COVERING THE EUROPEAN UNION FROM AN INTERGOVERNMENTAL TOWARDS A SUPRANATIONAL PERSPECTIVE? Abstract This article investigates the cross-national prevalence of five news frames in quality papers' coverage of the Treaty of Lisbon (EU Constitution). Three frames were identified in earlier studies: economic consequences, conflict, and human interest. Two additional frames were identified and composed: power and nationalisation. During the seven-month period leading up to the signing of the Treaty of Lisbon (December 2007), we analysed 341 articles from four quality papers: Le Monde (France), De Volkskrant (The Netherlands), De Standaard (Dutch speaking community of Belgium), and Le Soir (French speaking community of Belgium). Our results show that although significant differences between newspapers were found in the amount of framing, overall they reflected a similar pattern in the adoption of the news frames. The economic consequences frame, followed by the power frame, appeared most prominently in all of the newspapers' coverage. However, the conflict and nationalisation frames recurred in a significantly lesser degree. These findings indicate that the meaning behind the EU Constitution as a symbol of supranational unity could have led to a shift from a domesticated, conflict oriented coverage as found in previous studies to a more unified portrayal of the EU within and between the quality papers under study. ANNA VAN CAUWENBERGE DAVE GELDERS WILLEM JORIS Anna Van Cauwenberge is a Ph.D. Candidate at the Centre for Media Culture and Communication Technology, University of Leuven; e-mail: anna.vancauwen-berge@soc.kuleuven.be. Dave Gelders is former Assistant Professor at the Leuven School for Mass Communication Research; e-mail: dave.gelders@ gmail.com. Willem Joris is Researcher at Leuven Public Management Institute, e-mail: willem. joris@soc.kuleuven.be. LO O O O rN $ £ 3 a 0 £ I * (R 0 C This article examines the portrayal and use of frames in reporting on the European Constitution (Treaty of Lisbon). With this flagship treaty incorporating all existing EU treaties, the European Union's intention was to construct a coherent and flexible set of rules to cope with further enlargement and complexity of this supranational, multi-level political institution. The Constitution was intended to become the ultimate symbol of a united European Union, a "Constitution for European citizens" (European Council 2001, 23-24). However, French and Dutch referenda made it clear that the EU Constitution (then called The Treaty Establishing a Constitution for Europe) was not perceived as a "Constitution for European citizens" with results of 54.9 percent and 61.5 percent against. In this exploratory study, four quality newspapers from three countries that were in the centre of this EU Constitution debate will be examined: France and The Netherlands, the first countries to hold up the ratification process based on the results of the referenda, and Belgium, which hosts the capital of the EU, Brussels. In the aftermath of the Irish rejection of the European Constitution (Treaty of Lisbon) in June 2008, a survey1 was conducted among the Irish assessing, among other things, the reasons for non-participation in the referendum, the respondents' views about the campaign and the reasons for the "yes" or "no" vote (European Commission 2008). One of the most noteworthy findings is the statement that a lack of information is the main reason expressed by a quarter of the "no" voters to explain their preference. Over half of the people who did not vote in the referendum said this was due to a lack of understanding of the issue. These findings confirm the tendency, indicated by several surveys, that although a minority of the European citizens considers support to the EU negative (14 percent in 2008), only half trusts the EU and feels engaged with it (European Commission 2008; 2007). Several authors refer to this lack of interest and active support in terms of a democratic deficit (Meyer 1999; Schlesinger 1999; Ward 2001; 2004), legitimacy deficit (Meyer 1999; Baetens & Bursens 2005) and communication deficit (Brügge-man 2005) of the EU. It comes down to the fact that European citizens do not define y their rights and citizenship on the European level, but within the boundaries of their own nation. Despite the increasing transfer of economic and political policies from the national to the European level, there does not seem to exist a "European public sphere" (for example Kleinstüber 2001; Trenz 2004; Wimmer 2005). In this discourse a public sphere is considered vital for the healthy functioning of a democracy. Curran (1991) describes a public sphere as "the space between government and society in which private individuals exercise formal and informal control over the state: formal control through the election of governments and informal control through the pressure of public opinion" (Curran 1991, 29). Crucial in this process is the role media play in the construction and (re)presentation of a public sphere, by the distribution of information and the provision of an independent forum for public debate (Meyer 1999; Risse & Van de Steeg 2003). Basic criteria are described for the construction of a European public sphere, and comprise communication in different countries, on identical topics, at the same time, and with the use of the same frames (Brüggeman 2005). These findings raise questions regarding the European Constitution's portrayal in the media. Previous national and cross-national studies investigating media ^ coverage of the European Union draw largely on media analyses that measured a relatively limited set of content features. The scarce amount of EU news, concentrated around major European "eye-catching" events such as European summits fO and the introduction of the Euro, reported mainly through the national view; are ^ the most commonly identified characteristics of EU news (Peter et al. 2003, 2004; de Vreese et al. 2006; Meyer 2005; Drieskens & Fiers 2005). However, the way in which EU news is framed in news media (Semetko & Valkenburg 2000; de Vreese et al. 2001; d'Haenens 2005) plays a significant role in how news users reflect on these topics (Valkenburg et al. 1999). Besides an important influencing factor on opinion formation, greater visibility of EU topics in the news is also related to increased knowledge about the EU and an increase in political participation (de Vreese & Boomgaarden 2006a, 2006b). As political elites and journalists make sense of reality by providing certain ways to think about politics (Kinder 2007), researchers are not only interested in what news is covered, but also how this news is covered. Framing the EU Entman formulated a much cited definition of framing that defines framing as "to select some aspects of a perceived reality and make them more salient in a communication text, in such a way as to promote a particular problem definition, causal interpretation, moral evaluation, and/or treatment recommendation" (Entman 1993, 52). To grasp structure and organise large and complex streams of information, journalists use media frames to highlight certain aspects by excluding others, encouraging citizens to understand events and issues in a particular way. Obviously, news can be framed in a variety of ways. A distinction is made between issue-specific frames and generic frames. Where issue-specific frames appear in relation to specific issues or topics, generic frames have a more general application ranging from different topics, time, and cultural contexts (de Vreese 2003). This article studies the frequency of generic frames dealing with the EU Constitution. As stressed by de Vreese et al. using a common event to investigate frames in a cross-national manner brings not only better measures for cross-national comparison of framing of a common European event but also a reinforcement of the frame as a generic frame (de Vreese et al. 2001, 110). Semetko and Valkenburg (2000) point out that when focussing on media frames adopted in the portrayal of EU affairs, literature seems to indicate five commonly-used generic frames: conflict, economic consequences, responsibility, human interest, and morality. The conflict frame stresses conflicts between individuals, groups and institutions. Although this frame is typically applied in election coverage, portraying candidates and campaigns in terms of winning and losing, it is also one of the most commonly used frames in EU reporting. This frame is found especially in serious news outlets due to the more frequent coverage of economic and political news in these outlets. The economic consequences frame is often used in EU coverage as it approaches an issue by drawing attention to the consequences it will have economically on an individual, group, institution, or geographical entity. The responsibility frame presents an issue by attributing responsibility for its cause or solution to an individual, group or institution such as the government. A frame regularly connected with tabloid news is the human interest frame. Specific to this frame are the personal and emotional touches given to a specific issue, event, or topic. In search of more public attention, emotional, dramatic and personal angles are brought into the narrative of a news story. The morality frame gives a religious or moral swing to a news issue. Given the professional code of journalistic objectivity, news media mostly make indirect use of the morality frame by quotation or inference. In this study, two frames are analysed research identifies as commonly used in relation to the coverage of news and more specifically news about the EU, namely the conflict and the economic consequences frames (de Vreese et al. 2001; Semetko & Valkenburg 2000). We also added the human interest frame to investigate to what degree coverage of the European Constitution is made more accessible and comprehensible, by inserting more emotional and/or personal elements. Still this frame is expected to recur less frequently in quality papers' coverage of the EU Constitution. An inductive method was chosen to identify and compose two supplementary frames: power and nationalisation. Discourse analysis revealed the power discourse to be dominant in news about the European convention (d'Haenens 2005). We expect the power frame to be also prominent in news about the EU Constitution. The power frame can be defined as a frame that emphasises relations between parties/persons/states and the mutual power division. Previous research monitoring news content on the European Union also revealed the clear presence of a domestic angle. However, in-depth interviews with journalists of the EU press corps revealed that the tendency to nationalise EU-related news was reversed when reporting on the Constitution (Gleissner & de Vreese 2005). To test this indication we composed a nationalisation frame. This frame focuses on the own country / national politicians / national parties. This exploratory study will analyse the five above mentioned frames in four high-standing quality papers from three different countries: Le Monde (France), De Volkskrant (The Netherlands), Le Soir and De Standaard (respectively the French and Dutch speaking community of Belgium). Quality papers' less volatile character, compared to audiovisual media, and their better suited platform for elaborate, y in-depth coverage of political news are the main reasons for choosing for this medium. Even though the influence of television news can not be underestimated, high-standing quality papers remain a key reference, not in the least for other media outlets. France, The Netherlands, and Belgium as "EU capital" were, after the rejection of the European Constitution in the French and Dutch referenda, in the middle of the European Constitution debate. Hence, we expect a significant amount of coverage in the four selected quality papers. Although all three are original EU £ member states geographically located in the centre of the EU, they historically differ in their attitude towards the Union. France tends to be rather negative and critical towards the EU whereas Belgium has a more positive attitude. The Netherlands are positioned somewhere in between (European Commission, 2006). Research Question and Method C > The goal of this study is two-fold. First, we want to assess the prevalence of five generic news frames in four ---i.-------- -------- „ ---- i--i quality newspaper's coverage of the EU Constitution (Treaty of Lisbon) during a seven-month period leading up to the establishing of the Treaty of Lisbon (June-^ December 2007) in multiple countries. During this period, an agreement was made on the content replacing the rejected EU Constitution (The Treaty Establishing a Constitution for Europe). On 21 and 22 June 2007, European leaders decided to LH install an Intergovernmental Conference to further amend the temporarily so-called ^ "Reform Treaty." At the beginning of October, the draft version was completed and by mid-October, heads of state and government gave their approval after a final round of negotiations and concessions. In all three countries studied, a large majority of political parties were in favour of the Treaty of Lisbon. Based on previous findings on the framing of EU-news, we expect the conflict frame, economic consequences frame and power frame to be more present in the coverage of the EU Constitution than the human interest frame and nationalisation frame. Our first hypothesis is: The conflict frame, economic consequences frame and power frame are more frequently used in quality paper's coverage of the EU Constitution, than the human interest frame and nationalisation frame. Second, we want to examine whether the use of these news frames varies significantly between the selected quality papers from France (Le Monde), The Netherlands (De Volkskrant) and Belgium (Le Soir and De Standaard). Selection of these national quality news papers is based on the perceived position as a prominent quality newspaper, a high circulation rate, and regular reporting on EU affairs. Of the four newspapers, Le Monde is the most well-known. With an average circulation of 340,131, it is one of France's most prominent newspapers (OJD 2009). Perceived as a liberal, centrist-left newspaper, it is most commonly known for its international outlook and grounded political and economic analyses. With its focus on world news, foreign news covers 25 percent of the total newspaper coverage. In The Netherlands, De Volkskrant has a circulation of 263,845, the highest circulation rate among the Dutch quality papers (Cebuco 2009). This progressive leftist newspaper is appreciated for its mix of idiosyncratic news choice and original angles. The two Belgian quality newspapers have a similar circulation rate: 90,535 (Le Soir) and 95,940 (De Standaard) (CIM 2009). Both can be described as having no clear affinity towards the right or left. Le Soir and De Standaard are the leading quality newspapers in the French speaking community and the Dutch speaking community, respectively, of Belgium. Because of the different stances the three countries traditionally take on the EU, we could expect the power frame, conflict frame and nationalisation frame to be more used by Le Monde, than by De Volkskrant, Le Soir and De Standaard. These three frames focus on relations, power division, and conflicts between the own nation and other member states. The emphasis lays on an intergovernmental approach, instead of a supranational one which portrays the EU more independent of the different member states' interests and mutual relations. However, a European overview of the news coverage of the 2004 European parliamentary elections (de Vreese et al. 2006) showed that in the French press 45 percent of protagonists were EU actors, compared to less than 30 percent in Belgium, and 10 percent in the Netherlands. Furthermore, the coverage in Belgium and France turned out to be rather neutral, whereas a more negative tone appeared in the Dutch press. These contradictory findings make it difficult to formulate clear predictions about differences in the adoption of frames by the quality papers under study. Therefore we chose to formulate a research question: Do the French, Dutch and Belgian quality papers vary in the use of frames when covering the EU Constitution? We selected the articles by consulting the newspapers' online archives. A keyword search with the words "Reform Treaty," "Treaty of Lisbon," "EU Constitution," "European Constitution," "Constitution Europe," "EU Treaty" and "European Treaty" yielded 341 articles published within the seven-month period (June 1, 2007-December 31, 2007). We subjected the total 341 articles to a content analysis instrument that contained 27 closed questions by measuring article characteristics, visibility, tone, and framing of the articles. Inter coder reliability tests (Krippendorff's a) were conducted on a randomly selected sample of 10% of the news articles and ranged between .80 and 1.00 inter coder agreement. News Frame Measurement To measure the extent to which frames recurred in the news articles, we used a list of five sets of questions (17 in total). Each set of questions was meant to measure one of five news frames: human interest, conflict, economic consequences, power and nationalisation (see Table 1). The measurement of the conflict, economic consequences, and human interest frames is based on the categories developed by Semetko and Valkenburg (2000). To measure the power frame, we used a set of questions based on the power discourse scale as used by d'Haenens (2005) and tested by Van der Schoot (2002). The power discourse starts from the principle that "every community wants to achieve power and therefore needs to serve certain political interests" (d'Haenens 2005, 425). To measure the nationalisation frame, we composed a scale of two questions: (1) Does the article mention a connection between the EU and the country in question? (2) Does the article quote or articulate the ideas of national politicians or persons active on a national level? All the above questions had to be answered with "totally agree" (1), "more or less agree" (2), or "do not agree" (3). To investigate the extent to which the framing questions reflect underlying dimensions, we conducted a principal component analysis with varimax rotation on the 17 framing questions. The component analysis provided us with five clear-y cut news frames: human interest, conflict, economic consequences, power and nationalisation. The component analysis of the recurrent news frames explaining 66 percent of the total variance in the articles under study (N = 341), can be seen in Table 1. Only the component loadings higher then .50 are withheld for inclusion in the scales. To assess the internal consistency of the five frame scales, we used Cronbach's alphas. The alpha values were as follows: human interest frame scale, a = .69 (3 ^ items); conflict frame scale, a = .68 (3 items); economic consequences frame scale, a = .81 (3 items); power frame scale, a = .81 (3 items); and nationalisation frame, a = .82 (2 items). Inter-item correlations among the five frames ranged from r = -.047 between the nationalisation and power frames and r = .475 between the power and conflict frames. An overall average inter-item correlation of .258 was found. Five multi-items scales were composed by averaging the scores on the individual items in each component and transforming them into a scale ranging from .00 (frame not present) to 1.00 (frame present). (R 0 Table 1: Varimax Rotated Component Analysis of Articles in Le Monde, De Volkskrant, Le Soir and De Standaard (N = 341) r^ Framing items Human interest Conflict Economic Consequences Power National Human interest 1. Does the story provide a human example or "human face" on the issue? .817 -.057 .061 .142 .055 2. Does the story employ adjectives or personal vignettes that generate feelings of outrage, empathy/caring, sympathy, or compassion? .771 .180 .057 -.027 -.032 3. Does the story emphasise how individuals and groups are affected by the issue/problem? .664 .411 .090 .107 .035 4. Does the story go into the private or personal .415 -.329 .028 .211 .194 lives of the actors? 5. Does the story contain visual information that .215 -.025 .086 .423 -.039 might generate feelings of outrage, empathy/ caring, sympathy or compassion? Conflict 6. Does the story reflect disagreement between parties/individuals/groups/countries? .098 .606 .391 .285 .263 7. Does one party/individual/group/country .150 .807 .145 .068 .073 reproach another? 8. Does the story refer to two sides or to more than .031 .590 .161 .243 -.116 two sides of the problem or issue? .098 .370 .670 .183 .183 9. Does the story refer to winners and losers? Economic Consequences 10. Is there a mention of financial losses or gains now or in the future? .044 .165 .845 .125 .084 11. Is there a mention of the costs/degree of .073 .133 .745 .235 .239 expense involved? 12. Is there a reference to economic consequences .061 .034 .847 .112 .007 of pursuing or not pursuing a course of action? Power 13. Does the article mention the relationship of one or more EU states with one or more states .009 .131 .202 .761 -.369 (EU or otherwise)? 14. Does the article mention the power of one state .025 .234 .201 .869 -.020 within the EU regarding decision-making? .080 .358 .189 .687 .346 15. Does the article mention relative power? Nationalisation 16. Does the article mention a connection be- tween the EU and the individual country? -.013 -.013 .225 -.023 .852 17. Does the article articulate the ideas of or led the word to national politicians or persons that .107 .060 .116 -.116 .881 are active on a national level? Results Characteristics of Format and Content Between 1 June 2007 and 31 December 2007, 341 articles were published by the four quality newspapers under study. Le Monde published the most articles (n = 107), followed by De Volkskrant (n = 93), De Standaard (n = 71), and Le Soir (n = 70). The coverage concentrated on key moments in the EU Constitution debate. On 21 and 22 June, articles increased significantly when EU leaders at a European summit decided to hold an intergovernmental conference for the renegotiation of the EU Constitution. This resulted in 117 articles (34 percent) published between 15 and 27 June 2007. The second time a significant number of articles were published occurred at the time of the European summit on 18 and 19 October. This yielded 36 articles (11 percent) within a period of three days (18, 19 and 20 October). The signing of the Treaty of Lisbon on 13 December was covered by 23 articles (7 percent) published between 13 and 19 December. These results confirm earlier findings of EU news concentrating around major European events. Of the 341 articles, 215 articles (63 percent) are long (more than 400 words) and 75 percent of the articles are written by the newspapers' own journalists. Only a quarter of the news articles make use of illustrations, of which two-third are coloured photographs. When looking at the placement of the articles, 212 (62 percent) appeared in the foreign section of the newspaper and 89 (26 percent) under the heading of opinion/analysis. Thus, a majority of the articles (57 percent) are news reports with background briefing. However, an analysis of the four newspapers individually revealed that De Volkskrant differs from the other newspapers by publishing more articles under the opinion/analysis section (n = 34; 36.6 percent) than in the foreign section (n = 31; 33.3 percent). This can largely be explained by the specific character of this newspaper, reserving proportionally more space for opinion than the other newspapers under study. When focussing on the tone of the coverage, most of the articles (n = 286; 84 percent) are neutral, a tendency also y found in former research. Use of News Frames To investigate whether the conflict frame, economic consequences frame, and power frame are more frequently used in quality paper's coverage of the EU Constitution than the human interest frame and nationalisation frame, we conducted t-tests on the overall mean scores of the five frames. As shown in the bottom row of Table 2, all frames differ significantly from each other (p < .000)2. To analyse variation in the use of frames by quality papers from France, The ^ Netherlands, and Belgium, we conducted a multivariate analysis of variance (MANOVA) with the framing condition as a within-story factor and the quality newspaper as a between-story factor, as illustrated in Table 2. The Wilks Lambda of overall differences among groups was statistically significant (p < .000, F (15, 920) = 10.383, rp = .65) indicating that the use of news frames varies significantly between the quality papers under study. Furthermore, univariate between-subjects tests revealed that the use of each news frame, except for the power frame, significantly differs between the different quality papers3. To detect differences in the adoption of the five frames by the different newspapers, post-hoc pair wise tests with Games-Howell were conducted. 0 £ C > 00 -šT Table 2: Mean Scores of Cross-National Recurrence of Five News Frames in Four Quality Newspapers Newspaper Human interest Conflict Economic Power National Le Monde (n = 107) M SD .16a .21 .49a .25 .89a .18 .68 .31 .47a .36 De Volkskrant (n = 93) M SD .29b .26 .55 .24 .91a .15 .67 .31 .66b .39 De Standaard (n = 71) M SD .31b .24 .64 b .28 .86 .23 .70 .36 .26c .30 Le Soir (n = 70) M SD .16a .24 .46a .27 .79b .27 .67 .31 .17c .27 Total (n = 341) M SD .23v .25 .53w .26 .87x .21 .68y .32 .42z .39 Note. Adjustment for multiple comparison Games-Howell * Column values with different subscripts (a, b, c) were significantly different from each other at least at p < .05. ** Row values with different subscripts (v, w, x, y, z) were significantly different from each other at least at p < .000. We expect the conflict frame, economic consequences frame, and power frame to be more recurrent in quality papers EU coverage, than the human interest frame and nationalisation frame. When comparing the mean scores of recurrence of five news frames in the quality papers, in the bottom row of Table 2, we see that overall the four newspapers used the economic consequences frame significantly more (M = .87; p <.000) than the other frames under study, attributing financial and economic losses or gains to the introduction of the EU Constitution. The power frame (M = .68; p <.000), followed by the conflict frame (M = .53; p <.000), are the second and third most used frames. This means that in the portraying of the EU Constitution regular reference was made to the different stances member states took during the long-standing rounds of talks and negotiations leading up to the signing of the Treaty of Lisbon, focussing on the disagreements and tensions between the different actors, exercising power to make their national interests safe. As expected, we found the nationalisation frame (M = .42; p <.000) and human interest frame (M = .23; p <.000) to appear in a far lesser degree in quality papers' coverage of the EU Constitution. When the human interest frame did appear, it was mostly through photographs of smiling and hand-shaking politicians next to the newspaper article. Our hypothesis, namely, that the conflict frame, economic consequences frame, and power frame would be more used in quality papers' coverage of the EU Constitution is hereby confirmed. When interpreting these results however, one can determine a remarkable discrepancy between the economic consequences frame on the one hand, and the conflict and power frame on the other. We will come back on this in the conclusion and discussion section. Our research question asked whether there were differences between French, Dutch and Belgian quality papers in their use of frames when reporting on the EU Constitution. To analyse this, we must look at the column values of the four quality papers in Table 2. Post-hoc comparisons revealed that, when analysing the four quality papers separately, the economic consequences frame was the frame most adopted by all quality papers under study. De Volkskrant (M = .91; p < .05) and Le Monde (M = .89; p < .05) used this frame significantly more than Le Soir (M = .79; p < .05). The second most-used frame in the portrayal of the EU Constitution was the power frame. However, no significant differences were found in the adoption of this frame by the four newspapers meaning that all four newspapers made similar use of the power frame. Taking a look at the conflict frame post-hoc comparisons revealed significant differences in the adoption of this frame between De Standaard (M = .64; p <.05), and Le Monde (M = .49; p <.05), and De Standaard and Le Soir (M = .46; p <.05). This is the third most adopted frame by all four quality papers separately. Attribution of the nationalisation frame reflected large differences among the newspapers. De Volkskrant used this frame most frequently (M = .66; p <.05), Le Soir made the least use of it (M = .17; p <.05). Significant differences were found between all newspapers, except between De Standaard (M = .26) and Le Soir (M = .17), meaning that the two Belgian newspapers did not differ significantly in the use of a national touch in their articles on the Constitution. The human interest frame appeared in all quality papers to a minor degree. De Standaard (M = .31; p <.05) and De Volkskrant (M = .29; p <.05) made significantly more use of this frame than Le Monde (M = .16; p <.05) and Le Soir (M = .16; p <.05). Our results show that Le Monde can be perceived as a quality paper which, by using mostly the economic consequences frame, emphasises financial and economic consequences the introduction of the European Constitution brings with it. In addition, the power frame appears also frequently in the coverage, as is the case for all quality papers. Mutual relations between EU member states and division of power (for example the seat division of the European Commission) recur prominently in the coverage of Le Monde. Attribution to the nationalisation frame and conflict frame is made less often, and the human interest frame is almost non-existent in the coverage of Le Monde. The Dutch quality paper De Volkskrant differs from Le Monde in its frequent use of the nationalisation frame, which appears almost as often as the power frame. The conflict frame is also more prominent in the coverage than is the case for Le Monde. Hence, we can ascertain that in its portrayal of the EU Constitution De Volkskrant focuses in large part on the own nation, and relations and division of power between the member states, instead of using a more supranational angle, as found in the coverage of Le Monde. The two Belgian quality ^ papers, Le Soir and De Standaard, display a similar use of the nationalisation frame which is significantly less than Le Monde and De Volkskrant. However, De Standaard does differ significantly from Le Soir and Le Monde in its use of the conflict and human interest frame, two frames that by use of a conflict or emotional/personal angle try to grasp the attention of the reader. For both frames De Standaard has the highest scores in comparison with the other three quality papers. The two Belgian newspapers thus demonstrate clear differences in the use of frames when covering the EU Constitution. De Standaard shows more similarities with De Volkskrant, Le Soir tends more to Le Monde. Implications of and explanations for above findings U 0 £ C (6 il o LQ are discussed in the conclusion and discussion section. Despite the significant findings between the four quality papers, they all display a more or less similar pattern in their use of the five news frames. The economic consequences frame was the most utilised frame in the newspapers studied, followed by the power frame, the conflict frame, the nationalisation frame, and the human interest frame. Only De Volkskrant differed slightly from this pattern in that the nationalisation frame appeared almost as frequently as the power frame in this newspaper's reporting on the EU Constitution. Conclusions and Discussion The aim of this study was to analyse the recurrence of five news frames in the coverage of the European Constitution on the one hand, and to cross-nationally examine differences in the adoption of news frames of four quality papers: Le Monde (France), De Volkskrant (The Netherlands), De Standaard (Dutch speaking community of Belgium) and Le Soir (French speaking community of Belgium) on the other. To this end, we formulated one hypothesis and one exploratory research question. First, we expected, in line of previous research, the conflict frame, economic consequences frame, and power frame to be more frequently used in quality papers' coverage of the EU Constitution, than the human interest frame and nationalisation frame. Second, we wondered in what way differences in framing could be found between the French, Dutch and Belgian quality papers. The component analysis of the five recurrent news frames, for the first time examined in this combination, explained 66 percent of total variance in the 341 articles under study. Our results monitoring characteristics of format and content clarified that Le Monde devoted considerably more attention to the EU Constitution than Dutch and Belgian quality papers, a finding which is in accordance with the substantial higher visibility of the 2004 EP elections in France media outlets when compared to the Belgian and Dutch media (de Vreese et al. 2006). A possible explanation for this finding could lay in Le Monde's trademark as an outward, international focussed newspaper with foreign news covering a quarter of the total news coverage. De Volkskrant differed from the other newspapers by publishing significantly more opinion and analysis articles about the EU Constitution rather than news reports with background briefing, as was the case for the other newspapers. Our hypothesis was validated: the economic consequences frame, conflict frame and power frame appeared more frequently in the quality papers under study than the human interest frame and nationalisation frame, which we found back in a far lesser degree. These results confirm the indication based on in-depth interviews with EU journalists (Gleissner & de Vreese 2005) of a reversed tendency to nationalise EU related news in case of reporting about the EU Constitution and consequently the possible impact the symbolic meaning of unity, that goes behind the EU Constitution, could have on the way it is covered. In contrast with findings from previous research, the conflict frame appeared less often in the coverage of the EU Constitution. This is surprising since this is one of the most frequently found frames in political and economic news, especially in serious media outlets (Semetko & Valkenburg 2000), and one of the most dominant news selection criteria (Galtung & Ruge 1965). The symbolic meaning of the EU Constitution could be an explanation for the moderate use of the conflict frame. The economic consequences frame, mainly used in relation to financial and economic issues, such as the introduction of the Euro, recurred most prominently in the articles. This finding reinforces the generic character of the economic consequences frame as it confirms that this frame is also prominently used in relation to non-financial and economic news. Hence, this demonstrates that overall economic considerations could be one of the most important evaluation criteria when judging and reporting about the EU. It also shows that coverage was mostly focussed on content related aspects of the EU Constitution, than is the case for more attention drawing frames such as the human interest frame, conflict frame and nationalisation frame. The second part of our analysis, differences in framing between the four quality papers, leaded to some interesting findings. Le Monde and Le Soir on the one hand, and De Standaard and De Volkskrant on the other displayed on some points remarkable similarities. Le Monde and Le Soir were characterised by a minimal use of the human interest frame and conflict frame, in contrast with De Standaard and De Volkskrant. Hence, the latter made more use of attention drawing elements in the coverage such as emotions, personalisation, and the outlining of conflicts between member states and politicians. Apart from that, Le Soir and De Standaard showed a limited use of the nationalisation frame, compared to De Volkskrant, which applied this frame frequently in its coverage of the EU Constitution. This could be partly explained by the fact that the rejection of the EU Constitution (The Treaty Establishing a Constitution for Europe) in the Netherlands in 2005 still caused a lot of public debates during the renegotiation of its successor, the Treaty of Lisbon. This was less the case in Belgium and France. Our results show that, although significant differences were found in the amount of framing between the different quality papers, they all resemble more or less an identical pattern in the adoption of the news frames. The economic consequences frame was the most used news frame by all quality papers separately, followed by the power frame, conflict frame, nationalisation frame, and human interest frame. This illustrates that the meaning behind the European Constitution as a symbol of supranational unity could not only have led to a shift to a more unified portrayal of the EU in the news within the newspaper, but also between the newspapers we analysed. Since we analysed only one quality paper for each country (and two for Belgium), some caution is required when interpreting and generalising our findings. Whether these findings could also be translated to other quality papers' coverage of the EU could be subject of further research. As Semetko and Valkenburg (2000) pointed out, the prevalence of news frames in the news depends largely on the sensationalist or serious nature of the outlet. Consequently, it would be interesting not only to investigate serious media outlets but also to examine the portrayal of the EU Constitution in sensational media, thereby identifying possible differences ^ between these two types of media outlets. The political/ideological identity of a newspaper could also be connected to the frames that are used; an explanatory factor we didn't analyse in our study. Finally, the nationalisation frame we used in our study is just one first step in the construction of a measurement scale. Hence, further refinement and testing is required. The results of this cross-national study deliver evidence for the possible development of a European public sphere. During the same period we found in the selected quality papers Le Monde, De Volkskrant, Le Soir and De Standaard a similar use of frames in the coverage on the same topic: the EU Constitution (Treaty of LQ Lisbon). Coverage on the same topic, within the same period, and with the use of 0 £ C > identical frames, is considered a basic requirement for the development of a European public sphere (Brüggeman 2005). The importance of these findings lays in CO the meaning a European public sphere could have for the functioning and further ^ evolution of the European Union. Mass media as pillars of the public sphere which form the connection between citizens and the EU, play an influencing role in the knowledge acquisition and opinion-making about the EU on the one hand, and the political participation in the EU on the other. The cross-national recurrence of the same set of frames in our study illustrates the possibility of a launch of a supranational public sphere which, by the diverse use of frames, could also provide a pluriform coverage, necessary for the developing of a balanced and pluriform opinion on EU topics. Notes: 1. The Flash Eurobarometer survey was conducted from 13 to 15 June 2008, at the request of the European Commission Representation in Ireland. 2000 respondents, selected randomly, aged 18 and older, were interviewed by telephone (Eurobarometer 2008). 2. Conflict frame - human interest frame (t = 18.677, df = 340, p < .000), conflict frame - power frame (t = 9.145, df = 340, p < .000), conflict frame - economic consequences frame (t = 24.502, df = 340, p < .000), conflict frame - nationalisation frame (t = -4.656, df = 340, p < .000), human interest frame - power frame (t = 23.789, df = 340, p < .000), human interest frame - economic consequences frame (t = 41.019, df = 340, p < .000), human interest frame - nationalisation frame (t = 8.145, df = 340, p < .000), power frame - economic consequences frame (t = 11.668, df = 340, p < .000), power frame - nationalisation frame (t = -9.362, df = 340, p < .000), economic consequences frame - nationalisation frame (t = -21.582, df = 340, p < .000). 3. Power frame (F = .134; df = 3, 337; p < .940), economic consequences frame (F = 5.710, df = 3, 337; p < .001), conflict frame (F = 7.213, df = 3, 337; p < .000), nationalisation frame (F = 34.747, df = 3, 337; p < .000), human interest frame (F = 10.281, df = 3, 337; p < .000). References: Baetens, Marleen and Peter Burssens. 2005. Mass Media Coverage and the Reduction of the Legitimacy Deficit in the EU. The Belgian Case. Brüggemann, Michael. 2005. How the EU Constructs the European Public Sphere. Seven Strategies on Information Policy. Javnost - The Public 12, 2, 57-74. Cebuco. 2009. Oplagen Dagbladen. Centrum voor Informatie over de Media. 2009. Cijfers verklaardop erewoord. Curran, James. 1991. Rethinking the Media as a Public Sphere. In P. Dahlgren and C. Sparks (eds.), Communication and Citizenship: Journalism and the Public Sphere in the New Media Age, 27-57. London: Routledge. De Vreese, Claes H. 2003. Framing Europe: Television News and European Integration. Amsterdam: Aksant. De Vreese, Claes. H. and Hajo G. Boomgaarden. 2006a. Media Effects on Public Opinion About the Enlargement of the European Union. Journal of Common Market Studies 44, 2, 419-36. De Vreese, Claes H. and Hajo G. Boomgaarden. 2006b. News, Political Knowledge and Participation: The Differential Effects of News Media Exposure on Political Knowledge and Participation. Acta Politica 41, 317-341. De Vreese, Claes H., Jochen Peter, and Holli A. Semetko. 2001. Framing Politics at the Launch of the Euro: A Cross-National Comparative Study of Frames in the News. Political Communication 18, 2, 107-122. De Vreese, Claes H., Susan A. Banducci, Holli A. Semetko, and Hajo G. Boomgaarden. 2006. The News Coverage of the 2004 European Parliamentary Election Campaign in 25 Countries. European Union Studies 7, 4, 477-504. D'Haenens, Leen. 2005. Euro-Vision: The Portrayal of Europe in the Quality Press. Gazette: The International Journal for Communication Studies 67, 5, 419-440. Drieskens, Edith and Stefaan Fiers. 2005. Nieuws in de Europese Unie, de Europese Unie in het Nieuws? In M. Hooghe, K. De Swert and S. Walgrave (eds.), Nieuws op televisie: Televisiejournaals als venster op de wereld, 139-157. Leuven: Acco. Entman, Robert. 1993. Framing: Towards Clarification of a Fractured Paradigm. Journal of Communication 41, 4, 51-58. European Commission. 2008. Flash Eurobarometer 245: Post-Referendum Survey in Ireland: Analytic Report. Brussels: Directorate-General Communiation. European Commission. 2007. Standard Eurobarometer 67: Public Opinion in the European Union. Brussels: Directorate-General Communication. European Commission. 2006. The European Citizens and the Future of Europe: Qualitative Study. Brussels: Directorate-General Communication. European Council. 2001. Presidency Conclusions: European Council Meeting in Laeken 14 and 15 December2001. SN 300/1/01 REV 1, Brussels (14 December 2001). Galtung, John and Mari H. Ruge. 1965. The Structure of Foreign News. The Presentation of Congo, Cuba and Cyprus in Four Norwegian Newspapers. Journal of Peace Research 2, 1, 65-91. Gleissner, Martin and Claes H. de Vreese. 2005. News about the EU Constitution: Journalistic Challenges and Media Portrayal of the European Union Constitution. Journalism 6, 2, 221-242. Kinder, Donald R. 2007. Curmudgeonly Advice. Journal of Communication 57, 1, 155-162. Kleinstüber, Hans J. 2001. Habermas and the Public Sphere: From a German to a European Perspective. Javnost - The Public 8, 1, 95-108. Meyer, Christoph O. 1999. Political Legitimacy and the Invisibility of Politics: Exploring the European Union's Communication Deficit. Journal of Common Market Studies 37, 4, 617-639. Meyer, Christoph O. 2005. The Europeanization of Media Discourse: A study of Quality Press Coverage of Economic Policy Co-ordination since Amsterdam. Journal of Cultural Media Studies 43, 1, 121-148. Office de Justification de la Diffusion. 2009. Quotidiens Nationaux. Peter, Jochen, Edmund Lauf, and Holli A. Semetko. 2004. Television Coverage of the 1999 European Parliamentary Elections. Political Communication 21, 4, 415-433. Peter, Jochen, Holli A. Semetko and Claas H. de Vreese. 2003. EU Politics on Television News. A Cross- National Comparative Study. European Union Politics 4, 3, 205-327. Risse, Thomas and Marianne van de Steeg. 2003. An Emerging Public Sphere? Empirical Evidence and Theoretical Clarifications. Paper presented at the conference "Europeanization of Public Spheres, Political Mobilisation, Public Communication and the European Union," Berlin, 20 June. Schlesinger, Philip. 1999. Changing Spaces of Political Communication: The Case of the European Union. Political Communication 16, 3, 263-279. Semetko, Holli. A. and Patti M. Valkenburg. 2000. Framing European Politics: A Content Analysis of Press and Television News. Journal of Communication 50, 2, 93-109. Semetko, Holli A., Claes H. de Vreese, and Jochen Peter. 2000. Europeanised Politics - Europeanised Media? European Integration and Political Communication. West European Politics 23, 8, 121-141. Trenz, Hans-Jörg. 2004. Media Coverage on European Governance: Exploring the European Public Sphere in National Quality Newspapers. European Journal of Communication 19, 3, 291-319. Valkenburg, Patti M., Holli A. Semetko, and Claes H. de Vreese. 1999. The Effects of News Frames on Readers' Thoughts and Recall. Communication Research 26, 5, 550-569. Van der Schoot, A. J. 2002. Europe's Identity: A research Project on the Portrayal of European Identity in Television News and Current Affairs Programs in the Netherlands, Germany, France and Great Britain. Master's degree thesis, University of Nijmegen. Ward, David. 2001. The Democratic Deficit and European Union Communication Policy. An Evaluation of the Commission's Approach to Broadcasting. Javnost - The Public 8, 1, 75-94. Ward, David. 2004. The European Union Democratic Deficit and the Public Sphere. An Evaluation of EU Media Policy. Amsterdam: IOS Press. Wimmer, Jeffrey. 2005. Counter-Public Spheres and the Revival of the European Public Sphere. Javnost - The Public 12, 2, 93-110. U S i 0 C (6 LO