THE MULTIDIMENSIONAL MODEL OF ARGUMENTATIVE ANALYSIS: AN INTRODUCTION

ABSTRACT

In this paper, we advocate for a multidimensional approach to the analysis of argumentation by considering its cognitive, discursive and multimodal grounding. First, we briefly discuss the main theoretical premises of such an approach, assuming a multidisciplinary perspective. Second, by drawing on different traditions of argumentation studies, we introduce the five dimensions we consider relevant for a holistic analysis of argumentative practices – functional configuration, macrostructure, schematization, socio-affective grounding and argumentative orientation. Finally, we illustrate the functioning of the model through a multidimensional analysis of an argumentative move extracted from a Brazilian television political interview.

Argumentation; Discourse; Cognition; Language; Multidimensionality; Multidisciplinarity

RESUMO

Nosso objetivo, neste artigo, é discutir um modelo multidimensional de análise que considere a ancoragem discursiva, cognitiva e multimodal da atividade argumentativa. Em primeiro lugar, apresentamos sucintamente as principais premissas teóricas de tal abordagem, partindo de uma perspectiva multidisciplinar. Em segundo lugar, por meio de um diálogo com diferentes tradições dos estudos argumentativos, introduzimos as cinco dimensões que consideramos relevantes para uma análise holística das práticas argumentativas – a configuração funcional, a macroestrutura, a esquematização, a ancoragem socioafetiva e a orientação argumentativa. Por fim, ilustramos o funcionamento do modelo por meio de uma análise multidimensional de um movimento argumentativo extraído de uma entrevista televisiva com um político brasileiro em contexto de campanha eleitoral.

Argumentação; Discurso; Cognição; Linguagem; Multidimensionalidade; Multidisciplinaridade

Introduction

The study of argumentative reasoning and practices constitutes a heterogeneous field. This is true not only in terms of the different theoretical assumptions and methodologies that characterize the field, but also in relation to the dialogues that argumentation theories establish with different disciplines, such as philosophy, mathematics, computer sciences, law, psychology, cognitive sciences, linguistics and discourse analysis. Thus, it is not hard to infer that the many approaches will pursue different objectives: the evaluation of the validity or the soundness of arguments; the description of their functioning and persuasive potential; the understanding and organization of the reasoning processes embedded in the practices; the development of students’ skills in terms of discussion and debate performance; and even the engineering of software aimed at interpreting, producing and assessing arguments.

An issue that usually emerges in such heterogeneous fields is the relative independence and the rare convergence between the developments achieved in each of the approaches. It is not different in the argumentation field, even though there are some efforts – with a high degree of success – in pursuing this dialogue, such as Informal Logic ( JOHNSON; BLAIR, 2017 ; WALTON, 2013 ) and Pragma-dialectics ( EEMEREN; GROOTENDORST, 2004 ; EEMEREN; HOUTLOSSER; SNOECK HENKEMANS, 2007). Both these approaches, however, tend not to dive deep into the linguistic, the discursive and the cognitive aspects of argumentation.

Thus, our aim, in this paper, is to discuss the functioning of the analytical model we have been developing to describe, evaluate and explain argumentative texts. It is inspired by several sources from four fields of study already mentioned en passant: Argumentation Theory, Cognitive Sciences (especially, the embodied social paradigm in its weak/simple version), Linguistics (mainly, Cognitive Linguistics and Systemic-Functional Linguistics) and Critical Discourse Studies. Hence, it takes into consideration the cognitive processes underlying argumentation, i.e., the reasoning processes employed in grounded practices of argumentation; its multimodal realization in terms of language and image; and the discursive coercions that structure argumentation as a socio-semiotic practice.

Initially, we discuss our view of argumentation and the reasons why we claim that a multidisciplinary approach to argumentation that considers its intrinsic relation with cognition, language and discourse is relevant and fruitful for the better understanding of its nature and functioning; then, we discuss the five dimensions we have devised so far for consistently analyzing argumentative practices,² considering the multidisciplinarity involved; afterwards, we will partially apply the model under discussion in the analysis of a brief topic developed in a Brazilian television political interview whose guest was running as candidate in São Paulo mayoral election in 2012; finally, we will produce some remarks, discussing the road traversed so far and the possible paths that lie ahead.

For a multidisciplinary approach to argumentative practices

We assume, along the lines of Leitão (2012 , p. 26, author’s italics, our translation),³ that argumentation is characterized as a

[…] discursive (essentially verbal), social (of cultural nature, contextually dependent), cognitive (implying reasoning processes needed to support and critically evaluate claims), dialogic (simultaneously responding and anticipating responses), dialectical (characterized as a critical exam of divergent arguments) and epistemic (since it allows the construction of knowledge) activity.

Besides that, we draw on Niño and Marrero (2015)⁴ in considering that there are many functions to argumentation and that they are closely related to processes of belief construction. Thus, from our perspective, arguing may respectively aim at:

• i.convincing (belief formation or revision), understood as the perlocutionary effect of adhering, locally or globally, totally or in varying degrees, to certain conceptions of reality as a consequence of assuming them as reasonable or sound, based on the varying strength of the link from Data to Claim, through the Warrant, considering possible Rebuttals and the reliability of the Backings ( TOULMIN, 2006 [1958]; TOULMIN; RIEKE; JANIK, 1984 [1978]);

• ii.persuading (belief defensibility and decision making), understood as the perlocutionary effect linked to practical action and thus towards making decisions that result in a change on the course of reality – and not on the conception of reality, as in the first case. Thus, the soundness is tied to the strength of the link between Values, Goals, Consequences and Circumstances surrounding the proposed Action ( FAIRCLOUGH; FAIRCLOUGH, 2012 ) and the decision-making process associated with it;

• iii.preserving ideology/reinforcing discourses (belief maintenance),⁵ understood as a perlocutionary effect of confirming a certain conception of reality, e.g., a given moral stance against abortion, by presenting Claims and Data which are already shared and agreed upon by the members of the endo-group ( us ), as well counter-rebutting arguments which are already conceived as invalid and unsound by the same group. This process is, thus, linked to the reinforcement of the belief systems of the arguers, with positive effects in terms of construing internal identification and building knowledge repertoire, but also negative effects in terms of bias against the perspectives of others ( them ).

Some consequences stem from assuming this functional complexity. First, by recognizing three different functions, we understand that there are distinct kinds of argumentative practices and, hence, at least three functional configurations of argumentative types. These argumentative types are instantiated in texts through the coercions of both genres (such as interviews, debates, editorials) and interactive situations (the Ground composed by the discursive event, which encompasses participants and their shared knowledge, time, place and setting). This complex interaction will lead to distinct patterns of reasoning and multimodal (especially, through verbal and visual languages) realization.

The second one can be predicted from what has been just argued: these different types of reasoning, and even different types of biases and fallacies, which emerge from the above coercions, are both the result and the source of the employment of distinct combinations of cognitive operations, which, in turn, manifest themselves through a complex number of construal⁶ operations in language, such as schematization, metaphorization, categorization, focalization, granularization, among others ( HART, 2014 ; GONÇALVES-SEGUNDO, 2017a, 2017b). As a result, different construal patterns can emerge and be prototypically associated with the functions of convincing, persuading and preserving ideology/reinforcing discourses as well as with the different genres and situation types in which these processes are embedded.

Third, by recognizing belief maintenance as a function of argumentation, we assume that it is inevitable to consider discourse – and thus ideology – as relevant factors in the structuring of argument. However, we do not mean that only in terms of the third function: both belief formation/revision and belief defensibility/decision-making are processes that are inscribed in social practices, which, in turn, are constituted by orders of discourse ( FAIRCLOUGH, 2003 ). These orders of discourse are formed by discourses (socio-semiotic patterns of representation), genres (socio-semiotic patterns of action) and styles (socio-semiotic patterns of identification), which can function in an organized and correlated way, cementing path for hegemony. Understanding how hegemony is discursively construed in argumentation is a central aspect in terms of a critical discursive stance, but also of an argumentative and a rhetorical critique. Besides that, as Fairclough and Fairclough (2012) argue, discourses are often constitutive parts of premises in argumentation.

It is essentially because of these three consequences that a multidisciplinary approach is needed and, along with it, a multidimensional model which can tackle the complexity sketched above. Let us, then, systematize how multimodality, discourse and cognition may be drawn upon in order to provide more nuanced descriptive and explanatory tools for argumentative analysis:

• i.The consideration of cognition is relevant, since our argumentative competence is sustained by our cognitive systems, through several operations, among which we can include:

a.our main forms of reasoning – causal, analogical or symptomatic (EEMEREN; HOUTLOSSER; SNOECK HENKEMANS, 2007). They involve – among other processes – Force Dynamics, projections between knowledge domains, type-token relations, phenomena largely studied by Cognitive Sciences (Psychology and Linguistics, mainly) ( TALMY, 2000 ; WOLFF; BARBEY, 2015 ; VEREZA, 2010 , 2013 ; GONÇALVES-SEGUNDO, 2014, 2015, 2017a, 2017b, 2018a; ITKONEN, 2005 ; BARSALOU, 1999 ; LANGACKER, 2008 , among others);

b.our perspectivization ability, which encompasses the establishment of an epistemic stance regarding the representations and evaluations construed discursively. This can be done not only through the flexibilization of the reality status of a proposition and of our commitment regarding the utterances we produce, but also through the incorporation of different voices to (de)authorize what we (or the others) propose. This implies, in a certain way, the recognition of a Theory of Mind ( SPERB; JOU, 1999 ). In order to account for all this complexity, we assume a simple/weak version ( CLARK, 1996 ) of a embodied social model of cognition,⁷ which accounts both for modal and amodal structures ( BARSALOU, 1999 ), such as image schemas and frames.

• ii.A detailed account of the linguistic and visual strategies instantiated in texts allows us to understand the (micro)construction of argumentation, considering not only the elaboration of standpoints and claims for action, but also the articulation of the argumentation schemes which link Data to Claims, through Warrants, or Objectives, Values and Circumstances to Actions. Hence, it is relevant to consider this aspect, since the construal of actual utterances are, on the one hand, the point of tension between hegemony and alternativity, and, on the other hand, the window to the cognitive processes employed and to the structuration types (situation types, discourses, genres and styles) that organize argumentative practices. Several linguistic phenomena ought to be considered for such an approach, among which we may highlight inter-sentential relations (MOURA NEVES, 2007 ; HALLIDAY, 2004 ; LANGACKER, 2008 ), referentiality ( KOCH, 2014 ), evidentiality and modality ( MARÍN-ARRESE, 2011 , 2013 ; BEDNAREK, 2006 ; CARIOCA, 2011 ; MIRANDA, 2005 ; GONÇALVES-SEGUNDO, 2015 ), quantification and intensification ( MARTIN; WHITE, 2005 ; GONÇALVES-SEGUNDO, 2011), attitude and engagement ( MARTIN; WHITE, 2005 ), among others. To account for this aspect of analysis, we draw upon a dialogue between Cognitive Linguistics, given its orientation towards interpretation and discursivity, and Systemic Functional Linguistics, given its emphasis on production and textuality.

• iii.Finally, we should account for the discursive coercions, grounded on the shared conceptual models of distinct epistemic communities, concerning the valid and the legitimate argumentative processes ratified in different social and historical conditions. These coercions encompass the configuration of genres in distinct institutional and daily life fields, as well as identificational and representational pressures in the textual and discursive construal. To account for this aspect, we draw mainly – but not only – on Critical Discourse Studies bibliography ( FAIRCLOUGH, 2003 , 2010 ; VAN DIJK, 2003 ; HART, 2014 ; RESENDE; RAMALHO, 2006 ; MELO, 2012 ; GONÇALVES-SEGUNDO; ZELIC, 2016 ; CALDAS-COULTHARD; IEDEMA, 2008 , among others).

That said, as we consider having briefly, but consistently, sketched the reasons behind a multidisciplinary approach to the analysis of argumentative practices, we will discuss the five dimensions proposed so far for a holistic analytical model on argumentation.

Dimensions of argumentation

As Gracio (2010) , Plantin (2008) and others point out, contemporary studies on Argumentation tend to follow the steps of two major works, both published in 1958: Perelman and Olbrechts-Tyteca’s The New Rhetoric: A Treatise on Argumentation and Toulmin’s The Uses of Argument. From the first one emerged a whole set of theories aiming at describing argumentative processes focusing on the strategic role of argumentative schemas (an aspect of logos ) and the importance of the socio-affective grounding, encompassing both ethos and pathos, to convince/persuade, but leaving aside the structuring and the functional configuration of arguments (another aspect of logos ). These last aspects – especially the latter – were the cornerstone of the Toulmin’s model, which also highlighted the relevance of socio-discursive factors (through the polemic and not so clear notion of field )⁸ to the argumentative process. His proposal was – and is, by some authors – regarded as normative, insofar as it proposes an ideal diagram of argumentative configuration, in which utterances – and rules of inference – fill roles for supporting, dialectically and dialogically,⁹ a Claim.

Pragma-dialectics and Informal Logic – the most comprehensive approaches on argumentation that we can pinpoint – managed to tackle, although in different degrees, both traditions and to provide coherent tools to describe and evaluate arguments. In a certain way, our model aims at doing the same and, as such, we draw a lot on both approaches, although we do not intend, by any means, to provide a theory of argumentation, as these perspectives surely aim at. It is, nevertheless, the attempt of correlating discursive, linguistic and cognitive aspects that, somehow, distinguishes our proposal from theirs and may, perhaps, complement some of their theorization, although it may lose some of their consistency, for we lack their philosophical underpinning.

That said, we would argue that a holistic argumentative model should consider, at least, the five dimensions below, analyzed in an integrated fashion, aiming at possible correlations:

• i.Functional Configuration.

• ii.Macrostructure.

• iii.Schematization.

• iv.Socio-affective Grounding.

• v.Argumentative Orientation.

In the next sub-sections, we will briefly discuss each of these dimensions, defining them and commenting on some relevant categories, before the analytical application.

Functional Configuration

The analysis of the functional configuration of argumentation presupposes understanding how a claim (or claims) is (are) supported in different fields of social activity, in distinct genres, considering the functions of convincement, persuasion or preservation of ideology/ratification of discourse. It aims at arriving at an abstract pattern of argumentation which shows the essential and the facultative functional components that constitute an instance of a potentially sound argumentative move, considering the aforementioned discursive factors.

Toulmin’s (2006 [1958]) and Toulmin, Rieke and Janik’s (1984 [1978]) layout of arguments provide good starting points for this dimension. Enhancements proposed to the model, discussed by several theoreticians, are also welcome, since they provide better tools to the analysis of effectively produced argumentative texts – a serious concern in the development of our model. Although we do not have space to provide a critical assessment of all those collaborations, it is relevant to discuss some of the developments we consider useful – or necessary – to enhance this approach. Also, we would like to pinpoint one caveat: we see Toulmin’s model and its development to be better suited to analyze the belief formation/revision (convincement) and, perhaps, the belief maintenance (ideology preservation) functions of argumentation. We would need an alternative practical reasoning configuration to tackle issues of belief defensibility/decision-making (persuasion). Fairclough and Fairclough (2012) proposal seems to be rather comprehensive and useful for that objective. We will not discuss it further here, though. That said, we should, at least, sketch our view on Toulmin’s proposal.¹⁰

First, we assume that Toulmin’s model (2006 [1958]) can be replicated both vertically and horizontally. That means that we do not assume that it can only deal with simplex argumentation. It means that Data (or Grounds), “ specific facts relied on to support a given claim”¹¹ (TOULMIN; RIEKE; JANIK, 1984 [1978], p. 37, our translation), can also be contested and, thus, be provisionally treated as Claims that need to be supported. This is what is called horizontal expansion. Thus, the factual status of a Datum is actually subject to linguistic construal and, as a consequence, to conceptualization.

The Warrant that bridges the Data to the qualified (or modalized) Claim is interpreted as an inferential move.¹² Thus, it is often implicit. We also assume, with Langsdorf (2011) , that Backings should not be restricted to Warrants. They can also act as support to Data and, in our view, even to Rebuttals. Briefly, we take Backing as being the evidential grounding of a proposition, be it a Datum, a Warrant (when challenged or explicit) or a Rebuttal, responsible for the vertical expansion of the model.

Finally, we understand Rebuttals dialogically, along with Slob (2006) . When anticipated by the arguer, they can be weakened through concessive patterns, strengthening the link between Data and Claim. Their anticipation can also be considered relevant in terms of the selection of the modal qualifier applied to the Claim. The modal (or Qualifier) works as a device that builds a commitment scale on the reality status of the Claim based both on the Data presented and the possible Rebuttals, considering the strength of the Warrant (and its possible Backing). When Rebuttals are anticipated, the modal force tends to weaken from certainty to probability or possibility in texts oriented to belief formation, and from necessity/obligation/interdiction to expectancy or permission in texts oriented to practical reasoning. Rebuttals can also work as undermining propositions, either by creating conditions for the soundness of the Data-Claim link or by defeating the proposed bridge. They can do so by aiming at the Datum itself, the Warrant, the application of the Warrant towards the case at issue, the Backing, or the Claim itself.

Macrostructure

The macrostructural analysis involves understanding how the different component utterances fit together in the support of a standpoint. We view structuration through dialectical/dialogical lenses, proposing that the different combination patterns of Data, Rebuttals and Claims reveal the way the Protagonist¹³ is responding to real or projected arguments and counterarguments from the Antagonist.

Freeman’s (2011) work concerning the diagramming of macrostructural patterns is well-known and certainly more complex and detailed than the model we present here. His model stems from different sources, from Logic to the Toulmin’s model, and his discussion on the nature of convergent, multiple, divergent and serial structuring is certainly influential to our view. Our main source, however, lies in Pragma-dialectics (EEMEREN; HOUTLOSSER; SNOECK-HENKEMANS, 2007), since the main types proposed in their account – although not exhaustive, but reasonably comprehensive – are already tuned to a dialectical/dialogical perspective.

Our intent is to build a convergence between the diagramming tradition – as we can see in Freeman (2011) and even in Walton and Macagno (2016) – and the dialectical/dialogical account provided by Pragma-dialectics, using graph structures to account for argumentation as product and process both in written and oral texts.

In a brief attempt to show the convergence, we will define and illustrate the diagram referring to each of the four main patterns of macrostructuration recognized in Pragma-dialectics (EEMEREN; HOUTLOSSER; SNOECK-HENKEMANS, 2007):

• i.Subordinative argumentation occurs when the Protagonist anticipates or responds to a (concrete) criticism on their argument, based on the judgement that the Datum (2)¹⁴ they selected to support the Claim (3) is unacceptable (or cannot be taken as assumed, but at issue). They must, then, turn the Datum into a Claim (2), presenting further Data (1) to support it, in order to win over the Antagonist (possible) resistance.

Figure 1
– Subordinative Argumentation

• ii.Multiple Argumentation occurs when the Protagonist recognizes the validity of the Antagonist criticism (3) over the Datum (1) or the Warrant that supports the Claim (2),but believes in the soundness of the Claim. They search, then, for new Data (4) to support it. In implicit discussions, the Protagonist may anticipate the weakness of some of their arguments, developing a network of argumentative chains, with distinct degrees of focus, to persuade/convince the other(s).¹⁵

Figure 2
– Multiple Argumentation

• iii.Cumulative Coordinative Argumentation occurs when the Protagonist supplements their argumentation with further Data, insofar as they anticipate (or are criticized by the Antagonist) that a single reason is insufficient to ground the proposed Claim.

Figure 3
– Cumulative Coordinative Argumentation

• iv.Complementary Coordinative Argumentation happens when the Protagonist produces a Counter-Rebuttal, anticipating or responding to the Antagonist counterargument, procedure that has positive effects towards the original convincement/persuasion process.¹⁶

Figure 4
– Complementary Coordinative Argumentation

Schematization

Schematization is perhaps one of the most researched topics on the field. Perelman and Olbrechts-Tyteca’s Treatise on Argumentation (2002 [1958]) discloses a large inventory of schemas, each with several subcategories – those based on associations (quasi-logical arguments, arguments that appeal to reality and arguments that establish reality) and on dissociations.¹⁷ Informal Logic also has its version (best summarized and discussed in WALTON; REED; MACAGNO, 2008) as does Pragma-dialectics (EEMEREN; HOUTLOSSER; SNOECK-HENKEMANS, 2007), perspective that organizes schemas in three major categories: symptomatic, causal and analogical reasoning.

Macagno (2015) has discussed a means-end categorization of schemas, which seems to be the best fit for our approach, as it aims – in a rather open-ended way – to understand and organize schemas, considering a “strict interaction between the pragmatic and the reasoning dimension of discourse” ( MACAGNO, 2015 , p. 199). In this proposal, the author already establishes a distinction between argumentation oriented towards the desirability of course of action (practical argument) and argumentation oriented towards the acceptability of a judgement (epistemic argument). Both these types of argumentation may have claims supported internally (e.g., with Data concerning the action proposed itself and its positive consequences towards the aimed goals, or with Data showing the soundness of a certain view of reality, in analogy with another situation) or externally (e.g., appealing to the authority of the source of the action proposed or to the credibility of the source of the viewpoint defended). Further work is, of course, needed to explore the viability of the categorization and to expand its reach and usefulness – and even to test its core assumptions – but it seems to be a very important step in the same direction we imagine: one that can tackle argumentation in real discursive events considering its distinct functions.

Moreover, we would say that the study of argumentation schemas can be enriched by considering cognitive and cognitive-linguistic factors – the cognitive systems and operations that allow reasoning to occur and the construal operations instantiated in texts to build this reasoning. We could cite the role of Force Dynamics to understand causal reasoning ( TALMY, 2000 ; WOLFF; BARBEY, 2015 ; GONÇALVES-SEGUNDO, 2015 ), the role of cross-mapping between knowledge domains to comprehend analogical reasoning and its intersections with metaphor ( VEREZA, 2010 , 2013 ; ITKONEN, 2005 ) and the role of type-token relations in frame structure to understand symptomatic reasoning ( ZIEM, 2014 ; LANGACKER, 2008 ; BARSALOU, 1999 ).¹⁸

Socio-Affective Grounding

The study of the socio-affective grounding of argumentation concerns the role of the arguer’s construction of authority, credibility and appeal as well as the role of the values, beliefs and emotions of the audience in the processes of convincement, persuasion and ideological preservation. In other words, it concerns matters of ethos and pathos . We draw mainly on Amossy (2005 , 2018), Maingueneau (2005) and Meyer (2007) works to discuss this dimension.

According to Maingueneau (2005) , we should consider a pre-discursive ethos , involving our knowledge about the arguer and the stereotypes about his social identity, and a discursive ethos , construed during the discursive activity itself. The construction of the discursive ethos may, then, ratify or rectify the previous conceptions (of the audience) about the arguer, process that contribute to the maintenance or to the deterioration of his or her credibility.

The discursive ethos concerns not only direct textual references to the arguer – examples of this kind will be shown in the analysis below – but also the possible inferences about the arguer’s character as a result of his or her semiotic behavior and of his or her bodily dispositions. This inferential process is tied to socio-historical values and has, as its source, the way the arguer speaks, dresses, gesticulates in comparison to what is expected in a given practice.

In terms of pathos , we assume, with Meyer (2007) , that it encompasses the audience’s questions about the issue in discussion, the emotions that they experience in face of the questions and the possible answers and, finally, the values that justify their answers to these questions – in our terms, it involves the argumentative usage, in the processes of convincing, persuading and preserving ideology, of beliefs, emotions, values, goals and circumstances that surround and characterize the (often heterogeneous) audience. Thus, it is from pathos that the arguer can extract hierarchies of what is preferred and what is acceptable and relevant to their audience.

In the multidisciplinary approach we envisage, the study of ethos and pathos should be complemented by psychological and cognitive studies concerning epistemic vigilance (SPERBER et al ., 2010) and emotions (LEWIS; HAVILAND-JONES; BARRETT, 2008), in order to better handle the basis for the construal of appeals to authority, testimony, experience and emotion in daily and institutional argumentation.

Argumentative Orientation

The study of argumentative orientation is the one to which the Cognitive and the Functional paradigms in Linguistics can mostly contribute. This dimension, which is usually defocused in the main argumentative approaches, deal with the role of conceptual and lexical units as well as grammatical schemas in the construal of Data, Claims, Rebuttals and Backings, in epistemic argumentation, as well as Claims to Actions, Circumstances, Objectives and Values, in practical argumentation. Thus, it concerns the way language prompts conceptualizations and how these conceptualizations affect reasoning, inferencing, judgement on the soundness of arguments and even favor bias. Several phenomena may be studied here in terms of their association with argumentation and discourse: inter-sentential relations, referentiality, evidentiality, polarity, modality, quantification, intensification, appraisal, speech acts, among others.

The application of the model: argumentation in a Brazilian television political interview

We will illustrate the potential of the model discussed here through a brief analysis of an excerpt from a Brazilian television political interview whose guest was running for mayor in São Paulo city. The fifteen minutes interview was held during the news show SPTV, produced and transmitted by Globo Network of Communications. The interviewer is the journalist César Tralli, and the guest is the former mayor (candidate in 2012) Fernando Haddad. The excerpt we consider for the analysis is shown below both in Portuguese and in English (my translation):

Original Version (Portuguese) Translated Version (English) César Tralli – O senhor foi ministro da Educação durante seis anos e meio e por três anos foi duramente criticado nas falhas do Enem. Foram fraudes e erros que acabaram prejudicando a vida de milhões de estudantes. Isso não compromete a sua imagem de administrador? Fernando Haddad – Olha Tralli eu sou o ministro da Educação que mais expandiu a educação superior no país, com o ProUni, com a expansão das federais, com o novo Fies. Sou o ministro da Educação que mais expandiu a educação profissional no país. Eu sozinho construí 224 escolas técnicas, que é mais que a soma de todos os meus antecessores. Eu melhorei a qualidade do ensino fundamental no país depois de uma queda drástica nos anos 90. O Brasil hoje figura nos relatórios internacionais como caso de sucesso porque saiu da inércia. Não pelo patamar que atingiu mas porque está no rumo certo. Então, tanto a Unesco, a Onu quando a UCDE, que são os países ricos, reconhecem o esforço que o Brasil fez. Agora, se houve um crime contra o Enem, e foi um crime, não foi uma fraude. Um criminoso foi identificado, julgado e punido com cinco anos de cadeia. Eu gostaria que a oposição, ao invés de me criticar, se solidarizasse comigo. Porque houve um crime, e o culpado foi identificado e punido com cinco anos de cadeia. Imagina na cratera do Metrô, se fosse identificado um sabotador, nós iríamos nos solidarizar com o José Serra 19 , que era o governador à época. Mas não, o que aconteceu lá foi um erro, foi um homicídio culposo. Não foi o caso do que aconteceu no Enem, uma pessoa de fora da administração pública e dentro de uma gráfica que é a mais moderna do país cometeu um crime, foi identificado e punido. 20 César Tralli – You were Education Minister for six years and a half and for three years you were heavily criticized for the failures in ENEM (National High School Exam). There were frauds and mistakes that ended up harming the lives of millions of students. Doesn’t that jeopardize your reputation as an administrator? Fernando Hadda d – You see, Tralli, I’m the Education Minister who most expanded higher education in the country, with ProUni (University for All Program), with the expansion of federal institutions, with the new FIES (Student Loan Fund). I’m the Education Minister who most expanded professional education in the country. I, alone, built 224 technical/vocational schools, which is more than the sum of all my predecessors. I improved the Elementary School quality after a dramatic fall in the nineties. Brazil figures nowadays in the international reports as a case of success for we left inertia. Not because we achieved an ideal level, but because we are in the right path. Thus, both UNESCO and the OECD recognized the effort Brazil made. What happened, though, against ENEM was a crime, not a fraud. A criminal was identified, judged, and punished with five years in jail. I’d like that the opposition, instead of criticizing me, supported me, because there was a crime, and the guilty party was identified and punished with five years in jail. Imagine if a saboteur were to be identified in the episode of the subway crater... We would support José Serra who was the governor at the time. But no... in this case, there was a mistake, a culpable homicide. That was not the case with ENEM. A person who was not part of the public administration and who acted as an insider in the printing office, which is the most modern in the country, committed a crime, was identified, and punished. 1920

Original Version (Portuguese)

Translated Version (English)

César Tralli – O senhor foi ministro da Educação durante seis anos e meio e por três anos foi duramente criticado nas falhas do Enem. Foram fraudes e erros que acabaram prejudicando a vida de milhões de estudantes. Isso não compromete a sua imagem de administrador? Fernando Haddad – Olha Tralli eu sou o ministro da Educação que mais expandiu a educação superior no país, com o ProUni, com a expansão das federais, com o novo Fies. Sou o ministro da Educação que mais expandiu a educação profissional no país. Eu sozinho construí 224 escolas técnicas, que é mais que a soma de todos os meus antecessores. Eu melhorei a qualidade do ensino fundamental no país depois de uma queda drástica nos anos 90. O Brasil hoje figura nos relatórios internacionais como caso de sucesso porque saiu da inércia. Não pelo patamar que atingiu mas porque está no rumo certo. Então, tanto a Unesco, a Onu quando a UCDE, que são os países ricos, reconhecem o esforço que o Brasil fez. Agora, se houve um crime contra o Enem, e foi um crime, não foi uma fraude. Um criminoso foi identificado, julgado e punido com cinco anos de cadeia. Eu gostaria que a oposição, ao invés de me criticar, se solidarizasse comigo. Porque houve um crime, e o culpado foi identificado e punido com cinco anos de cadeia. Imagina na cratera do Metrô, se fosse identificado um sabotador, nós iríamos nos solidarizar com o José Serra 19 , que era o governador à época. Mas não, o que aconteceu lá foi um erro, foi um homicídio culposo. Não foi o caso do que aconteceu no Enem, uma pessoa de fora da administração pública e dentro de uma gráfica que é a mais moderna do país cometeu um crime, foi identificado e punido. 20

César Tralli – You were Education Minister for six years and a half and for three years you were heavily criticized for the failures in ENEM (National High School Exam). There were frauds and mistakes that ended up harming the lives of millions of students. Doesn’t that jeopardize your reputation as an administrator? Fernando Hadda d – You see, Tralli, I’m the Education Minister who most expanded higher education in the country, with ProUni (University for All Program), with the expansion of federal institutions, with the new FIES (Student Loan Fund). I’m the Education Minister who most expanded professional education in the country. I, alone, built 224 technical/vocational schools, which is more than the sum of all my predecessors. I improved the Elementary School quality after a dramatic fall in the nineties. Brazil figures nowadays in the international reports as a case of success for we left inertia. Not because we achieved an ideal level, but because we are in the right path. Thus, both UNESCO and the OECD recognized the effort Brazil made. What happened, though, against ENEM was a crime, not a fraud. A criminal was identified, judged, and punished with five years in jail. I’d like that the opposition, instead of criticizing me, supported me, because there was a crime, and the guilty party was identified and punished with five years in jail. Imagine if a saboteur were to be identified in the episode of the subway crater... We would support José Serra who was the governor at the time. But no... in this case, there was a mistake, a culpable homicide. That was not the case with ENEM. A person who was not part of the public administration and who acted as an insider in the printing office, which is the most modern in the country, committed a crime, was identified, and punished.

• ¹⁹²⁰

First step: considerations about the discursive coercions

In tune with our multidisciplinary agenda, it is extremely important to consider discursive aspects to describe and explain argumentation. We cannot, for reasons of space, delve too deeply into this aspect, but we will make some remarks we consider relevant, especially because there will be repercussions on the analysis itself.

Among the discursive coercions to be considered, the Volóchinovian (2017) notion of field and Chouliaraki and Fairclough’s (1999) and Fairclough (2003) concepts of social practice and order of discourse are always relevant to the analysis. Orders of discourse are constituted by three major patterns – discourses, conceived as ways of representing reality, i.e., how entities, properties, events, circumstances, values and reasons are correlated and interwoven in given texts; genres, understood as ways of (inter)acting, which constitute an essential part of our daily and institutional life and are closely tied to the existence of fields of social activity, such as economy, politics, education and religion, and their correlate social practices (e.g., in education, we could cite teaching, evaluation, enrollment, and so on); and styles, defined as ways of being, as consequence of the social identities we assume in the different practices we take part and that require the adoption of different behavioral patterns (in language, in gesture, in facial expressions, in clothing, among others). Besides that, it is of special importance to consider the discursive practice, as a moment of the social practice, which includes the processes of production, distribution, consumption and interpretation of texts.²¹

We will restrict ourselves to elaborating only two aspects of the set discussed above: field and genre. Our interview is a text that is connected to two fields of social activity: politics and media. In terms of politics, we must consider that the interview occurs during the mayoral campaign of the largest and richest city of the country. Moreover, in 2012, several alleged corruption schemes involving the federal government were being investigated and denounced. Haddad was the Worker’s Party (PT) candidate to the mayoral office in São Paulo and it was his party that oversaw the federal government. Finally, there were already signs of an economic crisis, which led to a political crisis, culminating in the impeachment of President Dilma Rousseff in 2016. In terms of media, we should point out that Globo Network of Communication is the largest communication company in the country, owning the TV Network with the highest penetration power in Brazil. It is, without any doubt, a relevant political player in the country.

In terms of genre, the interviews in SPTV news show are, by and large, argumentative texts with a very specific pattern: the interviewer builds an argumentative move, by presenting Data that support a Claim, which is construed as a question with an implicit argumentative orientation. The guest, then, has the opportunity of building Claims and presenting Data that act as Rebuttals towards the interviewer’s standpoint. We can say, thus, that these interviews are structured by representational conflicts and constitute an overlapping domain between belief formation/revision (convincement) – basically, the guests aim at convincing the audience about the relevance and the merits of their proposal and about their qualification to be mayor – and belief defensibility/decision-making (persuasion) – essentially, the guest’s goal to obtain votes from the audience. As a result of these complex coercions, we may argue that there is a constant preoccupation with the management of the candidate’s reputation ( ethos ) before the public. For this illustrative analysis, we focus on an argumentative move that directly and explicitly touches this subject matter.

Applying the multidimensional analysis: an illustration

We start the analysis by the two dimensions we consider basic – functional configuration and macrostructure –, since they enable the reconstruction of the argumentative move, a necessary step for the analysis of the other three dimensions. We will restrict this analysis in terms of topical salience regarding the epistemic problem in discussion: is Haddad’s reputation jeopardized or not?

Drawing on Toulmin’s model, we can reconstruct Tralli’s and Haddad’s argumentative moves as follows:²²

Figure 5
– Functional Configuration Analysis

As mentioned in the section above, it is important to see that the representational conflict is guaranteed by the operation which turns a Claim into a question. This opens up the possibility of Rebuttal and, thus, allows the guest to construe his perspective on the subject matter aiming both at convincement and persuasion.

Haddad’s Rebuttal is the most relevant part of the argumentative move here, both in terms of genre structure and topical salience. For this reason, we will focus on it. His Rebuttal aims at two different aspects:

• i.the first one concerns the sufficiency of the application of the Warrant: by showing every positive thing he has done in his term as a Minister of Education, Haddad seems to construe Tralli’s argument as insufficient to support the implicit claim (derived from the negative polarity question) that his image as an administrator is jeopardized. In other words, the frauds, mistakes and failures in ENEM would not erase all the good his administration had done before and the people/voters would acknowledge that. Tralli would be committing, then, in Haddad’s construal, the fallacy of hasty conclusion;

• ii.the second one concerns the Data itself: Tralli construes the events surrounding Brazilian National High School Exam as frauds, mistakes and failures, but he does not provide evidence that validates this assessment. He takes the evaluation for granted, as if there were not any other alternative conceptualizations. Haddad aims, then, at reconceptualizing this perspective by stating that the mentioned harmful events were, actually, crimes against the people and the administration. Thus, he construes Tralli’s move as a case of the problematic premise fallacy.

It is important to point out here that we are not stating that fallacies were, in fact, committed. We are saying that the linguistic construal of the Antagonist’s (Haddad) counter-argumentation may generate (in the audience) conceptualizations that the Protagonist’s move is fallacious: first, for drawing hasty conclusions (through the inappropriate application of Warrants to the question at issue or through the neglect for relevant criticism) and, second, for presenting problematic premises (false Data or unsupported Data). This way of understanding fallacies is more in line to the multidisciplinary approach envisaged here, since it focuses on argumentation as a process.

That said, we can observe through a more fine-grained lens how this move is structured. The graph below shows our proposal for the macrostructure of the whole fragment. The comments on the macrostructural dimension will already be accompanied by considerations pertaining the other dimensions: schematization, socio-affective grounding, and argumentative orientation: We include, then, the final excerpt of Haddad’s turn, which is not topically salient in terms of the epistemic problem in discussion (is Haddad’s image jeopardized or not?), but is rhetorically relevant in relation to the campaign, since it deals with self-positive representation and other negative representation ( VAN DIJK, 2003 ) in an electoral context.

Figure 6
– Macrostructural Analysis 23

1: You were Education Minister for six years and a half and during three years you were heavily criticized for the failures in ENEM (National High School Exam).

2: There were frauds and mistakes that ended up harming the lives of millions of students.

3: Your reputation as an administrator is jeopardized.

4: I’m the Education Minister who most expanded higher education in the country, with ProUni (University for All Program), with the expansion of federal institutions, with the new FIES (Student Loan Fund).

5: I’m the Education Minister who most expanded professional education in the country.

6: I, alone, built 224 technical/vocational schools, which is more than the sum of all my predecessors.

7: I improved the Elementary School quality after a dramatic fall in the nineties.

8: Brazil figures nowadays in the international reports as a case of success.

9: Both UNESCO and the OECD recognized the effort Brazil made.

10: What happened though against ENEM was a crime, not a fraud.

11: A criminal was identified, judged and punished with five years in jail/A person who was not part of the public administration and who acted as an insider in the printing office, which is the most modern in the country, committed a crime, was identified and punished.²⁴

12: I’d like that the opposition, instead of criticizing me, supported me.

13. We would support José Serra if a saboteur were to be identified in the subway crater episode.

As we can see (in green), Tralli uses a cumulative coordinative structure to support the negatively oriented question about Haddad’s supposedly jeopardized reputation as an administrator. This already shows that the interviewer somehow anticipates the insufficiency of each of the propositions alone to achieve a sound argument in support of the Claim.

In terms of argumentative argumentation, the lexical units selected by Tralli draw upon conceptual networks that evoke the public administrator’s agency and responsibility towards the events and the harms derived from them. The term fraud implies a deliberate action, on the part of the public administration, that threatens the rightful conduction and realization of the exam. Mistake suggests that the administration is responsible, by incompetence or carelessness, for the problems that happened in the exam. Failure , in turn, points to the wrongful application, elaboration or systematization of the whole process, assessment that could even lead to conceptualizations that the exam did not achieve its expected objectives.

Besides that, negative polarity questions are usually associated with an authorial stance that suggests that the propositional content is expected, in spite of the present circumstances that usually seem to contradict that expectation. In this case, what is expected is that Haddad’s image is, in fact, jeopardized; the present circumstances, in turn, would be his candidacy. The reasoning (and possible its effects) would be the following: “to be a candidate, one must not have his image jeopardized. Having committed so many mistakes in his term as Minister, why would Haddad try to be mayor of the biggest city in the country?” Thus, we can see this grammatical structure as a conflict-construing strategy that demands a counterargument from the Antagonist. This is, then, a fine example of how grammar can be studied and considered in terms of its argumentative orientation in texts.

All these linguistic cues instantiating a cumulative coordinative structure supporting a Claim that directly threatens the guest’s positive face ( BROWN; LEVINSON, 1987 ) certainly harms the projected ethos being construed by the candidate throughout not only the interview, but also the campaign. Discursive factors point to the need of saving face as well as guaranteeing an ethos of competence.

As the analysis allows hypothesizing, this is, apparently, one of the main aims of the Rebuttals (in yellow, as shown by the diamond-shaped squares, linked to Data (2) and to the Warrant that connects (1 & 2) to the Claim (3)), both composed of cumulative coordinative and serial patterns, although in different ways.

The central node of the Rebuttal that aims at the assessment of the events as frauds and mistakes – (2) – is (10) What happened, though, against ENEM was a crime, not a fraud . In this utterance, Haddad reconceptualizes the evaluation of the event, construing it as a crime . This strategy reduces the responsibility of the administration, since it backgrounds the candidate’s agency from enabling or causing the harmful events to not blocking them. Being accused of causing something harmful tends to have a greater moral impact, leading to negative evaluations of the agent’s behavior, then not blocking it ( NAGEL; WALDMANN, 2012 ). This reconceptualization is even more foregrounded by the parenthetical remark that the printing office was the most modern of the country. By hiring the most modern service possible, Haddad may cue conceptualizations that his actions were reasonable and correct. Thus, what happened would really put him under the label of victim, beside the millions of students affected.

This proposition – (10) –, then, is supported by (11) through a serial structure: A criminal – a person who was not part of the public administration and who acted as an insider in the printing office, which is the most modern in the country – was identified, judged and punished by five years in jail. This pattern signals that Haddad is possibly anticipating resistance against his correction ( fraud/mistake to crime ), turning it into a Claim itself.

As we can see, this Rebuttal – (10) – also acts as Data, along with (13) We would support José Serra if a saboteur were to be identified in the subway crater episode, to ground a claim for action – (12) I’d like that the opposition, instead of criticizing me, supported me – that, presumably, is not relevant for the discussion at issue: Haddad’s reputation as an administrator. In spite of that, though, the genre coercions and the context of situation – the electoral campaign – makes the move rhetorically relevant, as it aims both at construing the image of the opposition as unfair and uncapable of empathy and at building Haddad’s image as just and empathetic, as well as a victim of biased criticism.

A close look at the (dis)analogical schema that construes this part of the argumentation is relevant. In order to show how it functions, we will draw on the cognitive-linguistic notion of domain. Domains can be understood as set of multimodal knowledge that is drawn on to build conceptualization cued though language and other modalities (GONÇALVES-SEGUNDO, 2017b; LAGACKER, 2008; ZIEM, 2014 ). In metaphors and analogies, there are source domains and target domains ( KÖVECSES, 2010 ; GENTNER; SMITH, 2012 ). The target domain is at issue and its conceptualization is affected by the source domain’s network of conceived objects and of relations between concepts, presumably better known and/or more concretely experienced by the audience.

In metaphors, the conceptual structure of the source domain is used to understand the target domain, so that the source imagery is transferred to the target. In argumentative analogies,²⁵ the relations between the elements in the conceptual structure of the source are used as basis of comparison to reconceptualize the relation between elements in the target domain, usually aiming at changing our view of this domain. In disanalogies, some aspects of the projected structures are dissociated, in order to foreground discrepancies. Therefore, the existence of an analogy is implied.

Chart 1
– (Dis)Analogy between the Subway Crater Episode and the ENEM Events

Source Domain	Projection Criteria	Target Domain
Subway crater episode	Event	Problems in the application of ENEM
Deaths	(Negative) Consequences	Harms in the lives of students
Population	Affected	Students/Government
State government	Level of governance	Federal Government
Government Agents/José Serra (no saboteurs)	Responsibility	Criminals/Printing Office
Mistake/Culpable Homicide	Evaluation of the event	Crime/Victim
Condemnation	Expected Reaction (from the population)	Empathy

Source:

The two events – the subway crater episode and the problems in the application of the National High School Exam – can be compared, since there are a lot of common criteria which can be applied to them, as we can see in the italicized column. What the Antagonist Haddad does, however, is to show that, despite the common denominator, there is significant difference between the events, since, in terms of responsibility, the events surrounding ENEM had an external perpetrator – a criminal acting as an insider in a printing office –, whereas there were no saboteurs in the subway crater episode.

By construing this difference, Haddad targets the opposition, as he imparts responsibility to the State government concerning the deaths caused by the event, possibly aiming at generating attitudes of condemnation. The disanalogy operates insofar as he construes the problems in the application of ENEM as being the result of a crime, in a process that turn not only the population, but also his administration into victims, a strategy oriented towards obtaining empathy. Thus, it is through establishing grounds from comparison and, then, foregrounding the differences that Haddad executes a rhetorical move linked to pathos (a socio-affective category of analysis): by bringing to memory the subway crater episode, when 7 people died/were killed, Haddad may activate antagonistic attitudes towards the opposition, and empathetic attitudes towards himself, especially considering his textualized disposition for empathy and fairness.²⁶ The move highlights Haddad’s values, which support the Claim for empathy through an appeal to justice and reciprocation schema, which can be sketched as follows: “if someone acts in order to benefit another person in a given time, this same person should return the beneficial action to the other, when they need”.

The other Rebuttal construed by the Antagonist – the coordinated segments (4), (5), (7) and (8) –is also complex in its nature and it is more tied to the socio-affective category of ethos . It is an explicit construal of a positive self-image through values of competence and capacity ( I’m the Education Minister whomost expandedhigher education in the country; I’m the Education Minister whomost expandedprofessional education in the country; Iimprovedthe Elementary School quality after a dramatic fall in the nineties ), self-sufficiency ( I,alone, built 224 technical/vocational schools ) and high productivity and efficiency ( more than the sum of all my predecessors ).

It is basically a chain of Data– construed through a symptomatic scheme²⁷ that bridges the mentioned actions to the properties ascribed above, and the properties themselves to actually being a good administrator – that seems to aim at defeating the implicit Claim that his image is compromised, procedure that is done through contesting the sufficiency of the implied Warrant, as discussed in the beginning of this section.

The macrostructure is also rich to be observed. The Antagonist uses four cumulative coordinative arguments to support his Rebuttal, even backing (through serial structuring) some of them with numbers – see how (6) I, alone, built 224 technical/vocational schools supports the Claim/Data (5) that he is the Education Minister that most expanded professional education in the country – or through the authority and credibility of recognized international institutions – observe how (9) Both UNESCO and the OECD recognized the effort Brazil made backs (8) Brazil figures nowadays in the international reports as a case of success. This effort to give support to some of the coordinated propositions acts not only as an anticipation of possible criticism, but also as a means of self-promotion, since the interview itself can be used as a means of gaining the electorate’s trust and support (and, obviously, the consequent votes), in an implicit attempt of persuasion.

Final remarks

In this paper, we aimed to introduce the multidimensional model of argumentative analysis that we have been developing. To do so, we have drawn on several different sources from multiple disciplines, especially from Argumentation Studies, (Critical) Discourse Studies, Cognitive and Functional Linguistics, and Cognitive Sciences (namely, the embodied social paradigm in its weak/simple version, that acknowledges both modal and amodal structures).

The model considers five dimensions of analysis: 1. functional configuration; 2. macrostructure; 3. schematization; 4. socio-affective grounding; and 5. argumentative orientation. These dimensions can and should be developed individually (although this has already been done and it is still being done in several argumentation theories, with or without a multidisciplinary emphasis), but they should be especially approached in terms of possible correlations and combinations. In doing so, we can begin to deepen and enhance our understanding on the emergence of the perlocutionary effects of convincing, persuading and/or preserving ideology/reinforcing discourses.

In this perspective, we intended to show, albeit in a brief way, how to approach a single argumentative move in terms of the five dimensions and sketched possible ways of integrating these same dimensions in the analysis, considering linguistic, discursive and cognitive processes. Further research is needed to make any sort of generalization possible, covering corpora of multiple genres in several fields, encompassing social actors/arguers and audiences with distinct social identities and assuming different discourses. Experimental studies may also be useful in terms of understanding the dynamics of argumentation in terms of production, distribution and interpretation, as well as its perlocutionary effects.

REFERÊNCIAS

• AMOSSY, R. A apologia da polêmica. Coordenação de tradução: Mônica Magalhães Cavalcante. São Paulo: Contexto, 2017 .
• AMOSSY, R. O ethos na intersecção das disciplinas: retórica, pragmática, sociologia dos campos. In: AMOSSY, R. (org.). Imagens de si no discurso: A construção do ethos . São Paulo: Contexto, 2005 . p. 119-144.
• BARSALOU , L. W. Perceptual symbol systems . Behavioral and Brains Sciences, v. 22 , p. 577 - 660 , 1999 .
• BEDNAREK, M. Epistemological positioning and evidentiality in English news discourse: A text-driven approach. Text & Talk, Berlin, v.26, n.6, p. 635-660, 2006 .
• BROWN , P. ; LEVINSON , S. Politeness: Some universals in language usage . 2. ed. New York: Cambridge University Press, 1987 .
• CALDAS-COULTHARD, C. R.; IEDEMA, R. (org.). Identity trouble: Critical discourse and contested identities. Basingstoke, England: Palgrave Macmillan, 2008 .
• CAMERON, L. A dynamic model of empathy and dyspathy. Milton Keynes: The Open University, 2013 . (Living with Uncertainty, 6.).
• CARIOCA, C. R. Aspectos semânticos da evidencialidade nos trabalhos acadêmicos de grau. In: NOGUEIRA, M. T.; LOPES, M. F. V. (org.). Modo e modalidade: Gramática, discurso e interação. Fortaleza, Brasil: Edições UFC, 2011 . p. 143-163.
• CHOULIARAKI, L.; FAIRCLOUGH, N. Discourse in Late Modernity: Rethinking Critical Discourse Analysis. Edinburgh, Scotland: Edinburgh University Press, 1999 .
• CLARK , A . An embodied cognitive science? Trends in Cognitive Science, Maryland Heights, v. 3 , n. 9 , p. 345 - 351 , 1999 .
• CLARK, H. H. Using Language. Cambridge, England: Cambridge University Press, 1996 .
• CROFT, W.; CRUSE, A. Cognitive Linguistics. New York: Cambridge University Press, 2004 .
• DOURY , M . Preaching to the converted: Why argue when everyone agrees? Argumentation , Amsterdam, v. 26 , n. 1 , p. 99 - 114 , 2012 .
• EEMEREN, F.; GARSSEN, B.; KRABBE, E.; SNOECK-HENKEMANS, F.; VERHEIJ, B.; WAGEMANS, J. Handbook of Argumentation Theory. Dordrecht: Springer, 2014 .
• EEMEREN, F.; GROOTENDORST, R. A systematic theory of argumentation: The pragma-dialectical approach. New York: Cambridge University Press, 2004 .
• EEMEREN, F.; HOUTLOSSER, P.; SNOECK-HENKEMANS, A. F. Argumentative indicators in discourse : A pragma-dialectical study. Dordrecht: Springer, 2007 .
• FAIRCLOUGH, N. Critical Discourse Analysis: The critical study of language. 2. ed. Harlow, England: Longman, 2010 .
• FAIRCLOUGH, N. Analysing discourse: textual analysis for social research. London, England: Routledge, 2003 .
• FAIRCLOUGH, N.; FAIRCLOUGH, I. Political Discourse Analysis: A method for advanced students. London, England: Routledge, 2012 .
• FERREIRA, F. M. Analogia e argumentação no debate parlamentar: o caso da criminalização da LGBTfobia. 2018. 259f. Tese (Doutorado em Filologia e Língua Portuguesa) - Faculdade de Filosofia, Letras e Ciências Humanas, Universidade de São Paulo, São Paulo, 2018. Disponível em: http://www.teses.usp.br/teses/disponiveis/8/8142/tde-27022019-134324/ . Acesso em: 8 jun. 2020.
  » http://www.teses.usp.br/teses/disponiveis/8/8142/tde-27022019-134324/
• FIORIN, J. L. Argumentação. São Paulo: Contexto, 2015 .
• FREEMAN, J. Argument structure: Representation and theory. Dordrecht: Springer, 2011 .
• FREEMAN, J. B. Systematizing Toulmin’s warrants: An epistemic approach. In : HITCHCOCK, D.; VERHEIJ, B. (org. ). Arguing on the Toulmin model: New essays in argument analysis and evaluation . Dordrecht: Springer, 2006. p. 87 - 101 .
• GENTNER, D.; SMITH, L. Analogical reasoning. In : RAMACHANDRAN, V. S. (org. ). Encyclopedia of Human Behavior . 2. ed. Oxford: Elsevier. 2012 . p. 130 - 136 .
• GONÇALVES-SEGUNDO, P. R. Argumentação e perspectivação conceptual: possibilidades teórico-analíticas. In: SEMINÁRIO INTERNACIONAL DE ESTUDOS SOBRE DISCURSO E ARGUMENTAÇÃO, 4., Ilhéus. Anais [...], Ilhéus: Editus, 2018a . p.922-934.
• GONÇALVES-SEGUNDO, P. R. Discurso e prática social. In: BATISTA Jr, J. R. L.; SATO, D. T. B.; MELO, I. F. (org.). Análise de Discurso Crítica para linguistas e não linguistas. São Paulo: Parábola, 2018b . p. 78-103.
• GONÇALVES-SEGUNDO , P. R. Orientação argumentativa e cognição: A dinâmica de forças no debate acerca dos rolezinhos . Signo , Santa Cruz do Sul, v. 42 , n. 73 , p. 200 -212, 2017a.
• GONÇALVES-SEGUNDO , P. R. A relevância da noção de perspectivação conceptual ( construal ) no âmbito dos estudos do texto e do discurso: Teoria e análise . Letras , Santa Maria, v.27, n. 54 , p. 69 -100, 2017b. DOI: 10.5902/2176148529571 .
• GONÇALVES-SEGUNDO , P. R. Argumentação e falácias em entrevistas televisivas: por um diálogo entre o modelo Toulmin e a perspectiva textual-interativa . Revista Linha D’Água , São Paulo, v. 29 , n. 2 , p. 69 -96, 2016 . DOI: 10.11606/issn.2236-4242.v29i2p69-96 .
• GONÇALVES-SEGUNDO, P. R. A permeabilidade da Dinâmica de Forças: da gramática ao discurso. In: LIMA-HERNANDES, M. C. et al. (org.). Linguagem e cognição: Um diálogo interdisciplinar. Lecce: Pensa Multimedia Editores, 2015 . p. 163-185.
• GONÇALVES SEGUNDO, P. R. Convergências entre a Análise Crítica do Discurso e a Linguística Cognitiva: Integração conceptual, metáfora e dinâmica de forças. Veredas, Juiz de Fora, v.18, n.2, p. 32-50, 2014 .
• GONÇALVES SEGUNDO, P. R. Tradição, dinamicidade e estabilidade nas práticas discursivas: um estudo da negociação intersubjetiva na imprensa paulistana. 2011. 571f. (Doutorado em Filologia e Língua Portuguesa) – Universidade de São Paulo, São Paulo, 2011. 2. v. DOI: 10.11606/T.8.2011.tde-25042012-161141.
  » https://doi.org/10.11606/T.8.2011.tde-25042012-161141
• GONÇALVES-SEGUNDO, P. R. Posicionamento epistêmico e argumentação: articulações entre evidencialidade, modalidade epistêmica e provas retóricas. In: PIRIS, E. L.; RODRIGUES, M. G. S. (org.). Estudos sobre argumentação no Brasil hoje: modelos teóricos e analíticos. Natal: EDUFRN, 2020 . No prelo.
• GONÇALVES-SEGUNDO , P. R. ; RODRIGUES , R. B. R. Envolvimento e empatia: a solidariedade construída nas colunas de aconselhamento em revistas . Revista do GEL , São Paulo, v. 13 , p. 211 - 236 , 2016 . DOI: 10.21165/gel.v13i2.835 .
• GONÇALVES-SEGUNDO, P. R.; ZELIC, H. C. Relacionar-se é investir: Ideologia, cognição e metáfora no discurso sobre relacionamento em revistas femininas para o público adolescente. In: NASCIMENTO, L.; MEDEIROS, B. W. L. (ed.). Análise do Discurso e Análise Crítica do Discurso: Heranças, métodos, objetos. Saarbrücken, Germany: Novas Edições Acadêmicas, 2016 . p. 64-91.
• GRÁCIO, R. A. Para uma teoria geral da argumentação: Questões teóricas e aplicações didácticas. 2010. 434f. Tese (Doutoramento em Ciências da Comunicação) - Instituto de Ciências Sociais, Universidade do Minho, Braga, Portugal, 2010.
• HALLIDAY, M. A. K. Introduction to functional grammar. 3. ed. Revisado por Christian Matthiessen. London, England: Hodder Arnold, 2004 .
• HART , C . Discourse, grammar and ideology: Functional and cognitive perspectives . London, England: Bloomsbury, 2014.
• ITKONEN, E. Analogy as structure and process: Approaches in linguistics, cognitive psychology and philosophy of science. Amsterdam: Philadelphia: John Benjamins, 2005 .
• JOHNSON, R.; BLAIR, J. A. Lógica Informal: uma visão geral. EID&A: Revista Eletrônica de Estudos Integrados em Discurso e Argumentação, Ilhéus, v.14, p. 195-215, jul./dez. 2017 . Tradução de Paulo Roberto Gonçalves-Segundo et al .
• KOCH, I. As tramas do texto. São Paulo: Contexto, 2014 .
• KÖVECSES, Z. Metaphor: a practical introduction. 2. ed. New York: Oxford University Press, 2010 .
• LANGACKER, R. Cognitive grammar: a basic introduction. New York: Oxford University Press, 2008 .
• LANGSDORF, L. Argumentation as contextual logic: An appreciation of backing in Toulmin’s model. Cogency, Santiago, v.3, n.2, p. 51-78, 2011 .
• LEITÃO, S. O trabalho com argumentação em ambientes de ensino-aprendizagem: um desafio persistente. Uni-pluri/versidad, Medellín, v.12, n.3, p. 23-37, 2012. Disponível em: http://aprendeenlinea.udea.edu.co/revistas/index.php/unip/article/viewFile/15151/13196 . Acesso em: 8 jun. 2020.
  » http://aprendeenlinea.udea.edu.co/revistas/index.php/unip/article/viewFile/15151/13196
• LEWIS, M.; HAVILAND-JONES, J. M.; BARRETT, L. F. (org.). Handbook of Emotions. 3. ed. New York: London: The Guilford Press, 2008 .
• LINDBLOM, J. Embodied social cognition. New York: Springer, 2015 .
• MACAGNO, F. A means-end classification of argumentation schemes. In: EEMEREN, F.; GARSSEN, B. (org.). Reflections on theoretical issues in argumentation theory. Heidelberg: Springer, 2015 . p. 183-201.
• MAINGUENEAU, D. Ethos , cenografia, incorporação. In: AMOSSY, R. (org.). Imagens de si no discurso: A construção do ethos . São Paulo, Brasil: Contexto, 2005 . p. 69-92.
• MARÍN-ARRESE, J. Stancetaking and inter/subjectivity in the Iraq inquiry: Blair vs. Brown. In : MARÍN-ARRESE, J. M.-A.; CARRETERO, M.; ARÚS, J.; VAN DER AUWERA, J. (org. ). English modality: Core, periphery and evidentiality . Berlin: Boston: Mouton de Gruyter, 2013. p. 411-445.
• MARÍN-ARRESE, J. Effective vs. epistemic stance and subjecttivity in political discourse: Legitimising strategies and mystification of responsibility. In : HART, C. (org. ). Critical Discourse Studies in context and cognition . Amsterdam: Philadelphia: John Benjamins Publishing Company, 2011 . p. 193 -223.
• MARTIN, J.; WHITE, P. The language of evaluation: appraisal in English. New York: Hampshire: Palgrave Macmillan, 2005 .
• MATEUS , Elaine ; RESENDE , Viviane de Melo . O sistema posição-prática como categoria epistemológica: contribuições para análise de discurso crítica . Alfa: Revista de Linguística (São José do Rio Preto), v. 59 , n. 3 , p. 445 – 470 , dez. 2015 . DOI: 10.1590/1981-5794-1509-1 .
• MELO, I. F. de (org.). Introdução aos estudos críticos do discurso: teoria e prática. Campinas: Pontes, 2012 .
• MEYER, M. A retórica. São Paulo: Ática, 2007 .
• MIRANDA, N. S. Modalidade: O gerenciamento da interação. In: MIRANDA, N. S.; NAME, M. C. (org.). Linguística e cognição. Juiz de Fora: Ed. da UFJF, 2005 . p. 171-195.
• NAGEL, J.; WALDMANN, M. R. Force dynamics as a basis for moral intuitions. In: ANNUAL CONFERENCE OF THE COGNITIVE SCIENCE SOCIETY, 34., Austin. Proceedings […], Austin, TX: Cognitive Science Society, 2012 . p. 785-790.
• NEVES, M. H. de M. Texto e gramática. São Paulo: Contexto, 2007 .
• NIÑO, D.; MARRERO, D. The agentive approach to argumentation: A proposal. In: EEMEREN, F.; GARSSEN, B. (org.). Reflections on theoretical issues in argumentation theory. Heidelberg: Springer, 2015 . p. 53-67.
• PERELMAN, C.; OLBRECHTS-TYTECA, L. Tratado da argumentação: A nova retórica. São Paulo: Martins Fontes, 2002 [1958].
• PINTO, R. C. Evaluating Inferences: the nature and the role of warrants. In : HITCHCOCK, D.; VERHEIJ, B. (org. ). Arguing on the Toulmin model: New essays in argument analysis and evaluation . Dordrecht: Springer, 2006. p. 115 - 143 .
• PLANTIN, C. A argumentação: história, teoria, perspectivas. São Paulo: Parábola, 2008 .
• RESENDE, V. de M.; RAMALHO, V. Análise de Discurso Crítica. São Paulo: Contexto, 2006 .
• SLOB, W. H. The voice of the other: A dialogico-rhetorical understanding of opponent and of Toulmin’s rebuttal. In : HITCHCOCK, D.; VERHEIJ, B. (org. ). Arguing on the Toulmin model: New essays in argument analysis and evaluation . Dordrecht: Springer, 2006. p. 165 - 180 .
• SPERB, T. M.; JOU, G. I. Teoria da Mente: diferentes abordagens . Psicologia: Reflexão e Crítica , Porto Alegre, v. 12 , n. 2 , p. 287 - 306 , 1999 . DOI: 10.1590/S0102-79721999000200004 .
• SPERBER, D. et al . Epistemic Vigilance. Mind & Language, Oxford, v.25, n.4, p. 359-393, 2010 . DOI: 10.1111/j.1468-0017.2010.01394.x.
  » https://doi.org/10.1111/j.1468-0017.2010.01394.x.
• TALMY, L. Towards a cognitive semantics . Cambridge: MIT Press, 2000. v. 1.
• TOULMIN, S. Os usos do argumento. São Paulo: Martins Fontes, 2006 [1958].
• TOULMIN, S.; RIEKE, R.; JANIK, A. An introduction to reasoning . 2. ed. New York: Macmillan Publishing Company, 1984 [1978].
• VAN DIJK, T. Ideología y discurso. Barcelona, Spain: Ariel Lingüística, 2003 .
• VEREZA , S . Metáfora é que nem...: Cognição e discurso na metáfora situada . Signo , Santa Cruz do Sul, v. 38 , n. 65 , p. 02-21, 2013 . DOI: 10.17058/signo.v38i65.4543 .
• VEREZA, S. O lócus da metáfora: linguagem, pensamento e discurso. Cadernos de Letras da UFF, Niterói, n.41, p. 199-212, 2010 . Disponível em:
• http://www.cadernosdeletras.uff.br/joomla/images/stories/edicoes/41/artigo10.pdf . Acesso em: 8 jun. 2020 .
  » http://www.cadernosdeletras.uff.br/joomla/images/stories/edicoes/41/artigo10.pdf
• VERHEIJ, B. Evaluating arguments based on Toulmin’s scheme. In : HITCHCOCK, D.; VERHEIJ, B. (org. ). Arguing on the Toulmin model: New essays in argument analysis and evaluation . Dordrecht: Springer, 2006. p. 181 - 202 .
• VOLOSHINOV, V. Marxismo e filosofia da linguagem: Problemas fundamentais do método sociológico na ciência da linguagem. Tradução de Sheila Grillo e Ekaterina Vólkova Américo. São Paulo: Ed. 34, 2017 .
• WALTON, D. Methods of Argumentation. New York: Cambridge University Press, 2013 .
• WALTON , D. ; MACAGNO , F. Profiles of dialogue for relevance . Informal Logic , Windsor, ON, v. 36 , n. 4 , p. 523 -562, 2016 . DOI: 10.22329/il.v36i4.4586 .
• WALTON , D. ; REED , C. ; MACAGNO , F. Argumentation schemes . New York: Cambridge University Press, 2008.
• WOLFF , P. ; BARBEY , A. K. Causal reasoning with forces . Frontiers in Human Neuroscience , Lausanne, v. 9 , n.1, p. 01-21, 2015 . DOI: 10.3389/fnhum.2015.00001 .
• ZIEM, A. Frames of understanding in text and discourse: Theoretical foundations and descriptive applications. Amsterdam: Philadelphia: John Benjamins Publishing Company, 2014 .

• 2
  Many of the dimensions considered have, of course, already been theorized by several branches of argumentation theory. Not all of them deal with every dimension proposed neither have a particular interest in the cognitive and linguistic (or multimodal) aspects. We’d also like to warn the reader that it will be impossible to consider all the tradition on argumentative studies in this text; we will mainly focus on the ones with a greater impact in our proposal and a major influence in Brazil.
• 3
  Original: “ uma atividade discursiva (essencialmente verbal), social (de natureza cultural, contextualmente dependente), cognitiva (implica raciocínios necessários à fundamentação e avaliação crítica de afirmações), dialógica (simultaneamente responde a, e antecipa respostas da parte de outros), dialética (caracteriza-se como exame crítico de argumentos divergentes) e epistêmica (possibilita construção de conhecimento) ” ( LEITÃO, 2012 , p. 26).
• 4
  The terminology is perfectly extracted from both the quoted authors; the conceptions of formation, defensibility and maintenance are reframed and redefined in terms of the approach developed here. We will not discuss the particularities of the differences – which are not that sensible, by the way. We refer the reader to Niño and Marrero (2015) paper for a “purer” view of the concepts.
• 5
  Niño and Marrero (2015) discuss, in a rather engrossing way, this kind of argumentation, based on an example extracted from Doury (2012) .
• 6
  Construal is a central concept in Cognitive Linguistics. It can be defined as the semantic structuring of experience enacted in utterances. Construal guides conceptualizations and, thus, the reconstruction of meaning by interpreters. See Croft and Cruse (2004) for details.
• 7
  When applied to language and to semiotic practices, such a model requires language to be explained “in terms of its symbolical and interactive properties, considering its integration to other cognitive systems – such as memory, attention and categorization–and to sensory-motor systems – such as vision, audition, proprioception –, on the one hand, and in terms of its cultural and socio-historical grounding, which functions as a source of experiences and stimuli to the (dynamic) learning of language, including the different patterns of discourse activity in real contexts of interaction, on the other hand” (GONÇALVES-SEGUNDO, 2017, p. 72, our translation). For a detailed explanation of different cognitive models, see Lindblom (2015) .
• 8
  For details on the vagueness of the notion of field, its problems, but also its potential, see Freeman (2006) . We’d like also to make the following remark: the notion of field , proposed by Toulmin, was not conceived as a discursive phenomenon. This reading is being proposed here, in connection to our multidisciplinary and multidimensional model.
• 9
  A dialogical (VOLÓCHINOV, 2017) reading of Toulmin is also presented in terms of our model. It is not original, though. Slob (2006) also proposed a dialogical reading of Toulmin’s model, expanding the diagram horizontally and vertically in order to account for the different voices that cross and constitute an argumentative text. Our proposal is not identical to his, though, as it embeds the model on an argumentative situation – similar to what is proposed by Plantin (2008) or Gracio (2010) – conceived as an epistemic or practical problem, for which there are alternative answers and flows of argument supporting each of them. There is no room here to go into further detail.
• 10
  For reasons of space, we assume that the reader has a minimal familiarity with Toulmin’s layout of arguments. Also, we will focus on exposing our view, rather than justifying it. For a recent discussion on the model, see Eemeren et al . (2014). For its application in textual analysis, see Gonçalves-Segundo (2016) .
• 11
  Original: “ fatos específicos que sustentam uma determinada alegação ” (TOULMIN; RIEKE; JANIK, 1984 [1978], p. 37)
• 12
  There are several studies concerning the nature of Warrants. There is no room here for a review of the pertinent literature. We refer the reader to Freeman (2006 , 2011 ) and Pinto (2006) for debates and proposals on the theme.
• 13
  Protagonist (or Proponent) and Antagonist (or Opponent) should be understood as dynamic discursive and argumentative roles associated with different stances towards a subject matter. As roles, they are performed by social actors in the course of an argumentative interaction. For details, see Amossy (2017) . In the graphs below, the Protagonist utterances are represented in green, whereas the Antagonist utterances are in yellow.
• 14
  The numbers represent the functional component being referred to in the Figure below. The square to which the tail of the vector is linked is conceived as Data, while the square to which the head of the vector is connected is considered the Claim. Thus, the same square, such as (2), can be both Data (in terms of the Claim (3)) and Claim (in terms of the Data (1)).
• 15
  The diamond-shaped square represents the Rebuttal. The Rebuttal, as we have stated briefly in 2.1, can be oriented towards the Data, the Warrant, its application to the case at issue, the Backing or the Claim (see VERHEIJ, 2006 , for an even more fine-grained proposal). In the example, the line connects the Rebuttal to the Warrant (the vector that unites 1 to 2). The X signals the scope of the Rebuttal; in the case, the Warrant.
• 16
  The dashed arrow that links Rebuttal (4) to Claim (2) indicates that (4) is not directly in support of (2),but does lend strength to it by attacking Rebuttal (3), which would itself weaken the adhesion to (2). In the example, Rebuttal (3) undermines the Warrant from (1) to (2), while Rebuttal (4) defeats Rebuttal (3) itself.
• 17
  Fiorin (2015) has recently published a book which mainly discusses this dimension of argumentation.
• 18
  We further develop these connections in Gonçalves-Segundo (2018a).
• 19
  We must make a contextual note in order to enable a better understanding of the text. José Serra was São Paulo State governor during the time in which the subway crater episode happened. His political party (PSDB) is one of the most relevant in the country and is often associated with a center-right position. Haddad’s party (PT) is usually associated with a center-left position and was, by then, in power in the national level (presidency). Both parties have a recent history of being hegemonic in the country. The other parties tended to gravitate around them. In the present moment (2019-2020), however, we saw the strengthening of new political players, specially what we call the Bullet, the Bull and the Bible groups in Congress, usually linked to other parties and to a more conservative discourse/agenda. PSL, the party our current president, Jair Bolsonaro, was filiated to when elected, is certainly trying to cement the power it recently acquired in the last national election.
• 20
  All SPTV interviews with the main mayoral candidates are orthographically transcribed and are available, along with the video, in the show’s website: http://g1.globo.com/sao-paulo/eleicoes/2012/noticia/2012/09/fernando-haddad-do-pt-e-entrevistado-pelo-sptv.html . Access on: May 16, 2020. It is this version that we are using here in the analysis.
• 21
  For a detailed discussion on the relation between social practices and discourse, see Mateus and Resende (2015) and Gonçalves-Segundo (2018b).
• 22
  In reconstructing the argumentative move, it is often the case that the analyst must paraphrase the original text, which is not the best action in terms of descriptive precision. Nonetheless, it is usually a necessary procedure to understand the concatenation between the propositions that compose a move. The other dimensions of analysis, such as argumentative orientation, compensate this (possible) local fragility. Thin lines indicate implicit text. Italics signals inferential reasoning. In the excerpt under analysis, no Backing is given for the Warrant or the Data. That is why the space is left blank.
• 23
  In the macrostructural analysis, we tend to conserve, as much as possible, the original linguistic formulation. In some cases, though, adjustments are needed to avoid repetition and to maximize clarity in the diagramming. When relevant, the original elements can be recovered in the analysis of the argumentative orientation.
• 24
  The textual segment inserted after the slash (/) corresponds to a paraphrase that specifies the preceding segment.
• 25
  It is important to stress the adjective ‘argumentative’ here, since we are not talking about other uses of analogical reasoning, such as in learning (by extending a procedure used in case X, for example, to case Y, based on perceived similarities). For a detailed account on analogies, see Itkonen (2005) . For an account on argumentative analogies, see Ferreira (2018) .
• 26
  For more details on empathy and dyspathy in interaction and language, see Cameron (2013) and Gonçalves-Segundo and Rodrigues (2016) .
• 27
  According to Pragma-dialectics (EEMEREN; HOUTLOSSER; SNOECK-HENKEMANS, 2007), symptomatic argumentation schemas link what is stated in the standpoint (Claim) to what is presented as its symptom, sign or expression. In the example above, the actions executed by Haddad are signs of competence, capacity, self-sufficiency, productivity. These characteristics are considered, in terms of common sense, signs of a good administrator. Different to analogical schemas, symptomatic argumentation tends to rely more on already established relations between entities, properties, actions and circumstances. In cognitive terms, they tend to ratify the existent frames/domains than rearrange them. See Gonçalves-Segundo (2018a) for an initial discussion about this issue.

Publication Dates

• Publication in this collection
  05 Aug 2020
• Date of issue
  2020

History

• Received
  07 Aug 2018
• Accepted
  26 Feb 2019