovereign decisions such as treaty ratifications "are to be taken by or [must be] directly accountable to the pouvoir constituant of the states which are the citizens organized as People." Different conclusions have been drawn on the implications of this account for the politicization of the EU constitutional order. Some scholars take it to imply that the member states' peoples must participate in a direct and unmediated fashion in EU constitutional politics (Cheneval, 2011, pp. 133-137). Other authors accept that constitutional decisions of this sort may be made by elected representatives, but affirm the positive role that direct democratic mechanisms can play in incentivizing representatives to aim at constitutional decisions that all those subject to it can accept (Crum, 2012, pp. 46-52). Either way, the subjects exercising constitutional agency always ought to be the member state's demoi.
One straightforward reason to find this strategy appealing is that locating constituent power in the demoi of EU member states seems to be descriptively accurate, inasmuch as member states are often said to remain the "masters of the treaties" (Grimm, 2016, pp. 56-58); so, from the point of view of realism, the first account has much going for it.
Yet, as several commentators have pointed out, the focus on domestic pouvoirs constituants faces sizable obstacles. If constituent power refers here, as it typically does, to the more or less unrestrained competence to establish and revise constitutional orders, does this mean that member states' constituent powers also operate in an unbound fashion when they engage constitutional politics at the EU level? And if that is the case, what happens when conflicts between various pouvoirs constituants arise?
The first account of constituent power in the EU has no answers to these questions. Indeed, its assumption that the constituent power of the member states' peoples simply extends beyond the state, as if "the involvement in EU constitutional politics represented just another mode of intra-state constitutional revision" (Patberg, 2017, p. 206), veils these problems. At work here seems to be the contestable Schmittian assumption that the allocation of constituent power is an either-or question, such that any exercise of constituent power beyond the state is bound to nullify constituent power at the state level (see Cohen, 2012, Chapter 2).
Aiming to overcome such a view, a second account of constituent power in the EU that has been proposed in the recent literature on the topic offers a possible way forward. The suggestion is to conceive of constituent power in the EU as pouvoir constituant mixte, understood as a supra-state constituent power in which individuals participate in their dual capacity as members of the member states' demoi, on the one hand, and EU citizens, on the other.
One prominent author who has developed such a view is Jürgen Habermas. In his recent work on the topic, Habermas specifically imagines a complex European constituent assembly in which delegates, represented in their two personae, deliberate over the constitutional features of the EU (Habermas, 2011(Habermas, , pp. 68-70, 2017. The thought is that in such an assembly individuals are "able to address the other side with the aim of striking a balance between their respective interests" (Habermas, 2017, p. 174), and so can devise constitutional norms and rules that prevent potential conflicts between the member states' pouvoirs constituants, while ensuring that individuals' normative expectations vis-à-vis the EU are given adequate expression.
Importantly, Habermas does not want to show how supra-state constituent power can operate as a real social force, but the extent to which certain elements of the current EU constitutional order can be justified from an "idealised participant's perspective," and which of these elements require revision (Habermas, 2011, p. 69). His strategy is to start from the presumption that the EU has already assumed a democratic character as a result of the Treaty of Lisbon, and then conduct a rational reconstruction of the practice of law-making and implementation in the post-Lisbon Treaty EU. Analogous to his more generalized reconstructive approach, as developed in Between Facts and Norms (1996), this entails asking what hypothetical delegates to a European constituent assembly would decide in an idealized rational dialogue about the EU's constitutional foundations.
No doubt, the notion of pouvoir constituant mixte presents a significant advance in theorizing about constitutional agency in the EU. It can powerfully overcome the problems noted in connection with the first account of constituent power in the EU. Yet, if one seeks an action-guiding model of EU constitutional politics, Habermas's account is the wrong place to look. For, as it conceives of supra-state constituent power in purely hypothetical terms, as a post hoc justificatory device for existing institutional arrangements, it effectively accepts that constitutional norms can plausibly be articulated in isolation from procedures of political contestation-without the participation of those to whom these norms are supposed to apply.
On a strong reading, one might say that Habermas's perspective does not so much provide a solution to the problem that the constitutionalization of the EU polity has taken place in a largely apolitical fashion but a justification for the apolitical process of constitutionalization that has so often been lamented. Although Habermas is generally sympathetic to a "more democratic" EU, he stops short of offering prescriptions for the politicization of the EU constitutional order.
That Habermas conceives of EU constitutional politics in purely counterfactual terms raises another problem. Even if his approach of rational reconstruction could settle all questions to do with the constitution of the EU at a particular moment, constitutional norms must-at least to some extent-be adapted to an evolving historical context. In relegating citizens to being passive recipients of norms, however, Habermas's account leaves little room for shifting preferences about their political communities and forms of collective organization to bear on decisions concerning constitutional design. 1 To understand this worry better, consider Habermas's interpretation of the domestic dimension of the pouvoir constituant mixte. This is almost exclusively conservationist, in the sense that Habermas imputes to the citizens of member states' a strong and stable commitment to conserve the "revolutionary constitutional achievements of the past" (Habermas, 2011, p 70;see Von Achenbach, 2017, p. 196). Yet, even if citizens are currently so inclined, they might at some point in the future wish to reinvent themselves as a political community, perhaps in a "co-evolution with their other persona" (Niesen, 2017, p. 190). An account of constituent power that allows for such shifts cannot, however, remain tied exclusively to philosophical reflection; citizens must be made active participants to constitution-shaping, able to translate re-articulations of member state communities into constitutional arrangements.
Moving beyond a purely counterfactual perspective, a number of scholars have asked how Habermas's notion of pouvoir constituant mixte might inspire the exercise of constitutional agency in the real world. 2 Niesen (2017, pp. 190-191), for example, toys with the idea that the Habermasian approach might hold the potential to empower "future subjects of constitutional reform," by providing them with a "new language in which to ground transformative European citizen engagement." Similarly, Patberg (2017, p. 210) interprets Habermas's model as inviting reflection on how the pouvoir constituant mixte can be "brought into the world via a conscious decision taken by the demoi of the member states." This shift of emphasis from the counterfactual to the phenomenological is helpful for normative theorizing on the politicization of EU constitutional politics, as it alerts us to the importance of thinking about how normatively desirable forms of constituent power could look in reality. But Patberg's and Niesen's reflections are also too broad and general to glean from them a more concrete account of constituent power in the EU. The most urgent questions to be explored in greater detail concern the possibility of feasible organizational models for supra-state constituent power in the EU; the ideational resources that can support and sustain the collective pursuit of supra-state constitution-making; and the channels from which the sociological legitimacy of such efforts-roughly, the extent to which exercises of constituent power in the EU are considered legitimate by European citizens-can be expected to flow. The latter point appears of special relevance for a theory of constituent power that aspires to be democratic in the sense of allowing space for contestation and dynamic change. That there is a "deep relationship" (Arato, 2016, p. 131) between sociological legitimacy and a democratic understanding of constitutional change and revision follows directly from the earlier elaborated commitment to transcending a purely counterfactual perspective on constituent power.
My burden in the remainder of the article is to address these questions and articulate an empirically informed model of how supra-state constituent power may be exercised in the EU. To that end, I will explore a hitherto neglected form of supra-state constituent power that rests on transnational partisan networks, in which like-minded individuals from different political spheres coordinate with the aim of shaping the constitutional order of Europe.

TRANSNATIONAL PARTISANSHIP AND CONSTITUENT POWER
Reflecting on the prior section's discussion, we may single out several desiderata that an account of constituent power in the EU must meet: 1. It must be a mixed form of constituent power whose participants act both in their capacity as members of states and as members of a supranational polity.
2. It must allow for participation and contestation.
3. The institutional properties of constituent power must be specified, in particular: • a feasible model of organization.
• Ideational foundations that help justify and sustain the collective pursuit of supra-state constituent power.
These desiderata relate both to what may be called the entitlement and the agency (or capacity) dimension of constituent power. The entitlement desiderata (1 and 2) derive from the suggestions put forth by Habermas (on the superiority of mixed accounts of constituent power relative to unitary ones), and from the work of Patberg and Niesen (and their emphasis on the importance of practices of participation and contestation). The agency desiderata (3), on the other hand, demand approaching the subject matter from an angle that is quite different from that taken by all these authors, paying particular attention to the practical questions of how a supra-state constituent power can be realized.
The reason for this shift towards a greater emphasis on practical agency is simple. If the debate around constituent power in the EU is to move fully beyond a counterfactual perspective, headway must be made on precisely those applied questions that existing accounts touch upon only address or eschew altogether. In short, a model of pouvoir constituant mixte that is capable of guiding political practice has to spell out in more details how real-world citizens could act as shapers of the EU's constitution.
To develop such a model, I follow a strategy of theory-building that is historically oriented and reconstructive. Tak Let us look more closely at this form of political agency and explore whether it may satisfy the aforementioned desiderata.

Organizing constituent power: Partisan networks
Conceptually speaking, the transnational partisan networks developed by Christian Democratic parties were what White (2014, p. 388) calls "low-density networks" (on network theory more generally, see Kadushin, 2011). This means that they centered on a few nodal individuals whose activities bridged the different national parties. What makes the language of network especially appropriate (rather than speaking of individual transnational encounters) is that transnational interactions between those nodal figures gradually became regularized and to some degree even institutionalized. Transnational associations such as the Nouvelles Équipes Internationales (NEI), founded in 1947, gave further structure and continuity to the interactions of partisan elites.
The primary function of the Christian Democrats' transnational networks was to facilitate dialogue about shared commitments in more or less informal exchange to foster discussion between politically like-minded individuals from different national spheres. The topics addressed ranged from general reflections on the present state and possible future of the jointly espoused political project, to more specific questions to do with connecting principles to practice (Gisch, 1990;Kaiser, 2007, esp. Chapter 5). Here the more general nature of partisan practice makes itself visible: partisanship is not just about formulating political proposals but also about implementing them (Rosenblum, 2008;White & Ypi, 2016).
Insofar as networked transnational partisan coordination is about implementing political aims, it can be harnessed to realize objectives that single states cannot achieve on their own-for example, by helping national parties synchronize their actions and activities in a way that allows them to execute larger tasks. From the point of view of EU constitutional politics, this is one of the most attractive features of this form of cross-border organization, and the reason for why it can facilitate the exercise of supra-state constituent power. There are two different, sequential roles that the Christian Democratic partisan networks performed in this connection-roles we may consider in turn.
The first is preparatory: transnational partisan networks can over time lay the groundwork for the future enactment of constituent power beyond the state, helping individuals to overcome difficult challenges associated with exercising power transnationally or supranationally. Chief amongst these challenges is a lack of agreement among members of the network as to the goals to be pursued in concert. Likewise, the national parties that form the network may simply lack the means to translate their shared aims directly into decisions, for instance, because they are not in government in their respective states. Transnational party networks can contribute to solving these problems, in that they provide platforms for cross-national dialogue aimed at devising common strategies to gain power in the future and/or developing a shared vision that all national parties uphold.
To see what this means in practice, consider one transnational party network that may be straightforwardly interpreted as having prepared later exercises of constituent power-the Secrétariat International des Partis Démocratiques d'Inspiration Chrétienne (SIPDIC). SIPDIC was a transnational organization of Catholic parties that was founded in 1926 by the Italian left-wing Catholic Luigi Sturzo with the primary aim of fostering lasting peace between the peoples of Europe. The organization disintegrated in 1939, in part because many democratically minded Christian parties were put under severe pressure by fascist regimes, but also because the parties involved were unable to agree on a shared agenda of political objectives (Kaiser, 2007, p. 116). Yet, although SIPDIC failed as a transnational organization, the cross-border exchanges that occurred under its aegis proved crucially important for the Christian Democrats' constitution-shaping activities after World War II.
There are three senses in which this is true. First, as historians of Christian Democracy (Kaiser, 2007, p. 117) have pointed out, the transnational dialogue SIPDIC facilitated helped overcome a general reluctance among Christian Democratic parties to organize transnationally-a reluctance rooted in the view that the Catholic Church is the only legitimate international representation of Catholic interests and ideas. This was an important precondition for effective transnational organization after the War.
Second, the deliberations among SIPDIC's members also promoted a collective learning process, showing to all involved that party cooperation could only be effective if it combined strictly pro-democratic parties with a shared value system and was extended to democratic liberal and conservative Catholics to broaden its basis and strengthen its potential influence-and that it required some shared core strategic policy objectives. (Kaiser, 2007, p. 117) Though this learning process did not result in programmatic agreement in the interwar years, as I have noted, it prepared the way for the postwar emergence of an ideological consensus among Christian Democrats on European unification.
Third, though there was no direct organizational continuity between the networks formed around SIPDIC and the transnational associations that formed after World War II, key figures of postwar Christian Democracy-notably Konrad Adenauer and Robert Schuman-attended SIPDIC meetings and congresses and engaged in the association's deliberations about the unification of Europe. It is not implausible to assume that their participation at these events shaped their views on achieving European unification, facilitating future agreement about collective goals-though it was only the mutual exchanges in the later founded NEI that "created the basis of personal confidence among … such politicians as Schuman, De Gasperi and Adenauer, which … made possible the policy associated with their names" (Gisch, 1990, p. 484).
Importantly, as a practice of reasoning that involves individuals who are both members of nation-states and potential members of a future supranational European order, the latter deliberations within SIPDIC may be interpreted as instantiating the sort of two-personae exchanges envisaged in the Habermasian conception of pouvoir constituant mixte. In a sense, one may say that the success of SIPDIC was undermined by the fact that the interests tied to the first persona-being a member of a particular nation-state-were of greater significance to many of the participants than the interests associated with the (hypothetical) second persona. The French Popular Democrats, for one, mainly sought to use SIPDIC's network to foster alliances against revisions of the Versailles Treaty and, thus, also against the German Centre Party and Germany more generally. This created an atmosphere of mistrust among the parties and triggered numerous debates that were primarily about the assertion of national interests, rather than about goals shared by all network members. This was to change in the postwar era, where, as we shall see, Christian Democratic partisans were better able to bring their national commitments and transnational or supranational visions into equilibrium.
I now turn to the second role transnational partisan networks can serve, which is even more relevant to the question of constituent power than the first. This may be called their executive role and refers to instances where members of partisan networks are capable of putting their shared goals into effect. At least two conditions must be met in order for this to be possible. The first is that members of the network agree on the aims and goals they wish to pursue in concert.
Why this is crucial is not difficult to see: disagreement over possible courses of action often has action-impeding implications, and sometimes even erodes mutual trust amongst those who associate with the purpose of acting together.
The example of SIPDIC shows that transnational partisan networks are not immune to these problems.
The second condition may likewise be inferred from the above discussion: this is that most or all of the parties forming the network are in a position that permits them to exert significant influence on general norms and laws, for example that they are in government in their respective states. Again, it is easy to see why this is important. If transnationally shared aims on the founding or constitutional revision of a supra-state order are to be realized, individual members of the partisan networks must also be capable of authoritatively shaping (national) laws and constitutions. Holding office may further be seen as warranting the (sociological) legitimacy of constituent agency: the weighty ambitions that those who seek to enact constituent power typically have arguably require public support, and having gained sufficient electoral strength to be in government may plausibly be interpreted as signaling that such support is available.
When these two conditions-agreeing on shared goals and having access to power-are satisfied, transnational partisan networks can provide a springboard for the cooperative implementation of shared projects across nation-states.
What does that involve exactly? One important thing transnational partisan coordination can ensure in this connection is that certain key decisions are taken more or less simultaneously in each state: for instance, that all members of a future or present supranational community enact a major constitutional change at roughly the same time, or agree to the core constitutional features of a new, supra-state political order.
The cross-border exchanges enabled by partisan networks may likewise facilitate coordinating strategic reactions to temporary political failures or crises, as well as to national and transnational dissent and resistance facing partisans. Reactive actions of this sort are instrumental to the realization of transnational or supranational political projects, and they fall squarely in the category of executive action, as they involve jointly taking authoritative steps aimed at keeping on track the implementation of a shared political project.
The Christian Democrats' transnational partisan networks were able to exercise this executive function mainly in the period after 1950, when Christian Democratic parties were the dominant political force in all six member states of the ECSC and EEC. Being connected to the relevant legislative and executive mechanisms in these states, the national parties were capable of realizing politically their shared supra-state project of building the constitutional foundation of "core Europe"-and their transnational network was a major enabling factor in these exercises of constituent power.
The first thing to note is that transnational party cooperation, notably, via the consultations and congresses that took place in the newly established NEI and the more informal Geneva Circle, promoted the homogenization of policy objectives among Christian Democrats and more generally allowed them "to develop their own peculiar notion of 'Europe'" (Gehler & Kaiser, 2001, p. 780). 4 The ideological consensus that emerged-most emphatically perhaps at the Sorrento congress in April 1950-was not only a distinctively anticommunist but also, to some extent at least, an antiliberal one: it not only opposed the a-religious materialism of Stalin's Russia, but also linked the horrors of the two world wars to the ostensibly liberal roots of modern nationalism (Acanfora, 2015;Gisch, 1990, pp. 480-484). At any rate, these shared ideological commitments also led Christian Democrats to converge on preferences about European integration-a convergence that, as one historian notes, was a "precondition for developing common ideas for the constitutionalisation of core Europe and for preparing concrete European policy-making" (Kaiser, 2007, p. 240).
The transnational deliberations that paved the way for this ideological convergence, insofar as they have been documented, may again be interpreted as reflecting the sort of two personae logic that the exercise of mixed constituent power demands. A highly structured way of preparing transnational exchanges made this possible. For example, prior to the NEI's 1954 Bruges congress, which culminated in passing the "Manifesto of Bruges," which outlined the Christian-Democratic project of a European common market, the Belgian Christian Social politician Robert Houben, in his capacity as rapporteur for the Bruges congress, developed a detailed questionnaire on issues of European economic integration that he sent out to each national branch of the Équipes (Gehler & Kaiser, 2001, p. 792). The answers national party representatives provided-qua citizens of European nation-states-in turn served as the basis for deliberations at the congress, where participants also addressed each other qua members of a future European political order that transcends individual nation-states.
How did transnationally organized Christian Democrats eventually bring their shared vision of Europe to bear on constitutional decisions, enacting constituent power? To be sure, the influence of transnational party networks remained mainly indirect. Though they undoubtedly "saw it as their role to influence the process of integration in the direction of their own philosophy and programme" (Gisch, 1990, p. 481), Christian Democrats never intervened directly via the NEI or the Geneva Circle in the negotiation and implementation of economic treaty clauses. Neither in connection with the ECSC treaty, nor in connection with the EEC treaty, did transnational party networks exercise unmediated power (Kaiser, 2007, p. 222).
But this does not mean that their influence was negligible. The opposite is the case. For this was a period where, as we saw, national governments and ministries were largely controlled by Christian Democrats, and so the transnationally agreed political commitments made their way into the intergovernmental negotiations without the need for the direct political interference of party networks. Not least because key Christian Democratic figures such as Konrad Adenauer at the top of their respective party hierarchies were at the forefront of transnational coordination, the transnationally agreed objectives could be powerfully channeled into intergovernmental decisions.
With all that in mind, we may turn now to the question of whether this model of transnational organization aimed at exercising supra-state constituent power satisfies the desiderata specified at the outset of this section. How does the transnational partisan account of constituent power fare in this respect?
So far, it seems clear that it can satisfy desideratum (1), which emphasizes that the supra-state constituent power in Europe ought to be a mixed form of constituent power, whose participants act both in their capacity as members of states and as members of a (present or future) supranational polity. The cross-border communication enabled by transnational partisan networks can indeed induce dialogue between the two Habermasian personae. There furthermore appears to be good reason to think that constituent power exercised via transnational partisan networks is in principle a feasible organizational model (thus satisfying desideratum 3a). It may be considered feasible not only because it has been powerfully demonstrated to be workable provided that the requisite political will is available. Also, and perhaps more importantly, as it obtains its norm-creating power and legitimization at the member-state level it does not demand a complex supra-national institutional structure to operate effectively. This leads to a third point: in relying on electoral support in national political arenas, the model also specifies from which channels its sociological legitimacy can be expected to flow (thus satisfying desideratum 3c).

Justifying and sustaining constituent power: The role of ideology
Two of the above-stipulated desiderata have not been addressed yet. These are desideratum (2)-which demands that an action-guiding model of constituent power must allow for participation and contestation-and desideratum (3b)-which highlights the fact that such a model will have to be underpinned by ideational foundations that help justify and sustain the collective pursuit of supra-state constituent power. I will take these in reverse order, speaking first to (3b). Before doing so, however, a more fundamental question needs handling. Why exactly does it matter that supra-state constituent power can rely on ideational foundations that support its justification and coordinated pursuit?
As far as justification is concerned, one might say that, insofar as supra-state constituent power is not exercised directly by citizens but indirectly via elected representatives, as it is the case in the model the present article concentrates on, it can claim to be exercised in a democratic fashion only if as its exercise can be justified to citizens (Böckenförde, 2006, p. 103). For justifications to be accepted, though, the reasons given for particular constitutional decisionsfor example concerning the shape a future supranational order will take-must be rendered meaningful and intelligible to citizens. Here it helps a great deal to be able to draw on ideational foundations-ideas, values, and rhetorical figures-that have resonance with a wider public because they form part of political ideologies that are deeply rooted in the relevant societies (see White & Ypi, 2016, Chapter 3). 5 As far as support for the continual exercise of constituent power is concerned, a common ideational substrate is important because it can motivate those who aim at enacting constituent power (and others) to want the intended constitutional change to happen, encouraging them to put special efforts and energies into realizing their ambitions over an extended time. Theorists of parties sometimes speak in this connection with the motivational benefits of partisanship, understood as the capacity of ideological commitments to help agents endure motivational and epistemic obstacles to the achievement of their political project (White & Ypi, 2016, pp. 85-96). Such resilience is indispensable when major acts of constitutional transformation are aimed at.
A model of constituent power that centers on partisanship appears naturally well suited to offer something in the way of ideational resources that facilitate the justification and sustained collective pursuit of constitutional founding or reform. For partisanship may be seen as a powerful underpinning not just to the sort of political commitment that reinforces the motivation of agents to act as constituent power, but also to the kind of justificatory activity necessary to endow the exercise of constituent power with democratic legitimacy (Rosenblum, 2008;White & Ypi, 2016). Ideology is not something to look to with skepticism when it comes to constituent power, in other words; rather, it should be regarded as providing socially rooted normative ideas that can be harnessed for the successful enactment of constitutional reform.
A minimum requirement that must be met in order for an ideology to provide a suitable ideational underpinning for supra-state constituent power is that it can be reframed in transnational terms, as transcending particular geographical settings rather than reflecting the viewpoint of specific local constituencies (White & Ypi, 2016, pp. 202-203). 6 Moreover, when it comes to supra-state constituent power in the EU, presumably the simple fact that a political ideology can be presented in transnational terms is not sufficient for it to serve the aforementioned supporting roles; it must also be linked to diagnostic and prescriptive claims about Europe in particular, the geographical setting whose transformation is the ultimate intention. This involves clarifying vis-à-vis citizens how reshaping the political constitution of Europe follows from specific transnational political commitments. Equally, it entails drawing, at least provisionally, lines of demarcation, delimiting where Europe begins and where it ends, and giving reasons for why the proposed boundaries are plausible. Without localizing transnational ideologies in this way, it is certainly difficult to imagine how the constitutional transformation of Europe could be justified to citizens.
A glance at the Christian Democratic model of supra-state constituent power outlined here helps to better understand what it means to transnationalize a political ideology, to localize that ideology in the context of Europe, as well as to employ it as a justificatory resource.
Above I have noted that an ideological consensus emerged among Christian Democrats in the postwar era, one that had not only anticommunist but also anti-liberal pretensions. This consensus was structured around views that go back to the advent of Christian Democracy as a political movement, most importantly the doctrine of personalism, which sees the human being as constantly embedded in groups and natural communities that are not necessarily coextensive with the nation-state (hence the opposition to the species of political liberalism that centers on powerful secular, and centralized nation-states), and the notion that Europe is a spiritual realm at the origin of which Christianity lies (hence the opposition to the materialist world view that characterizes communism and, to some extent at least, also liberalism) (Forlenza, 2017, pp. 268-269). It is easy to see how these ideas could ground a transnational vision for a unified Europe that is, at the same time, still local enough to appeal to shared historical understandings of what Europe is (Kalyvas & van Kersbergen, 2010, p. 196).
In fact, much as the twin ideas of personalism and Europe as a distinctively Christian realm supported the project of establishing a unified Europe with shared transnational or supranational political institutions, they also opened the door to very narrow conceptions of where the boundaries of Europe lie. 7 In particular, the notion of Europe as a spiritual space that is, at bottom, a Christian civilizational project often went hand in hand with idealizations of the premodern and pre-national Catholic West or Abendland. From this, it is only a short step to the kind of anti-Protestantism that conceives the boundaries of Europe as congruent with Europe's Catholic countries, not to mention the tendency to assert the primacy of Catholicism over non-Christian religions (Forlenza, 2017, pp. 272-273;Rosenboim, 2017, pp. 252-257).
Be that as it may, what cannot be denied is that the Christian Democratic rationale for an integrated Europe was ultimately a powerful one, and one that resonated sufficiently with mass audiences to serve the purposes of political justification. It is, of course, true that Christian Democratic politicians, even if they held this view, rarely publicly presented their plans of an integrated Europe in terms of creating a democratic Carolingian Empire that restores the natural order of European history. Nonetheless, the idea of Europe "as an ideal and moral fatherland understood in no way as in opposition to, but rather as a natural development of, the traditional [national] fatherland" was a recurring trope in the political rhetoric (and writing) of key Christian Democratic figures (Forlenza, 2017, pp. 276, 279). This trope could be usefully employed to justify the integration of nation-states into a larger supranational order that required delegating sovereignty to new European bodies.
The distinctive political ideas that undergird transnational Christian Democracy are complex, and so are the ways in which they have been put to use. The main point to note is that Christian Democratic ideology provided a potent ideational foundation for the exercise of constituent power aimed at creating a united Europe after World War II; the fact that "the core ideological concepts of Christian democratic politics seemed ready-made for European integration" (Kalyvas & van Kersbergen, 2010, p. 196) considerably facilitated the coordinated enactment and justification of the major constitutional changes the members of the transnational network of Christian Democrats intended. No doubt, then, desideratum (3b) is also satisfied by the Christian Democratic model of supra-state constituent power.

Constituent power without participation?
This leaves us with desideratum (2), which plays up the importance of popular participation and contestation. As discussed in the article's first section, a primary normative requirement that must be met by accounts of supra-state constituent power in the EU is that they include citizens directly in the exercise of constituent power; citizens ought to be given the opportunity to participate actively in constitution-making activities and be able to contest past constitutional arrangements on their own terms. Can the proposed strategy of grounding constituent power in transnational partisanship deliver on this concern?
One way of looking at the practice of transnational partisanship is to see it as almost inevitably elitist. Accordingly, it may well be that its component parts-national and local parties-are participatory, but the transnationally pursued coordination will tend to remain driven by political elites who possess the requisite cross-border contacts (as well as the linguistic abilities to communicate with their counterparts from other countries). If true, this would signal reason for the utmost pessimism on the capacity of the transnational partisanship-facilitated model of supra-state constituent power to satisfy desideratum (2). The unavoidably elite-centered nature of transnational partisanship precludes inclusive participation and contestation.
The assumption that elitism is intrinsic to transnational partisanship certainly finds confirmation in the Christian Democratic transnationalism that this article engages with. While the history of Christian Democracy may plausibly be interpreted as characterized by a gradual, if sometimes reluctant, espousal of mass democracy (see Müller, 2011, pp. 141-143), there is a very clear sense in which many of the key figures behind the constitutionalization of core Europe "espoused a top-down culture of public administration in which decision making was largely remote from the people," and "believed in supranationalism as something done by well-connected elites of high-minded planners and bureaucrats" (Forlenza, 2017, pp. 278-279). It is easy to see the secretive and elitist structure of the transnational associations we have encountered earlier-the NEI and the Geneva Circle-as giving institutional expression to this attitude. Citizens are thereby relegated to the status of passive recipients of constitutional choices, who, despite retaining the capacity to vote out of office the parties who collaboratively exercise constituent power, have close to zero influence on the decisions of elites.
Should we infer from this that the Christian Democratic model of supra-state constituent power fails to satisfy desideratum (2), and hence that the model is inadequate as a blueprint for constituent power in the EU? A modest qualification of this view would highlight that, even if citizens were largely excluded from participation, the parties whose elites associated transnationally with the aim of shaping Europe's constitutional order still enjoyed widespread electoral support domestically, in the six original member states of the ECSC and EEC. The acquiescence of mass publics would at least seem to weaken the concern that a failure to satisfy desideratum (2) disallows the model as a whole.
One might be tempted to go one step further and draw a normative conclusion from this observation: namely that desideratum (2) should generally be framed more loosely, being accommodated to circumstances where discretionary decision-making by political elites is widely considered legitimate by citizens. The intuition is that, while it cannot be denied that participation matters, it would be implausible to assert its ultimate importance in cases where constituent power is endowed with what I have called sociological legitimacy (3c). Is the way forward then to think of the satisfaction of desideratum (2) as in-principle desirable but not necessary? If so, we would have to say that there is nothing problematic about the elitist character of the Christian Democratic model of constituent power.
But such a position would be spurious. It is possible to accept that broad popular participation in acts of constituent power can be less important in some moments than in others, while retaining the view that a model of constituent power that lacks participatory channels altogether is insufficient. A loosening of desideratum (2)  In this light, it would seem that there are still good reasons to think that the way in which Christian Democrats have organized supra-state constituent power leaves much to be desired, despite the fact that their transnational mode of coordination has many other attractive features possibly worth emulating.

REVISIONS TO THE CHRISTIAN DEMOCRATIC MODEL
These initial reflections on the shortcomings of the Christian Democratic model of networked constituent power instructively reveal that some revisions to the model are required if it is to be normatively defensible and capable of providing a point of orientation for supra-state constituent power in the EU today. While the model has many attractive features, and has been demonstrated to be in principle feasible, it cannot be adopted without modifications.
In these final paragraphs, I want to suggest three necessary amendments. The first has already been evinced in the prior section: in contrast to the elitism of the Christian Democrats, contemporary efforts to harness transnational partisan networks for the exercise of constituent power beyond the state ought to involve citizens in a participatory fashion.
The normative argument for conceiving constituent power in participatory terms has already been rehearsed.
The main concern, to repeat, is that constitutional acts cannot plausibly be said to be the product of a decision by the people if the latter are merely the recipients of decisions that are made without their involvement. Another question is the practical one of how transnational partisan networks could be designed in a sufficiently participatory and inclusive fashion. There are no doubt serious obstacles in the way. Chief amongst these is the challenge of scope: it is difficult enough for a small number of political elites to coordinate effectively across borders; these difficulties are multiplied when larger numbers of citizens are involved, especially if most or all of them are supposed to have a say on decisions.
How could these hurdles be overcome? What is certain is that some democratic division of labor is unavoidable. Even the most inclusive and participatory model of transnational partisan coordination we may be able to imagine cannot continually include all its members in its decision procedures. The individuals involved are too geographically dispersed to ensure that all of them may have a seat at the table; too many potential voices will exist there to be convened in the same forum; and if ordinary citizens are to be involved in the process, most probably their time constraints will also constrain their participation. Yet even if mechanisms of delegation or representation will necessarily be part of transnational partisanship, there are still diverse ways of including citizens.
One would be to place the emphasis on national nodes of the partisan network as sites of popular participation, and link these nodes together in a democratic fashion. The local chapters of the participant parties could accordingly provide venues in which citizens engage in deliberations about the larger constitutional project they seek to realize and select representatives who stand up for them in the transnational fora where decisions are taken (Wolkenstein, 2016, p. 316). The organizational model adopted by the emerging transnational party DiEM25 may be interpreted along these lines: its aim is to connect the views and ideas of local activists and supporters to the transnational level via democratic representatives, relying both on participatory forms of party membership and intra-party accountability.
The second modification I propose is relevantly related to the first but cannot be assimilated to it. This is that, contrary to Christian Democratic practice, future transnational partisan projects that seek to shape the EU's constitution must be ready to submit actions to public approval.
As we have seen earlier, many Christian Democratic partisans who were engaged in the transnational project of constitutionalizing core Europe tended to be sceptical of mass politics. If one consequence of this was that their transnational groups remained very exclusive (think of the Geneva Circles), another was that they refused to submit their constitutional decisions to public approval. Relying on the acquiescence of mass publics, they proceeded in a mode described by one commentator as "emergency regime, with regular steps of constitutional significance taken by executive decision, legitimacy sought in securitising narratives, and dissent discredited as resurgent nationalism" (White, 2015, p. 312). Political justification was intermittently offered, but direct popular authorization was rarely sought.
Transnationally enacted constituent power, it seems, will have to be careful not to regress to a form of emergency politics if it is to be democratically defensible. While it is true that moments of constitutional change are exceptional moments where politics departs, to some extent at least, from the constraints of existing political and legal frameworks, agents of exceptional action can plausibly claim that their decisions were democratic only when they have obtained approval from a wider public. One straightforward way in which this may be achieved is by inviting the participation of citizens, as illustrated a few paragraphs earlier. But there are also other ways in which public approval might be secured.
Calling elections in the immediate aftermath of constitutional change is one possibility, but it is arguably strategically unwise if the aim is to realize an ambitious project of constitutional change. After all, as the capacity of transnational partisan networks to put their constitutional-political aims into effect depends on the electoral support their component parties have achieved nationally, and early elections at potentially polarized junctures jeopardize the longterm feasibility of the constitutional project as a whole. An equally risky alternative is the referendum, which likewise may destabilize partisan networks and eventually impair their ability to act as agents of constitutional change. Yet it is also possible to imagine institutional devices that are less prone to induce uncertainty, and perhaps even better suited for constitutional politics. The kind of constitutional crowdsourcing that has been tried out in Iceland would seem to provide a model that avoids the pitfalls of more plebiscitary forms of seeking public approval that I have just discussed (Landemore, 2015). An initial worry about this suggestion is that it smuggles substantive normative commitments into the theory of constituent power and thereby preempts the judgments of citizens as to what sort of constitution they prefer and what reasons they provide in support of it-a criticism analogous to that which I have leveled at accounts of constituent power that leave no space for citizen participation. But this is not my intention. My point is not that the Christian Democratic idea of Europe as discussed above is objectionable from the point of view of justice, but that it is arguably out of sync with the realities of contemporary Europe and hence impotent as an ideational underpinning for constituent power. The concern is pragmatic, not moral, in substance: widespread mobilization and effective justification seems difficult to achieve with a specifically Catholic story of Europe unity.
That this might be the case has certainly been sensed by many Christian Democratic parties in the last two decades or so. Many have "completely erased any reference, even perfunctory, to religion" in their political agendas (Kalyvas & van Kersbergen, 2010, p. 203), which has also informed their defence of an integrated Europe. Key Christian Democratic politicians today rarely invoke ideas of a spiritual "community of fate" when they speak of the EU; the dominant diction is rather one that asserts the lack of alternatives to EU integration (White, 2015). The task for future transnational partisan projects aimed at constitutional change is to develop an ideational substrate that is capable of challenging this understanding of Europe without relapsing into an analogous emergency rhetoric or an excessively narrow conception of Europe's boundaries. Current attempts to present constitutional change as the only means to secure the EU's survival-the strategy of DiEM25-would seem to fall short of these requirements. Even if effective on a short timescale, such moves risk excluding those who are uncertain as to whether survival is a goal worth pursuing in the first place, and may justify circumventing public approval when critical decisions are made. To be sustainable, transnational partisan projects must employ a broader, and more long-term vision of Europe.

CONCLUSION
Recent times have seen a spirited debate on constitutional politics in the EU, with many scholars and activists suggesting that the constitutional order of Europe is insufficiently politicized. The guiding thought is that the EU can claim to be genuinely democratic only if its citizens can be considered the source of its constitution, and the subjects with whom sovereignty ultimately rests. In search for a model of supra-state constituent power that could allow citizens to become shapers of the EU constitutional order, this article looked for inspiration to the transnational networks of Christian Democratic partisans, which played a central role in the emergence of the European Community in the postwar era. The Christian Democratic model has several shortcomings that require modification, and it is certainly a very demanding way of organizing constitutional agency, requiring as it does that political power can effectively be wielded by the parties to the transnational network. But it also signals that there are in principle feasible modes of organizing constituent power beyond the state that can be democratically legitimate. And if DiEM25's current efforts to create a transnational party are any indication, it seems plausible to conclude that transnational partisanship is not an outdated form of political agency, either.
To some readers, the suggested model of EU constitutional politics might seem excessively demanding. For not only requires it of those who seek to democratically reinvent the EU to organize transnationally, it demands that they mobilize support and win elections in many, possibly even a majority, of EU countries. Arguably this sets the bar very high, and makes bottom-up constitutional change driven by transnational partisanship look unlikely. But it is open to question whether constitutional reform initiated through transnational partisan coordination must take the exact same shape today that it did in the founding years of what eventually became the EU (meaning that parties of the same stripe hold office in all member countries). It is at least imaginable that, once a critical mass of citizens supports the idea of constitutional change, governing parties that were not initially in favor of that idea become willing to cooperate in transnational constitutional politics. And, of course, one might also reasonably ask whether there is anything wrong with models of constitutional change being demanding. That constitutional change should be easy is not a notion commonly entertained at the level of nation-states, and there is little reason to think that we should revise that intuition when it comes to exercises of supra-state constituent power.

ACKNOWLEDGEMENTS
For very helpful comments on a first draft I thank Markus Patberg, whose work has in large part inspired this arti- is part of a larger research project on transnational partisanship and democracy in the EU.

NOTES
1 Some would disagree that the method of rational reconstruction "relegates citizens to passive recipients of norms." After all, it seeks to uncover the presuppositions that underpin empirical practices from a participant perspective (e.g., Patberg, 2014, 2018). Yet, even if norms are formulated on the basis of an interpretation of citizens' "participant perspective," so long as these norms are considered valid independently of political practice it is difficult to see how citizens are anything other than passive recipients of norms. To be clear, this is not a blanket critique of rational reconstruction-a method that no doubt has many advantages-but a specific point about the importance of theorizing the link between norms and the political processes in which the specific meaning and implications of those norms are discussed and negotiated (on this point, see Ypi, 2012, Chapter 2).