Title of the Paper [16 point font] Emotions in Argumentative Narration. The Case of the Charlie Hebdo Attack SARA CIGADA Department of Linguistic Sciences and Foreign Literatures Università Cattolica del Sacro Cuore Largo Gemelli 1 20123 Milan Italy sara.cigada@unicatt.it Abstract: This paper studies emo- tional inferencing triggered by emo- tion terms, using Pragma-Dialectics and the Argumentum Model of Topics. The corpus, in French, is an excerpt of a video-recorded testimony in which a middle school teacher evokes her experience of being in class the day after the Charlie Hebdo attack, thus presenting a case of argumentation in context. The analy- sis focuses on the argumentative structure and on the rhetorical strate- gies that trigger emotional inferenc- ing. The emotional inferencing derives from a Locus of Ontological Implication, which links a situation and an emotion (and vice-versa), while the culture-bound elements tend to be part of the endoxon. Résumé: L’article étudie l’inférence déclenchée par l’usage de termes d’émotion, en utilisant comme mo- dèle de référence la Pragma- dialectique et le Modèle AMT (Ar- gumentum Model of Topics). Le corpus est constitué par l’enregistrement vidéo d’une ensei- gnante de collège dans la banlieue parisienne, qui évoque son expérience en classe le jour d’après l’attentat à Charlie Hebdo en janvier 2015. Nous avons analysé ce corpus en tant qu’exemple d’argumentation en contexte, en en considérant la struc- ture inférencielle et les stratégies rhétoriques qui déclenchent l’émotion. L’émotion dérive d’un Topos de l’Implication Ontologique, qui relie une situation à une émotion (et vice-versa), tandis que les élé- ments culturels ont tendance à appa- raître dans la fonction d’Endoxon. Keywords: Argumentum Model of Topics, emotional inferencing, emotion lexicon, French discourse, ontological implication, rhetorical strategies, Rodol- phus Agricola mailto:sara.cigada@unicatt.it 402 Cigada © Sara Cigada. Informal Logic, Vol. 39, No. 4 (2019), pp. 401–431 1. Introduction This paper explores the role played by emotions in knowledge- oriented argumentation. It aims to fill a knowledge gap concerning the function of emotions as an argumentative hint and an inferen- tial trigger. My analysis stems from “emotion terms” (Plantin 2004, p. 269) – in French, “termes d’émotion” (Plantin 1999 and 2011) – in an argumentative narration (Olmos 2017; Thierry 2008). The Argu- mentum Model of Topics (AMT) (Rigotti and Greco Morasso 2010) combined with Pragma-Dialectics (van Eemeren and Grootendorst 2004) form the theoretical frame for this contribu- tion. The choice is appropriate for the purposes of the analysis because the AMT is particularly suited to the semantic analysis of micro-excerpts in which linguistic structures displaying emotions (more precisely lexicon in the case of emotion terms) convey explicit and implicit information and work together with the other linguistic structures to rationally support a standpoint (Palmieri 2014). The corpus is an excerpt from a video-recorded testimony in which a French middle school teacher, Martine, evokes her experi- ence of being in class the day after the 2015 terror attack on the offices of the Parisian magazine, Charlie Hebdo. Data are particu- larly relevant for exploring emotions in knowledge-oriented argu- mentation as they refer to unexpected tragic events arousing dis- rupting, dysphoric emotions. It is worth highlighting that Mar- tine’s discourse does not aim to move the audience, but rather to share professional competence and experience about the manage- ment of highly emotional situations in an educational context with peers. In this paper, I am not referencing the opposition between “emotive” and “emotional” (“die Emotive” and “die Aussagen”, see Marty 1908, pp. 363-364 and passim, and Caffi 2007, p. 139, 177). In fact, as already pointed out by Plantin (2011, pp. 137- 142), we have too little linguistic evidence for this distinction. Rather, I analyze linguistic and rhetorical structures related to emotions insofar as they contribute to argumentative sense-effects. Emotions in argumentative narration 403 © Sara Cigada. Informal Logic, Vol. 39, No. 4 (2019), pp. 401–431 Given this, I will use a nonspecific “emotional” as a version of Plantin’s “émotionné.” The AMT allows the analyst to show the different ways in which emotion terms used in narration activate inferential struc- tures (Ungerer 1997, “emotional inferencing”). In the current paper, in the reconstruction of the so-called “Y structure,” emotion terms appear, with different functions, both on the right side, i.e. in the procedural-inferential component of the Y structure (maxim), on the left side, i.e. in the material-contextual component (data), and in the final conclusion. In any case, emotion terms convey information that forms a coherent part of the discourse meaning. Different communities described in narrative discourse may have good reasons to react in emotionally different ways to the same, unique event according to their personal and communal convic- tions, depending on whether or not they share the endoxon. Alt- hough based on current models (Plantin 2011; Micheli 2014) and partly consistent with them, this description of emotions in argu- mentation is much more precise and accords linguistic devices more relevance. In fact, the reconstruction shows that co-occurring linguistic structures (rhetorical strategies) take part in triggering emotional inferencing. Consistently with the models I refer to, I suggest that emotions displayed in discourse should be described in a “reasonableness” model in which rhetorical devices are accounted for by the same method as the other components of discourse (Rigotti and Palmieri 2016; cf. Gobber 2019, p. 292). 2. Emotions in discourse and in argumentation: accounting for linguistic data 2.1. The termes d’émotion This paper considers a single layer of the complex phenomenon of “emotion” in discourse, i.e. the presence of direct or indirect “emotion terms,” as an important first step towards analyzing the implicit emotional component in discourses that do not mention emotions either directly or indirectly. 404 Cigada © Sara Cigada. Informal Logic, Vol. 39, No. 4 (2019), pp. 401–431 I start by considering emotion terms. Firstly, emotion terms are considered with the procedure identified by Plantin and further applied and developed by Micheli. Plantin considers emotions to be an intrinsic element of discourse, and Micheli (2010) agrees that emotions should not be considered as external elements “help- ing” to persuade, but that they should be analyzed as the actual object of argumentative constructions. Plantin’s methodology is lexicon-based and consists of identifying “emotion terms,” i.e. words that mean emotions. In French, the language I focus on, as in English, these words may be nouns, adjectives, verbs, adverbs and interjections (e.g. joy, afraid, I hope, desperately, ouch!). The emotion terms occur in an “emotion sentence,” which “asserts or denies that a particular individual or experiencer is in the grip of a particular emotion or psychological state” (Plantin 2004, p. 269; “experiencer” for Fr. lieu psychologique) on the basis of a specific “reason” (Fr. source) that emerges in discourse as the cause of that emotion. This kind of analysis is useful to describe the layer of emotions that are explicitly mentioned in the discourse. In addi- tion, Plantin suggests analyzing “indirect” emotion terms (e.g. to tremble, to wring one’s hands), i.e. words that indicate typical physiological or attitudinal reactions associated with specific emotions. The same should be done with idioms that indicate emotions through metaphor (e.g. her heart was on fire; his eyes betray darkening storm). As both indirect emotion terms and emo- tion idioms are highly language- and culture-dependent, they are often comprehensible within a specific community, but barely (or not at all) outside of it. This is relevant for the intersubjective understanding of emotion terms. The addressee is able to correctly interpret and understand the emotions attributed to individuals on the basis of the reasons told in a discourse (obviously, the individual to whom an emotion is attributed may be the speaker himself. See Plantin 2004, p. 269). On this basis, a reader is able to understand the reasonableness of emotions that are, for example, attributed to characters in a novel or to real persons in a narrative excerpt, e.g. in a news story. The reasons given to support the experiencer’s emotions are usually rhetorically elaborated so as to create euphoric or dysphoric ef- fects. The linguistic devices are described by Caffi and Janney Emotions in argumentative narration 405 © Sara Cigada. Informal Logic, Vol. 39, No. 4 (2019), pp. 401–431 1994, Ungerer 1997, Plantin 1999 (cf. Cigada 2006), Micheli 2008, Plantin 2011, Cigada 2016. However interesting, these descriptions do not explain the inferential procedure by which emotion arising from “situational” elements is transferred to the experiencers. Indeed, the intersubjective understanding of emotion terms not only requires a mutual linguistic interpretation of emotion terms themselves1, but also a community dimension in the understanding and evaluation of the situation. As Micheli (2008, pp. 144-148) points out, the arguability of emotions is founded in their cognitive reasons (“a set of beliefs and judgments regarding this situation”, Micheli 2010, p. 7 and passim). It is relevant that the addressee shares the evaluation backing the emotional interpretation of a situation; however, doing so presupposes sharing both “contextual beliefs and judgments” and a specific ontology that justifies the inference from that situation to that emotion. 2.2. “Emotional inferencing” revisited using the Argumentum Model of Topics The situation that underlies and justifies an emotion is often re- ferred to in terms of “topics” (Ungerer 1997; Micheli 2008; Plan- tin 1999, 2011 and 2018), understood as a set of linguistic- cognitive devices that are able to trigger “emotional inferencing” (Ungerer 1997, p. 310) such as: discursive orientation to the inter- locutors, proximity in time and space, quantity and analogy, which are represented via the deictic system, quantification, evaluative lexicon and so on. According to Ungerer, emotional inferencing in news stories is based both on “principles” (including proximity, animacy, rank and number) and on “strategies” (emotional evalua- tion, intensity, negative content, homocentrism); principles and strategies are expressed by respective “linguistic triggers/cues of emotional inferencing” such as, for instance, deictic items, kinship terms, “disaster vocabulary,” and so on (1997, pp. 310-319). Start- ing from Grice’s maxims, Ungerer notes that “although the infer- 1 Plantin makes some almost implicit references to the fact that emotion terms in French may be different to those in other languages but gives no systematic account of this (Plantin 1999, p. 206 and p. 218 note 4). 406 Cigada © Sara Cigada. Informal Logic, Vol. 39, No. 4 (2019), pp. 401–431 encing is carried out by the reader, the maxims must also be ob- served by the author/editor to become effective.” He suggests that these might be found in the practical guidelines given to the jour- nalists or in manuals about “newsworthiness.” In fact, “the sub- stance of the maxims, which are at the heart of the inferencing process, must come from elsewhere” and not from the news stories themselves (Ungerer 1997, p. 310). I think that it is a worthwhile exercise to investigate “emotional inferencing” using Pragma-Dialectics combined with the Argu- ment Model of Topics, which is specifically designed to study inference and inference triggers in argumentative discourse. In Plantin’s model, emotions are identified and accounted for as the standpoint of argumentative processes, as there are “good reasons” that justify them and make them inter-subjectively under- standable. In Ungerer’s model, a reader’s emotions are the result of the content and linguistic features displayed by discourse. AMT is useful in providing a more precise insight into the phenomenon. 3. Situated argumentation: Martine’s testimony in its context The corpus is situated in many respects. As a discursive situation, Martine’s testimony is an autobiographical narration, that is, a kind of discourse in which emotions are likely to play an important role. Some emotions are discursively auto-attributed; others are attributed to other individuals. Her testimony takes places within the wider context of an international life-long learning program based in Italy and targeted at Lebanese French-speaking teach- ers—S’éduquer pour éduquer (Educating oneself to educate)— which took place from 2014 to 2017. I took part in the program as a trainer during the first year and subsequently as the project lead- er. Over the years, there was some turnover among participants and some of us took on different roles, thus creating a complex discourse community. The main group of stakeholders in the project was constituted of approximately forty Lebanese teachers from six French-speaking schools, both male and female, Maro- nites, Orthodox and Muslims, teaching different subjects across the full age range from kindergarten to high school (see Moeller 2010 on French-speaking schools in the context of Lebanese edu- Emotions in argumentative narration 407 © Sara Cigada. Informal Logic, Vol. 39, No. 4 (2019), pp. 401–431 cation system). Even if the teachers have no voice in the corpus analyzed in this paper, they were its main intended audience. The witness and author of the discourse is Martine, a French teacher in her forties. Her institute, a middle school situated in the suburbs of Paris, draws pupils from different cultural, religious and social backgrounds. As she had told me about her school on a previous occasion, I specifically asked her to video-record her testimony for the project. In the video, among other things, she recalls her experience of being at school the day after the terror attack on Charlie Hebdo2. In terms of the corpus, she is the speak- er. Within the context of the whole project, this session aimed to exchange ideas about the role of the teacher in potentially conflic- tive situations, discussing whether and how they are called to mediate between opposing positions and whether it is acceptable for them to take part in the dialogue supporting one of the parts. More generally, the issue at stake in the video is whether the teacher should be neutral or play an active role in the transmission of the tradition of their community3. The other stakeholders were the Italian NGO that was hosting the project, Fondazione AVSI, and its local partners (Caritas Lebanon and La Femme Libanaise). I myself was part of a group of about fifteen Italian trainers who evaluated and selected the 2 On 7th January 2015, two French terrorists attacked the offices of the satirical weekly Charlie Hebdo in Paris killing twelve people. Al-Qaeda took responsi- bility for the attack. France24 wrote the following about the attack: “Charlie Hebdo’s caricatures frequently caused outrage by poking fun at religion and politics. They were not popular with everyone, but in killing the country's most famous cartoonists the gunmen had targeted a core French value: freedom of speech” (http://webdoc.france24.com). 3 In recent years, Lebanon has experienced massive legal and, according to the laws of the country, illegal immigration from Syria and other Middle East countries, resulting in social tensions. The teachers from the South and East regions of Lebanon had been reporting terror attacks in their towns or villages, which caused a general state of insecurity. Schools were accepting thousands of foreign pupils scarred by war traumas. Schools, traditionally open to all, were facing internal and external pressures and reflecting violent socio-political conflicts. The vast majority of pupils at the French-speaking Christian school Saints-Cœurs in Baalbek, for example, were Muslim. For a description of the six schools, their statistics, geographical situation, social and religious context, see Marta Piccinelli 2019. http://www.france24.com/en/20150107-charlie-hebdo-satirical-weekly-islamists-prophet-cartoons-deadly-attack http://www.france24.com/en/20150107-charlie-hebdo-satirical-weekly-islamists-prophet-cartoons-deadly-attack http://webdoc.france24.com/ 408 Cigada © Sara Cigada. Informal Logic, Vol. 39, No. 4 (2019), pp. 401–431 video, supported by two experts in international cooperation. One of the most important reasons for this choice was the excerpt I am studying in this paper: the autobiographical reference to an ex- tremely violent situation at school and the deadly emotions con- nected with it. This description of the communicative context aims to explain why the video-recording analyzed in this paper should be consid- ered as dialogical in nature, even if it consists of a monologue: in reality, it singles out one moment of a highly dialogical process. The whole of the project, in fact, may be considered as the pro- gressive construction of a communicative exchange between the stakeholders (Palmieri and Mazzali-Lurati 2016) in which this specific communicative act has been designed with the purpose of activating dialogue between the participants. Martine shows a strong commitment to argumentation and gives her speech a clear argumentative structure. She starts by presenting her thesis about the issue at stake, which is the meaning the schooling, and presents her supporting arguments (minute 00’00’’ to 01’26’’). Then she explains in more detail what she has in mind by means of two autobiographical anecdotes, the second of which is the corpus4. 4. From the video recording to the transcriptions: methodolog- ical questions and the analyst’s position In terms of analytic methodology, the use of public video- recording lowers the observer’s paradox. The video was recorded with the purpose of being watched and discussed during that pro- ject session so all the audiences mentioned in the previous para- graph were considered and accounted for by the speaker. Martine set no limits on the use of the video but requested to remain anon- ymous. 4 This narrative excerpt is explicitly put forward as an “exemplum,” in the sense described by Plantin (2018): “The illustrative example facilitates the under- standing of a concept or a law, by introducing a (typical) instantiation of the concept or the law: ‘A migratory bird is a bird that ... So the swallow...’ Moreo- ver, if the example chosen is (presented as) typical of the phenomenon, it renders the time-consuming and precarious work of checking a large number of cases unnecessary” (p. 264, s.v. Example). Emotions in argumentative narration 409 © Sara Cigada. Informal Logic, Vol. 39, No. 4 (2019), pp. 401–431 The setting is very simple. The speaker is sitting alone in a classroom and addresses the video-camera. The whole recording is 07’18’’long. The analysis considers the section from 02’58’’ to 03’50,’’ which corresponds to the opening of the part in which Martine recalls how the shooting at the Charlie Hebdo magazine affected her pupils (02’58’’ to 06’25’’) ten months previously. The transcription is a simple textual transcription as is usual in Pragma-Dialectics and AMT. This kind of transcription fits well with the purpose of the analysis, which focuses on the lexical and semantic layer of discourse. However, the whole video has been transcribed according to the French transcription conventions ICOR. This transcription will be useful for further inquiries and for crosschecking emotions in the spoken discourse. As Plantin (2011) explains, when analyzing emotions in a discourse, the analyst risks self-identification and self-confusion with the addressee (pp. 194-196). Checking other discourses pro- duced in the same context and community, in order to avoid the risk of emic interpretation, is a key part of the methodology. This kind of checking allows one to measure and define the different emotional reactions to an argument for different individuals. It is important but not sufficient to say that one needs to know “the set of beliefs and judgments which are most commonly associated with this particular emotion” (Micheli 2010, p. 7) because this association changes from one time and one community to another, so that the “good reasons” for an emotion are not absolute, but should be considered in relation to a specific context. Lebanese teachers, for instance, displayed empathy towards Martine’s feel- ings about the Charlie Hebdo attack and appreciated the way she managed the class. One of them, however, commented on Mar- tine’s evident distress saying: “Elle est française, elle a pas l’habitude” (She is French, she is not used to this). After which, they went on discussing the best time and approach for talking to pupils after terror attacks.5 5 The corpus also consists of the c. 2h30’ video-recording of the training session with the Lebanese teachers based on the discussion of Martine’s testimony and transcripts of interviews given to participants. Lebanese teachers aligned them- selves with the “reasonableness” side (dialogue as a radical alternative to violence) and, subsequently, with the teacher’s standpoint of “non-neutrality.” 410 Cigada © Sara Cigada. Informal Logic, Vol. 39, No. 4 (2019), pp. 401–431 Here is the original French excerpt: J'ai aussi beaucoup appris de ce qui s'est passé dans notre éta- blissement l'an dernier à la suite des attentats du mois de janvier. Comme il y avait un certain nombre de tensions entre communau- tés dans notre établissement, et qu'un certain nombre d'élèves s'était félicité des attentats à Charlie Hebdo, nous avions reçu comme consigne de ne pas aborder le sujet en classe, de l'éviter le plus possible ou de nous contenter tout au plus d'un rappel géné- rique à la tolérance. Il se trouve que j'avais, cette année-là, une classe très particulière, avec beaucoup d'élèves dans des situa- tions extrêmement complexes d'un point de vue personnel, quelques délinquants, et dans laquelle nous avions déjà été ame- nés à intervenir pour des tensions entre communautés confession- nelles. Donc, inutile de vous dire que la tension était grande ce matin-là, le matin du 8 janvier avant mon entrée dans cette classe. In the English version, the first introductory utterance has been put in parenthesis as it is not considered in the analysis. Some foot- notes and comments have been inserted, and the corpus has been divided into three main utterances. The two emotion terms that will be analyzed in this paper are marked in bold. In the analysis, references to the French text will be made where necessary in order to take into account the specific contribution of rhetorical strategies (Jacobs 2006, p. 428). In fact, one cannot pretend that translation preserves discursive sense effects: [I also learnt a lot from what happened in our school last year 6 , follow- ing the attack in January.] (Utterance 1) As there was a certain number of tensions between the communities in our school, and a certain number of pupils had expressed appreciation for the attack on Charlie Hebdo, we had been In fact, Martine’s discourse provides an example of the teacher’s role in conflict mediation, not as an external director, but as an active promoter of the choice in favor of dialogue (“conflict should not have the last word”, Greco 2017, p. 29) and as a guarantor of space to think (Perret-Clermont 2005). In a sense, this is not only the short-term premise of the project S’éduquer pour éduquer, and nor it is just that of the school as an institution, but it is instead an essential basis of human reasonable get-togetherness. 6 Martine refers to the past school year 2014-2015, while recording her video at the beginning of the new school year 2015-2016. Emotions in argumentative narration 411 © Sara Cigada. Informal Logic, Vol. 39, No. 4 (2019), pp. 401–431 given the order not to address the subject in the classroom, to avoid it whenever possible or at the very most to confine our remarks to generic reminders about tolerance. (Utterance 2) That year, as it happened, I had a very particular kind of class, where there were a lot of pupils with extremely complex per- sonal circumstances, as well as some young offenders, and we had already been forced to intervene due to tension between the different religious communities. (Utterance 3) So needless to say, tension was high on that morning, the morning of January 8 th , before I entered the class. 5. Two examples: emotion as an argument and as a standpoint In the following sections I explain how, in the first example, an emotion term plays the role of datum in Utterance 1 (5.1). I then analyze emotion in the role of the conclusion in the same Utter- ance (5.2) and in Utterance 3 (5.4). 5.1. Emotion as an argument: “had expressed appreciation” (s’était félicité) According to Plantin’s classification of emotion terms, se féliciter should be considered an indirect term, i.e. a term representing the typical7 expression of an emotion (joy, satisfaction). Utterance 1 is characterized by a polyphonic phenomenon because the argumen- tation is reported; Martine reports the reasoning of the school. That is why this passage should be considered as a piece of argumenta- tion and not as an explanation. As for argumentative structure, the speaker explicitly (cf. connective as) reports the reasons behind the order not to address the subject of the attacks. The issue at stake is therefore: “What should we do to prevent more tensions arising in our school, which are highly likely and have already started to manifest themselves in the light of yesterday’s Charlie Hebdo attack?” The issue might in principle admit alternative solutions. In terms of a pragma-dialectical analytical overview (van Eemeren 2018, pp. 90-91; van Eemeren and Grootendorst 7 As mentioned, “typicality” should be referred to a specific community, both for the connection between an event and the emotional response to it and for the way the emotional response is expressed. 412 Cigada © Sara Cigada. Informal Logic, Vol. 39, No. 4 (2019), pp. 401–431 2004, pp. 120-121), the standpoint is formulated in the following three passages and is supported by a coordinative argumentation (i.e. the two arguments work together to support the standpoint): 1. [it is reasonable] to avoid the subject whenever possible or at the very most to confine our remarks to generic reminders about tolerance 1.1a as there was a certain number of tensions between the commu- nities in our school 1.1b and a certain number of pupils had expressed appreciation for the attack on Charlie Hebdo In some cases, such as this one, “components of coordinative argumentation are part of a complex datum within one inferential configuration under the same locus” (Rigotti and Greco 2019, p. 233). In fact, these coordinative arguments should be described as inferentially linked to each other in the sense that both are neces- sary to support the standpoint (Palmieri 2014, p. 40); “1.1a and 1.1b would represent different premises within the same inferential configuration; they would both be included within one AMT reconstruction” (Rigotti and Greco 2019, p. 231). The argument 1.1a reveals one datum (the contextual situation of the school with a permanent conflict between opposing communities), while 1.1b describes another datum, that is, the recent reactions of some students who belong to a specific community anticipating the probable beginning of a new and stronger conflict event. That the students belong to a specific and “opposed” community is inferred from their emotional reaction to the attack. The school’s conclu- sion is triggered by the conjunction of the two arguments because the decision not to allow discussion about the attack was made subsequent to the students’ reaction in the context of a permanent conflict. The datum presented in 1.1b becomes relevant to the school’s conclusion not to allow discussion about the attack due to another inference in which the emotion term had expressed appreciation appears. The analysis of an emotion playing the role of argument has already been proposed by Rigotti and Palmieri (2016, p. 55), who apply the AMT to Solomon’s judgment. In the context of Solo- mon’s judgement, the emotional reaction of the true mother to the king’s proposal that the child be divided in two is taken into ac- Emotions in argumentative narration 413 © Sara Cigada. Informal Logic, Vol. 39, No. 4 (2019), pp. 401–431 count by Solomon as a datum included in the material starting point. This permits Solomon to conclude that that woman must be the mother. Emotion enters the argument supporting the stand- point. In the excerpt, the following Y structure can be reconstruct- ed (Figure 1): Locus: Ontological Implications (between the ontology of a sit- uation and what the situation implies: the habitudo links the sec- ond pole to the first one, see Discussion in paragraph 6) Maxim: Euphoric emotions are the sign that one is experiencing a positive situation Major Premise (Endoxon): Terror attacks like the attack on Charlie Hebdo are horrific situations Minor Premise (Datum): These students have expressed appre- ciation for the attack on Charlie Hebdo First Conclusion: These students have expressed appreciation for a horrific situation Final Conclusion/ Standpoint: These students experience the Charlie Hebdo attack as a positive situation Discussion. As we can see from this Y reconstruction, emotion is a part of the material starting point and can be found both in the datum (the manifestation of euphoric attitude) and in the endoxon (as a relevant component in the evaluation of attacks in general). Nonetheless, the inference is only possible on the basis of the maxim that links euphoric emotions to a specific kind of situation. I am not discussing here the problem of a symptomatic inference (cf. Rigotti and Greco 2019, p. 249 and 259; Plantin recurrently refers to emotions as “syndromes” and see Plantin 1999, pp. 209- 211 on the “topics” of emotions). I will come back to this point in section 6. So, the final conclusion that “These students experience the attack on Charlie Hebdo as a positive situation” is one of the effec- tive meanings of 1.1b. 414 Cigada © Sara Cigada. Informal Logic, Vol. 39, No. 4 (2019), pp. 401–431 Figure 1: Emotion as an argument Endoxon: Terror attacks like the attack on Charlie Hebdo are horrific situations Datum: These students have expressed appreciation for the attack on Charlie Hebdo Locus Ontological Implication Maxim: Euphoric emotions are the sign that one is experiencing a positive situation First Conclusion/ Minor Premise: These students have expressed appreciation for a horrific situation Final Conclusion: These students experience the attack on Charlie Hebdo as a positive situation 5.2. Emotion as an implicit standpoint However, one should evoke another relevant Y structure (Figure 2) in which emotion occurs in the final conclusion (as set out in 5.4): Locus: Ontological Implications (between the ontology of a situation and what the situation implies: the habitudo links the first pole to the second one, see Discussion in paragraph 6) Maxim: Those who experience horrific situations feel trauma- tized immediately afterwards Major Premise (Endoxon): Terror attacks like the attack on Charlie Hebdo are horrific situations Minor Premise (Datum): We experienced the attack on Charlie Hebdo just yesterday First Conclusion: We have just experienced a horrific situation Final Conclusion/ Standpoint: We feel traumatized Emotions in argumentative narration 415 © Sara Cigada. Informal Logic, Vol. 39, No. 4 (2019), pp. 401–431 Figure 2: Emotion as an implicit standpoint Endoxon: Terror attacks like the attack on Charlie Hebdo are horrific situations Datum: We experienced the attack on Charlie Hebdo just yesterday Locus Ontological Implication Maxim: Those who experience horrific situations feel traumatized immediately afterwards First Conclusion/ Minor Premise: We have just experienced a horrific situation Final Conclusion: We feel traumatized Discussion. Under the same locus and with the same endoxon as in Figure 1, another relevant Y structure is reconstructed. The datum is almost completely ostensive (it is contextual evidence); none- theless, it is explicitly evoked by the mention of the attack. This second inference completes the meaning of 1.1b. The contempora- neous activation of these two Ys creates two pairs of conflicting meanings—the first combining the standpoint “These students experience the attack on Charlie Hebdo as a positive situation” and the endoxon “Terror attacks like the attack on Charlie Hebdo are horrific situations,” and the latter combining the standpoint “We feel traumatized” and the datum “These students have expressed appreciation for the attack on Charlie Hebdo.” The association of 1.1a (past conflict) and 1.1b (present disrup- tive conflict) suggests terminating confrontation in order to pre- vent dangerous consequences. The school’s reasoning is based on the Locus of the Opposite (between the community rejecting the attack and the community supporting it) and the Locus of Termi- nation and Setting up. As Rigotti and Greco (2019) put it: when one uses the locus from termination and setting up, he or she is comparing the current situation with a future possible world. Following this comparison is an evaluation about the opportunity 416 Cigada © Sara Cigada. Informal Logic, Vol. 39, No. 4 (2019), pp. 401–431 to set up (start or begin) an activity or to cease (‘terminate’) an ex- isting state of affair (p. 263). 8 5.3. First results from the example “had expressed appreciation” (s’était félicité) Regarding emotion and procedural-inferential structure: The role of the emotion term had expressed appreciation that occurs in the datum is crucial in triggering the inferences that invite the audi- ence to accept the conclusion (the standpoint of the school). In fact, although occurring in the left material-contextual side of the Y structures, this verb plays an important role in activating the relevant loci and maxims and functions as an “invitation to infer” (Rocci 2017, p. 103). It is thanks to this element that the audience is able to evaluate the kind of tensions at play and their intensity and to compare them with those previously mentioned (in 1.1a). Thus, the reasonability of such a strong preventive measure as forbidding teachers to talk about Charlie Hebdo is validated (cf. Rocci commenting on Jacobs 2006, in Rocci 2017, p. 103-104). Regarding the discursive strategies and sense effects: As Scott Jacobs points out, “strategy improves the quality of reasoning” (2006, p. 428). That is why the linguistic devices contributing to Martine’s argumentation are systematically taken into account here. It must be said that reporting the school’s reasoning does not coincide with sharing it as Martine clearly shows by her wording. In fact, Martine’s discourse in general is characterized by an un- derstated attitude, a high level of self-control (evidenced by the transcription of the recording, which is not analyzed here), and a non-emotional lexicon. For instance, she uses the word tensions to describe the conflicts and the quantification a certain number (un certain nombre, two occurrences) to quantify both the conflicts and the pupils welcoming the attack. Mitigation strategies contrib- ute to the atmosphere of tension in her narration because they are interpreted as a typical cautionary reaction to a risk situation. Nonetheless, with the exception of just this one passage, her point of view and first-person perspective is clear throughout her testi- mony. In this passage, by contrast, she adopts a specific perspec- 8 On this locus, see also Pollaroli and Rocci 2015, p. 187. Emotions in argumentative narration 417 © Sara Cigada. Informal Logic, Vol. 39, No. 4 (2019), pp. 401–431 tive, which is manifested linguistically when she reports the deci- sion to not allow debate about the attacks (nous avions reçu): recevoir la consigne {x₁, x₂}, in fact, is the semantic conversive of donner la consigne {x₂, x₁}, highlighting the receiver’s passive role (Mel’chuk 1988, p. 33; Gatti 1992; Rigotti and Cigada 2013, pp. 100-1019). Moreover, this role is represented by the plural nous (we, the teachers): so Martine as an individual “disappears” in the category targeted by the order10. The argumentative struc- ture of her narration appears in explicit connectives (as…., and…, we had been given the order…). The utterance is completely de- void of singular first-person pronouns or possessives, which, conversely, appear both before and after this passage thus creating an effect of distance from the school’s reported standpoint. The argumentative relevance of these linguistic hints is the activation of polyphony and the discursive emergence of Martine’s personal thesis about the issue. It could be represented under another endoxon, such as something along these lines: “Whatever happens, teachers are committed to managing their class, fulfilling the institutional raison d’être of educating pupils to become good citizens (who share the principles and laws of the Republic) ac- cording to the orders given by the Ministry and the school’s prin- cipal unless they contradict the institutional raison d’être.” Mar- tine’s personal conclusion is based on a maxim derived from the Locus of the Final Cause: “If the end must be reached, no possible obstacle may be considered as a good reason to stop the action,” so that all the problems and risks mentioned as data cannot be con- sidered as a good reason not to act11. 9 French recevoir la consigne is a Lexical Function Oper2, i.e. the conversive (Conv21) of Oper1 donner la consigne. The English version we had been given the order exploits a different syntactic structure which is possible in English but not in French, that of passive diathesis with the receiver in the subject’s role. 10 This would be an actantial shield in Caffi’s terminology (Caffi 2007, p. 107). I am grateful to an anonymous reviewer for highlighting this point. In the excerpt, mitigation strategies are consistent with the argumentation of emotions. 11 This interpretation is consistent with the preceding part of the testimony where she had already evoked a strong personal commitment to her institutional task (cf. video-recording 00’00’’ to 01’26’’), but I am not completing this Y structure as it is not fully relevant to my point. 418 Cigada © Sara Cigada. Informal Logic, Vol. 39, No. 4 (2019), pp. 401–431 5.4. Emotion as a standpoint: tension Let us consider the occurrence of the emotion term tension in Utterance 3 of the excerpt in the standpoint function. Martine’s discourse, in fact, narratively justifies a “personal high-tension situation.” The AMT Model will allow the complex reasoning supporting this final conclusion to be reconstructed. The totality of the excerpt is considered as argumentatively answering the issue at stake (“Shall I talk about yesterday’s attack in my class?” i.e. “Do I agree that the combination of risks is a good reason to act against a relevant institutional commitment?” but also, more generically, “How shall I manage this situation?”). Utterance 1, introduced by “as,” concluded with the (reasonable but still not-completely- shared-by-the-speaker) order not to talk about the attacks with pupils, thus indicating Martine’s misalignment with the received order. This part constitutes a first datum resulting from two op- posed institutional commitments: should she discuss it with the pupils, she would be disobeying an explicit institutional order12; and should she not discuss it with the pupils, she would be failing in any teacher’s general educational commitment. Even if she does not agree with the school’s conclusion based on the Locus of Termination and Setting Up, Martine still shares the school’s interpretation of the attitudes of certain students (had expressed appreciation) as being the opposite of the expected one thus iden- tifying with one of the opposing communities. The second part of her narration (Utterance 2, il se trouve que) adds further elements to the material-contextual situation, specify- ing the circumstances of Martine’s class and, subsequently, her personal involvement in the situation, that is, she is responsible for what can be told in class (going against the ideological neutrality of laïcité), but also for what can be done (by problematic pupils and young offenders who have already been proven to resort to violence instead of dialogue). Martine’s personal emotional reac- tions to the attack and to the pupils’ reactions are also relevant elements (they are explicitly evoked later in the discourse). In fact, 12 This remains implicit in the video but is relevant to the understanding. In fact, Martine asked me never to mention either her real name or that of her school. Emotions in argumentative narration 419 © Sara Cigada. Informal Logic, Vol. 39, No. 4 (2019), pp. 401–431 when she enters the classroom, she has not yet made a decision about the issue at stake and she is still weighing up the conflict (tension) situation. How does the narration justify Martine’s tension before she entered the classroom that morning? Let us turn to the Pragma- dialectical analytical reconstruction, moving on from Utterance 1, which has already been analyzed, to focus on Utterances 2 and 3. 1. So needless to say, tension was high on that morning, the morn- ing of January 8th, before I entered the class (Utterance 3) 1.1a as there was a certain number of tensions between the communities in our school, and a certain number of pupils had expressed appreciation for the attack on Charlie Hebdo, we had been given the order […] (Utterance 1) 1.1b that year, as it happened, I had a very particular kind of class 1.1b.1a where there were a lot of pupils with extremely complex personal circumstances 1.1b.1b as well as some young offenders 1.1b.1c and we had already been forced to intervene due to tension between the different religious communities In Pragma-dialectical terms, 1.1b represents a second argument and a subordinative argumentation: “subordinative argumentation is typically adopted when the argument justifying the standpoint does not constitute a shared premise” (Palmieri 2014, p. 39), so one needs to provide further arguments to support that premise (cf. Rigotti and Greco 2019, p. 233). In fact, as it is not a shared prem- ise that Martine’s class was of a very particular kind, so the speak- er needs to provide some arguments supporting this standpoint. The AMT highlights the fact that the procedural-inferential process that allows for the standpoint to be determined is based on a Locus of Ontological Implications: “This locus builds on the relation between the nature of an entity […] [in our case, the on- tology of a situation] and what this nature implies.” (Greco et al. 2018, p. 452). As Rigotti and Greco (2019, p. 254) put it, “[onto- logical implication] is active in reasoning processes such as the following: “We must thank him, because he has done a good job.” 420 Cigada © Sara Cigada. Informal Logic, Vol. 39, No. 4 (2019), pp. 401–431 The implication of being grateful, in this case, depends on the “ontological” relation between doing [or better receiving] some- thing good and being grateful.” In the current case, the implication of being tense depends on the “ontological” relation between having to face a complex and dangerous situation and being tense: the habitudo links the second pole to the first one (see Discussion in 6). Ontological implication is combined with another locus, specif- ically, Proportion (“All the More,” cf. Rigotti and Greco 2019, pp. 262-263). In fact, given the comparison between the past and present experiences of the difficulty in fulfilling the teacher’s institutional commitment in an everyday context and in the present context (of the previous day’s attack and some pupils’ euphoric emotions), it becomes reasonable to predict that, in her specific class, which is particularly problematic, it will be not only difficult but almost impossible to manage the situation. Therefore, the maxim derived from the combination of these two loci might be sketched as follows: If someone is committed to managing a situation but present circumstances (which are worse than normal) probably make it impossible, there is an implication that the person will be tense about the situation. In order to make this maxim work in the actual context, some material requirements, i.e. the endoxon and the datum, need to be filled in. Major Premise (Endoxon): Teachers are committed to manag- ing their classes, fulfilling their institutional duty of educating pupils to become good citizens (who share the principles and laws of the Republic), according to the orders given by the Ministry and the school’s Principal. Minor Premise (Datum): 1.1a as there was a certain number of tensions…, and a certain number of pupils had expressed apprecia- tion for the attack…, we had been given the order… + polyphony effect (the reported conclusion contradicts the general institutional duty of a teacher) + Martine felt traumatized by the attack + 1.1b that year… I had a very particular kind of class (such that it was difficult to manage on a daily basis). The Premises in the material starting point support the First Conclusion: Martine is committed to managing a probably unman- Emotions in argumentative narration 421 © Sara Cigada. Informal Logic, Vol. 39, No. 4 (2019), pp. 401–431 ageable situation. Thus, combining the Maxim with this First Conclusion, the reasoning arrives at the conclusion of the Stand- point. Final Conclusion/ Standpoint: It was reasonable to feel very tense that morning before I entered the class (Donc, inutile de vous dire que la tension était grande ce matin-là, le matin du 8 janvier avant mon entrée dans cette classe), as in Figure 3. Figure 3: Emotion as a standpoint Endoxon: Teachers are committed to managing their classes, fulfilling their institutional duty of educating pupils to become good citizens (who share the Republic’s principles and laws), according to the instructions given by the Ministry and the school’s Principal Datum: there was a certain number of tensions; a certain number of pupils had expressed appreciation for the attack; we had been given the order…; the order contradicted the general institutional duty of a teacher; M. felt traumatized by the attack; M. had a very particular kind of class (difficult to manage on a daily basis) Loci Ontological Implication and All the More Maxim: If P is committed to managing S, but present circumstances (which are worse than others) probably make S unmanageable, there is an implication that P will be tense about S First Conclusion/ Minor Premise: M. was committed to managing a probably unmanageable situation Final Conclusion: It was reasonable to be very tense that morning before entering the class 5.5. First results from the example tension Regarding emotion and the procedural-inferential structure: Even though emotional contextual elements come from the material- contextual (left) side of the Y structure, the connection between that kind of situation and the speaker’s emotions, which activates 422 Cigada © Sara Cigada. Informal Logic, Vol. 39, No. 4 (2019), pp. 401–431 the inference, is on the right side of the Y structure. It is relevant to analyze the meaning of this “ontological implication” and the maxim that can be derived from it in more depth. In fact, Walton et al. (2008, p. 307) declare that, “some [loci] are based on logical-semantic properties and are necessarily true; others are on- ly plausible.” This indirectly suggests that, in order to verify the hold of the argument schemes, one has to analyze in depth the se- mantic structure of the inferential connections (maxims) on which they are based. In fact, a careful semantic analysis helps to identi- fy the conditions of validity of these connections (Rigotti and Greco Morasso 2010, pp. 492-493). Regarding discursive strategies and sense effects: In Utterance 3, the tension is presupposed as the definite article la shows, and the focus is on its intensity. The emotion term la tension is interesting because its occurrence in the conclusion represents the third occur- rence of the lexeme in the excerpt. However, the meaning of this last occurrence is manifestly different from the previous ones which denoted conflict between pupils. The two meanings are linked by polysemy so that the same word is used to denote both the situation and the implied emotion. This adds consistency and effectiveness to Martine’s narration. From a discursive point of view, the connective so (donc) shows the status of Utterance 3 as a conclusion, while needless to say (inutile de vous dire) underlines, through the rhetorical device of praeteritio, that this conclusion is considered self-evident by the speaker. The move is strengthened by the direct appeal to the audience (inutile de vous dire instead of inutile de dire) and by the speaker’s self-deixis (mon entrée dans cette classe) but also by the insistence on evidence about the precise timing (ce matin-là, le matin du 8 janvier, avant mon entrée). The syntactically trans- formed phrase with an abstract noun naming the action (mon entrée dans cette classe instead of avant que je ne rentre dans cette classe),13 not a preferential structure (Gardes Tamine 2015; Rigotti and Rocci 2006, p. 234) in the French language, also con- tributes to the pathetic effect suggesting the reification of the 13 Not preserved in English translation. See Rigotti (1979, p. 229) about Shaumyan’s syntactic transformations. Emotions in argumentative narration 423 © Sara Cigada. Informal Logic, Vol. 39, No. 4 (2019), pp. 401–431 action. The fact that the emotion term tension is not explicitly attributed to the speaker is consistent with the general understate- ment characterizing Martine’s discourse, thus adding to the discur- sive construction of pathos, via understatement as a mitigation strategy and de-focalisation of the speaker.14 Details in description play the role of adding more data to the material starting point, thus strengthening the speaker’s emotional conclusion. Greco states that “the reconstruction of material- contextual premises, especially if they are left implicit and consid- ered as taken for granted, gives a perspective on what is or is not inter-subjectively shared by the interlocutors” (Greco et. al. 2018). In the excerpt, narration plays the specific role of inserting into the discursive context all those elements that, according to the speak- er’s point of view, are not shared by the audience like details about the class. At the time, addressing an audience of colleagues (teachers), Martine does not stress the endoxon (their shared pro- fessional commitment), which remains implicit. The right side of the Y also remains implicit. 6. Discussion The AMT emerges as particularly relevant to the understanding of what has been sketched as “emotional inferencing.” I outline four specific points regarding this in the examples and offer some hypotheses well aware that they require further investigation and discussion. Result 1: Y’s Procedural-inferential component in Figure 1, 2 and 3: It is, unfortunately, possible to imagine a reader not sharing the endoxon in Figure 1, “Terror attacks like Charlie Hebdo are horrific situations” but accepting, for example, another endoxon such as “Terror attacks like Charlie Hebdo are desirable situa- tions.” Nothing in the procedural-inferential structure would change, but the first conclusion would be that “These students have expressed appreciation for a desirable situation.” The differ- ence would depend on the material-contextual starting point and would coincide with the different communal opinion about a 14 I am grateful to an anonymous reviewer for pointing to mitigation and de- focalization. 424 Cigada © Sara Cigada. Informal Logic, Vol. 39, No. 4 (2019), pp. 401–431 specific situation. The procedural-inferential organization of the reasoning does not depend on the content of the endoxon but, rather, on the maxim (consistent with Danesi and Rocci 2009, pp. 203-212 on culture-bound argumentation). Additionally, both endoxa would require further justification as one might wonder on what basis they are shared among a community. In Martine’s discourse, there are no explicit arguments supporting one endoxon or the other. In Figure 1, the Locus of Ontological Implication links the situation and the emotion; the emotion’s extreme (the display of euphoric emotion) is taken as the inference-trigger and the situa- tion’s extreme (the experience of a positive situation) as the in- ferred conclusion, so the inferencing operates from the implied term to the implying one. Locus, in fact, coincides with a habi- tudo: “habitudo highlights the relationship itself, thus involving both relata. The habitudo, by the way, turns out to precisely coin- cide with the locus” (Rigotti and Greco Morasso 2010, p. 494 note 4). As such, the link may be reversed and “read” in the opposite direction. In fact, “each locus can be read in two directions, de- pending on which pole of the habitudo is taken as a starting point and which one is taken as a target point (e.g., cause to effect or effect to cause)” (Rigotti and Greco 2019, pp. xiii and 210). In Figures 2 and 3, in fact, the link between the two extremes of the habitudo goes from the implying term (in 2, experiencing a horrific situation; in 3, facing a probably unmanageable situation) to the implied one (in 2, dismay; in 3, tension). Thus, the implying extreme plays the role of inference-trigger while the emotion is inferred as the ontologically implied extreme. In examples 2 and 3, the logical form of the inference is a modus ponens: “when one experiences a horrific situation, one feels traumatized, and P is experiencing this horrific situation.” In Figure 1, a form of symptomatic reasoning occurs. This maxim “activates the logical form of false modus ponens, which is usual in symptomatic argumentation” (Rigotti and Greco Morasso 2010, p. 495; Rigotti and Greco 2019, pp. 211-212). Result 2. Trying to sketch the locus. Homocentrism (interest) in ontological implication: The link (habitudo) between “one being involved in a specific situation” and “one’s emotional response” Emotions in argumentative narration 425 © Sara Cigada. Informal Logic, Vol. 39, No. 4 (2019), pp. 401–431 has been accounted for as an ontological implication, that is, as an intrinsic locus, that is a locus in which the two extremes of the habitudo belong to the same possible word (Rigotti and Greco 2019, p. 96). Therefore, when describing emotions as one extreme of an ontological implication, I maintain that the situation at hand is affecting someone who is involved in it. This fits well with one of the most relevant features mentioned in Ungerer’s model, that of the “tendency of homocentrism which seems to permeate our thinking and feeling” (Ungerer 1997, p. 313). In fact, emotion as an ontological implication depends on the human perspective on the situation. This relates to a pertinent question concerning the asymmetry of emotion in orientation and in time. Emotions in fact tend to be oriented towards the positive (Rodolphus Agricola 1542, ch. 3,3,1; van der Poel 1997, p. 225) and the future. This needs to be investigated further in relation to the definition of the inference-triggering of an “euphoric” vs “dysphoric” emotion. In fact, the so-called “polarity” of emotions seems to be a fallacious simplification. Human (personal) involvement could appropriately be de- scribed in terms of “interest,” a specific dimension of one’s “rea- sonableness-in-context” (“it is reasonable that x evaluates any situation in which he is involved, from the point of view of any personal gain he has in that situation”). A pragma-dialectical account of reasonableness may provide adequate backing, as suggested by Rigotti (2011, p. 269): Contrasting it [reasonableness] to the apparently close notion of rationality can bring to light important differences, [in fact] “In ordinary language use of the word reasonable is not limited to verbal behavior but covers also non verbal behavior… The scope of reasonableness seems to be wider than that of rationality. One can, for example, speak quite well of reasonable desires, but not so easily of rational desires.” (van Eemeren 2011, p. 29, note 9). The example that is proposed exceeds the domain of behaviors, both verbal and non verbal. Indeed desires belong to a class that embraces emotions (fear, rage…), feelings, interests and other psychological states such as suspicion, doubt, confidence beside many other non-rational facts of human experience. And, if the category of reasonableness can be predicated on such facts, they acquire citizenship in argumentation theory, and strategic maneu- 426 Cigada © Sara Cigada. Informal Logic, Vol. 39, No. 4 (2019), pp. 401–431 vering amounts to taking into account a considerably wider area of human experience. The very etymology of “interest” (Lat. mea inter-est) refers to “one’s being involved” in a situation (Ernout and Meillet 1985, p. 320 and 567; Forcellini 1965, s.v. intersum). Thus, the work that needs to be done in order to understand emotional inferences within the AMT is an outlining of what the relevant semantic components, both of the situations evoked and of the human responses to them, are and under which conditions they may be taken for granted. This should be done via a corpus- based analysis of cases. It is likely that some habitudo will turn out to be highly culture-dependent and very strongly held within the community.15 Result 3. The cognitive dimension of emotional inference: In the examples considered, maxims activating emotional inference take the form of “Under xyz conditions affecting P’s interest, it is reasonable that P feels like E” (or the converse symptomatic: “When P displays E, it is reasonable to suppose that his interest is being affected by xyz conditions”). This reformulation can possi- bly account for the cognitive component of emotions, which has been extensively explored by Plantin and many others. For in- stance, to return to Figure 3, “If P is ‘committed to managing a situation’ [this is Martine’s specific interest towards the situation] and circumstances make her judge [cognitive component] that this will probably be impossible, it is reasonable that P feels tense about the situation.” This obviously reminds us of the third book of Rodolphus Agricola’s De Inventione Dialectica (1542), where Agricola explains that rhetorical discourse does not need to sug- gest or recommend emotions to the audience because emotions are the natural consequence of situations.16 Therefore, being moved is an ontological implication when one comes to know about a given 15 Maybe the so-called “connotational frames” (Jurafsky and Martin 2018, p. 376) could be compared with the AMT to use data, mined and tagged on this basis, for sentiment analysis. 16 According to Agricola, situations are made up of good or bad events (res) regarding guilty or innocent persons (personae) (van der Poel 1997). Emotions in argumentative narration 427 © Sara Cigada. Informal Logic, Vol. 39, No. 4 (2019), pp. 401–431 situation (see also Macagno and Walton 2014, pp. 66-67), accord- ing to one’s opinion about people and events. Result 4. Synchronization effect on the audience: As shown in Figure 2, an emotion (dismay) experienced by the speaker might be the conclusion of an inference triggered by a situation (the attack on Charlie Hebdo). This emotional meaning, though implic- it, is relevant to the meaning as it is understood by a specific audi- ence (conflicting contents, as pointed out in 5.2). This means that the audience is able to understand Martine’s emotion and its con- tribution to the meaning of the whole passage. I borrow from Martina Drescher (2003, pp. 129-130) the notion of emotional synchronization, not only as an empathic reaction, but more wide- ly as any emotional reaction of an audience to the situations and emotions presented in a discourse (see Amossy 2008 about “sym- pathy” and ethos). Different audiences “tested” with our corpus synchronized in different ways, but all of them understood Mar- tine’s emotion as a consistent element of her discourse; a previous- ly mentioned example of this is the reaction of the Lebanese teacher who commented on Martine’s distress “She is French, she is not used to this.” This discursive reaction to Martine’s testimony shows the teacher’s specific “interest” towards her and her situa- tion, understanding Martine’s emotion but not aligning with it. Moreover, according to Agricola’s description of affectus (van der Poel 1997), audience’s emotions consist not only of empathy and/or synchronization, but of emotional evaluation of any situa- tion that is in play in the discourse. Roughly speaking, applying the AMT allows us to give an account of the (subjective!) “emo- tional layer” we perceive in any discourse we are interested in. 7. Conclusion My analysis stems from Plantin’s account for emotion terms and considers a situated argumentation (the day after the 2015 attack on Charlie Hebdo; a middle school in the suburbs of Paris; the ongoing dialogue between Martine and her problematic class; Martine’s testimony addressing Lebanese colleagues in the inter- national project S’éduquer pour éduquer). I analyzed the emotion- al component of this corpus from the argumentative vantage point 428 Cigada © Sara Cigada. Informal Logic, Vol. 39, No. 4 (2019), pp. 401–431 using Pragma-Dialectics and the Argumentum Model of Topics to study the phenomenon of “emotional inferencing” (Ungerer 1997). I considered a narrative corpus both because in narratives it is taken for granted that emotions must be reasonable (in the sense that we expect the text to give reasons supporting emotions) and because narration (including the emotions it contains) may be used as an argument as is the case in my corpus. I analyzed two emotion terms: an indirect (to express apprecia- tion) and a direct one (tension) as well as an implied (but not expressed) emotion. For each case, I used the Locus of Ontologi- cal Implication, considering the habitudo in the two possible direc- tions: from the situation to the emotion or from the emotion to the situation. I noticed that emotional inferencing appears on the right side, that is, in the procedural-inferential component of the Y structure. On the left side, that is, in the material-contextual component, we find the situational information instantiating the locus. Emotion terms may occur on both sides and help to trigger the emotional inferencing. The other linguistic and rhetorical structures represent mitigation strategies that turn out to be consistent with the emo- tional inferencing thus contributing to the sense-effects. References Amossy, Ruth. 2008. Dimension rationnelle et dimension affective de l'ethos. In Emotions et discours. L’usage des passions dans la langue, ed. Michael Rinn, 113-126. Rennes: PUR. Caffi, Claudia. 2007. Mitigation. Amsterdam: Elsevier. Caffi, Claudia, Richard W. Janney. 1994. Toward a pragmatics of emo- tive communication. Journal of Pragmatics 22: 325-373. Cigada, Sara. 2006. Connectif et relation entre locuteurs. In Sýndesmoi. Connettivi nella realtà dei testi, eds. Giovanni Gobber, Maria Cristina Gatti and Sara Cigada, 97-173. Milano: Vita e Pensiero. Cigada, Sara. 2016. Analyzing emotions in French discourse: (Manipula- tive?) shortcuts. In Case studies in discourse analysis, eds. Marcel Danesi and Sara Greco, 390-409. Munich: Lincom Europe. Danesi, Marcel, Andrea Rocci. 2009. Global linguistics: An introduc- tion. Berlin/New York: Mouton de Gruyter. Emotions in argumentative narration 429 © Sara Cigada. Informal Logic, Vol. 39, No. 4 (2019), pp. 401–431 Drescher, Martina. 2003. Sprachliche Affektivität: Darstellung emotionaler Beteiligung am Beispiel von Gesprächen aus dem Französischen. Berlin: de Gruyter. van Eemeren, Frans H., Rob Grootendorst. 2004. A systematic theory of argumentation: The Pragma-Dialectic account. Cambridge: Cam- bridge University Press. van Eemeren, Frans H. 2018. Argumentation theory: A Pragma- Dialectical perspective. Cham: Springer. Ernout, Alfred and Antoine Meillet. 1985. Dictionnaire étymologique de la langue latine (4 e éd.). Paris: Klincksieck. Forcellini, Egidio. 1965 (1671). Lexicon Totius Latinitatis, eds. F. Corradini and J. Perin. Padova : Gregoriana. Gardes Tamine, Joëlle. 2015. Le style, entre langue et rhétorique. Nouvelle Fribourg 1. URL accessed March 2019 . Gatti, Maria Cristina. 1992. Dalla semantica alla lessicologia. Introduzione al modello senso-testo di Igor Mel’chuk. Brescia: La Scuola. Gobber, Giovanni. 2019. The scarlet letter of “post-truth”: the sunset boulevard of communication. Church, communication and culture 4(3): 287-304. Greco, Sara. 2017. La dimensione inferenziale nell’argomentazione. In Argomentare per un rapporto ragionevole con la realtà, eds. Paolo Nanni, Eddo Rigotti and Carlo Wolfsgruber, 25-53. Milano: Fondazione per la Sussidiarietà. Greco, Sara et al. 2018. The analysis of implicit premises within chil- dren’s argumentative inferences. Informal Logic 38(1): 438-470. ICOR, Clapi website. URL accessed March 2019: . Jacobs, Scott. 2006. Nonfallacious rhetorical strategies: Lyndon John- son’s Daisy ad. Argumentation 20 (4): 421-442. Jurafsky, Dan, James H. Martin. 2018. Speech and language processing (3 rd edition online draft). URL accessed March 2019: . Macagno, Fabrizio and Douglas Walton. 2014. Emotive language in argumentation. Cambridge: Cambridge UP. Marty, Anton. 1908. Untersuchungen zur Grundlegung der allgemeinen Grammatik und Sprachphilosophie. Vol. 1. Halle a. S.: Niemeyer. http://www.nouvellefribourg.com/universite/le-style-entre-langue-et-rhetorique/ http://www.nouvellefribourg.com/universite/le-style-entre-langue-et-rhetorique/ http://clapi.ish-lyon.cnrs.fr/ http://clapi.ish-lyon.cnrs.fr/ https://link.springer.com/journal/10503/20/4/page/1 https://web.stanford.edu/~jurafsky/slp3/ed3book.pdf 430 Cigada © Sara Cigada. Informal Logic, Vol. 39, No. 4 (2019), pp. 401–431 Mel’chuk, Igor A. 1988. Paraphrase et lexique dans la théorie linguis- tique Sens-Texte. In Lexique et paraphrase, eds. Gabriel G. Bès, Ca- therine Fuchs, 13-54. Lille: Presses Universitaires de Lille. Micheli, Raphaël. 2008. Emotions as objects of argumentative construc- tions. Argumentation 24: 1-17. Micheli, Raphaël. 2010. L'émotion argumentée. L'abolition de la peine de mort dans le débat parlementaire français. Paris: Editions du Cerf. Micheli, Raphaël. 2014. L'émotion dans le discours, Louvain-la-Neuve: deBoeck. Moeller, Esther. 2010. Une union éternelle ? Les catholiques libanais et les écoles françaises au Liban 1900-1950. Documents 45: 43-68. Olmos, Paula (ed.). 2017. Narration as argument, Cham: Springer. Palmieri, Rudi. 2014. Corporate argumentation in takeover bids. Am- sterdam: Benjamins. Palmieri, Rudi and Sabrina Mazzali-Lurati. 2016. Multiple audiences as text stakeholders: A conceptual framework for analyzing complex rhetorical situations. Argumentation 30: 467-499. Perret-Clermont, Anne-Nelly. 2005. Lo spazio per il pensare. In Discorso: Una proposta per l’autoformazione e apprendimento degli insegnanti, ed. Carlo Pontecorvo, 21-34. Roma: Carocci. Piccinelli, Marta. 2019. Il francese del Libano: per una verifica sociolinguistica dell’uso in contesto scolastico. Master thesis at the Università Cattolica del Sacro Cuore, Brescia. Plantin, Christian. 1999. La construction rhétorique des émotions. In Rhetoric and argumentation, ed. Eddo Rigotti, 203-219. Tübingen: Niemeyer. Plantin, Christian. 2004. On the inseparability of emotion and reason in argumentation. In Emotion in dialogic interaction, ed. Edda Weigand, 265-276. Amsterdam: Benjamins. Plantin, Christian. 2011. Les bonnes raisons des emotions. Bern: Peter Lang. Plantin, Christian. 2018. Dictionary of argumentation, College Publications. van der Poel, Marc. 1997. Rodolphe Agricola. Ecrits sur la dialectique et l’humanisme. Paris: Champion. Pollaroli, Chiara and Andrea Rocci. 2015. The argumentative relevance of pictorial and multimodal metaphor in advertising. Journal of Ar- gumentation in Context 4/2: 158-200. Rigotti, Eddo. 1979. Principi di teoria linguistica. Brescia: La Scuola. http://www.icar.cnrs.fr/pageperso/cplantin/documents/1999b.doc Emotions in argumentative narration 431 © Sara Cigada. Informal Logic, Vol. 39, No. 4 (2019), pp. 401–431 Rigotti, Eddo. 2011. Frans H. van Eemeren: Strategic maneuvering in argumentative discourse. Extending the Pragma-Dialectical Theory of Argumentation (Review). Argumentation 25: 261-270. Rigotti, Eddo, Sara Cigada. 2013. La comunicazione verbale. Santarcangelo di Romagna: Maggioli. Rigotti, Eddo, Sara Greco Morasso. 2010. Comparing the argument model of topics to other contemporary approaches to argument schemes: The procedural and material components. Argumentation 24: 489-512. Rigotti, Eddo and Sara Greco. 2019. Inference in argumentation. Cham: Springer. Rigotti, Eddo and Rudi Palmieri. 2016. Solomon's wise judgment: A case study of argumentation in context. In Case studies in discourse analysis, eds. Marcel Danesi and Sara Greco, 37-61. Munich: Lincom Europe. Rigotti, Eddo and Andrea Rocci. 2006. Le signe linguistique comme structure intermédiaire. In Nouveaux Regards sur Saussure. Mé- langes offerts à René Amacker, ed. Louis de Saussure, 219-247. Genève: Droz. Rocci, Andrea. 2017. Ragionevolezza dell’impegno persuasivo. In Argomentare per un rapporto ragionevole con la realtà, eds. Paolo Nanni, Eddo Rigotti and Carlo Wolfsgruber, 88-120. Milano: Fondazione per la Sussidiarietà. Rodolphus Agricola. 1542. De Inventione Dialectica libri III. Coloniae: Iohannis Kempensis. Thierry, Herman. 2008. Narratio et argumentation. In Argumentation et Narration, eds. Emmanuelle Danblon, Emmanuel de Jonge, Ekateri- na Kissina and Loïc Nicolas, 29-41. Bruxelles: Editions de l’Université de Bruxelles. Ungerer, Friedrich. 1997. Emotions and emotional language in English and German news stories. In The language of emotions, eds. Susanne Niemeier and René Dirven, 307-328. Amsterdam/Philadelphia: Ben- jamins. https://search.usi.ch/en/publications/16009/solomons-wise-judgment-a-case-study-of-argumentation-in-context https://search.usi.ch/en/publications/16009/solomons-wise-judgment-a-case-study-of-argumentation-in-context