Chapter One: Questioning the European Union’s “Democratic Deficit”

I. The different versions of the European Union’s “democratic deficit.”

The democratic characteristics of the European Union have been under scrutiny since its inception.[1] The famous phrase “democratic deficit” was coined by David Marquand[2] in 1979 to describe a simple idea: given the range and depth of the EU’s regulatory powers, the EU regulatory structure is insufficiently accountable from a democratic point of view.[3] The EU institutions themselves have raised such issues. The “classic” legitimating model is constructed around the paradigm of national parliamentary systems: the adoption and implementation of EU policies should be legitimized through the democratically elected European Parliament. Therefore, the “democratic deficit” depends on the role that the European Parliament plays in policymaking.[4] The European Commission has also acknowledged the public perception of the Union as “remote and at the same time too intrusive.”[5] It is telling that, even as it recommended wide consultation with civil society organizations, the Commission was quick to make clear that “it goes without saying that, first and foremost, the decision-making process in the EU is legitimised by the elected representatives of the European peoples.”[6] The European Council has also referred to the “democratic challenge facing Europe.”[7]

Similar formulations have made recurring appearances in the literature on the EU’s democratic deficit. For instance, it has been noted that “EC policymaking processes are largely dominated by bureaucracies and governments that provide little scope for parliamentary institutions (whether national parliaments or the EP) to intervene.”[8] According to a more recent and provocative account, the specter of “bureaucratic despotism” haunts Europe: “Behind the fig-leaf provided by the rhetoric of economic and political integration is to be found a major development, the rapid accumulation of power in Brussels,” in the process of which “democratic controls have suffered.”[9] If we carry through this argument, legislative acts that are increasingly produced at the EU level and affect the member-state legal orders are “tainted” with a lack of democratic legitimacy. However, this description overly relies on a model of legitimacy that is “uploaded” from national parliamentary systems onto the EU level. Hence, it ignores or underestimates other effects that EU law has on the national legal orders that might undercut the notion of the EU “democratic deficit.” Before proceeding to the bulk of this analysis, we need to briefly address two other versions of the EU democratic deficit.

According to the first version, the legitimacy of European integration is indirect: it proceeds from the democratic legitimacy of the member states.[10] The veto power of each member state in the Council was, in Joseph Weiler’s words, a key element for the legitimacy of Community decisions. Consequently, the shift from unanimity to majority voting in an increasing number of policy areas since the Single European Act and continuing in subsequent Treaty revisions—including the Lisbon Treaty—exacerbates the democratic deficit problem by weakening national parliamentary control of the Council.[11]

Three brief responses could be made to this argument. First, this account seems to ignore a fundamental reality of the institutional architecture of the European Union which includes both an intergovernmental component, where international features dominate (e.g., in the European Council, and the Council of Ministers), and a supranational component where supranational features are most salient (e.g., in the European Parliament, the Courts, and the Commission). The perception of the member states as the most important—albeit indirect—source of legitimacy is not sufficient to provide an adequate normative foundation for its supranational component,[12] which as we saw becomes increasingly salient as integration proceeds. Relatedly, it does not square with the foundational jurisprudence of the European Court of Justice on the autonomy of the Community legal order and cannot accommodate the function of EU law as creator and protector of rights of the citizens of the EU, even against the member states.[13] Second, the flipside of this argument would be that the veto power itself might contribute to the democratic deficit since it provides a small number of European citizens represented by their Minister in the Council with the ability to thwart the collective wishes of the rest of the Union.[14] Third, as will become apparent in subsequent chapters, the scheme in this book accommodates concerns about the role of member states in legitimizing the European project of regulatory integration by showcasing the importance of national regulatory authorities (“NRAs”). The literature has often neglected these national actors because of its narrower focus on more “traditional” bodies, such as national Ministers or Parliaments.

Another version of the democratic deficit emphasizes concerns about the future of the national welfare state in an increasingly integrated European market. The fear is that while the policymaking capacity of member states is being weakened, the EU itself lags behind member states in social protection.[15] In essence, these arguments do not pertain to the democratic deficit as such. They do not point to a lack of democratic processes, but rather to a lack of substantive policy outcomes linked to a specific agenda. Our concern is more about the former than it is about the latter. The idea of a democratic deficit in the EU denotes a failure of the institutional infrastructure to provide a framework of democratic contestation that would lead to the adoption of one or another substantive policy agenda. To argue otherwise seems to put the cart before the horse. Hence, what captures the core of this democratic deficit critique is the “classic” version of this claim: that is to say, EU policies are adopted and implemented without going through a legitimating process abiding by standards derived from analogous national institutions or, more abstractly, from the majoritarian, “Westminster” model of democracy.

II. Diverse notions of legitimacy.

One response to this broadly-held “classic” view of the democratic deficit was that what is perceived as a lack of democratic structures and processes within the EU, in contrast to those that prominently exist at the national level,[16] is, in fact, the flawed outcome of analyzing the EU against a democratic ideal that no modern government can meet.[17] As Andrew Moravcsik puts it, “the European ‘democratic deficit’ is a myth. Such criticisms rest on a vague understanding of what the ‘democratic deficit’ is, ignore concrete empirical data about whether one exists, and hold the EU to the impossible standard of an idealized conception of Westminsterian or ancient-style democracy.”[18] Moravcsik’s general “liberal intergovernmentalist” account of the EU reduces the latter to not much more than a traditional organization of sovereign states.[19] Thus, it downplays the supranational features of the EU, and is therefore subject to the same criticism discussed in Part I. However, if detached from his general theory about European integration, his specific assertion that the existence of a democratic deficit in the EU is overstated holds ground.

Indeed, the problem that accounts of the EU “democratic deficit” identify as distinctly “European” is not really so. Rather, it reflects a broader pattern of what could be called a “legislative democratic deficit” inherent in modern democratic systems, whereby easy associations between legislative performance and democratic preferences are viewed with skepticism.[20] The political science literature explains that the idea of “mandate representation”—according to which elections serve to identify a winning policy platform and select the politicians to be charged with implementing it fully—is at best fragile.[21] Moreover, incumbents make a series of decisions that cannot be placed on a single axis, are diverse, multi-layered and perhaps at times even internally inconsistent. In a purely majoritarian system, citizens have only one instrument to control policy, their vote. However, a vote in a general election is clearly insufficient to have an impact on the content of thousands of decisions that affect citizen interests.[22] Against this backdrop, we can better appreciate the remark that classic democratic deficit theory “generally overestimates the importance of legislatures in national systems [and] is inspired by a somewhat reductive view of democracy, which is implicitly equated with the mere voting of laws by parliamentary assemblies.”[23]

In assessing these accounts of a democratic deficit, in addition to the “decline of legislatures” thesis,[24] a version of which was analyzed above, we have to consider a parallel, and equally if not more important, related development: the rise of the administrative state.[25] These two developments are connected. As Wheare noted as early as 1968, “with a few important and striking exceptions, legislatures have declined in certain important respects and particularly in powers in relation to the executive government.”[26] Therefore, there is an even stronger case today that in the modern regulatory state the quality of the democratic process should factor in the administrative rulemaking process since administrative rules are increasingly important, in terms of both their volume and their scope.[27]

The traditional model of administrative law, premised upon an idealized perception of representative democracy, would conceive of the administrative agency as a mere “transmission belt” for implementing specific legislative directions in particular cases.[28] The “transmission belt” theory fails both empirically and normatively. Empirically, because the application of broad legislative directives frequently requires agencies “to reweigh and reconcile the often nebulous or conflicting policies behind the directives in the context of a particular factual situation with a particular constellation of affected interests.”[29] Normatively, because, beyond guiding principles, the complexity of modern problems requires policymakers to sift the changing scientific data in search of responsible regulatory solutions, for which parliaments have neither the time nor the expertise.[30] Therefore, we would want (and need) supplementary administrative policymaking. Furthermore, if the transmission belt theory were true, what would also be “transmitted” into administrative action would be the legitimacy inefficiencies of parliamentary institutions identified above. Last, even if the legislature exercises certain forms of oversight by means of parliamentary control (for example, hearings, commissions of inquiry) and its budgetary powers, this is far from ensuring the “transmission belt” ideal.[31]

A similar criticism would apply to a closely related construction which remains influential in European parliamentary systems. This is the “chain of legitimacy” idea. According to this view, administrative action is legitimate not only when (or because) it enacts specific legislative mandates, but because administrative institutions themselves draw their legitimacy from the public through this chain reaching down to the people. The picture is this: people elect their parliamentarians, who elect the Prime Minister, who selects her cabinet, which in turn appoints agency officials. This description should suffice to suggest its deficiencies. This chain of legitimacy seems to be too long, thus increasing the risk of an individual link breaking loose—in addition to the legitimacy lacunae associated with each individual stage. The picture becomes even more complicated if we take into account that many of these agencies are designed as independent authorities, thus challenging the very formal existence of this “chain of legitimacy” in the first place.[32]

The foregoing analysis confirms that exclusive reliance on the parliamentary democracy tradition in the EU context is misplaced:[33] it is misplaced both in overstating the role of parliamentary institutions and in underestimating the normative requirements for legitimate administrative action in modern democracies. But where does this leave us? The claim that the regulatory legitimacy problem exists at both the national and the European levels is valid,[34] yet it is still inadequate to fully address the democratic deficit question. Rejecting its uniqueness at the EU level offers a negative position, not a positive one. Hence, this book seeks to take the debate one step further and argue that the structure of the EU regulatory state may actually enhance certain aspects of the democratic accountability of national regulatory agencies at the member-state level through the creation of participatory channels for European stakeholders. Chapter Two will elaborate on how EU mandates may have had this effect; Chapters Three, Four and Five will then test the claim in three case studies. The focus of this chapter, however, is on a preliminary, and foundational, conceptual issue. That is, the democratic accountability-enhancing view of EU law presupposes a different model of accountability: one that does not rest upon the classic representative variant, the inefficiencies of which have already been pointed out. I shall call this the “deliberative-participatory model.”

Before analyzing the “deliberative-participatory” theory, we need to examine another model with increased appeal in the European context: the bureaucratic expertise model. According to this system, the regulatory state is legitimized because decisions are taken by an insulated body of independent experts. This “technical expertise” justification had found early support in the United States since the New Deal era.[35] The argument goes that, in today’s pluralistic world, mediating between competing groups of actors demands neutrality, and is facilitated by technical language that purports to apply in a relatively context-free manner.[36] This is probably the most salient normative justification for delegation to agencies. Delegation can reduce decision-making costs by helping overcome information asymmetries especially in complex technical areas of regulation. Legislators need expert information, but face both an increasing number of issues requiring their attention and limited resources. Agencies are therefore expected to develop and employ this expertise and thereby improve the quality of regulation thanks to their relative institutional competence.[37]

One of the most prominent supporters of the bureaucratic expertise model in Europe has been Giandomenico Majone who identifies two basic “logics” of delegation.[38] The first views delegation as a mechanism to reduce decision-making costs as described above. The second rationale is that delegation either to the regulatory branch in the domestic context or to supranational bodies, such as the Commission or the European Central Bank in the EU context, serves to enhance the credibility of policy commitments.[39] More specifically, the requirement of elections at regular intervals discourages the adoption of policies whose positive effects for the electorate will come after the next elections. Furthermore, a current majority cannot bind future majorities to long-term policies and, in fact, new governments emerging from elections can renege on policies that the previously elected officials had deemed optimal. The proposal is then to delegate these decisions to outside expert bodies that can choose and implement the optimal policies credibly and over time.[40]

However, even if we assume that these bureaucratic bodies act in a disinterested manner,[41] this should not suffice as a legitimating mechanism given that most purportedly “expert” decisions are in essence politically charged. It is indeed difficult to disentangle cleanly the technical from the political aspects of a regulatory question since many perplexing distributional questions are usually implicated in seemingly technical issues.[42] Furthermore, accounts favoring a purely technocratic model seem to assume that citizen preferences are either irrelevant or predetermined and exogenous to the political process. This view, however, seems to ignore the question that initiated this inquiry, namely the question of democratic legitimacy, thus failing to provide a satisfactory answer.

III. A deliberative-participatory model of legitimacy.

A. Habermas’s theory of deliberative democracy.

What is therefore required is an institutional design that acknowledges the reality of extensive administrative policymaking but self-consciously addresses the accountability challenges of the regulatory enterprise, and tries to bring citizens back into the policymaking process.[43] The model that arises as the most suitable alternative to evaluate the European regulatory integration is, I argue, a “deliberative-participatory democracy” model, a form of which was originally advocated by Jürgen Habermas. Under this model, the emphasis is shifted to political participation in deliberative processes:

civil society, which provides the social underpinning of autonomous publics, is as distinct from the economic system as it is from the public administration … [T]he integrative force of solidarity … should develop through widely expanded autonomous public spheres as well as through legally institutionalized procedures of democratic deliberation and decision making … The public opinion which is worked up via democratic procedures into communicative power cannot itself “rule” but can only channel the use of administrative power in specific directions.[44]

This theory of democracy ascribes a distinctive role to civil society, which is empowered to voice its opinions vis-à-vis the government. As Habermas puts it, “just those norms are valid to which all possibly affected persons could agree as participants in rational discourses.” The public use of communicative freedom and the discursively produced and intersubjectively shared beliefs have a motivating force that appears as a generator of power potentials. This influence generated by the communication circulating in the public sphere, however, “can be converted into political power only if it passes through the sluices of democratic procedure and penetrates the constitutionally organized political system in general.”[45] Habermas constructs a model of the circulation of power: “on the input side, influence generated in the public sphere is transformed through democratic procedures … into communicative power, which in turn is transformed through the legal programs and policies of parliamentary bodies into administrative power; at the output end, administrative programs create the necessary conditions for the existence of civil society, and hence of a vibrant political public sphere.”[46]

The most important feature of the model, in this context, is that public opinion is transformed into communicative power through a process that provides channels for political participation. Later chapters will argue that it is precisely these channels of public participation that EU law creates for national actors, thus providing them with access which enables them to influence regulatory decision-making at the member-state level. According to the Habermasian conception of democracy, the legitimating force of the democratic procedure can be found in the access to processes of such quality that facilitate publicly acceptable decisions.[47] His proceduralist paradigm of law, unlike the liberal and social-welfare models, does not favor a particular ideal of society or a particular political program.[48] Law’s addressees are also its authors, in the sense that they accept legislative outcomes even if they cannot endorse them substantively, because they consent to the lawmaking procedures.[49]

A caveat needs to be made at this point. There are two ways in which we could talk about a legitimating effect through civil society participation in the EU. The first one pertains to participatory processes at the EU level.[50] The second one refers to such processes established under EU mandates or influence at the member-state level. As already stated, my emphasis is on the second category, namely, on the processes legitimizing agency operations in the member states. Habermas himself does not claim that his theory depicts a current trend in the evolution of EU “central” governance toward this deliberative ideal.[51] And this would be a tenuous assertion given the current state of affairs. The claim here is much more moderate: my argument is that the Habermasian account could provide the basis for a new accountability model that I call deliberative-participatory. This model could accommodate and shed light on the accountability-enhancing effects of EU law with respect to specific agency operations at the member-state level.

B. The relationship of Habermas’s theory to other normative models of democracy.

It is useful to situate Habermas’s discourse theory of law and democracy in the broader context of other normative models of democracy. A first distinction would be between the republican and the deliberative model. Habermas describes the former as the system in which deliberation can rely “on a culturally established background consensus of the citizens, which is rejuvenated through the ritualistic reenactment of a republican founding act.”[52] Habermas accepts the basic premise of this model, that is, the idea of citizen communication and participation in politics, but emphasizes that his theory “does not make the success of deliberative politics depend on a collectively acting citizenry but on the institutionalization of corresponding procedures.”[53] In Habermas’s deliberative model there is no unified deliberating subject, concentrated in a single entity of the united citizens or their assembled representatives. Instead, popular sovereignty takes effect in the circulation of reasonably structured deliberations and decisions in the context of a decentered society.[54]

This perception of the citizenry not as an undifferentiated mass but as a network of voluntary associations participating in commonly accepted democratic processes brings us to a second necessary distinction between the deliberative and the pluralistic model. The pluralistic model, the leading exponent of which is Robert Dahl, recognizes that civil society is composed of a multiplicity of organizations with sets of preferences.[55] It is these organizations, and not the electorate in general, that exercise control over the government. The political system is reasonably open to these multiple interests if they feel strongly enough about an issue to mobilize pressure and campaign for their cause with the government.[56] Therefore, the responsiveness of the political branches to these mobilized interests will depend on the political, and electoral, power of the interest groups.[57] Dahl presented the advantages of this model in a way that reflects the Madisonian ideal: competing interests are represented in the multitude of interest groups that can balance each other out so that not one of them can dominate the political system.[58] The pluralistic model could be an even better justification of the bureaucratic state: agencies, which have fewer members than legislatures and regulate a narrower set of issues, may be the locus on which interest groups could focus more effectively in the course of promoting their policy preferences.[59]

Nevertheless, there are both empirical and normative objections to the pluralistic model. With respect to the former, an aspect of the pluralistic model is that interest groups can influence political decisions because of their ability to deliver votes. However, agency members are not directly elected, so the influence of organized interests would have to pass through the elected branches, which would in turn monitor and control the agencies. However, these oversight mechanisms do not always perform “the job pluralistic theory requires of them.”[60] Besides, this would assume that the behavior of legislators in the first place is dictated exclusively by their willingness to serve special interests, whereas public interest, constituent interest, and ideology also motivate legislative conduct.[61] Public choice scholars would then point to another way of powerful interest groups exerting direct influence on agency officials: even if the regulated entities cannot deliver votes to agency officials, they can still offer financial benefits and the prospects of lucrative positions in the private sector after the officials’ term in the agency is over.[62] However, this can be prevented through the adoption of conflict of interest rules and “revolving door” restrictions.[63] Furthermore, it has been argued that it is unlikely that regulators would make decisions hoping to improve their future employment: any secret deal between the regulator and the industry would not be enforceable, and the idea that an interest group would hire a formerly favorable administrator simply as a signaling device to present administrators who may be looking for a job later seems far-fetched.[64]

The pluralistic model of democracy could also arouse skepticism from a normative point of view. As Dahl observed, specific policies are instances of “minorities rule.”[65] The question poses itself: Which minorities rule? In other words, the pluralistic theory does not ensure in and of itself that all interests will be recognized. Some individuals may not find their interests represented by any politically active organization, or the organization that represents them might be systematically excluded from the governmental process.[66] If we subscribe to the theory that the groups that care the most and organize most intensively to change the status quo have the most influence on policymaking, then this is hardly reconcilable with the assumption or the hope that different groups will balance each other out. Instead, powerful industry groups enjoying the most resources will predominantly steer policymaking at the expense of other less organized and less “traditional” associations. This problem is likely to be exacerbated in countries with limited civil society capacity and in which new groups have not yet developed sufficiently to constitute an effective counterweight to the established interest groups. Assuming this “influence asymmetry” to be empirically true, the skepticism against the pluralistic theory further lies in that, as structured, this model seems to provide space for the normative accommodation of these asymmetries.[67]

Even if this discussion were to take on a different meaning, namely that these “minorities” are not privileged elite groups but less powerful sections of the society, the question of determining these groups would persist. One place to look at would be the famous footnote four of the Carolene Products case professing concern for the status of “discrete and insular minorities.”[68] However, as has been rightly noted, “discreteness and insularity” may be a source of bargaining advantage. Therefore, the concern should rather be protecting groups that possess the opposite characteristics, that is, groups that are “anonymous and diffuse” rather than “discrete and insular.”[69]

These “anonymous circuits of communication” are precisely within the framework of Habermas’s discourse theory. Democratic will formation is anchored in the informal streams of communication emerging from public spheres that are open to all political parties, associations, and citizens,[70] including “anonymous and diffuse” groups. In this context, we should distinguish the pluralistic theory as a normative model for democracy from social pluralism (or, better, plurality) as a sociological fact denoting a diversity of autonomous social organizations. The former, as noted previously, suggests that a pre-existing distribution of power among competing, more or less organized interest groups is, and perhaps even should be, mirrored at the policymaking level. The latter, instead, may be a precondition for a robust and diverse deliberating public sphere[71] in which, however, not just organized interests but also diffuse and anonymous groups and individual citizens can participate.

In addition to illuminating Habermas’s theory by contrast, the discussion about the pluralistic model serves another important function: it points to a challenge commonly facing both the pluralistic and the deliberative-participatory models. The deliberative-participatory account constructs an ideal picture of a rational discursive process in which the best argument wins. This alone does not suffice to ensure that this process is open and accessible to all groups and citizens—powerful and weak alike. However, the normative underpinnings of the deliberative-participatory model advocated in this chapter, contrary to the strand of the pluralistic model critiqued above, invite such broad participation. Whether this is the operational reality on the ground is a distinct issue that also pertains to questions of institutional design.[72] These questions underlie both the following section sketching the fundamental operative criteria of the deliberative-participatory model and the approach to the case studies in subsequent chapters.

Prior to the description of these operative criteria, a third distinction between Habermas’s theory and other related theories of democratic legitimacy through deliberation proposed in the EU context is in order. One of the most interesting theories is the “directly-deliberative polyarchy” model proposed by Cohen and Sabel, and Gerstenberg and Sabel:[73]

In a deliberative polyarchy local-, or, more exactly, lower-level actors (nation states or national peak organizations of various kinds within the EU; regions, provinces or sub-national associations within these, and so on down to the level of whatever kind of neighbourhood the problem in question makes relevant) are granted autonomy to experiment with solutions of their own devising within broadly defined areas of public policy. In return they furnish central or higher-level units with rich information regarding their goals as well as the progress they are making towards achieving them, and agree to respect in their actions framework rights of democratic procedure and substance as these are elaborated in the course of experimentation itself.[74]

This is an interesting model, certain aspects of which could map well onto the decentralized EU regulatory scheme described in Chapter Two. For the purposes of the analysis here, however, the following clarification is important. Gerstenberg and Sabel seem to prefer to situate deliberation outside the existing state apparatus to actors that are granted autonomy. Likewise, in a 1997 article co-authored by Cohen and Sabel, the role of administrative agencies is essentially reduced to providing the “infrastructure for information exchange” between units to facilitate benchmarking and continuous improvement.[75] It is no surprise then that they criticize Habermas’s claim that “communicative power cannot supply a substitute for the systematic inner logic of public bureaucracies. Rather, it achieves an impact on this logic ‘in a siege-like manner’”:[76] their argument is that in Habermas’s account “even the most radical extension of the public sphere would be of limited consequence precisely because the technical demands, to which administration, parliament, and party must in turn respond, set limits – but which ones? – to the direction that might issue from a more encompassing, unrestricted discussion among citizens.”[77]

Their “radical” response might look like an appealing normative account to some but it is of questionable feasibility. It certainly does not reflect the scope of the impact EU law has on NRA operations. What EU law does—as subsequent chapters will clarify—is employ state structures, in this context national administrative agencies, and mandate specific participatory processes with the aim of channeling public input and transforming popular participation into specific state-sanctioned policy outcomes. I will also argue that this may have long-term, even transformative, implications for the nature of administrative governance in the member states. This is not a small feat in itself: Habermas recognized that law—in this context we could say regulatory policy—can function as a popularly inaccessible administrative code. But, as he reconstructs it, when law is formulated within a civil society of free associations, through deliberating political bodies, it is the best way of translating popular will into public policy.[78] But to claim that this would create and multiply autonomous decision-making units outside the established public policymaking mechanisms is to claim too much.

This is the basic distinguishing criterion between the deliberative-participatory model endorsed here and the two last theories to be examined: Robert Cover’s work on “nomos and narrative” and the theory of “autopoiesis.” As the previous analysis suggested, Habermas claims that the communicative power of citizens influences the production of legitimate law; it is therefore “jurisgenerative.”[79] This formulation may, for some readers, invoke a connection with Robert Cover’s famous 1983 article on “jurisgenesis”: Cover argued that “the creation of legal meaning—‘jurisgenesis’—always takes place through an essentially cultural medium. Although the state is not necessarily the creator of legal meaning, the creative process is collective or social.”[80] He was interested first in the jurisgenerative processes of insular communities in their struggle to define and maintain the independence and authority of their “nomos,” their normative universe; he also described how in the workings of groups dedicated to radical transformations of constitutional meaning as it affects the application of state power, the law grew through common symbols and discourse, through shared narratives and interpretations.[81] Habermas was similarly interested in the role of voluntary associations and a robust public sphere outside of formal state institutions. However, a fundamental difference from Cover’s theory is already apparent in the language Habermas uses to describe this jurisgenerative power: “a jurisgenerative communicative power must underlie the administrative power of the government.”[82] The distinction is in the same vein as the one we drew from Gerstenberg and Sabel’s work. Habermas focuses on the interface between the public sphere and state power and examines how associations can influence the production of state law. In Cover’s scheme, on the contrary, state institutions, and precisely courts, are characteristically “jurispathic”: they usually destroy legal meaning.[83] Briefly put, “Habermas’s idea of jurisgenesis is thus decidedly more ‘statist’ than Cover’s.”[84]

Last, Habermas’s account should be distinguished from the “autopoiesis” theory, which is located in a broader social-theoretical framework, that of “systems theory.” The “autopoiesis” strand found a prominent exponent in Niklas Luhmann, who described modern societies as systems of communication that have become differentiated into various functional subsystems through a process of social evolution.[85] Systems are “autopoietic” in the sense that “the states of the system are exclusively determined by its own operations. The environment can eventually destroy the system, but it contributes neither operations nor structures.”[86] In another essay, Luhmann explained that “everything that is used as a unit by the system is produced as a unit by the system itself. . . . Autopoietic systems, then, are sovereign with respect to the constitution of identities and differences.”[87]

This system/environment relation and the division of politics and law into different, recursively closed systems is the focus of Habermas’s criticism against the autopoietic theory. Normatively, Habermas rejects Luhmann’s account on the grounds that law “drop[s] any further-reaching claim to legitimation. There is no output the legal system could deliver in the form of regulations: interventions in the environment are denied it. Nor is there any input the legal system could receive in the form of legitimation: even the political process, the public sphere, and political culture present environments whose language the legal system cannot understand.” This “autism” especially affects the political system. Furthermore, Habermas stresses that this assumed mutual indifference between law and other social subsystems does not correspond with empirically observed interdependencies.[88]

This account of autopoietic theory in Between Facts and Norms has been described as a caricature that exaggerates the distance between Habermas’s and Luhmann’s views in that it neglects another aspect of Luhmann’s theory: the idea of “cognitive openness” and “structural coupling.”[89] As Luhmann explains, systems are operatively closed but cognitively open, in the sense that communications within one system may, and ordinarily do, refer to events in the system’s environment. Moreover, the idea of “structural coupling” is that one system presupposes specific states or changes in another system and relies on them.[90] However, even this qualification of a strong version of self-referential autopoietic systems does not alter the fact that each system develops its own codes which do not admit of mutual translation.[91] A form of communication proper to one system is just “noise” to another system. The system “has to contend with perturbations, irritations, surprises, and disappointments channeled by its structural couplings.”[92] But this is precisely the point of differentiation. In a deliberative-participatory model the “voice” of societal actors should not be perceived as “noise” or “irritation,” but as an essential precondition for the development of legitimate law. In turn, the law opens up institutional opportunities and encourages voluntary associations to supply this influence.[93] This interface is at the core of accountable policymaking.

IV. From theory to practice: Bringing the “deliberative-participatory democracy” theory into the administrative sphere.

A. A question of scope: Is Habermas’s theory applicable to the administrative state?

Having presented the deliberative model of democracy in Habermas’s work and situated it in the landscape of normative models of democracy, we need to resolve what might seem a problem with the application of this model to the administrative state. The skeptic would say that it is a fundamental problem because it is one of scope of application. More specifically, one line of criticism of Habermas’s theory has pointed to its “very traditional emphasis on legislation.”[94] It is true that Habermas’s main concern in Between Facts and Norms is parliamentary lawmaking.[95] However, skeptics would mainly point to Habermas’s earlier two-volume work, The Theory of Communicative Action.[96] In that work, bureaucratic spheres are presented as “systems” whose internal operations, as well as their relations to their environment, are steered by “media” that circumvent the process of reaching understanding through communicative action.[97] In Habermas’s words, state and media-steered subsystems are “consolidated and objectified into norm-free structures. . . . [I]n modern societies, economic and bureaucratic spheres emerge in which social relations are regulated only via money and power. Norm-conformative attitudes and identity-forming social memberships are neither necessary nor possible in these spheres; they are made peripheral instead.”[98] However, if administrative operations are irreducible to mechanisms of communicative interaction and citizens cannot “meddle” with the “steering media” of administrative systems, if the public sphere is reduced in essence to contributing only “taxes” and “mass loyalty” to the administrative system in exchange for “organizational performances” and “binding decisions,”[99] then at first sight Habermas’s own language seems to defeat the project of applying his theory to the modern administrative state.

Nevertheless, this would be a hasty conclusion. Between Facts and Norms no longer views the administrative system in terms of “power as command.” Rather, in multiple places it highlights that communicative power is translated into administrative power through the medium of law.[100] To be more specific, using Bernard Peters’s model, Habermas depicts the circulation of power as a system of concentric rings.[101] At the “core” or the “center” of the system is represented each of the three branches of government: the parliamentary bodies, the judicial system and the “familiar institutional complexes of administration.” Beyond the center, an “inner periphery” develops out of various institutions equipped with rights of self-governance or with other kinds of oversight and lawmaking functions delegated by the state (e.g., universities, public insurance systems, professional associations, foundations, etc.). Last, the “outer periphery” encompasses the civil society.[102] Therefore, when communicative power reaches the core of the political system, it influences the administration as well. As Habermas puts it, “binding decisions, to be legitimate, must be steered by communication flows that start at the periphery and pass through the sluices of democratic and constitutional procedures situated at the entrance to the parliamentary complex or the courts (and, if necessary, at the exit of the implementing administration as well).”[103] Contrary to what critics suggest on this count,[104] this evolution from The Theory of Communicative Action to Between Facts and Norms should be welcomed. It should also be expected: as scholars pointing out these “inconsistencies” note, Between Facts and Norms is a very different project investigating the explicitly normative principles of the constitutional state.[105]

Nonetheless, the skeptic would probably still remain unsatisfied. She would point to the very formulation of the last passage quoted in the previous paragraph to highlight the asymmetry between the “parliamentary bodies” and the “implementing administration.” She would also quote in support the following: “the implementing administration [should be precluded from employing] the sorts of reasons that support legislative and judicial decision making. These normative reasons belong to a universe in which legislature and judiciary share the work of justifying and applying norms. An administration limited to pragmatic discourses must not change anything in this universe by its contributions; at the same time, it draws therefrom the normative premises that have to underlie its own empirically informed, purposive-rational decision making.”[106] Limiting the executive branch to pragmatic discourses and to “employing administrative power according to the law,”[107] the critic would continue, dries out the role of the citizens’ communicative power in the administrative sphere and puts us back to the starting position of the “transmission belt” theory, which both this chapter’s earlier analysis and Habermas himself[108] reject.

However, these objections do not preclude the application of Habermas’s deliberative model to the administrative state. What Habermas tries to do is offer a discourse-theoretical understanding of the separation of powers, which is in essence a functional account of the separation of powers. In this sense—namely, seen from a level of abstraction and “independently of any historical legal order and of any concrete form of institutionalization”[109]—the function of “administration” or, to make the distinction clearer, “implementation” does not coincide with the concrete form of the executive branch or the administrative apparatus. It would be fair to question the definitional clarity of terms such as “implementation”; however, Habermas makes clear that if the implementation of programmatic goals entails a further development of law, traditional legitimating structures would not suffice. Instead, “the logic of the separation of powers must then be realized in new structures, say, by setting up the corresponding forms of participation and communication or by introducing quasi-judicial and parliamentary procedures, procedures for compromise formation, and so on.”[110] Habermas is well aware that modern administrations take on tasks that entail much more than the technical implementation of a statute. He therefore advocates shifting the functional separation of powers into the administrative system itself and introducing new elements of participation and control. He also argues that standards of effectiveness should not substitute for standards for the legitimacy of administrative action.[111] Instead, the administration needs to address the normative questions before it in a rational way. In Habermas’s words,

[this] cannot be contained within the professional confines of a normatively neutral task fulfillment. Consequently, procedural law must be enlisted to build a legitimation filter into the decisional processes of an administration still oriented as much as ever toward efficiency. To this extent, my image of the democratically “besieged” fortress of the state apparatus was misleading. Insofar as the administration cannot refrain from appealing to normative reasons when it implements open legal programs, it should be able to carry out these steps of administrative lawmaking in forms of communication and according to procedures that satisfy the conditions of constitutional legitimacy.[112]

This passage should settle the issue of scope. The deliberative-participatory model can find its normative underpinnings in Habermas’s theory. However, it still needs an account of its fundamental operative elements in practice. Habermas’s response to this question is not conclusive. He supports a “democratization of the administration” but concludes that the specific form of innovations would result from the “interplay of institutional imagination and cautious experimentation.”[113] The following section will provide a first sketch of the building blocks of the “deliberative-participatory” model. The case studies in subsequent chapters will also revolve around these features.

B. The fundamental operative elements of the “deliberative-participatory democracy” model in the context of the administrative state.

When it comes to lawmaking processes in the context of legislative bodies, Habermas notes that they must be “regulated in the light of the discourse principle in such a way that the necessary communicative presuppositions of pragmatic, ethical, and moral discourses, on the one hand, and the conditions for fair bargaining, on the other, can be sufficiently fulfilled.”[114] As observed in the literature, in using the word “sufficiently,” Habermas acknowledges that the idealized version of rational discourse will never be fully realized within legislative bodies due to important limits, such as the majority rule, and the pressure for timely action.[115] Besides, establishing representative bodies for deliberation and decision-making is, according to Habermas, the natural outgrowth of the inability of all citizens to join in the shared exercise of such powers.[116] Indeed, with some notable exceptions,[117] citizens cannot ordinarily be directly involved in the lawmaking process.

This means, in turn, that administrative agencies might provide an effective alternative locus for more direct forms of citizen participation in policymaking. As Jerry Mashaw has noted, “[a]dministrators at least operate within a set of legal rules (administrative law) that keep them within their jurisdiction, require them to operate with a modicum of explanation and participation of the affected interests, police them for consistency, and protect them from the importuning of congressmen and others who would like to carry logrolling into the administrative process.”[118] Similarly, Mark Seidenfeld has remarked that the agencies’ place in government, their structure, and their policymaking procedures suggest that the administrative state holds the best promise for achieving the civic republican ideal of inclusive and deliberative lawmaking.[119]

The question then arises of what kind of administrative law we need to enact and support this deliberative-participatory ideal. I identify four basic building blocks: open access, transparency, reason giving, and judicial review.

1. Open Access.

The deliberative-participatory model would require as a minimum that when an agency decides to adopt a policy, it give all interested parties a reasonable and equal opportunity to submit their comments on that policy before it is enacted. This language raises a number of issues.

First, the notion of “interested parties” does not suggest that individual citizens or citizen groups need to prove that they are directly or potentially affected by the agency policy to be granted access and send in their comments. Their interest is assumed in that they took the time to share their opinions with the agency.

Second, a “reasonable” opportunity entails that citizens are given a reasonable time frame within which they can participate in the rulemaking process. For instance, a two-week window for comments on a multi-page draft rule in a complex technical regulatory domain would fail the test of reasonableness.

Third, the idea of “equal” opportunity for participation is more challenging to define. It implies that one group should not be granted more privileged access to policymaking processes over another. Nevertheless, equal opportunity for participation does not mean equal influence on the actual final policy outcome. The latter will depend on a number of factors, including the policy issue at stake, for instance, the type of represented interests and their proximity to the regulated area, or the quality of the comments submitted. These factors already suggest that industry groups, which generally possess more resources and organized regulatory departments, are in a better position to provide more and more sophisticated comments, and might thus have a greater impact on the content of the final rule. This is not by definition incompatible with the foundational premises of the deliberative-participatory model, whereby the “best argument should win” in an open and transparent discursive process.[120] However, it does raise concerns and suggests that, as a matter of institutional design, it would be desirable under the deliberative-participatory model that agencies actively reach out and elicit the opinion of less salient, yet potentially affected, groups. Similarly, the deliberative-participatory model would accommodate, and even invite, reforms aiming at addressing gaps in the level of sophistication of comments submitted by individual members of the public.[121] Last, as a matter of research design, the deliberative-participatory model also requires that we keep track of the extent of the influence coming from different groups in the case studies.

The “equal access” prong could also serve as the criterion against which to evaluate rulemaking procedures, such as “negotiated rulemaking” (also known as “regulatory negotiation”). In the United States, this process found its first extensive justification in an article by Philip Harter and was endorsed by Congress.[122] Agencies use this procedure to bring interested parties into the rule-drafting process at an early stage, prior to beginning a rulemaking proceeding under the Administrative Procedure Act (APA). In negotiated rulemaking, the agency, with the assistance of one or more neutral advisers known as “convenors,” assembles a committee of representatives of all affected interests to negotiate a proposed rule. The goal of the process is to reach consensus on a text that all parties can accept.[123] An extended discussion of the merits and disadvantages of this process is beyond the scope of this section.[124] However, it is useful to focus on one account describing negotiated rulemaking as negotiated compromise and arguing that it “fundamentally alters the dynamics of traditional administrative rulemaking from a search for the public interest, however imperfect that search may be, to a search for a consensus among private parties representing particular interests.”[125]

This process of compromise does not automatically fail the deliberative-participatory standard. Habermas himself suggested as much: in complex societies there are cases, he noted, when the proposed regulations touch on diverse interests without any generalizable interest or clear priority of one value being able to vindicate itself. In such cases, he observed, “there remains the alternative of bargaining, that is, negotiation between success-oriented parties who are willing to cooperate.” Nonetheless, he was quick to stress that the discourse principle should be brought to bear indirectly, namely, through procedures that regulate bargaining from the standpoint of fairness.[126] In this context, the “equal access” prong puts the emphasis on ensuring that the negotiating group represents all affected interests. But this is precisely where the challenge lies. For example, consumer federations purport to represent the interests of the consumer, but consumers are not a homogenous entity.[127] Similarly, it has been noted that even in the case of labor-management issues in the unionized sector, where the problem of representativeness does not appear to be of central importance, the ability of managers to represent shareholders may be questioned, only a small percentage of the workforce is unionized, and union leaders are not always good spokesmen for the rank and file. The problem is exacerbated in the case of negotiations about environmental pollution, in which “the diverse, geographically scattered individuals who breathe the air and drink the water cannot always be represented effectively by standard environmental groups.”[128]

To sum up, the “equal access” requirement mandates caution regarding the compliance of regulatory negotiation as a rulemaking technique with the deliberative-participatory model. In the US context, this caution is arguably unwarranted because the agency will ordinarily publish the draft rule coming out of the committee in a notice of proposed rulemaking, and at that point anyone can participate by submitting comments. However, in practice it will be difficult at that stage to amend significantly a draft rule based on the consensus achieved by the committee. At any rate, a negotiated rulemaking scheme in a different context not including a public consultation component would entail a clear risk of introducing through the back door elements of corporatism that run explicitly counter to the deliberative-participatory model.

Fourth, the access granted to citizens under the deliberative-participatory model aims at ensuring public input on a policy to be adopted. That is to say, the public is brought into this discourse as a group of citizens; the agency does not simply conduct a consumer satisfaction survey. This distinction is important especially in the public utilities sector during the last three decades. The “New Public Management” theory that originated in the UK but spread to other European countries (often, Chapter Two will suggest, via EU law) led to the liberalization of markets, opening them up to private actors, and to the privatization of previously state-run services: this shifted the emphasis from “citizens” to “customers” of public services.[129] However, this customer model has its limitations and underplays the democratic character of agency action.[130] This transformation of “citizens” into “customers” should not relax the interface between the administration and the citizens in the production of policy. Public surveys constitute useful tools for agencies to sense public sentiment but, as even their proponents acknowledge, they cannot replace public participation in rulemaking.[131] Chapter Five returns to these questions in more detail.

The fifth and last question to be considered in this subsection is: how much access is too much access? This question has also been posed in the terms of a tension between participation and deliberation,[132] a tension which, one could contend, might be obscured by our use of the term “deliberative-participatory” model. Participation is a precondition for deliberation: as the number of participants increases—and this might be the outcome of the agency reaching out to various groups itself—so does the possibility that more information and suggestions coming from more diverse sources will become available to the agency, thus potentially improving the quality of regulation.[133] However, it is argued in the literature that at some point increased participation stops being complementary and becomes detrimental to deliberation: a vast influx of comments may lead to information overload, encouraging poor analysis and superficial examination of alternatives.[134]

Although the numbers from our case studies do not suggest such a problem,[135] a general response is in order. We should distinguish between instances of mass participation in which the quality of the majority of comments is relatively poor and others in which the quality of the majority of comments is high. The latter, optimistic, case manifests the peak of the deliberative-participatory model: stakeholders are mobilized and contribute comments in high numbers and quality. This should not be a problem: if an agency is overwhelmed with input of high quality, this would suggest that the policy at stake has important implications and should be considered in depth before a final decision is made. However, the skeptics about the possibility of increased participation and deliberation being combined are concerned about the first scenario. A policy of differentiated treatment of comments may offer a solution in this case: a mass electronic postcard campaign should not be accorded the same weight as detailed comments that address the questions at hand in a sophisticated way and recommend solutions. This is by no means to say that mass comments should be disregarded.[136] However, a differentiated treatment is permissible since, as will be noted below, the deliberative-participatory model does not oblige the agency to respond to every individual comment but only the most important ones. This type of treatment is also confirmed empirically.[137]

2. Transparency.

If open access lies at the core of the deliberative-participatory model, transparency is its necessary precondition. This is why in the context of the European Union the right to transparency has been described as a second-generation participation right laying the ground for the emergence of the third-generation right of civil society participation in administrative policymaking.[138] Transparency should be required at every stage of the policymaking process. The notice of the agency’s intention to adopt a new policy and the draft rule itself must be publicized as widely as possible so that all potentially interested parties can take notice and participate in the process. At the end of the process, the agency must publish all the comments it has received (subject to restrictions for the protection of business secrets, restrictions that should nonetheless be narrowly construed) together with its justification for the final content of the rule pursuant to the third requirement analyzed right below. Publicizing the comments received allows interested parties and other actors, e.g., the media, to assess the influence that specific comments have had on the final rule. Moreover, this publicly available file can form the record on which citizens may base their application for judicial review of the published rule. Ideally, the administrative authority would publish the comments as soon as it receives them so that other parties can respond to them as well as to the original call from the agency, thus enhancing the deliberative nature of the enterprise.

3. Reason giving.

For the administrative policymaking process to be truly deliberative, the interaction between the agency and the citizens must be a two-way street. Administrators and citizens must be part of a give-and-take process in which citizens submit comments and administrators provide feedback to participants.[139] To ensure the latter, the deliberative-participatory model must introduce a reason-giving requirement so that agencies justify the reasons for their specific policy decisions.[140] The “deliberative” explanation could be added to other rationales proffered in the literature for the reason-giving obligation.[141]

The question that naturally arises is: what kind of reasons should the agency provide to comply with this requirement? In the US context, as Jerry Mashaw observes, a focus on an agency’s reasons for a regulation has become the hallmark of judicial review under the APA.[142] For instance, in one case the Supreme Court rejected the rescission of a rule because of the agency’s failure to consider amending its rule rather than rescinding it.[143] In another case a court of appeals struck down a regulation because the agency did not disclose the scientific data upon which it relied and did not find the facts on which it based its choice among the evidence in the record.[144] But does the agency have to respond to each individual comment it receives? The answer is no and is derived from balancing the deliberative and participatory aspects of the normative model that this chapter defends. An obligation to address every single comment submitted would place an unbearable logistical burden on the agency and would vindicate those maintaining that deliberation and participation are irreconcilable goals. The ideal of participation would suggest that it is not realistic to expect widespread participation and a detailed agency response to all the comments submitted. On the other side, however, the ideal of deliberation would require that the agency address the fundamental issues raised by a high number of comments or by fewer comments of higher quality.[145]

4. Judicial Review.

The role of courts in securing the deliberative-participatory promise has already emerged at previous points and is supported by scholars advocating a deliberative model of democracy. Habermas indicates that “the communicative and participatory rights that are constitutive for democratic opinion- and will-formation acquire a privileged position [in judicial review].”[146] Mark Seidenfeld’s civic republican model explicitly provides that the reviewing court’s proper function is to ensure that the agency has interpreted the statute in a deliberative manner: this means that the court must make sure that the agency permitted open discourse, responded to all significant comments, and explained adequately how its decision furthers the public interest.[147]

Judicial review of agency action along these lines might therefore facilitate the proper operation of the deliberative-participatory model. However, for purposes of analytic clarity we need to draw a distinction here. Judicial review could serve an instrumental purpose in supporting the operative elements identified above: open access, transparency, reason giving. Strictly speaking, it does not constitute in itself an independent conceptual building block of the normative model advocated in this chapter. Viewing judicial review as a sine qua non of the deliberative-participatory model would be unnecessarily (and perhaps even falsely from a theoretical standpoint) tied to the US experience of courts expanding their reach under the “arbitrary and capricious” standard.[148] In fact, even in the US literature there have been prominent voices challenging the tendency to view the value of notice-and-comment processes through the lens of their judicial vindication.[149]

Therefore, participation and judicial review should be unbundled in such a way as to accept that it is tenable that an administrative system complies with the deliberative-participatory ideal even when the three prongs of the model (open access, transparency, reason giving) are not judicially enforceable. This thesis acquires increased significance in the European context in which, as both the following section and subsequent chapters will point out, judicial review of administrative policymaking processes is, for the most part, not equally developed. An empirical question concerns the extent to which these systems meet the deliberative-participatory ideal notwithstanding the lack of stringent judicial review in the US fashion. The answer to this question will determine whether judicial review is operationally, albeit not definitionally, a requisite part of the deliberative-participatory model.[150]

V. European administrative law under siege. Is the United States a model?[151]

Administrative law in Europe is changing.[152] Subsequent chapters will argue that EU law is a major driving force for this transformation: more specifically, in the example of electronic communications, EU law is moving the domestic administrative systems in a direction closer to the deliberative-participatory model. If we look closer at the operative elements of this model identified earlier (open access, transparency, reason giving, judicial review), we will notice a striking similarity with the informal rulemaking process as organized by the US APA. Section 553 of the APA stipulates that, when making rules or regulations having binding effect on private parties, the agency must provide notice of its proposal, an opportunity for affected parties to comment, and “a concise general statement” of the basis and purpose of the rules. Furthermore, the APA empowers a reviewing court to hold unlawful and set aside agency action, findings, and conclusions found to be “arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.”[153] Does this suggest that the US APA model should be the operational model for the transformation of administrative law in European countries?

This question invokes the discussion about the US “adversarial legalism” model potentially spreading to Europe.[154] The theory of “adversarial legalism” describes American methods of policy implementation and dispute resolution as more adversarial and legalistic compared with the systems of other economically advanced countries. According to the same account, Americans more often rely on legal threats and lawsuits and US laws are generally more complicated and prescriptive, adjudication more costly, and penalties more severe. As to governance, compared with European democracies, regulatory decision-making in the United States entails more legal formalities, more interest group participation, and more aggressive judicial review.[155] Comparative law scholars have asked whether this “adversarial legalism” is spreading across Europe. The argument is that increases in economic liberalization and political fragmentation have undermined traditional, opaque, informal approaches to regulation and have generated functional pressures and political incentives to shift toward a more formal, transparent, adversarial legal style, similar to that found in the United States.[156]

My preliminary response would be that EU law, by moving the regulatory system closer to the deliberative-participatory model described in this chapter, seems to accept the basic premises of the “adversarial legalism” picture in terms of adopting formal, open, transparent and participatory regulatory processes. However, there is one fundamental part of the “adversarial legalism” description that does not capture fully the current state of affairs in all European countries; this is the lack of extensive litigation of the sort we witness in the United States. As I have already noted and will have the opportunity to explore further in the case studies, with the exception of some cases in the UK, European courts do not enforce routinely these participatory requirements to strike down measures on the basis, for example, of lack of effective consultation.[157] This does not ipso facto preclude the application of the deliberative-participatory model. Nor does it suggest that a more robust judicial review in this direction is unlikely to develop in Europe in the future. After all, the US APA introducing rulemaking procedures was passed in 1946, and it was not until a few decades later that US courts beefed up those provisions and imposed increasingly more obligations on US agencies with respect to participatory processes. However, what it does suggest is that a US “adversarial legalism”-style regulatory system is currently not in (full) operation in Europe.

To sum up, in the search for a model of democratic accountability, the account proposed in this chapter moves beyond the paradigms of the French republican tradition, or a utopian Westminster type of representative democracy, or a German hierarchical bureaucratic model of legitimacy through expertise in favor of an operational deliberative-participatory model that provides avenues for stakeholder input in administrative proceedings. From this perspective, we end up not just questioning the existence of the so-called “EU democratic deficit” in the specific context of regulatory policy, but actually arguing for a potential “democratic surplus” through the initiation, encouragement or formalization of public consultation procedures at the member-state level. Therefore, the emphasis of the following chapters is on this access to processes created under EU mandates or influence that may promote the deliberative-participatory ideal. The book elaborates on this idea by using the electronic communications sector as a case study in the example of France, Greece, and the UK. Chapter Two outlines the EU initiatives in the area of electronic communications with an emphasis on institutional and procedural mandates, and explains how these mandates may have an accountability-enhancing effect at the member-state level. The “democratic surplus” thesis is then examined in the context of three country cases: France (Chapter Three), Greece (Chapter Four), and the UK (Chapter Five). Chapter Six concludes by exploring what the country cases can teach us in terms of the direction that the decentralized EU regime should take in future reforms.

[1] See Stephen Sieberson, The Proposed European Union Constitution – Will it Eliminate the EU’s Democratic Deficit?, 10 Colum. J. Eur. L. 173, 174 (2004).

[2] David Marquand, Parliament for Europe 64-66 (1979).

[3] Damian Chalmers & Adam Tomkins, European Union Public Law: Text and Materials 65-66 (2007).

[4] See, e.g., the European Parliament Resolution on the democratic deficit, (1988) OJ C 187/229 of 18.7.1988.

[5] See Commission of the European Communities, European Governance: A White Paper, COM (2001) 428 final, Brussels, 25.7.2001, p. 3.

[6] Commission of the European Communities, Communication from the Commission: Towards a reinforced culture of consultation and dialogue – General principles and minimum standards for consultation of interested parties by the Commission, COM (2002) 704 final Brussels, 11.12.2002, p. 4.

[7] Presidency Conclusions, European Council Meeting in Laeken, 14 and 15 December 2001, SN 300/1/01 REV 1.

[8] Juliet Lodge, EC Policymaking: Institutional Considerations, in The European Community and the Challenge of the Future 26, 30 (Juliet Lodge ed., 1989).

[9] Larry Siedentop, Democracy in Europe 104, 119 (2001).

[10] See Giandomenico Majone, Europe’s ‘Democratic Deficit’: The Question of Standards, 4 European Law Journal 5, 6 (1998) (describing these “traditional arguments”). This account reflects the position of the German Constitutional Court, which, ruling on the constitutionality of the ratification of the Lisbon Treaty, wrote: “The empowerment to exercise supranational competences comes, however, from the Member States of such an institution. They therefore permanently remain the masters of the Treaties. In a functional sense, the source of Community authority, and of the European constitution that constitutes it, are the peoples of Europe with their democratic constitutions in their states.” (BVerfG, 2 BvE 2/08 Judgment of June 30, 2009, para. 231). Similar language had been used in the Brunner case, in which the German Constitutional Court ruled on the Maastricht Treaty (BVerfGE 89, 155, Judgment of October 12, 1993).

[11] J.H.H. Weiler, The Transformation of Europe, 100 Yale Law Journal 2403, 2473 (1991).

[12] Majone, supra note 10, at 12.

[13] See ECJ Case 26/62, Van Gend en Loos, Judgment of February 5, 1963; ECJ Case 6/64, Costa v. Enel, Judgment of July 15, 1964; Majone, supra note 10, at 12-13.

[14] Weiler himself acknowledges this point, supra note 11, at 2467.

[15] For a presentation of these arguments, see Majone, supra note 10, at 6-7, Oliver Gerstenberg, Law’s Polyarchy: A Comment on Cohen and Sabel, 3 European Law Journal 343, 347-348 (1997), Oliver Gerstenberg & Charles F. Sabel, Directly-Deliberative Polyarchy: An Institutional Ideal for Europe?, in Good Governance in Europe’s Integrated Market 289, 300 (Christian Joerges & Renaud Dehousse eds., 2002), Fritz W Scharpf, Democratic Policy in Europe, 2 European Law Journal 136 (1996).

[16] See Stephen Sieberson, The Treaty of Lisbon and its Impact on the European Union’s Democratic Deficit, 14 Colum. J. Eur. L. 445, 446 (2008).

[17] Andrew Moravcsik, In Defence of the “Democratic Deficit”: Reassessing Legitimacy in the European Union, 40 Journal of Common Market Studies 603, 605 (2002); see also Andrew Moravcsik, Despotism In Brussels? Misreading the European Union, 80 (3) Foreign Affairs May/June 2001 (vehemently critiquing Siedentop’s take on the question of democratic deficit).

[18] Andrew Moravcsik, The Myth of Europe’s “Democratic Deficit”, Intereconomics: Journal of European Economic Policy 331, 332 (November-December 2008).

[19] See Andrew Moravcsik, The Choice for Europe: Social Purpose and State Power from Messina to Maastricht (1998), describing European integration as “a series of rational adaptations by national leaders to constraints and opportunities stemming from the evolution of an interdependent world economy, the relative power of states in the international system, and the potential for international institutions to bolster the credibility of interstate commitments” (id. at 472) and concluding that the EC has been, for the most part, the deliberate creation of statesmen and citizens seeking to realize economic interests through traditional diplomatic means (id. at 501).

[20] See Susan Rose-Ackerman, Rethinking the Progressive Agenda. The Reform of the American Regulatory State 34-38 (1992). See also Yves Mény & Yves Surel, The Constitutive Ambiguity of Populism, in Democracies and the Populist Challenge 1, 7-11 (Yves Mény & Yves Surel eds., 2002).

[21] Bernard Manin, Adam Przeworski & Susan Stokes, Elections and Representation, in Democracy, Accountability, and Representation 29, 33 (Adam Przeworski, Susan C. Stokes, Bernard Manin eds. 1999) (noting that “mandate representation occurs when what politicians and voters want coincides or when politicians care only about winning elections, and to win they must promise and implement policies that are best for the public. But short of this happy coincidence, politicians may have incentives to … deviate from the mandate.” The authors explain that this might also be attributed to incumbents gratifying special interests that helped them get elected by financially supporting their campaigns). See also Susan C. Stokes, What Do Policy Switches Tell Us about Democracy?, in Democracy, Accountability, and Representation 98 (Adam Przeworski, Susan C. Stokes, Bernard Manin eds. 1999) (studying presidential elections in Latin America between 1982 and 1995, observing that politicians regularly deviated from their mandates, and noting that although violations of mandate are not inconsistent with representation narrowly construed, they should still raise alarms about the quality of democracies in which they are endemic).

[22] See Manin, Przeworski & Stokes, supra note 21, at 50. See also John Dunn, Situating Democratic Political Accountability, in Democracy, Accountability, and Representation 329, 336 (Adam Przeworski, Susan C. Stokes, Bernard Manin eds. 1999) (noting that “[a]t the very least, an analysis of how accountability can reasonably be hoped to work must take in a sophisticated division of political labor … and a far wider range of transactions than giving or witholding one’s vote, and winning or losing elective office.”).

[23] See Renaud Dehousse, Constitutional Reform in the European Community: Are there Alternatives to the Majoritarian Avenue?, 18 West European Politics 118, 122 (1998).

[24] See Valentine Herman & Juliet Lodge, The European Parliament and the “Decline of Legislatures” Thesis, 13 Journal of the Australasian Political Studies Association 10, 10-12 (1978) (noting a decline in Parliaments’ performance of their legislative, financial/budgetary, control and representative functions); Jean Blondel, Comparative Legislatures 5-7 (1973). This term was used by Lord Bryce (James Bryce, Modern Democracies 335 (vol. II, 1921)) to describe a disappointment with legislatures that had already appeared in the 19^th century works of A. Lawrence Lowell, Governments and Parties in Continental Europe (vols. I, II, 1896) and James Bryce, The American Commonwealth (vol. I, 1888).

[25] Over sixty years ago, Justice Jackson characterized the rise of the administrative state as “probably the most significant legal trend of the last century” (Federal Trade Comm’n v. Ruberoid Co., 343 U.S. 470, 487 (1952) (Jackson, J., dissenting)). The phrase “rise of the administrative state” can be found over 500 times in the literature (Jonathan R. Siegel, Law and Longitude, 84 Tul. L. Rev. 1, 57 n.285). For the rise of the administrative state in Europe, see Giandomenico Majone, The Rise of Statutory Regulation in Europe, in Regulating Europe 47 (Giandomenico Majone ed., 1996). Notably, the administrative state has been accorded constitutional status in the U.S. constitutional theory; see Bruce A. Ackerman, The Storrs Lectures: Discovering the Constitution, 93 Yale L. J. 1013, 1051-57 (1984) and Bruce Ackerman, We the People: Transformations 255ff. (1998). See also Cass R. Sunstein, Constitutionalism After the New Deal, 101 Harv. L. Rev. 421, 447-48 (1987) (agreeing that the New Deal “altered the constitutional system in ways so fundamental as to suggest that something akin to a constitutional amendment had taken place”). Contra Gary Lawson, The Rise and Rise of the Administrative State, 107 Harv. L. Rev. 1231, 1231-32 (1994) (declaring the post-New Deal administrative state unconstitutional, but acknowledging nonetheless that the essential features of the modern administrative state have been taken as “unchallengeable postulates by virtually all players in the legal and political worlds”).

[26] Kenneth C. Wheare, Legislatures 148 (2nd ed. 1968).

[27] See Susan Rose-Ackerman, American Administrative Law Under Siege: Is Germany a Model?, 107 Harv. L. Rev. 1279, 1295-96 (1994). See also Cornelius M. Kerwin, Rulemaking. How Government Agencies Write Law and Make Policy 13-20 (3d ed. 2003) (compiling data to document the “formidable volume” of rules). There are similar trends in other countries; see, e.g., the observation of a parliamentary committee in the UK that “[s]econdary legislation makes up the majority of the law of this country. When implemented it affects every sphere of activity” (House of Lords Merits of Statutory Instruments Committee, What happened next? A study of Post-Implementation Reviews of secondary legislation (HL 180 2008-09), [1]).

[28] See A. A. Berle, Jr., The Expansion of American Administrative Law, 30 Harv. L. Rev. 430, 431 (1917) (noting that “administrative law is the law applicable to the transmission of the will of the state, from its source to the point of its application”).

[29] Richard B. Stewart, The Reformation of American Administrative Law, 88 Harv. L. Rev. 1669, 1684 (1975). See also Jerry L. Mashaw, Due Process in the Administrative State 22 (1985) (noting that congressional statutes themselves require administrative agencies to make trade-offs between, for example, the need for public health or safety and the need for employment, product diversity, and a vibrant economy).

[30] Bruce Ackerman, The New Separation of Powers, 113 Harv. L. Rev. 633, 694 (2000).

[31] See Susan Rose-Ackerman, Regulation and Public Law in Comparative Perspective, 60 U. Toronto L.J. 519, 526 (“just as the legislature does not have the time, expertise, or foresight to write detailed statutes ex ante, so, too, it lacks the ability to provide comprehensive oversight ex post”).

[32] The various puzzles posed by independent agencies and their position in the constitutional edifice have been addressed in the literature on both sides of the Atlantic. See, e.g., in Europe, Mark Thatcher, Independent Regulatory Agencies and Elected Politicians in Europe, in Regulation through Agencies in the EU. A New Paradigm of European Governance 47 (Damien Geradin & Rodolphe Munoz & Nicolas Petit eds., 2005), Conseil d’Etat, Les autorités administratives indépendantes (Rapport Public 2001 – Jurisprudence et avis de 2000), Etudes & Documents No 52, at 251-462 (2001).  In the US literature, see Symposium: The Independence of Independent Agencies, 1988 Duke L. J. 215-99 (1988), Marshall J. Breger & Gary J. Edles, Established by Practice: The Theory and Operation of Independent Federal Agencies, 52 Admin. L. Rev. 1111 (2000), Martin Shapiro, The problems of independent agencies in the United States and the European Union, 4 Journal of European Public Policy 276 (1997).

[33] See Francesca Bignami, The Democratic Deficit in European Community Rulemaking: A Call for Notice and Comment in Comitology, 40 Harv. Int’l L.J. 451, 452 (1999). See also Peter Lindseth, Democratic Legitimacy and the Administrative Character of Supranationalism: The Example of the European Community, 99 Colum. L. Rev. 628, 684 (1999) (arguing that “parliamentary democratization largely misses the point, because the democratic deficit is really an accountability deficit at the subordinate technocratic level due to the lack of transparency and legally-enforceable participation rights in the regulatory process”).

[34] See Giandomenico Majone, Regulatory Legitimacy, in Regulating Europe 294, 294-295 (Giandomenico Majone ed., 1996). See also Yves Mény, De la démocratie en Europe: Old Concepts and New Challenges, 41 Journal of Common Market Studies 1, 9 (2002) (“there is certainly a case for emphasizing the weakness of popular input in European institutions, but the same kind of critique should also be addressed to national systems”).

[35] See James M. Landis, The Administrative Process 23-24 (1938) (noting that “[w]ith the rise of regulation, the need for expertness became dominant; for the art of regulating an industry requires knowledge of the details of its operation … Efficiency in the processes of governmental regulation is best served by the creation of more rather than less agencies. And it is efficiency that is the desperate need”).

[36] Bronwen Morgan, Technocratic v. Convivial Accountability, in Public Accountability. Designs, Dilemmas and Experiences 243, 246-48 (Michael W. Dowdle ed., 2006) (presenting this technocratic account, and later rejecting it as inadequate).

[37] See, e.g., Mark A. Pollack, Learning from the Americanists (Again): Theory and Method in the Study of Delegation, in The Politics of Delegation 200, 207-08 (Mark Thatcher & Alec Stone Sweet eds., 2003).

[38] Giandomenico Majone, Two Logics of Delegation: Agency and Fiduciary Relations in EU Governance, 2 European Union Politics 103 (2001).

[39] This is one of the different political reasons for delegation suggested in the literature in addition to the functional/prudential arguments. For instance, it has also been argued that delegation serves legislators’ self-interest in reelection because they can supply facilitation services to constituents in approaching the bureaucracy. (Morris P. Fiorina & Roger G. Noll, Voters, Legislators and Bureaucracy: Institutional Design in the Public Sector, 68 (2) American Economic Review, Papers and Proceedings 256 (1978)). Another reason for delegation—summarized as the blame-avoidance thesis—is the possibility of passing on hard choices for unpopular policies to agencies (Susan Rose-Ackerman, Introduction, in Economics of Administrative Law xiii, xv (Susan Rose-Ackerman ed., 2007)). Of course, the different rationales for delegation may overlap (Mark Thatcher & Alec Stone Sweet, Theory and Practice of Delegation to Non-Majoritarian Institutions, in The Politics of Delegation 1, 4 (Mark Thatcher & Alec Stone Sweet eds., 2003).

[40] See Giandomenico Majone, Two Logics of Delegation: Agency and Fiduciary Relations in EU Governance, 2 European Union Politics 103, 105ff. (2001), Giandomenico Majone, Theories of Regulation, in Regulating Europe 28, 40-44 (Giandomenico Majone ed., 1996), Majone, supra note 10, at 16ff.

[41] Which should not be taken for granted, see Renaud Dehousse, The legitimacy of European Governance: The need for a process-based approach, 1 Cahiers Européens de Sciences Po 1, 13-14 (2001).

[42] See Stewart, supra note 29, at 1684. See also Cass R. Sunstein, Factions, Self-Interest, and the APA: Four Lessons Since 1946, 72 Va. L. Rev. 271, 281 (1986) (noting that technical information will rarely be conclusive and its usefulness will often depend upon value judgments); Morgan, supra note 36, at 257 (noting “a certain disingenuousness about the evasion of the value-laden issues [through the use of expert language and the technical rationalization of the reasons for a decision]”). A similar response to Majone’s rather optimistic picture was that while regulatory policies are purely Pareto efficient, the current reality is rather different, for many such policies have identifiable winners and losers; see Andreas Follesdal & Simon Hix, Why There is a Democratic Deficit in the EU: A Response to Majone and Moravcsik, 44 Journal of Common Market Studies 533, 543 (2006).

[43] See Ackerman, supra note 30, at 694 (further suggesting that “[m]odern constitutions must take constructive steps to lay bare the crucial dimensions of normative bureaucratic judgment and to discipline its exercise by a host of techniques ranging from public participation to judicial oversight”). As subsequent chapters will show, our scheme is along these lines not just as a normative question of proper institutional design, but also as a descriptive account of the way in which the decentralized regulatory system works in the EU.

[44] Jürgen Habermas, Three Normative Models of Democracy, in Jürgen Habermas, The Inclusion of the Other. Studies in Political Theory 239, 249-250 (Ciaran Cronin & Pablo De Greiff eds., 1998).

[45] Jürgen Habermas, Between Facts and Norms. Contributions to a Discourse Theory of Law and Democracy 107, 147, 327 (William Rehg trans., MIT Press, 1996).

[46] Ciaran Cronin & Pablo De Greiff, Editors’ Introduction, in Jürgen Habermas, The Inclusion of the Other. Studies in Political Theory vii, xvii (Ciaran Cronin & Pablo De Greiff eds., 1998).

[47] See Erik Eriksen, A Comment on Schmalz-Bruns, in Debating the Democratic Legitimacy of the European Union 304, 306 (Beate Kohler-Koch & Berthold Rittberger eds., 2007).

[48] Habermas, supra note 45, at 445.

[49] Hugh Baxter, Habermas’s Discourse Theory of Law and Democracy, 50 Buff. L. Rev. 205, 288 (2002); see also Habermas, supra note 45, at 285 (“a consistent proceduralist understanding of the constitution relies on the intrinsically rational character of a democratic process that grounds the presumption of rational outcomes”).

[50] Recent trends at the EU administration level indicate that the EU, in particular, the European Commission, might be moving toward recognizing a right of civil society participation in European governance—Francesca Bignami calls this the third generation of participation rights before the Commission (Francesca Bignami, Three generations of participation rights before the European Commission, 68 Law and Contemp. Probs. 61, 72 (2004)). This was quite clearly indicated in the 2001 White Paper on Governance (supra note 5), in which participation figures prominently as one of the principles underpinning good governance in the Union. The Commission stresses that “the quality, relevance and effectiveness of EU policies depend on ensuring wide participation throughout the policy chain – from conception to implementation” (id., at 10). In 2002, the Commission issued the “Communication on Consultation,” which reiterates its commitment to encouraging “more involvement of interested parties through a more transparent consultation process” and provides general principles and standards for consultation that help the Commission to rationalize its consultation procedures (supra note 6, at 19-22). This emphasis on stakeholder participation is a recurring theme in Commission policy documents (see, e.g., the European Commission’s Third strategic review of Better Regulation in the European Union, Brussels, 28.1.2009, COM(2009) 15 final, p. 7). Related provisions were included in the rejected Constitutional Treaty (Article I-47 of the Constitutional Treaty on the “principle of participatory” democracy) and currently in the Treaty of Lisbon (Article 11 TEU). For an interesting account arguing that the scope of participation rights in EU administrative law is too narrow and should be extended to new situations and new types of procedures, in particular those that generally fall within the category of rulemaking, see Joana Mendes, Participation in EU Rule-Making: A Rights-Based Approach (2011).

[51] Habermas acknowledges the discussion about the “oft-bemoaned democratic deficit.” What he puts forth is, in essence, a normative account: he asserts that what is needed to promote a “European democracy” is the development of a European-networked civil society, a European-wide integrated public sphere in the ambit of a common political culture (Jürgen Habermas, Does Europe Need a Constitution? Response to Dieter Grimm, in Jürgen Habermas, The Inclusion of the Other. Studies in Political Theory 155, 159-161 (Ciaran Cronin & Pablo De Greiff eds., 1998)). In a more recent (and, one could say, more pragmatic) account, Habermas advocates a policy of graduated integration which would bind only member states in which citizens had voted in favor of a politically constituted Union (Jürgen Habermas, Europe: The Faltering Project 78 (Ciaran Cronin trans., Polity, 2009). Chapter Six revisits the question of civil society organization and participation at the EU level.

[52] Habermas, supra note 44, at 246.

[53] Id. at 248.

[54] Habermas, supra note 45, at 136. This could respond to Rubin’s criticism of deliberative democracy that “the image of civil society as a whole deliberating about some issue is an unproductive metaphor driven by the premodern image of democracy” (Edward L. Rubin, Getting Past Democracy, 149 U. Pa. L. Rev. 711, 749 (2001).

[55] See Nicholas R. Miller, Pluralism and Social Choice, in The Democracy Sourcebook 133, 133 (Robert A. Dahl, Ian Shapiro & José Antonio Cheibub eds., 2003) (“The fundamental postulates … are that (1) all societies are divided along one or more lines of fundamental conflict or cleavage that partition its members into different sets, and (2) the preferences of members of society, with respect to alternative public policies, are largely determined by the set to which those members belong”).

[56] John F. Manley, Neo-Pluralism: A Class Analysis of Pluralism I and Pluralism II, in The Democracy Sourcebook 381, 382 (Robert A. Dahl, Ian Shapiro & José Antonio Cheibub eds., 2003).

[57] See Robert A. Dahl, Pluralist democracy in the United States: Conflict and Consent 130-31 (1967).

[58] Id. at 24.

[59] For the pluralistic justification of the administrative state, see Mark Seidenfeld, A Civic Republican Justification for the Bureaucratic State, 105 Harv. L. Rev. 1511, 1521-23 (1992) (who argues, nonetheless, that the pluralistic democracy theory fails to fully justify the broad grants of policy-setting authority characterizing the present administrative state).

[60] Id. at 1524, 1523-26 (elaborating on the argument made in the text).

[61] See Rubin, supra note 54, at 745; Daniel A. Farber & Philip P. Frickey, The Jurisprudence of Public Choice, 65 Tex. L. Rev. 873, 894-901 (1987); Richard A. Posner, The Federal Courts: Crisis and Reform 265-67 (1985).

[62] See George J. Stigler, The Theory of Economic Regulation, 2 Bell J.Econ. & Management Sci. 3, 11-12 (1971).

[63] See Susan Rose-Ackerman, Corruption and Government. Causes, Consequences, and Reform 75-77 (1999).

[64] Steven P. Croley, Public Interested Regulation, 28 Fla. St. U. L. Rev. 7, 30 (2000) (noting additionally that “[i]t seems more likely that the future employment prospects of administrative regulators depend entirely on the regulators’ experiences with regulatory issues, not on particular decisions that were friendly to an interest group or groups. If so, regulated interests might well seek to hire those administrators who were most aggressive against them.… To that extent, future employment opportunities would make administrators less friendly towards groups who might hire them later—exactly the opposite of what the wisdom informed by “revolving door” imagines”).

[65] Robert A. Dahl, A Preface to Democratic Theory 128 (1956).

[66] Rubin, supra note 54, at 742-43.

[67] Cf. John F. Manley, Neo-Pluralism: A Class Analysis of Pluralism I and Pluralism II, in The Democracy Sourcebook 381, 391 (Robert A. Dahl, Ian Shapiro & José Antonio Cheibub eds., 2003) (noting that even updated accounts of pluralism –“Pluralism II”- are incompatible with substantive equality).

[68] United States v. Carolene Prods. Co., 304 U.S. 144, 152 n.4 (1938) (“whether prejudice against discrete and insular minorities may be a special condition, which tends seriously to curtail the operation of those political processes ordinarily to be relied upon to protect minorities, and which may call for a correspondingly more searching judicial inquiry”).

[69] Bruce A. Ackerman, Beyond Carolene Products, 98 Harv. L. Rev. 713, 723-24 (1985).

[70] Habermas, supra note 45, at 171 and 136 (referring to the “anonymous form” of popular sovereignty).

[71] Cf. Jürgen Habermas, On the Internal Relation Between the Rule of Law and Democracy, in The Inclusion of the Other. Studies in Political Theory 253, 259 (Ciaran Cronin & Pablo De Greiff eds., 1998) (referring to democratic procedure which provides legitimating force to the lawmaking process “in the context of social and ideological pluralism”).

[72] See, e.g., William E. Forbath, Habermas’s Constitution: A History, Guide, and Critique, 23 Law & Soc. Inquiry 969, 1000-01 (1998) (noting that forging and sustaining an institutional role for such associations in the deliberation and bargaining that attend policymaking frequently requires affirmative state intervention and support. In fact, “the broader, more diffuse, or more subjugated the potential ‘publics,’ the more pressing the need for state support,” id. at 1001).

[73] See Joshua Cohen & Charles Sabel, Directly-Deliberative Polyarchy, 3 European Law Journal 313 (1997) [hereinafter Cohen & Sabel, Directly-Deliberative]; Joshua Cohen & Charles F. Sabel, Global Democracy?, 37 NYU Journal of International Law and Politics 763 (2005); Oliver Gerstenberg & Charles F. Sabel, Directly-Deliberative Polyarchy: An Institutional Ideal for Europe?, in Good Governance in Europe’s Integrated Market 289 (Christian Joerges & Renaud Dehousse eds., 2002).

[74] Cohen & Sabel, Directly-Deliberative, supra note 73, at 291-92.

[75] Cohen & Sabel, Directly-Deliberative, supra note 73, at 335 (further noting that “[i]nstead of seeking to solve problems, the agencies see their task as reducing the costs of information faced by different problem-solvers: helping them to determine which deliberative bodies are similarly situated, what projects those bodies are pursuing, and what modifications of those projects might be needed under local conditions”).

[76] Jürgen Habermas, Further Reflections on the Public Sphere, in Habermas and the Public Sphere 421, 452 (Craig Calhoun ed., 1992).

[77] Cohen & Sabel, Directly-Deliberative, supra note 73, at 339.

[78] See John P. McCormick, Weber, Habermas, and Transformations of the European state: Constitutional, Social, and Supranational Democracy 216 (2007).

[79] Habermas, supra note 45, at 147 (employing the concept introduced dogmatically by Hannah Arendt).

[80] Robert M. Cover, Nomos and Narrative, 97 Harv. L. Rev. 4, 11 (1983).

[81] Id. at 25, 39-40.

[82] Habermas, supra note 45, at 147 (emphasis added).

[83] Cover, supra note 80, at 40.

[84] Baxter, supra note 49, at 266 n.276.

[85] Hugh Baxter, Autopoiesis and the “Relative Autonomy” of Law, 19 Cardozo L. Rev. 1987, 1993 (1988).

[86] Niklas Luhmann, Operational Closure and Structural Coupling: The Differentiation of the Legal System, 13 Cardozo L. Rev. 1419, 1424 (1992).

[87] Niklas Luhmann, Essays on Self-Reference 3 (1990).

[88] Habermas, supra note 45, at 51 and 335.

[89] Hugh Baxter, System and Lifeworld in Habermas’s Theory of Law, 23 Cardozo L. Rev. 473, 607-09 (2002).

[90] Luhmann, supra note 86, at 1432 (further explaining that “[w]alking presupposes the gravitational forces of the earth within very narrow limits, but gravitation does not contribute any steps to the movement of bodies. Communication presupposes awareness states of conscious systems, but conscious states cannot become social and do not enter the sequence of communicative operations as a part of them”).

[91] See Habermas, supra note 45, at 335. Cf. Marleen Brans & Stefan Rossbach, The Autopoiesis of Administrative Systems: Niklas Luhmann on Public Administration and Public Policy, 75 Public Administration 417, 428-29 (1997) (noting that in Luhmann’s view, “politics and “government” are functionally and analytically differentiated, and that the relationship between the two is not easily translatable into a particular institutional boundary).

[92] Luhmann, supra note 86, at 1432.

[93] Other theorists have tried to modify the autopoietic theory to account for intersystemic effects by introducing, for example, the idea of “interference” between law and society (Gunther Teubner, Law as an Autopoietic System 64-99 (Zenon Bankowski ed., Anne Bankowska & Ruth Adler trans., Blackwell 1993). Habermas refutes this attempt as inconsistent with the theoretical architecture Teubner desires on the grounds that it tacitly presupposes the kind of communication that systems theory must exclude (Habermas, supra note 45, at 52-56; see also William Rehg, Introduction, in Habermas, supra note 45, at xxiii).

[94] Niklas Luhmann, Quod Omnes Tangit. Remarks on Jürgen Habermas’s Legal Theory, in Habermas on Law and Democracy. Critical Exchanges 157, 166 (Michel Rosenfeld & Andrew Arato eds., 1998).

[95] That is, how public influence is transformed into communicative power and “enters through parliamentary debates into legitimate lawmaking” (Habermas, supra note 45, at 371).

[96] 1 Jürgen Habermas, The Theory of Communicative Action. Reason and the Rationalization of Society (Thomas McCarthy trans., Beacon Press, 1984), 2 Jürgen Habermas, The Theory of Communicative Action. Lifeworld and System: A Critique of Functionalist Reason (Thomas McCarthy trans., Beacon Press, 1987).

[97] Baxter, supra note 89, at 572.

[98] 2 Habermas, supra note 96, at 154.

[99] Baxter, supra note 89, at 588.

[100] Habermas, supra note 45, at 150 (further noting that “[a]dministrative power should not reproduce itself on its own terms but should only be permitted to regenerate from the conversion of communicative power”).

[101] For the analysis that follows in the text, see Habermas, supra note 45, at 354-56.

[102] Habermas explains that the “outer periphery” branches into “customers” and “suppliers.” “Customers” are groups that attempt to influence the political process more from the standpoint of particular interests; on the contrary, “suppliers” give voice to social problems, make broad demands, and articulate public interests or needs. Naturally, as Habermas acknowledges, the distinction is not a sharp one.

[103] Habermas, supra note 45, at 356.

[104] See Forbath, supra note 72, at 998 (noting that “Habermas violates his own injunctions. He sternly excludes the unruly norms and practices of democracy from the economic and administrative realms; then, at critical points, he smuggles them back in. The result is a deeply ambiguous and ambivalent account of the relationship between democratic principles, practices, and public spheres on the one hand, and the realms of the economy and public administration on the other”).

[105] See Baxter, supra note 89, at 570-71, 589-90.

[106] Habermas, supra note 45, at 192 (emphasis in the original).

[107] Id. at 188.

[108] Id. at 191.

[109] Id.

[110] Id. at 193; see also id. at 438 (noting that “[w]e can no longer focus on the abstract, general statute as though it were the sole support for the institutional separation of the legislative, adjudicative, and executive branches of government. Even during the so-called liberal period, the institutional separation of powers by no means fully coincided with the functional separation. To be sure, the differences emerged more clearly as the welfare state developed. In speaking of ‘legislature,’ ‘judiciary,’ and ‘administration’ in overly concrete terms, one disguises the logic of a functional separation of powers”).

[111] Id. at 391, 429 (also noting that legitimation problems cannot be reduced to the inefficiency of administrative steering).

[112] Id. at 440.

[113] Id. at 440. Forbath (supra note 72, at 998) has criticized Habermas for the “deep ambivalence” that runs through his works and impedes it from helping to answer some of the very questions Habermas hopes to illuminate, including the forms of public administration we need today.

[114] Habermas, supra note 45, at 171.

[115] Baxter, supra note 49, at 284.

[116] Habermas, supra note 45, at 170.

[117] See, e.g., Doctors for Life Int’l v. Speaker of the National Assembly, 2006 (12) BCLR 1399 (CC) (South Africa), in which the South African Court proffered an expansive reading of participation rights within the context of legislative lawmaking. The Court’s opinion also drew from international treaties and from the public law of other jurisdictions to argue for an enforceable right to public participation in lawmaking under the South African Constitution. See also, on this case, Athanasios Efstratios Psygkas, Revitalizing the “Liberty of the Ancients” Through Citizen Participation in the Legislative Process. Thoughts on Doctors for Life International v the Speaker of the National Assembly & Others, 5 Annuaire International des Droits de l’Homme/International Yearbook on Human Rights 719 (2010); Karen Syma Czapanskiy & Rashida Manjoo, The Right of Public Participation in the Law-making Process and the Role of Legislature in the Promotion of this Right, 19 Duke J. Comp. & Int’l L. 1 (2008).

[118] Jerry L. Mashaw, Prodelegation: Why Administrators Should Make Political Decisions, 1 J.L. Econ. & Org. 81, 99 (1985).

[119] Seidenfeld, supra note 59, at 1576. See also Robert B. Reich, Public Administration and Public Deliberation: An Interpretive Essay, 94 Yale L.J. 1617, 1631-40 (1985) (noting the potential of public administrators to foster public deliberation).

[120] Of course, the story would be different were these groups to use their resources in order to exert illegitimate influence on agency officials through under-the-table transactions.

[121] See, e.g., Mariano-Florentino Cuéllar, Rethinking Regulatory Democracy, 57 Admin. L. Rev. 411, 491-95 (2005).

[122] Philip Harter, Negotiating Regulations: A Cure for Malaise, 71 Geo. L.J. 1 (1982); Negotiated Rulemaking Act of 1990, 5 U.S.C. §§ 561-570.

[123] See Jeffrey S. Lubbers, A Guide to Federal Agency Rulemaking 212-17 (4th ed. 2006).

[124] See, e.g., Laura I. Langbein & Cornelius M. Kerwin, Regulatory Negotiation: Claims, Counter Claims and Empirical Evidence, 10 Journal of Public Administration Research and Theory 599 (2000) (studying the quality of the experience of the participants in both regulatory negotiations and conventional rulemaking and finding that on a wide range of criteria negotiated rulemaking received higher ratings than did conventional processes). But see Cary Coglianese, Assessing Consensus: The Promise and Performance of Negotiated Rulemaking, 46 Duke L.J. 1255 (1997) (arguing that, with respect to both saving time and eliminating litigation, negotiated rulemaking so far has not proven itself superior to informal rulemaking).

[125] William Funk, When Smoke Gets in Your Eyes: Regulatory Negotiation and the Public Interest-EPA’s Woodstove Standards, 18 Envtl. L. 55, 97 (1987) (referring to the case study of the EPA’s proposed rule to establish emission limitations for residential woodstoves under the Clean Air Act).

[126] Habermas, supra note 45, at 165-67 (adding that these procedures should “provide all the interested parties with an equal opportunity for pressure”).

[127] William Funk makes this point with respect to the role of the Consumer Federation of America (CFA) in the woodstove negotiation cited previously (supra note 125, at 95, noting that, the CFA “may have represented the interests associated with the mentality of a Consumers Reports reader, but it did not appear to lobby on behalf of poor, rural folk for whom the rule will provide little benefit and perhaps significant burden”).

[128] For both examples, see Susan Rose-Ackerman, Consensus Versus Incentives: A Skeptical Look at Regulatory Negotiation, 43 Duke L.J. 1206, 1210 (1994).

[129] A characteristic example is the “Citizens’ Charter” in the UK (The Citizen’s Charter: Raising the Standard, Cm 1599 (1991)), which uses the term “citizens” in its title but the core of its content pertains to consumer rights with respect to public services. The term “Citizens’ Charter” was for this reason described as a misnomer and the term “public customer’s charter” was suggested as more ‘fitting’ (Robin Hambleton & Paul Hoggett, Rethinking Consumerism in Public Service, 3 Consumer Policy Review 103, 111 (1993)).

[130] Cf. Morgan, supra note 36, at 258 (referring to the literature “excoriating the replacement of the citizen by the consumer as the prototypical ‘figure’ of political discourse” and noting that “[m]uch of this criticism has a neo-republican inflection, assuming that technocratic modes of accountability have the effect of marginalizing the active creation of community. Active political citizens are, it is feared, replaced by apolitical, passive consumers”).

[131] See Carol Harlow, Public Service, Market Ideology, and Citizenship, in Public Services and Citizenship in European Law. Public and Labour Law Perspectives 49, 54-56 (Mark Freedland & Silvana Sciarra eds., 1998) (noting that New Public Management techniques are not necessarily undemocratic because they are derived from the market, but acknowledging that policymaking involves choices which New Public Management sees as appropriate for management and inappropriate for citizen input); see also Lester W. Milbrath, Citizen Surveys as Citizen Participation Mechanisms, 17 Journal of Applied Behavioral Science 478, 489 (1981) (supporting the use of citizen surveys, but noting that they constitute a snapshot in time and are no substitute for the creative policymaking that can occur in good face-to-face discussion).

[132] See, e.g., Diana C. Mutz, Hearing the Other Side. Deliberative versus Participatory Democracy 17 (2006) (asking whether deliberation and participation can really be part and parcel of the same goal and noting that the prototypical deliberative encounter—a calm, rational exchange of views in near-monotone voices—might not convey the passionate enthusiasm potentially associated with political participation).

[133] Heterogeneity can be conducive to higher-quality deliberation; see, e.g., Cass R. Sunstein, Deliberative Trouble? Why Groups Go to Extremes, 110 Yale L. J. 71, 76 (2000).

[134] See Jim Rossi, Participation Run Amok: The Costs of Mass Participation for Deliberative Agency Decisionmaking, 92 Nw. U. L. Rev. 173, 211-41 (1997).

[135] See infra, Chapters Three, Four, and Five. More specifically, the data from public consultations in the three country cases during the period studied (i.e., from the inception of each regulatory agency until 2010) show that the average number of responses submitted per public consultation was 13.2 in France, 8.2 in Greece, and 22.35 in the UK.

[136] See Cuellar, supra note 121, at 485 (noting that a large number of such missives, notwithstanding their lack of sophistication, can signal to the agency’s political leadership the political cost of proceeding with a certain kind of regulation). See also Nina A. Mendelson, Rulemaking, Democracy, and Torrents of E-Mail, 79 Geo. Wash. L. Rev. 1343, 1380 (2011) (“We should strongly encourage agencies to engage comments on the value-laden questions more seriously, including the comments of lay persons submitted in large numbers”); Nina A. Mendelson, Should Mass Comments Count?, 2 Mich. J. Envtl & Admin. L. 173, 183 (2012) (arguing that large volumes of comments should at least “trigger an agency to engage in further deliberation and investigation. They should also prompt a brief response in the rulemaking documents. At a minimum, if agencies decide that the public comment game—at least for comments filed in large volumes by ordinary citizens—is not worth the candle, complete candor with the public is essential”). But see Cynthia Farina et al., Rulemaking vs. Democracy: Judging and Nudging Public Participation that Counts, 2 Mich. J. Envtl & Admin. L. 123, 130-45 (2012) (arguing, in particular, that the types of preferences expressed in mass comments “may be good enough for electoral democracy, but they are not good enough for rulemaking, even when rulemaking is heavily laden with value choices” (id. at 137), and further pointing to the unreliability of those comments as a gauge of citizen value preferences).

[137] See Cuellar, supra note 121, at 484 (“Differences in sophistication … might … affect who has a realistic chance of being part of the process shaping how agency staff use their limited discretion. Being deprived of sophistication shuts someone out of that discussion”).

[138] See Francesca Bignami, Three Generations of Participation Rights Before the European Commission, 68 Law & Contemp. Probs. 61, 68-72 (2004).

[139] Cf. Home Box Office, Inc. v. F.C.C., 567 F.2d 9, 35-36 (D.C. Cir. 1977) (“a dialogue is a two-way street: the opportunity to comment is meaningless unless the agency responds to significant points raised by the public”).

[140] Martin Shapiro characterized the judicial demand for more complete and persuasive explanation of agency decisions as a conversion of administrative law from a pluralist to a deliberative basis (Martin Shapiro, Who Guards the Guardians? Judicial Control of Administration 168 (1988) (“[t]here is a swing of the pendulum from synoptic adjudication to prudential deliberation for the agencies, and the only way courts can check on this kind of agency discretion is by themselves engaging in prudential deliberation”).

[141] See, e.g., Jerry L. Mashaw, Reasoned Administration: The European Union, the United States, and the Project of Democratic Governance, 76 Geo. Wash. L. Rev. 99 (2007) (explaining that a right to reasons has an instrumental purpose facilitating hierarchical, legal, and political accountability, but also has a particular moral force in a democratic polity recognizing the individual as the basic unit of social value that should be given reasons to affirm law as serving recognizable collective purposes).

[142] Id. at 110.

[143] Motor Vehicle Mfrs. Ass’n v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29 (1983). According to the court, “[n]ot having discussed the possibility, the agency submitted no reasons at all” (id. at 50).

[144] United States v. Nova Scotia Food Prods. Corp., 568 F.2d 240, 252-53 (2d Cir. 1977).

[145] See Cuellar, supra note 121, at 421 n.37 (suggesting in a similar vein that “the agency may not ignore qualitatively important dimensions of the problem raised in the course of the notice and comment process (i.e., by a cogently written comment, or by some substantial proportion of the reasonably intelligible comments in the aggregate.)”).

[146] Habermas, supra note 45, at 264.

[147] Seidenfeld, supra note 59, at 1547-49.

[148] See Patricia M. Wald, Judicial Review in Midpassage: The Uneasy Partnership Between Courts and Agencies Plays On, 32 Tulsa L.J. 221, 233-34 (1996) (“‘Arbitrary and capricious’ has turned out to be the catch-all label for attacks on the agency’s rationale, its completeness or logic, in cases where no misinterpretation of the statute, constitutional issues or lack of evidence in the record to support key findings is alleged. Frequently the arbitrary and capricious charge is grounded on the complaint that the agency has departed from its prior rationale in other cases without admitting it or explaining why. Sometimes the agency is rebuffed because it did not give adequate consideration to an alternative solution. But most often the court simply finds the agency’s explanation for what it is doing ‘inadequate.’” (internal citations omitted)).

[149] See Richard J. Pierce, Jr., Seven Ways to Deossify Agency Rulemaking, 47 Admin. L. Rev. 59, 86 (1995) (“I see the primary benefits of the notice and comment procedure as independent of judicial review. Agencies are more likely to make wise and well-informed policy decisions if they solicit, receive, and consider data and views from all citizens who are likely to be affected by a policy decision. . . . Indeed, I see judicial review of the resulting policy decision through application of the malleable duty to engage in reasoned decisionmaking as the source of net social costs rather than net social benefits”).

[150] This seems to be the general position advocated by Hermann Pünder, Democratic Legitimation of Delegated Legislation – A Comparative View on the American, British and German law, 58 Intl. & Comp. L.Q. 353, 375 (2009) (arguing that judicial control is “necessary for the legitimising effect of public participation and to prevent ‘agency capture’ by strong interest groups”).

[151] I reverse here the title of the article by Rose-Ackerman, supra note 27.

[152] See Matthias Ruffert, The Transformation of Administrative Law as a Transnational Methodological Project, in The Transformation of Administrative Law in Europe 3 (Matthias Ruffert ed., 2007).

[153] 5 U.S.C. §§ 503 (b)-(c), 706(2)(A).

[154] The term “adversarial legalism” appears in the works of Mirjan Damaska and Robert Kagan, see Mirjan R. Damas̆ka, The Faces of Justice and State Authority: A Comparative Approach to the Legal Process (1986), Robert A. Kagan, Adversarial Legalism. The American Way of Law (2001).

[155] Robert A. Kagan, Should Europe Worry About Adversarial Legalism?, 17 Oxford Journal of Legal Studies 165, 167-68 (1997).

[156] See R. Daniel Kelemen & Eric C. Sibbitt, The Globalization of American Law, 58 International Organization 103 (2004), R. Daniel Kelemen, Suing for Europe. Adversarial Legalism and European Governance, 39 Comparative Political Studies 101 (2006), and more recently R. Daniel Kelemen, Eurolegalism: The Transformation of Law and Regulation in the European Union 7ff. (2011) (arguing that European integration is encouraging the spread of a European variant of adversarial legalism that is “more restrained and sedate” than the American version. Kelemen calls this hybrid “Eurolegalism”).

[157] See, e.g., Rose-Ackerman, supra note 27, at 1294 (“[t]here is no judicial review of the adequacy of representation”); id. at 1300 (“[a]s a consequence of the anemic quality of judicial review, German executive-branch policymaking tends to lack formality and accountability. No judicially enforceable statute constrains administrative policymaking”).