Understanding the Off-the-recOrd as a sOcial Practice : german Press-POlitics relatiOns seen frOm france

In this article I aim to understand the structuring effects of the institutional and social context on journalist-politician interactions in Germany. In particular, this article focuses on different informal relations within political information circles and, more specifically, on the “off-the-record” practice. Bringing in some exploratory comparative results from France, I propose a sociological model of analysis of these practices. It requires observing and analyzing the complex interplay between political communication and journalism practices in political institutions. At the same time, these interactions have to be understood through their long-term transformations. The traditional explicative variable used by scholars—the effect of a national democratic culture as more or less respectful of journalists’ independence—has to be deconstructed. I propose here a historical sociology of political communication in Western democracies. My study focuses on the (West) German case, understanding the structure of the interactions in their continuity (from the Weimar Republic to the postwar Bundesrepublik). This example of German journalism is probably a limited case.

This project grew from a sociologist's enigma. 1 There is one image that is strongly anchored in savvy common sense: in Germany closeness and complicity between journalists and politicians do not hold sway.In all national (Altmeppen and Löffelholz 1998;Weischenberg, Malik, and Scholl 2006;Kepplinger 2011) or comparative international studies (Köcher 1989;Hanitzsch and Berganza 2012), German journal-ists are those who claim to be the least subject to competition and commercialization constraints, and their professional beliefs are often permeated by a strong critical stance.These ideal representations of the profession are corroborated by practical reality: political interviews are often very tense (Lemieux 2000:222-244).The virulent altercation between the former Chancellor Gerhard Schröder and journalists on the night of the 2005 election is often cited as an example of this (Wegmann and Mehnert 2006).German television is presented as having a "structural independence" (Bourgeois 2009).When it comes to political news, German journalists possess an important tool to regulate the production of their own information: the Bundespressekonferenz (BPK).Created in 1949 in the West German Federal Republic, it brought together in Bonn and then in Berlin (since 1999) the full spectrum of parliamentary journalists, together with the association of foreign journalists (Krüger 2005).The BPK is a unique institution that is run entirely by journalists.The spokesperson for the German government comes to explain the government's policies to journalists three times per week.As a symbol of their independence, these conferences are opened, moderated, and closed by a member of the BPK management committee without any intervention on the part of the government's spokesperson.Thus, we would be in the presence of a deeply "democratic journalistic culture" (Bourgeois 2009:39), where both sides (journalists and politicians) are apparently hermetically autonomous.
A previous comparison of German and French academic traditions (Hubé 2005) showed us that the methods and traditions of analyzing the media ultimately produce results that are hard to compare.In France observation and microsociological description of practice take precedence over any quantitative and more openly normative analysis of democratic practices in Germany.Journalists are given questionnaires concerning abstract cases.Where a researcher departs from this analysis of the perceptions of a journalist's role and holds in-depth interviews, the picture is much more varied (Meyen and Riesmeyer 2009).Some evidence makes no sense at all if these ethical principles apply.How would it be possible to explain the November 2003 call by the main press entities to no longer second-read their interviews if no political influence has ever been exercised?Likewise, how to explain that during an interview with the deputy director of a conservative newspaper, we were interrupted by a close friend who happened to be the former president of the West German Federal Republic?Furthermore, how is it that the former anchor of the second public channel (ZDF) was appointed spokesperson for Angela Merkel's government in July 2010, and his predecessor was nominated to be the administrator of the public radio and television channel in the region of Bavaria?Most recently the Federal Constitutional Court in Karlsruhe (Bundesverfassungsgericht) issued a judgement in March 2014 criticizing the overly political nature of appointments to the board of people for parties and ministries between 2003 and 2015, and observations of BPK meetings in 2010 and 2015.Its analysis also uses the archives of the Bundespresseamt (Bundesarchiv Koblenz) 1949-1985, the Vereinigte Presseabteilung der Reichsregierung (Bundesarchiv Berlin) 1918-1933, of the Bundespressekonferenz (Berlin) 1949-1985, of the Verein der ausländischen Presse 1970-1975, and an analysis of the administrative documents filed in the documentation department of the Bundespresseamt.ZDF, which distorts the distribution of editorial jobs, systematically arranged as a "ticket" comprising a journalist close to the government party and a deputy from the opposition party (Robert 2016).Paradoxically, the list of such situations is quite long.With a critical perspective on her own work, the journalist and former president of the BKP Tissy Bruns (2007) has reminded her colleagues about the norms of independence that used to characterize relations between the two groups.She insists on numerous informal circles and "off-the-record" conversations as opposed to the principles of journalistic practice.By explicitly referring to these same circles, Nikolaus Wegmann and Ute Mehnert have underlined that "beyond personal acquaintances, there exists a structural connection between politics and the media" (2006:148-149).
Whether it is due to the strong institutionalization of interactions or the apparent absence of competitive struggles, as well as the relative issue and nonconflictual centered media coverage of politics (Esser, Reinemann, and Fan 2000;Hubé 2008), this example of German journalism is probably a limited case.I hypothesize that its explanation is to be found in the specificity of the configuration of German politics.The social weight applied by the control of actors (journalists and politicians) taken by strong relations of interdependence, in a rather limited place (the arena of federal politics) as well as in a more weekly competitive journalistic space, leads to a stronger harmonization of regulatory practices for these interactions.Insisting on the German case is important because few other studies and publications have focused on this work of legitimation (Mergel 2002).To the contrary, the French case has already been the object of historical investigations on state communication both before (Georgakakis 2004) and after (Ollivier-Yaniv 2000) World War II, as well as on the transformations of journalism (Delporte 1999;Lévêque 2000;Kaciaf 2013).If the French case is well known to us, it must be reanalyzed in its more institutional dimensions based on the results found in the German case (Hubé 2013(Hubé , 2016)).

Some background information on german journaliSm
After 12 years of control and "Aryanization" of the German press, the postwar press defined itself in 1945 in terms of a fear of the state, which strongly shaped professional representations.Emancipation occurred in three stages: control by Allied troops; creation of the Federal Republic of Germany in 1949; and a jurisprudence that strengthened this "independence."The German press complies with a "democratic-corporatist" model (Hallin and Mancini 2004) in which the political parties and civil society (associations, trade unions) are given the role of monitoring the state and media and jointly managing their regulation.For example, the journalists' associations allow a journalist who becomes a press officer to keep his/her press card, because that journalist is continuing to publicize information.The Western Allies occupying Germany (1945)(1946)(1947)(1948)(1949)) wanted a broadcasting system that was publicly owned and regionally organized, separate from both the market and the central state.Control was to be exercised not by the state but by "socially relevant groups" (gesellschaftlich relevante Gruppen) (churches, associations, trade unions).Content was produced by regional broadcasting bodies (Rundfunkanstalten) federated from 1950 as the ARD, which has been the name of the first public channel since 1952.The regional bodies represent a Land or a wider region and produce programs broadcast by a local channel, known as the third channel.The second television channel ZDF (Zweites Deutsches Fernsehen) was set up in 1963 on the model of a single centralized channel also dependent on the Länder.Since the late 1980s the broadcasting scene has been dual, with private broadcasting-national and funded by advertising-and highly regionalized public broadcasting with little funding from advertising.Radio is also highly regionalized.On the same line, the daily press is more regionalized than in other countries.News dailies are more supraregional than national.One of the main political dailies, the Süddeutsche Zeitung, has a largely Bavarian readership.Even Bild-the best-selling tabloid in Europe with 2.35 million copies and 9.96 million readers in 2016, according to the German Audit Bureau of Circulation2 -also publishes local pages.The daily press is also sold by subscription and is funded by advertising.Although the media do belong to major groups (Springer, Bertelsmann, etc.), these groups' businesses are almost exclusively media and publishing companies.In terms of political control, the Press Council (Presserat) was set up in 1956 to resist interference from the government, headed then by Konrad Adenauer.When in 1960 that government sought to create a second television channel, the Länder and the Social Democratic Party of Germany (SPD) took the matter to the Constitutional Court in Karlsruhe, which defends the principles of Staatsfreiheit (freedom from the state) and Staatsferne (distance from the state).The Bundespressekonferenz (BPK), which I address in greater detail, has since 1949 been attended by the political journalists of all the German media present in the capital.In 2013 it had 939 members.
to w a r d S a H i S to r i c a l S o c i o lo g y o f P o l i t i c a l l e g i t i m at i o n In Berlin, like everywhere else, professional politicians and journalists meet and spend time with each other.This type of interdependence, where one needs the other to exchange information against publicity, is consubstantial with a social space of practices that is strongly heteronomous-what is commonly referred to as the public sphere (Bourdieu 2005;Schudson 2005).One of the central issues for politicians has to do with the access to the market of symbolic political goods that they contend for with journalists.The means of access to this public sphere are structured very differently depending on the social and political configuration.In other words, the approach that I am proposing brings us out of a media-centrism that is widespread in the sociology of journalism and communication (Schlesinger 1990) to focus particularly on the coproduction of political discourses (Davis 2007).More generally, it brings us to consider political communication as an enterprise (amongst others) of political legitimation, supported by actors more or less professionalized "in communication," discussed and contested by different actors interacting with each other, where the media is not only a "mirror" of current discussions but an actor in those policies (Benson 2013).

journaliStS and PoliticianS aS rival-aSSociateS
In France, as in Germany and Russia, journalists and politicians are rival-associates (Neveu 1989;Legavre 2011), who participate together in the work of symbolic political production.As associates, they participate in highlighting politics, its rules and frameworks, as well as the issues at stake on the political agenda.As rivals, they follow divergent interests and expectations: politicians seek to ensure a strong visibility, a positive image, and an advantageous framework; journalists seek to maintain independence within this framework, including a critical mind-set right up to the possibility of leading an investigation.Relying on a dance metaphor, many works have helped to highlight how these two groups are involved in a tango, a rumba when it is not a danse macabre (e.g., Strömback and Nord 2006;Ross 2010).These works focus on such relations in their strategic context, by seeking to understand the power games (Strömback and Nord 2006), the calculations and anticipations of the strikes against each other (Bernier 2000;Cohen, Tsfati, and Sheafer 2008), as well as the representations and expectations from one group on the other (Charron 1994;Ross 2010).These works try to understand how actors establish and maintain distance from their sources and seek to comprehend the framework of these interactions.These investigations focus on different locations in France and in different countries: lobbies and corridors within parliaments (Tunstall 1970;Dunwoody and Shields 1986;Charron 1994;Bernier 2000;Lemieux 2000;Burgert 2010), within the European Commission (Bastin 2002;Baisnée 2007), during the follow-up to political campaigns (Crouse 2003;Kaciaf 2014), or on the military terrain (Gatien 2009), where confidences can be heard.This helps to illustrate that the interactions between journalists and politicians are neither random nor only due to a professional ethic, but are part of a permanent negotiation on the contours of what can be said, done, and published (Legavre 1992(Legavre , 2014;;Perloff 1998;Davis 2007).More recently, these works have sought to comprehend the role played by press agents and/or communications officers in the emergence of a new public relations democracy (Davis 2002;Juhem and Sedel 2016).In addition, the permanent tension involved in these exchanges is also linked in many ways to the public debates that give them structure, as they focus on the normative association of independence and manipulation, as well as their corollaries, a government by/with the media, and/or a lowering of the quality of journalism (Cook 1998;Lemieux 2000;Bennett 2011).
This outlook on these interactions has the merit of removing journalism from its media-centrism by taking the work done by journalists and their sources on an equal basis.All seek to understand the processes involved in highlighting the visibility of the media (Kaciaf and Nollet 2013;Desrumaux and Nollet 2014).They are often case studies decontextualized from historical anchoring or institutional structures.They slightly deconstruct a perception of universalizing practice (Hanitzsch 2007).Although it is true that communications instruments and the quest for publicity play an important role for the exercise of power in representative democracies (Manin 1997), this practice is not identical in all places and in all historical configurations.Moreover, in each historical era the forms of political struggle are intimately linked to the types of media available at the time, and conversely, the development of the media is very much dependent on the evolution of regimes, legislation, and, more generally, on the social balance of power.

examining tHe Structuration of national Public SPHereS
Understanding the work of legitimizing politics on national public spaces relies on analyzing and grasping the structural transformations in these public spheres (Habermas 1989;Bourdieu 2014).The state, like journalism, is an institution that is a social construct, the end result of a long process.Examining the exchanges between press and politics means seizing the coproduction, the interdependence, and the rationalization of the political work oriented towards the media as a relational arrangement between (at least) two groups of actors.The ways of accessing the public sphere are the result of a historical process.It is important to broaden the framework of observation to the whole chain of transactions, from mediations to intermediations, that today contribute to, interfere with, facilitate, or thwart the construction of political meanings.As suggested by Daniel Hallin andPaolo Mancini (2004, 2012), understanding the functioning of media systems in a social space entails analyzing and describing the structure of the media's economic markets, as well as the historic emergence of a journalistic profession, while at the same time insisting on more political criteria such as the role of the state and what they refer to as "political parallelism" between the media and politics.For these authors one of these four variables (at least) cannot be understood in isolation from the three others.The rise of an autonomous journalistic profession is not given in abstracto but is the result of a process made up of ruptures and struggles between agents within the state, of transformations from capitalism, and so on (Chupin, Hubé, and Kaciaf 2012;Roudakova 2012).This is why such variables of the media system must be added with the characteristics of the political system.The form of government, for example, "must be understood as a structuring element-and not as an isolated dimension of social life" (Schudson 1994:532) of political interactions within the public sphere.It is important not only to understand these interactions at the individual level of actors, but to analyze them at the "meso level" of the institution or rather the "inter-institutional level" (Benson 2004:280; see also Schlesinger 1990;Benson 2013) between the two semiautonomous fields of political journalism and politics.The sociology of journalism must be analyzed through the lens of the sociology of the state (Schudson 1994;Benson 2004Benson , 2013)).Understanding these relations in their historicity makes it possible to open the black box of the coproduction of media goods by going beyond, on the one hand, a reifying reading in terms of the (in)dependence of journalistic power and, on the other hand, in terms of power of journalism.
underStanding germany from france (and vice verSa) When examining the practice of off-the-record and spin doctoring in Germany (Esser et al. 2000), one can only be surprised by the differences from France and Britain.Close relations occur in Germany within a highly institutionalized framework.Obviously, there as elsewhere, there are rules (Legavre 1992(Legavre , 2014;;Charron 1994;Bernier 2000;Crouse 2003;Gatien 2009).In addition, it is often the case that confidential information is given in order not to remain confidential too long.The whole "art" of journalists and politicians is to play with this rule.On both sides of the Rhine, as Jean-Baptiste Legavre (1992) describes so well, off-the-record (unter drei in German) is indeed a tool of mutual recognition and coordination for the two groups of associates/rivals.The proper use of these rules is a sign that one belongs to the political microcosm.The difficulty of this tool is that it requires and assumes the strategic ability to know whether or not it is possible to use information obtained in this way, and what "period of immunity" may elapse before that information can be used.The off-the-record tool also imposes social constraints: an obligation for politicians to provide confidential information in return for good relations with journalists and, on the other hand, an obligation for journalists to receive this information by agreeing to attend press conferences, press announcements, lunches and breakfasts; and it raises the issue of the countergift (rewarding one's source by interviewing them or agreeing not to pass on the information straightaway).Journalists do an apprenticeship as part of their professional socialization.Any slips they make are always a source of anecdotes among colleagues.The situations in France and Germany seem to be set apart from international experience and yet quite different from each other.In France off-the-record operates rather vaguely, with no strong codification of places and times of interaction.There are indeed rules defined in practice and by practice and renegotiated in the interactions (Legavre 1992).Conversely, in Germany one can only be struck by the extreme codification and proceduralization of off-therecord, which is far from common in Western journalism (Legavre 2014).
I intend to examine the difference between the two countries by highlighting what appears to be most obvious in the comparison.The highly formalized separation between confidential information and official information is based on the institutionalization and codification of the procedures of exchange.From an international and comparative perspective (Legavre 2014), the most surprising is that they respected to this extent.The question that arises from all this is the following: how can we explain that, in the end, collective profits from calculability and predictability linked to codification prevail without discussion over interests (Bourdieu 1990), including the individual interests of journalists and political actors to break the rules-in the name of the right to information for citizens, of the economic competition between newspapers, or of the quest for mediatization at all cost from politicians?
The German case presents a second particularity.In 1999 Germany's federal institutions moved from Bonn to Berlin at the same time that the Green-Social Democratic coalition was ascending to power, with the reputation for being very skilled in communication.This move brought with it the massive arrival of new journalists.The interactions between the different groups found themselves temporarily marked and seemed to prove right the reading of a new media republic (Bruns 2007).The temptation existed for all sides to redefine the rules, as I have elaborated on in an article on the effects of political alternating on the economy of symbolic goods (Hubé 2016).
Note that the division of Germany (especially the building of the Wall in 1961) and reunification in 1989-1990 were also times of political upheaval.However, they did not have a structuring effect on the politics-media arrangements.The rules of BPK membership were, admittedly, changed in 1961 to exclude all journalists working for any publication considered not to "respect the constitution."3But these East German journalists kept their accreditation to the Bundestag.From 1972, under Chancellor Willy Brandt's Ostpolitik, BPK journalists and foreign correspondents supported the East German correspondents' request to join the Foreign Press Association (Verein der Ausländischen Presse, VAP), which shares building headquarters and meetings with the BPK.From the start of 1974 they were full members of the VAP. 4 Their work was de facto integrated with that of all other correspondents.These journalists were socialized in the political-journalistic community of Bonn and the BPK.After reunification the number of BPK members rose (from 506 in 1990 to 576 in 1991), but the change does not mean that the new members came from the East. 5The German newspaper market did indeed change after 1990, but not in a way that directly influenced the configuration under study here.r e S P e c t fo r o r d e r-a g e r m a n c H a r a c t e r i S t i c ? a n a ly z i n g P r e S S -P o l i t i c S r e l at i o n S w i t H i n a S o c i o lo g y o f i n S t i t u t i o n S Relations between journalists and their sources occur within a set of practical constraints and rules to be obeyed.A facile culturalist explanation would be that respect for order is fundamentally German, which is hardly satisfactory.Rather than consider a practice as belonging to a "culture," it should be analyzed as the product of the internalization of essential struggles within a social configuration and of the structural mechanisms regulating these struggles, and the subjectivization by actors of these rules as an objective element (Bourdieu 1995:343), in this case to preserve journalists' independence from politicians.Three explanatory factors (at least) may be advanced for this structuring of the field: the history of states since 1945, the configuration of the political-journalist sphere, and the rather noncompetitive economic structure of the media.
germany ye ar zero, or How do we live toge tHer?
After 1949 the Constitutional Court's work of defining a foundation of basic rights had the twofold consequence of "relativizing the state" in the organization of collective life and the welfare state, and of underpinning the state as the guarantor of that life (Le Gloannec 2001).This "democratic corporatist" model (Hallin and Mancini 2004) allocated a monitoring role to political parties and incorporated bodies (associations, trade unions).Consequently a foundation of jurisprudence and collective beliefs was built within relations involving great tensions and was then institution-alized in the modus vivendi we see today.The years when the Federal Republic of Germany was being created, and more particularly the Adenauer era, were years of high tension and intense conflict between journalists and governments.This "finding a path towards the modernization" of relations, in a more democratic direction, occurred via these conflicts (Krüger 2005:12).From the first Federal Constitutional Court's Lüth decision in 1958 to its decisions in 1990, including the Spiegel affair6 in 1962 (decided in 1966) (Doerry and Janssen 2013), in every case the judges defended the exercise of journalism in a manner highly respectful of their freedom of information and comment (Hubé 2008:37-48).The BPK has consistently intervened to defend journalists as a group and not as individuals.This defense was not always obvious and was constructed a posteriori after each crisis by the proclamation of strict rules demarcating fields of action.
The BPK was set up by political correspondents who had lived through the Weimar Republic and its fall.Weimar was characterized by an extreme politicization and polarization of publications engaged in fighting for their particular worldview (Weltanschauung) (Fulda 2009:223).In Berlin at that time there were already places of socialization run by both journalists and the government (Krüger 2005).But even meeting nearly every day did not manage to calm a situation that was still in the shadow of the post-Second World War journalists and government spokespeople.Moreover, they attempted to "correct" the errors of the earlier period in order to avoid arriving at a situation that "as in the time of the Weimar Republic, could lead to the formation of public opinion being monopolized by a [press agency] enjoying irregular economic advantages,"7 for example.The BPK later emerged as a compromise between journalists' desire to make up for their inadequate numbers and lowly position in their news organizations (wages below average) and Adenauer and the Allies' desire to have structures to publicly communicate parliamentary activities.The federal government "delegated" to the BPK the management of accreditations (Krüger 2005:39).The BPK also accepted the financial arrangements made by the city of Bonn and the government to help with correspondents' accommodation, on the condition that these financial supports were made on general principles, while at the same time organizing the Bundespresseball (federal press ball), the annual social event for politicians and the media, to raise money.At the same time, to protect the group, the BPK accepted the Kanzlertee (chancellor's teatime), on condition that Adenauer invited a variety of journalists; the BPK did not ask journalists to keep quiet about their party affiliations, because it would in this way display its pluralism and democratic openness.It also accepted former journalists who had become public relations officers for political groups and refused to exclude those who were more obviously close to the government.It brought pressure to bear on government politicians and vilified any official spokesmen who refused this politicization or were too selective of questioners.The Spiegel affair in 1962, 13 years after the birth of West Germany and a rather long period of tension and conflict between the two groups, was the turning point at which the BPK's modus operandi became institutionalized and the necessity to defend the constitutional principle of freedom of the press crystallized (Doerry and Janssen 2013).By adopting a common position journalists formalized the representation of their professional role as missionary and critical and no longer solely as supporters and responsible builders of an emerging democracy (Krüger 2005:92).These moments of crisis enabled the group to define itself anew around this duality: in public the two groups were clearly separate; in private nothing prevented political affinity, as long as it did not spill over into the first framework, in other words, took the form of comment and not information.

tHe (reStricted) configuration of excHangeS
What made this equilibrium of exchange possible and sustainable was also the restricted number of actors in the federal capital.This reluctance to speak about backstage relations was due to the coziness typical of Bonn and then transferred to Berlin.Journalists' knowledge of and closeness to the political game enabled them to decipher events without having to reveal backstage information and thus expose politicians to competition (Esser et al. 2000).
Being a parliamentary journalist is a prestigious position in an editorial team, but at a distance from the head office.These journalists have a privileged status close to that of European journalists based in Brussels (Bastin 2002;Baisnée 2007). 8he head of the Berlin office is hierarchically equal to the head of the Politik section.The Berlin offices focus on the daily production of political news in the most conventional sense."In the main, we work on politics.The other stories, we do them only, let's say, on the side when we feel like it." 9The division of tasks within these offices corresponds above all to a highly legitimist division of politics: each political party in the Bundestag (Die Linke, SPD, Grünen, FDP [until 2013], and CDU/CSU) must be covered by a journalist.At least four persons are essential to provide comprehensive coverage of parliamentary activities, and the offices rarely contain more than ten journalists.The second division of tasks is ministerial.The political journalists get the great offices of state (justice, home and foreign affairs, government) and social ones (education); the economic journalists have the ministries of the economy, finance, and agriculture.On the basis of our figures compiled from the 2010 Bundespressekonferenz members' directory, there were 73 freelance journalists (not at-tached to a publication) and 908 journalists working for 270 editorial offices. 10On average an editorial office had 3.2 members, but variation is high, ranging from 23 working for the Spiegel and the ZDF to 128 offices with a single journalist.Only 21 editorial offices (7 percent) have more than ten members, most of them international press agencies (4), public radio and television channels (8), and magazines.This small number of journalists per outlet works within a restricted set of actors, where their value for politicians is "to be able to build close contacts with a small number of journalists and engage in a controlled exchange of information" (Grunden 2009:291).
This particular configuration is also valid for elected officials.At the federal level, like at the level of the federal states, access to a political mandate goes through a list system for half of all elected officials-which makes them dependent on the parties-in a context where holding several mandates concurrently is relatively rare for politicians (22 percent in the Bundestag in 2012; in France since the 2012 reform, the figure for the lower house is now only 76.4 percent, down from 90.8 percent in 2007) (Baloge and Hubé 2015).The combination of these factors has consequences for the occupation of the media and political terrain, in other words access to the market for symbolic political goods.Even though, "in order to exist in politics, one has to be seen.In order to be seen, one has to move about and the only way to do this is to speak out" (Gaxie and Lehingue 1984:12; see also Juhem and Sedel 2016).The means of access to the media space are structured differently in the two countries.In France, an elected official with more than one mandate has fairly easy access to the regional daily press in a number of ways: as a local elected official, and/or département, regional, or national elected official.If their actions are not reported in the Paris media, they can easily be in the local press.Their access to media-covered political goods is thus direct and permanent, placing journalists in a situation of relative subordination (Frisque 2010).German MPs have less access to the regional media sphere than their French peers, for example.Indeed, the public sphere of individual federal states restraines media and political configuration, where the exchanges of information and informal relations take place essentially at the level of the Landtag (state-level parliament) (Grunden 2009;Burgert 2010).The journalistic coverage is focused on the political issues of the state, from which MPs are more or less "absent."Moreover, the local public sphere is often restricted around a unique regional newspaper and two public audio-visual channels.However, all the politicians and journalists I have met up until now in Berlin complain about these difficulties: how to exist through the media for a Berlin politician and how to justify one's added value of working in Berlin as a correspondent for one's title?
Journalists posted to Berlin reveal in interviews that their "added value" in being in Berlin is not to cover local matters, which they could just as well do from elsewhere, but rather to approach problems at a more general level and provide a local interpretation of the laws passed.So they do not primarily or exclusively cover local elected officials.In practice, the journalists form a pool of generalists working for the regional press.The Saarbrücker Zeitung has created a press agency for the regional press, BMS (Berliner Medien System GmbH) working with three exclusive client newspapers-Pfälzische Merkur, Lausitzer Rundschau, and Trierischer Volksfreundand a list of subscribing clients.Its local competitor, Die Rheinpfalz, shares its offices with the correspondents of the Stuttgarter Zeitung, Rhein-Neckar-Zeitung, and other dailies in southwest Germany.These correspondents' groups are unofficially but very formally organized as a decentralized editorial office and allocate the topics to be covered among themselves each day.Although each journalist is independent in what they write for their newspaper, this pooling system enables them to give wider coverage to more topics.In both cases the correspondents find themselves covering parliamentary affairs at a general level of policy topics so as to make policy sense at Land level.This is less to do with discussing the effects of a given law on the Land but rather with showing the consequences of a given measure (say, in the budget) on the Bund-Land budget balance, in other words so that an article of interest to a reader in Saarland will also interest a reader in the state of Bremen.The head offices, in Saarbrücken or Bremen city, then have only to find someone locally to interview for specific details.In addition, journalists are organized in informal associations and discussion groups of "provincial" journalists.This also stops them from focusing too much on their own regions so that they can distinguish and justify their posts to their central offices.
In other words, to gain a clear picture of exchanges between press and politics, these exchanges would need to be placed, even more than in this article, within the legal and political framework (and its practical ramifications) where the exchanges of political goods take place (Kocks and Raupp 2014).German politicians are forced into a policy of occupying a Land-level political market different from that occupied by French politicians, which because of the country's centralism, accumulation of political resources, and nationalization of political and media life (Kaciaf 2013) has an easier task focusing on the political center.

tHe economy of journaliStic Practice
From 1949 to 1999, Bonn, known as the Raumschiff (spaceship) among journalists and politicians, was a small town where everyone knew everyone.There were only a small number of parliamentary journalists.When the federal government institutions moved to Berlin in 1999, there was a massive influx of journalists who had not been in Bonn.So in January 1998 there were 768 BPK members in Bonn, and 933 members in Berlin in January 2000 11 and 3,000 journalists in total accredited to the Bundestag, according to the government spokesperson of the time (Heye 2002:287).Commentaries at the time of the move were euphoric, hailing Berlin's return to its old status as media city.The social configuration within which journalists worked was apparently modified, and one might have supposed there would be fiercer competition between publications.In fact, the earlier economy of exchanges was main-tained.The lack of, and desire to avoid, competition for off-the-record interviews (Hintergrundgespräche) is due to the strong historical institutionalization of this practice and the continued weak economic competition among newspapers.
However, caring little for expense in the boom years 1999-2000, all the newspapers opened offices in Berlin.The German dailies hoped to find a new readership and began to compete more openly, as happens in other European countries (Bourdieu 1998;Hallin and Mancini 2004:251-295;Duval 2005;Hubé 2013).Examples are the Bonn General-Anzeiger, Frankfurter Rundschau, Süddeutsche Zeitung, Frankfurter Allgemeine Zeitung, and Financial Times Deutschland, which from 1998 to 2000 launched specific pages or sections devoted to the city of Berlin.Radio and television channels, including three 24-hour political news channels (NTV, N24, and Phoenix), also entered the market in the capital.But this fad for Berlin did not live up to its hopes.In the 2000s the Berlin market was saturated.Nine daily newspapers had head offices there.There were three regionals (Berliner Zeitung, Berliner Morgenpost, Tagesspiegel), three tabloids (B.Z., Berliner Kurier, Bild-Zeitung), and three nationally distributed supraregionals (Die Welt, taz.dietageszeitung, Neues Deutschland).According to the German Audit Bureau of Circulation, in 2003 their total circulation in the Berlin market was roughly 1.08 million (city population 3 million) (Hubé 2008:251;Röper 2004:8).Furthermore, the former East Berlin market remained impermeable to these papers.Failing to find a market, and as the economy turned down, all the papers that had set up specific editorial offices in Berlin closed them.Nor did the free press manage to gain a foothold, neither in Berlin nor in the other cities.This is partly due to the fact that the economics of the press is based on advertising and subscriptions in a highly regionalized market: finding a scoop to boost daily sales is not a concern.This failure hardly gave journalists any incentive to alter their behavior, go for sensationalism, and heat up the economic competition between papers (Hubé 2008:246-250; see also Esser 1999).Political journalism in Berlin appears to have maintained some autonomy with respect to the journalism field more openly affected by commercial constraints and competition.
unter drei n g e r m a n y : e f f e c t S o f t H e S e fa c to r S n P r e S S -P o l i t i c S i n t e r a c t i o n S Any examination of the practical production of political information in the federal capital and the capital cities of the Länder immediately reveals a "structural connection" (Wegmann and Mehnert 2006:148).Journalists, communications workers, and some researchers agree that there are and must be off-the-record interviews (Hintergrundgespräche) at the initiative of politicians or off-the-record circles (Hintergrundkreise) at the initiative of journalists.All of these are places and occasions where there is an "informal interpenetration of politics and journalism" (Rinke et al. 2006;see also Hoffmann 2003;Grunden 2009).When French correspondents are asked to name their sources, they all mention these circles: "We all have Hintergrundgespräche.
We have the direct German stuff in the press club with German journalists and two or three foreign colleagues … It's institutional!Every two or three weeks.It depends on the news." 12But they quickly add how difficult it is to use the information."When people read in the newspapers 'a government source,' everyone knows where it comes from.It comes from those places.So you have to be careful when you bring the news out." 13 This surprise on their part is, on reflection, "typically French," as we pointed out in the introduction.Some observers see informal circles as being the fourth arena of parliamentary work (the nonpublic side of parliament) after public debate, public explanations (press conferences), and public media (Sarcinelli 2005:237).This is another reason why in February 2017 the Berlin administrative court asked the chancellor's office to publish the dates of and names of those invited under the constitutional right (Grundrecht) to transparency.Still, as commentators have pointed out, there is no obligation to reveal the topics of conversation, and nothing prevents informal circles from meeting or phoning each other.Strict application would make journalists' work almost impossible or pointless. 14The surprising thing is not, therefore, that nonpublic zones exist, but rather that they are so well preserved from public scrutiny and not least from leaks.In other words, how does the competitive world of journalism, seeking readers, viewers, and scoops (Bourdieu 1998:20), manage to maintain such a collective discipline?Far from being "cultural," the reason is to be sought in the historical structuring of the interactions between the two fields.In this case, too, the number of journalists is sufficiently large that each journalist taken individually has an interest in not keeping quiet so as to "beat the competition" or "get a scoop," while not risking reprisals since "everyone" heard the same thing. 15 And yet they do not.The specific German feature is to be found in the control exercised over off-the-record by its strict codification and in compliance by the various actors.
off-tHe-record aS Second nature?
The publicly asserted distance cannot only be explained by the constraint of a public grammar required by social distancing conventions (Lemieux 2000) or by the purely political doxa of the separation of powers (Schudson 2005).It is as if in practice the vagueness of off-the-record is clarified by procedures.Rather than situations of "controlled relaxation" (Wouters 2010) where the rules may be allowed some play, the informal relations between German journalists and politicians are formalized, regulated, and institutionalized.Compliance with the rules of confidentiality operates as "second nature," a self-imposed constraint that journalists, communications staff, and politicians work with, within a straitjacket of rules of conduct (Wouters 2010:164).In other words, it is not the interactions that are informal but more the 12 Interview with Libération's Berlin correspondent, April 18, 2004. 13Interview with AFP's Germany desk manager, Berlin, April 21, 2004.
14 "'Unter drei'-die Sache mit den Hintergrundgesprächen," Tagesspiegel, February 19, 2017. 15When I observed a press conference followed by an off-the-record briefing, there were more than 50 journalists in the Room (field notes, January 25 and 26, 2010).The chancellor's office's Hintergrundkreise are attended by 12-18 journalists (Heckel 2009:122).language used in these highly codified places and times.However constraining the rule may be, it does lessen the risk of betrayed confidence.
The codification process operates in two ways.First, in the definition of public arenas: each place has its type of interaction, type of information provided (public information arenas versus confidential arenas), and the roles played by the protagonists.When asked, a former Bundespresseamt (Federal Press Office) head of department from 2006 to 2009 clearly distinguished between "press conferences where we give the facts and nothing but the facts, and everything is public; and Hintergrundgespräche where we give explanations, but everything is off-the-record."Airplane journeys are identified as confidential places "when you can observe Merkel without a filter and only centimeters away."During these trips there is no doubt about what is at stake in the interactions."These conversations are always unter drei-the code that politicians choose, under which it is strictly impossible to write what is said but which directly gives us better information and situation assessments" (Heckel 2009:123).This testimony emphasizes the contractual and procedural dimension of the interactions.That is the second way of regulating this social space: codification of the publication of press releases.Unter drei corresponds to situations where confidences may strictly not be published.It is numbered three because it comes after two others laid down in the BPK's bylaws, all subject to sanction: exclusion from the group.§ 16 (1) Information is given at press conferences: unter 1. for general use [on] 16or unter 2. for use with no mention of source or name of informant [off] or unter 3. confidential [background].
(2) Informants may state how their information is to be handled.Association members and press conference participants are bound by this classification of the information.If no statement is made, the material is considered to be for general use.Any breach of these rules concerning the classification of information may lead to exclusion from the association or withdrawal of accreditation as permanent guest.
In other words, this classification of degrees of confidentiality refers to the number of intermediaries or channels officially used to obtain the information: unter 1 is an interpersonal communication from A to B; unter 2 is information heard on channel A to B via C, namely a "close source" to be protected; unter 3 is information from A to B obtained by C who learnt it via D, namely a close source of a close source.So in symbolic terms (the rhetoric of journalist objectivity [Tuchman 1972]), a journalist cannot use this information without losing credibility, because the information could not be checked-it would be considered hearsay.
The usual recurring question for a journalist is when and if the information can be divulged.French off-the-record is rarely unter drei: it is given en toute confidentialité by a politician with the aim-or at least the knowledge-that the journal-ists will release it, while at the same time the politician seeks to ensure themselves a protection whereby they can say, if a controversy blows up, that they did not say it.An impossible task, of course; one solution for journalists is to be found in the place and time of the conversation and the number of journalists aware of the information.Most of the time the breach of off-the-record is tacit, since the actors know-that is, have internalized-the boundaries of what is possible and also know each other.But in order to be certain, journalists often get together after off-the-record conversations to agree what they have heard and whether it can be released, as long as a colleague seeking a competitive scoop does not release it first.The centralization of power and the media (and consequently the number of players in the game) and the desire in the journalism field for higher reading figures and scoops are without doubt the structural principles of this practice.This codification of places and uses for press releases is only of interest to the actors as a way of objectivizing a procedure and placing the relationship of trust on a longterm basis.I have used Niklas Luhmann's (1988) term "decided trust" because the relations between journalists and their sources are neither natural nor familiar, but are rather part of the practical rules of the political field in Berlin.For a journalist, exclusion is a sword of Damocles.But what maintains the symbolic order is the fact that this exclusion is the work of journalists alone.In that way, they protect the procedure and maintain a refined system of inclusion and exclusion, selecting new entrants and excluding undesirables, in other words, those who do not follow the rules.As in high society in the nineteenth century, this code works "to regulate social mobility and competition for status" (Wouters 2010:164).Admission to the BPK is by written application, considered by the management board subject to written objection from a BPK member.The Berliner Press Club was set up in 1952 by Berlin journalists and publishers.It succeeded the Verein Berliner Presse, an association highly active during the Weimar Republic as a place for socialization and exchange for the Berlin world of politics and journalism, the ancestor of the BPK.New club members had to be sponsored by at least two existing members, their application had to be unanimously approved by the management board, and applications were only accepted from eminent Publizisten, whether newspaper managers, editorialists, or writers for major publications (Bonk 2007a:32).Within living memory, the code has nearly always been complied with.The archives of the Berliner Presse Club, which Christina Bonk examined, contain only four disputed cases in 55 years of existence (1964, 1968, 1972, and 1975) (Bonk 2007b:70-71).Breach of confidence in these circumstances has consequences both for the culprit and the group.Trust can only be guaranteed because the organization ensures relative independence for these spaces by bringing "into play new, impersonal, motives for action" (Luhmann 1979:93; see also Bourdieu 1990).So breaching this trust amounts to unraveling the "internally guaranteed security" provided by the organization (Luhmann 1979:93), that enables politicians to frequent journalists.In other words, to avoid reintroducing mistrust the journalists' reaction is constrained by the need to maintain the well-functioning of the organization.In France the person "punished" does not long remain so, and once the politician's anger has passed, the practical consequences for the offender are, with very few exceptions, extremely limited.
wHen every tHing cHangeS, notHing cHangeS With power changing hands in 1998 and the move to Berlin in 1999, the exchanges between journalists and politicians were unsettled (Hubé 2016).The case is of interest because it temporarily introduced a certain fluidity into the political game, placing the model under stress.Two elements were fundamentally different from Bonn.First, space-the size of the city is often put forward by interviewees as an explanatory factor for changes.Journalists have to timetable their days more effectively and be more selective about the events to cover.Each side attempts in vain to gain an advantage.After 1999 one might have expected the ground rules to be altered.Requests from the media for interviews increased in number.Politicians were thus in a position to choose among publications or at least to pretend they could choose.Second, the actors themselves.Gerhard Schröder's first term, coinciding with the shift of capital city, was a time of potential modification to the manner of regulating groups and the game, since the number of interdependent players increased (Elias 1991:98)-and especially, because the chancellor himself went in for a personalization and celebrification of his role.A former Christian Democratic Union (CDU) communications officer who then entered the government communications department said, "the beginning of his term was an electric period, when everyone was keyed up, both journalists and politicians … but then things calmed down in the mid-2000s.That was the transition period." 17His remarks have been confirmed by a number of journalists interviewed.
Two crises erupted within five months, both for the same reason: politicians' practice of using the interview and the forms of coercion in their power (insisting that a sentence be altered or refusing to give an interview), which at a time of tension between the two groups looked like censorship and thus a threat to the proper workings of democracy.On November 28, 2003, the SPD's general secretary asked to reread and rewrite his interview, which the tageszeitung refused.This ordinary incident (having an interview reread is routine among journalists) received extraordinary treatment: all the major newspapers in Berlin reacted with a single voice in their editorials.Four months later a further spat with politicians became an opportunity to reassert the ground rules.This time two key actors were opposed: the chancellor's office and the tabloid Bild-Zeitung.The former refused to let journalists from the daily Bild (and the magazine Stern) join official delegations.In early March 2004 the government's spokesperson officially confirmed that no further interview would be granted, since it was considered that the campaign against the chancellor was "odious."This time not only the BPK, the journalists' union, but also the speaker of the Bundestag and the opposition party groups objected to this statement.The degree of mobilization matched the importance of the issue involved in this government retaliation.The journalists saw the refusal as an act of censorship.They spoke of Germany's darkest hours and the constitutional freedom of the press."This moment is to be taken seriously.This is not a joke any more.It's about the freedom of the press," said the editor of Stern, quoted by a colleague. 18The chancellor had touched a livewire of postwar politics: press freedom (Kopper 2003;Robert 2014).
The decision was seen as particularly illegitimate because Chancellor Schröder had made so much of his communication.It was immediately perceived by actors as a new practice of "disciplinary public relations" (Disziplinierungs-PR) 19 that meant that editorial offices had no choice but to stand together.It put a temporary end to the benefit of the journalists and the Bonn modus vivendi, to a process of redefining practice that had begun with the issue of interview correction four months earlier.What this case shows is that a change of power as a time of hesitation and fluidity offered all the actors a chance to attempt to impose new standards, during the first term of the media chancellor (1998)(1999)(2000)(2001)(2002).But the fluidity came to an end during the 2003 political crisis.The hysteresis of the political-journalistic field was the determining factor.*** German journalism is probably an extreme example of confidential exchanges of information and nonpublic relations between interdependent groups motivated by partially opposing interests.The added-value of looking at the issue from a foreign viewpoint is that it encourages us to look at the institutional, economic, and political structures that help in understanding these relations in their historicity.These exchanges are constructed by acts and particularly by precise moments of tension that, when resolved, temporarily reduce uncertainty and put these exchanges on a longterm footing.With the move to Berlin this extreme case provides what social sciences rarely have the chance to experiment with: the change in an institutional and economic configuration that only slightly modifies the structure of exchanges.The lesson of this case is that ultimately the codification of relations between journalists and politicians wins out over their interest in upsetting this equilibrium of practices, because it is so internalized.Adding, of course, that this system is a contingent one, historically and socially constructed.