Fulfilling High Hopes? The Legitimacy Potential of the European Citizens’ Initiative


 The European Citizens’ Initiative (ECI) constitutes the Union’s first genuine instrument of direct democracy. The first experiences with regard to its deployment have however proved disappointing, criticism of the instrument is mounting, and suggestions for reform abound. This raises the question whether the ECI holds any genuine potential to enhance the legitimacy of the EU - or whether it should at least in this regard be considered an ineffective, misguided experiment instead. The current paper provides a response by triangulating three types of data: legal insight and argument, empirical information with regard to the functioning of the instrument in the first five years of its operation, and a rational assessment of the procedure, now and in the future. This allows for a comprehensive reflection on the impact the ECI has had, and what impact it may yet have in the long or medium term, on the overall legitimacy of the EU.


Introduction
A core insight in European Union studies is that legitimacy constitutes an important requirement for the EU to be effective and durable. There exists a broad consensus that the EU suffers from a democratic deficit that affects its legitimacy.1 Especially after the rejection of the Constitutional Treaty by French and Dutch voters in 2005, both the European Commission and the European Council became fully aware of the need to take action. Accordingly, they sought to enhance the Union's legitimacy, primarily through strengthening the roles of the European Parliament and national parliaments in legislative decision-making.2 These investments in representative democracy are warranted by the fact that democracy arguably constitutes the central means for legitimising political power.3 In the EU context, legitimacy and democratic legitimacy have even been considered interchangeable concepts.4 In line with this observation, Beetham and Lord have analysed the EU's legitimacy from the vantage point of representative democracy.5 As asserted by Grimm however, the problem of democratic government is that the people are the source of all public authority, but cannot govern themselves. 6 The European Citizens' Initiative (ECI), as the Union's first 'experiment in direct democracy', could be seen as an alternative institutional vehicle for enhancing EU legitimacy, similar to instruments existing at the national level.7 In a notable break with the past, where they had, as such, no role to play in the decision-making trajectory, it allows citizens to invite the Commission to submit a legislative proposal on a matter they deem important.8 According to Commissioner Maroš Šefčovič, the ECI is to 'foster a European public space, widen the sphere of public debate across Europe and bring the EU closer to the concerns of the citizens'.9 In its first Report on the functioning of the instrument, the Commission underscored that the instrument 'aims to further strengthen the democratic legitimacy of the Union'.10 Whether the ECI can ever fulfil these high hopes, even in a modified form, remains to be seen. It initially triggered a modest, and inevitably speculative scholarly debate.11 At present, it has become possible to provide an empirical assessment of the key question: does the ECI indeed possess the potential to enhance the legitimacy of the EU, or should it be discarded as an ineffective, misguided experiment instead? To this effect, the current paper will provide a measured, contextual analysis of, first, how the instrument and its operationalisation holds up against criteria of democratic legitimacy and, secondly, how the Union's legitimacy has been impacted -or how it may yet be impacted in the near future.
The bare numbers hitherto are already quite impressive: a total of more than nine million citizens have supported an ECI, with three campaigns even collecting more than one million signatures each.12 Despite this impressive tally, the Commission, the Ombudsman and NGOs are actively monitoring the efficacy of the instrument, contemplating reforms to improve its accessibility and usefulness.13 Up until now however, few studies have focused squarely on its legitimacy-enhancing potential.14 The current paper aims to fill this gap, seeking to evaluate the relevant procedural characteristics, as well as the empirical evidence for its realisation. The analysis is a contextual one, in that it is based on a set of normative criteria developed in the legal and the political science literature on legitimacy, incorporating also relevant findings from the field of social psychology. We hereby triangulate three types of data: legal insights and argument; empirical information with regard to the functioning of the ECI and similar instruments in national states; and rational assessments of the procedures that have been established.
The paper consists of three parts. It starts by giving a succinct overview of the ECI's origins and procedural framework, presenting a useful reminder of the original intentions, and valuable background for our analysis (II). Next, we develop the normative criteria that inform our evaluation, arranged under the headings of input and throughput legitimacy (III). We subsequently apply those criteria to the operation of the European Citizens' Initiative (IV). While scrutinizing the legal arrangements, particular attention is devoted to how the Commission has interpreted the rules, and the consequences of these interpretations. Hereby, we draw from the experiences acquired during the first five years of its operation (2012)(2013)(2014)(2015)(2016)(2017). On the basis of these findings, we consider what influence the instrument has so far exerted, and may still exert in the long or medium term, on the overall legitimacy of the EU (V). The final section concludes (VI).

Conception and Development
The ECI's legal basis is contained in articles 10 and 11 of the Treaty on European Union (TEU), and Article 24 of the Treaty on the Functioning of the European Union (TFEU). Notably, Article 10 TEU holds that the functioning of the Union is predicated on the principle of representative democracy, and that every citizen has the right to participate in the EU's democratic life. Article 11(4) of the Treaty sets out the core concept, stipulating that '[n]ot less than one million citizens who are nationals of a significant number of Member States may take the initiative of inviting the European Commission, within the framework of its powers, to submit any appropriate proposal on matters where citizens consider that a legal act of the Union is required for the purpose of implementing the Treaties'.15 Article 24 TFEU calls upon the Parliament and the Council to adopt regulations detailing the further conditions and procedures, including the minimum number of Member States from which the citizens concerned should come. In December 2009, the Commission published a Green Paper, inviting comments from relevant stakeholders. During this consultation phase, several of the Commission's ideas on operationalisation of the ECI attracted immediate criticism. The main complaint was that the envisaged country threshold, requiring that signatories come from at least one-third of Member States, was too demanding.16 The draft of the first Regulation was tabled in March 2010. It sparked off a protracted period of negotiations between the Commission, Council and Parliament, striving to reach consensus on the instrument's practical modus operandi. Particularly contentious proved the eligibility criteria for signing an ECI, the timeframe for signature collection, and the methods of verifying their authenticity -issues that would also stir up trouble later, as further elaborated upon below. In November 2010, final agreement was attained, and the Regulation was officially adopted in February 2011. Additional patience was required though, as it only became fully operational from 1 April 2012 onwards.
The first official report on the functioning of the ECI, issued in 2015, listed a number of challenges in the implementation of the regulation, ranging from technical or logistical aspects to issues of a more political nature.17 It triggered a process of review, collecting ideas on ways to improve its implementation via consultations with, and contributions from various stakeholders and interlocutors, including advisory bodies and EU institutions. The Parliament also adopted a resolution calling for a substantial regulatory overhaul.18 In September 2017, the Commission adopted a draft for a recast instrument, followed in mid-2018 by the official second report on the ECI's functioning.19 At the time of writing, the eventual contents of an amended Regulation, and even the actual materialisation thereof, remain highly unsure.

Procedural Framework
The ECI procedure as currently laid down in Regulation 211/2011 consists of several steps, which deserve to be briefly mentioned. First, a so-called 'Citizens Committee' must be formed, consisting of at least seven natural persons holding the nationality of one of the EU Member States, eligible to vote in the EP elections. These persons should come from different Member States. The Citizens Committee then is to register the proposed Initiative with the Commission. It must specify inter alia the exact title, subject matter, underlying objectives, and received or projected sources of funding. If the proposal meets the criteria of Regulation 211/2011, the Commission shall indicate this in its response, to be provided within two months upon receipt, and proceed to officially register the ECI. At this stage, the limitative grounds of rejection are that the proposed Initiative falls manifestly outside the framework of the Commission's powers to submit a draft legal act for the purpose of implementing the Treaties; that it is manifestly abusive, frivolous or vexatious; or that it contravenes any of the Union's values as set out in Article 2 TEU. Should the Commission proceed to a rejection, it explicitly has to invoke one of these grounds in its response to the Citizens Committee. After registration of the proposal, the organisers can commence the collection of the individual statements of support.20 Only individuals holding the nationality of an EU Member State who are eligible to vote in EP elections may endorse an ECI. Crucially, the signatories should come from at least one quarter of the total number of EU Member States (currently: seven countries). Moreover, the number of signatories per country should at least comprise the specific minimum stipulated in Annex I of the Regulation. These thresholds mean to ensure a sufficiently broad backing for proposed Initiatives across the EU. In addition, heed should be taken of the timeframe for attaining the required minimum of one million signatories, namely twelve months maximum. At the end of this twelve-month circulation period, the organisers are to request the designated authorities in the pertinent Member States to verify and certify the statements of support. Hereafter, the proposed Initiative can be submitted to the Commission. Ordinarily, the latter will then proceed to publish it in the register of ECIs that have fulfilled all applicable thresholds. The organisers must be received 'at an appropriate level', and shall be granted the opportunity to present their Initiative at a public EP hearing.
Within three months after submission, the Commission has to communicate its legal and political conclusions, outline the action it intends to take, if any, along with the reasons for its decision. Should the organisers wish to appeal that decision, they may bring proceedings before the EU's General Court under Article 263 TFEU. Alternatively, in case of perceived maladministration, they may file a complaint with the European Ombudsman on the basis of Article 228 TFEU. By now, both avenues have been tried and tested, as further discussed below.

Antecedents of Legitimacy
As noted above, legitimacy can be defined as the perceived 'right to rule'.21 Arguably, this right to rule is seriously contested in the case of the EU.22 The core of the problem is that the Union is said to lack 'diffuse' democratic legitimacy: legitimacy of the political order as a whole, its key regime principles, or the political community.23 Such diffuse legitimacy is essential, because it functions as a generalized 'reservoir of support' for authorities.24 It consequently leads to a general willingness to obey rules beyond immediate self-interest, regardless of the perceived merits of these rules.25 The lack of this diffuse legitimacy in the EU is contended to lead to a necessity of obtaining more specific legitimacy, i.e. concerning a particular political decision or political actor, primarily in particular decision-making processes.26 Over time, various European integration scholars have sought to identify the antecedents of and solutions for the EU's democratic legitimacy gap. Surveying the relevant literature, both Wimmel and Schmidt have suggested three 'legitimacy variables' that may foster democratic legitimacy: input, output, and throughput.27 This typology, derived from the systemic view of legitimacy developed by Scharpf, has been fruitfully employed to analyse the legitimising potential of particular decision-making instruments.28 According to this systemic perspective, the first factor leading to legitimacy concerns participation, or the direct and indirect opportunities for citizens to influence policies.29 The ECI, given its character of 'citizeninitiated agenda-setting tool' should primarily be seen as an instrument for enhancing the EU's input legitimacy, as it allows citizens to directly channel their demands into the political system.  181-199, at 188-189, similarly distinguishes between three objects of legitimacy: the EU's political system, its institutions, and particular policy decisions 24 As argued by Beetham and Lord (supra n. 5), the EU also taps two additional sources of legitimacy, an indirect and a technocratic type. In the present analysis, we bracket both of these and focus on direct, i.e. democratic legitimacy. In our view, procedural legitimacy is an addition to input and output legitimacy. In addition, as follows from our focus on direct democracy, the indirect, parliamentary, and technocratic vectors are not relevant to our analysis. Also, the input-throughput-output distinction is fairly close to the four legitimising means suggested by Føllesdal (supra n. 3, 448-449): democratic rule, participation/ consent, and output. 28 Lenaerts  A second factor on which legitimacy has been argued to depend is throughput, or 'the procedural mechanisms of decision-making in the European multi-level system'.30 Throughput quality is argued to be an important requirement for building EU democratic legitimacy, since 'bad throughput (…) regularly undermines public perceptions of the legitimacy of EU governance, regardless of how extensive the input (…)'.31 This view equally transpires from the social psychology literature about legitimacy, which has shown that the perceived procedural fairness in specific decision-making processes is an important predictor of compliance across a great variety of social settings, amongst which law-making.32 Even if the ECI should be primarily understood as an instrument on the input side of politics, the quality of the processes that follow citizens' inputs also seem an important requirement for the ECI to affect legitimacy.
A third factor affecting legitimacy is output.33 Legitimacy in this perspective derives from the quality of the results of a political system -notably effectiveness of EU policy outcomes for the people.34 Output legitimacy, however, seems not very relevant to our analysis of the ECI's legitimising potential, because the instrument only provides for a potential intermediate output: a legislative proposal by the Commission, which enters the regular legislative and decision-making process.
Our analysis will hence revolve around input and throughput legitimacy. In their general form though, these variables do not suffice for our evaluative purposes; they must be detailed into specific, inter-subjective yardsticks that can guide our analysis.35 In order to do so, we combine the most recent legal and political science insights on the notion with more specific literature on representative and direct democracy.36

Conceptualising Input Legitimacy
Input legitimacy refers to the 'participatory quality of the process leading to laws and rules as ensured by the "majoritarian" institutions of electoral representation'.37 The main standards under this heading have been derived from representative democracy literature, some of which are also pertinent to our analysis.38 A first specific standard is equal access, e.g. the question whether all citizens are equal before the law.39 In the case of the ECI, equal access would imply that citizens in various Member States face the same conditions for tabling an Initiative.40 The argument here is basically pluralist in nature: to what extent do various groups in society have equal access to a political system -in this case to the process of agendasetting by the Commission, which despite the ECI retains its unique right of initiative? As such, the ECIbeing an instrument of direct democracy -can be seen as a further check on the centrality of a technocratic actor -the Commission -in the Union's legislative process, complementing the growing importance of the European Council in EU agenda setting.
In addition, input legitimacy requires several constructive preconditions, particularly the presence of a thick collective identity, and a European demos.41 These preconditions are crucial for a democratic 30 Wimmel,supra n. 23,191. 31 Schmidt, supra n. 27, 3. 32 Tyler, supra n. 25. 33 Wimmel, supra n. 23, 191. 34 Schmidt, supra n. 27, 2; Scharpf, supra n. 22, 11. 35 A.I. Applbaum, 'Culture, Identity, and Legitimacy', in J.S. Nye and J.D. Donahue (eds.), Governance in a Globalizing World (Brookings Institute, 2000), 319-329, 325. 36 It must be noted that at this point, we cannot make firm judgments about the relative importance, sufficiency/necessity and mutual interdependencies of these criteria. We bracket these complicated questions, as these have not received unequivocal answers in the underlying normative literature on EU legitimacy; providing or testing these answers would go beyond the applied nature of our analysis. . 40 A second standard for representative democracy, competition, in the sense of regular, free, and fair electoral competition, is not relevant for our purposes, although it must be noted that competition is relevant for referenda. 41 Scharpf, supra n. 22.
Brought to you by | Radboud University Nijmegen Authenticated Download Date | 1/4/19 3:22 PM system to function properly, because they attenuate the threatening character of majority rule.42 Arguably, these important preconditions are absent in the EU; 'a sense of European identity and loyalty is embryonic at best'.43 This absence of identity leads to the importance of alternative legitimising mechanisms. Lord and Beetham stress the importance of 'legitimation by constraint': the idea that legitimacy is more likely when power is credibly limited, and majorities restrained, for instance through separation of powers or use of super-majorities.44 At the same time, the challenge is not to compromise policy performance, which carries a Janus-like quality.45 Given the fact that instruments of direct democracy harbour the risk of use by popular majorities seeking to reduce minority rights, and possibly even a risk of capture by minority special interests, an important standard seems to be that, in line with pluralist theory, the demands inserted into the political system should be representative of the general consensus which defines the limits of political actions and the framework of policy outcomes46-our second evaluative standard.47 This criterion is meant to balance direct democratic legitimacy, flowing straight from EU citizens to the EU decision making machinery, with indirect democratic legitimacy, which in the EU is guaranteed by its quasi-federal character, providing democratically mandated and scrutinized national governments a central position in decision-making.
A first way of preventing unrepresentative demands from being put on the agenda through the ECI lies in the legality of the popular demands put on the political agenda.48 Since all primary law at least has been derived from the unanimous preferences of the governments representing the European peoples, this is the closest we get to guaranteeing that politics reflects the wishes of the majority of Europeans. Therefore, for the ECI to be legitimate, so-called qualifiers limiting the range of topics that may be covered by the proposal are required.49 A second way of preventing truly unrepresentative demands is by ensuring broad support of the demands inserted into the political system through the ECI. This can be done by requiring a high signature quorum for Citizens' Initiatives, also ensuring a broad geographical basis.50 Third, the specific risk of interest group capture, which is argued to be high for instruments of direct democracy, should be prevented.51 Mainly so as not to bias usage, citizens should be able to bear the financial costs of registering their initiative and obtaining signatures. Finally, checks and balances in the EU decisionmaking process may function to correct any small minority positions channelled into the political system. Given the ECI's character of instrument for agenda influence, the regular direct and indirect democratic checks and balances the EU has institutionalized, are relevant. The Initiative is nothing more than a right to place a particular topic on the agenda, subjected to further decision-making, with the Parliament and the Council providing a further democratic check on the process.

Conceptualizing Throughput Legitimacy
Throughput legitimacy concerns the internal processes and practices of governance, or what goes on in the 'black box' after an input into the EU policy-making is provided.52 Throughput legitimacy implies 'the perception of those addressed by a rule or a rule-making institution that the rule or institution has come into being and operates in accordance with generally accepted principles of right process'.53 Social psychologists have demonstrated that procedural legitimacy is an important predictor of compliance across a great variety of social settings.54 Regarding the ECI, two sub-processes within the Commission are relevant: decision-making about registration and follow-up to an Initiative by the Commission.
The legitimacy of such processes is predicated on three specific standards. A first throughput standard is accountability, or the extent to which decision-makers are responsive to input demands, and can be held responsible for their decisions.55 We view accountability as a mechanism consisting of three consecutive steps: 1) the actor is obliged to inform the forum about his or her conduct -in this case, the way it dealt with a Citizens' Initiative; 2) the forum must be able to interrogate the actor about its conduct; and 3) it must be possible for the forum to pass judgment about that conduct, backed by the possibility of imposing a sanction.56 Accountability can operate either horizontally or vertically.57 In our analysis, the former refers to other EU institutions, the latter to citizens directly. A second throughput standard is transparency of the governance process: citizens must have access to information about the decision-making process and particular decisions.58 A third throughput standard is legality, so as to prevent the abuse of power by key decision-makers.59 This translates into the requirement that the Commission behaves in line with the procedural registration and decision-making requirements contained in Regulation 211/2011.
In sum, the empirical analysis will be based on seven specific criteria, listed in table 1 below. We now proceed to our evaluation on that basis, applying the respective criteria to the operation of the ECI.

Analysis -Applying the Legitimacy Criteria to the ECI's Operation
In the first five years of the ECI's operation, 48 Initiatives were admitted and started with the collection of the necessary signatures.60 22 registration requests were refused for falling outside the ambit of the Commission's powers -a delicate issue we will return to below. By the end of 2017, four Initiatives have managed to cross both the signature and the geographical threshold, namely Right2Water; One of Us, Stop Vivisection and Ban Glyphosate. The topics of the various Initiatives are highly diverse, ranging from the recognition of water as a human right, safeguarding media pluralism, to expanding the Erasmus programme. In what follows, we apply our legitimacy criteria to the ECI procedure and its actual operation during the first quinquennat.

The ECI's Operation and Input Legitimacy
Criterion 1. Equal access. The legal framework stipulates that particular requirements to sign an ECI (ID number, address, date of birth, etc.) are set by national governments. As a result, significant variation between the Member States exists. In Table 2, we list the requirements up until January 2018.61 In September 2013, the Commission updated Annexes II and III of the Regulation, thereby amending several requirements. Six countries implemented changes (Luxembourg, the Netherlands, Spain, France, Ireland and Slovenia). Luxembourg and Ireland relaxed their conditions. Conversely, France moved from a moderately strict to a strict position, joining countries such as Austria and Italy which both have the maximum score of six out of six possible requirements. In March 2015, a second round of changes was announced. Luxembourg added two requirements, while Bulgaria removed two (see Table 2).
Many civil society groups push for abolishing ID demands specifically, deeming these the most important hurdle.62 That claim can at present not be assessed with full accuracy. A first explorative evaluation can be conducted though, as the four successful Initiatives, Right2Water, One of Us, Stop Vivisection and Ban Glyphosate, published detailed figures with regard to the collected signatures on their respective websites. The latter three experienced no problems reaching the thresholds: they ticked the box in 21, 20 and 22 countries respectively. The Right2Water Initiative is more interesting as it struggled to get enough signatures in at least seven countries. The relevant numbers show the minimum threshold was crossed in 5 of the 9 countries without ID-requirements (55.5%), but only in 8 of the 19 countries with ID requirements (42.1%). Admittedly, this campaign constitutes a single case only. Facts and figures on a more extensive range are needed to gauge the requirements' true impact. Indeed, alternative explanations can be thought of: some Member States have a longer tradition with national agenda initiatives than others, and one may even simply expect that some topics are relevant and salient in certain Member States, but not in others. Therefore, more successful ECI cases or more detailed data at the Member State level will be needed to assess this matter thoroughly, before watertight conclusions can be drawn.
The broader point remains however that a certain amount of inequality between the Member States exists, in the extent to which the ECI provides them with access to the Commission's legislative agenda. In reality, this inequality may even be more extensive, as successful use of the ECI also presumes the presence of sufficient resources and adequate organisation. Positive follow-up by the Commission is, moreover, likely to be aided by the presence of useful connections with the Commission, EP and other institutional actors -qualities that are not equally distributed across individuals and the groups that represent them.
Criterion 2: Qualifiers. It matters a great deal which topics an Initiative may concern. Not all conceivable topics can be proposed. An ECI may only deal with issues concerning a policy area where the EU 60 Eight were withdrawn by the organisers and not re-submitted. 61 No alterations were made since then. 62 cf Berg and Thomson, supra n. 14.
Brought to you by | Radboud University Nijmegen Authenticated Download Date | 1/4/19 3:22 PM has competence; and the Commission must have the power there to propose the adoption of a legal act. Moreover, as flagged above, it must not be manifestly abusive, frivolous or vexatious; or contravene any of the Union's values contained in Article 2 TEU. These provisions clearly serve to position the ECI squarely within the existing indirect democratic mandate underpinning European integration.
Yet, as Altman notes, such qualifiers can be an important 'entry hurdle'.63 What is crucial is how the Commission interprets its powers. So far, it has done so in a remarkably restrictive way.64 For example, 63 Altman, supra n. 50, 19. 64 In the same sense J. Organ, 'Decommissioning Direct Democracy? A Critical Analysis of Commission Decision-making on while the TEU and the ECI Regulation do not explicitly rule out that it can call for Treaty amendments, the Commission elected to deny that it possesses competence in this regard. Already in the first two years, 17 (42.5%) proposals were not admitted, allegedly for lying beyond the Commission's powers. While the percentage fell to 30% by the end of 2017, it continues to underscore one main point: the Commission clearly possesses an agenda-controlling power here, and the exercise is not a simple technocratic one. 65 In the conclusion of an own-initiative inquiry, the European Ombudsman has issued a series of critical recommendations which did not lead to meaningful alterations.66 Some of the committees that saw their proposals rejected instigated judicial procedures in response. In the early cases litigated, the Court of Justice proved unwilling to overturn the Commission's decisions.67 While two more recent rulings suggest that annulment is not impossible, the judiciary continues to shy away from a meticulous scrutiny of the underlying reasoning.68 Thus, the practice overall suggests that the Commission adopted a rigid approach which is not so easy to correct, unless the institutional remedies and prerequisites are recalibrated radically.
Criterion 3: The signature quorum. As mentioned, an ECI is successful if it attracts one million valid signatures. This may seem a high number, but given that the EU has 508 million inhabitants, the threshold is actually very low: less than 0.2% of the population needs to sign in order for an Initiative to pass. To indicate just how low this number is, comparatively speaking, consider the following.69 Of the 28 EU Member States, 12 have agenda initiative provisions. Across the 28, there exist no less than 23 different national procedures, as some Initiatives on e.g. constitutional matters have higher signature quorums than e.g. those on ordinary legislative matters. Of these 23 procedures, just two feature a signature quorum lower than the ECI's, namely the Italian agenda initiative (0.08% of the population) and the British e-petitions (0.16% of the population). Obviously, whether a quorum is high or low also depends on the circulation period: if the period to collect signatures is very short, even a relatively low quorum may be hard to reach. In 16 of the 23 procedures the circulation period is six months or lower. The ECI can, instead, circulate for 12 months, which is again rather permissive. In sum, the formal signature quorum can actually be considered relatively low.
One could retort that the ECI also imposes a geographical condition, which makes it more difficult to pass the signature quorum. At present, a successful ECI requires a minimum number of valid signatures from at least seven Member States. Geographical requirements are not unique to the ECI, but also known to e.g. Romanian agenda initiatives and Swiss and Australian citizen-initiated referendums.70 They function as safeguards against the tyranny of the majority;71 yet in the present case, they actually seem to be a safeguard against the tyranny of the minority in the sense of one large country providing the required signatures. In practice, the geographical requirement can make the procedure rather cumbersome. Organisers typically collect 25% extra signatures in order to ensure a campaign has enough valid signatures. One of the successful Initiatives, the Right2Water Initiative, struggled to pass the threshold in enough Member States (see Appendix 1). Indeed, on 22 April 2013, the organisers had already collected 1,465,489 signatu-res, but only passed the minimum requirement in five Member States. The campaign collected another 400,000 before enough Member States were covered.72 Early on, the Right2water campaign had mainly been successful in Germany and Austria. Indeed, on 22 April, 82.1% of the total number of signatures came from Germany, and another 4.5% came from Austria. As such, one can argue that the Initiative was heavily skewed, especially towards German signatures. However, when the signature collection period ended on 3 September 2013, the proportion of German signatures had declined to 74.8%, and the Austrian to 3.5%. It was still skewed, but the organisers had realized the need to campaign outside Germany and Austria, attracting broader support for their cause.
Criterion 4. Interest group capture. To avoid interest group capture, all donations exceeding EUR 500 must be registered. Appendix 1 provides an overview of the donations of the first 15 Initiatives. Strikingly, though it was originally expected that a successful ECI would cost one euro per signature, even the successful Initiatives attracted at best EUR 100,000 in funding. The case of the (withdrawn) Initiative EU Directive on Dairy Cow Welfare is instructive. The Initiative collected no less than EUR 345,567 in funding -of which EUR 90,934 was donated by Ben & Jerry's -but was finally withdrawn. Ben & Jerry's Social Mission Manager offered as the rationale for the withdrawal that at the time the instrument was a waste of 'valuable time and resources'.73 Her concerns mainly referred to the inadequate signature collection process; but the mere fact that the ECI was not reintroduced once the collection process had been improved, shows that organisations may feel there are more efficient ways to influence EU decision-making. This comes as no surprise; after all, even when an Initiative is successful, this by no means guarantees that action is taken by the Commission. This extra check constitutes a second important safeguard against interest group capture. In sum, '[w]hen the ECI Regulation was being developed, many feared that powerful interests would use the ECI for their own ends. (...) These fears have proven groundless'.74 Overall then, several criteria for input legitimacy are met by the ECI framework relatively well. The signature threshold is fairly low, and the circulation period seems appropriate. The same holds for the geographical requirements: while they do make it more difficult to pass the threshold, they certainly do not make it impossible, and ensure diversity in the country-base from which the support is derived. The ECI procedure also seems to do a good job in guaranteeing that Initiatives are not captured by special interests. At the same time, quite some differences exist regarding the conditions for signing an Initiative, which presents us with at least a hint of unequal access. Furthermore, one could note that the geographical requirements may well ensure regional diversity, but nevertheless introduce an additional hurdle. Most important are the qualifiers. Deciding whether or not a topic is allowed is by no means a technocratic exercise and greatly affects the instrument's actual usage. Lastly, the fact that special interests have not captured the ECI because they think it inefficient, ironically, also highlights that those initiating an ECI have to overcome a substantial number of hurdles, while the outcome is merely a request to the Commission to introduce legislation.

The ECI and Throughput Legitimacy
Criterion 5. Accountability. As argued above, it matters greatly to which extent decision-makers can be held responsible for their decisions. Applied to the ECI, we need to assess whether the main actors can be held to account in forums where they are obliged to explain and justify their conduct (horizontal accountability). In fact, the EU features the preferred three-tier mechanism, whereby the Commission is firstly obliged to elaborate on how it deals with ECIs through official decisions and periodic reports, and in lawsuits brought before the Court of Justice. The Parliament and the Ombudsman are also able to subject its 72 The other two successful ECIs passed the geographical threshold early on. 73 See Compassion in World Farming/World Society for the Protection of Animals, 'Press Release -Supporting Better Dairy urges European Commission to rethink the European Citizens' Initiative', <https://www.ciwf.org.uk/includes/documents/ cm_docs/2012/p/pr201207.pdf > (last visited 10 July 2018). 74 Berg and Thomson,supra n. 13,73. Brought to you by | Radboud University Nijmegen Authenticated Download Date | 1/4/19 3:22 PM practices to a closer scrutiny, and proceed with a more elaborate interrogation and critical review. Lastly, the Court, the Parliament and the Ombudsman are able to pass judgment on those practices. Ultimately, the Court may decide to annul a Commission decision, and the Parliament could adopt political sanctions (with a motion of censure as ultimum remedium). While not immediately efficacious, the Ombudsman's 'naming and shaming' strategy should not be discounted here either. 75 Less is clear about the vertical accountability of the ECI process. As noted above, the ECI instrument was officially evaluated at the end of 2015 and in mid-2018, but for present purposes these occurrences are rather less persuasive, since the reviews were conducted by the Commission itself. However, on several occasions we have seen an ad hoc creation of forums of vertical accountability where the Commission attended and responded to organisers' complaints about the process. The Commission has also shown that it is willing to listen to and assist organisers facing difficulties.76 For example, it generously extended the deadline for the first Initiatives. In July 2012, it offered to host ECIs on its own servers, alleviating the troubles reported by some organisers. In September 2013, the Commission considerably streamlined the registration process. In early 2015, it introduced the practice of partially registering Initiatives, which by 2018 it intends to cement in a recast Regulation.77 Overall then, the Commission seems fairly responsive when it comes to technical issues: it actively engages with the arguments of critics, and adapts its procedures where necessary. However, it may be doubted whether this willingness to listen and act remains as high once an ECI has collected enough signatures. Indeed, this is the moment where most national initiatives become problematic, as it often is unclear what the rights of organisers are.78 In case of the ECI, organisers are granted three important certitudes: an official hearing at the Parliament, a motivated reaction from the Commission, and the option to contest the latter by proceeding to the Court. While perhaps not optimal, this package of rights is more extensive than for many national equivalents (where often no judicial remedies are available). From this perspective therefore, the level of accountability still seems adequate. One cautionary remark does need to be made, as the significant variety in treatment that the successful Initiatives received during the official hearing is striking. Indeed, while the hearing on the first of these (Right2Water) resulted in an open, high quality debate, the two subsequent ones met with a much less favourable reception.79 This suggests that in practice, the European institutions involved are able to influence the substantive quality of the hearing in a significant way, which is clearly a matter of concern.
Criterion 6. Transparency. Most of the ECI's procedural steps are fairly transparent. The Commission's website presents information on all ongoing, withdrawn and rejected Initiatives. It provides the legal and technical information about the registration process, the final verdict of the Commission, and the official letters of rejection. Interestingly, while the official reason for rejecting a registration request has always been the same (namely that the proposed Initiative falls manifestly outside the ambit of the Commission's powers), the accompanying letters by the Commission provide more detailed background information. The procedure therefore seems to be most transparent up until the end of the signature collection phase.
The Right2Water, Stop Vivisection, One of Us and Ban Glyphosate campaigns showcase how this transparency is admirably retained afterwards. The Right2Water Initiative was accepted by the Commission, and each of the follow-up steps remained transparent: each further step taken was published online; all reports of meetings and press releases were archived, and made publicly available. Stop Vivisection and Ban Glyphosate resulted in mixed responses, and again all information was made publicly available. The One of Us Initiative was rejected, and did not trigger legislative action. Initially, communication was rela-tively minimalistic and slow here. After complaints by the organisers however, the Commission issued a detailed press release outlining the reasons why the Initiative was not adopted.80 Criterion 7. Legality. This criterion requires an evaluation of whether the Commission conforms to the applicable procedural standards. Its initial scrutiny of a proposed ECI, described above, is limited to assessing whether the proposal is not manifestly abusive, frivolous, vexatious, or contravenes the Union's values as set out in Article 2 TEU, or manifestly fall outside the ambit of the Commission's powers. The Commission clearly enjoys sizeable discretion to determine whether a proposal should be deemed unlawful or inadmissible. At the same time, this does not mean that no fetters should be deemed to apply on that discretion. Moreover, the case can be made that in the exclusion of the overwhelming majority of proposed Initiatives, an inconsistent practice has been followed that falls foul of the appropriate legal benchmarks.81 As remarked though, in its first pronouncements on the Commission's assessments, the Court was unable to find fault with the Commission's assessments. While more recent judgments did lead to annulment, the Union's judiciary is unlikely to engage in a more searching review in the near future.82 At this point, it may therefore suffice to point out that the rejection grounds employed are in conformity to those mentioned in the implementing Regulation; that the Commission fully complied with its duty to respond to successful Initiatives; that it supplied a proper motivation for its decisions in virtually all dossiers; and that it duly adhered to the applicable timeframes. On these counts, ECI practice has so far lived up to the legality criterion.

The ECI's Legitimising Potential for the EU -Looking Back and Thinking Forward
In light of the foregoing, the somewhat surprising conclusion is that the ECI should be considered to command a reasonable measure of legitimacy already. To many, however, this assertion may reflect a counter-intuitive 'paper reality', at odds with the unsatisfactory experiences of the various campaigners. On this cue, it could justifiably be asked whether the instrument may ever succeed in attenuating the legitimacy deficit of the EU -i.e. the weak foundation of its perceived, and continuously contested, 'right to rule'.
The core benefits of the ECI remain clear: it enables citizens to place items on the agenda that are otherwise ignored or overlooked, due to entrenched political preferences. Accordingly, it has been described as a form of ex ante non-majoritarian input in the policy process.83 Of course, the specific institutional structure and decision-making process should be accounted for here, keeping in mind the EU's unusual, quasi-federal setup. In this respect, it is noteworthy that the very same instrument has a different constitutional place and significance in a system as the present one, where the monopoly to come up with proposals officially lies with the Commission (but is de facto overtaken ever more frequently by the European Council84) -than in national systems, where a more central position is allocated to the representative body (compared to the still less powerful role played by the European Parliament). Still, the ECI is evidently capable of expanding and strengthening democratic participation in the Union's everyday life, exactly because of the comparatively weaker position of the EP. It thus also forms a complement to the (indirectly legitimising) role played by national governments, based on the legitimacy derived from representative democratic systems at the Member State level, as well as to the model of technocratic legitimacy underpinning the centrality of the Commission (which until the birth of the ECI was not controlled in a direct democratic fashion). In sum, it results in the creation of a richer palette of vectors of legitimacy than was previously the case.
In addition, the modus operandi of the ECI allows for a transnational dialogue and solidarity to take root, contributing to the establishment of a (sense of) collective community. It stimulates the creation of a common deliberative space, and the construction of durable interaction between grassroots movements, governmental and non-governmental actors -as illustrated by the 'Stop TTIP' Initiative, among others.85 Consequently, even when the results repeatedly led to disappointments, the operationalisation of the ECI seems to have had a positive impact on the legitimacy of the EU, and brought Europe (somewhat) closer to its citizens.86 What is more, in line with the general literature on instruments of direct democracy, we could expect it to serve as a less direct, preventative tool -signalling possible 'red lines' for EU regulation.
When reviewing recent practice, one could object that the new agenda-setting opportunities have surreptitiously been seized by single issue movements and NGOs ('instrumental capture'). Prima facie, the successful campaigns convey the impression of representing a large number of persons, originating from different corners of the Union. Yet, if the signing up of the required number within the fairly short time frame can in practice be mainly attributed to the efficiency of the organisers, often already established organisations (as in the cases of Right2Water, One of Us, Stop Vivisection and Ban Glyphosate), the true allegiance and sympathy of that million-plus may be wafer thin. Additionally, given the powerful and democratically legitimate checks and balances in the Union's legislative procedure, and the various ways to ensure representativeness of demands inserted by the ECI, we should qualify the risks and downsides of such instrumental capture. One may even argue that it allows for minorities to have access to EU decisionmaking in a way that was previously unthinkable, adding to the complexly balanced political system the Union has developed into. So, whereas interest group access to the agenda may compromise representative democracy, as has been noted in the general literature on direct democracy, the EU's unusual setup offers adequate checks and balances to prevent minority dominance over the legislative process.87 Lastly, it should be pointed out that the legitimising potential of the ECI appears greatly hampered by an inherent design flaw that -even when the procedural framework may be considered sensible, balanced and functional -is an inevitable cause of disillusion. The crude general message that has crept into many minds is that the citizens are now able to express their wishes, and will be listened to as long as their ideas are sufficiently supported by others. Yet, the Union simply does not possess competencies across the board, so that some requests are bound to fall on deaf ears; even in areas of shared power, it often cannot lawfully proceed in the way envisaged by the organisers.88 Confronted with flawed presumptions and false expectations, the Commission can then only be misunderstood, and accused of selectively following-up those Initiatives that are in line with its own priorities -a classic case of blaming the messenger. Even 85 See e.g. J. Organ, 'EU Citizen Participation, Openness and the European Citizens Initiative: The TTIP Legacy', (2017) 54 Common Market Law Review 1713-1747. 86 Cf C. Berg, 'Commission ignoring the ECI's positive potential for democracy', <http://www.euractiv.com/section/eupriorities-2020/opinion/commission-ignoring-the-eci-s-positive-potential-for-democracy> (last visited 10 July 2018), opining that: 'All serious ECIs united campaigners across borders, brought the EU directly into the daily lives of ordinary citizens and introduced new actors to the EU policy arena. Several ECIs even promoted overt pro-European integration goals, such as Fraternité2020 to strengthen EU youth mobility programs, For a High Quality European Education for All to expand the EU educational model, and Let Me Vote to support EU citizen mobility by strengthening democratic participation. Even ECIs that put EU policy into question help to strengthen the European project by encouraging healthy democratic debate.' when the Commission's mediating role is slightly more pronounced, and its discretion a bit larger than it would acknowledge, the negative outcomes strengthen the belief that the EU-at-large does not genuinely care, and that the ECI was merely devised as a public relations exercise.89 In result, the choice for institutional mediation, though arguably offering a healthy safety-valve, is cast in a similarly bad light. Despite performing its new tasks in good faith, taking pains to make clear that the extant rules and conditions need to be properly complied with before it can offer any follow-up at all, the Commission thus involuntarily inflicted damage on its reputation as an honest broker. Whereas the public perceptions are nigh impossible to counter, the solution might reside in a radical adjustment to the procedural framework not contemplated in current reform suggestions: a conscious curbing of the existing discretion in the ECI Regulation -either by explicit amendment, or by issuing guidance on how to determine when a subject matter is 'manifestly abusive, frivolous or vexatious', and 'manifestly outside the Commission's competence to submit a proposal for a legal act for the purpose of implementing the Treaties'.90 If that proves untenable, perhaps a reconsideration is in order of the minimalist conception contained in Article 11(4) TEU, to better cater to the apparent needs and concerns of Union citizens -especially when the EU continues to be attacked even when it is fully entitled to resist an application (a tendency from which its legitimacy obviously does not stand to gain at all).

Conclusion
The ECI was launched in April 2012 as a daring experiment in direct democracy. Up to 2018, more than nine million citizens have taken part. This paper has analysed the instrument's legitimising potential visà-vis the EU, by evaluating its operation against seven normative criteria under the rubrics of input and throughput legitimacy. Applying these criteria to the first years of its operation, we may conclude that the ECI can lay claim to some input legitimacy, partly due to its relatively thresholds for participation and the available methods for guaranteeing that special interests are unable to capture the instrument. This goes some way towards countering the criticism of the instrument's usefulness and viability.
On the downside however, geographical requirements, while ensuring diversity, impose an extra obstacle. Moreover, the conditions for signing up to an Initiative differ across countries, which presents at least a hint of unequal access. How the Commission has hitherto interpreted its own powers constitutes a further cause for anxiety. Given that an ECI may only deal with topics where the Commission is competent to introduce legislation, a restrictive interpretation may constitute a significant entry hurdle; and indeed, an uncomfortable number of proposed ECIs have not been accepted for registration. Hence, a risk looms large of too many topics being excluded, and relatively non-controversial topics being admitted. The Parliament, or from a different angle the Ombudsman, may pressure the Commission to engage in follow-up nonetheless, but even then, a positive outcome remains far from assured. Conversely, concerning throughput legitimacy, the ECI procedure features a laudable measure of transparency. In addition, the applicable legal standards have not been seriously violated, even when the preliminary scrutiny practice does lack a certain consistency (as sanctioned by the Court). Finally, judging from recent practice, the procedure appears to offer a sufficient measure of accountability as well. Overall then, the question whether the ECI possesses at least the potential to enhance the legitimacy of the EU, can be answered in the affirmative.
89 Compare e.g. the incisive comment of Berg, supra n. 86: 'Any Euroscepticism generated by the ECI comes from the Commission's own actions that limit citizen participation and increase the gap to citizens, not by the ECI instrument itself.' 90 As proposed in the study of the European Citizen Action Service, The European Citizens' Initiative Registration: Falling at the First Hurdle? (ECAS, 2014), 17-18. In comparison, the recommendations of the European Ombudsman (supra n. 66) that the Commission should 'endeavour to provide reasoning that is more robust, consistent and comprehensible to the citizen' and 'in its formal response to an ECI that has obtained one million signatures, should explain its political choices to the public in a detailed and transparent manner' arguably focus on the symptoms, and the procedural adjustments suggested there amount to little more than marginal tinkering. It was subsequently noted however that, whilst the instrument itself has been operationalised in a legitimate manner, and clearly possesses the ability to enhance the Union's legitimacy, actual experience suggests that the latter has remained underdeveloped. The causes are believed to lie not so much in the instrument itself, which grants citizens unprecedented access and recognition. It rather appear to be the high hopes and demands of ECI organisers on the one hand, and the Commission's necessarily restrained application of the rules on the other, that produced a palpable sense of disenchantment. The risk that too many proposals would come to naught has indeed materialised, damaging the legitimacy of the EU and its institutions, even while the underlying motives for the exclusions and rejections cannot legitimately be questioned. Arguably this quandary can only be resolved by a radical adjustment of the procedural framework or -albeit politically by no means palatable -the instrument's primary law foundations.