Framing TTIP in the wake of the Greenpeace leaks: agonistic and deliberative perspectives on frame resonance and communicative power

ABSTRACT

Although never conceived as a tool of direct democracy, the European Citizens’ Initiative (ECI) raised hopes that it would involve citizens more directly in EU decision making. Previous research has suggested that one contribution of the ECI is its effect on fostering public deliberation on EU issues, raising questions about the ECI’s potential as a tool for social movements to generate communicative power in relation to EU issues. This article draws on agonistic and deliberative perspectives to argue that communicative power generation can be seen as a process where ECI organizers use social media to advance specific understandings of their concerns and channel those understandings into mainstream mass media. The article analyses this by investigating how frames constructed on the Stop TTIP campaign’s Facebook page have resonated in twelve online news sites in four European countries in the wake of the Greenpeace leaks.

KEYWORDS:

• European Citizens’ Initiative (ECI)
• European Union
• TTIP
• communicative power
• frame resonance
• agonistic democracy
• deliberative democracy

1. Introduction

Although at times described as an instrument of direct democracy, the European Citizens’ Initiative (ECI) raises intriguing questions from the perspective of deliberative democracy. This is particularly true in relation to its potential to foster transnational deliberation and thereby contribute to the generation of communicative power. In this context, social media and their wider impact on the public sphere play a key role. There is already a considerable literature on social media’s transformative impact (e.g. Rasmussen 2013), specifically as mobilization tools for social movements. However, too little attention has been paid to the resonance of social movement frames in mass media discourse. This is unfortunate because framing and frame resonance, as this article argues, constitute core ingredients in the deliberative-democratic concept of communicative power. As a power resource that is specific to the public sphere, communicative power plays a key role as a counterweight to the administrative power held representative institutions (Habermas 1992). The concept is also important in discussions about the role of deliberation in transnational democracy (e.g. Bohman 2005, 2007, 2010).

In the context of communicative power generation, social media present an opportunity for social movements to advance specific understandings of their concerns and ultimately to channel those understandings into mass media discourse. This is particularly relevant to discussions about the democratizing impact of the European Citizens’ Initiative (ECI), which has attracted considerable criticism for its demanding legal requirements as much as for its non-binding character. Even legally admissible initiatives that manage to collect the required one million verified signatures do not automatically result in legislative proposals by the European Commission (Bouza García 2015). This has resulted in claims that the ECI is as demanding as a genuinely direct-democratic citizens’ initiative in terms of formal requirements, but at the same time as weak as a mere petition in terms of its legislative impact. Despite such objections, however, both formal and informal¹ initiatives/signature-collection campaigns have had considerable success in raising awareness for certain issues, explaining why the ECI has been interpreted as an instrument of transnational deliberative democracy (Conrad 2016; Greenwood and Tuokko 2017).

The argument advanced in this article is that the generation of communicative power in public deliberation can be understood by tracing how frames constructed and advanced by ECI organizers resonate in mainstream mass media discourse and ultimately impact on decisions made by the EU’s legislative and executive institutions. Empirically, this process is illustrated by an analysis of the extent to which frames advanced by the ‘self-organized’ Stop TTIP initiative have resonated in mass media coverage on TTIP in four EU member states in the wake of what has become known as the ‘Greenpeace leaks’ in May 2016. The Transatlantic Trade and Investment Partnership (TTIP) between the EU and the United States (De Ville and Siles-Brügge 2016), along with the EU-Canada Comprehensive Economic and Trade Agreement (CETA), has been among the most contentious projects in European integration in recent years (Oleart and Bouza 2018a). As such, TTIP has received considerable academic attention, including a special issue in this journal, which has focused on TTIP and CETA as testimony to the broader phenomenon of ‘contentious market regulation’ (Laursen and Roederer-Rynning 2017; Hübner, Deman, and Balik 2017; Dominguez 2017). In line with a neoliberal discourse (De Ville and Orbie 2014), the European Commission has argued that both TTIP and CETA will bring considerable opportunities in terms of economic growth and employment (Garcia-Duran Huet and Eliasson 2017). Opponents of TTIP, on the other hand, have drawn attention to the considerable (perceived) risks associated with the two trade agreements, mainly in terms of consumer and environmental protection (De Ville and Siles-Brügge 2017), but also as a potential threat to democracy and the rule of law (Siles-Brügge 2017; Oleart and Bouza 2018b).

The case of TTIP is of particular interest because the transnational campaign against TTIP succeeded in mobilizing significant public opposition against the proposed agreement, even though Stop TTIP was never formally registered as an official ECI by the European Commission. Having collected well over three million (albeit unverified) signatures, the initiative has been described as one of the few episodes of ‘transnational politics’ (Young 2017) where there has been a close connection between parliamentary and extra-parliamentary opposition in the EU (Crespy and Parks 2017). As such, the episode of the Greenpeace leaks in May 2016 is relevant as the best illustration of such ‘transnational politics’.

It should be emphasized that the case of Stop TTIP is indeed chosen for purely illustrative purposes. While the empirical analysis certainly contributes to the literature on EU trade policy (Meunier and Nicolaidis 1999; Young and Peterson 2014), the article’s main ambition is theoretical. We demonstrate how the literature on framing and frame resonance can inform our understanding of the Habermasian concept of communicative power generation. Further empirical research on other cases, not least from contexts beyond EU politics, will be necessary to be able to generalize from our findings. In this article, we use perspectives of deliberative and agonistic democracy to create a link between framing, frame resonance and communicative power generation. The ECI presents an institutional opportunity for activists to promote particular understandings of contentious issues that may be fundamentally opposed to the ‘official’ discourses of the European institutions. Activists’ successes in advancing frames that resonate in the public sphere at large are seen as key ingredients in the generation of communicative power, which in turn also has implications for the continued democratization of the EU.

The following section provides the theoretical groundwork for our account of communicative power generation. While the concept of communicative power clearly has roots in the Habermasian tradition of deliberative democracy, our understanding of the process can also be enhanced by taking up aspects of Mouffe’s understanding of agonistic democracy. Using this insight, the third section spells out the connection between framing/frame resonance and communicative power generation in the context of the ECI. Methodological choices are introduced and justified in section 4, before sections 5 and 6 delve into the empirical analysis. Section 5 presents the findings of the social media analysis, identifying the most salient frames through which the transnational Stop TTIP campaign has made sense of TTIP on its Facebook page. In section 6, these findings are then contrasted with the findings of our media content analysis of the debate on TTIP on twelve online news sites in France, Germany, Spain and the UK. Finally, the concluding discussion returns to the overarching question of the extent to which framing and frame resonance can indeed be regarded as processes of communicative power generation.

2. Deliberative and agonistic democracy

Given our interest in communicative power generation, our theoretical point of departure is that the differences between deliberative and agonistic perspectives on democracy and the public sphere are to an important extent exaggerated. In fact, aspects of both kinds of theories can be combined usefully to make sense of what communicative power is and how it is generated in the public sphere. We make this claim by reference to two observations. On the one hand, Habermas’s two-track model of the public sphere already contains a certain ambiguity regarding the relationship between rational argumentation and other forms of contestation. On the other hand, agonistic perspectives underline that while it may not always be possible or even necessary to settle normative conflicts through rational deliberation, channelling them in appropriate ways is nonetheless important.

In Habermasian deliberative theory, deliberation is seen foremost as a way of enhancing the democratic legitimacy of decisions made by representative institutions, since the most fundamental starting point for deliberative theories is that representative democracy in one way or another creates gaps between citizens and decision makers. Consequently, maybe the most important role of the public sphere is to probe the democratic legitimacy of the decisions made by the institutions of the political system, in particular as regards the extent to which the latter are based on good arguments (and justifications) and not merely reflections of the self-interest of the involved decision makers.

There is however a certain ambiguity regarding the consensus orientation of public deliberation in that the Habermasian two-track view of democratic politics implies an at least potentially antagonistic relationship between institutionalized decision making and deliberation in the public sphere. Along those lines, this paper introduces a normative dimension rarely addressed by deliberative democrats, namely the importance of political conflict for a vibrant democratic debate in the public sphere. Whereas deliberativists emphasize rational deliberation and the possibility of reconciling diverging normative worldviews, the agonistic critique broadly holds that conflict based on diverging normative worldviews is a constitutive element of democracy. Democracy (and pluralism) therefore also require the recognition of conflictual relations between actors holding (potentially) irreconcilable normative world views. The recognition of such conflict is important because, as agonists presume, an orientation towards rational consensus – as required by deliberativists (e.g. Cohen 1997, 74–5) – excludes specific perspectives, overlooks the importance of counterhegemonic discourses and is therefore biased towards the status quo (Mouffe 2005, 2013). For our argument, it is however also important to note that Mouffe’s agonism advocates a particular way of channelling conflict, where opponents recognize one another as legitimate participants of the debate and thus as adversaries rather than enemies. For agonists, the public sphere is a place where counterhegemonic discourses can be articulated and should thus be thought of as a vibrant (and indeed ‘agonistic’, as opposed to ‘antagonistic’) sphere of contestation where different hegemonic political projects can be confronted (Mouffe 2005, 3).

Given this shared emphasis on the public sphere as an arena for contestation, there is arguably more that unites than separates Mouffe’s agonistic and Habermas’s deliberative account of democracy (cf. Karppinen, Moe, and Svensson 2008). In fact, the Habermasian understanding of deliberative democracy already touches on the relevance of conflict in the generation of communicative power and, by extension, democratic legitimacy. Both perspectives therefore offer a fruitful basis for conceptualizing communicative power, all the more so as Habermas views the role of civil society in the public sphere as one of identifying, staging and amplifying concerns to be addressed in public deliberation.

3. The ‘two faces of framing’: frame resonance and communicative power

In Habermasian democratic theory, public deliberation generates communicative power. Through a process of argumentation, actors in the public sphere not only challenge the normative validity of institutional decision making, but also identify, stage and amplify concerns that may not be visible to the institutions of the political system. This process may seem akin to the kind of contestation promoted by an agonistic understanding of democracy; at its core, however, it is about raising awareness for and initiating deliberation about perceived injustices (Bohman 2010).

3.1. From communicative freedom to communicative power

Despite the evidently central role that communicative power plays in Habermasian deliberative theory, the concept has proven difficult to define (Flynn 2004) and consequently still lacks a commonly accepted operationalization. As a power resource, communicative power can be said to presuppose both the existence and exercise (on the part of citizens) of communicative freedom. Bohman defines communicative freedom as ‘the exercise of a communicative status, the status of being recognised as a member of a public’ (Bohman 2010, 432; emphasis added), which rhymes well with deliberative as well as agonistic understandings of democracy. The generation of communicative power is virtually inconceivable without the presupposition of the exercise of communicative freedom. As a matter of fact, Bohman views individuals’ ability to ‘initiate deliberation’ about perceived injustices as one of the hallmarks of the normative ideal of transnational democracy (Bohman 2007).

But the exercise of communicative freedom clearly does not in and of itself constitute communicative power. Instead, communicative freedom turns into communicative power only to the extent to which it has an impact on institutional decision making. In Bohman’s words, communicative freedom ‘is transformed into communicative power when it is incorporated into institutionalized processes of decision making’ (Bohman 2010, 432, emphasis added). The ECI has been interpreted as an instrument of transnational deliberative democracy precisely because it represents an institutional opportunity for channelling deliberation into the EU’s legislative institutions (Conrad 2016). From this perspective, communicative power generation is as a process beginning with (1) the perception of an injustice, continuing with (2) the exercise of communicative freedom for the purpose of initiating deliberation, and culminating with (3) the channelling of deliberation from the public sphere into the legislative institutions of the political system.

3.2. Framing and frame resonance as the missing link

Goffman originally coined the term ‘frame’ as a metaphor to highlight the complexity of social reality, basically to emphasize that perceptions of reality depend on highlighting certain aspects while toning down others (Goffman 1974; cf. Entman 1993). Frames can be described as ‘ready-made interpretative packages’ that draw attention to the aspects that a given observer finds most relevant in describing and/or explaining a specific phenomenon. The framing perspective takes frames as socially constructed by actors that can use frames strategically to define issues in particular ways.²

The discussion on the relationship between communicative freedom and communicative power above highlights the importance of the relationship between framing and frame resonance. In the same way that the exercise of communicative freedom does not in and of itself generate communicative power, the success of framing efforts, for instance by social movements/activists, depends crucially on frame resonance, that is: social movement frames need to resonate in wider forums of the public sphere in order to turn into communicative power with an impact on institutionalized decision making.

Baumgartner and Mahoney (2008) described this as the ‘two faces of framing’. At the individual level, actors frame issues in order to legitimize their claims, encouraging particular interpretations and understandings of reality. Such partisan frames are deliberately partial, address an audience and are intended to gain a hegemonic position in the public sphere rather to convince opponents. The second face of framing refers to the perception of such issues by the wider public and/or policy makers, indicating that the second face of framing is closely connected to the fundamental question of frame resonance in the public sphere.

This second face of framing is therefore crucial in linking framing (as a specific exercise of communicative freedom) to communicative power generation. In order to generate communicative power, framing processes have to be sufficiently convincing and persuasive to resonate with wider audiences, questioning the normative appropriateness of the issue at hand, and resulting in a broadly accepted image of a need for action.

3.3. Social movements, mainstream media and frame resonance

But how do frames travel from social media into mainstream media discourse? Cammaerts (2018) speaks of ‘circuits of protest’ in this context, conceiving of frame resonance as a four-step process in which frames are (1) created by social movements, (2) communicated both internally and externally (e.g. via social media), (3) migrate into the mainstream media, and finally (4) resonate with the public at large.

We do not view social media as the only – or even necessarily the most important – channel of communication available to social movements (in this case: ECI organizers). Consequently, we do not propose a unidirectional, let alone monocausal, understanding of frame resonance. The frames that social movements advance via social media are clearly also advanced via other channels, in other situations and contexts, presumably also in direct communication with journalists. But this research is not (primarily) interested in why these frames resonate, but that they resonate, and especially what their resonance means in terms of communicative power generation.

Given our focus on mainstream media discourse, our reflections on this dynamic are closely connected to the role of journalists. Due to their respective roles as ‘gatekeepers’ and ‘critical watchdogs’, not least in European integration (Trenz, Conrad, and Rosén 2009), journalists have the power (and indeed responsibility) to determine which speakers are given voice. This can occur in different forms, depending on the respective roles played by journalists in terms of news reporting, analysis or opinion making, especially in democratic-corporatist countries such as in Northern and Central Europe (Hallin and Mancini 2004; Conrad 2014). Voice can be given directly to speakers representing social movements, whether through direct or indirect quotes, interviews, or even guest commentaries. But voice can also be given more indirectly, most importantly when perspectives advocated by social movements are adopted by journalists and incorporated into leaders, editorials or commentaries.

The extent to which this practice of giving voice to social movements takes place – whether directly or indirectly – is arguably dependent on the ‘news value’ or ‘newsworthiness’ (Lippmann 1922; Galtung and Ruge 1965) of a given item, which is also connected to the journalists’ gatekeeper role. In our case, the likelihood of the adoption of social movement frames in mainstream media discourse increases if they are articulated in the context of a mass demonstration against TTIP or, as in this case, in the context of an event such as the Greenpeace leaks, which news media can be assumed to have an interest in reporting on.

The remaining question is whether one can speak of frame resonance already if frames advanced by social movements have reached the mainstream by means of a simple quote in news reporting, or whether the frame has to be incorporated into journalistic opinion making and/or analysis. From our perspective, giving voice in the form of direct or indirect quotes already constitutes an instance of frame resonance, due to the fact that the perspective advanced by a social movement was considered important enough, by the gatekeeper, to pass through the gates of news production and thereby to reach a wider audience.

4. Methodological discussion: cross-country media content analysis

Communicative power generation, understood here as a process of framing and frame resonance, needs to be studied both from the perspective of (a) the frames advanced by social movements (in our case: ECI organizers) and (b) the frames adopted in mass media discourse. In the first step of our empirical analysis, we therefore analyse the frames employed by the Stop TTIP campaign on its Facebook page. Although social media do not replace traditional channels of communication (e.g. press releases, demonstrations, conferences), political scientists have identified social media as one of the main channels through which civil society organizations and social movements encourage the politicization of issues, including transnational ones (see Barisione and Michailidou 2017 for an overview, Barisione and Ceron 2017 on austerity protests, and Ruiz Soler 2018 on TTIP and Schengen). For the purposes of this research, Facebook was chosen because it is currently the most popular social medium in terms of users (Hutt 2017).

Our sample of Facebook posts only includes posts that contain an identifiable message about the nature and/or consequences of TTIP (and/or CETA, for that matter)³, or that propose steps to be taken to stop the two agreements. The time frame for the analysis spans the month of April 2016, i.e. the month prior to the publication of the Greenpeace leaks on May 1. This time frame was chosen so as to avoid that the frames used by Stop TTIP were influenced by the conclusions from the Greenpeace leaks. The sample contains 21 posts (status updates) and was analysed inductively, with frames being constructed and integrated in successive rounds of coding.

The main part of the empirical analysis, on the other hand, consists of an analysis of the extent to which these frames, often viewed as alternative frames to the ‘official’ discourse of institutional actors, resonated in mass media discourse, and not least what kind of opposition they met from other frames in mass media discourse. In this second step, the findings of the social media analysis are contrasted with the findings of an analysis of material from twelve online news sites in four EU countries, namely France, Germany, Spain and the United Kingdom. The news outlets included in the analysis were chosen on the basis of their respective political orientation, corresponding to different ‘general political tendencies’ (Hallin and Mancini 2004, 27). For each country, one leading conservative/economy-oriented, one leading liberal/mainstream and one leading left/alternative publication was selected, based on the type of information the respective publications aim to provide and the public they aim to reach (see Table 2). The media selected are arguably not exact counterparts, but they do play a similar role in the media landscape of their respective countries.

Table 2. Analysed publications by country and political orientation (number of sampled articles in parentheses).

	Conservative/economic	Liberal/mainstream	Left
France	Les Échos (7)	Le Monde (5)	Libération (2)
Germany	Frankfurter Allgemeine (10)	Süddeutsche Zeitung (9)	die tageszeitung (9)
Spain	El Economista (4)	EL PAÍS (6)	eldiario.es (9)
UK	Financial Times (1)	BBC (2)	The Guardian (4)

The articles included in our sample were published in the online version of the selected news outlets during the three-day period following the publication of the so-called ‘Greenpeace leaks’, i.e. between May 1 and 3, 2016. This episode may not necessarily be representative for the wider reporting on TTIP during the almost four years of negotiation. However, it is certainly a relevant episode to analyse how (and to some extent why) frames travel from the realm of critical civil society organizations to the mainstream media. The total mass media sample consists of 68 articles stemming from all sections of the selected publications and were selected on the condition that their title or subtitle included the search word ‘TTIP’ in its different forms.

The sampled articles were coded using the NVivo software in close interaction between the two authors. Beyond basic information including the country, name and orientation of the respective newspapers, coding was done at two levels, namely (1) the titles and subtitles of the sampled articles, and (2) the main body of the sampled articles. This distinction was used to capture the importance of a given frame within an article and thus to take into account the hierarchy of frames. Similar to the social media analysis, frames in this part of the analysis were developed inductively and integrated in successive rounds of coding. The reliability of the coding frame and the coding itself was ensured by ongoing interaction and discussion between the two coders/authors.

5. Frames advanced by stop TTIP on facebook

The analysis of the frames used on the Facebook pages of the European Stop TTIP campaign identifies two broad categories of frames, namely frames referring to the process of the TTIP negotiations and frames that emphasize the content and perceived consequences of the proposed agreement (see Table 1). While the former draw attention e.g. to the secretive character of the TTIP negotiations, the latter category consists exclusively of what we refer to as threat frames. These portray TTIP as a matter of threat to e.g. Europe’s regulatory standards. Stop TTIP sees the risk that European standards will be lowered or bypassed to meet American demands, which could ultimately result in allowing e.g. genetically modified vegetables and meat on European markets. The posts also underline the high salience of the potential impact of Investor-State Dispute Settlement (ISDS) courts on the rule of law in EU member states, since the Stop TTIP campaigners largely look at such courts as a form of private justice. A key argument in this context is also that ISDS courts would have a detrimental effect on European legislative standards, largely for fear of ISDS lawsuits. The posts also introduce the idea that TTIP favours the interests of corporations over those of citizens, and that the agreement is therefore a threat to democracy.

Table 1. Frames used on the Stop TTIP Facebook page, April 2016.

Category	Frame	Used in # of posts	Used in % of posts
Content	Threat to regulatory standards	5	24
	Threat of private courts	4	19
	Corporations over citizens	3	14
	Threat to democracy/popular sovereignty	3	14
Process	Lack of democratic standards	4	19
	Lack of support, mass opposition	3	14

Furthermore, the campaign frames the TTIP negotiation process as well as the CETA ratification process as being characterized by a fundamental lack of democratic standards, both with regard to the involvement of parliaments and citizens. One key element in this way of framing TTIP is the lack of transparency in the process, as well as the fact that citizens in the member states hardly had any direct say in the negotiation and/or ratification of TTIP and CETA.

6. Frame resonance: framing TTIP in twelve online news outlets in four EU countries

To what extent and in which ways did the frames identified in the Facebook material resonate in the twelve online news sites analysed here? Similar to the social media analysis, the analysis of the mass media material identifies two different categories of frames, i.e. process frames and content frames. However, the content frames in the mass media material also include various opportunity frames, including economic opportunities in terms of jobs and growth, but also opportunities for the EU to play a global leadership role in terms of promoting European labour and environmental standards in international trade (see table 3 below) – aspects that the Stop TTIP social media campaign explicitly denied.

Table 3. Frames identified in the media content analysis.

Main category	Subcategory/ Master frame	Context-Specific Frame
Process frames	Process	Lack of Transparency
		‘Horse race’ (i.e. difficult negotiation process)
		Mass opposition and/or lack of public support
Content frames	Threat	Threat to regulatory standards
		Threat of private justice (i.e. private arbitration courts)
		Corporations versus citizens
	Opportunity	Jobs and growth
		Global leadership

6.1. Process frames

The category of process frames includes three frames, namely ‘lack of transparency’, ‘horse race’ and ‘mass opposition’. The ‘lack of transparency’ frame focuses on the secretive character of the TTIP negotiations, while the ‘horse race frame’ emphasizes disagreements between executive actors participating in the negotiations, thus questioning whether the TTIP negotiations can be finished successfully.⁴ The ‘mass opposition’ frame, in turn, focuses on non-executive and/or civil society actors’ opposition to TTIP.

6.2. Content frames: threat

The first frame in the subcategory of threat frames sees TTIP as a threat to European regulatory standards in fields such as environmental and/or consumer protection, food safety or the precautionary principle. The ‘threat of private justice’ frame emphasizes the role of the proposed ‘private arbitration courts’ for Investor-State Dispute Settlement (ISDS), which are seen as a challenge to the rule of law in EU member states. Similarly, the ‘corporations vs. citizens’ frame portrays TTIP as a project for the benefit of big multinational corporations, whose interests will be prioritized over the interests of citizens, thus posing a threat to democracy in Europe.

6.3. Content frames: opportunities

Among the opportunity frames, references are made to TTIP as an economic opportunity that might generate jobs and stimulate growth. This ‘jobs and growth’ frame thus highlights economic output, but without considering environmental or consumer safety concerns. The ‘global leadership’ frame, in turn, views TTIP as an opportunity for the EU to strengthen its position in international trade and thereby to promote European standards in a globalized economy.

6.3.1. Frames in the titles and subtitles of the sampled articles ⁵

This part of the analysis is distinguished from the analysis of the frames used in the main body of the text, based on the assertion that the use of frames in the titles/subtitles provides information about the hierarchy of frames: frames employed in the titles/subtitles suggest what the analysed publications consider to be at stake in the given context (for a complete overview, see Appendix 1). Three frames clearly stand out in this regard, in addition to which they correspond very closely to the frames identified in the social media material: the ‘lack of transparency’, ‘horse race’ and ‘threat to regulatory standards’ frames (see Table 4). However, there are fairly clear country-specific differences: The German sample is considerably more balanced than the other three in the sense that no single frame stands out as particularly dominant. Here, the three most salient frames appear in the titles of 21% (horse race and threat to regulatory standards) and 11% (lack of transparency) of the coded articles, respectively. The German sample also differs from the other three in the sense that the negotiations’ lack of transparency is addressed considerably more rarely (in 11% of the articles) than in the Spanish (68%), French (64%) and British (43%) samples.

Table 4. Most frequent frames in titles/subtitles (whole sample, by country).

	Lack of transparency	Difficult negotiation/ ‘horse race’	Threat to regulatory standards
GER	11%	21%	21%
UK	43%	57%	43%
FRA	64%	36%	14%
ESP	68%	32%	53%

% of articles in respective national samples

Although the emphasis both on a lack of transparency and on the threat to regulatory standards suggests the strong resonance of social media frames in mass media discourse, the sampled articles overall also strongly emphasize that EU and US negotiators are still a long way from reaching an agreement on TTIP. Consequently, the ‘horse race’ frame (which is not employed in the social media material) is the most salient frame in the titles/subtitles of the British articles (57%), but it is also important in the French and Spanish samples (36% and 32%, respectively). This indicates a certain deviation from the emphasis placed in the Stop TTIP Facebook campaign, which had claimed (at least implicitly) that the European Commission was sacrificing European regulatory standards in the wake of a non-transparent negotiation process. On this point, there is also a clear difference between the French and Spanish articles on the one hand, and the German and (albeit to a lesser extent) the British articles on the other: The titles/subtitles in the French and Spanish articles strongly emphasize the negotiations’ lack of transparency (mirroring the framing promoted by Stop TTIP) and place significantly less emphasis on the seemingly irreconcilable differences between European and American negotiators. The negotiations’ lack of transparency, by comparison, plays a considerably smaller role in the German titles/subtitles – especially in the conservative FAZ, where it isn’t taken up at all.

When looking at the material from the perspective of the different political orientations of the newspapers, it is apparent that the negotiations’ lack of transparency plays a considerably more central role in the framing of the left and liberal/mainstream publications than in the economic/conservative ones. In the former, the frame is employed in close to half of the sampled articles titles/subtitles, whereas it only appears in a third of the titles/subtitles of the latter (see Table 5).

Table 5. Most frequent frames in titles/subtitles (whole sample, by newspaper orientation).

	Lack of transparency	Horse race	Threat to regulatory standards
Left	46%	29%	25%
Liberal/mainstream	45%	36%	41%
Ecoomic/Conservative	32%	27%	27%

In terms of frame resonance, the analysis of the titles/subtitles therefore suggests that the frames advanced by the Stop TTIP campaign do indeed resonate in the analysed material, but they arguably resonate more strongly in the Spanish and French than in the British and German articles, and they resonate more strongly in the left and liberal/mainstream than they do in the economic/conservative publications.

6.3.2. Frames in the main body of the articles

With regard to the frames employed in the main body of the sampled articles, patterns are fairly similar across countries, and the frames employed by Stop TTIP resonate clearly (for a complete overview, see Appendix 2). Four frames stand out as particularly relevant in all four countries, namely the ‘threat to regulatory standards’, the ‘horse race’, the ‘lack of transparency’ and the ‘corporations over citizens’ frames. In fact, the first three are the most frequent in all four analysed countries. The ‘threat to regulatory standards’ frame is the most frequent in the UK, France and Germany, while the ‘horse race’ frame is the most salient in the Spanish material (see Table 6).

Table 6. Most frequent frames in main body of articles (by country).

	Threat to regulatory standards	Horse race/ difficult negotiation	Lack of transparency	Corporations over citizens
FRA	100%	93%	57%	29%
GER	68%	61%	50%	32% (not top 4)
ESP	68%	95%	79%	53%
UK	100%	86%	43%	71%

% of all articles in the respective national samples

One qualification needs to be made as regards the German material, which is the only part of the overall material where the ‘corporations over citizens’ frame – although salient (32% of the German articles) – is not among the four most frequently employed frames. Much greater emphasis is placed on this aspect in e.g. the British publications. As the Guardian writes on 3 May 2016, ‘the main driving forces behind TTIP have been multinational corporations and business lobby groups, who stand to gain from harmonized regulations.’

Similar to the frames in the titles/subtitles of the articles, the frames employed in the main body of the articles indicate the salience of the issues identified in the Greenpeace leaks and thereby primarily draw attention to the perceived threat of TTIP in terms of European regulatory standards, most importantly in the fields of environmental and consumer protection. Literally all articles in the French and British samples employ this frame, which is furthermore featured in two thirds of the German and Spanish articles. However, the ‘horse race’ frame is similarly important (and even more important in the Spanish case), which deviates at least to a certain extent from the frames advanced by the Stop TTIP campaign. This suggests that the analysed publications place greater emphasis on the conflict-ridden nature of the negotiation process, without necessarily implying that the outcome of the negotiations will lead to an undermining of European standards.

It is also important to look at frames that might be expected to matter, but that turn out to play virtually no role. The clearest example may be the ‘economic opportunity’ frame, which could be expected to play an important role, at least in the economic publications. However, it only appears in a relatively small share of the analysed articles: 23% of the economic/conservative, 32% of the liberal/mainstream and 21% of the left/alternative articles (see Table 7). Strikingly, the conservative/economic publications place less emphasis on TTIP’s economic opportunities than their liberal counterparts. In fact, even when the frame is present, it often merely provides a backdrop, as exemplified by an article in Les Échos (2 May 2016): ‘when the negotiations were launched, a study published by the Centre for Economic Policy Research (CEPR) highlights that the European economy would generate an additional 120 billion Euros thanks to TTIP’⁶. However, in the case of economic/conservative publications such as the German FAZ or Les Échos in France, it has to be noted that economic opportunities play a more central role in editorials and commentaries than a lack of transparency or the undermining of European standards. This finding highlights the distinction made above between frame resonance in opinion-making as opposed to news-reporting journalism. While civil-society actors are given voice in news reporting (and to some extent also in guest commentaries), their frames are not necessarily adopted in opinion-making articles, in particular in the economic/conservative publications. This finding therefore also explains why our analysis suggests a fairly balanced debate in Germany overall, despite the impression that public opinion in general appears to have been highly sceptical of TTIP, focusing on contentious issues such as the role of ISDS courts in the TTIP proposals.

Table 7. Most frequent frames in main body of articles (by newspaper orientation).

	Threat to regulatory standards	Horse race/ difficult negotiation	Lack of transparency	Corporations over citizens
Economic	82%	87%	50%	18%
Liberal	82%	77%	68%	50%
Left	71%	79%	50%	58%

This is further underlined by the very different conclusions drawn in the opinion-making articles of left-alternative publications such as the German taz or the Guardian in the UK. There, the lack of transparency in the negotiations is taken as part and parcel of the risks that TTIP represents in terms of European standards. A middle-ground position is taken by the liberal/mainstream SZ, which highlights the potential economic benefit of the agreement, but also emphasizes that TTIP can only succeed if the process is based on transparency, informed public debate and broad public support. As the SZ’s editorialist Heribert Prantl writes, the supporters of the agreement have to make sure that ‘the critical points can be discussed with knowledge and substance’ instead of accusing opponents of spreading ‘conspiracy theories’ (SZ, Das Geheimstabkommen, 2 May 2016). But even balanced arguments of this kind strongly suggest frame resonance: despite a commitment to the potential benefits of TTIP, it is made clear that the concerns raised by opponents have to be taken seriously.

From a comparative perspective, it is worth noting that despite certain parallels across countries, the frames employed in the twelve publications also reveal clear country-specific patterns. For instance, the threat of private ISDS courts, which was one of the most salient ways of framing TTIP in the German debate, was virtually absent in the Spanish, French and British media.

Overall, our analysis reveals that the frames advanced on the Stop TTIP campaign’s Facebook page resonated in the mass media, and they did so not only in publications on the left of the political spectrum, but – to varying degrees – also in liberal/mainstream and in economic/conservative outlets. Similar to the social media material, the mass media material displays a strong emphasis on challenges to European standards and a lack of transparency in the negotiation process. This critical perspective on TTIP points to the crucial role of the Greenpeace leaks in raising awareness for and, indeed, initiating deliberation on contentious aspects of TTIP that may otherwise not have surfaced in mainstream media discourse at all, at least not without the framing efforts of the Stop TTIP campaign.

7. Concluding discussion: social movement framing and communicative power generation

Executive actors have traditionally dominated EU news at the national level (Koopmans and Statham 2010), which has largely left the administrative power held by EU institutions unchecked. Through a combination of deliberative and agonistic theory, we have argued that the generation of communicative power beyond national borders is an indispensable ingredient in the democratic legitimacy of EU decision making, not least in the field of trade policy. The ECI is an interesting tool in this context, since it encourages civil-society actors to engage in transnational collective action beyond the ‘Brussels bubble’ (Sánchez Salgado and Demidov 2018; Bouza and Oleart 2018). The role of agency is central in this context, highlighting why framing and frame resonance are the key ingredients in the generation of communicative power, in particular as regards the voice of civil society actors who often lack access to the mass media and therefore need to rely on social media to advance their ideas in the public sphere. The episode of the Greenpeace leaks serves as a useful illustration of how communicative power was generated by activists in the context of the TTIP negotiations.

As our empirical analysis has shown, the main frames developed by the Stop TTIP campaign have resonated in the mass media material, where a similar image of TTIP as a threat to European standards has been constructed. Despite the degree of frame resonance achieved in the wake of the Greenpeace leaks, it is worth noting that Stop TTIP had used the frames identified in this analysis already long before, making TTIP appear – as argued in some of the liberal and even financial newspapers – as a confirmation of the darkest fears of TTIP opponents.

In their roles as gatekeepers and watchdogs, journalists are evidently not passive transmitters of the frames propagated by critical civil-society organizations. Consequently, the frames used in news reporting, analysis and opinion-making pieces originate from an abundance of sources. Journalistic framing obviously also reflects ‘general political tendencies’ (Hallin and Mancini 2004, 27), which explains why the economic opportunities of TTIP – which Stop TTIP had explicitly denied – are emphasized at least in parts of the mass media material, in news articles as well as in analytical and opinion pieces. However, it has to be noted that while the potential economic benefits of transatlantic free trade played a considerable role in the German discourse, the ‘jobs and growth’ frame was virtually absent in France, Spain and the UK. This may be attributable to the observation that frames such as those advocated by Stop TTIP were not present in the mainstream media before May 2016 (for the Spanish case, see Bouza and Oleart 2018). In part, this may be because the editorial opinion of mainstream news outlets such as Le Monde and EL PAÍS was largely in favor of TTIP.

The results of our research confirm that frame resonance in mass media outlets has contributed to the generation of communicative power about TTIP. Therefore, we may ask to what extent social media as a campaigning tool play an extraordinary, possibly even a decisive role for social movements such as the transnational Stop TTIP campaign. As discussed, it would be problematic to assume a simple causal relationship between social media framing and effects on mass media discourse. Social media activism has only been one element in the mobilization of opposition against TTIP, and arguably not even the most visible one in terms of mass media coverage, for instance when compared to the visibility of the many Europe-wide demonstrations against TTIP. Nonetheless, the media content analysis clearly shows that frames employed by Stop TTIP on Facebook did resonate in the mass media coverage of the Greenpeace leaks. The mass circulation of these frames in mainstream media may very well represent a turning point in the debate on TTIP, suggesting the transnational generation of communicative power, at least in the short episode analysed here.

From a normative perspective, the simultaneous agonistic and deliberative debate on TTIP in the wake of the Greenpeace leaks is good news for the democratic legitimation of EU decision making. The Greenpeace leaks thus provided a backdrop against which frames already used by critical civil-society organizations were allowed to resonate in the public sphere at large, thus making the debate on TTIP more inclusive in terms of issues, ideas and participants.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. Formal initiatives are those that passed the legal admissibility test of the European Commission and were registered as official initiatives on the Commission’s website. By contrast, informal initiatives are those that – such as most notably ‘Stop TTIP/CETA’ – were refused formal registration by the European Commission, but carried out a signature-collection campaign in the EU’s member states nonetheless.

2. Along these lines, Gamson and Modigliani (1989, 143) saw a frame as a ‘central organizing idea or story line that provides meaning to an unfolding strip of events, weaving a connection among them. The frame suggests what the controversy is about, the essence of the issue’.

3. In its social media campaign, Stop TTIP tends to treat TTIP and the EU-Canadian ‘Comprehensive Economic and Trade Agreement’ (CETA) as two sides of the same coin, which is underlined by the fact that CETA is often constructed as a ‘Trojan horse’ to introduce TTIP through the backdoor. A physical Trojan horse is also frequently present at demonstrations against both TTIP and CETA.

4. The label ‘horse race’ was chosen as a reference to the journalistic practice of covering elections from the perspective of the different candidates chances of winning rather than from the perspective of their substantive proposals.

5. The discussion of the findings of the media content analysis will focus primarily on the most frequent frames found in the sampled articles. A complete overview of all the frames found in the material is included in the appendices.

6. ‘Au moment du lancement des négociations, une étude publiée par le centre for economic policy research (CEPR) de londres souligne que l’économie européenne dans son ensemble générerait 120 milliards d’euros supplémentaires grâce au ttip.’

Appendices

Appendix 1. Frames in the titles/subtitles of the sampled articles.

		France				Spain				UK				Germany
Frame	Total	LeM	Éch	Lib	Total	País	Eco	Dia	Total	FT	BBC	Gua	Total	FAZ	SZ	taz	Total
Lack of transparency	28	3	4	2	9	4	3	6	13	0	1	2	3	0	2	1	3
‘Horse race’/ Difficult negotiation	21	2	3	0	5	3	1	2	6	0	1	3	4	2	2	2	6
Threat to regulatory standards	21	1	1	0	2	3	4	3	10	0	1	2	3	1	4	1	6
Lack of support/ mass opposition	3	0	0	0	0	1	0	1	2	0	0	0	0	1	0	0	1
Private justice	2	0	0	0	0	0	0	0	0	0	0	0	0	0	1	1	2
Business as usual	2	0	0	0	0	0	0	0	0	0	0	0	0	1	0	1	2
Greenpeace leaks as obstacle	2	0	0	0	0	0	0	0	0	0	0	0	0	1	1	0	2
Corporations vs. citizens	0	0	0	0	0	2	1	1	4	0	0	0	0	0	0	0	0
Unidentifiable	4	0	0	0	0	0	0	0	0	0	0	0	0	1	1	2	4
Miscellaneous	4	0	0	0	0	0	0	0	0	1	0	0	1	3	0	1	4

Appendix 2.

Frames in Main Body of Articles (whole sample).

	FAZ	SZ	taz	Total	FT	BBC	Gua	Total	Éch	LeM	Lib	Total	Eco	País	dia	Total
Horse race	7	5	5	17	0	2	4	6	7	5	1	13	4	5	9	18
Regulatory standards	6	6	7	19	1	2	4	7	7	5	2	14	4	5	4	13
Lack of transparency	4	7	3	14	0	1	2	3	4	2	0	6	3	5	7	15
Corporations vs citizens	0	5	4	9	0	2	3	5	2	0	2	4	2	4	4	10
Economic opportunity	3	3	4	10	0	2	0	2	2	0	0	2	0	2	1	3
Business as usual	4	6	4	14	0	0	0	0	0	0	2	2	0	0	0	0
Private justice	2	4	2	8	0	0	0	0	0	1	0	1	0	1	0	1
Lack of support	3	0	4	7	0	0	0	0	0	0	0	0	0	1	1	2
Obstacle	2	3	2	7	0	0	0	0	0	0	0	0	0	0	0	0
Other threats	1	2	2	5	0	0	1	1	0	0	0	0	0	0	0	0
Global leader	3	1	1	5	0	0	0	0	0	0	0	0	0	0	0	0
Other opportunity	0	1	0	1	0	0	0	0	0	0	0	0	0	0	0	0