AGENTS OF TRANSNATIONAL DEBATE ACROSS EUROPE
THE PRESS IN EMERGING EUROPEAN PUBLIC SPHERE
Abstract
This article aims at assessing the theoretical and empirical role of the national press in the emerging European public sphere. The study draws on Europeanisation as the emerging framework for transnational communication across European nation states. It assumes that the press itself may perform as a political actor and make a substantial contribution to Europeanisation by advocating European integration and by broadening its scope to include the perspectives of all EU member states and the EU itself. In order to discern the influence or role of the media - its "voice" - this study analysed the content of editorials of 28 newspapers in seven European countries along two dimensions. First, the receptiveness of the press towards European perspectives is investigated by measuring the degree to which its editorials feature European scopes. Second, the study examines newspapers' attitudes about European integration as a political project. The overall findings point to a remarkable representation of European perspectives, and substantial support for EU integration, by the national press in Germany, France, Italy and Spain. The newspapers in the Netherlands and Switzerland were somewhat more parochial, but still supportive. We also see that the United Kingdom (UK) media deviate substantially from these patterns. This study concludes that, in contrast with the findings of earlier studies, the press must be regarded as a significant agent of Europeanisation fostering transnational linkages of public debate.
BARBARA PFETSCH
Barbara Pfetsch is Professor at the Institute for Media and Communication Studies, Freie Universität Berlin; e-mail: pfetsch@zedat.fu-berlin.de.
O
rN si
o
00 o o rN
LO
£
A 3
a
0 £
I *
(R 0 C
K
fN
Introduction
In light of the recent stagnation of political reform in the EU, European politics appears as a rather silent diplomacy of national political elites who struggle to find ways and means to revitalise the constitutional process. In fact, voices calling for more public debate and democracy across Europe have become cautious. From a scholarly point of view however, the claim of a democratic deficit within the EU, its lack of a public sphere, and the formation of a European identity are still pressing issues. This state of affairs in EU politics corroborates how urgently an inclusive public discussion about common matters is needed. The European public sphere is widely regarded as a legitimate source for information about the European political project and it is seen as an instrument that may help to build a collective identity which may in turn trigger a sense of belonging to a common European community (Kielmannsegg 1996). Most scholars agree that the national mass media would be the primary forum within which a European public sphere would appear - if it is ever to emerge. In fact, the mass media are at the center of a rich strand of research that seeks to assess the degree and forms of Europeanisa-tion of national public spheres (Peters et al. 2005, Wimmel 2005, van de Steeg 2006, Kleinen von Königslöw et al. 2005, Adam 2007, Trenz 2006, Gerhards 2000, LeTor-rec et al. 2001, deVreese 2003, Brantner et al. 2005, Kevin 2003, Eilders and Voltmer, 2003, Koopmans and Pfetsch 2006, Pfetsch et al. 2008).
In most of the studies that look at the media in the context of EU political communication, the roles and functions of the media remain rather vague. In particular, research has largely failed to recognise and declare that the media fulfil two functions. As theories of the political public sphere point out, the mass media are the institutionalised forum of debate, which serves as a central linkage between the public and the political structure (Neidhardt 1994). In this function, they are conveyors of information about issues and actors according to their professional norms and values. However, the media are not merely serving as a channel of com-y munication and forum for exchange; as political scientists emphasise, the media must also be seen as political actors themselves, who legitimately raise their own voice (Page 1996, Cook 2006, Pfetsch and Adam 2008) and thereby impact political will formation and public opinion (Page et al. 1987). They do so in particular by assigning relevance to issues for public debate and by expressing their own opinions in specifically designed sections of media outlets. It is argued here that the media can make a significant contribution to the emergence of a European public sphere by: (1) shifting the scope of issues from national to European angles, thereby focusing on European perspectives; and, (2) speaking in favour of European integration, thereby supporting the political project of Europe.
Against this background, this study investigates the role of the national media in European public debate theoretically and empirically. It starts out by discussing the research on the emerging Europeanisation of the public sphere, pointing out some theoretical and empirical shortcomings in defining Europeanisation and sorting out the role of the media therein. The second part of the article is devoted to investi-
___________________________________________r______________________________________
gating the voice of national press in seven countries with regard to two questions: First, how open is the national press for transnational European scopes which feature the view of the EU or other EU countries? By measuring the receptiveness
for European angles, the study allows for conclusions regarding the press as an agent of transnational debate and motor of Europeanisation. Second, how supportive CO of European integration is the national press? By assessing the media's evaluation ^ of the EU political project, this study seeks to determine whether the press acts as conveyor of "good news" about Europe or - to the contrary - as a purveyor of Euro scepticism. Empirically, the voice of the press is analysed on the basis of 4740 editorials of 28 newspapers in seven European countries, specifically the United Kingdom, France, the Netherlands, Germany, Italy, Spain and Switzerland.
Europeanisation of National Public Spheres and the Role of the Media
The urgency of an inclusive European public sphere is widely agreed upon from a democratic point of view (Kielmannsegg 1996, Scharpf 1999). A scholarly debate, however, has now begun with controversies about how it should be theoretically conceptualised and measured (Neidhardt et al. 2000, Risse 2002). It has been widely acknowledged that a European public sphere must be a "mass-mediated" public sphere because the media provide for the only inclusive channels of communication between Europe and the public. At the same time, a significant strand of research highlights the centrality of the media in a potentially emerging European public sphere (Gerhards 1993 and 2000, Schlesinger 1997, Schlesinger and Kevin 2000; Kunelius and Sparks 2001, Kevin 2003, Koopmans and Erbe 2004, Trentz 2006). However, all efforts to establish transnational mass media that could maintain the political functions of a democratic European public sphere have failed. If there are supranational European media to be detected, they are not at all so inclusive as to reach the broad (mass) political public. Rather, they are confined to a limited audience of political, cultural, or business elites, or take the form of non-political media that specialise in sports and music (Kevin 2003, 38-41). Linguistic boundaries, cultural heterogeneity, and the fact that media systems are strongly bound to national mass audiences are crucial - and perhaps insurmountable - barriers to the formation of a unified European public sphere.
In consideration of the non-existence of a general European media infrastructure, further studies point out that Europeanisation "is for the most part dependent on the output of the national media" (Kevin 2003, 52). Thus far, there are two strands of empirical studies that yield rather contradictory findings and conclusions regarding the role of the media in the Europeanisation of public debate.
First, it has been assumed that Europeanisation of national public spheres would occur when nationally-based mass media shift their focus away from the national political arena towards the European level. According to this perspective, increasing salience for European issues and actors in the national media and growing references to transnational contexts would be indicative for a Europeanised public debate. Relying on rather simple measures, such as the salience of European issues and actors in national news coverage,3 a series of studies by Peter de Vreese (2004), Kevin (2003), and Gerhards (2000) find rather low levels of coverage of European issues and actors in the media of Western European countries. For instance, Peter and de Vreese (2004), who analyse the representation of EU stories in the television news of five European countries,b find that - except for Denmarkc - the proportion of EU-related stories in television news has been marginal during periods
of routine news. The proportions only increase slightly in the context of major European political events.d In light of a rather low level of visibility of European issues, this research concludes that the media have never left the nation state and that there is no meaningful European public sphere to speak of. Thus, the (national) media in this strand of research appear as obstacles to a European public sphere. From early on it has been argued that the logic of the media - and in particular the professional news values, the goal of attracting large national audiences, and their disinterest in administrative, policy-driven coverage - results in a lack of interest in European issues and angles (Gerhards 1993). As a consequence, the national media were held responsible for keeping public political debate within the boundaries of the nation state.e
Second, a different picture emerges from studies that investigate transnationali-sation in terms of direct and indirect references to European aspects in larger media coverage of Europe. Trenz (2006) finds a remarkably high visibility of implicit and explicit European aspects in reports on European politics. He interprets his findings as "banal" Europeanism and as a clear indication of an ever-growing European public sphere. Similar conclusions were drawn from a series of case studies that look at the coverage of single European issues or policies, such as the Haider case (van de Steeg 2006, Berkel 2006), the corruption scandal of the European Commission (Trentz 2000) or BSE (Berkel 2006).
Considering the vast array of contradictory findings and interpretations as well as methodological shortcomings of measuring Europeanisation (for a detailed discussion see Neidhardt 2006), one might well become sceptical about this trajectory of research. Do we have evidence after all that the media do matter if we look at the larger picture of the emergence of a truly European public sphere that can be connected with the functions of political legitimacy, the inclusion of European citizens, and the meaning of European democracy? At this point the work of Michael Bruter (2005) is important because it brings the media back as relevant political actors (Page 1996, Cook 2006) that possess the potential to move public opinion y (Page et al. 1987). Bruter (2005) provides empirical evidence that political communication about Europe in the media does impact the level of European identity by individual European citizens. However, according to Bruter, it is not just the salience of European issues in the media, but the explicit disposition of positive aspects or good news about Europe (as opposed to bad news) that stimulates European identity and thus provides an important basis for a European community of communication. If this finding is taken further, we may indeed conclude that the media's voice and their performance are critical variables for the emergence of a European public sphere. If we grant the media an explicitly political role, their performance in transnational political debate across Europe, and their position about European integration need to be further investigated.
Political Functions of Press Commentary in Europeanisation: Research Questions
This study aims at assessing the political role of the media in transnational communication across Europe by focusing on the press. We have chosen to investigate newspaper editorials and not television news, since we wanted to capture the deliberative aspect of European public debate and not the simply the visibility of EU
issues. The objective was to include those mediated formats which would contain at least some degree of discursive political content that can be attributed to specific LH political profiles of media outlets (Peters et al. 2005, Kleinen von Kleinigslöw et al. ^ 2005). The analysis of newspaper commentary and editorials is one way to study the media as political actors. We assume that this format allows one to clearly identify the attitudes of the specific media title and set it apart from other voices in public debate. Regarding Europeanisation the press may choose to emphasise parochial or transnational frames since they may support or oppose European integration.
With regard to European public debate, newspapers fulfil crucial functions, which Eilders and Voltmer (2003, 9-10) discuss with respect to: (1) agenda setting and second level agenda-setting (or framing); and, (2) opinion formation. The agenda-setting role of the media (Dearing and Rogers 1996, Ghanem 1997) is usually linked to the study of news coverage, which is largely dependent on input from external sources and information.' The media's task here is to select issues and events for news coverage according to their professional protocol and rules. On the other hand, the media also act on their own account when they publish opinion pieces and editorials. In these formats, they are taking the liberty to deviate from the news agenda that is dictated by external events. Within the recognised format of editorials and commentaries, they select issues and assign relevance to them as topics for public deliberation (Neidhardt et al. 2004). This mechanism not only applies to issue selection but also to framing (Entman 1993). As Juan Diez Medrano (2003) shows, press commentary is a most obtrusive format for framing Europe in public debate.
Concerning the Europeanisation of public debate, the crucial empirical question which is addressed here is, To what degree does the national press use its political role to either contain public debate within national boundaries or - to the contrary - to open it up for transnational European angles? The power to transnationalise public debate is directly related to the function of framing issues with regard to national or transnational scopes. The media may constrain an issue within the realm of the national debate, or alternatively, resonate with scopes of other countries or supranational or even global actors. The active role of the press with regard to Eu-ropeanisation may therefore vary between the promotion of rather parochial views (which may even lead to nationalistic closures of public debate) or of truly transnational European points of view in favour of deliberative European democracy.
The most genuine and active function of the media as political actors refers to opinion formation. This implies that the press not only reports about the positions of other actors, but take their own stance on issues by commenting on the opinions and actions of non-media political actors. This function is the predominant purpose of editorials. National media express their political preferences insofar as they expose a more or less stable and coherent commentary line (Neidhardt et al. 2004). In most European countries the press is usually bound to the left-right spectrum of politics regarding domestic issues and policies. However, as European politics do not necessarily fall within such a left-right cleavage (Mittag and Wessels 2003, 419), there is more room for autonomous media opinion. Therefore, one would expect that the freedom of the media to come up with their own position about European issues may be higher than this freedom for national issues (Eilders and Voltmer 2003, 11). Against this background, the research question which is addressed here
is whether the press uses its power to advocate European integration as a political project or to side with rather Eurosceptic views. In light of empirical findings by Bruter (2005) that "good news" about Europe enhances European identity it is consequential indeed whether the media join the supporters or the opponents of European integration.
European Public Sphere as a Structure Communicative Linkages
For measurement and empirical data this study draws on the Europub.com-project.g This research set out to measure the Europeanisation of public political communication across seven European countries.11 The study rests on the assumption that public debate can be determined by investigating the contents of political statements or "claims" of actors (such as governments, parties, civil society actors, media etc.) as well as the patterns of transnational communication. Europeanisation was assessed by measuring the relative density of public communication within and between different national and supranational political spaces. The center of the communicative space is the national public sphere of each country. The next level of communication refers to other national European public spaces, which comprise the EU member countries. The third level includes the transnational, European political space, in which the European institutions and common policies are situated. The degree to which public spheres can be deemed "national," "transnational" or "European" depends on the density of communicative linkages within and between these spaces. In fact, Europeanisation is measured in terms of horizontal (between EU member-states) and vertical (between EU member-states and the EU level) communication linkages that are made by the various actors (Koopmans and Erbe 2004). Since our approach to public debate allows for measuring the contents and the reach of communication for each actor category, we were able to single out the genuine voice of the media.
Many studies of Europeanisation seem to expect that transnationalisation of public debate occurs as a given and on a general level of politics or with regard to y all policies. This conjecture however, largely ignores the institutional structure of the EU and the differences in the actual policy competences of European actors. It is unrealistic therefore to expect an overall high and stable degree of European debate across all policy fields and issues. Instead, we may expect public communication to resonate with patterns of political institutions and their decision making power. This relates to the actual distribution of power between the various European and the national levels and to the intergovernmental or supranational nature of the European decision-making process. As a consequence, the study design does not capture public communication on a general level but seeks to examine seven issue fields. They are selected to represent various institutional settings and levels of national and European governance. The seven issue fields addressed are: (1) European integration, including six substantive policy domains which vary systematically in terms of political integration; (2) Monetary politics (i.e. currency politics and interest rate); (3) Agriculture (i.e. subsidies, livestock and dairy quotas, animal disease control), representing issue areas that are character-
. ......................................„---------------------------
ized by a high degree of EU involvement; (4) Immigration (i.e. entry and exit); (5) Troops deployment, which mark the policy areas with increasing EU competences, while national decision-making is still predominant; (6) Retirement and pension
schemes; and, (7) Primary and secondary education domains, which have largely remained under the umbrella of national or regional decision making, and where the role of the EU is very limited.
The research design, which varies by issue areas or policy domains with respect to EU competences, allows us to observe whether the shift of political decision-making and responsibility away from the nation state is followed by the opening up of communicative space to supranational angles, actors and interpretations. It also enables us to overcome the shortcomings of a number of studies of Europeanisation which look at only one issue (Van de Steeg 2006, Wimmel 2005) or one country (Brantner et al. 2005) and then draw broad and general conclusions about the Europeanisation of public sphere.
Methodology and Data
We conducted a quantitative content analysis of press editorials to study the openness and the attitude towards European integration. The opinion articles examined related to the seven issue fields identified above in the press of the United Kingdom, France, The Netherlands, Germany, Italy, Spain and Switzerland during the period between 2000-2002.i The country selection includes six EU member countries of which we expect at least some degree of Europeanisation due to the political significance of EU membership. We also included Switzerland as non-EU member-state because we wanted to get some hint as to whether Europeanisation of public debate is bound to formal EU membership.
In order to capture the range of ideological positions in the media system of each country, the variation in styles of addressing the political preferences of the audience, and the regional aspects of the media system, we chose four daily newspapers of different types in each country under study: a centre-left as well as a centre-right quality newspaper, a tabloid newspaper, and a regional newspaper featuring a specific regional identity (see Appendix A1). Since not all media systems fit into such a clear-cut set of theoretically defined dimensions, we had to make compromises in the selection of newspapers for the study.j In countries where the media landscape does not feature a clear-cut tabloid press, newspapers that can be regarded as functional equivalents were selected for the study. In countries where we could not identify such a functional equivalent newspaper, we chose a second regional press title. We recorded 5294 editorials for the three years period between 2000 and 2002. Regarding the distribution of commentaries, we analysed between 595 and 1004 cases per country.k (United Kingdom 910, France 678, Netherlands 732, Germany 1004, Switzerland 681, Spain 595, and Italy 694). Of the 5294 editorials which relate to the seven issue fields under study, 89.5% (4739) were included in this analysis since they contain explicit political statements.' With respect to the relative distribution of issue fields and countries (see Appendix A2), the issue of European integration has been the most salient one in five of the seven countries under study. Interestingly, the issue of European integration was quite frequently represented in the Swiss press. By contrast, European integration was rather low on the UK media's priority list.
The general approach to data collectionm was quantitative content analysis of newspapers editorials. The coding unit and unit of analysis was the single editorial, which is regarded as one claim by the journalist.11 The coding recorded the
event that triggered the editorial (initiating event) and its scope. If the journalist addresses his demands to actors or institutions in criticism or support, the variable addressee was coded. We also qualified the object actor whose interest is involved by the statement, and the content of the demand or the issue and the argumentative framing that supports it. In order to determine whether or not we are dealing with Europeanised claims, we coded the geographical or polity level at which the actors and institutions mentioned are situated (e.g., European or national scope), as well as, in the case of national or sub-national actors, the country where they are based (e.g., Germany, France).
In the analysis we used one indicator to represent the openness of the press for European perspectives, and another one for the position of the media towards EU integration. For the first indicator, we computed a summary variable that captures - for each editorial - the degree to which it refers to European (as opposed to national) scopes. An editorial is regarded to be fully Europeanised if a European scope is apparent (1) in the initiating event that triggers the editorial; plus (2) the most important issue that is being discussed; plus (3) the most important addressee; plus finally, (4) the most important object actor who is seen to benefit or suffer from the matter. Thus, if a claim is fully Europeanised it scores a value of four. By contrast, if none of the four dimensions refers explicitly to a European or EU scope, the variable takes a 0 value.o
The second indicator refers to the position of the press towards European integration. For each editorial we recorded whether the journalist was positive overall about European integration (+1) or negative overall (-1). A positive value for this variable signifies the commentator's political support of European integration, while a negative value is recorded if European integration is opposed or rejected. Since one can assume that the editorials represent the political and ideological line of the outlet, the aggregation of editorial opinion was taken to represent the position of the newspaper.
y The Openness for European Scopes in the Press
This study looked at the degree to which the editorials in national newspapers feature communicative linkages that transcend the national realm in order to assess the performance of the press to open public debate for European angles.
Table 1 shows the findings relating to the openness of editorials towards European scopes? in all seven issue fields across countries. At first glance it is evident that there are strong differences between the countries under study regarding the level of EU references vs. non-EU references. Almost 60% of all analysed editorials in Spain contained one or more linkages towards the EU, whereas in the UK these linkages were present in only 37% of editorial. In general, a sharp contrast appeared
UK, the Netherlands, and Switzerland on the other hand. At the time, the French print media in our study were most open to European dimensions: Only 42% of all editorials excluded EU scopes, while 29% were strongly Europeanised (all four EU scopes). Similar levels of fully-Europeanised editorials were observed for the press in Spain, Italy and Germany. This means that in these countries at least one in four editorials under study was committed to transnational European linkages. By contrast, in the UK, almost two thirds of the newspapers under study excluded
Table 1: Openness towards European Scopes by Country (EU summary scope, index*)
	DE	CH	ES	IT	UK	FR	NL	all	N
	%	%	%	%	%	%	%	%	
All issue fields (including European integration)									
No EU-Scope	47,0	48,6	41,2	41,3	63,0	42,5	56,3	50,2	1970
1 EU-Scope	8,4	21,5	9,9	4,7	14,1	7,8	10,0	10,7	421
2 EU-Scopes	8,9	5,7	11,5	12,3	10,3	8,0	6,1	8,9	351
3 EU Scopes	12,0	6,1	14,4	17,7	7,2	13,3	7,6	10,8	426
All EU-Scopes	23,6	18,1	23,0	24,1	5,3	28,5	20,1	19,3	759
Total	100,0	100,0	100,0	100,0	100,0	100,0	100,0	100,0	
Means (1-4)	1.56	1.23	1.68	1.79	0.77	1.77	1.25	1.38	
N (editorials)	593	442	243	644	806	527	672		3.927
Issue fields without EU integration									
No EU-Scope	65.8	80.5	54.6	58.5	74.8	61.0	70.2	67.6	1968
1 EU-Scope	7.8	5.6	10.4	6.2	12.4	7.9	9.5	8.9	259
2 EU-Scopes	6.1	1.5	10.9	11.7	6.2	6.5	4.8	6.7	195
3 EU Scopes	7.1	3.4	11.5	9.7	4.3	10.1	5.0	6.8	197
All EU-Scopes	13.2	9.0	12.6	13.9	2.4	14.4	10.4	10.0	291
Total	100	100	100	100	100	100	100	100	
Means (1-4)	.94	.55	1.17	1.14	.47	1.09	.76	.83	
N	424	267	183	453	679	367	537		2910
* (4=fully Europeanized editorial, 0=no European scope)
European dimensions. Instead they featured national, and to a small degree supranational, aspects. The UK press' rather strong concentration on national angles is corroborated if we notice that only 5% of the press commentaries were fully Europeanised. The Dutch newspapers ranked second with respect to the level of national scope. However, they were not as self referential as the British press. Approximately 20% of Dutch newspaper editorials featured all EU scopes, compared to 56% of editorials that were restricted to non-European aspects. A similar pattern was evident for the newspapers in Switzerland, in which 49% of editorials appeared as non-European compared to 18% of fully Europeanised statements.
The findings so far refer to all seven issue fields that were covered in the study including EU integration. One could argue therefore, that the level of Europeanisation mirrors the a priori selection of a policy area which by definition contains a high level of EU references. In order to control for this effect, we calculated the level of Europeanisation in editorials excluding the issue field of EU integration. The results (lower part of Table 1) reveal the same national patterns, except for the fact that the level of European scopes was much lower. About one third of the editorials now contain at least one European scope, while roughly two thirds took a national perspective. Once again, the press in the UK, the Netherlands and Switzerland stood out by their extremely low levels of European scopes. On the other hand, the print media in Spain, Italy, France and Germany featured a considerable number of EU perspectives, ranging between 33% (Germany) and 45% (Spain) of editorials with at least one or more EU scopes.
The level of Europeanisation drops considerably if the most integrated policy field is excluded; this suggests that the differences between the countries not only mirror national idiosyncrasies, but also the degree to which the press in each country covers certain policy areas in their editorials.
Table 2: Openness towards European Scopes by Issue Field (EU summary scope, index*)
	Monetary	Agricult	Immigrat	Troops	Pension	Education	EU- Integration	al	z
	%	%	%	%	%	%	%	%	
No EU-Scope	16,6	45,1	73,9	85,3	91,3	96,0	0,2	50,2	1970
1 EU-Scope	17,9	13,1	10,1	6,7	3,4	2,0	15,9	10,7	421
2 EU-Scopes	15,1	10,1	5,8	5,2	1,5	1,2	15,3	8,9	351
3 EU Scopes	17,9	12,8	5,8	2,6	1,1	0,3	22,5	10,8	426
4 EU-Scopes	32,4	18,9	4,5	0,3	2,6	0,5	46,0	19,3	759
Total	100,0	100,0	100,0	100,0	100,0	100,0	100,0	100,0	
Means (1-4)	2.32	1.47	.57	.26	.20	.07	2.98	1.38	
N	614	328	398	659	265	646	1.017		3.927
* (4=fully Europeanized editorial, 0=no European scope)
If we compare the openness for European scopes across issue fields, we see significant variation. On the one hand, monetary policy and European integration issues cannot but be debated within an overall European frame of reference. Commentaries regarding these issue fields were completely Europeanised. On the other hand, the media discussed pension plans and education issues within a national reference frame only. Immigration and troop deployment were treated y almost exclusively with national scopesq as well. The most interesting issue with respect to EU scope is agriculture. Although agricultural issues are addressed in an integrated EU policy, press editorials were equally split between only national scopes and an EU scope. Although the political decision-making power about agriculture rests with Brussels, the press also supported national perspectives in their editorials. Only half of the newspaper editorials about agriculture related to a European scope, and only 20% were completely Europeanised. This finding does not represent the division of labor in political decision-making between the national and the European level, and it can be interpreted as a manifestation of ^ the power of the media to set their own priorities in framing issues other than the political actors. Thus, the media's role in public debate may not only work in favour
0
C >
of transnational linkages. Instead of opening up the national debate for European frames, the press may comment about truly European issues in light of national perspectives. This has become evident at least to some degree in the editorials about agricultural policy.
Concerning the potential of the press to avoid Europeanisation, one may assume that the tabloid press in particular advocates the closure of public debate. The f^ expectation that the tabloids appear as agents of national perspectives is based on
Table 3: Openness towards European Scopes by Newspaper Type (EU summary scope, index*)
	Quality press	Regional and Tabloid Press	all	N
	%	%	%	
No EU-Scope	48.4	52.9	50.2	1970
1 EU-Scope	9.8	12.1	10.7	421
2 EU-Scopes	8.7	9.4	8.9	351
3 EU Scopes	12.0	9.0	10.8	426
All EU-Scopes	21.2	16.5	19.3	759
Total	100	100	100	
Means (1-4)	1.48	1.24	1.38	
N	2378	1549		3927
* (4=fully Europeanized editorial, 0=no European scope)
their rather populist appeal which they must uphold to keep their audience. In our content analysis, we found that the tabloid press, as well as the regional press, advocated national views to a higher degree than the quality press (Table 3). In fact, the national broadsheets were more inclined to stress European dimensions. However, the differences between quality papers on the one hand, and tabloids and regional papers on the other hand, were rather moderate. We did not find significant differences between the openness of quality papers and conservative outlets and left-wing newspapers towards European scopes. Instead, the differences between countries and issue fields were much stronger than the variation across newspaper types. We also analysed the variation over time in the Europeanisation of newspaper editorials and hardly found any evidence for systematic changes over the three year period of 2000 to 2002.r
At the beginning of this study we argued that the media can contribute to the emergence of a European public sphere by shifting the scope of issues from national to European angles. We conclude the descriptive analysis about the openness of press editorials towards European scopes by finding that the media work in both ways. They may open up public debates for European scopes, and they thereby act as a motor of Europeanisation. They may also work in the other direction by interpreting European policies with explicitly national angles. Regarding the emergence of a European public sphere, our analysis shows that the potential of the media to foster transnational communication is conditional, since it rests on differences between countries, policy fields and newspaper types.
Evaluation of European Integration
The second question regarding the performance of the press in Europeanisation of public debate refers to opinion formation. Here we ask whether the voice of the press supports "good news" about Europe which, according to Bruter (2005), enhances the formation of European identity at the citizen level. Considering our second indicator, the analysis shows that the journalist revealed a positive attitude towards European integration in 55% of editorials containing evaluations, Moreover, the level of truly negative comments about European integration was below five percent across all countries, except for the UK (Table 4). If we compare across
countries, the most positive attitude towards European integration was found in the Italian and French press, where four out of five editorials were favourable towards EU integration. It seems that at the beginning of the 2000s there was a strong consensus in the media in these countries that EU integration was an overall positive project. There was a sharp contrast however, between the southern European countries and the UK media, where almost every second press editorial declared an overtly negative attitude towards European integration.
Table 4: Editorial Position towards EU Integration by Country
	DE	CH	ES	IT	UK	FR	NL	all	N
	%	%	%	%	%	%	%	%	
Negative (-1)	2,9	4,0	0,3	1,1	49,0	3,0	6,7	9,0	200
Mixed/ambivalent	30,9	45,7	69,0	14,4	34,8	18,7	41,4	35,6	794
Positive (+1)	66,2	50,4	30,7	84,5	16,2	78,3	51,9	55,4	1.234
Total	100,0	100,0	100,0	100,0	100,0	100,0	100,0	100,0	
Positive - negative (%)	63,3	46,4	30,4	83,4	-32,8	75,3	45,2	46,4	
Means	0,63	0,46	0,30	0,83	-0,33	0,75	0,45	0,46	
N	385	278	316	368	296	300	285		2.228
U
S
i
0
£ *
i *
IR 0
C
«
fN
m
The attitudinal space between the most positive and most negative newspapers was filled by the press in Germany, Switzerland and the Netherlands. In the Swiss and Dutch media, the positive voices overrode the negative ones by 46 percentage points; in the German press the difference amounted to even 63 percentage points. Finally, the Spanish press was most idiosyncratic about the attitudes of the commentators towards European integration; Although the print media in Spain ranked among the most Europeanised newspapers in terms of scopes, the vast majority of opinions (69%) were mixed, undecided or ambivalent. In addition, the Spanish press' share of 31% of editorials with positive attitudes is moderate if we compare it with the other national media.
Table 5: Editorial Position towards EU Integration by Issue Field
	Monetary	Agricult	Immigrat	Troops	Pension	Education	EU- Integration	al	N
	%	%	%	%	%	%	%	%	
Negative	12,2	8,8	3,4	2,0	5,0	0,0	8,2	9,0	200
Mixed/ambivalent	42,5	55,8	48,3	31,4	60,0	50,0	27,9	35,6	794
Positive	45,3	35,4	48,3	66,7	35,0	50,0	64,0	55,4	1.234
Total	100,0	100,0	100,0	100,0	100,0	100,0	100,0	100,0	
Positive - negative* (%)	33,1	26,6	44,9	64,7	30,0	50,0	55,8	46,4	
Means	0,33	0,27	0,45	0,65	0,30	0,50	0,56	0,46	
N	640	181	89	51	20	12	1.235		2.228
It is hardly surprising that positive evaluations by journalists predominantly concentrated on debates in the issue field of European integration. As the data in CO Table 5 indicate, two thirds of editorials about this field held positive opinions. It ^ is also quite plausible that the issue fields that are typical for a predominantly national debate, namely pension and education, were characterized by a high level of mixed or undecided attitudes regarding European integration. The highest level of approval was discovered in monetary policies, immigration and troop deployment. Interestingly enough, journalists strongly supported the involvement of the EU in troop deployment which, at the time, was not at all an area of strong EU competences. By contrast, EU competences in agriculture policy were seen with mixed feelings. In this issue field with far reaching EU competences we saw almost 60% of undecided or ambivalent editorials. Moreover, facing 9% of press commentaries with negative attitudes, the critical potential is comparatively high. This finding indicates that European policies are judged quite differently depending on contexts and national conditions. Thus, the fact that an issue field is politically integrated does not trigger media support of European integration in this area; In agriculture specifically, political integration invites dispute and contention in the media.
Table 6: Editorial Position towards EU Integration by Newspaper Type
	Quality press	Regional and Tabloid Press	all	N
	%	%	%	
Negative	4,6	1 6,4	9,0	200
Mixed/ambivalent	35,1	36,6	35,6	794
Positive	60,3	47,0	55,4	1.234
Total	100	100	100	
Positive- negative (%)	55,7	30,6	46,4	
Means	0,56	0,31	0,46	
N	1.403	825		2.228
In the same vein, we found that the attitudes towards European integration varied across newspaper types. The journalists of the tabloid press held more negative opinions on average than the commentators of the quality newspapers. We see in Table 6 that the majority of journalists' opinion pieces in the quality press was positive (60%) and only 5% percent were negative. In contrast, the proportion of negative statements in the regional and the tabloid press was four times as high: we observed 16% of claims opposing European integration, compared to 47% supporting it.
Patterns of Media Performance in Europeanisation
Combining both dimensions of the empirical analysis - the level of Europeanisation and the evaluation of European integration - four patterns regarding the role of the national press in emerging European public sphere could be identified: (1) The press may down-play or ignore its potential to set up transnational linkages and contain political debate mostly within national boundaries on the one hand, and comment rather negatively about European integration on the other hand. This role
pattern can be observed for the press in the UK, which has been the most parochial in scope, and which was the most critical about European integration during the three years under study. We also observed this pattern of press performance with respect to agricultural policy: In this issue area, the media revealed a tendency to narrowcast an otherwise European debate with national perspectives. At the same time, the number of critical voices about European integration was above average in this issue field. (2) The opposite pattern of media performance comes to the fore when the press uses its potential to open up public debate for European angles and speaks up openly in favour of European integration. This behaviour was displayed by the national press in Germany, France and Italy. (3) Another pattern for the press in Europeanisation is to be open in scope, yet indifferent or moderately positive about Europe as a political project. We found this pattern of media behaviour in Spain. (4) A complementary pattern to the previous one is one in which press performance is rather narrowly focused on national angles, yet supportive towards European integration. This attitude was found in the Netherlands and in Switzerland, where the press was low in Europeanisation, but moderately positive in their support of European integration.
Our findings point to variation in press performance regarding transnational communication since we see differences across countries and issue fields. On the basis of bivariate descriptive tables we are unable to assess the significance and strength of the factors that lead to variation in the level of Europeanisation and in the evaluation of European integration. Therefore, we chose to test the explanatory factors by computing Multivariate Analysis of Variance (MCA) with country, issue field, newspaper type and year as independent variables and EU scope as a dependent variable.
According to Table 7 our model explains 60% of the variance of the level of European scope. The strongest and most highly significant effect is caused by the issue field (Beta .745). This means that the propensity of the press to open up for European angles depended strongly on the issues of the debate. At first glance, this finding does not seem spectacular at all, since it mirrors the design of our study which anticipated EU governance structures. However, the finding that integrated issue fields draw a Europeanisation of public debate is highly consequential. It means that we can expect the transnationalisation of public debate if institutional arrangements foster EU competences. With regard to the media, it seems that if the EU is in charge of decision-making, the national press is prone to open up the debate for arguments and references EU level actors or actors in other EU countries. Thus, it becomes clear that the Europeanisation of public debate is a function of institutional structures of European governance.
^	With respect to the openness towards European scopes, we also find a significant
effect of the country (Beta .230) in which the media appeared. This clearly confirms the differences between the national media as regards their level of Europeanisation. Compared to the national cleavage and the issue field, the newspaper type (Beta .072) and the year (.125) are hardly sources of variation of European scopes. Interestingly, the print media within each country shared their tendency or their reluctance to support European angles. Finally, the degree to which the media opened up their
0 £
C >
discursive space was stable during the three years under study.
Regarding the evaluation of European integration, the explanatory power of the multivariate model is lower, yet the four independent variables in the MCA still
Table 7: Multivariate Analysis of European Scopes and Editorial Positions (MCA)
Dependent Variable	Independent Variables				Strength of each model
	Country	Issue field	Paper type	Year	
	Beta	Beta	Beta	Beta	R-square
EU Summary scope	.230	.745	.072	.125	.60
Editorial Position to EU integration	.506	.189	.128	.061	.32
accounted for 32% of the variance (R-square). Interestingly, the effect of the country (Beta .506) was by far the strongest source for variation. The issue field (Beta .189), the newspaper type (Beta .128) and year (.061) had very weak effects. This means that - despite newspaper type and the issue - the attitudes of the press towards European integration reflect national attitudes. The picture here is rather clear: At the time, UK journalists clearly opposed European integration, while French, German and Italian journalists supported it, regardless of whether they were writing about monetary politics or immigration, or whether they belonged to the staff of a quality newspaper or a tabloid. Ideological cleavages that appear between the media in domestic politics did not play a role in media opinion on European integration, nor did the type of newspaper (broadsheets and tabloids).
Discussion
Our study focuses on the debate about the emergence of the European public sphere and assesses the role of the national press in this process. It highlights the central role of the media in public debate - not only as mediators of the issues and opinions of political sources, but also grants them a political actor's role. The opinion page in newspapers is the legitimate forum in which to play this role. Therefore, in the empirical part of this study, editorials and press commentaries were analysed as indicators for the autonomous voice of the media.
Regarding Europeanisation, the national media may act as proponents of an inclusive European public sphere if they convey transnational linkages in public communication across Europe. However, they may also prevent the development of such Europeanisation if they decide to constrain public debate within national perspectives. The potential of the national media to open up or to contain public debate within national angles gains relevance with respect to the normative democratic theory based argument which demands an inclusive European public sphere. We confront this discussion and argue that the political role and the autonomous voice of the media must be included in an explanation of an inclusive public sphere across Europe. From a theoretical point of view, the media may act as crucial proponents of transnational communication. The empirical question, however, is whether and to what degree they have acted as such proponents and whether the press has reacted to the increasing political decision-making power of the EU by opening up their national perspective on policies for transnational scopes.
In our study of the newspapers of seven European countries, we were able to detect different patterns of how the press performed its role in European commu-
nication in the early 2000s. In France, Italy, Spain and Germany we saw a considerable level of openness towards European angles. The press in the Netherlands and Switzerland was more withdrawn, yet we detected some degree of transnational European linkages. Moreover, regarding media opinion, we found strong support of EU integration among the print media in France, Italy, Spain and Germany and also, although to a lesser degree, in the Netherlands and even Switzerland. The UK press was the only exception; it maintained rather national angles and opposed EU integration. We found that the openness of mediated political debate for EU scopes was largely a residual of the issue at (political) stake, while the evaluation of EU integration by the press followed national boundaries. This finding gains significance because it sheds light on the conditions required for the emergence of a European public sphere. Thus, we can expect Europeanised debate to emerge in the media if the political decision-making power lies with the EU and if the consequences of EU politics matter to politics within the member states. In other words, if issues are politically contentious within a nation's public, and yet decided in Brussels, chances are high that the media engage in transnational public debate. In almost all countries of this study except for the UK, the press redirected their national focus and monitored the angles of EU actors and EU neighbouring countries. In our conception of Europeanisation a redirection of communicative linkages is the first step towards a sustainable European communication structure which fulfils the political requirements of an inclusive democratic public sphere. Only if this structure for transnational communication has been established can we expect common deliberations at the citizens' level.
However, the media are not redirecting their scope unconditionally. Our study shows that transnational public debate about issues in which the EU has no stake is not to be expected in the media. Across all countries in our study, the press focused on national perspectives for those issues where the decision-making power rests with national institutions. This is not to say, however, that the media would not support EU integration. Our findings for the early 2000s point to rather positive press attitudes about European integration. However, we must be careful about inferring that this support is more than of temporary nature. Media opinion is £ part of the general political culture of a country, and the media resonate with the ^ national political cleavages and the "ups and downs" of public opinion.
0 £
Notes:
1.	Often also, conclusions are drawn from secondary analyses of data that were gathered for other purposes and are not always suited to grasp the intricacies of the European multi-level polity (e.g., Gerhards 2000; Eilders and Voltmer 2003).
2.	France, Germany, the Netherlands, Denmark, and the United Kingdom.
3.	The Danish level of television coverage of EU affairs amounts to 19 percent of all news stories in Q routine periods, and 25 percent in summit periods.
£	4. Such as summit meetings. It is a general finding that the visibility of European issues and actors is
highly event driven (Trentz 2006).
5. Following this line of argument, it can be tested whether the media are rightly accused of misrepresenting or playing down European issues and actors. For instance, in Germany we found ^q	that the press is more Europeanised and less polarized about European politics than the political
C^	parties, the interest groups and government actors (Koopmans and Pfetsch 2006).
6.	In the terminology of public sphere sociology (Neidhardt 1994), the media here deploy an arena for the exchange of political and societal actors who want to be heard in public and therefore raise
their voice publicly (or with the help of public relations strategies).	CO
7.	This study is part of the larger project, "The Europeanization of Political Communication and Mobilization in European Public Spheres (EUROPUB.COM) that has been funded by the European Commission in the context of its 5th framework programme (Project No. HPSE-CT2001-00046). For a detailed description of the project and its methods, see http://europub.wz-berlin.de.
8.	For a detailed description of the theoretical approach to Europeanisation see Koopmans and Erbe (2004).
9.	This was a rather optimistic period in the history of the European Union in which the Euro was newly established and on the political level the enlargement and the EU constitution were expected to become successful reform projects. However, since the media in their editorials express their principal ideological positions we are confident that over a three year period of content analysis we were able to capture rather stable long term attitudes towards Europe and European politics and not temporary, cursory fashions of short term media orientations.
10.	While we were able to detect a left and a right quality paper in all countries under study, we faced some difficulties with regard to the tabloids. There are different cultural notions about tabloids in the various countries and some media systems hardly include newspapers that would fit in the category at all.
11.	In order to restrict the coding effort to manageable proportions, we used a sampling strategy: In those countries with a low number of commentaries, France, the Netherlands and Switzerland, we selected every day of the year. In the United Kingdom, Italy and Spain - countries that yield
a medium number of commentaries - we registered the commentaries on every second day. For Germany, we sampled every fourth day. On the days of the sample we looked through all newspapers and selected all commentaries related to our seven issue fields.
12.	The rest showed no political references or speech act that allowed for further recording.
13.	The editorials were coded by native speakers in country teams which were carefully trained before coding and supervised throughout the whole coding period. For the reliability test, coders in each country team coded a random sample of seven commentaries from the Scotsmen, the Times and the Guardian of the year 2002. The inter-coder reliabilities were measured as the average match between the coders. The overall reliability calculated on the core variables of the analysis turned out to be highly satisfactory with an average match of 75% which corresponds to a reliability correlation of .87. The reliability scores for each specific variables can be obtained by the author.
14.	The Codebook for content coding of commentaries was developed by Adam et al. (2002). It is available at https://europub.wz-berlin.de.
15.	This is usually the national level, yet in the policy field of troops deployment and immigration, there are a limited number ofreferences to the supranational level.
16.	Communicative linkages between the national and EU level of politics mark a vertical form of Europeanisation, and communicative linkages between national and other EU-countries as horizontal form of Europeanisation. Both types of EU scopes are collapsed in the present analysis and contrasted with either only national references or supranational linkages. Supranational linkages here include references between EU-countries and International Organisations or bilateral references of EU-countries to non-EU-countries like the US or Russia or references between non-EU-countries.
17.	We know from other analyses that the level of supranational and multilateral references is comparatively high in troop deployment and still above average in immigration.
18.	Data are not displayed here, but can be obtained from the author.
References:
Adam, Silke. 2007. Symbolische Netzwerke. Der Einfluss der nationalen Ebene auf europäische
Öffentlichkeit. Köln: Halem. Adam, Silke, Barbara Berkel, Julie Firmstone, Emily Gray, Ruud Koopmans, Barbara Pfetsch, and Paul Statham. 2002. Codebook for content coding of commentaries/editorials. <https://europub. wz-berlin.de>
Berkel, Barbara. 2006. Konflikte als Motor einer europäischen Öffentlichkeit. Eine Inhaltsanalyse von Tageszeitungen in Deutschland, Frankreich, Großbritannien und Österreich. Wiesbaden: VS. Brantner, Cornelia, Astrid Dietrich and Florian Saurwein. 2005. Europeanisation of National
Public Spheres: Empirical Evidence from Austria. First European Communication Conference, Amsterdam.
Bruter, Michael. 2005. Citizens of Europe? The Emergence of a Mass European Identity. Houndmills: Palgrave Macmillan.
Cook, Timothy E. 2006. The News Media as Political Institution: Looking Backward and Looking
Forward. Political Communication 23, 159-171. Dearing, James W. and Everett M. Rogers. 1996. Agenda-Setting. Communication Concepts 6. Thousand Oaks: Sage.
De Vreese, Claes H. 2003. Framing Europe. Television News and European Integration. Amsterdam: aksant. Diez Medrano, Juan. 2003. Framing Europe: Attitudes to European Integration in Germany, Spain, and
the United Kingdom. Princeton: Princeton University Press. Eilders, Christiane and Katrin Voltmer. 2003. Zwischen Deutschland und Europa. Eine empirische Untersuchung zur Europäisierung der meinungsführenden deutschen Tageszeitungen, Medien und Kommunikationswissenschaft 51, 2, 250-270. Entman, Robert. 1993. Framing: Toward Clarification of a Fractured Paradigm. Journal of
Communication 4, 4, 51-58. Gerhards, Jürgen. 1993. Westeuropäische Integration und die Schwierigkeiten der Entstehung einer
europäischen Öffentlichkeit. Zeitschrift für Soziologie 22, 96-110. Gerhards, Jürgen. 2000. Europäisierung von Ökonomie und Politik und die Trägheit der Entstehung einer europäischen Öffentlichkeit, Kölner Zeitschrift für Soziologie und Sozialpsychologie. Sonderheft 40, 277-305.
Ghanem, Salma. 1997. Filling the Tapestry: The Second-Level of Agenda-Setting. In M. McCombs, D. L. Shaw, and D. Weaver (eds.), Communication and Democracy: Exploring the Intellectual Frontiers in Agenda-Setting Theory, 3-14. Mahwah: Erlbaum. Kevin, Deirdre. 2003. Europe in the Media: A Comparison of Reporting, Representation, and Rhetoric in
National Media Systems in Europe. Mahwah: Earlbaum. Kielmannsegg, Peter Graf. 1996. Integration und Demokratie. In M. Jachtenfuchs and B. Kohler-Koch
(eds.), Europäische Integration, 49-71. Opladen: Leske und Budrich. Kleinen-v. Königslöw, Katharina, Michael Brüggemann, Stefanie Sifft and Andreas Wimmel. 2005. The Europeanization of Public Spheres: Each Country on its Own? 55th Annual Conference of the International Communication Association, New York. Koopmans, Ruud and Barbara Pfetsch. 2006. Obstacles or Motors of Europeanization? German
Media and the Transnationalization of Public Debate. Communications 31, 2, 115-138. Koopmans, Ruud and Jessica Erbe. 2004. Towards a European Public Sphere? Vertical and Horizontal Dimensions of Europeanised Political Communication. Innovation 17, 97-118. ^	Kunelius, Risto and Colin Sparks. 2001. Problems with a European Public Sphere: An Introduction.
Javnost—The Public 8, 1, 5-20.
C	Le Torrec, Virginie, Philippe Blanchard; Guillaume Garcia and Charles Patou. 2001. Framing Europe:
News Coverage and Legitimacy of the European Union in Five Countries. European Community ^	Studies Association Seventh Biennial International Conference, Madison, Wisconsin.
Mittag, Jürgen and Wolfgang Wessels. 2003. The "One" and the "Fifteen"? The Member States between Procedural Adaptation and Structural Revolution. In W. Wessels, A. Maurer and J. OQ	Mittag (eds.), Fifteen into one? The European Union and Its Member states, 413-457. Manchester:
C^	Manchester University Press.
U
S
i 0 £
Neidhardt, Friedhelm. 2006. Europäische Öffentlichkeit als Prozess. Anmerkungen zum
Forschungsstand. In W. R. Langenbucher and M. Latzer (eds.), Europäische Öffentlichkeit und	^
medialer Wandel. Eine transdisziplinäre Perspektive, 46-61. Wiesbaden: VS.	CO
Neidhardt, Friedhelm. 1994. Öffentlichkeit, öffentliche Meinung, soziale Bewegungen. In F.
Neidhardt (ed.), Öffentlichkeit, öffentliche Meinung, soziale Bewegungen (Special issue of Kölner Zeitschrift für Soziologie und Sozialpsychologie), 7-41. Wiesbaden: VS.
Neidhardt, Friedhelm, Ruud Koopmans and Barbara Pfetsch. 2000. Konstitutionsbedingungen politischer Öffentlichkeit: Der Fall Europa. In H.-D. Klingemann and F. Neidhardt (eds.), Zur Zukunft der Demokratie (WZB-Jahrbuch 2000), 263-293. Berlin: Sigma.
Neidhardt, Friedhelm, Christiane Eilders, Barbara Pfetsch. 2004. Einleitung: Die Stimme der Medien - Pressekommentare als Gegenstand der Öffentlichkeitsforschung. In C. Eilders, F. Neidhardt and B. Pfetsch, Die Stimme der Medien. Pressekommentare und politische Öffentlichkeit in der Bundesrepublik Deutschland, 11-38. Wiesbaden: VS.
Page, Benjamin. 1996. The Mass Media as Political Actors. Political Science and Politics 29, 1, 20-25.
Page, Benjamin I., Robert Y. Shapiro and Glenn R. Dempsey. 1987. What Moves Public Opinion? American Political Science Review 81, 23-43.
Pfetsch, Barbara and Silke Adam. 2008. Die Akteursperspektive in der politischen
Kommunikationsforschung. Fragestellungen, Forschungsparadigmen und Problemlagen. In B. Pfetsch and S. Adam (eds.), Massenmedien als politische Akteure. Konzepte und Analysen, 9-26. Wiesbaden: VS.
Pfetsch, Barbara, Silke Adam and Barbara Eschner. 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.
Peter, Jochen and Claes H. de Vreese. 2004. In Search of Europe. A Cross-National Comparative Study of the European Union in National Television News. The Harvard International Journal of Press / Politics 9, 3, 3-24.
Peters, Bernhard, Stefanie Sifft, Andreas Wimmel, Michael Brüggemann and Katarina Kleinen-von Königslöw. 2005. National and Transnational Public Spheres: The Case of the EU. European Review 13, 1, 139-160.
Risse, Thomas. 2002. Zur Debatte um die (Nicht-)Existenz einer europäischen Öffentlichkeit. Berliner Debatte Initial 13, 15-23.
Scharpf, Fritz W. 1999. Demokratieprobleme in der Europäischen Mehrebenenpolitik. In W. Merkel and A. Busch (eds.), Demokratie in Ost und West, 672-94. Frankfurt a. M.: Suhrkamp.
Schlesinger, Philip. 1997. From Cultural Defence to Political Culture: Media, Politics and Collective Identity in the European Union. Media, Culture & Society 19, 3, 369-391.
Schlesinger, Philip and Deirdre Kevin. 2000. Can the European Union Become a Sphere of Publics? In E. O. Eriksen and J. Erik Fossum (eds.), Democracy in the European Union: Integration through deliberation?, 206-229. London: Routledge.
Trentz, Hans-Jörg. 2000. Korruption und politischer Skandal in der EU. Auf dem Weg zu einer europäischen politischen Öffentlichkeit. In M. Bach (ed.), Die Europäisierung nationaler Gesellschaften (Special issue of Kölner Zeitschrift für Soziologie und Sozialpsychologie) , 332-359. Wiesbaden: Westdeutscher Verlag.
Trenz, Hans-Jörg. 2006. "Banaler Europäismus." Eine latente Kategorie der Europäisierung politischer Kommunikation. In W. R. Langenbucher and M. Latzer (eds.), Europäische Öffentlichkeit und medialer Wandel. Eine transdisziplinäre Perspektive, 192-212. Wiesbaden: VS.
Van de Steeg, Marianne. 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, 609-634.
Wimmel, Andreas. 2005.Transnationale Diskurse in der europäischen Medienöffentlichkeit: Die Debatte zum Beitritt der Türkei. Politische Vierteljahresschrif, 46, 3, 459-483.
Appendix A1: Newspapers under Study
	Quality Press		Regional Press	Tabloid Press
	left	right		
Germany	Süddeutsche Zeitung	Frankfurter Allgemeine Zeitung	Leipziger Volkszeitung	Bild-Zeitung
Spain	El Pais	Abc	La Vanguardia	El Mundo
France	Le Monde L'Humanite	Le Figaro	Ouest France	
Italy	La Repubblica	Il Corriere della Sera	Il Mattino La Nazione	
Netherlands	De Volkskant	Het Algemeen Dagblad	De Limburger	De Telegraaf
United Kingdom	The Guardian	The Times	The Scotsman	The Sun
Switzerland	Neue Zürcher Zeitung	Le Temps	Le Matin	Blick
Note: In the case of the Netherlands, De Telegraaf fits best the Dutch conception of tabloid. In Italy instead of a tabloid, a second regional newspaper was chosen. In France no tabloid paper could be analysed. L' Humanité, the paper of the Communist party that was analysed neither fits the category of a regional nor of a tabloid paper. Therefore it was omitted in the analysis.
Appendix A2:
Distribution of Editorials with Claims in Issue Fields by Country
U
S
i
0
£ *
i *
IR 0
C
«
o
	DE	CH	ES	IT	UK	FR	NL	all	N
	%	%	%	%	%	%	%	%	%
Monetary politics	18.1	14.8	25.0	18.9	20.6	13.8	11.1	17.4	825
Agriculture	8.3	4.1	9.9	2.8	10.5	9.5	12.1	8.3	394
Immigration	10.4	10.0	14.3	7.1	7.2	6.2	14.3	9.8	465
Troops deployment	14.6	6.6	7.9	18.9	21.8	24.3	13.1	15.8	749
Retirement/pensions	11.6	15.7	4.7	4.3	2.8	9.5	5.0	7.5	354
Education	9.7	9.4	9.9	19.0	21.6	8.2	22.7	14.9	705
European integration	27.3	39.5	28.3	29.0	15.6	28.6	21.6	26.3	1248
Total	100	100	100	100	100	100	100	100	100
N	835	562	516	673	835	601	718	100	4740