3. E-Democracy and the European Public Sphere

Even with direct parliamentary European elections and after fostering the initiative and controlling rights of the European Parliament, the EU is still regarded by many—including European policymakers—to suffer from what has been coined a “democratic deficit” (Grimm 1995). This is due to the fact that the EC, with its growing competences as a European government, has no direct liability to European citizens, but is enacted and controlled by a multilevel system of policymaking, as a rather indirect mode of democracy. At the same time, the EU acts as a body representing European citizens, and functions as “… a uniquely large and complex body of specialized decision-making, often operating outside the control of formalized and territorially bound systems of representative democracy” (Michailidou and Trenz 2013: 260).

In a summary of the democratic deficit discussion and with a view to the current EU crisis and widespread criticism of the EU institutions’ crisis management, Habermas (2015: 547) refers to the term by noting an “increasing distance separating the decision-making processes of EU-authorities from the political will formation of European citizens in their respective national arenas”. Thus, the “democratic deficit” of the EU institutions continues to be a central feature of discussions about the future and the further development of the European Union. It becomes more relevant with the obvious problems of how to arrive at an integrated widely accepted European solution with regard to problems such as the financial crisis or the EU refugee policy. The central question is then, to what extent can the European institutions evolve into a European government with extended responsibilities (particularly with regard to social welfare and transfer politics), and how can this European government be democratically legitimised? Or is it unrealistic to expect this because the given diversity of Europe in terms of political culture, economic power and development of social welfare cannot simply be overcome by democratic structural reforms (e.g. strengthening the European Parliament and a European government elected by the Parliament)?

In a recent seminal dispute between public intellectuals about the future of Europe, both positions are prominent: On the one hand, it is held that there is a need to expand EU competences and that this has to (and can) be done by democratising the EU political system, in order to overcome the democratic deficit as well as the symptoms of the EU crisis (Habermas 2014a, b, 2015; Offe 2013a, b). On the other hand, there is the notion that it is the crisis itself which makes it obvious that further integration of Europe, as well as a way out of the legitimisation crisis by giving more competences to Brussels, is not an option (Scharpf 2014, 2015; Streeck 2013, 2015). Both positions hold that EU politics, as executed by the European Commission and the European Council, lacks democratic legitimation and responsiveness to the European citizenry. However, while from the one perspective (Scharpf, Streeck) this leads to the demand of restricting the competences of the EU to the advantages of national governments, keeping the EU in the status of a “regulatory state” (Majone 1996; see Lindner et al. 2016), from the other perspective (Habermas, Offe) this leads to demands of expanding the competences of the Union at the costs of national governments and at the same time strengthening the democratic legitimation of the European institutions by making them subject to direct elections and control by the European citizenry.

Especially, the Euro crisis has led to vivid debates about the legitimacy of the EU institutions’ policy, the dominance of the European Central Bank (ECB) over the Parliament and the relation between national and European sovereigns, in the context of a discussion about EU fiscal policy, which cannot be discussed in detail here (see contributions in de Witte et al. 2013; see also Streeck 2015). It is, however, striking with regard to the Euro crisis that for both positions held with regard to the right fiscal policy, the democratic deficit “strikes back”. Those who support the austerity and neoliberal programme of forcing Greece into a process of lowering the level of social welfare are confronted with the accusation that technocratic institutions are overruling democratically legitimised governments in Member States without themselves being backed up by democratic legitimisation. For those in favour of reacting to the crisis by installing a European fiscal and social policy (including a fiscal union), backing by a transnational solidarity consensus is needed, but is not in sight, given the apparently even deeper separation of the EU citizenry and the public sphere by predominant national interests.

Fritz W. Scharpf’s (1999) influential distinction between input and output legitimacy helps to understand the particular relevance of the democratic deficit in times of crisis. Due to their transnational character, the EU institutions’ legitimisation cannot be rooted in strong channels of information from citizens to the EC (input legitimacy) and thus must rely on legitimising its policies by the quality of its output, i.e. its decisions and regulations being in the best interests of, and thus supported by, the citizenry. The fact that in the latter respect the means of the EU institutions are also restricted has a special bearing in times of crisis. The missing input legitimacy becomes all the more problematic the weaker output legitimacy becomes, with apparent difficulties in establishing a consensus on a European way out of the fiscal crisis, or a joint European policy to solve the refugee problem. In a situation where strong decisions have to be taken at the EU level (beyond national interests), input legitimacy is urgently needed. For this reason, some regard a rearrangement of the EU institutional setting to be necessary. In order to (re)establish the bridge between the European citizenry and the European political elites, a strong European Parliament is needed. This implies a European electoral contestation between European (not nationally defined) political parties, on the basis of which the European Commission would be transformed into a government which is accountable to the Parliament representing the European citizenry. Offe (2013a), regarding the growing competence of institutions like the ECB, considers it to be detrimental to the project of European integration that those institutions which are “farthest remote from democratic accountability” have the “greatest impact on daily life of people”. He regards this to have developed into a “deep divorce between politics and policy” at the European level. Politics is based on often populist national mass politics with limited implications for the lives of people, whereas policymaking becomes an elitist matter that “has no roots in, no links to nor legitimation through politics” (Offe 2013a: 610). Also, in Habermas’ view the democratic deficit has deepened in the course of establishing the Fiscal Compact and the European Stability Mechanism, because the European Parliament alone did not benefit from the increase of competences of the EU institutions (Habermas 2015: 551).

As a way out of the democratic deficit as well as of the crisis of EU integration, Habermas (2015)—despite the current climate against it—considers far-reaching institutional reform to be necessary. Due to the transnational character of the European Union, the democratic legitimisation of the institutions has to be backed up by a double “sovereign”, represented in two chambers: on the one side the Parliament (citizenry) and on the other side the European Council, which he would like to see as the second legislative “leg” alongside the Parliament, representing as a second sovereign the European Member States and their peoples (“House of States”). He regards this to be the way to take account of both the transnational character of the European Union and European citizens’ interests in having their ways of living and wealth protected by their national governments. This is in line with the conceptualisation of “transnational” democracy in Articles 9–12 of the Lisbon Treaty (Bogdandy 2012). The concept of “people” is reserved for the nation state, whereas individual citizenship (with individual political rights) is seen as the foundation of democratic legitimacy of the European Union.

On the other hand, it is argued that the conditions for expanding the competences of the European Parliament and deepening integration are not given. The crisis, requiring strong decisions on redistribution of resources, which have to be taken according to the majority rule, reveals that European solidarity is weak. The crisis has obviously brought about a reorientation towards national interests. As Offe (2013b: 75) puts it: the bank crisis has been transformed into a crisis of state finances (via the obligations that have been taken over by national governments to save the banks), which has turned out to cause a crisis of European integration, where rich countries force poorer countries into an austerity policy in order to re-establish trust in the financial industry. This has widely led to “a renationalization of solidarity horizons” in the European Union (Renationalisierung der Solidaritätshorizonte). Thus, what is needed for institutional reform and a further integration of the EU is lacking more now than ever before (Scharpf 2014; Streeck 2015). More generally, it is believed that the heterogeneity of Europe with regard to local, regional and national ways of living and economies only allows for a democratic European constitution that acknowledges these differences by way of far-reaching autonomy rights, which, with regard to financial constitutional questions, implies low mutual obligations of financial solidarity among partners (Streeck 2013).

Thus, beyond any debates of the actual problems of European policymaking, the discussion points to the more fundamental problems of European citizenship and the European public sphere. It can be argued that especially in times of crisis, it would be necessary to legitimise far-reaching decisions that will deeply influence living conditions in the European Member States through a vivid process of deliberation about pro and cons, needs, demands and duties. This, however, appears to have even less chance of being fostered, precisely due to the crisis mechanism that leads to focusing on national interests (Scharpf 2014; Streeck 2015).

Is there enough homogeneity and a European citizenship that can motivate European integration, and is there a European public sphere that can provide the fundaments for joint democratically legitimised European political action? In the following, we will first discuss the question of the European “demos” and then turn to actual research on the state of the European Public sphere.

It has always been a major pillar of the legitimisation of the European integration project that it will bring about increasing prosperity and general welfare through stimulating economic growth. Thus, it cannot come as a surprise that in an economic situation causing obvious difficulties in achieving consensus about common solutions at the European level, citizens expect their national governments to look at national economies first and protect them against a loss in welfare. Accordingly, based on Eurobarometer data, a study of the average European’s identification with Europe as a part of their identity as citizens showed that “a sense of being European” dropped significantly in many European countries during the financial crisis in the period of 2005–2010. The general decrease in identification with Europe was strongest in those countries which suffered most in terms of decrease of per capita GDP or increase in unemployment as a result of the recession caused by the financial crisis, namely, the Baltic states, Great Britain, Italy, Ireland, France and Greece (Polyakova and Fligstein 2016).

It is this observation of weak European solidarity and the predominance of national perspectives that feeds the so-called “no demos” discussion among scholars of European politics. The debate dates back to the 1990s and starts from the notion that in order to work, democracy needs to be rooted in a “demos”, a political (as opposed to an ethnic) community which is rooted in “a strong sense of community and loyalty among a political group”—this being, as it were, the socio-cultural prerequisite of democracy (Risse 2014: 1207). The assumption that a “demos” of this kind does not exist at the European level, but only at the national level, and that the different “demoi” of the Member States do not form a meta-national demos, implies that democracy at the transnational level cannot (and must not) be based on input legitimacy, but mainly on the quality of the output of the political system. Authors such as Scharpf (see above) hold that due to the heterogeneity (cultural as well as economic) of living conditions in the Member States, there is no basic consensus—or subjectively felt citizenship—which could function as a cultural backbone holding the community together against conflictive majority decisions in the (reformed) European Union. The acknowledgement of majority decisions that might be against their own interests (at least until the next elections) can only be expected on the grounds of an implicit cultural consensus based on shared citizenship.

In other words, no pre-political community exists for integration at the European level that is comparable to integration at the level of the nation state. Europe is not a nation state but can be thought of as a “mixed commonwealth”, in which national and supranational identities coexist with each other. Europe “… possesses aspects of a nation, but it is a rather watered-down version of it […]. It relies on a body of treaties that provides a framework of ‘constitutionality’ but without a constitution. It offers membership, but subordinated to the stronger Member State form. Its members are related, but with a link much weaker than that of ordinary polities. Such a link is based on some commonalities—which ground a very vague shared political identity among its members, but not comparable with political identities at nation state level” (Lobeira 2012: 516).

This type of observation is not contested by those who are optimistic about the possibility of European citizenship; however, they hold that it neglects the specific character of transnational compared to national citizenship (Habermas 2015). Instead, it is argued that European democracy is not in need of a “demos” in terms of a cultural (national) community, but that European citizens’ commitment to the fundaments of the European political constitution is sufficient to establish a new form of “citizenship”. This “constitutional patriotism”, together with a well-functioning European democracy, would be sufficient as a solidarity fundament for the European Union (see also Lindner et al. 2016).

With the current conflicting mode of policymaking in the EU, this position is confronted with new scepticism. A prominent observer from abroad, the US philosopher and communitarian thinker Amitai Etzioni (2013), considers neither the democratic deficit nor a weakness of the European political system to be Europe’s current main problem. From his perspective, what proves to be crucial in this crisis is what he calls the “communitarian deficit”, i.e. the lack of a post-national sense of community or European citizenship: “The insufficient sharing of values and bonds—not the poor representative mechanisms—is a major cause of alienation from Brussels’ and limits the normative commitment to make sacrifices for the common good” (Etzioni 2013: 312). He holds that there needs to be more than constitutional patriotism to establish enough solidarity to solve the problems of economic disparities among the European Member States: “Membership in a more interdependent EU involves not just rights but burdensome duties (such as bailing out the Greeks) that will only be voluntarily met if citizens feel the value of communal obligation to those beyond their national borders” (Etzioni 2013: 315).

Contrary to this communitarian view, it is believed that European citizenship cannot be understood according to national citizenship coupled with cultural identity as it emerged with the constitution of the nation state in early modernity. Trans national identity or citizenship and the related sense of belonging “… involves starting from a different standpoint, one that sees belonging as an identity ‘in the making’ and that imagines it to be ‘deterritorialised’ and set in a transnational dimension” (Scalise 2014: 52). Indeed, there are at least weak indications from some qualitative research that a “mixed identity” can be found in Europe. Based on interviews with a group of 40 people from a local community in Italy, Scalise (2014) undertook to reconstruct the narratives about Europe that emerge and are shared among “average” European citizens, and highlights this specific type of identity and citizenship in the following way: “Different narratives of Europe are shared among Europeans: stories related to the cultural and historical roots of the continent, institutional and ‘official’ narratives of the EU, biographical stories weaved together with collective memories. Multi-level stories, a mixture of values and references coming from the local and national heritage and linked to the European post-national plot. In the broad range of the narratives which have emerged, the influence of the local context, where the stories originate, can always be identified. The stories of Europe are embedded in the regional territories. […] There is a dynamic relation between the local, national, supranational and transnational dimensions. These levels interact in the European identity construction process” (Scalise 2014: 59).

To what extent this can be regarded as an indication of a general European identity is of course an open question and has to be confirmed by much broader research approaches. Beyond this, however, proponents of a further integration of the EU base their cautionary optimism with regard to the “Europeanisation of European citizens” in the further development of the discourse about Europe and, thus, in the further development of the European public sphere (see the next section). From this perspective the development of European identity and solidarity depends on the chances and opportunities to discuss and define what is in the common European interest via a common European political discourse. For this—and the democratic deficit comes into perspective again—it is necessary for the European Parliament to function as a European public space, which foremost implies that societal interests and political debate on the “common good” are not organised alongside national party lines but are fostered by European (transnational) party groups (Habermas 2014a: 94). A European party system is a precondition for overcoming the national restriction of perspectives and would prepare the ground for a will formation at the European level, i.e. in the light of shared (and not nationally divided) normative principles of social justice, and with regard to shared assessments of the problems at stake and the way out (Offe 2013a: 606 f.). The constitution of the EU generally has the effect that European citizens in their national contexts are not confronted with alternatives of European policies to be discussed publicly, but are just affected by the results of EU policies decided on by the EC and the Council of Ministers. From this perspective, identity is not culturally given (as supposed by a communitarian perspective; see Etzioni, above), but evolves in a political process. In this way, citizenship must (and can) emerge out of debates and conflicts about the public good—as was the case for national identities in the conflictual emergence of the nation state (Habermas 2015). Thus, it is important to what extent the EU polity allows for a vivid political discourse among citizens. From this perspective, it clearly makes a difference whether the citizenry is consulted by means of e-participation methods (see below), or to what extent institutional innovations, like the European Citizen Initiative, add to the set of citizens’ rights by giving citizens a voice in lawmaking (Ene and Micu 2013).

In this respect the contestation of European issues in the context of the fiscal or refugee crisis is regarded by many as not necessarily an indication of disintegration, but as an indication of the Europeanisation of politics. Scholars of European politics thus speak of a “politicisation of the European integration” with positive conno tations, meaning that there is an observable tendency to publicly address the issues and problems of European multi-level democracy (Wendler 2012; de Wilde and Zürn 2012; Hooghe and Marks 2008). The contestation of European integration has for around a decade been observed to be based in conflicts around cultural identity (Hooghe and Marks 2008; Kriesi et al. 2007), in the lack of compatibility between national and supranational institutions (Schmidt 2008) and in resource and distribution conflicts in the context of regulation of the European market (Hix 2009; Majone 2002). The so-called “permissive consensus”, characterised by wide implicit EU-scepticism, but where citizens do not engage with EU issues and leave the playing field to political elite with the effect of the de-thematisation of Europe in national public spheres, has come to an end. Underlying conflicts have now come to the fore and made Europe a public issue (Hooghe and Marks 2008).

Politicisation of European integration is then regarded as being driven by an expanding public discourse that provides for transparency of decision-making, includes civil society and provides for critical feedback to decision-makers, thus having a “democratising function”. The contestation of Europe in the actual crisis is also regarded as being a sign of a functioning Europeanised public sphere with a potential for democratic reform of the European polity (Statham and Trenz 2012, 2015). However, insofar as this reform does not take place, the weakness of the EU institutional system will be further exposed in the public sphere, which will foster scepticism even further (ibid.)

Despite the obvious fact that politicisation of the issue of European integration in the Euro crisis is accompanied by national interests and the dominance of national stereotypes in national public debates, many observers (Risse 2015c; Hutter and Grande 2012; Rauh 2013) hold that politicisation, when coupled with an opening of national public spheres in terms of Europeanisation (see below), can be regarded as an indicator of increased awareness of the relevance of Europe for Europeans. It depends on the framing of EU issues whether or not the growing politicisation of EU affairs increases the sense of community in Europe. In this respect, it is also held that politicisation of European integration is clearly induced by the growing authority of the European institutions since the 1980s. It is therefore believed that it is decisive to actively address and deal with the problem of the growing authority of the EU institutions and the need to back this up by fostering their democratic legitimacy (de Wilde and Zürn 2012).

An overview of recent research about European identity even concludes that there are indications for a gradually growing identification of citizens with Europe as well as the Europeanisation of national public debates. In a 2013 Eurobarometer poll, 59% of polled citizens showed some degree of identification with the EU while only 38% identified exclusively with the nation state. No significant divisions could be found in creditor and debtor countries of the Eurozone in this respect (Risse 2014: 1208 f.). There are also indications that identification with Europe does not mainly come as a symbolic attitude. According to Kuhn and Stoekel (2014), polling data shows that the more people identify with Europe the more they are also prepared to support policies of economic governance with re-distributional effects to overcome the Euro crisis. Thus, the crisis and related conflicts do not necessarily lead to reduced solidarity. Risse (2014: 1210) summarises that available opinion poll data challenges the “no demos” thesis, leading to an optimistic notion that “… the European polity is more mature than many scholars assume. A sense of community does exist among Europeans and this community might even be prepared to accept redistributive consequences”. In the same vein, based on opinion polls and long-term panel research with citizens from six European countries, Harrison and Bruter (2014) conclude that there is evidence that the politicisation of European issues can be seen as a cause as well as a result of the emergence of a European identity. This implies that the more people appropriate themselves to the European polity, “the more politicized is their perception of their—thereby appropriated‐system” (Harrison and Bruter 2014: 166).

However, these optimistic conclusions are not uncontested, and with a look at nationalist and populist EU-scepticism all around Europe, not least after the Brexit vote, it can reasonably be argued that neither politicisation nor an increase in Europeanisation of media reporting in the current crisis (if observable) will have positive effects on identities. The crisis clearly brings new forces and actors to the fore that are not supportive of European integration and offer views that focus on national interests and thus help to strengthen national identities (Checkel 2015; Checkel and Katzenstein 2009). Based on a media analysis in six European countries up to 2012 (thus including the Euro crisis), Grande and Kriesi (2015) report a substantial increase in politicisation as well as Europeanisation of public discourses, but are sceptical with regard to the effects of this on the identification of citizens with Europe. They hold that since negative framing of the European integration goes across the left–right party political cleavage, politicisation under the given political structures will contribute to more EU-scepticism.

There is, however, consensus—also among observers with a more pessimistic view on the current state of European solidarity and citizenship—that the European public sphere has a strong bearing on the development of a European identity: “It is in public debate that collective identities are constructed and reconstructed and publicly displayed thereby creating political communities” (Pfetsch and Heft 2015: 30). It is therefore decisive to understand to what extent a public sphere in Europe exists.

According to a recent review of scholarly discussion on the European public sphere, the following can be regarded as being consensus among researchers: “The concept of a utopian European public sphere, defined as a singular supranational space that echoes the national public sphere, is nowadays rejected in the literature under the evidence of a missing common European identity, the lack of significant purely European media, and communication difficulties, namely language differences” (Monza and Anduiza 2016: 503).

The European public sphere is almost exclusively conceptualised as the Europeanisation of national public spheres. Europeanisation is then observable by a change of national public spheres in three respects: (1) European issues, policies and actors are visible in the “national” public spheres, i.e. in mass media coverage of political issues, (2) there is reference in national media not only to EU policymaking actors (vertical) but also to actors from other European Member States (horizontal) and (3) the same issues are addressed in the different national public spheres and similar frames of reference or claims and arguments are put forward. In all these respects, there is apparently a consensus in research that a Europeanisation of public spheres has taken place. Mass media studies have shown that over the past 15 years, national publics have become more European in terms of the visibility and salience of EU issues and actors, the presence of other Europeans in national public spheres, as well as with regard to the similarity of interpretative frames of reference or claims across borders, without the existence of European-wide media (Koopmanns and Zimmermann 2010; Hepp et al. 2012; Sicakkan 2013; Risse 2014, contributions in Risse 2015a). It can be said that positions claiming that the emergence of a European public sphere is impossible, due to structural or mainly language barriers (e.g. Grimm 1995), are almost no longer visible (Risse 2015b: 3). Meanwhile, the Europeanisation of national public spheres has also been found by many issue-specific media studies, such as in the discourse on the EU Diversity directive in France (Dressler et al. 2012), the media coverage of the EU’s growth and job strategy (de la Porte and van Dalen 2016), media coverage of the discussion about a common EU foreign and security policy (Kandyla and de Vreese 2011) or by analysis of references to the “European citizen” in national media (Walter 2015). Additionally, a methodological approach which differs from the usual news content analysis supports the notion of horizontal integration of the mediated European public sphere. Data provided by Veltri (2012) via a network analysis of information flow between Central European high-quality newspapers (UK, Germany, Spain, France) indicate that from 2000 to 2009 the information flow became less dense among the newspapers analysed, but that a more balanced network of information flows among European newspapers took shape that can be interpreted as a signal of a qualitative transformation of the European communicative exchange in the direction of horizontal integration. Beyond these general findings, in recent years research on the European public sphere has brought about results that help to understand in what respects we can speak of Europeanisation and also what its limits are. Despite growing interest in Internet-based public communication and the role of social media (see below), the vast majority of empirical work regarding the European public sphere is still on the coverage of European issues, actors and institutions in (national) mass media (mainly press). The vast amount of studies published in recent years cannot be covered in detail in this chapter. In the following, we briefly sketch the most interesting findings.

As with the debate about European identity, the notion of a “politicisation” of Europe and the future of European integration is prominent in discussions and research on the European public sphere. Despite the strong current of EU-scepticism that comes with it, politicisation is believed by many to indicate a vitalisation of the European public sphere (Statham and Trenz 2015). “Politics is back”, as a volume with contributions on the latest state of debate about the European public sphere puts it (Risse 2015a). Europe—its future, its mode of policymaking and its democratic legitimacy—is an issue of vivid public debate more than ever before. A few empirical studies are available on the effects of the crisis and the subsequent politicisation of European integration on the “European public spheres”. Their results show a growing dominance of national perspectives and interest in public discourse on the EU, but do not necessarily dismiss the notion of a European public sphere.

Findings of a large-scale media content analysis of newspaper and television news in the EU-15 (1999), EU-25 (2004) and EU-27 (2009) in relation to European Parliament elections show that media coverage of EU issues is dependent on the elites’ or parties’ positions in the respective countries (Boomgaarden et al. 2013). The more disputes among elites and national parties about European issues, the more Europe becomes visible in the national media—which, however, implies a strong position of EU-critical positions. It could be shown that “…increases in EU news visibility were strongest in a situation in which there was both increasing negativity about the EU in a country’s party system and increasing party disagreement about the EU” (Boomgaarden et al. 2013: 621). The authors conclude that “… ironically, euro-scepticism, in the form of elite polarisation, is one of the best chances for improving EU democracy by sparking news coverage of EU affairs” (Boomgaarden et al. 2013: 625).

A study (Monza and Anduiza 2016) focusing on exploring the visibility of the EU and European subjects in national media during the financial and Euro crisis started from the plausible hypothesis that with the strong consequences of EU policies, the salience of EU issues in the news should have been increased, especially in those countries that are subject to the EU austerity policy. The study—based on a set of articles with reference to the crisis, recession or austerity from leading newspapers in Germany, Greece, Italy, Spain, Poland, Sweden, Switzerland and the UK in the period of 2008–2014—found that the visibility of European actors (in terms of claims made related to the crisis as reported in the media as well as the addressees of claims) was surprisingly low when compared to national actors, indicating that the national perspective and national policymakers were dominant in the crisis discourse. The visibility of EU actors was highest in Germany (11% of the sample) and Greece (11%), and lowest in the UK (4%) and Switzerland (1%). Differences in the visibility of EU actors and issues were not correlated with the countries’ degree of negative effect from the recession. The relative prominence of European issues and actors in Germany and Greece can be explained by the German government’s leading role in debates on the Greek bailout and by Greece being the main addressee of the European institutions’ austerity policy.

An analysis of news coverage of the Euro crisis (2010–2013) and the 2009 parliamentary elections in online media news platforms, held by leading national newspapers or TV channels in 13 European countries (Michailidou 2015), found that in all countries covered by the study the EU was uniformly contested and criticised from the perspective of national politics, which in all cases were the key defining frame of media reporting. The results of the content analysis suggest that there is indeed a European-wide pattern of discussing the financial crisis and the role of EU politics, mainly made up by EU contestation, which refers to the issue of (lacking) democratic legitimation of EU politics, but mainly in a diffuse or emotional manner. Interestingly, especially concerning the online comments on EU news posted by readers, the study showed that, “…democracy is the most frequently used category to contextualise or justify not only Eurosceptic comments but evaluations of the EU polity across the entire ‘affirmative European to anti-European’ spectrum. What unites the user community is its anti-elitism and self-understanding of constituting the people’s voice that mobilizes in defence of the representative system of democracy or more frequently against the corrupt, decaying version” (Michailidou 2015: 332). What appears to be interesting about this finding is that the crisis apparently brought the issue of democracy into the centre of the debate, thus stressing the relevance of the “EU democratic deficit” issue for the European public sphere in times of crisis.

In the politicised and Europeanised national public spheres, the national perspective in times of crisis appears to be dominant, but a study on the few broadcasting formats at the transnational European level found the framing of European issues in terms of European solidarity to be dominant (Williams and Toula 2017). In an analysis of the debate programme “Talking Europe”, which is produced with the sponsorship of the European Parliament and the Commission and has been broadcast on “France 24” since early 2009, it was found “… that the solidarity frame is used to define problems and causes of issues and events as attributable to a lack of solidarity between EU members and also to present the solution of increasing solidarity as a means to enhance policy and practice. Moral judgments are introduced to cast blame on those actors who do not demonstrate solidarity. Problems framed in terms of solidarity deficits are then remedied through three-pronged solutions of integration, harmonization, and calls for greater solidarity” (Williams and Toula 2017: 8). The analysis focused on episodes dealing with the Eurozone crisis between January and November 2011.

All in all, empirical research on the effects of the crisis on the Europeanisation of national media publics appears to show mixed results. “Politicisation” is an indicator of European issues coming to the fore of national agendas, but this, of course, does not necessarily lead to issues being framed as questions of common European concern requiring European solutions. It depends on discursive structures and dynamics of whether politicised debates about Europe foster European common thinking and identities or renationalisation. Politicisation must come with Europeanisation: With a view to Brexit, it is interesting to note that research has shown that in the British public sphere Europe is highly politicised but Europeanised to a much lower degree (with respect to frames and visibility of European actors, Koopmanns and Statham 2010; see also Koopmanns 2015). It thus appears to be important whether or not pro-European elites use the politicisation of Europe for rethinking the democratic structures of the EU and/or actively engage in a discussion about options to address the democratic deficit (Risse 2015c). The fact that issues of European integration, European democracy as well as modes of European governance are found to be increasingly contested in the European public spheres does not necessarily imply dysfunctional workings of European political communication. Contestation and conflict are as much a necessary function of the public sphere as striving for consensus and compromise, or aggregation of political will. It can even be said that if empirical studies should find a lack of disagreement on central constitutional issues—as are at stake in debates on the legitimacy of EU policy interventions in times of crisis—this might not be taken as an indication that “desired consensus processes has run its course in the European public spheres”. It could also be seen as an indication that public arenas are not yet “fully developed and utilized in a truly democratic manner” (Føllesdal 2015: 254). In other words, as is held by a broad scope of scholars of European studies, the contestation of Europe as a democratic project is as much an indication of a failure of the European public sphere as the long period of the so-called “permissive consensus”—with a low level of discussion about European integration—was an indication of its functioning.

The discussion of media analysis of the European public sphere has brought up some shortcomings and deficits of research that should be taken account of when interpreting research results. Especially with regard to the increasing EU criticism in some new Member States, it has to be kept in mind that research has so far mainly focused on the old EU15, and there is not much data about the development of the public sphere in new Member States (especially in those countries currently being front runners of EU-scepticism). Data showing a widely Europeanised public sphere is mainly from central and western European Member States (with the exception of the UK). On the one hand, research would expect that “in the course of time differences in the Europeanisation of old and new Member States seem to vanish”, since they observe a “pattern of catch-up Europeanisation” (Kleinen von Königslow and Möller 2009: 101, cit. Pfetsch and Heft 2015: 45). Others hold a more pessimistic view with regard to this “modernisation story” and point to the different historical backgrounds of new Member States in Central and Eastern Europe, as well as cultural aspects like a strong and orthodox religious current in some countries that might persist (Checkel 2015: 236 f.).

Since media analysis is mainly done by using data from quality newspapers, which are read mainly by elites, the finding of a step-by-step Europeanisation found here might not apply for TV or other newspapers, which are the reference for the average public (Koopmanns 2015). It has also been stressed that we have to take into account that the focus of research on elite mass media communication neglects the relevance of new Internet-based communication networks mainly applied by social movements, which can be regarded as a Europeanisation of public spheres “from below” (Bennett et al. 2015). The focus of research on mass media might on the one hand overstate Europeanisation, but on the other might underestimate the diversity of publics and their segmentation, the latter coming into focus in research on political communication via the Internet.

Bohman (2004; see also Lindner et al. 2016) regarded the Internet as opening up a new mode of transnational political publics, due to the possibility of allowing for communication across the restrictions of time and space and also national and linguistic boundaries. He thus expressed some optimism that, while we find a decline of national public spheres with passive audiences and disenchantment with politics, the Internet could support the emergence of a transnational public sphere that is more inclusive and deliberative and is rooted in a transnational civil society. But such far-reaching expectations are now rarely put forward.

There is, without doubt, a growing importance and public visibility of so-called “dot.​com” protest platforms and social media-based exchange across national borders on humanitarian or environmental issues, which is held to show features of an emerging global civil society. Local protest movements can have outreach to the world, make their demands known and gather support globally—e.g. the movement of the outraged in Spain and Greece. The “Occupy Wall Street” movement managed to engage on a global level via social media. World economic summits and climate change summits are regularly accompanied by online mediated activities of NGOs. Thus, globalisation at the economic and governmental levels can be regarded as being complemented by an Internet-supported global civil society organising counter- or protest discourses (e.g. Frangonikolopoulos 2012). Bohman (2004; see Lindner et al. 2016) conceptualised this Internet-based transnational public sphere as consisting of multiple issue-related publics, thus creating a public of publics with a distributed rather than a centralised structure. Additionally, currently observers who are more sceptical with regard to the emergence of a trans national public sphere underline the capacity of Internet communication to induce global political communication in an “… indirect and networked sense‐not as a supra-national sphere, but as a multitude of mediated and unmediated discursive processes aimed at opinion formation at various levels, interconnected directly and indirectly” (Rasmussen 2013: 103). If this is a correct description of the Internet’s structure as a public sphere, then the decisive question is to what extent these multiple publics are related to or cut off from one another. Smith (2015) regards the ease of creating new websites or digital platforms (by everybody and for any purpose) as a political “double-edged sword”, as it makes it “both easier to create common realms open to all and to leave the common world and create one’s own little realm where no opposing viewpoints can be heard” (Smith 2015: 256). Thus, the Internet has the potential for both creating new public spaces and weakening the general public sphere.

A fundamental critique of the discourse on the virtual public sphere looks at the economic fundaments of social media and peer-to-peer networks. From this perspective the democratic potential of the Internet’s ability to allow for self-production of content, independent of the restrictions of the mass media, is called into question. The explosion of self-production and exchange of content is regarded as fundamentally based on a growing economy of transmission and exchange of data by providers such as Google or Facebook. With a view inspired by Michel Foucault’s analysis of governmentality, Goldberg (2010) concludes the following in his criticism of the scholarly discourse on the “virtual public sphere”: “On the internet there is no ‘debating and deliberating’ that is not also ‘buying and selling’[…]; participation is a commercial act. Every instance of participation involves a transfer of data which has been economized, driving the profitability and viability of the networking industry and of internet based companies like Google that cover infrastructure costs through innovative advertising, ‘freemium’ business models, and other methods” (Goldberg 2010: 749).

Jürgen Habermas, one of the most important thinkers of the model of deliberative democracy, appears to be rather sceptical as regards the potential of the Internet to foster a modernised, renewed democratic sphere of public discourse when postulating the decisive and indispensable function of a lively public sphere for modern democracy. When asked in an interview in 2014, “Is internet beneficial or unbeneficial for democracy?”, his answer was “neither one nor the other” (Habermas 2014b). He substantiates this notion by referring to what in his view was and is the central function of the public sphere for democracy, which allows for the simultaneous attention of an undefined number of people to be paid to public problems. Despite increased transparency and access to information for everybody as well as the option to make every reader an author of statements on the web, the web in Habermas’ view does not help to concentrate the attention of an anonymous public on a few political important questions. By opening up a vast scope of single-issue spaces, the web rather “distracts and dispels”. The web thus is a mare magnum of digital noises containing billions of communities as dispersed archipelagos and is not able to bring about a space of common (public) interests. In order to bring about concentration, the skills of good old journalism are still needed (all quotations from Habermas 2014b).

The conclusion by West (2013, following Dahlgren 2005) that the Internet may be best understood as an agent of mobilisation of sub-publics with regard to all kinds of issues as an “extension” for the mass media public sphere appears to catch a seminal feature of Internet political communication, but also underlines the restrictions of its democratic potential: “The ability of the internet to quickly rally people, as in the 2011 ‘Occupy’ movements, is difficult to contest. But, as subsequent events has (sic) shown, the ability of the new electronic media to transform those movements into lasting social change, or to use the new media as a public sphere whose discourse must be reckoned with, is not yet evident” (West 2013). In this respect the conclusion we drew a couple of years ago (Lindner et al. 2016) that the Internet is at best an emerging public sphere would still hold. However, substantial differentiation with regard to different modes and formats of political Internet communication is visible in the scholarly debates and empirical research in publications of the past few years.

The widespread use of new modes of political communication via the Internet indicates that Internet communication is indeed about to modify the public sphere from one mediated by mass media (and mass communication) to one mediated by a multitude of networks based on the interpersonal exchange and interactivity allowed by the Internet. From this point of view, the public sphere then exists of a network with nodes being made up by web-based spaces of political discussion, organised on websites or social media sites held by individuals, social interest groups, governmental authorities or political parties. Ideally, these different nodes are connected to one another so that the different issue-related or socially organised political communication spaces are not completely isolated but form some sort of new networked public sphere. As far as such networks also reach across national borders, one might speak of transnational public spheres emerging from below, rather than from above through the mega television networks offered to world audiences (Munteanu and Staiculescu 2015).

Transnational issue advocacy networks of Civil Society Organisations (CSOs) mainly mediated via social media (Twitter, Facebook, YouTube) and NGO websites are held to have the capacity “to engage large publics directly” and bring them into contact with government institutions, enabling people to coordinate action across national boundaries (Bennett 2012: 6). In the social science literature on the political relevance of social media, there are both expectations that social media have the potential to empower underrepresented interests, as well as more sober assessments which doubt that social media will help to reduce inequalities in the political sphere. Despite these far-reaching and contradictory expectations, a recent analysis of literature on interest groups’ use of social media concludes that “systematic, quantitative literature of social media use of interest groups is scarce” (van der Graaf et al. 2016: 121).

The new modes of communication via the Internet have obviously modified formats of mass communication. Nowadays, there is probably almost no mass media which do not host an Internet-based news site besides their print or broadcasting versions. These news sites regularly have comment sections, which offer the opportunity for online readers to comment on and discuss the news articles offered at the site. Thus, previously passive readers have the option to publicly express their political thoughts and ideas. A media content analysis including a study of reader’s comments via online news sites, published in 2015, found that at the beginning of the period of research in 2009 social media was not very well integrated in online news sites. Since then, all news sites have incorporated social media “sharing” functions, and in the EU “… readers’ participation through Web 2.0 functions has thus dramatically increased”. This was found to apply particularly “… in Southern Member States where internet availability and use was previously lagging behind the North-Western countries” (Michailidou 2015: 331).

Social media currently also function as a news source for mass media. Facebook and Twitter posts trigger mass media reports. Especially, online portals of mass media not only have their own Twitter or Facebook accounts but allow readers to forward news from online news portals. Mass media also regularly include social media posts in their news and reports about political issues. In this respect, there are channels that allow content from segmented and issue- or community-specific publics organised via social media to find its way into the general political public sphere.

Social media are increasingly used by interest groups and play an important role in political campaigning and organisation, and the coordination of political activities, since they are supportive of building communities around certain issues and interests by direct communication with supporters. There is no doubt about the growing importance of social media for political communication (Chalmers and Shotton 2013; Obar et al. 2012). Social media, or formats like political blogs, have changed the public sphere, not only by adding something to old forms of mass media public spheres but also by partially substituting them. Expectations from 10 years ago that political blogging would substitute traditional journalism still appear to be exaggerated. But notions predicting that web communication would not affect mass media journalism at all have proved to be dewy-eyed. Today, new mixed models of journalism are observed where leading newspapers incorporate “blogs, columns and news stories and where writers may be bloggers one day and reporters the next” (Zuckerman 2014: 158 for an analysis of digital journalism; see also Peters and Witschge 2015). The enormous popularity of comment sections has recently attracted intense interest among communication scholars (for an overview see Toepfl and Piwoni 2015). Surveys show an increased spread of comment sections on online news sites, and research indicates that user-generated content on comment sites influences readers’ perceptions of public opinion and can change the reader’s personal opinion.

Internet activism as a new form of protest is gaining influence in the public sphere. There is little doubt among researchers that meanwhile it is obvious that it is “the norm, not the exception, for political and activist campaigns to rely on social media, crowdfunding and other digital techniques as well as advertising, lobbying and conventional fundraising” (Zuckerman 2014: 158). Online communication is used by political actors and activists in many ways: for spreading information and news online, for e-mobilisation (using online tools to facilitate offline protests), for online participation (e.g. online petitions) and for organising movement efforts online, so that there are discussions of whether pure online activism might reduce the relevance of (offline) NGOs (Earl 2015). Thus, social media and online debates are regarded as having the potential to function as counter-publics to the established and published discourse (originally Fraser 1992; see Dahlberg 2011). Especially in developing countries, which often lack media channels for underprivileged groups, social media is seen as a means to empower the poor and increase the possibilities for them to influence or petition the government (e.g. Hoskins 2013). Impressive social movements and uprisings in recent years have shown that the Internet, and especially communication via social media, has been widely supportive of networking and the public campaigning of social movements. The attention to completely new forms of bottom-up spontaneous political activism fostered by the political use of social media was especially triggered by the revolutionary movements in North-African countries (the so-called Arab Spring) that led (albeit mostly temporarily)—to fostering democratic structures of public debate and governance in previously autocratic regimes (see contributions in Kumar and Svensson 2015; Özcan 2014). This perspective has not only been tempered by the observable autocratic or oligarchical backlash in most of the Spring-countries, but also by analysis that shows that years of offline planning, negotiation and organisation made the Arab Spring possible, suggesting that social media was nothing but a supportive tool that has been used for campaigning and organising counter-publics (Lim 2012; Bennett and Segerberg 2012).

The potential of social media to establish “counter-publics” is undoubtable, but this potential also has its downside, with effects which are detrimental to a democratic public sphere. Following the public sphere concept of John Dewey (2012: 1927), that publics emerge as soon as knowledge about a public problem evolves, it appears to be plausible that such knowledge is now more easily spread and thus potentially combines formerly unconnected individuals into concerned publics (see Farrell 2014) by organising all kinds of Internet fora, social media, etc. on any political question. It is, however, evident that new modes of political Internet communication not only have the capacity to support the emergence of counter-publics and the empowerment of civil society groups, but are also effective tools for campaigning by established political actors, institutions and groups. It was quite clear, before Donald Trump as US President made Twitter his preferred media of political communication, which has dominated mass media coverage of the elections in many ways (Enli 2017), that political Internet communication can also be regarded as a battlefield, with all kinds of manipulative strategies and tools applied to steer public opinion (Bradshaw and Howard 2017). The use of social media in electoral campaigning in Western democracies, often referred to as improving the options for civil society to connect to political representatives, is—as has been shown by analysis of the use of social media in US electoral campaigns (Kreiss 2012, also Towner and Dulio 2012)—far from being self-organised bottom-up support for candidates, but is “meticulously planned, tested, and crafted by highly bureaucratic, hierarchical institutions” (Wells 2014).

It is also noteworthy that the option for organised as well as individual actors to introduce their political thoughts or preferences into the public sphere and establish counter-publics is not bound to anti-establishment or grass-root world views. A study of the online news site of opinion-leading German newspapers, published in the aftermath of the 2013 national elections (Toepfl and Piwoni 2015), analysed journalistic articles as well as user comments regarding the new German Anti-Euro party “AfD” and found clear indications that while the news sections of the sites (journalists’ content) unanimously painted a dismissive picture of the new party, the comment sections were mainly used to challenge this mainstream consensus. The authors conclude that in the comment sections, “… a powerful counter (sub) public sphere had emerged. Remarkably, approximately 75% of comments supported a new party that just days before only 4.7% of the electorate had voted for. In essence, these findings thus showcased how an emergent collective of counter public-minded individuals were exploiting the comment sections of Germany’s opinion-leading news websites in order to create a highly visible—and therefore enormously powerful - counter-public sphere” (Toepfl and Piwoni 2015: 482).

All in all, the landscape of political communication has changed, which has in many ways empowered civil society to get access to the public sphere. However, this may not challenge existing structures and hierarchies as much as expected by e-democracy enthusiasts. As referred to above, Koopmanns and Statham (2010; see also Koopmanns 2015) found claims of CSOs being underrepresented in media reporting on European political issues when compared to institutional actors, based on a sample of quality journals in six EU Member States. This was especially the case for media reports on European issues. Interestingly, an analysis of websites conducted within the framework of the same study did not find a more balanced representation of institutional and civil societal actors (Koopmanns and Zimmermann 2010), leading to the conclusion that the Internet replicated power hierarchies that affected actors’ abilities to reach audiences. Without overstating this (and other) single findings, since the overall state of research on the empowering force of the Internet is still insufficiently developed, it can be summarised that there is an online space for political communication with many new features and options that go beyond or bypass mass media channels. It is, however, subject to debate as to what extent these features have the potential and are set into practice to democratise political communication and public discourse. It is quite clear that despite the democratic potential of many of its features, “…the internet and related technologies are increasingly identified as posing threats to democratic structures and participation in politics and society” (Dutton 2018: 4). Features that add to this picture are the misuse of personal data for political advertising, by personal profiling and micro-targeting (Kind and Weide 2017; Dubois 2017), as in the case of Facebook, where user data was obtained by the political consulting company Cambridge Analytica for personal profiling and selective political campaigning. Manipulative strategies are supported by the application of algorithms (social bots) to automatically spread messages in social media communities which are presented as having been posted by users, and thus falsely produce the perception that the message spread is shared by a vast majority of (fake) community members (Wardle and Derakshan 2017). A basic feature of political Internet communication that is massively opposed to any notion of a public sphere as a shared space of rational discourse is the tribal structure of social media communication. Any content is shared and distributed mainly among communities of like-minded people, who join the same filter bubble (Pariser 2011) of content. Anything that is posted by members of these bubbles (or by a bot pretending the content has been posted by many members) has a pre-established reliability bonus since it confirms the worldviews and identities held by members and is (factually or apparently) distributed by people “like us” in whom we can trust. Social media in this respect can be regarded as being tailor-made for spreading news and preventing it from being counter-checked by other sources. News is travelling within or in between peer-to-peer networks and not via media, which are able to create a public space in which content might be checked by gatekeepers and, can be, respectively, criticised in a public (i.e. open to everybody) manner. One effect of this, beyond any single attempts at disinformation or manipulation, might be to render a rational debate impossible, because citizens enclosed in their specific filter bubble do not see any possibility—beyond their peer community—to tell wrong from right, deceit from the truth or rational reasoning from emotional affect.

Despite the negative effects mentioned above, it still is the interactive quality of Internet communication that is the anchor of accounts that the virtual public sphere has the potential to foster the deliberative quality of public discourse, compared to mass media publics. Deliberative quality implies an open exchange among a broad spectrum of perspectives and views, without restrictions as regards access to the discourse, the right to speak, and the willingness to listen and rationally react to opposing perspectives. With regard to this, political social media sites and political blogs come into perspective. In a review of research on online deliberation, Freelon (2015) sees two perspectives being dominant in research. One research thread studies the deliberative content of online political communication, asking to what extent online political communication meets normative criteria such as civility, reciprocity and reason-giving. The other thread focuses on “selective exposure”, based on the assumption that “exposure to a diverse array of information sources is good for democracy, while the exclusive consumption of opinion reinforcing content is problematic” (Freelon 2015: 774). Many studies focus on specific case studies of Internet fora, blogs and others. According to Freelon, there is consensus that online political discussions mostly do not meet the quality criteria of deliberative content. As regards the “selective exposure” perspective, Freelon sees mixed results. Some studies support the notion that online debates reinforce the exclusive consumption of opinion-reinforcing content, while other studies cannot support the selective exposure thesis (Freelon 2015: 773 ff.). Liu and Weber (2014) come to similar conclusions for research on social media. Due to the enormous amount of literature available, it is impossible to undertake a systematic tour d’horizon through available research at this point. In the following, a few examples from recent studies are given to illustrate the “quality of content” as well as the “selective exposure” perspective.

Generally, the political “blogosphere”, which began in the late 2000s, gave rise to far-reaching expectations of the positive effects on democracy in terms of bringing about a new space for open and rational exchange across political affiliations. Seen from the perspective of established politics, the blogosphere should bring about a new space to learn about public worries, expectations and needs, thus supporting the functionality of the public sphere for the responsiveness of the political system. However, blogs often show features of political exchange among elites and/or well-educated publics, and rather than opening up spaces for deliberation across political communities or perspectives, they often appear to foster communication only among like-minded communities. As regards the quality of content, new social media and the so-called “blogosphere” have been diagnosed to show strong discrepancies along the lines of established politics and more informal use by citizens. While online media are often used by policymakers in a vertical manner of communication and “replicate the worst aspects of the established political communication system, with politicians running blogs that look like old-fashioned newsletters”, citizens’ initiatives use blogs and social media more as a means of horizontal communication among peers (Coleman and Blumler 2012: 146).

Empirical studies, mainly based on an analysis of hyperlinks between different political blogs, show contradictory results: There are indications that blogs have a potential to foster deliberation in terms of exchange on political issues across political affiliations, as well as examples of blogs that function as spaces for in-group self-assurance (see research overview in Silva 2014). A study of the leading political blogs in Romania (2013, 2014) found that other than the mass media commitment to neutrality, users of political blogs clearly tend to choose blogs that support their political thinking and position (Munteanu and Staiculescu 2015). A network analysis of 20 of the most popular political blogs in Portugal (Silva 2014: 200) during national election campaigns could not find support for the thesis that blogs tend to polarise political positions “… blogs managed by citizens interested in politics do engage in conservations and debates regardless of the ideology. We find right and left wing blogs linking to each other, thus indicating that they share issues and themes of debate, interests, and arguments”. Negative reactions among participants “… that intend to mock or show contempt, insult and hamper dissident voices” were found to be of minor relevance.

The example of the Norwegian Labour Party’s (MyLabourParty) websites, with blog-like articles and comments, shows that the extent to which online blogs or social media sites allow for open debates and political communication depends on their design and purpose. For inner party communication, blogs are used for distributing news among party members and supporters, while others are meant to reach out to a wider public. Analysing different online offerings from the Norwegian Labour Party (Johannessen and Følstad 2014), it was found that blogs whose contributors are mainly or only party members tend to be restricted in triggering debates when compared to sites that are also open to opposing political opinions. It has been shown by a broad network analysis of the online discussion forum of the Italian Five Star Movement that online discussion platforms provided by political parties and groups are not necessarily platforms for mutual self-assurance. The Five Star Movement owes its foundation to the exceptional success of a political blog run by its founder, Beppe Grillo, in 2009. The widely used online forum of the movement, according to Bailo (2015), did not show significant tendencies of fragmentation of the online community using the forum. Many users engaged in discussions on different topics, thus the debate was not structured in accordance with specific interests or values held. The author concludes that people “are more interested in engaging rather than convincing each other” and they come to the forum “mainly to socialise their ideas and be exposed to other’s thoughts on issues they are interested in” (Bailo 2015: 564).

As regards the quality of communication, the anonymity that is allowed for in Internet chats, fora or social media has always been held to be conducive to allow for a situation that comes close to the ideal of deliberative exchange of arguments implied in Habermasian discourse theory, because anonymity allows us to disregard hierarchical factors such as social status. It is, however, mainly the anonymity of communication that often gives way to idiosyncratic and untrustworthy talk, to bullying or the erratic dismissal of the arguments of other users. While anonymity can strengthen the focus of participants on the argument rather than the person, and thus increase deliberative quality, at the same time it implies a lack of social control that can lead to emotional and erratic behaviour, as has been found, e.g. by analysis of Twitter discussions on new abortion legislation in the UK by Jackson and Valentine (2014). This—with regard to a rational exchange of arguments as the core of a deliberative public sphere—destructive feature of Internet communication, especially of social media, is represented by the “troll”. The intervention of the troll in social media communication, or in the comment sections of mass media, by posting statements meant to destroy the mood of serious exchange of arguments by insulting and bullying participants, discrediting their credibility and spreading doubtful “news”, is a ubiquitous phenomenon. The role of the troll—originally an obscure niche existence—can even be said to have made it into mainstream political communication (Hannan 2018). Not only does the “puer robustus”, like US President Trump, stand for this but also the aggressive style of political communication introduced by many populist, right-wing movements in Europe. “Twitter wars” are meanwhile featured in quality mass media. Hannan (2018), in an instructive and pessimistic analysis, develops an account of trolling as the political mode of communication inscribed in the social media technology itself, undermining the value of “truth”: “Disagreements on social media reveal a curious epistemology embedded within their design. Popularity now competes with logic and evidence as an arbiter of truth. […] Lengthy detailed disquisitions do not fare well against short, biting sarcasm. They also do not fare well against comments that, however inane, rack up a far greater number of likes. In the mental universe of social media, truth is a popularity contest” (Hannan 2018: 33).

Looking at empirical studies of communicative practice in political fora and other spaces, the seriousness of the above sketched analysis revealing the anti-democratic aspect of Internet communication cannot really be questioned, but such studies can support the notion that there is another, democratic potential that can be realised given the right frame of conditions. This deliberative quality was found to be dependent on factors of political culture. A study using 15,000 comments from five national newspaper online sites conducted by Ruiz et al. (2011) found two models of audience participation in online fora of newspapers: in the first, “communities of debate” are formed based on mostly respectful discussions between diverse points of view. This model—more in line with deliberative norms—was found in Anglo-American newspapers (The Guardian and The New York Times). The second model of “homogenous communities” is characterised by expressing feelings about current events and has fewer features of an argumentative debate, less respect between participants and less pluralism, and was found in European newspapers (El País, Le Monde and La Republica). The authors regard this difference to be an effect of different cultures of journalism based on the political cultures of the respective countries. While a culture of “internal pluralism” is dominant in the Anglo-American case, with newspapers not being aligned with a particular political position, a culture of “polarised pluralism” is dominant in the European case, where “participants are mostly aligned with the ideological perspective of the newsroom: Citizens participate in the spaces provided by their news website of choice, mostly finding similar positions to theirs and editorial content that fosters political polarization” (Ruiz et al. 2011: 483).

Freelon (2015) has conducted a comparison across two technical platforms: Twitter hashtags and online newspapers comment sections. One of his central conclusions is that issue hashtags on Twitter made it more likely that discussions were of a more “communitarian”—meaning in-group and self-assurance—character, whereas comment sections on online news (which are more open to and are accessible by broader mass publics) were more likely to generate discourse with deliberative features such as openness to and exchange among a diverse and contradicting scope of arguments and statements. Research regarding the question of to what extent blogs, Internet fora and political social network sites can contribute to or foster features of an ideal public sphere (in terms of equality, inclusiveness and rationality of discourse) generally show mixed results.

It is well known that new populist movements rely very much on social media to organise and mobilise their members and followers (e.g. Januschek and Reisigl 2014). There are indications that social media fosters a “closed shop” in which those who are already convinced mutually reconfirm their ideology and their prejudices, rather than providing for a democratic and open rational exchange of arguments. Generally, this thesis is connected to the notion that while mass media normally provide for a mixed or balanced view of differing standpoints on political issues, the Internet (due to its ability to organise certain communities) is suspected “that recruitment, radicalization, and incitement are facilitated” via its tendency “to foster echo chambers where people are denied feedback contrary to their own views, which are therefore reinforced” (O’Hara and Stevens 2015). While O’Hara and Stevens reject this thesis as portraying a general feature of political Internet communication, there are indications that it holds true for the use of social media and website communication by extremist and populist movements (Warner and Neville-Shepard 2014).

In a broad review of research and scholarly discussion on the changes in news supply and consumption on the Internet, Tewksbury and Rittenberg (2012) found some evidence for fragmentation and polarisation of audiences alongside political predispositions, due to the multitude of specialised news sites. However, they argue that the fact that there is a multitude of specialised news sites and that some people restrict their information consumption to a certain set of news sites does not necessarily imply that they do not share common public knowledge as well as public agendas: “Fragmentation and polarization are ideas, still, more than observable realities. There is ample evidence that many people are specialising their news consumptions in ways that might lead to either or both outcomes. There is less evidence that knowledge and opinion are fragmenting and/or polarizing. Most of the uncertainty about the operation of these phenomena stems from a lack of research; it rarely lies with disconfirming studies” (Tewksbury and Rittenberg 2012: 143). On the other hand, they found evidence that the Internet offers more user control with regard to choice of content as well as with regard to contribution to news production, which can be regarded as “information democratization”. But also in this regard, it is not yet clear to what extent the potential will become a reality. Counter tendencies of fragmentation and the dominance of strong media companies on the web, as well as regulation on the content of the Internet are regarded as interfering with the democratic empowerment of the audience (Tewksbury and Rittenberg 2012: 144 ff.).

Political communication via social media extends the opportunities for individuals to post their own thoughts about any kind of public event and share it with friends or peer groups. In a more optimistic view, this is regarded as being in line with a general change of civic identities that has been observed for decades and represents a shift away from materialist to post-materialist and individualist values. More individualistic expression of self and weakening ties to formal organisations (parties, unions), and collectives (class) is regarded as being expressed as well as pushed by the use of social media. Individual choices made possible by the Internet allow connection to all kinds of cultures, social groups and preferences, and this comes at the cost of adherence to widely shared ideologies or bigger (public) formal organisations such as political parties (see, e.g. Wells 2014; Bennett and Segerberg 2012).

On the other hand, a clear danger when restricting oneself to these formats of political information and communication is the segmentation into peer groups or issue-related publics. In addition, one runs the danger of the complete loss of connection to any broader sphere of exchange among competing perspectives on contested issues of public (in the meaning of national or transnational) interest—which in the worst case would lead to idiosyncratic discussions and worldviews. As Zuckerman (2014: 165) puts it: “Social media allows the friends you follow online to participate in setting your political agenda, adding dots to the canvas that are in your immediate line of sight. We likely need a new class of tools and practices too help us step back and see our interests and perspectives in a broader context.”

The visibility of European issues in mass media has always been part of the focus of empirical research on the European public sphere. However, research on the relevance of political communication on the Internet for building a European public sphere or supporting the Europeanisation of national public spheres is scarce. What comes into focus first is the use of web-based communication by the European Commission. It is only recently that “issue publics”, organised via the web by civil society actors, has come into the focus of research with regard to their potential to “Europeanise” the public sphere. This also applies for the use of social media by Eurosceptic movements and political parties. The latter, as has been shown above, can be said to form real “echo-chambers” of “EU bashing”.

Following a programmatic turn to new and open forms of governance laid out in the White Paper on Governance (EC 2001), following the Irish “No” to the treaty of Nice (2001), the EC began to actively fund and set up citizen participation and public consultation activities through its “Plan D for Democracy Dialogue and Debate” (EC 2005) in 2005, as a response to the rejection of the constitutional treaty in the French and Dutch referenda. This was explicitly meant to strengthen the development of the European public sphere, also via means of e-participation (see Yang 2013; see also Lindner et al. 2016). Part of this strategy was to connect the process of EU policy formulation and legislation to the European constituency by inviting civil society actors and interest groups to participate in online consultations on issues under EU regulation via the EC’s web portal, “Your voice in Europe”. Another outcome was the set-up of citizen consultations and online fora. In the following, a brief overview of new research available on the citizen consultations and the online consultations of the EC is given.¹

Summarising research on the European Commission’s online consultations, “Your voice in Europe”, Dieker and Galan (2014) conclude that although many consultations are “open”, allowing any group to participate, the consultations—in terms of the effects at the European public sphere—at best contribute to establishing segmented and mainly expert public spheres. The consultations normally do not attract interest from groups beyond those interest groups already represented in Brussels. Due to the fact that participation in online consultations is resource-consuming, it is mostly professional and well-organised groups that participate. According to a study from 2011, business associations make up 39% of all participants in online consultations (Quittkat 2011, acc. to Dieker and Galan 2014). As regards the potential to contribute to a more inclusive mode of policymaking and to a European will-formation in the sense of (segmented) public spheres, the consultations are perceived to suffer from shortcomings. The consultation process lacks transparency with regard to clear information about the criteria for weighting contributions and deciding on whether they are taken into consideration or not. Contributions are not made accessible to participants and no exchange among participants about contributions is possible. The purpose of the consultations is to search for input to the policymaking process rather than public deliberation with or among the groups contributing. However, the function of transmitting demands and interests from civil society to the European institutions is regarded as being restricted, since agenda-setting lies solely with the European Commission, which decides about the issues that are made open for online consultation. Since online consultations take place in highly segmented public spheres with mainly expert and stakeholder communities participating, consultations are regarded as having a highly professional character which does not allow them to take up a Europeanising function in terms of active European citizenship (Dieker and Galan 2014, 245).

Between 2001 and 2010, 23 transnational citizen consultation projects supported by the European Commission have been conducted, involving participants from a minimum of three European countries. They included face-to-face meetings as well as online discussions on specific issues, including the social and political implications of brain research as well as more general issues such as the European constitution and the future of Europe (Yang 2013: 25 f.). The six transnational “Deliberative Citizens Involvement Projects” (DCIP) covered by the Plan D programme involved approximately 40,000 people. The online project “Speak up Europe” alone involved 300,000 users in discussions on European politics (Yang 2013: 27). An evaluation of these DCIPs with regard to their deliberative quality as well as impact has been undertaken by contributions in Kies and Nanz (2013a). The case studies presented support the notion that DCIPs have a “...potential to ameliorate the legitimacy of the EU and to promote a more substantial EU citizenship” (Kies and Nanz 2013b: 10). The interactive aspect of deliberation is held to be a feature that can support the experience of European citizenship. However, this study also holds formats applied by the EU to function in a suboptimal way, such as “Your voice in Europe”, which allow citizens to send comments to policymakers, since they provide no space for deliberation and interaction among citizens on the issues addressed (Smith 2013: 209). In the EC’s approaches to citizen participation, the study found a tendency—mainly due to the lack of common language—to reduce the role of citizens to posting statements or commenting on statements by policymakers, rather than engaging in a European citizens’ debate and jointly working out policy options to be forwarded to policymakers. Most disappointing, according to the authors, was the lack of any follow-up activities or visible impact of the deliberative experiments on policymaking (Smith 2013: 215; Kies et al. 2013: 74 f.). Friedrich (2013: 44 ff.), discussing EU governance innovations, attests a strong bias to expert involvement. The approaches for dialogue with CSOs failed to realise their potential to strengthen the ties between EU authorities and European civil society or to support the construction of a European demos, due to a lack of commitment and “discretionary” patterns of participation. It is concluded that as long as a regulated integration of DCIPs in EU policymaking processes is not provided for and as long as DCIPs are mainly held on broad topics such as the social and economic future of Europe rather than on concrete challenges and the problems of decision-making, there is a danger that they are increasingly perceived as being more of a promotional instrument than serious attempts to engage the European citizenry in EU policymaking (Kies and Nanz 2013b: 11 f.). According to this analysis the potential of public consultations at the EU level to contribute to a lively European space of debate about EU policy, which could contribute to a European public sphere, appears to be restricted at this point.

The roles of segmented publics which are emerging around European issues, be it via initiatives taken by the EU institutions or bottom-up by interest groups across the borders of Member States, are regarded to have the potential to serve as nodes for a European networked public sphere alongside mass media publics (see Lindner et al. 2016). Kriesi et al. (2010: 225) argue that due to their frequent cross-national character, interest groups, and business and professional organisations (rather than political parties) can be regarded as a “Europeanized type of political actor”. But such organisations engaged in consultations with decision-makers can hardly be regarded to be functioning as nodes of a political public sphere. Issue publics that are exclusively “based on the horizontal intermediation between bureaucrats, experts and organized interests fall way short of complying with democratic provisions of openness and equal access” (Eriksen 2005, cit. Pfetsch and Heft 2015: 33; see also Eriksen 2007 and Lindner et al. 2016), for which online consultations (see above) provide an example. However, as far as such issue publics involve a broad range of actors, or are organised bottom-up by civil society actors, they are held to be more inclusive (in terms of reaching the average European citizen) than mass media debates on European integration issues, that are often driven mainly by elites, with the general public in the position of an observer in the gallery (Pfetsch and Heft 2015).

An analysis of the capacities of Internet-based issue publics created by networks of civil society active in Fair Trade and Climate Change campaigning (Bennett et al. 2015), however, found sobering results regarding the capacity of such networks to support the Europeanisation of publics. European-level networks for the issues of Climate Change and Fair Trade identified by the study have been found to be weak (compared to the connectivity of nationally based networks). The study found a certain amount of “Europeanisation” as far as nation-based networks move into networks active at the European level. But national networks mainly remain separate from each other and from those networks organised around issue-related EU platforms. It was also found that EU-platform-related issue networks of NGOs were able to engage citizens with the issues at stake to a much lesser extent than their national counterparts. The authors regard their findings as supporting the notion that “…civil society organisations in the Brussels area often serve as substitutes for the voices of European citizens, creating a civil order without credible levels of public engagement, and thereby deepening the EU’s democratic deficit” (Bennett et al. 2015: 135). Thus, it appears that the problem of segmentation in the sense of restriction of publics to “epistemic communities” and experts (as addressed with regard to European sub-publics organised around European issues and particular EU regulatory activities, Eriksen 2007; see Lindner et al. 2016) is not easily ruled out by Internet-based networks organised by NGOs.

First results on the use of social media by interest groups active in EU lobbying are available from a large European-funded project on the activity of EU lobbying groups (www.​intereuro.​eu). Van der Graaf et al. (2016) revealed, based on a data set of groups active in EU lobbying provided by this project, that when regarding the scope and volume of social media use there is little evidence that social media use was able to change inequalities in power and social representation at the EU level. The study comprised around 500 interest groups with reported activity at the EU level. “Range” was measured by the presence of interest groups on 11 selected social media platforms, and “Volume” was measured by the activity of groups on Twitter and Facebook. As regards the volume of social media use, small interest groups (citizens, workers unions) prevail over internationally organised groups, as well as big companies. However, when it comes to “range”, large organisations and firms with big resources prevail. Thus, at least with regard to interest groups at the EU level, social media appear not to provide for a level playing field for democratic will formation. The authors conclude with regard to the “democratic effects” of the “online world of interest representation”: “Rather than representing a new playing field where pre-existing resource differences between groups play less of a role, our analysis underlines the importance of resources both when we consider the range and volume of social media use” (van der Graaf et al. 2016: 122).

When analysing online comments of readers in political blogs, news platforms and transnational websites in 12 European countries during the 2009 EP election campaigns, de Wilde et al. (2014) found patterns of communication similar to those in mass media communication in the blogosphere with regard to European issues. The study found that diffuse Eurosceptic evaluations dominate public debates across Member States. The majority of evaluations made, particularly those by citizens leaving comments online, were Eurosceptic, constituting a gap between them and political elites who intervened with EU-affirmative statements. More complex evaluations of EU politics on the side of citizens were missing. These diffuse negative statements however were mainly about actual politics (complaining about the democratic deficit) than against EU integration as such. All in all, the authors conclude that there is “little evidence of the potential for legitimation through politicization in online public spheres” (de Wilde et al. 2014: 779). However, the study could not support the often-purported notion of a fragmentation of audiences in online discourse: debates intensified with politicisation of the European integration issue, but pro and con arguments were related to each other. Less “ambivalent” findings as regards the fragmentation of publics can be expected to apply for social media communication in Eurosceptic and right-wing groups. The negative aspects of social media communication as addressed in the previous section are no doubt relevant for any attempt to appraise the Internet’s possible effects on the European public sphere. As the prominent case of the activities of Cambridge Analytica (not only in the US elections) show: micro-targeting obviously played a role in the British referendum to leave the European Union. And social media is the central means of right-wing populist movements across the Union to set up information echo-chambers for their followers in order to provide information to counteract the so-called mainstream media that are defamed as providing fake news. As not only the case of Brexit shows, social media meanwhile is a means for politicians as well as grass-roots campaigners to reach audiences directly and bypass the filters of mass media journalism, to an extent that is about to dominate election campaigns (Enli 2017; Mair et al. 2017; Toepfl and Piwoni 2015). An analysis of Twitter communication during the Brexit referendum campaign found that Twitter users supporting the Leave campaign were more active (they tweeted more frequently) than Remain users and showed a strong tendency to interact only with like-minded persons (Hänska and Bauchowitz 2017). Setting aside the unanswerable question of whether or not social media campaigning and communication did decide the British referendum on EU membership, social media are without doubt the media of choice for activist groups and individuals challenging the mainstream, and not surprisingly, this counts not only for democratic civil society organisations but also for populist movements and campaigns.

In view of the phenomenon of the organised spreading of dis-information via social media by using so-called social bots or by other means of campaigning, and also with a view to the Cambridge Analytica case, the European Commission has set up a High Level Expert Group on online dis-information to make suggestions on how to steer against these “post-truth” currents and provide for quality safeguarding in Internet communication. The measures suggested range from increasing the transparency of Internet platforms’ use of data for advertising purposes, to establishing independent fact checkers in news media and safeguarding the diversity of the European news media ecosystem. The observation the Expert Group starts from, is to state the obvious downside to the structure of Internet communication that originally gave rise to hopes for a more vivid public sphere: “an increasingly digital environment gives European citizens many new ways of expressing themselves and of finding and accessing diverse information and views. It also enables an increase in the volume of various kinds of disinformation in circulation” (High Level Expert Group 2018: 10).

It is quite clear that scholarly debate as well as research on the European public sphere and on European citizenship and identification with Europe as a political community has intensified over the past years, due to the symptoms of an actual crisis of the EU institutions and the idea of European integration. It is still believed by many that the perceived democratic deficit of the European Union indicates the need for fostering a European public sphere as a space of debate across public spheres which are established at (and restricted to) national Member States. Moreover, there is a consensus that the new modes of political communication via the Internet have to play a role in that respect. However, far-reaching expectations and optimism envisaging the Internet as a panacea to political disenchantment and as a way to establish new transnational spaces of European bottom-up political communication are scarce compared to a decade ago.

As regards the state of the European political system, it is argued on the one hand that precisely in times of crisis, it is necessary to legitimise far-reaching decisions that will deeply influence living conditions in the European Member States. These decisions are to be reached through a vivid process of deliberation about pro and cons, about needs, demands and duties. On the other hand, there is pessimism whether—in the actual crisis that leads to focusing on national interests—there is enough homogeneity in the Union and strong identification with the EU as a transnational political entity. It is the observation of weak European solidarity and predominance of national perspectives that actually feeds the so-called “no demos” discussion among scholars of European politics. The point of dissent here is whether Europe is in need of the development of a transnational cultural identity (which is held by many to be exclusively bound to the national state) or whether a political identity—i.e. the European citizens’ commitment to the fundaments of the European political constitution—is sufficient to establish a new form of “European citizenship”. Proponents of a further integration of the EU base their cautionary optimism with regard to the “Europeanisation of European citizens” in the further development of the discourse about Europe and thus in the further development of the European public sphere. In this respect, the development of European identity and solidarity depends on the chances and opportunities to discuss and define what is in the common European interest via a common European political discourse. This would include fostering the role of the European Parliament and a European cross-national party system.

In this respect, what has been coined the “politicisation of Europe” in the actual crisis is—despite the undeniable symptoms of a renationalisation of political discourse and EU-scepticism—regarded as offering the opportunity to strengthen European identity. Since citizenship evolves in a political process of debate and emerges precisely outside of debates and conflicts about the public good, the current conflicts about EU policies and democratic legitimisation are regarded as a result of stronger engagement of citizens with the idea of Europe. On the other hand, it is evident that the crisis brings new forces and actors to the foreground that are not supportive of European integration and offer views that focus on national interests and thus help to strengthen national identities. There is, however, consensus that the European public sphere has a strong bearing on the development of a European identity as a space of debate where collective identities are constructed, and political communities are created.

With regard to the state of research on the European public sphere it has been critically stressed that so far, the focus of research has been on elite mass media communication and that research has neglected the relevance of new Internet-based communication networks mainly applied by civil society actors. In this respect some change can be observed, as there is a growing interest in Internet-based political communication and its potential for establishing new public spheres. However, a decade ago optimism was widespread that, while we find a decline of national public spheres with passive audiences and disenchantment with politics, the Internet could support the emergence of a trans-national public sphere that is more inclusive, deliberative and rooted in a transnational civil society. Such far-reaching expectations are scarcely put forward nowadays. Political communication via social media is currently in the focus of research, but it is difficult to draw clear conclusions with regard to their role in supporting the emergence of a vivid political public sphere:

Since the overall state of research on the empowering force of the Internet is still insufficiently developed, the actual potential for the Internet to bring about a new “public sphere” is impossible to assess. It can be summarised that there is an online space for political communication with many new features and options that go beyond or bypass mass media channels. It is, however, the subject of debate as to what extent these features have the potential to democratise political communication and public discourse.

It is held by many researchers that, in principle, the use of interactive tools of e-participation at the European level can contribute to fostering the legitimacy of the EU and to promote a more substantial EU citizenship. However, it is observed that the role of citizens is often reduced to just posting statements or commenting on statements by policymakers rather than engaging in a European citizens’ debate and jointly working out policy options to be forwarded to policymakers. Also, the notion put forward in our report to the European Parliament a decade ago (Lindner et al. 2016), that public spaces established by consultation processes offered by the European institutions are often restricted to expert communities and at best help to establish segmented issue-related elite publics on the European level, as confirmed by recent research, is just about to emerge. First results of the research on the use of social media and Internet sites by civil society organisations active on the European level indicate that the restriction of publics at the European level to “epistemic communities” and experts is not easily ruled out by Internet-based networks organised by NGOs. Moreover, political communication via social media plays a destructive role by supporting filter bubbles and dis-information. This is widely used by anti-European populist movements all over Europe with, as is suggested by research on the role of Internet communication in the British EU referendum on the EU membership, significant detrimental effects to public deliberation.