Economic Citizenship in Online Financial News : Affect and Argumentation in Eurocrisis News Coverage

This paper evaluates the practices of online news media in representing citizens in the midst of the Eurozone crisis from the perspective of civic participation. Our special focus is on the interplay of argumentation and affect as crucial sources of political action. The empirical analysis examines affective and argumentative framing in two multimodal story types, captioned photo galleries and video reviews, and evaluates their capacity facilitating or hindering political agency. Video reviews construct news narratives that incorporate both affective and argumentative elements. Photo galleries, by contrast, emphasize affect at the cost of argumentation and voice resulting in alienation and silencing. The analysis gives grounds to suggest that dramatized visual news narratives about the financial crisis echo discourses of war on terror and unspecific threat. Such discourses legitimize pre-emptive action (austerity measures, police force), which, in effect, participate in producing what it is supposed to fight (economic inequality, social unrest, violence).


Introduction
The continuing financial turmoil has had an accelerating effect on the lives and possibilities of ordinary people in many ways, while the need to understand, explain and participate in economic discourses has become more pressing.In web-based media, discourses of financial crises are extensively realized in forms of multimodal texts that persuade audiences through selections of facts, analyses of processes and trends, comments and interviews as well as depictions of everyday reality.Online news packages, such as crisis «guides», timelines and chronicles, also include representations of citizens in forms of video interviews, photographs, and citations.These representations are to a significant extent included in the reports of street-level civic activism, such as demonstrations, protests, and violent clashes, prompting the question of the positions available for ordinary citizens in the face of the economic crisis and the role of the media in mediating citizens' voices and facilitating political agency.
The media plays a central role in making sense of economic phenomena.Specifically, media representations create the conditions of possibility for moral and political action, enacting possible forms of agency, which may or may not be followed up by media publics (Chouliaraki, 2008: 832).In this paper, we focus especially on the mediated practices and preconditions of economic citizenship, i.e. the roles and positions that members of the public assume and that are prescribed to them in mediated economic discourses.The aim of the paper is to evaluate how the different (audio)visual and verbal framings of the crisis in the Eurozone may guide and constrain the potential forms and practices of economic citizenship: What kinds of frames are evoked, and what kind of roles, identities and possibilities do they propose and prescribe for members of the public as economic citizens?
In web-based news media, the identity and personal voice of referents as well as the emotional intensity of depicted situations are conveyed by verbal as well as (audio)visual means.The use of images and audiovisual material within the financial crisis news relay requires careful attention and consideration.In addition to «true-to-life» documentation and inbuilt point-of-view, lens based images, and especially photographs (Sontag, 2003: 19, 21, 23), are capable of triggering our mental stocks of images, the collection of «memory freezeframes».Moreover, by chasing and depicting dramatic moments, visual news coverage activates emotions effectively.
The paper examines representations of citizenship especially in two multimodal story types: captioned photo galleries and video reviews.We analyze media frames and the way they tend to guide the perceptions and interpretations of members of the public as economic citizens along two different dimensions, one related to identification and the level of individual voice and reasoning (argumentative framing), and the other related to the emotional intensity of representation (affective framing); and introduce an integrative analytical tool that connects these dimensions.From a variety of visual themes that define and prescribe economic citizenship in the financial crisis imagery, we focus on the coverage of demonstrations, protests and riots, which emerge as prominent themes in addressing the citizen's role in the crisis.The analytical framework, which we call the Affect and Argument Map, is tested and illustrated through the empirical examples of video reviews and photo galleries related to online news about the financial crisis.
The rest of the paper is organized as follows.Section 2 introduces the theoretical framework of the study.Section 3 describes our analytical tool, developed and applied for evaluating emotional intensity as well as identification and personal voice in multimodal texts.Section 4 introduces research material and the analytical process of the study.Section 5 describes and discusses the central findings of the research, and finally, section 6 presents our conclusions of the current state of the study.

Theoretical framework: affect, argumentation, and framing
In current politically oriented media and discourse studies it is broadly acknowledged that both «reason» and «emotion», i.e. both logico-discursive and affective signification constitute important sources of moral and political action and that they are, in effect, indissociable in such discursive work (ten Bos & Willmott, 2001;Laclau, 2005;Massumi, 2010).While logico-discursive reasoning is generally regarded as the nucleus for democratic participation (e.g.Habermas, 1962Habermas, /1989)), studies on cognitive and affective responses to media messages suggest that emotional responses also play a role in public opinion and political judgment (Cho et al., 2003: 324).Silverstone (2007) incorporates this interplay in his notion of «mediapolis», by which he refers to the mediated public space of appearance, where we appear at others and others at us: «in the mediapolis, the issue is not just the rationality of human communication, but a recognition that communication is grounded in a feeling for the world, and in the condition of being in the world among others» (Silverstone, 2007: 43).Thus, the media may enhance political participation or agency through facilitating the processes of both identification (creating a sense of relatedness, connectedness) and affect (creating the sense of emancipation, freedom, the possibility of change).In this study, we examine the interplay of affect and argumentation in terms of media framing.

Media framing
Media frame can generally be regarded as a «central organizing idea or story line that provides meaning to an unfolding strip of events» (Gamson & Modigliani, 1987: 143).Frames are persistent patterns of selection, emphasis, and exclusion where certain aspects of reality are promoted over alternatives.The empirical analysis of media framing rests on the idea that media texts, or their particular attributes, can be interpreted as pointing towards certain meaning systems, thereby priming audience towards certain interpretations (Griffin, 2004: 384).Attributes, which are analyzed as cues or framing devices, vary according to the mode of communication.Thus, media frames operate by persuading audiences to interpret news by emphasizing certain ideational and interpersonal markers, using visual, verbal and auditory resources.
Visual framing may be understood as ways in which images or their attributes (e.g.camera angle, composition, depicted objects) work as framing devices through referring to culturally available signifying practices and meanings.Correspondingly, auditory framing and verbal framing refer to ways in which sounds (e.g. a piece of music, a tone of voice, a whistle of tear gas grenade) and linguistic choices (e.g.vocabulary, person, tense) are used to evoke such practices and domains.
It has been assessed that, compared with verbal messaging, part of the power of images as framing tools derives from their less intrusive nature, as well as the general impression of lens-based images referring closer to reality (Rodriguez & Dimitrova, 2011: 50).Following Messaris and Abraham (2003), visuals, and particularly lens-based images, have three characteristics that create both challenges and opportunities to their potential to frame news: 1. the analogical quality of images, 2. the indexicality of images, and 3. the lack of an explicit propositional syntax in images.In addition, images, and as Sontag claims specifically photographic images (2003: 19), have the ability to create a high level of emotional intensity and capacity.
These characteristics make the lens-based images, effective tools for framing, especially as it has been demonstrated that visual framing does not seem to affect the evaluation of journalistic objectivity by the audience (Brantner et al., 2011: 533).In addition, it has been argued that visual modes of communication, especially metaphor-rich «narrative visualizations» (Fry, Wilson & Overby, 2013) may have heightened capacity to influence economic behavior through evoking emotional resonance and appealing to intuitive reasoning.Therefore, the use of images in signifying complex and controversial topics such as the financial crisis and its implications should be carefully considered, and ideally, verbal content and visual imagery should be analyzed in conjunction.

Affective fact, simulation and meaning contamination
The relationship between affect and democracy is far from simple, and practices of expressing and representing affect in the media vary in terms of their democratic and moral potential.It has been argued that representations that merely dramatize or visualize the antagonistic, exclusive or unfair nature of social relations make little intervention in the world but may instead work to restate or reproduce the current social order.In order to facilitate democratic processes and the possibility of change, cultural or media forms would therefore need to reflect also the inherent contingency of social relations and structures, and the potentially «terrifying» experience of freedom and emancipation resulting from this contingency (Laclau, 1996;Gilbert, 2011).
The exhilarating fear inherent to freedom differs drastically from another kind of terror -the paralyzing, discouraging fear connected to the threat of arbitrary injustice and violence described by Massumi. Massumi´s (2005, 2010) idea of affective fact refers to the ways in which the creation of affect by the media precedes and intertwines with argumentative and empirically based reasoning in the creation of «facts», which may, once established, be uncritically used to justify pre-emptive action.It is argued that the media, and especially the rhetoric of neoconservatism in the media, evoke emotions first, based on fear, disruption, repetition, and «the looming uncertainty of illdefined threat» (Massumi, 2005: 8).After creating affect, and only then, media produces content and meaning for it (Massumi, 2005;Terranova, 2007).
Following Massumi, the mechanisms and rhetoric of affective fact are based on alien, unknown threat, and the bending of temporal dimensions.Thus, affective fact is especially about a state of alarm and preemption.It is concerned with future anterior, reversal indexicality, where the smoke is no more an indexical sign, a prediction of the fire.Instead, the smoke causes the fire (Massumi, 2005: 8).In other words, as Terranova (2007: 133-134) defines a mechanism of passage between affective and empirical facts: «we would be dealing, then, with a dispositif of power that grasps knowledge not as the result of abstract logical games affecting the sphere of truth, but as the secondary result of a certain hold on the body».Thus, it can be argued that within contemporary media tactics, the cognitive perception and the field of knowledge are frequently enclosed with strategic intervention in a public consciousness with affective facts.Consequently, one of the objectives of our research is to analyze the passage between affective and empirical facts within representations of individuals as economic citizens.Massumi (2010: 84) notes that «the affective reality of threat is contagious», meaning that once the affective fact of fear has been created, it works to signify objects and people as threatening with little regard to their actual or empirical ties to the (imagined) source of threat.This research discusses the possibility of meaning contamination on the level of narrative structure, conceptualizing it as internal resonance between individual visuals and other semiotic resources within news narratives.Our aim is to discuss how and to what extent representations may derive their meanings internally from the narrative whole that they are a part of.
Visuals, and especially loosely captioned still photographs, as semiotic resources in the news media are inexplicit and flexible in meaning, and may thus easily work as floating signifiers, losing part of their characteristic indexical gravitation on the «real» referent and gliding toward the state of simulation (Baudrillard, 1994).Simulation refers to mediation without the referent, it is «the generation by models of a real without origin or reality: a hyperreal» (Baudrillard, 1994: 1).In addition, simulation may be understood as «representation of already existing spectacles of mass fascination» (Chouliaraki, 2008: 833).Thus the concept draws attention to the ways in which, in contemporary mass media, signs may refer to existing cultural narratives and meanings more than to any real referents.

Affect and Argument Map as an analytical framework
These theoretical premises and initial observations from the empirical material guided us in the construction of our analytical framework, titled the Affect and Argument Map (Figure 1).Empirical phenomena that we wanted to subject to closer examination included especially the dominance of affective, dramatic imagery in the photo galleries of demonstrations, the tension between anonymous and identified referents, and the apparent differences in tone between photo slide shows and video reviews referring to similar events.We analyze the representations of citizens in financial news discourse in terms of affective framing and argumentative framing; analytical devices for describing these dimensions were gathered especially from the literature of multimodal discourse analysis and visual framing (Fahmy, 2004;O'Halloran, 2006;Kress & van Leeuwen, 2006;Bock, 2012).
Affective framing in this study refers to the ways in which visual, verbal, and auditory attributes evoke emotional states, practices, or atmospheres in representations of economic citizens.We analyze cues for emotional frames of each (audio)visual unit of texts on three levels based on the metafunctional principle (Halliday, 1994;Kress & Van Leeuwen, 2006).First, interpersonal emotion markers that work as cues for the relationship between the viewer and the depicted referent include contact (gaze), attitude/subordination (camera angle), distance, orientation, as well as verbal and auditive cues.Second, ideational emotion markers include cues to the representation of emotions: the emotional state of the represented citizens, signaled by facial expression, posture, movement, verbal and auditive cues, and the emotional atmosphere of the situation/text as a whole, signaled by verbal cues, auditive cues (e.g.music), and certain ideational motifs and effects (e.g.fire, smoke, clothing, gear).Third, we assess the interplay of visual components and other semiotic resources of each (audio)visual unit of texts, i.e. visual and intersemiotic compositions that relate interpersonal and ideational meanings to each other (e.g.salience, frames).Argumentative framing refers to the ways in which the attributes of multimodal news texts point to culturally available practices and forms of identification, such as socio-cultural roles and identities, or symbolic figures or characters.It includes the ways in which citizens are identified or left anonymous in media texts, as well as the presence (or absence) and quality of the voice.Argumentative framing is examined by assessing the amount and quality of identity markers shown or verbally indicated in a representation.These include personal and demographic markers (e.g.name, gender, age) as well as different indicators of personal voice: the audiovisual presence of citizens, citations, or visual presence of banderoles or other signs.

Research material and analytical process
The study is part of a larger project examining the signification of finance in online environments.For the project, we have compiled a dataset comprising extensive news packages such as virtual timelines of the financial crisis in the Eurozone during 2010-2013 in several news media.The videos and captioned picture galleries analyzed in this paper appear as parts of such overview presentations.
The analytical process was begun by identifying recurring visual themes of representing citizens in the complete dataset.These themes included, for example: Protestors clash with police; Pedestrians in front of important buildings; Demonstrators with banderoles; Sale or shop closedown; People in the employment office; Traders in front of curves; People at cash machines; People interviewed in the street; Supporters of parties and politicians; Voters at the polling station.Of these themes, we chose to focus on representations of demonstrations and clashes, partly due to their prominence in our dataset, and partly because of their notably strong emotional atmosphere.
News material analyzed in this article concerns demonstrations and protests against austerity measures, government-imposed wage cuts and tax increases in Athens.These measures were based on the requirements by foreign lenders (i.e.European Union, European Central Bank, and International Monetary Fund) in order to qualify Greece for bailout packages aiming to ease the economic crisis in the country and to prevent default.Articles, video reviews and photo slide shows illustrate events at the time of strikes in Athens in June and October 2011.In these protests, peaceful demonstrations eventually turned violent, led by various hard-line interest groups clashing with riot policemen, vandalizing the city.The final material includes four videos and four sets of photo galleries addressing demonstrations in Athens published in the online versions of The Guardian, The New York Times, The Washington Post and cbs Evening News (Table 1).A particular challenge in the analysis of multimodal texts is to account for the interplay of visual, written, and auditory attributes.In this study, analysis is a hermeneutical process where the examination of different attributes alternates with their interpretation in terms of discursive functions, and finally with their assessment with reference to cultural narratives and meanings.Another challenge in the analysis of complex (audio)visual texts concerns the inclusion of context.This study considers two contextual layers: first, the textual context, referring to the immediate connected or neighboring elements (such as other pictures in a slide show, headlines), which are likely to trigger internal resonance; and second, the wider cultural contexts that are evoked and made relevant.
The analysis proceeded as follows.We analyzed representations of citizens distinctively for each individual (captioned) photograph and each distinguishable shot of video, in terms of affective framing and argumentative framing.We used the list of attributes described above as a checklist while aiming at a holistic interpretation that would take into consideration the interplay of different markers.Evaluations of the emotional intensity and the level of argumentation and voice were placed on the analytical map and the representations of citizens were further interpreted with reference to cultural meanings and frames.At the second phase, we assessed the presentation (slide show or video) as a whole: its organization and narrative features.At this phase, we also reflected on the possibility and effects of meaning contamination.In the following sections, we present and discuss findings according to these two stages.

Placing representations of citizens on the map
After analyzing representations of citizens in each individual captioned photograph and shot of video in our material, we placed them on the Affect and Argument Map.The result is pictured in Figure 2. Our analysis shows that, overall, the level of argumentative voice remains relatively low throughout the sample of demonstration coverage that we analyzed.The vast majority of representations includes only a minimal amount of identity markers, such as visually discernible gender and age, leaving citizens anonymous, decontextualized, and silenced.The level of emotional intensity, by contrast, varies considerably more.The dimensions of the map create four fields or combinations of affective and argumentative framing.

Floating referent + high affectivity -citizen as affective character
The majority of the (audio)visual representations of citizens in our material falls in the category characterized by high level of affectivity and low level of identification and voice.The prototypical example is the masked protestor depicted in a highly affective setting, marked particularly by ideational cues such as fire, smoke, and riot gear.In such imagery, we argue, the figure of the protestor does not appear as an identifiable or relatable person who would evoke a sense of connection (Silverstone, 2007) but rather as an affectively tuned symbol or index.
Emotional intensity, in our material, is most often constructed through ideational markers that cue into the emotionally charged atmosphere of the situation, and only marginally through interpersonal means of creating contact or proximity between the viewer and the referent.In many pictures, tight cropping excludes all signs of the ordinary or the everyday, tuning emotional intensity to the extreme.With reference to identification, in the most extreme cases there are no identity markers visible or audible; a citizen may, for example, appear as a mere silhouette against fire or smoke.In other cases, there are visual or verbal elements that work as weak indications of the possibility -if not the content -of the voice or message, for example untranslated, uncommented banderoles or placards, or a brief reference to the context of the protests in a written caption.
The «citizen as affective character» category is extremely popular especially in all four sets of still photographs that we analyzed.This observation resonates with a long-standing critique of the practice of «aesthetic dramatization», resulting in the (over)emphasis of the emotional affectivity above the ideational relevance in news photographs (Schwartz, 1992;Caujolle, 1995in Salo, 2000: 28).Video reviews in our material include relatively less anonymous-affective representation; however, the video review of the cbs Evening News marks an exception, with its footage of the demonstrations in Athens falling entirely in this category.This video runs footage exclusively from the violent phase of the protests, with protestors throwing stones and police firing teargas, as background imagery without any reference to the motives or goals of the demonstration.
Our analysis suggests that cultural narratives through which events are told in such imagery -and furthermore through which they are interpreted -include elements from discourses of war, anarchy and chaos as well as simulations of war, anarchy and chaos through fiction and gamification.Preliminary observations on the social reinterpretation of this news material imply that in certain online communities, visual news coverage on demonstrations and protests during the Eurocrisis is in fact perceived within science-fiction, warrior or military saga frames (see e.g.http://gamingtrend.com/forums/offtopic/riot-police-gear).

Floating referent + moderateness -citizen as mundane character
Representations of citizens falling in this category include the images and footage of unidentified (peaceful) demonstrators and bypassers especially in the New York Times and Guardian video reviews.They depict ordinary people participating in familiar, institutionalized forms of political action: as demonstrators and strikers, forming a collective public.When these images represent large masses marching with banderoles, they evoke to some degree the cultural frames of mass power and uprising -however in a diluted form, since the message is missing.Another image type depicts members of the public confronting the financial turmoil at everyday errands, e.g. at cash machines, gas stations and employment offices.In terms of cultural framing, these pictures may be read as signifying especially the ways in which the ordinary has become extraordinary due to the financial turmoil and the accompanying political activity and unrest: cash is not available, the airport is empty, and the streets are filled with garbage.
Representations in this category lack the most blatant ideational emotion markers and effects; in addition, emotional intensity is leveled through the visual and auditory cues of ordinary everyday life.Consequently, these representations may be considered less distancing and alienating than those in the previous category.However, due to the low level of argumentation, they are also likely to fall short in terms of facilitating an active political and moral engagement.Members of the public are typically depicted in this category as generic representatives of civic activism or as indexes of the everyday effects of the crisis.As an image type in the news media these kinds of visuals may be even seen as falling somewhere between news photographs and illustrative symbolic images (Salo, 2000: 15-16).

Identifiable referent + moderateness -citizen as deliberative person
Notably, the material that we analyzed does not include representations classified under high and very high level of identification.Such representations, according to our criteria, would mean that citizens are named or otherwise described, contextualized and get to express their arguments, opinions, or emotions at length.However, the material includes some representations with moderate or relatively high level of argumentation and voice, especially in the video reviews by The Guardian and The New York Times.
In the street interviews on these videos, citizens are given the possibility to articulate their thoughts and feelings concerning the immediate situation, as well as the economic turmoil more broadly.In practice, screen and audio time given to citizens is limited, which restricts an in-depth deliberation of causes, consequences and possible solutions, and directs argumentation towards brief, responsive overview comments.However, even brief citizen comments may express relatively multifaceted arguments in condensed and organized form, as the following example illustrates: Example G/Video/2-4.Video review Guardian, Male demonstrator interviewed in the midst of the demonstration Video: Footage alternates between close shots of the referent speaking and long, wide, the slightly high-angle overall shots of Parliament square full of demonstrators with banderoles.Audio: The background bustle of crowds, music playing from loudspeakers, creating a calm, slightly melancholy atmosphere; male referent speaks Greek in a calm, determinate tone.«There is no way that we will stop this battle until this government falls, which is one of our demands.Our second demand is an end to this monopolistic policy, which has created all these problems and creates a dead end for workers.Our message is: we will battle till the end.»This example also illustrates the significance of auditive emotion markers in communicating the overall atmosphere of the situation.Especially in comparison with the mute -literally and figuratively -the messages of photographs, video footage with ambience sound, the bustle of crowds, and music, inevitably evokes the glimpses of the ordinary or the everyday, toning down the mute threat and emotional intensity that characterizes dramatic photographs.
In this category, citizens appear as deliberative actors capable of forming and expressing conceptual insights and opinions.Such representation, we argue, is likely to facilitate identification with the referents and evoke the processes of reasoning, thereby creating preconditions for political engagement.More specifically, they suggest political agency as a possibility, which may or may not be taken upon by the audience.
There are, notably, no still photographs in this category.There is of course no reason why captioned photographs could not display this combination, since identity and voice could in principle very well be communicated in the caption.The lack of argumentation and personal voice in the galleries of still photographs is thus not inherent to the genre; more likely, it reflects, at least partly, the conventionalized journalistic practices and processes, e.g. of separating images from their makers and the context of their making (Bock, 2012).

Identifiable referent + high affectivity -Citizen as compassionate person
No representation in our material depicts the combination of high level of argumentation and high affectivity.Principally, this category could in our context mean representations in which a named or contextualized person uses his/her voice in a heightened emotional state or setting.Ideally, such representations could potentially evoke a sense of compassion and connection by the viewer and thereby facilitate the emotional engagement.

Summary of the map
The levels of affectivity and argumentation within Eurozone crisis news seem to vary depending on applied semiotic resources.It appears that semiotic resources and narrative patterns used in audiovisual video reviews offer more versatile markers and cues for representing actors, the scenes of action, and especially articulated motivations and arguments of members of the public, when compared with captioned photo galleries.Hence, video reviews seem capable of representing coherent perceptions and stories with diversity at the levels of personal voice and reasoning as well as the atmosphere of the situations depicted.By contrast, photo slide shows often seem incapable of depicting referents as identified individuals with a personal voice.On the contrary, it seems that such visual news coverage downplays, ignores or is unwilling to emphasize individual characteristics of referents represented in the narrative, or giving them an opportunity to demonstrate their arguments and reasoning.Photo slide shows tend to focus on action, combat and confrontation, framing in only the abnormalities of «crisis reality» and thereby creating highly affective framings to impact and astonish viewers.

Narrative features and practices in representing citizens
Thus, it can be argued that the way in which the more aggressive and apparently violent rioters are equipped (face masks, tear gas masks, hoods, sticks) and represented (anonymously in surreal, emotionally intensive, affective imagery) tends to alienate them from mundane everyday life.Such anonymity, facelessness, and silence can naturally be considered, to a certain extent, the protestors' own choice.Nevertheless, we argue that the way in which the media choose to focus on this kind of (self-)representation, emphasizing the level of affect and ignoring or fully neglecting opportunities to include the personal voice and reasoning, has problematic implications.

Anonymity, muteness, alienation
All the photo galleries and one of the video reviews analyzed for this study represent citizens mainly as anonymous and mute.We argue that this kind of representation emphasizes the visual idea of demonstrations over specific demonstrations with particular arguments and identified voices.Images of demonstrators shouting slogans, mouths wide open, but silent, thus represent the idea of demonstrators with banderoles as symbols, carrying out no perceivable meaning (without translations and transcriptions in text).Citizens are muted and disguised in the way that some rioters prefer to perform in these setups, anonymously without revealing their motives and backgrounds.Thus, photo slide shows even seem to use auditive space, explicitly and implicitly, literally and symbolically, as an amplifier for emotional intensity and affective framing.Lacking the meaning potential of the auditive mode, photography derives its strength by cropping all sounds from the scenery, including the transcriptions of spoken argumentation in captions, and rebuilding the setting depicted based on the visual.
The photo gallery of The Guardian serves as an example.In addition to the headline («Violence erupts as Greece strike begins»), there is only one sentence (in the ingress) in the whole slide show presentation, which refers to the motives of action of the referents: «Anger over new austerity measures and layoffs erupts into clashes outside parliament as two-day general strike begins».This is an overly simplified verbal summary for the variety of actors and groups visible in the photo gallery.Bundling up all actors of a day like this is not only questionable journalism but also unpredictable.It is questionable especially in terms of the rights of the referent for his and her own identity, including visual and verbal markers, from physical appearance to his or her ideas and argumentative reasoning.Furthermore, it is unpredictable especially in terms of the audience as participants, the viewers and their perceptions and interpretations of the flows of events based on such representations.
Our analysis indicates that the relatively open meaning structure of photo slide shows may increase the semantic weight or impact of overly affective elements within the presentation, thereby enhancing the possibility of meaning contamination, the alienation of floating referents, and oversimplification of portrayed individuals and events.For example, particular photographs within a slide show may mediate emotionally decisive moments and, thus, create affective thematic cues and impacts, which guide interpretations also of the neighbouring images within the photo gallery as a whole.Whether intentional or unintentional, this kind of visual news coverage may have a tendency of producing particular «desktop editorial realities».

(Im)potentializing citizens
The preceding discussion gives reason to suggest that the journalistic choice to focus on affective representations of anonymous demonstrators, together with the flexible and contagious meaning structure of multimodal story types, is likely to affect also the representations of those citizens and demonstrators who would be more than willing to express their views.As demonstrators shouting possibly well thought-out slogans and carrying written banderoles are represented basically the same way as anonymous hooded rioters, without any level of personal voice, there is a risk that they, too, are distanced and alienated in a similar way.
From the perspective of democratic participation and considering the role of media in facilitating it, such representation does injustice to demonstrators and undermines their freedom to be heard (Silverstone, 2007: 40).Most likely, the primary reason for citizens to go out demonstrating would be the possibility to have their say and to be heard, read, and listened to.Furthermore, the motives, contents and nature of their actions during a demonstration are likely to have been very different from the ones that inspire more violent protestors and rioters.However, this distinction between the motives, arguments, opinions, and nature of action between different groups is not apparent, noticeable, or readable in the collections of photographs that we analyzed.
It may thus be argued that, especially, photo slide shows and galleries dominated by anonymous, affective characters tend to incapacitate members of the public and deprive them of political potential, which is ultimately based on the possibility of real, effective communication that includes the right of being heard.Such muted representations invalidate referents, including those who would most possibly like to have their say, compressing and framing them as floating referents, generalized symbols, illustrations or ideas of protesting citizens rather than as individuals, groups and collectives with relevant opinions and arguments.
To conclude in a more optimistic note, video reviews in our material seem to illustrate the possibility that meaning contamination in Eurozone crisis discourses may also work in the opposite direction of potentialization and personification.There is a level of internal resonance within video reports that creates a narrative fit between identified, deliberative, expressive citizens with a personal voice, and the other, anonymous referents.As a preliminary notion, it seems that comments and statements depicted and represented in the interviews by referents as well as their unprompted verbal messages from posters, banderoles and slogans may have a tendency of projecting potentiality and expectations for similar kinds of thinking, behavior and activity towards other referents in the representation as well.Further research and analysis of this kind of pre-personification and potentialization is one of the future objectives of this study.

Conclusion
In this paper, we set out to study media representations as the conditions of possibility for moral and political action.A special aim was to evaluate how different multimodal narrative and representational practices guide and constrain the potential forms and practices of economic citizenship in the context of Eurozone crisis discourses.
In current politically oriented media and discourse studies it is broadly acknowledged that both «reason» and «emotion», i.e. both logico-discursive and affective signification constitute important sources of moral and political action and that they are, in effect, indissociable in such discursive work (ten Bos & Willmott, 2001;Laclau, 2005;Massumi, 2010).Gilbert (2011), in an insightful essay on what democracy «feels like», connects democracy to an exciting, emancipatory, and even terrifying sense of potential, a realization that social relations are contingent and may actually be subject to dislocation and transformation through collective action.
However, this study gives reason to question the capability of certain established practices of signification in multimodal online journalism to evoke such political potential.Our analysis of the representation of demonstrations in online financial crisis news indicates that affective intensity and argumentative depth do not often meet in a productive way in multimodal representations of economic and political activity.Quite the contrary, highly affective representations, epitomized in our study by the archetypal still photograph depicting anonymous rioters, tend to be so strongly disconnected from everyday realities, concrete events, and political reasoning, that they incline towards simulation and the creation of an empty sense of threat and chaos (Baudrillard, 1994;Massumi, 2010).By contrast, representations of citizens with a more personal voice and argumentation, exemplified in this study especially by video interviews, frequently miss their chance to enhance an emotional engagement.
Our analysis and findings indicate that the analytical tool that we developed for this study, the Affect and Argument Map, is helpful in creating structured and nuanced understandings of the interplay of affective and argumentative framing in representations of social actors.As a special advantage, the tool is flexibly adjustable for the analysis of linguistic, (audio)visual as well as multimodal texts.The next challenge developing the tool will be to relate the interplay of affect and argumentation in a more systematic and theoretically justified manner to the question of the moral power of media and the potential of media texts to facilitate ethical and political action.
The material analyzed in this study focuses on strikes, demonstrations and protests in the face of austerity measures and represents, as such, a limited sample of the practices of representing economic citizenship in the online news about the financial crisis.The theme may be considered relevant to our study because of its prominence, and because it offers insights into the intersection of traditional means of public participation and current challenges of economic citizenship in the globally interconnected world.However, examination of additional visual and textual themes and materials is needed to give further indication of the usefulness of the map and to complement the picture of the affective and argumentative construction of economic citizenship in online environments.
The two story types that we examined -video reviews and the organized galleries of still photographs -seem to serve to some extent specialized functions in the meaning-making work of multimodal news packages.From the perspective of democratic participation and the representation and facilitation of citizenship, we identified both encouraging and discouraging examples of compiling and using these story types.However, our analysis indicates some differences, which may be normalizing and therefore require critical attention.First, it seems that galleries of still photographs, while providing the opportunity to present a large amount of selected images in a semi-organized manner, tend to overemphasize affect and emotional impact at the cost of adequate contextualization and anchoring to everyday reality, thereby creating internal «editorial realities» of their own and potentially gliding towards simulation.This is a risk especially when captions attached to photographs are very short and merely state the surface level elements visible in the picture.With regard to representing civic action and activism, such practices may be considered particularly harmful, as they often result in the silencing and alienation of the referents.
The use of more elaborate captions, which provide more background and more in-depth interpretations of the meaning of depicted events, emerges as one possible way of anchoring emotionally intensive pictures more closely to local social and political realities.However, such anchoring remains limited as long as the news media favors photographs with framings that exclude all elements of the ordinary or the everyday from the dramatized pictures.Enhancing the politicizing potential of organized photo galleries as a story type would therefore necessitate revision also in the visual culture and practices of photojournalism.
Video reviews embedded in news packages, by contrast, seem to lend themselves relatively well for constructing news narratives that incorporate both affective and argumentative elements, representing both the dramatic and the deliberative, and appealing to both reason and emotion.This capacity arguably stems from several sources, particularly the seamless multimodality of the audiovisual presentation, resulting from a more unified journalistic production process where textual and visual journalistic phases are not strictly separated; and the signifying potential of the audio layer, including e.g. the ambience sound and tone of the voice, which play an important role in admitting the familiar and the ordinary to slip into the representation of also more extraordinary situations.Even though audiovisual material as such does not guarantee these advantages, it seems that in video footage, it is harder to shut out the everyday or the real completely, or to create symbolically strong and affective interpretations as in still photographs.
Overall, it is argued in this paper that together with verbal messaging, visual choices and patterns create the general structure and intensity of news stories, evoking cognitive reasoning, mundane observations, conceptual insights, and affective impassioned interpretations.Audio landscape, in turn, plays a role especially in affective framing, creating the meaning and atmosphere of the situation in complex ways.

Figure 1 ,
Figure 1, the caption: The affect and argument map

Figure 2 ,
Figure 2, the caption: Empirical materials placed on the affect and argument map