Title of the Paper [16 point font] Emotions, Argumentation, and Argumentativi- ty: Insights from an Analysis of Newspaper Headlines in the Context of the Greek Crisis THIERRY HERMAN Faculty of Literature and Humanities Institute of Language Sciences and Communication & EFLE Universities of Neuchâtel and Lausanne. Pierre-a-Mazel 7, CH2000, Neuchâtel, Switzerland. thierry.herman@unine.ch DIMITRIS SERAFIS Faculty of Communication Sciences Institute of Argumentation, Linguistics and Semiotics Università della Svizzera Italiana (USI) Via G. Buffi 13, CH6900, Lugano, Switzerland dimitrios.serafis@usi.ch Abstract: The present paper exam- ines how discursive representations and emotive constructions underpin an argumentative dynamic that emerges from apparently non- argumentative statements, like those found in newspaper headlines. Our data comes from Greek broadsheet newspapers in the polarized context of the Greek crisis. First, we outline an analytic synergy that scrutinizes representational meaning and the semiotization of emotions in head- lines. We then move towards the reconstruction of the inferential passage, contained in the headlines, that unites the implicit standpoint with its supporting argument. Résumé: Cet article examine comment les représentations discursives et les constructions émotionnelles qui les sous-tendent créent une dynamique argumentative dans des énoncés apparemment non argumentatifs, tels que les titres des journaux. Pour ce faire, il s’inscrit dans le contexte polarisé de la crise grecque, nos données provenant de journaux nationaux de ce pays. Nous nous fondons sur une analyse qui allie d’un côté la signification que l’on peut tirer des représentations discursives et de la sémiotisation des émotions dans les titres de journaux avec, de l’autre côté, la reconstruction du mouvement inférentiel unissant le point de vue implicite et l'argument que les titres permettent de reconstituer. Le but de cet article est d’expliquer cette démarche méthodologique par le biais d’exemples représentatif mailto:thierry.herman@unine.ch mailto:dimitrios.serafis@usi.ch 374 Herman & Serafis © Thierry Herman and Dimitris Serafis. Informal Logic, Vol. 39, No. 4 (2019), pp. 373– Keywords: argumentativity, argued emotions, argumentum model of topics (AMT), Greece, Greek crisis, newspapers, print media discourse 1. Introduction Traditionally, argumentation is seen as either a product, a process, or a procedure (Tindale 1999) in which a claim (Toulmin 2003 [1958]) is supported by at least two components: a given premise (datum) and an often implicit linkage premise between the datum and the claim. In this sense, argumentation could be defined a minima as “a verbal act containing a set of propositions (premises) supporting another proposition (conclusion)” (Lewiński and Mo- hammed 2016, p. 1). In many French argumentation theories, however, argumentation may be defined as broadly as “exerting an influence on one’s opinion, attitude, even behavior” (Grize 1990, p. 41, translated by Amossy 2005, p. 87) through verbal means. The focus here is on the goal rather than on a mere description of a textual and/or logical process1 (see the introductory pages of Os- wald, Herman, and Jacquin 2018 for an overview of the different perspectives). Two main influences might explain such a difference. First, the seminal work of Chaïm Perelman and Lucie Olbrechts-Tyteca (1969 [1958]) whose emphasis on the rhetorical goal to gain “the adherence of minds” had a deep impact on the research on argu- mentation, especially after the horror of the World War II and the Nazi propaganda. Even if the authors’ work is loaded with reflec- tions on argumentation schemes, the main idea that argumentation aims to gain and exert an influence on audience’s minds seems to have had more of an impact on the readers than the typology of structural schemes. The explicit reference to a “new rhetoric” emphasizes the attempt to revise the Aristotelian rhetoric, and, indeed, many scholars have suggested that Perelman’s “argumen- 1 The aforementioned distinction does not generally imply that both definitions are not compatible. Some theorists choose to reduce the pragmatic criterion in favor of a structural criterion or the other way around. Still, the utterances labelled as “argumentation” by Grize and Amossy will not be considered as such by most argumentation scholars. The reverse is not true. We try further to distinguish argumentativity and argumentation to avoid such confusion. Emotions, Argumentation, and Argumentativity 375 © Thierry Herman and Dimitris Serafis. Informal Logic, Vol. 39, No. 4 (2019), pp. 373– 400 tation” is essentially a synonym for “rhetoric” (Amossy and Koren 2009, §7). The second main influence may come from linguistics during the so-called structuralist wave, which constrains later argumentation theories, in a way, to observing natural language rather than argumentative reconstructions. The natural logic of Jean-Blaise Grize and his group (1990, 1996) has greatly influ- enced French discourse analysis during the 1970’s and 1980’s along with Anscombre and Ducrot’s (1983) L’Argumentation dans la langue. Grize’s works, in particular, are fueled by the idea that language creates a schematization of the world that can influence an audience’s representations of the world. Describing different phenomena under the same aegis of “ar- gumentation” may be highly problematic. However, what is more important, in our view, is that analysts must be equipped to deal with texts that suggest standpoints without stating the explicit argumentative relationships between different textual elements (see Herman 2018). More specifically, we claim that argumenta- tion theories should take into account the layer of implicit argu- mentation, which Amossy (2005) calls the “argumentative dimen- sion of discourse” or, preferably, “argumentativity” (rather than merely “argumentation”) (see Amossy 2009a). As the author highlights, texts may develop an argumentative potential, that is “argumentativity,” and suggest a standpoint without actually ex- plicitly uttering an argument in favor of a conclusion (see Amossy 2009b). Moreover, we argue that this layer of argumentativity may be amplified by the discursive (or broadly semiotic) construction of certain emotions (e.g. fear, anger, detestation) that could be addressed in order to be legitimately felt by the audience in specif- ic contexts (see Plantin 2011, Micheli 2014). An apparently de- scriptive text, through that prism, may “argue” justifying the emo- tions the audience should feel in this context. As becomes evident, this view is clearly on the fringe of what has been traditionally conceived as “argumentation.” Under these premises, the present paper aims to sketch a methodological and analytical proposal in order to examine how implicit argumentative dynamics are (a) borne out in discursive representations and (b) complemented by emotive constructions in apparently non-argumentative linguistic wrappings such as news- 376 Herman & Serafis © Thierry Herman and Dimitris Serafis. Informal Logic, Vol. 39, No. 4 (2019), pp. 373– paper headlines. We will first explain how argumentativity may be envisioned, setting the threshold between non-argumentation, argumentativity, and argumentation (section 2). Then we will tackle the case of emotions, which are argued through discursive representations, as a prototypical case of argumentativity, and we will explain, following Micheli’s (2014) view, how this hidden argumentative layer in discourse may protrude in the discursive representation (section 2.2). After presenting our analytical framework (section 3), we will focus on our case study: Greek newspaper headlines during the Greek crisis (section 4). Given the methodological orientation of the present paper, two examples will illustrate our proposal. We indeed place our examination within the context of the Greek crisis. Greece has proven to be a special case among the countries of the European Union (EU) that faced significant finan- cial problems when crisis emerged in 2008 (see Vassilopoulou et al. 2014 and references therein). The austerity measures imple- mented by several Greek governments in accordance with the proposals of the so-called “Troika” of the International Monetary Fund (IMF) and the EU institutions caused massive indignation (see e.g. Goutsos and Polymeneas 2014, Serafis et al. 2018) and new divisions in the Greek political and party system (see Dinas and Rori 2013, pp. 271-276; Vasilopoulou and Halikiopoulou 2013, pp. 528-529). This profound socio-political unrest was crucially shaped by media constructions (see Lampropoulou 2014; Mitsikopoulou and Lykou 2015; Lykou and Mitsikopoulou 2017; Serafis and Herman 2018). Within the aforementioned context, we will first scrutinize discursive representations and emotive con- structions realized in headlines of the Greek newspapers Kathi- merini (conservative) and Ta Nea (center-left). Then we will track the opaque argumentative dynamic that lies at the core of the headlines by reconstructing the inferential passage that binds the main standpoint with the supporting argument. 2. Theoretical and methodological orientations Our introduction implies that (a) we can find some argumentative moves in apparently non-argumentative discursive structures and Emotions, Argumentation, and Argumentativity 377 © Thierry Herman and Dimitris Serafis. Informal Logic, Vol. 39, No. 4 (2019), pp. 373– 400 texts, and (b) emotions can make these moves more salient in crucial ways. This issue appears to be quite challenging in current debates. For example, in the “argumentation mining” field, the necessity to identify argumentation along with emotions is under- lined (see Stede and Schneider 2019). More specifically, as Wiebe et al. put it: “[i]s the rhetoric from a particular opposition group intensifying? What is the range of opinions being expressed in the world press [e.g.] about the best course of action in Iraq? A system that could automatically identify opinions and emotions from text should be an enormous help to someone trying to answer these kinds of questions” (2005:1). We aim at intensifying the respective debates, and indeed, we argue that newspaper headlines might be considered as argumenta- tive utterances under certain conditions and in certain contexts.2 In what follows we will try to set clear limits on the concepts we are going to discuss: (a) what is argumentativity, and how does it differ from argumentation? (b) is there any text without argumen- tativity? Our theoretical perspective is informed by Ruth Amossy’s (2012 [2000]) framework entitled as “argumentation in discourse.” This framework is clearly inspired by Grize’s natural logic and Perelman’s view on rhetoric. Following Perelman and Olbrechts- 2 Wiebe et al. (2005) comment on a revealing example from an article from the Beijing China Daily: “(Q1) As usual, the US State Department published its annual report on human rights practices in world countries last Monday. (Q2) And as usual, the portion about China contains little truth and many absurdities, exaggerations and fabrications” (p. 184). According to their terms, the authors consider sentence (Q1) as objective and (Q2) as subjective and they make an interesting remark about the element “And as usual” whose “subjectivity is highly contextual” and “amplified by the fact that ‘as usual’ is repeated from the sentence before” (Wiebe et al 2005, p. 22). Our analysis would not have drawn conclusions about the objectivity of Q1. Briefly, it would have been a two-fold analysis: (1) we would have considered the contextual knowledge about the relations between USA and China as potentially influencing the apparent objectivity; and (2) on a textual level, we would have considered the presence of “as usual” in (Q1) redundant because of the adjective “annual” and lightly incongruous because of “last Monday,” which would have been explained as a clue that may trigger an emotion for a subjective interpretation. 378 Herman & Serafis © Thierry Herman and Dimitris Serafis. Informal Logic, Vol. 39, No. 4 (2019), pp. 373– Tyteca (1969 [1958]), Amossy highlights the importance of rea- sonableness, co-construction, and social setting developed by the Belgian philosopher against the core concepts of truth, rationality, and validity traditionally associated with argumentation. As she says: As a result, argumentation does not simply rest upon solid, immu- table logical procedures the validity of which must be tested by the analyst. What is reasonable and plausible is always co- constructed by men and women engaging in verbal exchanges, and it is the dynamic of this co-construction realized in natural language and in a communicative framework, that has to be ana- lyzed” (Amossy 2009a, p. 317). In this line, and also following Grize (1990), Amossy considers that argumentation may be seen, more specifically, as the use of specific discursive means (and discourses) in order “to act upon an addressee by modifying (or strengthening) his representations of the surrounding world” (Amossy 2009a, pp. 313-314; see also Grize 1990, pp. 40-41). We would like to pay special attention to two aspects of this view- point. Firstly, by highlighting the significance of verbal and dis- cursive means, the theory of argumentation in discourse implies that “argumentation is an aspect of an overall ‘discursive function- ing’ that has to be analyzed in its intrinsic logic” (Amossy 2009b, p. 254) or, more specifically, that “the patterns of reasoning must not only be reconstructed, but also examined in their exact phras- ing, which is not an exterior garment, but the very body of argu- mentation” (2009a, p. 317). Therefore, it proposes a fruitful cross- fertilization between argumentation studies and approaches which belong to a language-based discourse analysis in order to show the discursive means by which one “tr[ies] to reach an agreement, to deal with dissent or to influence ways of experiencing the world” (Amossy 2009b, p. 254). We, in particular, attempt to offer an integrated analytical framework (see section 3) that will, firstly, draw on tools and notions coming from the systemic functional tradition in linguistics (SFL), and discourse analysis, in order to provide an in-depth textual analysis as a first, critical step of an Emotions, Argumentation, and Argumentativity 379 © Thierry Herman and Dimitris Serafis. Informal Logic, Vol. 39, No. 4 (2019), pp. 373– 400 argumentative approach.3 In brief, according to SFL, language is a source/system of meaning making within a socio-cultural context, since, at the same time, it has the intrinsic ability to represent the word in its ideational function/meaning, enact social relationships in its interpersonal function/meaning, and interweave ideational and interpersonal meaning(s) creating contextualized discourses in its textual function (Halliday 1973, 1978). Consequently, in the present case, the term discourse refers to the linguistic modes (social and cultural) that the world is organized and represented in various texts (see Fairclough 2003). Secondly, Amossy proposes the aforementioned cross- fertilization among argumentation and discourse analysis since, as we have already mentioned, she adopts a more dynamic concep- tion of argumentation. According to this conception, argumenta- tion is not only a property of texts that have an explicit “argumen- tative goal or objective” (Amossy 2009b, p. 254), that is, texts that explicitly aim to persuade and achieve consensus, like, for exam- ple, a political speech in the parliament. On the contrary, an argu- mentative potential and dynamic may be retrieved even in texts that simply “orient ways of looking at things and interpreting the world without putting forward any thesis” (Amossy 2009b, p. 254), that is, in apparently non-argumentative genres of texts, such as, for instance, literature and/or news media (see Amossy 2005). Such a broad conception of argumentation is based on the premise that a discourse “always answers some explicit or hidden question, or at least suggests a way of looking at the surrounding world” being “pervaded by a general argumentativity” (Amossy 2009a, p. 254, our emphasis). More specifically, “[t]he theory of ‘[a]rgumentation in discourse’ is based on the assumption that argumentativity pervades and partly regulates all verbal exchang- es” (Amossy 2009a, p. 313). In other words, argumentative poten- tial and obscure argumentative patterns may be tracked (in various 3 The significance of a descriptive textual analysis is highlighted in both New Rhetoric’s tradition and Amossy’s (2005) perspective. In section 3, we will become more specific in accordance with the tools we employ from the SFL. It is worth mentioning that the integration of tools coming from systemic- functional linguistics is part of the methodological proposal that we outline here. Amossy does not follow an SFL perspective. 380 Herman & Serafis © Thierry Herman and Dimitris Serafis. Informal Logic, Vol. 39, No. 4 (2019), pp. 373– degrees) even in apparently non-argumentative texts and structures (Amossy 2002, 2005) since “argumentativity is one of discourse’s constitutive dimensions, which means that even discourses that do not participate in any formal debate or disclose a manifest persua- sive purpose have an underlying argumentative dimension” (Amossy 2009a, p. 315, original emphasis; see also Amossy 2005). As we will show in our analysis (see section 4), even if, at first glance, much remains implicit in newspaper headlines, they nevertheless may construe dense, contextualized representations of the social action that takes place in specific moments in the Greek crisis. We grant the premise (and we will illustrate in our analysis) that, in these discursive representations, a standpoint may be sus- tained even if left implicit (see Zampa 2017, p. 131). A SF-based analysis will provide us with a scrutiny of the discursive represen- tations construed in the headlines under examination. 2.1. Argumentativity and argumentation By using the term “argumentativity,” our purpose is not necessari- ly to widen the notion of argumentation to all discourses, which is a danger already noted by Micheli and referred to as “generalized argumentativism” (2009, p. 20). Even though we will not consider each utterance as a form of argumentation (we will put forth two criteria that we use to avoid the trap of the “generalized argumen- tativism”), we argue that linguistic wrappings, as descriptive as they may appear, may trigger inferences in an argumentative way. Many reasons can support this position. First, there is no statement without a portrayal of the speaker, as Amossy acknowledges (2012); therefore, the image I may give of myself when I speak can influence how the audience will perceive what I say. Secondly, as Plantin (2011) and Micheli (2014) showed, emotions can be situated beyond simple emotional nouns or adjectives. For instance, the clause “Children are starving to death in Syldavia” is a descriptive statement that may however legitimize, through the expressed fatal outcome, pity or anger towards such a reality.4 We will return to this issue later. 4 We refer to the rhetorical dimensions of ethos and pathos here. However, we maintain that the notions of ethos and pathos should only be used under certain Emotions, Argumentation, and Argumentativity 381 © Thierry Herman and Dimitris Serafis. Informal Logic, Vol. 39, No. 4 (2019), pp. 373– 400 What these two claims clearly show is that texts do not only describe reality but may also influence ways of considering the world. The text itself (as part of its context) serves as an argument to justify an interpretation in an abductive approach of discourses (Eco 1992). The signs, here language, but we could broaden the palette to prosody, gestures, etc., may give arguments that support interpretations in a particular way. According to Jean-Blaise Grize, the way we use and arrange words in context can be described by the metaphor of “illumination”—“éclairages” in French5: “The discourse entities must be illuminated, which means that some of their facets must be highlighted and others hidden and every illu- mination colours what it illuminates, because it makes use of cultural preconstructs that are never neutral.” (Grize 2004, p. 42, translated by van Eemeren et al. 2014, p. 488). The notion of “illumination” is a relevant metaphor that highlights how lexical and syntactic choices constitute signs to be interpreted in an infer- ential way, precisely because of how they are “illuminated.” While Grize considers that it is difficult to decide whether a text is argu- mentative or not,6 Amossy radicalizes her theoretical point of view by considering, as we mentioned before, that argumentativity is inherent of any discourse. While we assume that any statement can be critically, hermeneutically, or interpretatively analyzed, we believe it would be difficult to consider that each statement, be- cause it may influence our beliefs, is necessarily a threat or a trap that aims at adhering unaware minds. It would be quite exhausting to think that every utterance exerts an influence of which we should be aware. Therefore, we propose two criteria that, in our view, may indicate that a text is fueled by argumentativity: (a) argumentativity may be triggered by textual salience and/or (b) conditions, specifically in the explicit presence of an opinion that the speaker is committed to (see Herman 2018). 5 According to Amossy, the official translation of “éclairages” is “points of view” in English (personal communication). 6 “Taken in itself, indeed, an utterance is neither argumentative nor non- argumentative. You tell someone ‘The train leaves at 11:45 am’. It may be an argument to urge him to leave you, as it may be a simple piece of information that he will take note of and use on another occasion” (Grize 1996, p. 22, our translation). 382 Herman & Serafis © Thierry Herman and Dimitris Serafis. Informal Logic, Vol. 39, No. 4 (2019), pp. 373– argumentativity can reasonably be taken into account because of the contextual knowledge. The first criterion echoes a discussion revolving around how linguistic/semiotic choices may be underlined in texts strengthen- ing in this sense the construed meaning. For instance, the notion of “foregrounding” (see Leech 2008, Leech and Short 2007), which according to the dictionary of stylistics “is […] the ‘throwing into relief’ of the linguistic sign against the backgrounds of the norms of ordinary language” (Wales 2014, p. 166). On the same discus- sion, Bonhomme (2005) proposes the idea of different degrees of “salience” provoked by figures of speech in a text. Finally, in the systemic functional perspective, Halliday uses the concept of “prominence as a general name for the phenomenon of linguistic highlighting, whereby some feature of the language of a text stands out in some way” (Halliday 1971, p. 340), while fore- grounding is a kind of motivated prominence that “contributes to the writer’s total meaning” (Halliday 1971, p. 339).7 In all the aforementioned traditions in linguistics, textual salience (or fore- grounding or prominence) is supposed to mark the constructed meaning; in this sense, it may give rise to (or better, or sharpen) an implicit argumentative potential, that is argumentativity. The second criterion is a contextual one. It is the knowledge of the context that strengthens the representation of the text and allows the addressee to assume, with more or less plausibility, that an apparently descriptive text may have an argumentative poten- tial, or better, may be intertwined with argumentativity. For exam- ple, the knowledge regarding the different positioning of the ex- amined newspapers, their political tendencies, and the very polar- ized and tense events in Greece are good reasons to believe that Greek newspapers’ headlines are more inclined to reveal their standpoints than a traditional news report from a news wire agen- cy. Because of the context, we could be suspicious about the prob- 7 An extensive discussion regarding these quite subtle questions goes beyond the scope of this paper. However, we would like to highlight that the way of saying things should be more or less salient, prominent, or foregrounded to trigger an argumentative analysis. Emotions, Argumentation, and Argumentativity 383 © Thierry Herman and Dimitris Serafis. Informal Logic, Vol. 39, No. 4 (2019), pp. 373– 400 able conscious or unconscious bias and intent or will to exert an influence on the audience.8 Next, we would like to emphasize a special aspect that enforces argumentativity in texts and discourses, and this one relates to specific interpretations tracked by emotions constructed in dis- course. We will show how different point of views (i.e. “illumina- tions” in Grize’s terms) might justify some emotions; we will show how they argue for and through an emotion. Our idea is that apparently descriptive journalistic headlines offer justifications for and through emotions particularly in the highly polarized context during the Greek crisis. 2.2. Emotions and argumentativity More specifically, we seek to show how the argumentative dynamic borne out in discursive representations is enforced in newspaper headlines. To that end, we appeal to the studies exploiting the category of “argued emotion” (Plantin 1999, 2004, 2011; Micheli 2010, 2014). We are seeking to show how argumentativity is materialized through emotive constructions in configurations of print media discourse, specifically in newspapers headlines. According to these studies, and echoing rhetorical pathos, “emotions themselves are accessible to argumentation,” and more specifically, they may become “the very object of argumentation” since speakers “argue in favor of or against an emotion” and support what they feel and why they should legitimately feel like this (Plantin 2004 as quoted in Micheli 2010, p. 5, p. 13). More specifically, emotions may become the very vehicle of achieving consensus on the reasonable (Perelman 1979, Amossy 2009a) and thus they may be an intrinsic part of argumentation as this is realized in various genres of discourse (see Cigada 2016, p. 390; see also Plantin 1998). According to recent research on an argumentative perception that could embrace an analysis of 8 This does not imply that argumentativity does not exist in the most descriptive texts and in neutral contexts. For example, it has been shown that even language dictionaries may manifest sexist, religious, or political ideologies (Chen 2017, Moon 2014). However, we highlight that argumentativity is less relevant in some situations. 384 Herman & Serafis © Thierry Herman and Dimitris Serafis. Informal Logic, Vol. 39, No. 4 (2019), pp. 373– emotions in discourse, the particular function of these emotive constructions is to enhance the premises on which a claim may be developed (see Schär 2018, p. 147). Following this research, we will show in our analysis that emotive constructions stem from and complement the discursive construction of reality, aiming to motivate the audience to make the choice proposed (implicitly or explicitly) by each newspaper. Moreover, as we will see in the analysis section, the presence of emotions may advance the “degrees of argumentativity” (Amossy 2005, p. 88), sharpening, for example, the discursive representation. Summing up, following the conceptual lines of Amossy’s (2010 [2000]) perception on argumentativity, we first pay special attention to a linguistic-based discourse analysis of newspaper headlines in order to unveil implicit aspects of the construed representational meaning. Then, we show how meaning is complemented by appealing to an analysis that focuses on how emotions are semiotized in discursive structures (i.e. in newspapers headlines). Based on these findings, we will move towards the reconstruction of the inferential passage that unites the main standpoint with the argument that supports it, unveiling, in this sense, the opaque argumentative dynamic that lies at the core of the headlines. 3. Analytical perspectives—a synergy For the aims of our analysis, we will present an approach employ- ing analytical tools by two pillars (see Serafis 2017, Ch. 3; Serafis and Herman 2018, pp. 187-189 for a preliminary elaboration). Namely: (a) Systemic Functional (SF) lexicogrammatical analysis (Halliday and Matthiessen 2004) and (b) rhetorical analysis of emotions semiotization in discourse (Micheli 2014). As we have already heralded, based on the results coming from this analytical combination, we will proceed to the reconstruction of inferential configurations emerging in the headlines by drawing on the Ar- gumentum Model of Topics (henceforth AMT) (see Rigotti and Greco Morasso 2010; Rocci 2017; Rigotti and Greco 2019). More specifically, firstly, we exploit tools from a SF analysis of transitivity, aiming to scrutinize the meaning construed upon the representation of social action in newspapers headlines (see Halli- Emotions, Argumentation, and Argumentativity 385 © Thierry Herman and Dimitris Serafis. Informal Logic, Vol. 39, No. 4 (2019), pp. 373– 400 day and Matthiessen 2004, Ch. 5). Transitivity involves configura- tions of representational meaning by interrelating three main ele- ments: the processes (realized by verbal groups), the participants (realized by nominal groups), and the circumstances (realized by various types of adverbial groups and prepositional phrases) (Hal- liday and Matthiessen 2004, pp. 176–177, p. 260). In addition to transitivity analysis, and following Micheli (2014), we conduct an analysis of emotions semiotization in dis- course in order to show the emotions revealed in different repre- sentations in transitivity configurations. We highlight the im- portance of emotions constructed by discursive representations because, as we will witness, emotive constructions complement the discursive construction of reality, aiming to guide the audience towards the implicit claim made by the newspaper. In this sense, “argued” emotions underpin newspapers’ attempts to express a standpoint and provide stimuli for actions (Plantin 2004; Micheli 2010). The process of emotion semiotization occurs through three independent modes coded as “said,” “shown,” and “argued” emo- tion (Micheli 2014: Ch. 1). A “said” emotion is explicitly realized by a verbal or a nominal group: “to fear,” “joy,” and so on. A “shown” emotion may be inferred by semiotic markers whose presence appears to be due to this emotion; there is a linguistic feature or marking that can be interpreted as a result of an emotion which appears to be (either truly or not) experienced by the speak- er/writer. For example, an elliptic clause or an exclamation mark are signs of a shown emotion. Finally, an “argued” emotion is construed through: (a) the representation of a situation in a text, in essence, the interplay of the elements in transitivity, and (b) social and cultural knowledge that ties up this situation with emotions (see also Plantin 2011). The text here is providing some justifica- tion for experiencing emotions. In this case, the emotion is not the cause of a semiotic marker/characteristic but the outcome one can infer from the represented situation. Transitivity analysis may consequently exemplify how emotions are semiotized in discourse. Furthermore, Micheli (2014: 114) gives seven “criteria” on which the speaker can support that the emotion “relies on good reasons.” Those are adumbrated in the following lines and are related to the 386 Herman & Serafis © Thierry Herman and Dimitris Serafis. Informal Logic, Vol. 39, No. 4 (2019), pp. 373– people involved (naming strategies, participant roles of agent or patient); to the expressed spatial and temporal distance (e.g. the closer in time an event, the more justified the emotion it triggers); to the potential consequences and to the probability of them hap- pening; to the causes and responsibilities emphasized; to the capacity to command or control the situation; to the allusion to resembling situations; or to the compatibility or not with the val- ues shared among a reference group. For example, the headline “three children have been found starved to death in our city” in an European newspaper may be seen as constructing an argued emotion through many criteria including: (1) victims are children who are socially considered “innocent” and need to be protected—the prominence is highlight- ed by the fact that they are the grammatical subject of this passive phrase; (2) the consequences are terrible (death); (3) the temporal and spatial distance are quite near, raising the degree of emotion we are supposed to experience; (4) this situation is not compatible with social values, in this case, children must be protected; (5) we were unable to control and avoid this terrible situation; (6) the causes and responsibilities are unknown although we should won- der about the parents’ responsibilities which are socially supposed to protect the children as well as the responsibilities of social institutions. For these reasons, this forged example offers many reasons that underpin an emotion that is, by the way, hard to label (possibilities include anger, shame, indignation, etc). The “argu- mentativity” of this headline, which appears to be descriptive, can be seen through the reconstruction of unexpressed arguments that justify an expected emotion in such situations. Based on the findings of the aforementioned analysis, we em- ploy the Argumentum Model of Topics (AMT) (see Rigotti and Greco Morasso 2010; Rigotti and Greco 2019) in order to track the internal inferential configuration that links the main standpoint with the argument that lies at the core of the headline(s). The AMT provides a quasi-Y structure that presents the main components of an inferential configuration. Namely, on the one hand, the proce- dural-inferential component which includes (a) the locus/loci, i.e. the ontological relation (e.g. cause-effect) the argument originates from and (b) the maxim, i.e. some general inferential principles— Emotions, Argumentation, and Argumentativity 387 © Thierry Herman and Dimitris Serafis. Informal Logic, Vol. 39, No. 4 (2019), pp. 373– 400 usually realized as prepositions—that spring from the locus and give rise to specific arguments (e.g. if the cause is present, so is the effect). On the other hand, the material-contextual component includes the shared socio-cultural premises that the participants (must) take into consideration. Namely, (a) the endoxon that refers to the shared opinion(s), accepted by the public in a specific socio- cultural setting, and (b) the datum, which is a premise of factual nature. The intersection of these two components (procedural- inferential and material-contextual) form a quasi-Y structure that better highlights the way towards the sustained stand- point/conclusion (see Rigotti and Greco 2019: 208-216). Next, we briefly present our data and the criteria of their selection before moving towards our analysis. 4. Data analysis In total, our data comprise 50 headlines coming from the following Greek broadsheet newspapers: Kathimerini, Ta Nea, Elftherotypia, and Ethnos. The headlines were selected from key dates in the emergence of the Greek crisis like, for instance, May 06, 2010—t he day after the voting in the Greek parliament for the approval of the first bailout program and the consequent austerity measures. We focused on the aforementioned newspapers since they are representative of the mainstream views appearing in the Greek context, in essence, center-right, right-wing in Kathimerini; and center-left, left-wing in Ta Nea, Eleftherotypia, and Ethnos (see Psychogios 2004: 477-493). We have chosen to focus on the ex- amination of newspaper headlines dealing with structures that, as we mentioned above, do not have an explicit argumentative aim (Amossy 2005). We nevertheless maintain, and will show in our analysis, that an argumentative potential (that is argumentativity) is borne out in newspaper headlines. On the one hand, headlines as a specific genre (Bell 1991) express newspapers topic(s), that is, they “summarize the text and specify the most important infor- mation.” On the other hand, their selection has significant “ideo- logical implications” (van Dijk 1991, p. 113; see also van Dijk 1988) since headlines express the chosen “angle” (Robin 2009) of the news, giving a perspective that may frame reality and sustain a 388 Herman & Serafis © Thierry Herman and Dimitris Serafis. Informal Logic, Vol. 39, No. 4 (2019), pp. 373– newspaper’s standpoint. After presenting our data and illustrating the criteria of selection, we offer representative examples of analy- sis compliant to the analytical synergy we outlined before (see section 3). (1) Η στυγνή βία προκαλεί σοκ ‘The brutal violence causes shock’ (Kathimerini, May 6, 2010) In a headline in Kathimerini on May 6, 2010, the nominal group Η στυγνή βία “The brutal violence” has the participant role “phenom- enon” in the “mental process” which is, consequently, realized by the verbal group προκαλεί σοκ “causes shock.” The adjective “brutal” assigns a further (negative) quality to the nominal type “violence.” It is also worth mentioning that there is no explicit reference to an individual or entity under the status of “shock.” This would be realized by the participant role “senser” (of the “mental” process) in the structure. Since we do not know who exactly the shocked individual (or group) is, this very absence tends to construe a kind of generalization in the sense that every- one is (or should be) shocked. Moreover, the choice of a “demon- strative deictic” (“the”) “embodied in the structure of the nominal group” (“the violence”), has “the function of identifying a particu- lar subset of the ‘thing’ that is being referred to” (Halliday and Matthiessen 2004, pp. 312–314). In this sense, “the violence” refers to a specific incident or kind of violence taking place within the specific period/context. What appears to be crucial for our understanding of the headline is the background knowledge that is activated here through the presence of the nominal “the violence.” Specifically, on May 5, 2010, massive demonstrations took place in the Greek capital (Athens) protesting against the first bailout program which was about to be voted in the Greek parliament in accordance to the Memorandum of Understanding (henceforth MoU), signed between the Greek center-left/socialist government (PASOK) and the “Troika” of the EU institutions and the IMF. During these demonstrations, three persons lost their lives when the branch of the Marfin Bank in Stadiou Street (Athens city- center) was set ablaze. Contextual knowledge about the violent events of May 5 makes the headline more coherent securing our Emotions, Argumentation, and Argumentativity 389 © Thierry Herman and Dimitris Serafis. Informal Logic, Vol. 39, No. 4 (2019), pp. 373– 400 interpretation in the specific context. Still, the absence of the specific social agent (who exercises “the violence”) or the one that experiences the “shock” construes extensive social diffusion of the violence and shock in the demonstration. This allows Kathimerini to implicitly represent the demonstrations (like the one on May 5, 2010) as the very place where “the violence” exists and the pro- testers (in general as the embodiment of the protest) as the agents of this “violence.” Kathimerini incriminates social disobedience to austerity programs in this manner. Complementing our analysis in transitivity with the analysis of emotions semiotization in this headline, we may witness shock coded as “said,” realized, respectively, by the nominal types βία “violence” and σοκ “shock.” The adjective στυγνή “brutal” and the (further) negative quality that it carries (in the nominal group Η στυγνή βία “The brutal violence”) further loads the specific emo- tion (i.e. shock). Since no specific social actor is represented as experiencing the emotion of shock in the transitivity structure, shock may be experienced by everyone in the specific context. Moreover, the “said” emotion, portrayed as having a general affect to the audience, can be coded as “argued” and one that further provokes the emotion of fear, in Micheli’s terms, that should be legitimately felt by the audience of the newspaper. According to the representational meaning, construed in the transitivity, shock is extensively disseminated in the society (audience) and thus (a) there is no easy way (for anyone) to control the situation (see the respective criterion of the [lack of] ability to control the situation) and, moreover, (b) according to the extra-textual knowledge, the “violence” stems from the anti-bailout protests (see the criterion of the people involved) and the consequences which are very close to time (see criterion of temporal and spatial distance) are very strong, i.e. the death of the three bank-employees (see the criterion of the possible consequences). Indeed, the (im)possibility of the audience to take control of the situation (see the respective criteri- on) is highly used here: the lack of control justifies and legitimizes the emotion of fear which is “argued” as a consequence of the shock. Having illustrated the meaning and emotive constructions revealed in the analyzed headline, we move now towards the 390 Herman & Serafis © Thierry Herman and Dimitris Serafis. Informal Logic, Vol. 39, No. 4 (2019), pp. 373– reconstruction of the implicit standpoint honed in the discursive and emotive construction. This could be: “The anti-austerity pro- tests should be avoided.” The argument that supports the afore- mentioned standpoint results from the discursive and emotive construction that, as we tried to unveil in our previous analysis, appears in the headline: “The anti-austerity protest encapsulates fatal violence and causes fear.” In order to exemplify how this very standpoint and argument are interconnected, we employ the AMT quasi-Y structure. More specifically, on the axis of the procedural-inferential components the locus from final cause reveals, in the sense of the maxim that focuses on the negative side effect of the action (i.e. participation in the anti-austerity demon- stration), “if action X has a negative side effect Y, then X must not be undertaken.” The procedural-inferential components are instan- tiated in the social context through the axis of the material- contextual premises. In this case, the endoxon could be the follow- ing bi-focal one: “the use of violence should be avoided” and the one amplified by the provoked emotions: “whenever possible [an individual or a group] tries to avoid confrontation with fear [and shock]” (Schär 2017, p. 149). The datum–resulting from the analy- sis of meaning and emotions implemented before—could be: “the anti-austerity demonstration of May 5, 2010 has resulted in the emergence of a brutal violence that caused the death of three citizens and consequently fear and shock.” At the intersection of the two axis, the first conclusion/minor premise would be: “Indi- viduals or groups going to anti-austerity demonstrations, such as the one of May 5, 2010, has the negative effect of fear and shock because of the emerging (fatal) violence.” Consequently, the final conclusion is: “anti-austerity demonstrations should be avoided” (see figure 1). Emotions, Argumentation, and Argumentativity 391 © Thierry Herman and Dimitris Serafis. Informal Logic, Vol. 39, No. 4 (2019), pp. 373– 400 Figure 1: The AMT representation in the case of Kathimerini on May 6, 2010 (2) Ψήφος εκτάκτου ανάγκης “Emergency vote” (Ta Nea, June 29, 2011) In the nominal group of the headline of Ta Nea on June 29, 2011, Ψήφος εκτάκτου ανάγκης “Emergency vote,” by the nominal type Ψήφος “vote” portrays the process of the actual voting of the middle-term agreement of the bailout program (including a series of austerity measures) inside the Greek parliament; thus, the nom- inal type is seen as a “process noun” (van Leeuwen 2008, p. 30; see also Halliday and Matthiessen 2004, p. 439). The action of voting (“vote”) undertaken by the MPs is perceived to be caused 392 Herman & Serafis © Thierry Herman and Dimitris Serafis. Informal Logic, Vol. 39, No. 4 (2019), pp. 373– by the nominal group εκτάκτου ανάγκης “emergency”; the specific conceptualization is realized in Greek by the nominal group in the genitive case which states the necessity of the ‘vote’ (see εκτάκτου ανάγκης [genitive case underlined]). Through the above represen- tation, the “vote” in the Greek parliament is imposed by an “emer- gency” occurring due to the financial turbulence in Greece. Con- textual knowledge offers us this more adequate interpretation: There were widespread statements addressed by governmental officials, according to which none of the MPs of the ruling social- ist party (PASOK) were disposed to vote for such large-scale austerity measures—included in the bailout program—baldly at the antipode of their pre-electoral manifesto which was promising an expansion of fiscal policies. PASOK MPs were attributing implementation of these measures to the fact that a financial col- lapse would follow a possible failure of the bailout agreement with Greece’s creditors (MoU). In this sense, the parliamentary majori- ty is conceptualized as acting (“vote”) against its will in order to overcome the tremendous financial difficulties and avoid a finan- cial collapse. The meaning construction could be exemplified as: PMs’ ruling majority votes (against its will) due to the (e.g. finan- cial) emergency. On the emotions analysis, anxiety and danger may be first realized by the nominal type “emergency” and thus coded as “said” in the headline. However, based on the precedent analysis, the socialist government and the MPs of the majority are implicitly conceived as the social entities who vote in favor of the austerity measures (see the criterion of the people involved), not according to their will but because of Greece’s emergency state, in order to overcome this state of emergency. In this sense, although they are (primarily) represented as not being compatible with the values of a background that favors socialist ideals of, for example, expan- sionary fiscal policies, the voting in favor of austerity measures is imposed by the need to rescue the country, and thus this is fully compatible with the dominant value according to which the gover- nors should ensure the financial survival and societal prosperity (see the criterion of compatibility with dominant values). So, the initially semiotized (“said”) emotions of anxiety and danger give their place to the emotions of sympathy and/or admiration—which Emotions, Argumentation, and Argumentativity 393 © Thierry Herman and Dimitris Serafis. Informal Logic, Vol. 39, No. 4 (2019), pp. 373– 400 are labelled as “argued”—in favor of the Greek government and the MPs that support it. Moreover, it is worth noting here that sympathy and admiration should be felt since, in an extremely short and polarized period (see temporal criterion), the government makes difficult decisions in order to avoid (financial etc.) dangers (see consequences criterion) caused precisely because of the actual socio-political and financial situation, that is, by the state of emer- gency. In this sense, the nominal choice “emergency” in this con- figuration argues the specific emotions. The standpoint implicitly expressed in the newspaper’s con- struction is the following: “the government (and the parliamentary majority) should be sympathized with and/or admired.” The sup- porting argument could be: “because the government is voting in order to overcome an extreme danger.” The inferential configura- tion emerging in this headline is adumbrated in the following lines and is founded on the results of the precedent meaning and emo- tions’ analysis. On the procedural-inferential component, the locus employed is the ontological implication, which gives rise to the maxim: “if X does what she is expected to do—especially under difficult conditions—then she deserves sympathy and admiration.” On the axis of the material-contextual components, the endoxon could be the following complex one: (a) “governments must en- sure the prosperity of the society” and (b) “in cases of emergency extra measures should be approved” and (c) “negative emotions should be avoided.” The datum in this case would be “the gov- ernment and the MPs’ majority vote for an austerity bailout pro- gram in conditions of (financial) emergency thus ensuring prosper- ity and avoiding negative emotions.” The minor conclusion could be: “the government, along with the majority of the MPs, do what they are expected to do, that is, they vote in order to end the state of (financial) emergency and the provoked anxiety/danger” and the major conclusion could be: “the government and the majority of the MPs should be sympathized with and/or admired” (see figure 2). 394 Herman & Serafis © Thierry Herman and Dimitris Serafis. Informal Logic, Vol. 39, No. 4 (2019), pp. 373– Figure 2: The AMT representation in the case of the headline of Ta Nea on June 29, 2011 5. Conclusion Summing up, this paper aimed primarily at sketching an analytical proposal in order to study instances of implicit argumentative potential emerging from apparently non-argumentative structures of language and discourse. Therefore, we examined headlines appearing in Greek mainstream and broadsheet newspapers; a genre of discourse (i.e. headlines) that is supposed to offer the main meaning construction of a newspaper regarding the socio- political situation without apparently putting forward some explicit Emotions, Argumentation, and Argumentativity 395 © Thierry Herman and Dimitris Serafis. Informal Logic, Vol. 39, No. 4 (2019), pp. 373– 400 argument(s) that may support a claim. Our analysis was located in an extremely polarized socio-political context, specifically, the Greek crisis. Despite the short data analysis, we have shown that the analyzed newspaper headlines may possess an opaque argu- mentative core in the discursive and emotive representations that emerge from these very linguistic wrappings. Since the paper was a methodologically-oriented one, we do not claim to provide ex- tended analytical findings through our analysis during the relevant period. However, we hope that the analytical integration we out- lined here will be applied to different datasets in future research. Headlines may sustain an implicit standpoint since argumenta- tivity there is an implicit argumentative potential and a dynamic that pervades them as one of the constitutive intrinsic characteris- tics of discursive wrappings (Amossy 2012) mostly appearing in public texts and discourses such as the one under examination here. Since much of the meaning construal in a headline remains implicit, a language-based discourse analysis appears to be crucial to provide us with a scrutiny of the meaning construction. To that end, a Systemic Functional (SF) analysis (Halliday and Matthies- sen 2004) underpinned our consideration and offered us a descrip- tive view of the representational meaning construed in the two above-analyzed headlines in relation to the contextual lines that shaped this very meaning construction. Moreover, we show that an emotive construction borne out in the discursive structure may amplify the argumentative dynamic of the headlines enforcing the argumentative premises and paving the way to the final conclu- sion. We drew on studies of rhetorical pathos (Plantin 2011) in order to frame this premise and we developed an analysis of semi- otization of emotions in discourse (Micheli 2014) in order to trace emotions in the headlines. We gave weight to the role emotions play in widening the argumentative potential of discursive repre- sentations. Following the criteria and the methodological tools provided by Micheli (2014), we show that emotions may enrich both the material-contextual and the procedural-inferential com- ponents of an inferential configuration. Overall, this bi-focal ex- amination better unpacked and clarified the premises on which the sustained standpoint and the supporting arguments are based. The reconstruction proposed by the Argumentum Model of Topics 396 Herman & Serafis © Thierry Herman and Dimitris Serafis. Informal Logic, Vol. 39, No. 4 (2019), pp. 373– (AMT, see Rigotti and Greco 2019) through the quasi-Y structure permitted us to unveil then the inferential configuration argument- to-standpoint that otherwise would remain obscure in the present- ed data. Although a more extensive analysis of a wider corpus would address these issues more precisely, our intention in this methodologically-oriented paper was primarily to provide a syner- gy in order to track implicit argumentation in instances of appar- ently non-argumentative discourses. In this sense, we wish to sketch new research avenues for contemporary debates on argu- mentation studies. Acknowledgements: We would like to thank Raphaël Micheli for his input on a previous version of this paper. We especially owe thanks to the two anonymous reviewers for their constructive comments. References Amossy, Ruth. 2002. Introduction to the study of doxa. Poetics Today 23(3): 369-394. Amossy, Ruth. 2005. The argumentative dimension of discourse. In Practices of Argumentation, eds. Frans H. van Eemeren and Peter Houtlosser, 87-98. Amsterdam: John Benjamins. Amossy, Ruth. 2009a. The new rhetoric’s inheritance. Argumentation and discourse analysis. Argumentation 23(3): 313-324. Amossy, Ruth. 2009b. Argumentation in discourse. A socio-discursive approach to arguments. Informal Logic 29(3): 252-267. Amossy, Ruth. 2012. L’argumentation dans le discours. Paris: A. Colin. Amossy, Ruth, and Roselyne Koren. 2009. Rhétorique et argumentation: Approches croisées. Argumentation et Analyse du Discours 2: https://doi.org/10.4000/aad.561. URL accessed 26 February 2019 https://journals.openedition.org/aad/561. Anscombre, Jean-Claude, and Oswald Ducrot. 1983. L’argumentation dans la langue. Bruxelles: P. Mardaga. Bell, Allan. 1991. The language of news media. Oxford: Blackwell. Bonhomme, Marc. 2005. Pragmatique des figures du discours. Paris/Geneve: H. Champion. Cigada, Sara. 2016. Analyzing emotions in French discourse: (Manipula- tive?) shortcuts. In Case studies in discourse analysis, eds. Marcel Danesi and Sara Greco, 388-408. Munich: Lincom Europa. https://journals.openedition.org/aad/561 Emotions, Argumentation, and Argumentativity 397 © Thierry Herman and Dimitris Serafis. Informal Logic, Vol. 39, No. 4 (2019), pp. 373– 400 van Dijk, Teun A. 1988. News as discourse. Hillsdale, NJ: Lawrence Erlbaum. van Dijk, Teun A. 1991. The interdisciplinary study of news as dis- course. In Handbook of qualitative methods in mass communication research, eds. Klaus Bruhn-Jensen and Nicholas Jankowski, 108-120. London: Routledge. Dinas, Elias, and Lamprini Rori. 2013. The 2012 Greek parliamentary elections: Fear and loathing in the polls. West European Politics 36(1): 270-282. Eco, Umberto. 1992. Les limites de l’interprétation. Paris: B. Grasset. van Eemeren, Frans. H., Garssen, Bart, Krabbe, Erik C. W., Snoek Henkemans, Fransisca A., Verheij, Bart, and Jean Wagemans. 2014. Handbook of argumentation theory. Dordrecht: Springer. Fairclough, Norman. 2003. Analysing discourse: Textual analysis for social research. London: Routledge. Goutsos, Dionysis and Polymeneas, George. 2014. Identity as space: Localism in the Greek protests of Syntagma square. Journal of Language and Politics 13(1): 675-701. Grize, Jean-Blaise. 1990. Logique et langage. Paris: Ophrys. Grize, Jean-Blaise. 1996. Logique naturelle et communications. Paris: Presses universitaires de France. Grize, J.-B. (2004). Le point de vue de la logique naturelle. Démontrer, prouver, argumenter [The point of view of natural logic. Demonstrat- ing, proving, arguing]. In L’argumentation aujourd’hui. Positions théoriques en confrontation, eds. M. Doury and S. Moirand, 35-44. Paris: Presses Sorbonne Nouvelle. Halliday, Michael A.K. 1971. Linguistic function and literary style: An inquiry into the language of William Golding’s the inheritors. In Lit- erary Style: A symposium, ed. Seymour Chatman. New York: Oxford University Press. Halliday, Michael A.K. 1973. Explorations in the functions of language. London: Edward Arnold. Halliday, Michael A.K. 1978. Language as social semiotic: The social interpretation of language and meaning. London: Edward Arnold. Halliday, Michael A.K., and Christian M.I.M Matthiessen. 2004. An Introduction to functional grammar (3rd revised edition). London: Edward Arnold. Herman, Thierry. 2018. Éclairages, dimension rhétorique et argumentation à l’épreuve des tweets de Donald Trump. Argumentation et Analyse du Discours 20. https://doi.org/10.4000/aad.2504. URL accessed 26 February 2019 : https://journals.openedition.org/aad/2504. https://journals.openedition.org/aad/2504 398 Herman & Serafis © Thierry Herman and Dimitris Serafis. Informal Logic, Vol. 39, No. 4 (2019), pp. 373– Lampropoulou, Sofia. 2014. Greece will decide the future of Europe: The recontextualisation of the Greek national elections in a British broadsheet newspaper. Discourse & Society 25(4): 467-482. Leech, Geoffrey. 2008. Language in literature: Style and foregrounding. London: Routledge. Leech, Geoffrey, and Michael Short. 2007. Style in fiction: A linguistic introduction to English fictional prose. London and New York: Longman. van Leeuwen, Theo. 2008. Discourse and practice: New tools for criti- cal discourse analysis. New York: Oxford University Press. Lewiński, Marcin, and Dima Mohammed. 2016. Argumentation theory. In International encyclopedia of communication theory and philoso- phy, eds. Klaus Bruhn Jensen, Robert Craig, Jefferson Pooley and Er- ic Rothenbuhler, 1-15. New York: John Wiley & Sons. Lykou, Christina, and Bessie Mitsikopoulou. 2017. The chronicle of an ongoing crisis: Diachronic media representations of Greece and Eu- rope in the Greek press. In: Greece in crisis: Combining critical dis- course and corpus linguistics perspectives, eds. Dionysis Goutsos and Ourania Hatzidaki, 113-150. Amsterdam: John Benjamins. Micheli, Raphaël. 2009. La construction de l’objet dans quelques approches discursives et interactionnelles de l’argumentation: Une approche métathéorique. In Argumentation: théorie, langue, discours: Actes de la section Argumentation du XXX. Congrès des romanistes allemands, Vienne, septembre 2007 (Rhethos), eds. Vahram Atayan and Daniela Pirazzini, 19-30. Frankfurt am Main: Peter Lang. Micheli, Raphaël. 2010. Emotions as objects of argumentative constructions. Argumentation 24(1): 1-17. Micheli, Raphaël. 2014. Les émotions dans les discours. Modèle d’analyse, perspectives empiriques. Bruxelles: De Boeck (“Champs linguistiques”). Mitsikopoulou, Bessie, and Christina Lykou. 2015. The discursive construction of the recent European economic crisis in two political magazines. On the Horizon 23(3): 190–201. Oswald, Steve, Herman, Thierry, and Jérôme Jacquin, eds. 2018. Argu- mentation and language-linguistic, cognitive and discursive explora- tions. Cham: Springer International Publishing. Perelman, Chaïm. 1979. The rational and the reasonable. In The new rhetoric and the humanities. Essays on rhetoric and its applications, 117-123. Dordrecht/Boston/London: D. Reidel. Emotions, Argumentation, and Argumentativity 399 © Thierry Herman and Dimitris Serafis. Informal Logic, Vol. 39, No. 4 (2019), pp. 373– 400 Perelman, Chaïm, and Lucie Olbrechts-Tyteca. 1969 [1958]. The new rhetoric: A treatise on argumentation. Notre Dame: Notre Dame University Press. Plantin, Christian. 1998. Les saisons des émotions. In: Forms of argu- mentative discourse, ed Marina Bondi, 3-50. Bologna: CLUEB. Plantin, Christian. 1999. Arguing emotions. In Proceedings of the fourth international conference of the international society for the study of argumentation, eds. Frans H. van Eemeren, Rob Grootendorst and J. Antony Blair, 631-638. Amsterdam: SicSat. Plantin, Christian. 2004. On the inseparability of reason and emotion in argumentation. In Emotion in dialogic interaction, ed. Edda Weigand, 269-281. London: Benjamins. Plantin, Christian. 2011. Les bonnes raisons des émotions: Principes et méthode pour l'étude du discours émotionné. Berne: P. Lang. Psychogios, Dimitris. 2004. Τα έντυπα Μέσα επικοινωνίας: Από τον πηλό στο δίκτυο [Print media: From the clay to the network]. Athens: Kastaniotis. Rigotti, Eddo, and Sara Greco Morasso. 2010. Comparing the argumen- tum model of topics to other contemporary approaches to argument schemes: The procedural and the material components. Argumenta- tion 24(4): 489-512. Rigotti, Eddo, and Sara Greco. 2019. Inference in argumentation. A topics-based approach to argument schemes. Cham: Springer. Robin, Christian. 2009. L’angle journalistique. Paris: CFPJ. Rocci, Andrea. 2017. Modality in argumentation: A semantic investiga- tion of the role of modalities in the structure of arguments with an application to Italian modal expressions. Dordrecht: Springer. Schär. Rebecca, G. 2018. An argumentative analysis of the emergence of issues in adult-children discussions. PhD dissertation, Università del- la Svizzera italiana. Serafis, Dimitris. 2017. The construction of the Greek crisis in the public sphere: Discourses and emotions (pathos) under critical perceptions. PhD dissertation, University of Neuchâtel and Panteion University. Serafis, Dimitris, and Thierry Herman. 2018. Media discourse and pathos: Sketching a critical and integrationist approach – Greek and French headlines before the Greek referendum of 2015. Social Semi- otics 28(2): 184-200. Serafis, Dimitris, E. Dimitris Kitis, and Argiris Archakis. 2018. Graffiti slogans and the construction of collective identity: Evidence from the anti-austerity protests in Greece. Text & Talk 38(6): 775–797. Stede, Manfred, and Jodi Schneider. 2019. Argumentation mining. San Rafael, California: Morgan & Claypool Publishers. 400 Herman & Serafis © Thierry Herman and Dimitris Serafis. Informal Logic, Vol. 39, No. 4 (2019), pp. 373– Tindale, Christopher. W. 1999. Acts of arguing: A rhetorical model of argument. Albany NY: State University of New York Press. Toulmin, Stephen E. 2003. The uses of argument (Updated edition). Cambridge: Cambridge University Press. Vasilopoulou, Sofia, and Daphne Halikiopoulou. 2013. In the shadow of Grexit: The Greek election of 17 June 2012. South European Society and Politics 18(4): 523-542. Vasilopoulou, Sofia, Daphne Halikiopoulou, and Theofanis Exadaktylos. 2014. Greece in crisis: austerity, populism and the politics of blame. Journal of Common Market Studies 52(2): 388–402. Wales, Katie. 2014. A dictionary of stylistics. London and New York: Longman. Wiebe, Janyce, Wilson, Theresa and Claire Cardie. 2005. Annotating expressions of opinions and emotions in language. Language Re- sources and Evaluation 39(2): 165–210. Zampa, Marta. 2017. Argumentation in the newsroom. Amster- dam/Philadelphia: John Benjamins.