Options for Improving e-Participation at the EU Level

In this chapter, Nielsen et al. propose options for improving e-participation at the EU level without changing underlying legal frameworks. In response to the challenges to e-participation, which arise out of current institutional designs, the authors make creative use of the research results presented earlier in the book to suggest ‘ low-hanging fruits ’ for practical reform. The challenges addressed include the relative weakness of individual citizens ’ participation compared to that of CSO ’ s, the weakness of the Parliament in the legislative process and the continued de-coupling of the EU policy process from the will of the European citizens. While the chapter proposes no easy ﬁ xes, it points to some obvious practical steps forward. To improve existing participation mechanisms, the authors recommend providing improved support to citizens using the ECI, investing in the back-of ﬁ ce support needed for the EP Petitions Portal to realize its potential and improving the scalability of Your Voice in Europe through advanced data analysis. They also make four novel proposals, ﬁ rst among which is to experiment with participatory budgeting in relation to the Regional and Social Funds. The chapter ends with a plea for a long-term vision of a uni ﬁ ed European participation structure to gather and harness the potentials of individual mechanisms.

This section of the book discusses options for improving e-participation at the EU level. The discussion makes creative use of the findings of the previous sections. The first step is to outline the challenges that the specific institutional arrangements of the EU present in relation to the development and implementation of e-participation. Our main emphasis, however, is on options for improving participation through digital means within existing institutional frameworks. Therefore, the second step of this section is to revisit already existing mechanisms and discuss options for additional ones.
To identify immediate options for strengthening EU institutions' rapport with European citizens through e-participation, we added two additional sources of information to the findings from the review of the state-of-the-art of research and the case studies (Part I and Part II of this book). The first additional source of information is experience. It is a well-established principle in pragmatist social science to rely on the first-hand experiences of the actors involved in a given social system, to identify the paradoxes and potentials for development of such systems (e.g. Flyvbjerg 2001). The other source of additional information is the outcome of creative brainstorming. Developing new tools and mechanisms for the functioning of institutions relies to a great degree on the ability of people who are positioned at the intersection between different institutional spheres to creatively combine elements of the different organizational practices to which they are exposed (see, e.g. Campbell 2004).
To allow our analysis to be influenced by these additional sources of information, we engaged with a small number of stakeholders to gather and generate ideas for immediate improvement of participatory practices at the EU level. All interviewees involved in the local-, national-and EU-level case studies were asked to elaborate on their ideas on how the tools, about which they were being interviewed, could be applied at the EU level. These inputs are reported in part in each individual case study. We also invited a small group of institutional and nongovernmental stakeholders and experts to a day of co-creation at a workshop held in Brussels on November 21, 2016. At this workshop, the authors of this report and the expert group discussed ideas for improving existing participatory tools at the EU level, as well as immediate options for going beyond these tools, for example, by adopting some of the new tools described in the case selection earlier in this report. Following these steps of stakeholder engagement, we have used the most clearly apparent consensus positions among the stakeholders as starting points for recommendations, which have been supplemented by the evidence gathered in the literature review and case sections of this report. Rather than attempt a systematic presentation and evaluation of all logically possible applications of the tools in use at national level, which were analysed in the previous section, we attempt to use the findings from the case studies to identify the 'low-hanging fruits' for EU-level participation. We take such low-hanging fruits to be those changes or additions to EU-level participation mechanisms that might make a significant difference without demanding changes to existing institutional mandates.
The section ends with a discussion of how a unified approach to e-participation could provide a common access point, not only to participation in the processes of the EU institutions but in the entire multilevel construction of European governance. Along the way, we seek to provide concrete suggestions for small steps towards such a unified approach and to take into account risks and potential pitfalls to be avoided.

EU's Institutional Architecture and the Need for e-Participation
Identifying suitable approaches and tools for e-participation at the European level demands first of all that we recapitulate EU's specific institutional architecture and the prevailing patterns of governance in the European Union, which need to be taken into account. Established institutional structures and procedures represent important enabling and constraining conditions not only for formal opportunities for citizens to influence European decision-making but also for the potential uptake of e-participation tools and practices. In many ways, the European Union is a political system sui generis. As such, the EU combines elements of a supranational body, a joint federation of states and a few characteristics of genuine statehood. At this stage, the EU is not a fully sovereign state, and whether it will ever be so is a matter of fundamental contestation (e.g. Nicolaïdis 2013). This unique setting is reflected in EU's institutional structure and its related decision-making processes. Among the most notable characteristics is EU's duality as a union of citizens and a union of Member States (Sturm 2010). This duality is expressed in the roles of the European Parliament-the representative body of the European citizens-and the Council of the European Union, which represents the Member State governments. Another striking feature of the European Union is its multilevel governance, which blurs in everyday practice the distinction between national, international and federal governance. EU's complex institutional design is not based on a constitutional blueprint but is the result of numerous integration steps and incremental reforms, often accompanied by contention and tough negotiations between the Member States (Wallace et al. 2010: 70ff.).
We identify three features concerning the relations between the European institutions and their constituency. These are coupled to the specific institutional setting and the procedural rules of the European democratic system, which can be regarded (and actually are regarded by the European policymaking bodies) as problems or deficits of the democratic set-up of the EU; it is against this background that e-participation is perceived as an option for strengthening the ties between European citizens and the EU policymaking bodies.
• Citizens have fewer rights to voice and consultation than civil society organizations (CSOs) Before the enactment of the Lisbon Treaty in 2009, the often-cited democratic deficit and the legitimacy crisis of the EU (see Part 1 Sect. 3.3.2) had triggered discussions on how to better involve European citizens in the decision-making processes of the EU. At first sight, this 'participatory turn' (Saurugger 2010) seems to be more than mere rhetoric, as the aim for more and better involvement of civil society and citizens has entered a number of official policy documents, most prominently EC's White Paper on Governance (Commission of the European Communities 2001). Yet, a closer look at both the debate and the formal framework within which such increased participation could take place cautions us not to expect too many advances in citizen participation. First and foremost, it is important to note that according to the Lisbon Treaty (TEU, Art. 10), the EU is explicitly based on principles of representative democracy. Second, Art. 11 of the TEU contains a number of provisions complementing the principle of representative democracy. In clause 1, citizens and associations are given a right to voice their views. And clause 2 requires the institutions to 'maintain an open, transparent and regular dialogue with representative associations and civil society'. Comparing the two provisions, citizen involvement is explicitly defined as voice and consultation and remains rather noncommittal. This gives citizens fewer rights in decision-making processes when compared to civil society organizations (CSOs), which receive a formal guarantee to be heard and involved in dialogue (Fischer-Hotzel 2010: 340). Against this background and taking into account the debate, Fischer-Hotzel (2010: 339) points out that for many, 'participatory democracy' at the EU level actually means 'associative democracy' and the inclusion of CSOs in the processes of decision-making. It is a common critique of (digital) participatory processes that they are monopolized by established political actors (parties, associations or movements), and that ordinary citizens are not heard as much. In 11 of the studied cases, we found that both established organizations and professionals are strongly involved in the digital tool; this includes all four of the cases at the EU level (Voice of Europe, European Citizens' Initiative, European Citizens' Consultation, Futurium).
• Improved legislative functions of the European Parliament, but still no right to directly initiate legislation or ability to effectively hold the European Commission politically accountable Structures for representation at the EU level have arguably improved considerably over time. The Lisbon Treaty addressed many of the institutional problems that were frequently raised in debates about the EU. Most importantly with regard to the democratic deficit and related legitimacy problems of the EU, the position of the European Parliament (EP), which is EU's only directly elected institution, was considerably strengthened. Clearly, the Lisbon Treaty has taken substantial steps towards an effective parliamentarization of the EU. For the most part, the EP has become an equal player in legislative processes and spending decisions, thus significantly increasing at least the formal democratic legitimacy of most European regulation (Oppelland 2010: 87f.). Still missing, however, is the right to directly initiate legislation-a function that continues to rest with the European Commission (EC). In addition to the improved legislative functions, the EP gained important electoral functions, as the President of the EC, and the European Commissioners need to be approved by a majority of the EP. Any nomination for EC presidency by the European Council has to take into account the majority situation in the EP. However, EP's ability to effectively hold the EC politically accountable remains weak, as the threshold to dismiss a Commission with a vote of non-confidence is extremely high (2/3 majority). It is unusual that the threshold for non-confidence is higher than the requirements for election. One rationale for this atypical design feature might be that the EC President does not have the prerogative to dissolve the European Parliament (Oppelland 2010: 88).
• Continuing de-coupling of the European political system from the processes of political will-formation of the European citizens While important institutional improvements have been achieved, the political processes of the European Union still do not sufficiently fulfil key functions of representative democracy. Most importantly, election campaigns for the EP continue to be primarily driven by national perspectives. In addition and related to this observation, the political parties and parliamentary factions in the EP are currently not divided into recognizable majority and opposition groups competing for different policy solutions. The dominance of informal grand coalitions of the largest parties in the EP makes it difficult for the public and the citizens to hold the members of the EP and their parties accountable. This points to the current most crucial deficit of the European Union, as emphasized by, e.g. Habermas (2008: 98f.): the continuing de-coupling of the European political system from the processes of political will-formation of the European citizens. Noteworthy improvements in this regard have been achieved with the introduction of the so-called 'spitzenkandidaten' (top candidates) of the main political party families participating in the EP elections in 2014. From this perspective, the next logical step would imply that not only the President of the EC would be backed by a majority of the EP but also that the EC President and Commissioners are more often than not elected from the parliament, thereby establishing a more visible linkage between parliamentary majority and the executive actions of the EC. However, care needs to be taken that this type of party politicization of EU politics remains compatible with the requirements of negotiation between different Member State interests (Lippert 2013: 13) and sufficiently takes into account the interests particularly of smaller Member States.
Howsoever one views the state-of-play of European integration, there are good reasons to explore pragmatic options for citizens to voice their concerns and ideas. The long-standing and continuing democratic deficits of the EU are rooted in a complex and mutually reinforcing mix of institutional design features, lack of a genuine European public sphere, and insufficient politicization of European politics as such. Redressing these problems is ultimately a constitutional matter and far outside the range of what e-participation can achieve alone. However, if properly designed and implemented, e-participation has the potential to contribute to promising solutions in the areas of accountability and transparency, transnationalization and politicization of public debates and the improvement of exchanges and interactions between EU decision-making and European citizens.

Challenges Specific to e-Participation at the EU Level
Beyond the above-mentioned problems in the relation between the EU policymaking system and its constituency are a number of other challenges arising from the specific institutional structures of the EU, which must be taken into consideration for any attempt at improving channels for e-participation at the European level.

Language
A major challenge to e-participation at the European level is language. With 24 official languages, translation is a major element of the running costs of the EU. Several of the European-level cases address this in different ways (Part II). The institutionalized mechanisms range from relying on English as a working language to full translation of all major content into all official languages. Your Voice in Europe clearly privileges English speakers by treating English as a de facto lingua franca. The platform provides all consultations in English and only a few additionally in one or more of the major languages (German, French, Italian, Spanish). Written contributions are accepted in all official EU languages. But with the Englishonly availability of core information, such as the questions to be answered, the platform has a clear choice built-in that decisively shapes a priori the demographic of possible participants. The European Citizen Initiative (ECI) strikes a middle ground. The platform provides all information about the mechanism in all official languages. The platform also accepts initiatives written in all official languages. Translation into other languages was previously the sole responsibility of the initiator, but with a new agreement reached in December 2018, initiators will now also be able to ask for help from volunteer supporters contributing through an online collaboration platform. 1 The previous system clearly favoured well-organized campaigns over more loosely affiliated individual citizens as initiators. With the addition of the collaboration platform, one can hope that individual citizens and small groups will be better able to be heard as others volunteer their resources. The European Parliament's Petitions Portal is clearly the most multilingual of the institutionalized mechanisms. The portal allows submission of petitions in all official EU languages, and summaries of the petitions are translated into all official languages and made available to the public. Furthermore, video of meetings in the Petition Committee, where petitions are discussed and petitioners are sometimes invited to make their case, are made available online with the option to select interpretation in each of the official languages. It should be noted that the working language in these meetings is typically English. It is also important to note that that the translation efforts of the EP, from which the e-participation platform benefits, would take place in any case, so that the e-participation mechanism is thus able to piggy-back on already allocated translation practices and resources.
There seems to be a pattern whereby e-participation platforms at the European level mimic the underlying institutional working mode with regard to working languages and resources committed to translation. This tendency is corroborated by the single example among our case studies in which a non-EU e-participation mechanism makes use of multiple languages, namely the Swiss e-voting platform. On this platform all information is made available in all official languages, which is traditional for the underlying Canton institutions. In this case, the translation workload is lightened considerably by the fact that the mechanism is a voting mechanism without deliberation.
It is a well-known dilemma of the European Union that full inclusiveness demands considerable investment in translation, while full efficiency privileges English as a working language. Considerable investments in new tools for digital translation have therefore been ongoing since at least the first Framework Programme for Research and Development (1984-87). However, the promises of digital translation have long seemed a mirage; always on the horizon and never quite as good as expected. Despite these setbacks, a new wave of optimism exists around translation software based on artificial intelligence and so-called deep learning (www.ec.europa.eu/citizens-initiative). One recent paper thus claims that a new version of Google's translation software was scored by observers to have a degree of fluency in the translation of random sentences from English to Spanish, which was very close to that of human translators (5.43 on average compared to 5.55 for humans). It is outside the scope of this report to assess the plausibility of such claims and the implied hopes for a more multilingual Internet that comes with it. However, there is no doubt that while digital translation into the major languages of the world are seeing massive investment, the European Union and its Member States will be forced to add their own investments on top of those of private actors if all of the official languages of the EU are going to have comparable degrees of support. Less ambitious results may be useful, of course. We would thus expect the availability of digital translation into the few most widely spoken languages in Europe to greatly improve the accessibility of EU-level e-participation mechanisms.

Multilevel Governance
An often-discussed challenge is the multilevel nature of European governance. It is well-known that governance complexity rises proportionally with the many levels of governance that the European system encompasses. With the upper echelons of this system having often only indirect connections with national representative democracies, it is easy to assume that e-participation at the EU level will automatically inherit the democratic deficit/'no demos' problems of the governance system as a whole. However, examples such as the UK Democratic Dashboard (see Part II, Chap. 2) show that it is possible to construct a common access point to a multinational and multilevel governance system, even if not all potential users have access to participation in all of the channels of participation. The digital infrastructure of the Five Star Movement similarly shows that a common infrastructure for local-, national-and European-level political participation can provide much-needed advice, guidance and overall structuration for citizens wishing to participate in decision-making (Part II, Chap. 10). Of course, the fact that the construction of such common infrastructures is technically possible means neither that it is necessarily, politically feasible nor that developing a well-functioning system is easy. Our point here is only that the constitutional difficulties of European democracy do not by necessity translate into roadblocks for a common European e-participation platform.

Digital Divide
As regards the so-called 'digital divide' (Mossberger et al. 2003(Mossberger et al. , 2007, there are good reasons to revisit some of the assumptions underlying the traditional discussion of the divide between advanced and less advanced regions of Europe, given the development infrastructures for Internet access over the last decades. Granted, Europe-wide patterns of exclusion of the elderly, citizens with lower levels of education and citizens with disabilities from digital platforms of e-government and e-democracy remain (as discussed, e.g. by Van Dijk 2012 andPanopoulou et al. 2014). But these patterns are not specific to the EU level: they affect opportunities for e-participation at all levels of government. More importantly, with regard to access to basic broadband they no longer map onto the underlying divides between richer and poorer regions of Europe (Negreiro 2015). While the digital divide as traditionally understood is thus a challenge to be addressed by any e-participation platform, this challenge is not exclusive to participation on decision-making at the EU level. Rather, it is a reminder that all efforts at increasing citizens' participation in policymaking should beware of an online-only strategy; face-to-face participation processes supported by effective mobilization efforts must remain in the toolbox.
If an EU-specific digital divide is to be considered a relevant challenge for e-participation at the European level, it is the cultural divides between Member States with a great deal of trust between governments and their populations regarding the sharing and recording of personal data such as ideological standpoints, and those Member States which-for good historical reasons-do not have the same degree of trust. This cultural divide concerning degrees of digital openness presents a real challenge to the plausibility of common European approaches to e-government under any form, including e-democracy and e-participation.
Keeping these qualifications in mind, the following sections will present and discuss suitable e-participation approaches at the EU level in greater detail, while taking into account some of the above-mentioned institutional characteristics and weaknesses.

Ways of Improving Existing Participatory Mechanisms
The EU institutions already have a range of well-established mechanisms for digitally supported citizen participation. In our co-creative workshop, we focused on three of them: the European Citizens Initiative, the Your Voice in Europe consultation platform and the European Parliament Petitions Portal. The consensus position among stakeholders and scientific observers with regard to these mechanisms seems to be twofold. There is a general agreement that these platforms and the underlying legal mechanisms hold great potential as first steps in the direction of opening up European decision-making and governance to citizens' participation. However, the net result of the legal mandates, their interpretation and their practical and technical implementation is that ordinary European citizens are still without simple and transparent channels to engage with the EU institutions.
To make this consensus position more concrete, the work underpinning this report repeated an exercise performed by Lironi (2016) to facilitate a SWOT analysis (Strengths, Weaknesses, Opportunities, Threats) for each tool. Using this technique in dialogue with a large number of individual stakeholders, Lironi et al. corroborated the existence of the above-outlined consensus positions. In order to go beyond the findings of Lironi et al., our analysis placed special emphasis on ways of improvement. We thus deployed the SWOT analysis in the above-mentioned co-creative workshop, where participants were able to feed off one another's ideas to come up with recommendations to improve the existing mechanisms.

Improving the European Citizens' Initiative
At the time of writing (February 2019), a vote by the European Parliament on a European Commission proposal to improve the European Citizens' Initiative (ECI) is imminent. A reform proposal was made by the EC in September 2017, which sought to respond to a number of critiques and suggestions put forward by stakeholders, including some of those discussed below; the multi-NGO ECI Campaign welcomed EC's proposal as an 'overdue step' in the right direction. 2 The proposal went through interinstitutional negotiations in late 2018 and thus currently awaits a parliamentary decision. Our analysis of the ECI in this section takes the ECI as it existed before EC's proposal as a reference point.
As the first transnational e-participation tool for policy agenda-setting, which has an institutional embedding as strong as the one provided by the ECI in the Lisbon Treaty's §4, the platform is both unprecedented, and still unparalleled in terms of the advancement of participatory democracy at transnational level (cf. Part I, Chap. 4).
In formal terms, the ECI gives citizens a powerful agenda-setting tool, given the conditions that a proposal must fall within the remit of the European Commission and gather 1 million signatures. Ideally, the mechanism would both allow citizens to take collective actions and allow decision-makers within the EU institutions to gain greater insight into citizens' concerns. Furthermore, the ECI platform has arguably grabbed the attention of organized civil society and thus created an existing user base, which could help to propel the platform forward in the case of a process of revisiting and expanding its reach.
Despite these positive notes, there has been widely shared criticism of the ECI mechanism as implemented. The consensus position here seems to be that not only could the digital platform itself and the digital support tools be made more userfriendly but more importantly that the underlying legal constraints and the way they are interpreted by the EC block a culture of open involvement and engagement with citizens. 3 Since the ECI was implemented, over 70 proposals have been submitted, of which only four have been successful in passing the entrance demands, with a fifth on its way. The submission of proposals has been declining, likely due to the poor success rates of their predecessors. Of the failed proposals, roughly a third failed to gather the necessary support, another third were retracted by the submitters, and the remainder were rejected on formal grounds.
From a constitutional point of view, this should come as no surprise. As mentioned above, the Lisbon Treaty is explicitly based on principles of representative rather than participatory democracy, and it favours organized interests over individual citizens. These principles are mirrored in the conditions under which the ECI functions. The demand that the proposal submitted must fall within EC's competencies to act places a heavy burden of regulatory insight on those wishing to formulate and submit proposals. And the demand that proposals must gather one million signatures, with its various technical and security requirements, places a burden of organizational capacity and resources on proposers. In terms of the abovementioned conditions for successful participation, a central weakness of the ECI is thus that although it provides an opportunity to participate, it fails to support this opportunity with a strategy for mobilizing and engaging citizens.
It is, therefore, no surprise that reform is now underway. However, the limited rights to 'voice' and 'consultation' established in §11 of the Lisbon Treaty could, in a practical context, be interpreted much more widely than is currently the case for the ECI. It would be legal and also practically possible to support citizens to formulate citizens' initiatives and ensure that proposals meet the terms of EC's competencies to act. The Finnish Open Ministry platform, for example, makes active use of volunteer experts, who support the formulation of citizens' initiatives to ensure that they fall within the remits of the body to which they are addressed (see Part II, Chap. 7). In lieu of such support having been provided for the ECI, NGOs have had to coalesce around the ECI Campaign, where they have attempted to provide citizens with some measure of support. 4 Part of the ongoing reform debate is, therefore, the question of where the responsibility lies for supporting citizens' participation via the ECI. Receiving little or no support from the EU institutions in their efforts to use the ECI mechanism, it would be no surprise if non-organized citizens suspect that the ECI in its original form was intentionally designed as a half-measure. The ECI has arguably set the stage for an unfortunate outcome where almost all proposals fail to qualify, either by falling outside the scope of EC's remit or by lack of mobilization and support.
Another important aspect of the ECI debate is what happens to proposals once they pass the qualification criteria. What exactly is the formal process for treating successful proposals, and how are they used in decision-making processes? The presence of such clarity is one of the most important conditions for the positive impact of participatory mechanisms and tools on decision-making and agendasetting processes.
The EC's earlier strategy of achieving improvements within the existing framework has also been effective to some extent and should be pursued further as a complementary path to current reform attempts. Our analysis thus points to several opportunities to improve the mechanism through decisive action by institutional leaders.
Among the most obvious opportunities for improvement are greater support for proposal formulation and better follow-up regarding the processing of proposals after submission. Furthermore, following the success of the Five Star Movement (Part II, Chap. 10), to support mobilization efforts by ECI initiators, the ECI digital platform could be broadened to allow organized civil society to use it as a mobilization and campaigning platform, for example, by integrating online community functions as well as functions to support offline meetings.
More broadly, it is important not to fall into the trap of believing that improving the ECI is simply a matter of finding the right technical or legal 'fix'. The challenge of opening up the ECI platform to active engagement with European citizens is more than a matter of the adoption of new digital tools and new legislation. Getting the ECI to work for citizens is just as much a matter of organizational culture and leadership commitment. If a relaunch of the ECI was to take place, it would be essential that the process should not be one-sided. Instead, the relaunch process itself should seek to embody a new openness and a willingness to engage in mutual learning along the way. It would, therefore, be important to open up the implementation process to user involvement and to work actively with local and national governments as well as NGOs to draw on their experiences (Table 13.1).

Improving Your Voice in Europe and the European Parliament Petitions Portal
In the following, we treat the common consultation platform of the European Commission alongside the European Parliament (EP) Petitions Portal. Although both these tools have a formal link to the decision-making process, there are still some important weaknesses to alleviate. Albrecht (2012) reviews the e-consultation practice at EU level with a focus on the Your Voice in Europe platform, building on analyses of other scholars (cf. Quittkat and Finke 2008;Quittkat 2011;Tomkova 2009). His main points are as follows: online consultations have become a well-established instrument regularly used by practically all Directorate Generals (DGs). This has increased existing participation opportunities and brought more frequent public participation, especially of diverse interest groups, resulting in broader input into EU policymaking and • Seize the opportunity to improve the legal structure to address weaknesses (impact; transparency) • (Use as a tool for) mobilization and campaigning.
• (Use as a tool to achieve) transparency in lobbying.

Threats
• Current crisis response is too chaotic to make good use of opportunities • The crisis of Europe, including the weaknesses of the ECI, lead to frustrated citizens • Pseudo-legitimacy • An ineffective ECI (and other mechanisms) easily backfires, leading to increased Euroscepticism and nationalism • An ineffective ECI leaves the door open for negative advice in the form of referenda, a type of input which is very difficult to handle the extension of its knowledge base. However, serious flaws include opaque and sometimes inadequate processing of contributions; a shift of focus on to closed question formats; little evidence of mutual learning; lack of feedback to participants on the use of their contributions, entailing frustration; one-way formats of communication with no opportunities to debate contributions; limited use of technologies (general purpose instead of specific e-participation and web 2.0 tools); and a lack of integration of new arenas for debate, e.g. the political blogosphere (Albrecht 2012: 15 ff.).
Albrecht advocates a model of deliberative e-consultations, which not only consist of collecting comments on a policy proposal but also allows for discussions both among the participants and with representatives of the EU institutions concerned. This is supported by our comparative case-analysis, where these two conditions-interaction with other participants, and interaction with decisionmakers-appear to be very important in order to have an impact on the agendasetting process. However, the implementation of such a model is confronted with a number of unresolved problems, such as how to adapt a face-to-face format to a large-scale setting, high costs, a minority of participants being willing to engage more deeply, the need to facilitate the process and to inform and support the participants and the reluctance of officials and policymakers to participate. With regard to improvements in technological support, natural language processing and argument visualization technologies are regarded as interesting candidates, although evaluation results to date are mixed. A third approach is to integrate e-consultations with 'third places' in new ways, i.e. social media platforms such as the blogosphere and popular social networking sites, in order to counter the dominating top-down flavour of existing EU channels. The assumption is that a good deal of exchange on these sites includes political talk and that the separation between political content and life world is increasingly blurred. Several EU projects have already experimented with linking e-consultations to social media (cf. Albrecht 2012: 19). Taken together, the strategies outlined show some promise to develop e-consultations to a model which is more open and effective than existing practice, and which will also enhance the quality and legitimacy of policy decisions with the help of tools such as Your Voice in Europe.
With regard to the EP Petitions Portal, Tiburcio (2015) examined 'The Right to Petition' in the European Parliament for the Committee on Petitions and made recommendations for the EU petition system. Tiburcio notes that recent studies on petitions tend to neglect the petitioning system of the European Parliament, referring to it as being a 'well-embedded process to deal with petitions' (Tiburcio 2015: 12). In his study Tiburcio comes to the following conclusion: [. . .] the petition system of the European Parliament compares well overall with the petition systems of Parliaments of Member States. In terms of conventional features, it scores well in all dimensions: ensures direct access (and not intermediate) by citizens; it's highly inclusive and open to both national citizens of Member States as nationals from third countries, if they reside within the EU territory; it offers possibilities for greater involvement of citizens, including through frequent hearings, followed by public debate in committee. (Tiburcio 2015: 40).
Nevertheless, the EP Petitions Portal is also a prime example of how the institutional peculiarities of the European Union can make it difficult to transfer experiences from the national to the European level. The limited powers of European parliamentarians to set the political agenda, combined with the subsidiarity principle, thus bear directly on the usefulness of petitioning MEPs. One workshop participant observed the difference that this creates in comparison with, for instance, the Dutch petitions platform petities.nl: 'The petition has to be about European laws and regulation and in particular the implementation of it by the Member States. It has to be about issues where the European Union has exclusive competence' (interview with researcher, our translation). This is one of the reasons why so many filed petitions are rejected because the subject they are addressing does not fit this condition. By early 2019, more than 10,000 petitions had been received. Of these, one third turned out to be inadmissible.
Working within these limitations, it would nevertheless seem reasonable to make some use of the inputs gathered through the platform. The Petitions Committee could, for instance, make an inventory of what people ask for in inadmissible petitions, which could be distributed among MEPs and perhaps even national parliaments. Taking this idea one step further, the Petitions Portal could gradually be expanded to serve as a multilevel petition system with connections between the existing local, national and European institutions. This would give an enhanced basis for understanding issues which are stirring among European citizens, compared with top-down tools such as consultations and polls.
Comparing the Petitions Portal to Your Voice in Europe is illustrative in a number of ways. Your Voice in Europe exemplifies a one-stop-shop for EC consultations, which is a great advance on previous decentralized approaches. However, the consultation formats have not yet been harmonized across the different DGs, which makes the process less transparent for users than it could be. Upfront clarity about use of the inputs gathered through the platform could be improved. And there is a lack of feedback to citizens about the outcomes. The EP Petitions Portal, by contrast, has relatively clear feedback mechanisms, although this often comes late. The Petitions Portal has even less upfront transparency about what citizens may expect to happen to their input than the consultation platform. Information management is a concern with regard to both platforms. Petitions produce not only quantitative data but also potentially vast amounts of qualitative statements. Aggregating such input is both time consuming and politically risky; especially because there is no mechanism for the approval or disapproval of aggregation choices made by Commission services and by the people who provided the input. This leaves a great deal of power to shape the outcomes of consultations in the hands of the secretariat functions of the DGs. Similarly, the lack of transparent curation of petitions submitted to the EP Petitions Portal, along with the lack of clarity about the use of the inputs submitted, gives leeway for selective interpretation to the EP Petitions Committee. In terms of user experience, both platforms suffer from typical ailments of online participatory tools: a lack of interactivity; a lack of deliberation; and a lack of mobilization efforts. Together, these weaknesses produce results that may be recognized across a broad array of e-participation cases: over-population by organized interests and elites; and a lack of publicity.
Despite these shortcomings, we believe that relatively simple measures could considerably improve the socio-technical functionality of the platforms.
Your Voice in Europe could: • Provide feedback via e-mail Once consultations are submitted, the results which are processed and fed into internal decision-making processes could easily be communicated to participants, thereby increasing their sense of transparency and involvement. For example, the synthesis report-which is a mandatory follow-up to each consultation-could be e-mailed to each participant in addition to publication on the website.
• Make use of data analytics to aggregate qualitative inputs Several data analytics companies, as well as DG Connect, have developed tools that help to make systematic and transparent decisions about the aggregation of qualitative data. Deploying such tools in the internal processing of results would help to improve the dependability of the process.
• Improve scalability through technological support The current difficulty of treating qualitative data represents a bottleneck, which from a resource perspective could actually serve as an incentive to maintain low participation numbers; how would EC services handle a 10-, 100-or 1000-fold increase in data? Big data technologies, such as machine learning whereby algorithms improve in step with the data amounts processed, might hold some answers to scalability.

• Open up back-end data
If consultation data and the tools used to process it were made available to the public, the process of making use of the input would become transparent and would establish a hotspot for public dialogue on EU policy decisions.
The EP Petitions Portal could: • Benefit from more back-office resources Whether through additional staff, additional technological support, or a mix of both, users would gain a much livelier experience of interacting with MEPs if the necessary back-office resources were available to ensure swift and qualified responses and interactions with users.
• Use simple tools to educate and mobilize Simple additional tools such as updates via text or e-mail, education on issues via video messages, ad hoc inputs via mini-polling and visualization of data and policy mechanisms would help to keep the attention of citizens and qualify their input.

• Provide communication and mobilization support to petitioners
Since the EP Petitions Portal is more successful than the two other established tools in attracting the attention of non-organized citizens, it would be highly useful to provide these citizens with basic tools to mobilize support for their petitions (handbooks, free publicity mechanisms, etc.). The Dutch petition platform, for example, contains information that addresses how to get traffic to the petition, start a campaign website or blog with more information, write and spread a press release, get in touch with local or national TV or radio broadcasting centres and place a widget (so people can sign the petition from another social network site or campaign site). The Portal might also be provided with crowdsourcing functionality in order for campaigns to collect finances to hire a public affairs professional or to collect citizen volunteers for support.
• Add various functions for online deliberation A lot of the pressure to respond directly to questions and petitions could be taken off back-office staffers and MEPs if options for deliberation between participants were added to the Portal, e.g. debate options, options for collaboration on petitions, voting both for and against, etc. This would make it possible for citizens and interest experts to share knowledge in the ongoing process of developing and sharing ideas for petitions. Wiki Melbourne is one case in which such functions were embraced with enthusiasm by citizens and officials alike.
Such deliberative functions can be more or less structured. The 5SM makes use of debate platforms that are open to everyone, while voting on proposals is for registered users only. The Petities.nl platform has a structure where users cannot comment directly on proposals, but must make counter-proposals; the 'debate' between opposing proposals is then settled by voting. How to balance openness and structure is a question to be settled through experiment and experience. The major criterion is not to reinvent the wheel, but to keep working on the platform to improve its usefulness and popularity while drawing on experience from others along the way. This is underlined in the comparative case analysis by the fact that the condition of sustainability-improving the tool over time-contributes considerably to impact on final decision-making (Table 13.2).

The Low-Hanging Fruits: Obvious Steps to Improve EU-Level e-Participation
The discussion about increasing openness and participation at the EU level often centers around regulatory reform. However, no matter which regulations are put in place, openness in administration is as much a matter of culture as it is one of the formal structures (Torfing et al. 2012). To address the space of possibilities available to European institutions within existing formal structures, we have put togetherwith great help from experts and stakeholders engaged in our efforts-the following four suggestions for 'low-hanging fruits' of participation, which institutional leaders should be able to harvest while relying only on their existing remits.

Experiment with Participatory Budgeting in Relation to the Regional and Social Funds
This idea is that given the positive experiences with participatory budgeting methods by European cities, regions and Member States, there must be areas of EU spending where such methods could help to enhance citizens' participation. Current best practices such as described in, e.g. the Belo Horizonte and Paris case studies (Part II, Chap. 10), show that an e-participatory element is essential for scaling-up such methods. At the same time, however, face-to-face interaction and a certain rootedness in local situations is characteristic of all successful cases of participatory budgeting. Finally, the general conditions for the success of e-participation also Structural separation between problem owners and tool developers apply here. For example, the Belo Horizonte case shows that failure to achieve a clear understanding among the participants concerning the mandate given to the process is lethal. In this case, the participation rate decreased enormously over time, in 2006 172,938 participants online and in 2013 8900. This was due to the failure to implement the winning project in 2008, after which people lost their trust in the procedure, despite other projects having been implemented. The question is therefore as follows: Given the complexity of the European decision-making process, can participatory budgeting even be conceived of at the level of the common European budget? Or should methods of participatory budgeting be seen as a means to making a connection between citizens and the EU at the local and regional level?
There is no doubt that setting aside a certain percentage of the total EU budget to be distributed by citizens would be a powerful symbolic gesture. However, there are many ways in which such a mechanism could go awry from the beginning. Allocating funds at the discretion of citizens would demand the implementation of some methodology to avoid simply reproducing current patterns of influence of different Member States. The participants at the workshop pointed to the Horizon 2020 EU research framework programme as an example of an allocation mechanism which is constructed to avoid simply reproducing national interests and focuses instead on the excellence and societal relevance of projects. The EU-funded CIMULACT project 5 provides an example where citizens have been involved, albeit indirectly, in the allocation of funds through the Horizon 2020 mechanism. Their role is to produce visions, priorities and calls for projects. A similar role might be conceivable if a budget was allocated to participatory budgeting at the EU level. One participant suggested such a mechanism might be thought of as an Erasmus programme for entrepreneurs, i.e. a platform where young entrepreneurs could submit ideas and compete for funding from participating citizens. Other participants underscored that such an allocation mechanism ought to be flexible and oriented towards pressing problems, such as-in these years-migration, climate change and improved education. Experiences from city-level cases show that participatory budgeting methods that start with small but realistic setups have a better chance of achieving longevity than those that make big promises, but do not follow through. For this reason, one participant suggested that it would be useful to start small and allow for an ongoing process of community-building to take place around the mechanism, which might then grow over time.
The EU already has well-established mechanisms for reallocating EU budgets to local initiatives and concerns through the Structural Funds. The Regional Development Fund as well as the Social Fund both already assign significant decisionmaking authority about the spending of these funds to the local or regional level. Building on lessons learned in cases ranging from Belo Horizonte to Paris (Part II, Chap. 10), it is not at all hard to imagine a unified framework for participatory budgeting being implemented as part of these budgeting processes. The availability of good projects and the willingness to back them would be crucial. But given the local development ambition of these funding programmes, and taking into account the much smaller scale of application, it would be feasible at this level to implement more open co-creation processes which reached organically from idea formulation to project application, funding decisions and implementation. In the wider perspective of regional development policy, it is conceivable that the input and throughput stages of participatory budgeting could bring together decision-makers, citizens and local organizations and businesses in a process that may help to create a common focus point for the community.

Expand Online Engagement with MEPs Beyond Petitions
This idea is to expand the palette of online engagement tools available to citizens to interact with Members of the European Parliament (MEPs) (and vice versa), beyond those that are currently available via the Petitions Portal and the EP website. Such tools could include: • Availability of voting records for each MEP • Public functionality to pose questions to MEPs and their staff • Consultation functionality for MEPs to gather input from citizens • Blogging functionality where MEPs can share work-in-progress and receive input from interested citizens Most of these tools already exist (see Part II): votewatch.eu is an example of how voting patterns can be recorded and made public; WriteToThem and Abgeordenetenwatch both include an example of how posing questions to MEPs could be implemented; Your Voice In Europe already has the functionality needed to enable MEPs to post online consultations; and some MEPs have already adopted personal online blogging platforms to share work-in-progress and engage with citizens. The technical challenge is thus very minor in providing such tools to MEPs.
For such additional tools to have an effect on the relationship between European citizens and their MEPs, such tools would have to be both technically and strategically integrated with social media. Our case studies of TheyWorkForYou in the UK and Abgeordenetenwatch in Germany show that such functionalities in and of themselves tend to reach mainly organized interests and journalists, while social media provide a bridge through which ordinary citizens may also become involved. This reflects a tendency in which social media have grown to act as central hubs for communication and social networking in contemporary society; hubs that enable decentralized production and co-creation of ideas and even societal movements (Skoric et al. 2016). For most contemporary organizations, this tendency has produced a shift in online presence strategies from an emphasis on drawing traffic to the organization's website, to a focus on producing content that gains traction on social media platforms. Making this same shift in the EP would imply providing MEPs and their staff with the tools needed to send their 'fish hooks' into the whirlpool of social media debates in order to draw citizens onto their own platform for debate and co-creation (Dahlgren 2013).
Our case interviews indicate that initially some parliamentarians will see this as an 'extra' workload. The argument could be made, however, that online engagement is not going away, but is rather a new element of the changing role of the parliamentarian: from that of a representative of societal groups to that of a figurehead for an 'affective public' (Papacharissi 2015). In any case, it is clear that increased online engagement will make new demands of MEPs and that-as argued earliersupporting services must go beyond a compliance mindset to one of exploration and co-creation. At the same time, for online engagement with MEPs to work, it is also necessary that parliamentarians and their parties accept a certain loss of control as the price to pay for a more vibrant interaction with (the most active parts of) their constituencies.

Create a Platform for Monitoring Member States' Actions During European Council Decisions
This idea is that the contributions and votes of each Member State in relation to decisions made in the Council of the European Union should be made publicly available in an easily accessible form. Our discussion (Part I, Chap. 3) of the democratic deficit of the EU touched briefly on the Council's 'black box' function in European decision-making. This function is one of many factors that makes the EU seem to many citizens to be an outside force acting on the conditions for national policy. While there is some truth to this perception, increased insight into the actions taken by national governments in the context of the Council would help to dismantle those elements of this perception that rest on illusion or disinformation. It would also help to hold national policymakers accountable for the positions taken in the Council.
Much of the information needed to establish such accountability is already available, through the common EU web-platform, civil society services such as votewatch.eu and the web portals of national governments and parliaments. It is thus possible for the highly intrepid citizen to put many pieces of the puzzle together and to get an outline of the positions taken by nationally elected politicians in the European arena. However, not only does this place an unfair, and for most people prohibitive, burden of information gathering and analysis; key information is simply not available through ordinary channels.
Providing clear insights into the contributions and voting patterns of Member States is less a technical problem and more a question of procedure and culture. Where the line between the two is drawn, i.e. how much additional information could be made available without formal changes to the rules, is outside the scope of this report. Nevertheless, providing such information qualifies as an 'easy' step towards overcoming the division between an opaque European policy arena and the national public spheres. It would also, quite naturally, strengthen the ability of European citizens to participate in an informed basis in other, more active forms of e-participation.

Enable Crowdsourcing of Policy Ideas for the European Commission
This idea is that there is a gap in the policy formulation processes of the European Commission, which could be filled by a mechanism for crowdsourcing policy ideas. 'Crowdsourcing' is a highly ambiguous term. In this context we mean a process of gathering ideas through informal and frank exchanges of experiences and views, which is not bound to a specific phase in the decision-making process at the European level. Even assuming that both the European Citizen Initiative and Your Voice In Europe were revamped and relaunched, there would still be a gap between the functions of these two mechanisms, where early-stage policy development could benefit from open and frank sharing of ideas between European Commissioners, their staff and citizens. This early stage of pathfinding is especially vulnerable to lobbying activities by organized interests. While an online debate platform would not be a safeguard against such dominance, it would at least provide ordinary citizens with a space to engage with EU institutions in an informal manner that is otherwise only possible for lobbyists and other organized actors. The web portal Debatingeurope.eu provides an example of how such a crowdsourcing approach to the interaction between European citizens and decision-makers could be structured in an informal manner. The creation of an informal crowdsourcing platform would help the EC to seize an otherwise missed opportunity to create a space for policy debate with a more transparent and ordered structure than the one currently provided for European citizens by social media. A crowdsourcing mechanism could also help to gather ideas for how the EC should interpret and weigh different expert and stakeholder inputs. For example, tools such as Futurium produce a wealth of expert ideas, but no clear synthesis. Here, a crowdsourcing mechanism could provide a space for followup discussions in the wake of foresight exercises, where less expert participants could become involved and help to develop ideas for policy strategy. By giving decision-makers and their staff a forum for gaining immediate feedback on tentative ideas and considerations, a crowdsourcing tool could also help to create more transparency in the policy formulation process, simply by making it possible to understand the thinking that went into more formal documents.
There are obvious risks to a more open platform. The Predlagam platform (Part II, Chap. 8) is an example to learn from. On this platform, which is an initiative of the Slovenian government, participants can add a proposal on current regulation or propose new regulation. There is room for voting and deliberation both between participants and with policymakers and some feedback from the government on the proposals since they are obliged to react. Impact is low, however, partly because many of the proposals would be difficult to achieve (see examples in the case study). Interesting criticism from one of the interviewees was that the format of the tool is too open and that it should be more structured, with more information given on what kind of input the government wants from citizens. This, of course, goes hand in hand with limiting the scope of participation and bureaucratizing the manner in which a proposal must be made. The interviewed researcher was of the opinion that the open structure of the Predlagam.vladi tool would not be an issue if there were sufficient staff to process the ideas: 'The policy process is very complex. And citizens should be aware how complex it is. I don't think that they should be fooled'. Taking the lessons learned in Slovenia into account, our proposal is not to develop a stand-alone crowdsourcing platform, which would in itself risk becoming a 'black box'.
However, a crowdsourcing platform would perhaps be the ideal starting point for a one-stop-shop for online participation in European policy processes. While an online crowdsourcing platform could provide valuable input in and of itself, its main usefulness from a citizens' perspective would be as a springboard for deeper involvement, e.g. through Citizens Initiatives, EC consultations or EP petitions. From the perspective of the EU institutions, the input gathered from crowdsourcing could serve as inspiration alongside more formal expert group and stakeholder consultations. This could also help to hone the framing of consultations opened on Your Voice in Europe.
There are good reasons to explore this idea. Lironi (2016) argue that crowdsourcing platforms may enhance participation by involving civil society beyond typical stakeholder groups, as well as reaching young people, which may contribute to a learning process where both citizens and decision-makers broaden their understandings of a given topic and the range of opinions that exist on that topic. This argument is at least partially supported by our case studies of the Finnish Open Ministry and Wiki Melbourne, which both reached young people to a greater degree than is typical in participatory exercises (Part II, Chaps. 7 and 8). However, both of these case studies also showed clear tendencies towards over-representation of white, male, highly educated citizens. The main expectation of a less formal crowdsourcing platform should therefore not be that it will create a representative picture of what 'people' think. Rather, crowdsourcing is an opportunity to broaden debates by going beyond the implicit bounds that may arise in the Brussels 'bubble'.
To reap the benefits of crowdsourcing, an explorative mindset combined with ongoing commitment is a prerequisite. It would be of the highest importance that the design and implementation of such a platform go beyond mere compliance with some underlying legal mechanism. And it would also be necessary for platform development and learning to take place under relative resource stability. The case of the Finnish Open Ministry platform shows this quite clearly. The Open Ministry platform builds on an underlying legal mandate for citizens' initiatives and provides online functions to submit these initiatives. However, in its first years, the platform went far beyond the minimum requirements necessary for those functions. The Open Ministry platform thus aimed to provide a deliberative environment in which many spillover effects could be gained from the process of formulating, debating and gathering support for citizens' initiatives. The platform provided facilities for commenting on proposals, debating their possible consequences, suggesting improvements and voting for or against proposals. The platform also provided support from volunteer legal experts to draft proposals compliant with formal criteria. Over time, however, the budget available for these activities was reduced, and as a consequence, the platform gradually reverted to a bare-minimum approach. The online activities of Open Ministry around legislative proposals are now more directed to supporting initiatives with signatures rather than on deliberating proposals, and citizens can no longer take advantage of the legal support. In the interview the researcher argued: 'There is a need for some sort of legal advice to ensure that the proposals actually achieve what they are supposed to. And that's a problem because of course most citizens don't have the knowledge that they would need to ensure this'. This case shows that it is possible for online exchange platforms linked to formal procedures to grow beyond a compliance mindset and embrace a more exploratory approach, but also that the long-term success of such an approach is highly dependent on sustained support.
Other cases, such as the Five Star Movement (Part II, Chap. 10) and the Icelandic experience of crowdsourcing a new constitution (Part II, Chap. 8), show that once an online platform grows beyond a compliance mindset, it may gain vibrancy and take a decisive role in the democratic community. The Five Star movement, centering in part on online crowdsourcing of policy ideas and strategies, has thus successfully mobilized a base of support that not only rivals existing political parties but has also placed the movement as the second largest Italian party. The Icelandic case similarly shows that a crowdsourcing platform can come to play a central role in public policy discourse, but also shows that a successful e-participation tool is in itself not enough to ensure policy impact. Care must be taken to balance formal and informal structures and to be clear about the ability of citizens to influence (or not) the process of policy formulation. The recipe for success here seems to be honesty and straightforwardness.
While these cases provide grounds for cautious optimism and concern regarding the plausibility of establishing a crowdsourcing platform to supplement the ECI and Your Voice in Europe, it is necessary to take into account the limitations of onlineonly platforms with regard to the facilitation of deliberation. Earlier in this book, we reviewed recent literature on policy crowdsourcing and found some critical warnings that are of the highest relevance in this context (Part I, Chap. 4). Face-to-face deliberative processes for the delivery of citizens' input to policy are often marked by high citizen interest, are often quite cost-effective and-when supported by effective mobilization strategies-may even provide superior performance regarding the inclusion of marginalized people and the overcoming of prejudice (Collingwood and Reedy 2012). However, deliberation is no silver bullet and only provides such benefits in settings that live up to other quality criteria at the same time. One wellestablished 'fact' among observers and stakeholders is that clear outcomes only come from deliberation when it is combined with some form of aggregation. The possibility to vote (quantitative aggregation) was thus present in nine of the twelve cases in which an actual impact on decisions was detected. But many online systems that claim to reproduce the deliberative situation quite simply do not. Many such systems support the construction of group identity and community very well, but fail to facilitate a respectful and consensus-oriented political dialogue (Kersting 2013). For this reason, when the purpose of a participatory process includes mutual exploration and co-creation, many expert observations point to the necessity of mixing online and face-to-face participation in processes; even if no one expects the potential gains from such processes to come easily (e.g. Kersting 2013;Nabatchi and Leighninger 2015). Our comparative case analysis also shows that the possibility to participate both online and offline is an important condition to create an impact on the agenda-setting process.
We expect that this latter recommendation will fall on fertile ground in the European Commission, where a recognition of the necessity of blending online and offline elements seems to pervade those DGs that are currently experimenting with online engagement. The Futurium platform 6 is one example. This platform is developed by DG Connect to enable co-creative processes involving policymakers in explorative and creative deliberation of possible futures in Europe (Part II, Chap. 9). This platform blends a wide range of online options for debates with offline meetings and events in order to enable structured deliberation and knowledge exchange. The CIMULACT project funded by DG Research and Innovation is another example. This platform is developed by a consortium of organizations to enable citizens and experts to co-create visions of a future Europe and to formulate priorities and calls for research and innovation to support these visions. This project blends online consultations with offline co-creation processes to enable the broadening of participation in the formulation of research and innovation policy.
Together, these two examples show that the idea of using crowdsourcing as a feed-in to policy already has support among EC decision-makers. Given the precedence of creating a one-stop-shop for online consultations across the DGs (i.e. Your Voice in Europe), it seems to be a natural next step to seek to establish a parallel or directly connected one-stop-shop for policy-crowdsourcing.

Cross-Cutting Issues: Towards a European e-Participation Infrastructure
In the above sections, we have focused on individual mechanisms and platforms and their shortcomings and opportunities for improvement. A key cross-cutting issue is that while tools such as the ECI, Your Voice in Europe and the EP Petitions Portal have the potential to serve as vibrant bridges between different spheres of public dialogue, they fail to do so due to a number of shared failings. Firstly, none of the existing mechanisms are supported by a clear and effective strategy for mobilizing citizens to participate (Part I, Chap. 4). Observers and stakeholders generally agree that in lieu of such support, these mechanisms easily come to serve as yet another platform for elite debate among 'the usual suspects', i.e. organized private interests and social movements (see also Part I, Chap. 3). Secondly, all three existing mechanisms fail to provide feedback on the impact of their contributions to participants. Altogether, not enough effort has been put into ensuring that participantscitizens as well as decision-makers-experience their engagement with these mechanisms as rewarding.
What is perhaps most striking from a cross-cutting perspective, however, may be the weakness of follow-up and learning efforts which characterizes the implementation of existing mechanisms and the development of new ones. As already stated, it may be prudent to pursue a strategy of ongoing improvement within existing formal mandates, which seems to be, for example, what the EC has been doing with the ECI. However, if we accept that a focus on implementation may be more productive in the short term than a constant return to the question of formal frameworks, this places a responsibility on the services to pursue an implementation strategy where the improvement efforts surrounding EU's institutional e-participation mechanisms lead the field. Such a strategy currently seems to be absent. Instead, the current implementations of e-participatory mechanisms seem in many respects to aim for the delivery of a bare minimum standard. As a consequence, decision-makers and citizens are forced to look outside institutionalized e-participation channels to build the bridges for dialogue that Europe needs. Parliamentarians are thus taking to commercial blogging and networking platforms, while NGOs are attempting to provide mobilization support around the ECI. The net total of these failings is very little actual forward momentum on the advancement of a participatory approach to European decision-making.
The core question for a strategy of improving participation while staying within existing formal frameworks seems to us to be: What is the common unifying vision? As long as each of the existing mechanisms and experiments, such as DG Connect's Futurium projects, or DG Research's various pilot projects (e.g. CIVISTI, VOICES, CIMULACT 7 ) remain stand-alone mechanisms with discrete functions and implementation programmes, the EU will remain an opaque jungle to the average citizen. If, on the other hand, a unifying vision of moving gradually towards an organic European participation infrastructure was agreed upon by all involved actors, the currently separate efforts of the different institutions and services to open up European decision-making could begin to build on one another rather than carving out separate corners of a bureaucratic universe. Such unity, of course, is easier to dream up than to achieve. For that reason, we have gathered a number of practical pieces of advice, which we hope may serve as support and inspiration to ongoing work to enhance citizens' participation in European decision-making.

Unify Platform Design Around the User
• A one-stop-shop for participation would provide synergy between the EU institutions Parallel efforts in different parts of the EU institutional system to enhance opportunities for e-participation would all benefit from integration into a 'one-stop-shop' platform. Contemporary platform design has long since abandoned the traditional approach of mirroring underlying organizational divisions because it puts an undue burden on the user to decode the internal logics of the organization. Why should it be up to each individual European citizen to understand the interfaces and overlaps between the ECI, EC's consultations and the EP Petitions Portal? Conversely, why should each participation experiment have to restart the process of mobilizing citizens for participation? Why not gather these and other participatory opportunities together in a common platform? The UK Democratic Dashboard is an example of such an approach. A one-stop-shop approach could significantly reduce the risk of citizens becoming dissuaded from participation because of a mismatch between their initial impulse to do so and the specific mechanism they turn to. • Multi-level integration would help citizens to navigate European democracy more confidently Participating in the European policy cycle is not only about participation in the policy process within EU institutions. Europe includes local, regional, national and transnational processes. In the long term, efforts to arrive at a unified participation infrastructure ought to include the ambition to integrate the multiple levels of European democracy in which each citizen is involved. Without it, the many separate channels of participation available to citizens all run the risk of failing to channel citizens' wishes and concerns to the right governance levels and arenas. Also here, the UK Democratic Dashboard could be a starting point for such integrative thinking.
• User-centric design can help to keep development focused on real-world usefulness Due in part to the influence of legal expertise in public sector organizations, public sector online services tend to prioritize compliance with formal frameworks over user experience. E-participation platform designers therefore need an explicit mandate to put user experience first. Of course, this is not to say that online platforms should be anything other than compliant with formal demands. However, in terms of design processes, achieving compliance with legal requirements should be a secondary objective downstream from the development of an engaging user experience. A similar note should be made about the approach of the technical staff and contractors developing online platforms and other digital support systems for citizens' participation. A unified approach should not be read as a technically unified 'super-system'. Rather, unification should be understood from the user perspective, as a unified form of access and a homogenous user experience. Underlying such an experience may well be a number of heterogeneous systems; from the user point of view, this makes no qualitative difference.

Integrate Participation Processes
• Formal and informal dialogue and consultation are points on the same spectrum From a legal perspective, the different existing participatory mechanisms are distinct processes with discrete flows of information. From the point of view of citizens and decision-makers, however, it is more intuitive and useful to consider existing mechanisms and emerging experiments as points on a spectrum. Strategy formation in the EC services and the European-level political parties could benefit equally from the opportunity to engage in informal dialogue with citizens. Such dialogues could help to build momentum around European citizens' initiatives, and direct citizens to participate in consultations. Those issues that have no place in either might be taken up in petitions aimed at parliamentarians. A myriad of other connections is conceivable, which could be much more easily drawn in an organically overlapping e-participation infrastructure than by discrete stand-alone platforms focused strictly on each mechanism. • Expert and stakeholder consultations and citizen participation are part of the same process Drawing on experiences from technology assessment and foresight, concepts such as 'Policy Making 3.0' have sought to consolidate the insight that policy formation processes that integrate evidence gathering, interest negotiation and democratic dialogue in a structured and transparent way can provide both greater input and output legitimacy. Again, a legally oriented approach to these processes will focus on the existing rules concerning expert and stakeholder consultation and seek first and foremost to ensure compliance with these rules. However, an approach which does not take into account the need for informal overlaps between these processes and risks pushing informal dialogues into the dark. A more integrated approach would present new challenges, but would also open up opportunities for more transparency in the policy formation processes of the EU.
• E-participation and e-government are parts of a greater whole E-participation and e-government should be viewed as part of the same movement towards a twenty-first century public sector. Europe cannot afford to consider e-government as a 'need to have' while e-participation is considered as merely 'nice-to-have'; both are equally necessary.

Learn As You Go
• Starting small can help build trust The road to the digital public sector of the future is by most accounts paved by trial-and-error and building on small successes rather than top-down planning of 'super-systems'. The guiding motto for the Swiss e-voting system-safety before speed-applies here in a broad sense: better to build on good and stable results towards a long-term goal than to overreach and fall short. To be sure, this is not a recommendation for further stand-alone experiments. As discussed before, sustainability and tenacity are essential for citizens to have trust in institutions' efforts to build platforms for participation and patterns of governance around them. With long-term commitment, stand-alone experiments can in some cases do more harm than good to the relationship between EU institutions and European citizens. • Co-creation beats perfect planning every time The standard operating mode of public sector institutions, including the European institutions, is to separate decision-making and implementation. In projects involving external contractors, this separation is most often a formal requirement. This means that well-proven approaches to the building up of online communities-e.g. starting small with simple services that are obviously helpful and easy to adopt and ongoing user involvement and crowdsourcing of ideasare very difficult to implement. This operating mode also makes it very difficult to engage in dynamic partnerships with, for instance, media organizations that might help to create traffic to participatory platforms. It is a standard complaint concerning public sector ICT-development that this separation between the project owner and the developer favours planning over agility and top-down decision-making over bottom-up co-creation. Most leading advisors (from the EC Expert Group on Public Sector Innovation (EC 2013) to the OECD (2015)) therefore agree that innovative solutions to, for example, e-participation demand a new mindset. Central to such a new mindset is a focus on co-creation and a reinterpretation of formal and informal rules governing development efforts.

Organizational Support Is Necessary
• High-level support and coordination None of the above recommendations will be possible without high-level, crosscutting political support, guidance and investment. A high-level coordination group with cross-institutional participation and authority could thus be established to ensure that the efforts of different administrative bodies towards a unifying vision and strategy on e-participation work in the same direction.

Final Remarks
We end, then, on a note that is perhaps more positive than the conclusions drawn in our literature review. In the literature review, we saw that e-democracy, as it exists today, has not lived up to earlier expectations. Despite the Internet having produced vast improvements in the access to information on political decision-making, there has been little to no forward momentum in the direction of an online sphere of deliberation on diverse ideas and shared decision-making. Instead, the many-tomany communication, which the Internet makes possible, has become structured in a way that has accelerated the circulation of intentional and unintentional misinformation and tends to lock citizens into patterns of pre-existing preferences reinforced in the 'echo chambers' of subpublics. But even as this sobering state-ofplay has weighed on our analysis, we have not taken for granted that the development of e-democracy in Europe has reached the end of the road. On the contrary, having delved into the detailed mechanics of specific cases of e-participation and e-democracy we have come to realize that current practice and best practice are still quite far apart.
There are basic rules of thumb for e-democracy implementation, identifiable in the comparison of empirical cases, which are too often ignored. Our analysis shows that e-participation cannot stand alone but must be embedded in an actual decisionmaking process, that the role of the participation mechanism in this process should be clear to participations, and that it is a good idea to tell participations what came out of their input once their participation is over. These and other examples show that where e-democracy fails is very often in the very same places that democracy in general fails. The core conclusion of our literature review-that e-democracy has failed to bring citizens much deeper into the decision-making processes of public institutions-thus seem valid. But our empirical findings indicate that this failure is less a consequence of the 'e' aspect of e-democracy than it reflects the underlying relationship between democracy as such and the decision-making culture of public institutions. As long as e-democratic solutions continue to be tacked on to existing institutionalized decision-making processes, without changing the business-as-usual of those processes, it is almost a matter of mechanical certainty that participating citizens will continue to be delegated to the limited roles that were available to them beforehand.
We take this insight to be good news for the prospects of e-democracy in Europe. Because when we see that the apparent failure of e-democracy to live up to its potential has little or nothing to do with anything inherent in the new world of digital communication and everything to do with the lack of openness of institutional decision-making processes, then we are able to land our investigation in the home court of European decision-makers: in the politics of European institutional design. In a nutshell, there is no technological mystery to be solved, 'only' an institutional one. The recommendations provided in this last chapter provide some obvious starting points.
Looking forward and looking for practical steps to advance e-democracy through institutional action, our recommendations should not stand alone. In our attempts to identify the low-hanging fruits of e-democracy, we have deliberately limited the scope of recommendations to actions that could be taken by the EU institutions within their existing institutional remits. Even without going into the possibilities that might arise in the event of a future revision of the Lisbon Treaty, European regulators could take a second look at the private sector actors who provide the contemporary world with its digital infrastructure. The question that looms over current debates on the future of democracy is: What can we as a society expect of Silicon Valley? Does there not accrue some public service obligation to commercial providers of digital communication platforms similar to that of the media of the past? When newspapers, radio stations and TV stations have had a monopoly positionwhether de jure or de facto-states have imposed editorial obligations for ensuring a fact-based and open public sphere. As social media platforms grow to become critical infrastructures for democratic governance, is there a positive argument for not imposing similar obligations on the Facebooks, the Twitters and the Snapchats of the world? We expect that this will be a second leg of future discussions, necessary to improve the prospects of e-democracy in Europe.