Europe in crisis? Testing the stability and explanatory factors of the Europeanization of national public spheres

Abstract

Presenting new data for a large-scale longitudinal content analysis of newspapers in six European countries between 1982 and 2008, the article explores what has previously been described as the process of ‘segmented Europeanization’ of national public spheres. Expanding on the theoretical concept of political discourse cultures, the article tests the explanatory power of national and editorial political discourse cultures on the degree of Europeanization of national public spheres. The analysis reveals the persistent stability of the pattern of ‘segmented Europeanization’ in which national public spheres increase their monitoring of Brussels but not their efforts to connect with other European countries. Still, the revealed national differences pale compared to the transnational gulf existing between quality and tabloid newspapers which may ultimately represent a far greater threat to the legitimacy of the European Union than the persistent lack of discursive exchange among the European community of nations.

Keywords

comparative longitudinal content analysis European public sphere political discourse culture political legitimacy tabloid and quality newspapers

The European public sphere revisited

The popularity of the European public sphere as a research topic may have waned somewhat in recent years (for an overview, see de Vreese, 2007; Kleinen-von Königslöw, 2011; Machill et al., 2006). Political developments and events of recent years have provided ample proof, however, that the validity of these empirical appraisals may already have expired while simultaneously the topic itself has not lost any of its political relevance. The European Union (EU) still appears to be in dire need of a European public sphere as a source of legitimacy (Habermas, 2009) as it is struggling for acceptance with the people and leaders of the European member states. The European people have clearly expressed their dissatisfaction with the EU in several failed referenda on the Constitutional Treaty and the Lisbon Treaty. And during the global financial crisis national political elites have also displayed a blatant lack of trust in the political stability of the European Union and the solidarity of its members, as each focuses primarily on finding national answers to the problems. The continuous political integration of the EU and its public sphere are not as self-evident and path-dependent as its orchestrators originally may have hoped, and thus a final, concluding answer to the question of the existence of a European public sphere (EPS) cannot be achieved yet.

Our research project has previously analysed the Europeanization of national public spheres in leading quality newspapers of six European countries between 1982 and 2003 and the results have been discussed in a number of publications.¹ The most important finding has been the pattern of ‘segmented Europeanization’, depicting a European public sphere where each country increases its surveillance of Brussels, but a shared European discourse among the European community of nations fails to develop. The following article adds to our previous work in two ways: first, it evaluates the stability of the original empirical findings over time and for a much neglected segment of the public sphere, the tabloid press. It provides an urgently needed update by presenting results for a new wave (2008) of the standardized content analysis and expands the original sample to tabloid papers for all six countries (Austria, Denmark, France, Germany, Great Britain and Poland). Second, it uses this fundamental extension of the sample to systematically test several hypotheses on probable explanations for the levels of Europeanization in different national public spheres, thus continuing previous efforts by Brüggemann and Kleinen-von Königslöw (2009).

The segmented Europeanization of national public spheres

Since the late 20th century a number of academic research projects have addressed the question whether the political integration of the EU has been echoed by the development of a European public sphere (e.g. AIM Research Consortium, 2006; Kantner, 2006; Koopmans and Statham, 2010; Liebert, 2007; Pfetsch et al., 2008; Tréfas and Lucht, 2010; Trenz, 2004; Triandafyllidou et al., 2009; Voltmer and Eilders, 2003; Wessler et al., 2008). According to Habermas and others, the existence of such a European public sphere, a space for the debate of European politics linking both European citizens and political institutions of the national and EU level, might represent an important step towards legitimizing the European Union as a political institution (Habermas, 1998; Peters, 2005).

The study presented here takes a first and foremost empirical view on the public sphere, defining it, as proposed by Dahlgren, ‘as a constellation of communicative spaces in society that permit the circulation of information, ideas, debates’ (2005: 148). Normative democratic theory attributes to the public sphere and its main communicative space, the mass media, a number of political functions for the political community, independent of its political or geographical reach, i.e. whether the public sphere is a national, European or global public sphere (Cohen, 1997; Ferree et al., 2002; Habermas, 1997; Peters, 2008). Of the multitude of political functions discussed, one appears to be particularly important but still realistic given the limited possibilities of mass-mediated public spheres: the transparency function, i.e. that public spheres should provide their participants with information on all political problems relevant to them.²

Given the fragmented nature of national public spheres, which are also constituted by a multitude of different media outlets, the fulfilment of such a function for the European political community does not necessarily depend on the existence of pan-European media with a significant audience reach.³ In line with the large part of European public sphere research, our project therefore focuses on tracing the ‘Europeanization of national public spheres’ (Gerhards, 2001: 152), i.e. whether the political communicative space as represented in national media outlets is changing in a way that enables it to also sustain a European public sphere and fulfil the transparency function for the new European political community. In our opinion this change would need to occur on two empirical dimensions. In the vertical dimension, public spheres should extend their focus to include the additional level of government as political competences are shifted from national to European institutions. And as the recent example of the Greek state deficit so strikingly demonstrated, political (and economic) developments in EU member states have an impact on the EU community as a whole and on each of its members. Correspondingly, national public spheres should expand horizontally to monitor political developments in the whole European community of nations and integrate opinions and arguments from speakers of neighbouring nation-states in their own public debates.

Our first empirical study pointed out that a Europeanization of national public spheres did occur between the early 1980s and 2003, however not evenly across the two dimensions and the six countries under analysis, but in a process best described as ‘segmented Europeanization’ (Kleinen-von Königslöw and Möller, 2009; Wessler et al., 2008). The separate public spheres increase their monitoring of Brussels’ politics over time, but the public discourse retains its national framing, there is no growing interest in the issues and opinions of the European neighbours and only a slight shimmering of European collective identity can be distinguished. The trends and levels of Europeanization display strong differences between countries and hence illustrate how the public spheres remain tied to their national containers.

This diagnosis of ‘segmented Europeanization’ is corroborated by results from several other multinational empirical studies. The EUROPUB-project, for example, also noted an increase in vertical Europeanization for political claims (claims addressed to the EU-level) while claims addressed to (or citations from) speakers from other European countries do not increase, and in some policy fields even decline (Koopmans et al., 2010). Research on the synchronization of national debates on particular events or conflicts – such as the Haider scandal in 2000 (Berkel, 2006; Van de Steeg, 2006), the introduction of the euro (de Vreese, 2001) – or on long-term issues such as the debate on the Constitutional Treaty (Adam, 2008; Tobler, 2010; Vetters et al., 2006) has provided evidence that national public spheres are somewhat open towards each other, and sometimes use the observation of public debates in other countries as a proxy to stimulate national discourse on specific issues. Still, as d’Haenens (2005) is able to show for coverage of the EU and Europe in general, substantial country differences remain concerning the use of news frames and discourse types, which probably reflect deeply entrenched differences in audience expectations. It seems therefore rather doubtful whether the cross-national exchange on particular issues has led to an increase in mutual observation and discursive exchange among the European nations. As our original cross-issue analysis only covered the time period between 1982 and 2003, our first research question to the new wave of data is thus:

RQ1: How has the Europeanization of national public spheres evolved between 2003 and 2008?

The role of the tabloid press in the Europeanization of national public spheres has not been explored much.⁴ Normative public sphere theory often includes criteria on the content and form of public debate which tabloid media may have difficulties fulfilling: in the view of many scholars (and journalists) the focus on ‘soft’ news and the ‘melodramatic’ emphasis on sensation and emotion prevent the popular press from contributing (positively) to the public sphere (see, for example, Gripsrud, 1992; see also the overview in Sparks, 2000). Focusing on the transparency function of a public sphere, and primarily on the question whether the communicative space sustained by national public spheres has extended vertically and horizontally, however, allows us to include tabloid papers in our analysis, as the vertical and horizontal extension of a public sphere is not tied to any particularly challenging quality standard of public debate. We are thus able to determine whether the parts of the European population which do not regularly read the quality press still have a chance to participate in a European public sphere through the tabloid papers, even though it may turn out to be a discourse on different issues and in different forms than those in the quality press.

RQ2: What shape does the Europeanization of public spheres take in the tabloid press? Is the pattern of ‘segmented Europeanization’ mirrored in tabloid newspapers?

The second aim of this article is to explore how one might explain this pattern of ‘segmented Europeanization’, in particular the different levels of vertical and horizontal Europeanization in the different national public spheres. Most EPS research so far has been descriptive in scope, with a few notable exceptions such as Boomgaarden et al. (2010), who investigate the impact of exogenous factors and events on the visibility of EU news (i.e. vertical Europeanization). In contrast this article does not focus on the impact of singular events but on the more general structural changes in the vertical and horizontal extension of public debate. It thus expands on the work of Brüggemann and Kleinen-von Königslöw (2009), who looked at the impact of political and media factors on the level of Europeanization in the quality press of five European countries, by applying a similar analysis to a wider sample of countries and newspapers and by adding an overall theoretical explanatory concept that provides a unifying frame for the potential impact of political and media factors.

Following Bernhard Peters’ observation that public spheres have a ‘social and cultural foundation’ (2008: 246), national public spheres can be seen as the articulation of the political discourse culture of the respective country. Political discourse cultures in turn should be understood as ‘a specific thickening of cultural patterns (systems of classification, discursive formations) of political communication that people of a definable group refer to in order to make sense of political actions’ (Hepp and Wessler, 2009: 186; my translation). Understanding these patterns as cultural patterns has two implications: on the one hand, ‘culture’ is seen as something that is articulated in everyday practices of political communication, for example in the everyday practices of journalists, their habits and reflections. On the other hand, these ‘cultural patterns’ denote a certain kind of inertia, the patterns are relatively stable and lie under the everyday changes of the news agenda. Depending on the characteristics of the political discourse culture of a given country a public sphere may be more open to change than others, i.e. some public spheres may react more promptly to changes in the political system while others may hold on longer to the traditional power players.

Parallel to national political discourse cultures there also exist patterns of political communication that are characteristic for specific types of newspapers and could best be called editorial political discourse cultures (see also Hepp et al., 2009). These can be shaped by differences between national political discourse cultures, but they also can be very similar across country lines, for example they may be part of a European political discourse culture specific for a certain type of newspaper such as European tabloids. These patterns of political communication as articulated in the daily production of the newspaper are minted by a multitude of factors: they may be related to the organizational structure of the paper, newspaper ownership and political leanings, even the layout of the newsroom may have an impact on the way political communication is produced. They are expressed in a set of (often unarticulated) beliefs, rules and habits of journalists, which then again translate into certain patterns of coverage of EU and European foreign affairs observable through the analysis of concrete newspaper articles. Taking into consideration these two overlaying types of political discourse cultures several hypotheses can be developed on how they might affect vertical and horizontal Europeanization.

The most important influence factor in the political discourse culture of a country is doubtlessly the nation’s political, cultural and social history: does a country still consider itself a world power (even though its practical power may have long since declined) and thus the EU only as a necessary evil; or has it co-founded the EU in an attempt to overcome its own shameful national history? Political discourse cultures should thus strongly depend on the power and size of a country, which may translate into different degrees of Europeanization: newspapers in small countries are more likely to discuss the affairs of neighbouring countries or the EU as they are (or consider themselves to be) more dependent on others (and the EU):

H1: Public spheres of small countries are more vertically and horizontally Europeanized.

Table 1.

Logistic regression of influence factors on indicators of vertical Europeanization.

	Visibility EU institutions		Focus on EU politics
	(N = 18,913)		(N = 18,913)
Potential influence factors	e ^β	e ^β _stand	e ^β	e ^β _stand
Year of analysis	1.27^***	1.38^***	1.31^***	1.45^***
Quality paper	2.48^***	1.57^***	2.63^***	1.62^***
Analyst paper	1.72^***	1.29^***	1.49^**	1.21^**
Ambassador paper	–	–	–	–
Caterer paper	.56^***	.75^***	–	–
Power/size	–	–	–	–
Length of membership	.88^***	.87^***	1.22^***	1.23^***
Focus on EU topic	216.08^***	2.50^***	n.a.	n.a.
Length of article	1.50^***	1.45^***	1.23^***	1.22^***
Commentary/editorial	–	–	–	–
External author	–	–	–	–
Constant	.01^***	.08^***	.00^***	.02^***
Nagelkerke R² adj.		0.35		0.08

Notes: e ^βlogistic effect coefficient; e ^β _stand standardized logistic effect coefficient; n.a. = not applicable.

^* p < .05, ^** p < .01, ^*** p < .001 (Wald).

The national political discourse culture may also be shaped by the length of membership in the European Union. Founding members of the EU may be more committed to the project and more accustomed to the workings of its institutions, thus discussing the EU (but also the affairs of other members) has become a routine part of their political discourse cultures to which journalists and readers are accustomed:

H2: Public spheres of long-standing EU members are more vertically and horizontally Europeanized.

The potential impact of editorial political discourse culture on vertical and horizontal Europeanization depends largely on which types of (transnational) editorial political discourse culture are identified and what characteristics this typology is based on. One solution is the very common traditional differentiation between quality and tabloid newspapers. Then one can argue that in general quality papers have larger resources at their disposal, resulting both in a larger staff (of foreign correspondents) and more news space, making it easier for journalists to insert coverage both of the EU and of European foreign countries in addition to reports on national events. Furthermore their own perception of their readers encourages them to consider the EU and foreign news as important to their (elite) audiences.

H3: Newspapers with a quality editorial political discourse culture are more vertically and horizontally Europeanized than tabloids.

As Brüggemann and Kleinen-von Königslöw (2009) have shown, another important influence factor on (vertical) Europeanization is the editorial mission of a newspaper to promote the EU. Any type of editorial mission will form an important part of the editorial political discourse culture of a newspaper, as it impacts on the beliefs, rules and habits of its journalists which then translate into a specific daily practice of reporting. This study therefore also tests the explanatory power of a typology of newspapers that is not based on formal characteristics such as readership, resources and form of distribution, but on the self-conception of its journalists towards coverage of the EU and European foreign affairs: how important do they consider EU and European foreign coverage in their own opinion, in the opinion of their colleagues and superiors, and in the opinion of their (imagined) audiences? How do they see their role regarding this type of news? Hepp et al. (2009) conducted an extensive study of EU and foreign editors’ self-conceptions through newsroom observations and qualitative interviews. Based on their material they identified the following four editorial political discourse cultures in relation to EU and foreign European news. The analyst emphasizes the need for an extensive, thorough analysis of all EU and European news, even going far beyond the perceived demand of its (elite) audience. Elements of this type of editorial political discourse culture can be found in some, but not all of the traditional quality papers. In other quality papers, the editorial discourse culture of the ambassador is more prevalent: here editors want to explain the political process to their readers combining a pedagogical approach with a distinct pro-EU mission. In the newsrooms of most (but not all) tabloids the editorial political discourse culture of the caterer can be found: here international news items are only of interest if they are shocking or sensational; the perceived interests of the readers come before all other considerations. And finally there is the reporter, an editorial political discourse culture most often found in regional newspapers where journalists strive to provide only a basic, compact coverage of important EU and foreign European news without any ambitions (or resources) to go beyond the strictly necessary.

H4: Newspapers with an analyst and ambassador editorial political discourse culture are more vertically and horizontally Europeanized than other newspapers.

The final hypothesis concerns the relationship between the two levels of political discourse cultures. Previous research suggests a very strong impact of national differences and thus national political discourse cultures on the levels of Europeanization (Brüggemann and Kleinen-von Königslöw, 2009; Pfetsch et al., 2008); however, a systematic comparison of the explanatory power of the different levels of political discourse cultures will enable us to find out whether this influence is in fact filtered and dampened by the impact of (transnational) editorial political discourse cultures or not:

H5: National political discourse cultures have a greater impact on vertical and horizontal Europeanization than editorial discourse cultures.

Methods

The empirical basis of this article is a standardized content analysis of newspaper coverage in six European countries – Austria, Denmark, France, Germany, Great Britain and Poland – in the years 1982, 1989, 1996, 2003 and 2008, with a sample of two constructed weeks per year of analysis. The countries were chosen in order to include sufficient variation on the independent variables ‘power/size of a country’ and ‘length of membership’: Austria and Denmark are comparatively small member states, dwarfed by France, Great Britain and Poland as larger and more powerful countries and finally by Germany, with a population of 82 million and a GDP of 35.479 per capita⁵ the most populous and strongest economy in the EU. Together with France it is also a founding member of the European community, joined by Great Britain and Denmark in 1973, and more recently by Austria (1995) and Poland (2003). Including late-comer Poland in the sample also enables us to explore whether the Europeanization of national public spheres hinges on formal membership or whether shifts in the national political discourse culture may already occur during the protracted admission process.⁶

As the main focus of the study is the vertical and horizontal extension of the public spheres, i.e. which levels of government and countries are part of the debate, and not its political framing or bias, the selection of newspapers was not based on the classic political left–right cleavage. Instead the quality and the tabloid paper with the greatest overall reach (on average for the whole period of analysis) were included for each country, as these can be considered as influential Leitmedien, which are closely observed by other parts of the public sphere. In addition this assured that we obtained representations of national public spheres as they are experienced by as much of the general population as possible, and not only by an elite audience. It should be noted that in the media systems of five of our analysed countries tabloid papers form an important part of the media system and the public sphere. The exception is France, where the role of the tabloid press is mainly performed by sports newspapers and weekly tabloid magazines. The newspaper Le Parisien/Aujourd’hui selected for the study used to have a sensationalist orientation similar to the other tabloids in the sample, during the period of analysis it has undergone a reorientation as a regional paper (for a more detailed account see Hallin et al., 2004: 97). The final newspaper sample includes Die Presse and Kronen Zeitung (Austria), Politiken and Ekstra Bladet (Denmark), Le Monde and Le Parisien/Aujourd’hui (France), Frankfurter Allgemeine Zeitung and Bild Zeitung (Germany), The Times and The Sun (Great Britain), Gazeta Wyborcza and Super Express (Poland, only since 1989 for GW and 1996 for SE).

Even though the study is interested in a rather basic aspect of the European public sphere, the vertical and horizontal extension of debate, it does have a focus on the public sphere and public discourse, hence the sample was limited to ‘discursive’ articles in the political section (quality papers) and the general news section (tabloid papers), i.e. articles concentrating on the exchange of arguments and opinions rather than ‘mere’ information or ‘facts’ such as editorials, commentaries, interviews, contributions by external authors, press reviews and analysis/background articles. In the case of the tabloids, all articles were included in the sample as tabloid articles almost always contain some form of opinion (even if not necessarily backed by arguments) and reliable criteria to exclude ‘pure news’ articles could not be developed for the tabloids. Thus the sample includes a total of 19,175 articles.

Each dimension of Europeanization was assessed through two indicators. For vertical Europeanization all references to EU institutions and the EU itself were coded (whether they were favourable, neutral or critical) as well as EU politics as the main topic of an article. Horizontal Europeanization could either occur through mutual observation, i.e. when other (Western) European countries were referred to in the lead of an article, or through an exchange of opinions, i.e. when speakers from other (Western) European countries were cited.⁷

For testing the hypotheses, each newspaper had to be assigned comparative index values for the independent variables both on the level of national discourse cultures and of editorial discourse cultures. On the level of national political discourse cultures, the power/size of the respective culture was determined based on GDP and population, resulting in an ordinal scale variable ranging from 1 for small (Austria, Denmark) to 4 for large countries (Germany). Length of membership was also an ordinal variable ranging from 1 for the founding members to 4 for Poland. On the level of editorial political discourse cultures, binary variables were used to discriminate between members (or not) of a particular editorial culture. For the editorial political discourse culture of ‘quality papers’, all traditional quality newspapers were assigned to the group. The classification for editorial political discourse cultures which reflect the self-conceptions of journalists regarding EU and foreign European news is based on the newsroom studies of Hepp et al. (2009), as described in the previous section. Most newspapers were assigned to just one editorial political discourse culture; however, two papers contained elements of two cultures and were thus included in both groups.⁸ There was no newspaper corresponding to the editorial political discourse culture of the ‘reporter’ in the sample; hence this type is excluded from the analysis.

A regression analysis was used to test the explanatory power of the potential influence factors on each of our indicators of Europeanization. As the dependent variables are dichotomous variables (for example an article either focuses on EU politics or not), logistic regression analysis was employed. Whereas a linear regression model would predict how an independent variable influences, for example, the number of articles on European politics, a logistic regression model predicts how the independent variables influence the chances of the article being on European politics or not.

In addition to the explanatory variables, several control variables were included in the models: year of analysis (allowing us also to reassert RQ1 concerning the stability of our results over time); length of article; and whether it was an editorial/commentary or a guest contribution. In order to control for the topic of the article as a potential impact factor, EU politics as focus of the article was introduced in three of the models (except for the model where it was the dependent variable).

Results

The stability of the segmented Europeanization of national public spheres

Vertical Europeanization: an increasing monitoring of Brussels until 2003

The previously observed trend of increased observation of EU institutions for the quality press has been reversed by 2008. While in some countries (Germany, Poland) the interest in EU institutions only stagnated, in France, Denmark, Great Britain and Austria it actually declined (see Figures 1 and 2).

Figure 1.

Share of articles referring to EU institutions between 1982 and 2008.

Figure 2.

Share of articles referring to EU institutions by country and type of newspaper.

The trend of vertical Europeanization is mirrored in the tabloids, though at a much lower level. In 1982 there was almost no attention paid to EU institutions: only 1.3% of all articles in the tabloids mentioned EU institutions, ranging between 0.2% in Austria (Kronen Zeitung) and 2.5% in Denmark (Ekstra Bladet). By 2003, the interest had tripled to 4%, but it still remained negligible compared to the 28% share of articles mentioning the EU in the quality press. However, as EU institutions continue to gain ground in the tabloids in 2008, one might speak here of a laggard, but continuing process of vertical Europeanization.

Of all countries in our sample, France emerges as the race leader, in the quality paper Le Monde on average one-third of all articles mention EU institutions, and the tabloid Le Parisien also lies above the mean of its newspaper type with a share of 4.4%. Austria comes in at second place for this indicator, the broadsheet Die Presse refers to the EU and its institutions in a quarter of its articles, but as with Le Monde its interest appears to have reached a plateau in 2003. The Austrian ‘Volkszeitung’ Kronen Zeitung displays only an average interest in EU institutions for a tabloid paper (3.6% of all articles), but here the trend of vertical Europeanization increases strongly in 2008. The two Danish newspapers differ somewhat for this indicator: the quality paper Politiken refers comparatively often to the EU (19% of the articles), Ekstra Bladet brings up the rear among the tabloids (2.8%). In both German newspapers the trend of increasing references to EU institutions sets in comparatively late but strongly. British papers mention the EU rarely (even though the indicator counted all references, even negative ones). While both papers start off with a clear trend of Europeanization, the curve drops off sharply in 2008 – during the global financial crises, EU institutions appear to play no role at all in Great Britain. Poland as a new member state maximizes its efforts to recover from its late start, the trend is particularly strong in Super Express, whose interest in EU institutions is highest among all tabloid papers.

The reversal in the trend of vertical Europeanization in the quality press by 2008 is confirmed for the second indicator: EU politics, which in 2003 had seemed to have conquered a fair share of the public debate, reaching 7% of all articles in the quality press, drops again below the 5% line. At the same time the share of articles on foreign politics in general declines continuously (from 9% in 1982 to 4% in 2008) until it reaches the same level as EU politics. The growing political importance of the EU appears to have come particularly at the cost of discussion of other topics of foreign politics. In this sense the EU has fulfilled one of its political goals: the contestation among the individual member states appears to have been partially replaced with the contestation with a single political actor, the EU.

In the tabloid press, however, EU politics as the main focus of an article seems to be a lost cause. The share of articles discussing EU politics never reaches 2%, by 2008 it has dropped back to 1.3%. Vertical Europeanization in the tabloid press remains limited to a casual observation of EU governance, EU politics never becomes the focus of discussion.

Horizontal Europeanization: continuous interest in European neighbours

Our previous study found a clear stagnation for horizontal Europeanization in the quality press and with the extension of the sample to 2008 this finding does not change. Though the share of articles discussing other European countries differs from one data point to the next, the overall level of mutual observation does not increase (see Figure 3).

Figure 3.

Share of articles discussing other (West) European countries.

This pattern is mirrored in the tabloid press. While the overall level of interest in the European neighbours is only slightly smaller (an average of 15% of articles compared to the 18% in the quality press), no trend over time can be identified: the political integration of the European countries is not accompanied by a growing familiarity with one another. The openness towards foreign countries in the tabloids may at first glance seem counter-intuitive; however, a closer look at the topics of these articles reveals that here the tabloid papers are even less interested in politics than in news of national origin: foreign VIPs, catastrophes or weird stories set in foreign countries are the focus of these articles. So the intensity of the European community of communication that is constructed for the readers of the tabloids is similar to that in which the broadsheet readers participate. However, it is based on non-political topics; the European neighbours are mostly used as a resource for soft news.⁹

For this indicator the countries analysed can easily be divided into two different groups. Germany and Austria display a strong interest in the politics of their neighbouring countries, on average 23% and 24% of articles focus on politics in (West) European countries (see Figure 4). This openness towards other European countries, however, is a particular trait of the quality paper FAZ, which is by far the most horizontally Europeanized of all the quality papers, the German tabloid Bild shows an only just about average interest in its neighbours. In Austria there appears to be a slight trend towards a renationalization: both in Kronen Zeitung and Die Presse national topics gain ground compared to articles on other European countries.

Figure 4.

Share of articles discussing other (West) European countries by country and newspaper type.

By contrast, France, Denmark and Great Britain discuss the affairs of their European neighbours in only 13% (F) or even just 8% (GB) of all articles. Poland has a special position here as topics from (West) European countries continuously gain ground between 1989 and 2003 – Poland is hence the only country in the sample undergoing a process of horizontal Europeanization.

For the second indicator, discursive integration, the difference between the newspaper types is more pronounced, in the tabloids only 8% of quoted speakers are from European countries, only half of the share in the broadsheets (16%). In particular, the public debate in tabloid newspaper appears to be more nationalized than that of the quality papers addressing a more elite audience.

Explaining the multi-segmented European public sphere

The article now proceeds to test the five hypotheses relating to the potential influence of national and editorial political discourse cultures through a logistic regression analysis (see Tables 1 and 2).¹⁰ For both indicators of vertical Europeanization the regression models reveal an important impact of editorial political discourse cultures: the chances of an article in a quality newspaper referring to EU institutions or focusing on EU politics are significantly higher than for tabloid articles by a factor of 2.5 for EU institutions and 2.6 for EU politics (H3). Curiously it is not the self-proclaimed mission of a newspaper to promote EU integration and make the EU understandable to its readers (ambassador newspapers) that increases vertical Europeanization but only the general policy of extensive, thorough political analysis of the analyst newspapers (H4).

National political discourse cultures play a lesser role in models of vertical Europeanization. Power/size of the country has no significant impact (H1), length of membership only a contradictory one: being published in a long-time member state of the EU apparently reduces the chances of an article containing references to EU institutions, but increases the chances of an article discussing EU politics as a main topic. For horizontal Europeanization both models confirm the negative influence of power/size: the smaller and less important the country, the higher the chances of foreign coverage and discussion (H1). There is no support for H2, in fact the length of membership has a negative impact on mutual observation: new members of the EU are apparently more interested in the affairs (but not voices) of their new neighbours than the established members.

Table 2.

Logistic regression of influence factors on indicators of horizontal Europeanization.

	Focus on other EU countries (N = 18,913)		Citing speakers from other EU countries (N = 18,913)
Potential influence factors	e ^β	e ^β _stand	e ^β	e ^β _stand
Year of analysis	.96^*	.95^*	–	–
Quality paper	–	–	1.71^***	1.31^***
Analyst paper	1.23^***	1.10^***	–	–
Ambassador paper	.57^***	.77^***	–	–
Caterer paper	.50^***	.71^***	–	–
Power/size	.82^***	.83^***	.73^***	.75^***
Length of membership	.83^***	.82^***	–	–
Focus on EU topic	6.19^***	1.36^***	6.54^***	1.38^***
Length of article	.90^**	.90^**	1.45^***	1.41^***
Commentary/editorial	.66^***	.86^***	.46^***	.75^***
External author	.50^***	.77^***	.27^***	.60^***
Constant	1.15^***	.17^***	.02^***	.06^***
Nagelkerke R² adj.		.07		.10

Notes: e ^βlogistic effect coefficient; e ^β _stand standardized logistic effect coefficient.

^* p < .05, ^** p < .01, ^*** p < .001 (Wald).

On the level of editorial political discourse cultures the classic distinction between quality and tabloid newspapers appears to be the most relevant for the chances of European speakers being cited (H3). The degree of mutual observation, however, seems more determined by the self-perception of journalists concerning EU and foreign coverage than by the resources available to them (H4): of the quality newspapers only those committed to a thorough analysis of foreign affairs – of the analyst type – discuss other EU member states particularly often. For those (quality) papers with a mission to promote the EU (the ambassadors), mutual observation appears to be not that important. And newspapers of the caterer type display an even greater disinterest in what is going in the neighbouring countries: with a logistic effect coefficient of just .50 the chances of an article focusing on other European countries are halved if it appears in a caterer newspaper – which includes not only most tabloids but also the British quality paper The Times.

Thus our data offer no support for our final hypothesis (H5), in all four regression models it is the editorial political discourse cultures that have the greater impact on the levels of vertical and horizontal Europeanization, while country differences only play a minor role. No matter in which European country a quality paper (or an analyst/ambassador newspaper) is published, it will always contain more information on the EU and other European countries than its tabloid counterpart.

Discussion

Coming back to our research questions the results of our latest content analysis confirm the general stability of the pattern of ‘segmented Europeanization’ which Wessler et al. (2008) had identified for the quality press: while the process of European integration leads to a growing interest in the EU in each of the analysed public spheres, they remain separate, displaying no signs of increasing interest in each other, or in the development of a common European discourse. In response to RQ1 we can ascertain that this national segmentation remains stable over time: the update of the original analysis for the year 2008 yielded no signs of an increase in mutual observation and discursive exchange. As confirmed by the regression models, national political discourse cultures continue to have a significant impact on levels of Europeanization, leading us to conclude that each nation still displays its own particular pattern of Europeanization.

Concerning RQ2 it can be said that the pattern of ‘segmented Europeanization’ holds true also for the tabloids, in which an increase in the monitoring of EU institutions is not accompanied by the development of a common discourse. We can thus conclude that the Europeanization of public spheres is best understood as multi-segmented: it is not only segmented in relation to nations, but also in relation to the type of media outlet (Hepp et al., 2009: 47). This pattern of multi-segmentation is corroborated by the results of the logistic regression analysis which attribute to editorial political discourse cultures a consistent influence on levels of vertical and horizontal Europeanization.

Editorial political discourse cultures have a particularly strong effect on vertical Europeanization (see Table 3). Even though tabloid references to EU institutions increase over time, the type of editorial political discourse culture retains its very strong impact on both indicators. Tabloid papers (and those of the caterer type) apparently do not feel a strong responsibility to inform their readers about EU institutions and politics. Contrary to expectations, however, a newspaper’s commitment to promoting the European project (ambassador papers) does not result in an above average number of references to or articles on the topic – instead it probably translates into a different way of discussing EU politics which could only be identified through a more detailed analysis of their coverage.

Table 3.

Overview of results for tested hypotheses.

Tested hypotheses		Europeanization
		Vertical		Horizontal
		Visibility EU institutions	Focus on EU politics	Focus on other EU countries	Speakers from other EU countries
Stronger Europeanization for. . .
National pdc	H1: small countries	False	False	True	True
National pdc	H2: long membership	False	True	False	False
Editorial pdc	H3: quality papers	True	True	False	True
Editorial pdc	H4: analyst/ambassador papers	(True)	(True)	(True)	False
H5s: National pdc has greater impact than editorial pdc		False	False	False	False

pdc = political discourse culture.

On the dimension of horizontal Europeanization, the use of editorial political discourse cultures based on the journalists’ self-perceptions concerning EU and European foreign news coverage has proved to be especially fruitful for a deeper understanding of the patterns of political communication. In the case of mutual observation, this typology laid open differences between newspapers that the traditional categorization of tabloid vs quality would have overlooked: intensive EU foreign coverage appears to be limited to a group of quality papers that are committed to an extensive analysis of foreign events even beyond their readers’ expectations and wishes (for more details on the journalistic self-perception of the analyst, see Hepp et al., 2009).¹¹ By contrast, a greater openness towards and interest in the arguments and opinions of European foreign speakers appears to be characteristic of all quality newspapers.

The extended sample has allowed us to assess that the impact of national political discourse cultures is less than originally supposed:¹² for both dimensions of Europeanization their influence is filtered and muffled by the stronger lenses of the different editorial political discourse cultures. This is particularly true for vertical Europeanization, which is only affected by the length of EU membership. Apparently newer member states have a more superficial interest in EU politics, thorough discussion of EU politics is far more likely to occur among the founding or early members such as Germany, France and Great Britain. It may be that casual references to political institutions are more easily adapted according to the news agenda, while a change in the focus of articles represents a more fundamental and thus more long-range change in national political discourse cultures. For horizontal Europeanization, the impact of national differences is somewhat stronger, as both mutual observation of and discursive exchange with European neighbour states are more important in the smaller EU member states.

What conclusions can we now draw from the results of this study concerning the state of the European public sphere and its potential to serve as a source of legitimacy for EU government?

First of all, the European public sphere has (for the present) stabilized. The update of the previous content analysis has shown that the trend of vertical Europeanization has not continued beyond 2003, in 2008 – the year of the global financial crisis – interest in EU institutions and politics has predominantly stagnated, if not ebbed. At the same time the level of horizontal integration has remained stable, no dramatic changes in observation and discursive exchange between European neighbours can be observed. Apparently the Europeanization of national public spheres is not a process ad infinitum, a plateau has been reached. However, this should not be problematic for the European public sphere serving as a source of legitimacy for the European Union: though the political relevance of the EU and its institutions continues to grow, national institutions also remain important as political actors, even if their role may be limited to the translation and adaptation of EU policy. Given the continuing low levels of opportunities for collective identification with Europe in public discourse (Kleinen-von Königslöw, 2010), the legitimacy of the EU may actually be improved by the presence of national political institutions as objects of collective identification and community bonding in public discourse.

Second, the European public sphere is multi-segmented. Even though our study employed a very basic definition of the public sphere, focusing on its transparency function and assessing transparency only for the vertical and horizontal extension of the debate, it revealed important differences between the scope of the public sphere as it is experienced through the lenses of newspapers from different nations and of different editorial political discourse cultures. Hence the multi-segmentation of the European public sphere may have a significant impact on its potential to serve as a source of legitimacy for the EU. If the European public sphere looks that different for the concerned citizens depending on whether they participate by means of a quality or a tabloid paper from this country or another country, how are they supposed to reach joint political decisions based on very disperse political knowledge? And taking into consideration that opportunities for collective identification with Europe are even rarer in the tabloid papers (if not non-existent, Kleinen-von Königslöw, 2010) the growing discontent of the common European people with ‘their’ European Union becomes understandable. The audience of the tabloid press may be obtaining more and more information on the EU and its institutions but these are not embedded in at least sporadic appeals to a common European identity, leaving them with no opportunities to make the European project their own.

Despite the decade that has passed since Gerhards’ (2001) first appraisal of the European public sphere, an empirical outing in the field still contributes important findings which fundamentally add to our understanding and evaluation of the phenomenon. The process of political integration in the European Union will not be concluded for some time, another round of applicants is waiting in the wings for admission to the EU, and the global financial crisis continues to challenge the European community of nations, leaving us wondering how much political and financial instability the European Union can sustain. Thus European public sphere research will retain its purpose for quite some time and should continue to trace the developments of its object of analysis and not mistake the currently observed apparent stability for a final state and destination.

Footnotes

Funding and Acknowledgement

This article presents results from the research project ‘The Transnationalization of Public Spheres in Europe’ directed by Andreas Hepp at the German Research Foundation’s Collaborative Research Centre ‘Transformations of the State’ at the University of Bremen and Jacobs University (former project heads were Hartmut Wessler [2005–2007] and Bernhard Peters [2003–2005]). I am greatly indebted to Michael Brüggemann with whom I have co-authored an earlier article from which I draw some of my arguments. I also thank Klaus Schönbach for his valuable comments and my former colleagues Johanna Möller and Swantje Lingenberg as well as the student coders for their contributions to the data collection.

Notes

References

Adam

(2008) Do mass media portray Europe as a community? German and French debates on EU enlargement and a common constitution. Javnost/The Public 15(1): 91–112.

AIM Research Consortium (2006) Understanding the LOGIC of EU Reporting in Mass Media. Analysis of EU Media Coverage and Interviews in Editorial Offices in Europe. Bochum, Freiburg: Projektverlag.

Berkel

(2006) Political parallelism in news and commentaries on the Haider conflict: A comparative analysis of Austrian, British, German and French quality newspapers. Communications 31(2): 85–104.

Boomgaarden

Vliegenthart

de Vreese

Schuck

(2010) News on the move: Exogenous events and news coverage of the European Union. Journal of European Public Policy 17(4): 506–526.

Brüggemann

Kleinen-von Königslöw

(2009) ‘Let’s talk about Europe’: Why Europeanization shows a different face in different newspapers. European Journal of Communication 24(1): 27–48.

Brüggemann

Schulz-Forberg

(2009) Becoming pan-European? Transnational media and the European public sphere. International Communication Gazette 71(8): 693–712.

Cohen

(1997) Deliberation and democratic legitimacy. In: Bohman

Rehg

(eds) Deliberative Democracy: Essays on Reason and Politics. Cambridge, MA: MIT Press, pp. 67–92.

Dahlgren

(2005) The internet, public spheres, and political communication: Dispersion and deliberation. Political Communication 22: 147–162.

D’Haenens

(2005) Euro-vision: The portrayal of Europe in the quality press. Gazette: The International Journal of Communication Studies 67(5): 419–440.

10.

De Vreese

(2001) Frames in television news: British, Danish, and Dutch television news coverage of the introduction of the euro. In: Hjarvard

(ed.) News in a Globalized Society. Gothenburg: Nordicom, pp. 179–196.

11.

De Vreese

(2007) The EU as a public sphere. Living Reviews in European Governance 2(3): 4–20.

12.

Ferree

Gamson

Gerhards

Rucht

(2002) Four models of the public sphere in modern democracies. Theory and Society 31(3): 289–324.

13.

Gerhards

(2001) Missing a European public sphere. In: Kohli

Novak

(eds) Will Europe Work? Integration, Employment and the Social Order. London and New York: Routledge, pp. 145–158.

14.

Gripsrud

(1992) The aesthetics and politics of melodrama. In: Dahlgren

Sparks

(eds) Journalism and Popular Culture. London: Sage, pp. 84–95.

15.

Habermas

(1997) Between Facts and Norms. Cambridge: Polity Press.

16.

Habermas

(1998) The Inclusion of the Other: Studies in Political Theory. Cambridge, MA: MIT Press.

17.

Habermas

(2009) Europe: The Faltering Project. Oxford: Polity.

18.

Hallin

Mancini

(2004) Comparing Media Systems: Three Models of Media and Politics. Cambridge: Cambridge University Press.

19.

Hepp

Wessler

(2009) Politische Diskurskulturen: Überlegungen zur empirischen Erklärung segmentierter europäischer Öffentlichkeit. Medien und Kommunikationswissenschaft 57(2): 174–197.

20.

Hepp

Möller

Kleinen-von Königslöw

. (2009) Political discourse cultures in Europe: Explaining the multi-segmentation of the European public sphere through a transnational and transcultural perspective. In: Carpentier

Pruulmann-Vengerfeldt

Kilborn

et al. (eds) Communicative Approaches to Politics and Ethics in Europe: The Intellectual Work of the 2009 ECREA European Media and Communication Doctoral Summer School. Tartu: Tartu University Press, pp. 45–58.

21.

Kantner

(2006) Die thematische Verschränkung nationaler Öffentlichkeiten in Europa und die Qualität transnationaler politischer Kommunikation. In: Imhof

Blum

Bonfadelli

. (eds) Demokratie in der Mediengesellschaft. Wiesbaden: VS Verlag für Sozialwissenschaften, pp. 145–160.

22.

Kleinen-von Königslöw

(2010) Europe for the people? The Europeanization of public spheres in the tabloid press. In: Trefás

Lucht

(eds) Europe on Trial: Shortcomings of the EU with regard to Democracy, Public Sphere and Identity. Innsbruck et al.: Studienverlag/Transaction, pp. 44–60.

23.

Kleinen-von Königslöw

(2011) Media and the European public sphere: Common misconceptions and shared conclusions on the representations of the EU and Europe. In: Lodge

Sarikakis

(eds) Communication, Mediation and Culture in the Making of Europe. Bologna: Il Mulino, pp. 41–62.

24.

Kleinen-von Königslöw

Möller

(2009) Nationalisierte Europäisierung. Die Entwicklung der polnischen Medienöffentlichkeit nach 1989. Nordost-Archiv. Zeitschrift für Regionalgeschichte 18: 219–234.

25.

Koopmans

Statham

(eds) (2010) The Making of a European Public Sphere: Media Discourse and Political Contention. Cambridge: Cambridge University Press.

26.

Koopmans

Erbe

Meyer

(2010) The Europeanization of public spheres: Comparison across issues, time, and countries. In: Koopmans

Statham

(eds) The Making of a European Public Sphere: Media Discourse and Political Contention. Cambridge: Cambridge University Press, pp. 63–96.

27.

Liebert

(ed.) (2007) Europe in contention: Debating the constitutional treaty. Special issue of Perspectives on European Politics and Society 8(3).

28.

Machill

Beiler

Fischer

(2006) Europe-topics in Europe’s media. The debate about the European public sphere: A meta-analysis of media content analysis. European Journal of Communication 21(1): 57–88.

29.

Peters

(2005) Public discourse, identity and the problem of democratic legitimacy. In: Eriksen

(ed.) Making the European Polity: Reflexive Integration in the EU. London and New York: Routledge, pp. 84–123.

30.

Peters

(2008) Public Deliberation and Public Culture: Theoretical and Empirical Writings 1993–2006, ed. H Wessler, with a preface by J Habermas. Basingstoke: Palgrave Macmillan.

31.

Pfetsch

Adam

Berkel

(2008) The contribution of the press to Europeanization of public debates: A comparative study of issue salience and conflict lines of European integration. Journalism 9(4): 465–492.

32.

Risse

(2010) A Community of Europeans? Transnational Identities and Public Spheres. Ithaca, NY and London: Cornell University Press.

33.

Sifft

Brüggemann

Kleinen-von Königslöw

. (2007) Segmented Europeanization: Exploring the legitimacy of the European Union from a public discourse perspective. Journal of Common Market Studies 45(1): 127–155.

34.

Sparks

(2000) Introduction: The panic over tabloid news. In: Sparks

Tulloch

(eds) Tabloid Tales: Global Debates over Media Standards. Lanham, MD: Rowman and Littlefield, pp. 1–40.

35.

Tobler

(2010) The Europeanization of national public spheres or domestication of the European Union in national discourses. In: Tréfas

Lucht

(eds) Europe on Trial: Shortcomings of the EU with regard to Democracy, Public Sphere, and Identity. Innsbruck et al.: Studienverlag/Transaction, pp. 61–83.

36.

Tréfas

Lucht

(eds) (2010) Europe on Trial: Shortcomings of the EU with regard to Democracy, Public Sphere, and Identity. Innsbruck et al.: Studienverlag/Transaction.

37.

Trenz

H-J

(2004) Media coverage on European governance: Exploring the European public sphere in national quality newspapers. European Journal of Communication 19(3): 291–319.

38.

Triandafyllidou

Wodak

Krzyzanowski

(2009) The European Public Sphere and the Media: Europe in Crisis. Basingstoke: Palgrave Macmillan.

39.

Van de Steeg

(2006) Does a public sphere exist in the European Union? An analysis of the content of the debate on the Haider case. European Journal of Political Research 45(4): 609–634.

40.

Vetters

Jentges

Trenz

H-J

(2006) Exploring the EU’s social constituency: Patterns of public claims-making in constitutional debates in France and Germany. ARENA Working Paper 18.

41.

Voltmer

Eilders

(2003) The media agenda: The marginalization and domestication of Europe. In: Dyson

Goetz

(eds) Germany, Europe and the Politics of Constraint. Proceedings of the British Academy 119: 73–197.

42.

Wessler

Peters

Brüggemann

. (2008) Transnationalization of Public Spheres. Basingstoke: Palgrave Macmillan.