Textual-intertextual analysis

Danish Council for Foreign Relations in Copenhagen.³ Figure 4 provides an overview of the temporal stages that are studied and the eight texts’ placement within them.

deficiency I propose to conduct a textual-intertextual analysis of the kind suggested by the rhetorical scholar Leah Ceccarelli. In a study, the purpose of which is similar to mine although it deals with a different type of meaning formation, Ceccarelli proposes that the close analysis of a text should be supplemented with evidence of the actual responses with which the text was met.

Ceccarelli’s contention is that it is possible to explain how texts work by exploring the connection between rhetorical strategies and their effects on historical audiences – that is, the relationship between texts and their intertextually established contexts (Ceccarelli, 2001, p. 6).

In my version of the textual-intertextual analysis there are eight texts, namely the two declarations and the six speeches, and the intertextual responses to these texts consist of articles and comments published in newspapers stemming from the five national contexts that are also

represented by the speakers.⁷ I have chosen to focus on the responses made in newspapers in order to include the broader level of public opinion formation in each of the national contexts.

Newspapers are often hailed as being of vital importance to the nation-builders of yore (Anderson, 1991, p. 46), and although the printed press has been experiencing some decline as a result of competition from the electronic media it still plays a central role as national arena for public opinion formation. Moreover, the production of news is today so standardised that most media follow more or less the same agenda; the coverage of one mainstream medium may in many respects be

considered representative of the media’s coverage as such. In the case of media coverage of political discussions the newspaper is arguably the most comprehensive medium: the papers’

traditional affiliation with specific political groups ensures that they take an interest in political issues and allows for some divergence between them. Although most newspapers have today cut their direct connections with political parties and as a matter of course adhere to the same news criteria as all other media, the papers remain the most politically involved of the news media. Many still overtly recognise political leanings, at least on their leader pages, and most devote more

attention to political developments and discussions than the other media do.

There is also a purely instrumental reason for choosing the newspapers as the source of intertextual responses to the eight texts: the existence of searchable electronic archives makes the newspapers’ coverage of the two declarations and six speeches readily available and allows the

7 In order to create total complementarity between texts and contexts, I should also have included responses from European news services. Such services exist – most notable is the European Voice that has an actual paper version, but internet services such as Agence Europe, EUObserver, and Euractive should also be mentioned. These services provide an inherently European level of mediation, but are mostly read by people working in or in close contact with the EU’s institutions and do not reach a broader public. Although the slight dispersion of the European news services justifies their exclusion somewhat, the main reason for my omission of them is, unfortunately, one of lack of access. Articles from these sources that dated back to May 2000 simply were not available.

generation of compatible sets of data. I have employed two such archives or databases, namely InfoMedia in the case of the Danish context and LexisNexis for the other four.⁸ From the totality of available sources 17 newspapers were selected: three from each of the French, German, and Spanish contexts and four from Denmark and England⁹ (see table 1).

Country Denmark England France Germany Spain

News-papers

Berlingske Tidende (BT), Information (Inf),

Jyllands-Posten (JP),

Politiken (Pol)

Financial Times (FT), The Guardian (Guar), The Independent (Ind), The Times (Tim)

Le Figaro (LF), Le Monde (LM), Libération (Lib)

Frankfurter Allgemeine Zeitung (FAZ),

Süddeutsche Zeitung (SZ), taz, die

tageszeitung (taz)

Cinco Días (CD), El País (EP), Expansión (Exp)

Context¹⁰ BT, JP, and Pol are Denmark’s three major national morning papers.

Inf is a smaller, intellectual paper included for its full coverage of EU-matters.

Guar, Ind, and Tim are major national dailies.

The economic newspaper FT is included for its European scope. The absence of tabloids is especially problematic in the English case as the tabloids hold a large share of the market and are very outspoken on European issues.

The French press is particularly politicised; Lib is socialist in inclination, LM reveres an ideal of balanced, in-depth coverage, but remains attached to left-of-centre goals, and LF has Gaullist

sympathies

The German press is extremely

regionalised. FAZ, SZ, and taz are among the few nationally read newspapers, but having national readerships does not rule out regional attachments as FAZ and SZ immediately reveal in their titles.

The Spanish press is also regionalised, but three national papers exist, EP being the largest.

Unfortunately, the other two (ABC and El Mundo) were not available in the LexisNexis database. CD and Exp are economic newspapers of the FT type.

Table 1: Presentation of the 17 newspapers included in the study

All the chosen newspapers belong to the so-called serious press; there are no tabloids among them and thus ‘popular’ opinion formation in the somewhat derogatory sense of the word is not represented. The lack of tabloid newspapers is, at least in some national contexts, a major setback for my study’s claim of representing mainstream opinion formation. However, I have prioritised cross-national compatibility over national representability, and since I did not have

8 Both are electronic databases available to subscribers; I have accessed them through the library of the Copenhagen Business School: http://www.cbs.dk/library.

9 Although the chosen newspapers are published in all of the UK they all have their base in London and do not necessarily represent Scotch, Welsh or other regional views. In order to avoid the risk of being accused of ignoring regional differences I term the newspapers and the viewpoints they represent English rather than British. However, the newspapers themselves do not seem to make this distinction, and I shall be referring to British views when the coverage does so, but not in my own characterisation of the coverage. A similar argument about the partiality of the represented nationality could be made in the case of the Spanish newspapers. These newspapers do not represent Spanish regional identities such as the Basque or the Catalan, but a term for Spain and the Spanish national identity in the restricted sense that could parallel England and English in the British case does not exist (at least not in the English language).

Therefore, all I can do is notice that the Spanish national context presented here is a majority position from which significant regional minorities actively delineate themselves.

10 The information about the national contexts is obtained from the European Journalism Centre’s website (www.ejc.nl), Kuhn (1995), and Seymour-Ure (1996).

access to newspapers of the more popular type from all countries I have chosen to disregard them altogether. The exclusion of popular media of both the printed and the electronic kind can also be justified on the basis of the serious newspapers’ special status as both strictly national media and as the media with the most thorough coverage of political issues – European and otherwise.¹¹

The papers are in a sense both the least and the most likely arenas for supranational opinion formation to emerge and therefore constitute a particularly interesting field of study. The papers offer one among several possible representations of the national publics – surely their image is a biased one, but just as surely it is influential. My study will not discuss internal differences of the various national contexts; the purpose of the investigation is to compare and contrast the different national versions of the debate. Hence, I will only consider features that are typical of the national representations and enactments of the European debate. The direction towards the

interrelations of the contexts is caused by my interest in European opinion formation, but it is also an attempt to cope with the vastness of each national context. Rather than claiming in-depth

knowledge of each of the five national contexts, I explicitly and exclusively focus on the contextual information that appears in the selected intertextual material. I explain the characteristics of the national contexts as these emerge from the transnational comparison rather than as a result of studies of each context in isolation. My study is textual and comparative and finds its limitation, but also its main justification in these two traits. At the national level, then, differences will be

smoothed over in order to establish a general impression and understanding of emerging

commonalities and remaining differences of the European debate as it is articulated in the national contexts.¹²

Having selected the 17 newspapers whose coverage of the two declarations and six speeches is to provide the intertextual dimension of the textual-intertextual analysis, I set specific temporal boundaries around the surveys for intertextual references. In order to facilitate searches that are both precise and exhaustive, the surveyed periods for each studied moment and turning point were limited to ten days (see table 2). When the period is limited to ten days one can set rather

11 Nevertheless, the socio-demographic leanings of my data should be recognised. The opinion formation processes that the newspapers reflect and participate in primarily include the intellectual and political elites who beforehand can be estimated to be the most likely to care about European issues and most likely to engage in European debate. This puts restrictions on what I can claim to be studying; I may point out the emergence of common European modes of discussing EU matters, but cannot ascertain the depth of such Europeanisation. However, the data does allow me to draw general conclusions on the limits of the convergence, the point being that if those most likely to discuss the EU in European terms do not do so, then there is little likelihood that such discussions are emerging elsewhere.

12 This disregard for admittedly salient features in the name of cross-national compatibility also extends to the issue of party politics. I shall not seek to explain the speakers’ proposals in terms of party affiliations, but only as they relate to and seek to establish specific national and common European positions.

precise search terms: the speaker’s name and “speech” in the case of the six moments and the name of the city in which the Council meeting was held and “declaration” for the two turning points. The results include all the relevant articles published within the ten days, and the sets of data are not unmanageably large. See appendix 9 for a list of all the articles of the eight ten-day samples; each sample will be presented in detail in the course of the textual-intertextual analysis.

Turning point/moment Survey period

Fischer’s speech, 12^th of May 2000 10/05/00 – 19/05/00 Aznar’s speech, 26^th of September 2000 24/09/00 – 03/10/00 Blair’s speech, 6^th of October 2000 04/10/00 –13/10/00 The Nice Declaration, 7^th-9^th of December 2000 05/12/00 – 14/12/00 Jospin’s speech, 28^th of May 2001 26/05/01 – 04/06/01 Prodi’s speech, 29^th of May 2001 27/05/01 – 05/06/01 Lykketoft’s speech, 23^rd of August 2001 21/08/01 – 30/08/01 The Laeken Declaration, 14^th-15^th of December 2001 12/12/01 – 21/12/01

Table 2: The turning points/moments and their corresponding survey periods