Abstract: Even though interpreting is to a great extent about the interlinguistic reproduction of arguments, argumentation theory is almost completely overlooked by interpreting studies, which partly explains the frequent production of pragmatically inappropriate interpreted texts. Against the theoretical gap, the paper puts forward a descriptive argumentation approach to political speeches with a view to their simultaneous interpretation, in the attempt to make the case for a systematic contribution of argumentation studies to interpretation theory and training.
Keywords: argumentative equivalence, simultaneous interpreting, source texts (STs), strategic manoeuvring.

1. Introduction
The omnipresence of argumentation in everyday verbal communication is the hinge of argumentation studies (van Eemeren et al., 1996, pp. 1-2) and the main reason for their flourishing, from the classical period to a renewed spate of interest last century. What is probably less evident and certainly less studied is the fact that the socio-professional needs of the globalised world repeatedly demand that argumentation be reproduced in another language to cater for interlinguistic communication needs. In this respect, the present paper analyses how argumentation is and should be reproduced in interlinguistic settings requiring the interpreting service.

Argumentation is “the dominant mode of discourse in many interpreted situations” (Marzocchi, 1997, p. 182) and interpretation consequently implies a continuous argumentative interaction, thereby requiring at least an intuitive knowledge of the appropriate contextual use of arguments on the part of interpreters (Marzocchi, 1997, p. 184). Notably, when the predominant focus of a communicative situation is on the discursive attempt to resolve a difference of opinion, the quality of the interpreter’s performance is assessed on the basis of his/her ability to convey the argumentative purpose of the original text, “possibly to the detriment of other kinds of equivalence or of received ideas concerning fidelity” (Marzocchi, 1997, p. 183). This particularly holds true for political argumentation, in which the systematic and subtle implementation of strategic manoeuvring (Zarefsky, 2009, p. 115) to overcome a conflict between different lines of action demands specific equivalence standards not only concerning the content of the message but also its persuasive and ethotic dimensions, which are less important in other communicative events such as specialist conferences.

However, despite the significant development of argumentation studies in the last few decades and the argumentative character of several interpreted situations, the extent to which knowledge of argumentation theory by the interpreter could benefit the overall comprehension of the original or source text and favour the production of the respective interpreted text has yet to be explored (Crevatin in Marzocchi, 1998, Preface, p. xiv). More precisely, the growing interest in argumentation has gone mostly unnoticed in interpreting scholarly settings, in spite of its multidisciplinary and interdisciplinary nature. This is partly due to the fact that interpreting research is a fairly young and largely unexplored discipline (Garzone & Viezzi, 2002, p. 2), and the contribution of argumentation makes up one of most overlooked albeit potentially fruitful domains.

The paper outlines the preliminary findings of a broader PhD project focusing on the empirical examination of recurrent argument schemes in a multilingual corpus of political speeches. By harnessing the hermeneutical and contrastive functions of argumentation analysis (Marzocchi, 1998, p. 8), the study uncovers substantial differences in speakers’ adoption of argument schemes, thereby making the case for enhanced language-specific, argumentation-driven interpreter preparation. In this respect, the present paper is primarily concerned with the applicability of argumentation concepts and methods to interpreter training.

The study of political argumentation in interpreting settings entails a specific focus on simultaneous interpreting, since it is the most widely adopted modality for the interpretation of political speeches. However, the remarks will also hold true for consecutive interpreting because, though progressively ousted by the simultaneous modality, it is still adopted and included in university curricula, and thus fits the training-oriented rationale of the study. In section 2, evidence of interpreters’ difficulty in reproducing the original arguments in the interpreted text is shown and discussed; sections 3 and 4 will respectively briefly present the reference corpus and illustrate the methodological underpinnings of the study; section 5 will explore the main findings, which are eventually discussed in section 6.

2. Interpreting is first and foremost a translational activity, involving a “source-text induced target-text production” (Neubert, 1985, p. 8).
Reflection on the relation between the source text (ST) and the interpreted text (IT) is therefore inescapable, and the ST-IT comparison, aiming at determining relations of equivalence and standards of quality, is at the heart of translation and interpreting studies.

Interpreting can be distinguished from other types of translational activity by its immediacy:
Interpreting is a form of Translation in which a first and final rendition in another language is produced on the basis of a one-time presentation of an utterance in a source language. (Pöchhacker, 2004, p. 11)

Therefore, unlike written translation, interpreting is characterised by ephemeral presentation and immediate production. These severe time and cognitive constraints intensify in the simultaneous modality, in which the interpreter listens to the ST through headphones and delivers the IT into a microphone, almost simultaneously, with a slight delay (décalage) between message reception and message production. Therefore, the hic et nunc nature of the activity confronts interpreters with the task of constructing a mental representation of the text as it progressively unfolds. This further challenges the attainment of an “acceptable” degree of equivalence, rendering interpreted texts more prone to substantial pragmatic shifts than translated texts, which are instead supposed to be produced after careful work. Indeed, what is required in interpreting is not equivalence in toto, but the equivalence of the communicative function (Viezzi, 1999, p. 147), or “pragmatic quality” (Kopczyński, 1994, p. 190) ensuring a high degree of “intertextual coherence” (Straniero Sergio, 2003, p. 147) between the ST and the IT.

However daunting, the task can be appropriately performed after the development of procedural competence (Riccardi, 2005, p. 755) and only by relying on extra-textual knowledge and anticipation, i.e. prediction “based on previously acquired contextual and discoursal knowledge” (Garzone, 2000, p. 73). This means that training and advance preparation play a major role in determining the success of an interpretation (Gile, 1995, pp. 144-145), in that they are vital resources making up for the lack of the necessary time to process a novel text. In other words, professional interpreting is not limited to the actual oral translation of a speech but covers a larger lapse of time, catering for the otherwise insufficient minutes or hours interpreters would have qualitatively to perform an unnatural (Riccardi, 2005, p. 756), extremely delicate and “unstable” activity, potentially engendering substantial pragmatic shifts (Colucci, 2011).

Building on its “instability” and based on the translational needs for compensation (Harvey, 1995) and reformulation (Falbo, 1999), interpreting has been compared to a chemical experiment, in which matter and energy remain unchanged before and after the operation, despite the likely alteration in their distribution (Snelling, 1999, p. 203); matter may be considered to refer to the propositional content of the original message and energy is identifiable with the pragmatic force of the ST. However evocative and vivid, though, the comparison is not a faithful description of the activity, but only a useful methodological suggestion, because interpreting is not subject to laws of physics; both matter and energy are systematically threatened by alteration, more often resulting in “dissolution”, i.e. omission of text segments or mitigation of the pragmatic force of statements, and less frequently leading to “aggregation”, i.e. arbitrary additions and “parallel formulations” (Straniero Sergio, 2003, pp. 159-160).

The quantities of matter and energy can be altered by various factors, which may be said to fall into two categories. On the one hand, there are intrinsic factors compounding the translational task, among which linearity or “the fact that the text becomes available only gradually” (Shlesinger, 1995, p. 193); co-text dependence (Garzone, 2000, p. 71), leading interpreters to lose sight of the context; and the specific language combination, posing targeted problems mainly deriving from the different syntactic rules of the language-pair in question. However challenging, though, these intrinsic obstacles are gradually overcome through the development of procedural competence.

On the other hand, there are “contextual” triggers of matter and energy dissolution and/or aggregation, and they are all ascribable to the interpreter’s scarce preparation or insufficient individual knowledge regarding the topic, the speaker and the type of text (Riccardi, 1998, pp. 173-174). Unlike intrinsic constraints, these factors have a more pronounced individual dimension and are directly linked to interpreter training, particularly to the need gradually to develop the “textual and discoursal competence” (Garzone, 2000, p. 73) enabling interpreters to tackle the speeches with a reasonable degree of confidence. In this respect, extra-linguistic knowledge-related mistakes can be considered a direct consequence of the scarce attention devoted to STs in interpreting studies (Garzone, 2000, p. 69), which is ascribable to a general underrating of pragmatics (Viaggio, 2002, p. 229) that, especially in the interpretation of political speeches, is partly determined by a marked neglect of argumentation theory (Marzocchi, 1998).

The consequences of incomplete interpreter curricula and scarce preparation of the genre, content and ethotic dimension of the ST are evident in the example provided in Table 1, showing the interpretation of an excerpt of Obama’s 2009 Inaugural Address, broadcast live on Italian television and performed by a professional interpreter. The example is drawn from CorIT, the television interpreting corpus developed at the University of Trieste. The Italian interpretation has been retranslated into English and displayed in the right column.

Rather than an experiment, Table 1 looks more like a chemical disaster. Despite the partially conjectural nature of mistake aetiology in Error Analysis (Falbo, 2002, p. 115), the origin of a mistake can generally be gleaned with a certain degree of precision during the comparison between the ST and the IT. All the more so when, as in the above passage, neither incomprehensible referents nor particular difficulties stand out. In this particular case, the ST delivery speed was at times considerable and the pressure deriving from interpreting live the first message of the newly-elected President of the United States cannot be overlooked. Neither can the peculiarities of TV interpreting; whereas in conference settings interpreters are generally given the texts of the speeches and therefore have at least a few minutes to prepare, in televised interpreted events they hardly ever have the opportunity to see the text (Straniero Sergio, 2003, pp. 169-170). However, despite being unquestionable compounding factors, information density, unhelpful employers and excessive delivery speed are not excuses for poor translation, because their potentially detrimental consequences can be partially obviated by attentive and selective listening (Palazzi, 2007, p. 264).

Therefore, given the groundlessness of the above hypotheses, the “dissolution of argumentation” is likely to have been triggered by the presence of a straw man argument in the ST. Since “when an opponent’s position is distorted or exaggerated in a straw man argument, the effect is often to divert the line of argument to irrelevant issues” (Walton, 2004, p. 22), the sudden shift towards irrelevance may have compounded comprehension and challenged relevant translation. In simpler terms, an interpreter, uninformed of the exaggerating implications of the straw man, has difficulties in grasping the argumentative move and may reasonably dread an impending launch of a personal attack by the speaker or, in Politeness terms, a face-threatening act (FTA) against another politician. Its translation would require enhanced attention to proper names, politeness strategies and the careful reproduction of the pragmatic force of the message, exposing the interpreter to the risk of committing an FTA against him/herself, i.e. staining his/her interpreter reputation.

This is only an example, but it corroborates the hypothesis that unawareness of the specific argument strategies adopted by source language speakers is bound to add a further obstacle to the attainment of a quality interpretation and put the interpreter at a disadvantage against the rhetorical abilities of politicians.

3. The corpus
The study is based on a recently assembled multilingual corpus that is composed of three hundred and thirteen political speeches on the current financial and economic crisis, which are almost equally divided into American, British and French speeches; a hundred and nine were delivered by Barack Obama, a hundred and one by David Cameron, a hundred and three by Nicolas Sarkozy and François Hollande (respectively fifty-three and fifty). All the speeches were delivered between 2008 and 2014 and make up a sample of the discourse on the financial and economic crisis that has monopolised political communication over the last few years, probably outranking the discourse on the war on terror, at least in the United States. The speeches have been selected for their being delivered within international settings requiring the interpreting service (e.g. the G20), but also national speeches have been taken into account, as they are often chosen for exam sessions in translation and interpreting faculties.

The corpus has been named ARGO, for its reference to argumentation and after Ulysses’ dog because, just as the dog recognised his owner when he returned to Ithaca twenty years after leaving, the corpus is meant to promote the recognition and internalisation of recurrent argumentative strategies before the interpretation.

4. The methodological scope of interpreting-oriented argumentation analysis
The interpreting-oriented argumentation analysis is therefore a ST analysis, having little to do with the ST-IT comparison and rather aiming at studying the argumentation of source language speakers with a view to its recognition and reproduction in translation, with the observation of the interpreters’ failures in reproducing argumentation patterns merely serving as a rationale for targeted argumentation analyses. The scope of this peculiar kind of analysis is determined by its specific informative, profession-oriented needs, which are only catered for by the descriptive study of content-related argumentation. More specifically, interest in argumentation in interpreting research is directed towards the findings of applied studies, focusing on both text analysis and the didactics of argumentation in the attempt to promote anticipation and enhance the argumentative competence of interpreter trainees (Marzocchi, 1998, p. 43).

Among all the crucial concepts in argumentation theory (van Eemeren, 2001), the present research was limited to the detection and description of argument schemes, because their study provides insights into the generalised content of arguments (Garssen: in van Eemeren, 2001, p. 72) and provides textual and contextual information (van Eemeren, 2001, p. 20), thereby catering for the descriptive and content-related need of interpreting research.

Yet, however insightful theoretical dissertations may be, the boundaries of interpreting-oriented argumentation analysis and the specific focus of the paper are best explained by means of an example. Take the formal scheme of the argument from distress (Walton, Reed & Macagno, 2008, p. 334):

Individual x is in distress.
If y brings about A, it will relieve or help to relieve this distress.
Therefore, y ought to bring about A.

It is not unrelated to political communication, but the relevance of its internalisation by the interpreter of political speeches is questionable, as it does not provide him/her with a particular advantage during the interpreting process, but only gives a content-abstract indication of what may or may not come up in a speech; moreover, interpreters are not interested in being told normatively how to argue, because their task is to reproduce the argumentation of others. Rather, they may benefit from knowing in advance the contextual implementation of the scheme by a given speaker.

(1) But it [the American Recovery and Reinvestment Act] will give millions of families resigned to financial ruin the chance to rebuild. (my emphasis) (Remarks by President Obama on the Mortgage Crisis, 18th February 2009)

Example (1) is one of the several enthymemic implementations of the argument from distress that have been found in the American sub-corpus. Particularly, the sentence rests upon a specific form of the argument, that can be named Argument from Middle-Class Distress.

The Middle Class is in distress.
If the government implements act x, it will relieve or help to relieve this distress.
Therefore, the government ought to implement act x.

Only content-related schemes have the potential to warn interpreters of the recurrent topicality of discursive practices legitimising specific political courses of action, like Obama’s recurrent leveraging the middle class topos to gain consensus for economic policies. In the following section, other examples drawn from the corpus are examined in context, highlighting the focus on strategic manoeuvring in political argumentation and its key role in supplementing the education, background knowledge and procedural competence of the interpreter.

5. Findings
In pragma-dialectical terms, this chiefly content-related analysis may be described in terms of highlighting the topical potential of a specific instance of argumentation. The practice is highly relevant to interpreters’ needs because, in the light of the predictability (Zarefsky, 2009, p. 115) of political argumentation, a descriptive account of the schemes recurrently used by a given speaker in a given context may raise text expectations in the interpreters’ minds during the training and/or preparation phases, thereby easing the inferential and translation processes during the interpretation.

However, the predictability of political argumentation concerns not only the topical dimension but the whole of argumentation (Zarefsky, 2009, p. 115), therefore also how it is presented and how it is adapted to audience demands, following the strategic manoeuvring categorisation (van Eemeren, 2010, p. 93).
Indeed, analysis of strategic manoeuvring in the corpus uncovers several argument recurrences; particularly, substantial disparities in the speakers’ adoption of argument schemes stand out. Unlike Obama, for instance, David Cameron presents himself as an expert on the economic crisis, but he also uses simple argumentation. For example, he repeatedly compares the world economy to a global race in which every country runs alone and aims at winning the race. By doing this, he generally abides by the following argument scheme:
The UK is running a global race.
It risks being outpaced by other countries.
Therefore, it must run faster than others.

The scheme alone can act as a cognitive support by helping interpreters anticipate the general content of several speech passages, but highlighting the recurrent presentational devices associated with the Global Race Argument provides a further and equally helpful lexico-syntagmatic support, reducing processing and decoding efforts and paving the way for a higher quality translation. Quite unsurprisingly, fast, quick, speed, win/lose, keep up with are the recurrent lexical indicators of the scheme, that is also accompanied by vivid and less predictable sentences like “the world is breathing down our neck” and “in this global race you are quick or you are dead”. Pointing out the presentational devices typical of a given argument scheme is instrumental, in that it warns interpreters of what the speaker is accustomed to saying, instead of limiting the research applicability to the indication of the generalised content of the arguments found in a specific discourse.

However, the sole focus on topical potential and presentational devices is not enough to give a faithful overview of the speaker’s argumentative routines, because Cameron actually does not always resort to the Global Race Argument when dealing with the world economy: the presence of the argument is considerable only in national addresses, while in international settings it seems to be replaced by its opposite, that can reasonably be named Argument from Multilateral Economic Cooperation. The argument, shifting the argumentation to a completely different ground, has the following scheme:
We are living in an interconnected world.
Crisis in one country affects all the other countries.
Therefore, cooperation is needed for the sake of world economy.

The context-dependent alternation between the Global Race Argument and the Argument from Multilateral Economic Cooperation corroborates the pragma-dialectical tenet that argumentation is systematically adapted to the audience. In this particular case, the standpoint is adapted, or rather overturned, too because, when dealing with the role of nations in the world economy, Cameron defends a different standpoint depending on the relevant audience. From an interpreter’s point of view, this means knowing in advance what the speaker generally says (and therefore will probably say) in a specific communicative context.

The adaptability, culture- and context-based heterogeneity of the above arguments are substantially confirmed by the analysis of ARGO. It is true that also a few similarities have been detected among the different sub-corpora, but they concern typically “political” arguments, namely the straw man argument, giving the opponents a bad name, the topos of history (Reisigl & Wodak, 2001, p. 80), comparing the present crisis to past predicaments, and the locus of the irreparable (Zarefsky, 2009, p. 123), urging the implementation of economic policies. However, apart from these distinctly “political” arguments, the speakers generally steer their speeches towards highly different thematic and argumentative corners, which is, however, also a consequence of the specific national implications of the global recession. Their choice of argument schemes differs substantially, as far as topical selection, adaptation to audience demands and presentational devices are concerned. For instance, in line with his focus on the middle class, Obama tells vivid stories of the resilience of the American people in the face of the crisis, showing a predilection for anecdotal arguments (Govier & Jansen, 2011, p. 75) breaking the flow of rational argumentation.

(2) When Bryan Ritterby was laid off from his job making furniture, he said he worried that at fifty-five, no one would give him a second chance. But he found work at Energetx, a wind turbine manufacturer in Michigan. Before the recession, the factory only made luxury yachts. Today, it’s hiring workers like Bryan, who said, “I’m proud to be working in the industry of the future”. (State of the Union Address 2012)

Cameron and Sarkozy opt for a more specialist discourse, presenting themselves as experts on the economic crisis. While Cameron explains the causes of the crisis and the looming consequences of certain policies with intricate slippery slope arguments (Walton, Reed & Macagno, 2008, p. 114), Sarkozy recurrently adopts an argument that may be named Argument from Need for Regulated Capitalism, defending the need for state intervention in the economy, which is perfectly in line with his interventionist political stance (Mayaffre, 2012, p. 15).

(3) L’idée de la toute puissance du marché qui ne devait être contrarié par aucune règle, par aucune intervention politique, était une idée folle. L’idée que les marchés ont toujours raison était une idée folle.

The very idea of a free, non-regulated market was a crazy idea. The idea of markets being always right was a crazy idea. (my translation, my emphasis)
(Sarkozy’s Speech in Toulon, 25th September 2008)

Rather, Hollande’s argumentation heavily relies on the argument from thrift (Reisigl & Wodak, 2009, p. 105).

(4) 2014, ce sera aussi l’année de décisions fortes. […] D’abord, je veux réduire la dépense publique. Nous devons faire des économies partout où elles sont possibles.

2014 will also be the year of tough decisions. […] First, I want to reduce public spending. We must save whatever we can. (my translation, my emphasis)
(Hollande’s New Year’s Greetings, 31st December 2013)

This is actually and formally an argument from sacrifice, enabling the speaker to anticipate the persistence of tough economic times and legitimise government policies.

6. Conclusion
The length of this paper prevents a thorough listing of the most recurrent schemes, but the examples are indicative of the variations in the speakers’ argumentative choices when faced with the same topic. The analysis shows that the most recurrent argument schemes in the ARGO corpus are culture-specific and context-dependent, and also personal or idiosyncratic in some way because, even though they share the same culture and communicative contexts, Sarkozy and Hollande resort to different argument schemes.

This has specific implications for the interpreter: in the light of the heterogeneity of arguments and given the interpreter’s difficulty in processing argumentative passages, the study suggests that previous knowledge of topic-related and speaker-related argument schemes renders certain aspects of STs comparatively predictable, and may therefore act as a cognitive and lexical support during interpretation. This fosters the systematic adoption of argumentation analysis as a source text research methodology, providing interpreting research with findings of direct training applicability, potentially enhancing the communicative skills of interpreter trainees by gradually strengthening their discourse competence. In this respect, a promising line of research could lie in the extension of the practice to other political topics, speakers and languages, in the attempt progressively to build up repositories of data-driven hints on the predictability of political STs.

Moreover, in addition to finding instrumental scope in interpreting research and training, argumentation theory also provides a theoretical and operational contribution to interpreting activity, suggesting a particular interpreter approach to ST argumentation in the attempt to attain an acceptable degree of argumentative equivalence in the IT. Just as the argumentation analyst takes a differentiated view of manoeuvring rather than viewing it as a monolithic whole (van Eemeren, 2010, p. 93), the interpreter of argumentative texts ought to see his/her activity in relation to what could be named the Strategic Manoeuvring Equivalence Triangle.

In order to recognise and reproduce the speaker’s illocution and perlocution, thereby guaranteeing the correct transfer of argumentation patterns, three aspects have to be taken into account:
– topical coherence, respect for the topical choice made by the source language speaker;
– respect for audience demands, even though the relevant audience may change from the ST to the IT. For example, in the European Parliament the ST audience corresponds to the target audience, as all the participants share the same context of situation and the same communicative interests. Yet, during an interpreting exam, for instance, the IT audience is composed of interpreting professors who are in charge of assessing the quality of the student’s performance. Or, in television interpreting, the audience shifts from the actual participants in the original event and the TV audience sharing the language and culture of the speaker to the TV spectators in the target culture, which renders the activity an example of documentary interpreting as opposed to instrumental interpreting (Viezzi, 2013, p. 384). However, respect for audience demands is paramount under any circumstances, because the interpreter’s task is to show the speaker’s attitude towards his/her audience.
– presentational coherence, or the preservation of the ST presentational devices. This third category is controversial, because the pragma-dialectical notion of “presentational devices” encompasses a variety of features that cannot be always reproduced in the IT. Think of alliteration, that is often bound to perish against the intrinsic differences between languages. However, its reproduction is not demanded despite the rhetorical mitigation in the IT, as interpreters are generally dispensed from the task of “translating the untranslatable”, to paraphrase Reboul’s assertion on the untranslatability of rhetoric (1991, p. 110). Yet the systematic study of ST argumentation highlights a number of presentational devices that can be more easily reproduced, such as Obama’s recurrent use of the historical analogy between the crisis and the Great Depression, or Sarkozy’s and Cameron’s habit of appealing to liberal and conservative presumptions (Zarefsky, 2009, p. 122).

Against this background, argumentative equivalence may be viewed as the faithful reproduction of the features of strategic manoeuvring into the IT. The focus on words and their contextual meaning, which is inherent in the analysis of strategic manoeuvring, is certainly not unknown to interpreting studies. For instance, Gile (1995, pp. 35-36) addresses the interpreter’s need to pay attention to both form and function in terms of dealing with content and packaging, while the need to adapt the argumentation to audience demands may be inferred from Kopczyński’s (1994, p. 190) thorough analysis of the situational variables of interpreted events and the considerable number of studies on the pragmatic aspects of conference interpreting (Schäffner, 1997; Setton 1999). However, the literature on interpreting lacks a specific focus on argumentation, whose distinctive features are only tangentially addressed and whose scattered, incomplete and non-harmonised study stands in the way of its internalisation by interpreters and further shapes the heterogeneous and often ambiguous metalanguage of interpreting studies (Gambier, 2008, p. 64). The focus on strategic manoeuvring, instead, provides a comprehensive and intuitive framework for understanding the threefold notion of argumentation and, by implication, the importance and nature of argumentative equivalence between the source and the interpreted text.

This does not mean that the swift internalisation of the notion of strategic manoeuvring and the superficial study of ST argumentation are destined to solve the ever-present problems of performing “unstable” interpreting activity. Rather, the relevance of the pragma-dialectical approach to STs in interpreting research has to be sought in its explanatory potential, highlighting the salient features of a given instance of argumentation, and in its methodological and operational guidance, shedding light on the best way to transfer the pragmatic and argumentative nuances of STs. This is, in substance, the contribution of argumentation theory to interpreting research and, consequently, to the interpreting profession: it promotes matter preservation by emphasising the ST topical potential; it helps preserve energy by showing the speaker’s strategies of adaptation to audience demands; and it promotes both energy and matter preservation by highlighting the distinctive presentational devices of STs, thereby preventing chemical disasters.

References
Colucci, C. (2011). Modalisation and pragmatics in simultaneous TV interpreting. A case study: American presidential debates. The Interpreters’ Newsletter, 16, 61-80.
Eemeren, F. H. van, Grootendorst, R., & Snoeck Henkemans, F. (1996). Fundamentals of argumentation theory. A handbook of historical backgrounds and contemporary developments. London/New York: Routledge.
Eemeren, F. H. van (2001). The state of the art in argumentation theory. In F. H. van Eemeren (Ed.), Crucial Concepts in Argumentation Theory (pp. 11-26) Amsterdam: Amsterdam University Press.
Eemeren, F. H. van (2010). Strategic maneuvering in argumentative discourse. Amsterdam/Philadelphia: John Benjamins Publishing Company.
Falbo, C. (1999). Morfosintassi e riformulazione. In C. Falbo, M. Russo & F. Straniero Sergio (Eds.), Interpretazione simultanea e consecutiva. Problemi teorici e metodologie didattiche (pp. 175-188). Milano: Hoepli.
Falbo, C. (2002). Error analysis: a research tool. In G. Garzone, P. Mead & M. Viezzi (Eds.), Perspectives on Interpreting (pp. 111-127). Bologna: CLUEB.
Gambier, Y. (2008). Stratégies et tactiques en traduction et interprétation. In G. Hansen, A. Chesterman & H. Gerzymisch-Arbogast (Eds.), Efforts and Models in Interpreting and Translation Research : a Tribute to Daniel Gile (pp. 63-82). Amsterdam/Philadelphia: John Benjamins Publishing Company.
Garssen, B. (2001). Argument schemes. In F. H. van Eemeren (Ed.), Crucial Concepts in Argumentation Theory (pp. 81-100) Amsterdam: Amsterdam University Press.
Garzone, G. (2000). Textual analysis and interpreting research. The Interpreters’ Newsletter, 10, 69-88.
Garzone, G., & Viezzi, M. (2002). Introduction. In G. Garzone & M. Viezzi (Eds.), Interpreting in the 21st Century. Challenges and Opportunities (pp. 1-11). Amsterdam/Philadelphia: John Benjamins Publishing Company.
Gile, D. (1995). Basic concepts and models for interpreter and translator training. Amsterdam/Philadelphia: John Benjamins Publishing Company.
Govier, T., & Jansen, H. (2011). Anecdotes and arguments. In E. Feteris, B. Garssen & F. Snoeck Henkemans (Eds.), Keeping in Touch with Pragma-Dialectics (pp. 75-88) Amsterdam/Philadelphia: John Benjamins Publishing Company.
Harvey, K. (1995). A descriptive framework for compensation. The Translator, 1 (1), 65-86.
Kopczyński, A. (1994). Quality in conference interpreting: some pragmatic problems. In M. Snell-Hornby, F. Pöchhacker & K. Kaindl (Eds.), Translation Studies. An Interdiscipline (pp. 189-198). Amsterdam/Philadelphia: John Benjamins Publishing Company.
Marzocchi, C. (1997). The analysis of argumentation and its relevance to interpretation research and theory. In K. Kinga & K. János (Eds.), Transferre Necesse Est. Proceedings of the 2nd International Conference of Current Trends in Studies of Translation and Interpreting, 5-7 September, 1996, Budapest, Hungary (pp. 179-185). Budapest: Scholastica.
Marzocchi, C. (1998). Interpretare il discorso argomentativo. Teoria, aspetti e problemi. Trieste: Scuola superiore di lingue moderne per interpreti e traduttori.
Mayaffre, D. (2012). Nicolas Sarkozy. Mesure et Démesure du Discours (2007-2012). Paris: SciencesPo. Les Presses.
Neubert, A. (1985). Text and translation. Leipzig: Verlag Enzyklopädie.
Palazzi, M. C. (2007). Tipologie di interpretazione e formazione degli interpreti. In M. T. Musacchio & G. Henrot Sostero (Eds.), Tradurre: Formazione e Professione (pp. 257-275). Padova: Cleup.
Pöchhacker, F. (2004). Introducing interpreting studies. London/New York: Routledge.
Reboul, O. (1991). Introduction à la rhétorique. Paris: PUF, Presses Universitaires de France.
Reisigl, M., & Wodak, R. (2001). Discourse and discrimination. Rhetorics of racism and antisemitism. London/New York: Routledge.
Reisigl, M., & Wodak, R. (2009). The discourse-historical approach (DHA). In R. Wodak & M. Meyer (Eds.), Methods of Critical Discourse Analysis (pp. 87-121). London: Sage.
Riccardi, A. (1998). Interpreting strategies and creativity. In A. Beylard-Ozeroff, J. Králová & B. Moser-Mercer (Eds.), Translators’ Strategies and Creativity (pp. 171-179). Amsterdam/Philadelphia: John Benjamins Publishing Company.
Riccardi, A. (2005). On the evolution of interpreting strategies in simultaneous interpreting. Meta: Translators’ Journal, 50 (2), 753-767.
Schäffner, C. (1997). Strategies of translating political texts. In A. Trosborg (Ed.), Text Typology and Translation (pp. 119-143). Amsterdam/Philadelphia: John Benjamins Publishing Company.
Setton, R. (1999). Simultaneous interpretation. A cognitive-pragmatic analysis. Amsterdam/Philadelphia: John Benjamins Publishing Company.
Shlesinger, M. (1995). Shifts in Cohesion in Simultaneous Interpreting. The Translator, 1 (2), 193-214.
Snelling, D. (1999). Introduzione alla simultanea verso la lingua straniera. In C. Falbo, M. Russo & F. Straniero Sergio (Eds.), Interpretazione simultanea e consecutiva. Problemi teorici e metodologie didattiche (pp. 202-206). Milano: Hoepli.
Straniero Sergio, F. (2003). Norms and quality in media interpreting: the case of Formula One press conferences. The Interpreters’ Newsletter, 12, 135-174.
Viaggio, S. (2002). The quest for optimal relevance. The need to equip students with a pragmatic compass. In G. Garzone & M. Viezzi (Eds.), Interpreting in the 21st Century. Challenges and Opportunities (pp. 229-244). Amsterdam/Philadelphia: John Benjamins Publishing Company.
Viezzi, M. (1999). Aspetti della qualità in interpretazione. In C. Falbo, M. Russo & F. Straniero Sergio (Eds.), Interpretazione simultanea e consecutiva. Problemi teorici e metodologie didattiche (pp. 140-151). Milano: Hoepli.
Viezzi, M. (2013). Simultaneous and consecutive interpreting (non-conference settings). In C. Millán & F. Bartrina (Eds.), The Routledge Handbook of Translation Studies (pp. 377-388). Abingdon/New York: Routledge.
Walton, D. (2004). Relevance in Argumentation. Mahwah (NJ): Lawrence Erlbaum Associates.
Walton, D., Reed, C., & Macagno, F. (2008). Argumentation schemes. Cambridge: Cambridge University Press.
Zarefsky, D. (2009). Strategic maneuvering in political argumentation. In F. H. van Eemeren (Ed.), Examining Argumentation in Context. Fifteen Studies on Strategic Maneuvering (pp. 115-130). Amsterdam/Philadelphia: John Benjamins Publishing Company.

Abstract: Combining the ScaPoLine (Nølke, Fløttum, & Norén, 2004; Nølke, 2009, 2011, 2013) with the (extended) pragma-dialectical approach (van Eemeren & Grootendorst 1984; 2004; van Eemeren 2010), I suggest a reconstruction of judgement motivations as critical discussions between a plurality of voices conveyed even in one and the same sentence. In particular, I present some illustrative examples of polyphonic strategic manoeuvring from a landmark judgment of the Italian Constitutional Court: n. 440/1995.

Keywords: Critical discussion, Italian Constitutional Court, legal discourse, polyphony, Pragma-dialectics, ScaPoLine, strategic manoeuvring.

1. Introduction
When reading legal texts such as judgement motivations, one encounters a plurality of voices carrying different views on the issue at stake. This happens not only – quite unsurprisingly – at a textual level, but also at the micro-level of the utterance (cf. Nølke, 2009, p. 12). With Nølke (2009, p. 12), I focus here on “polyphonie en langue, conçue comme le produit des éléments de la langue susceptibles de favoriser une certaine lecture polyphonique de la parole.” I maintain that the use of such polyphony has an argumentative significance and that this can lead to reconstruct such apparently monological texts as fully fledged critical discussions permeated by the striving for rhetoric efficiency known in Pragma-dialectics as strategic manoeuvring (van Eemeren, 2010). In other words, I suggest focusing on argumentative polyphony in judicial motivations and looking at it from the angle of rhetoric efficiency, since, as van Eemeren (2010, p. 153) pointedly writes, it is the “ample room left for strategic maneuvering [that] is, in fact, the basis of the legal profession.”

The matters addressed in the present contribution are of methodological order and can be broken down into two questions:
* Can an integration of the pragma-dialectical and a polyphonic approach provide useful insights into argumentation analysis?
* Does polyphony account for strategic manoeuvring in judgement motivations?

The fundamental suggestion put forward is therefore the integration of two theoretical pillars: the (extended) Pragma-dialectics on the one hand (see among others van Eemeren & Grootendorst, 2004; 1992; 1984; van Eemeren, 2010), and the linguistic polyphonic approach known under the acronym ScaPoLine – which stands for Théorie SCAndinave de la POlyphonie LINguistique – on the other hand (see among others Nølke, Fløttum & Norén, 2004; Nølke, 2006; Nølke, 2013).

2. Theoretical framework
2.1 The first pillar: Extended Pragma-dialectics
I will not dwell largely upon the first pillar here, since it is the specialty of the Institution hosting the conference from which the present volume results. Only two aspects are to be briefly recalled to the reader’s mind: the ideal abstract model of a critical discussion, in which argumentation and standpoint are staged, and the rhetorical component present in argumentation.

The former is articulated in four stages: a confrontation stage, where protagonists put forward a standpoint while antagonists cast doubt upon it, thus establishing a difference of opinion; an opening stage, where the common ground of the parties is established; the actual argumentation stage, where arguments are advanced in support of a standpoint; finally, the concluding stage, where the difference of opinion is either overcome or maintained (van Eemeren & Grootendorst, 2004, pp. 57-68). The four stages take place when a discussion about a difference of opinion begins; it is however important to keep in mind that Pragma-dialectics acknowledges the complexity of real life interactions by making clear that the logical order pictured above seldom coincides with the chronological one in a discussion, and that some stages of a critical discussion often take place implicitly. This is for example typical of the opening stage, which can mostly be elicited by the fact that a protagonist holding a standpoint directly proceeds to argue for it – and were they not to, it wouldn’t strike anyone as surprising if they were challenged to do so.

On its way towards a resolution of a difference of opinion, the critical discussion thus described is invariably carried by both a dialectical and a rhetorical component at every single stage. While the former component aims at reasonability, the latter strives for effectiveness. Extended Pragma-dialectics tackles the matter by christening this component strategic manoeuvring (van Eemeren, 2010) and pointing out its three simultaneously present aspects: presentational devices, topical potential and audience demand.

2.2 The second pillar: ScaPoLine
The second theoretical pillar is represented by a linguistic theory of polyphony developed by a French-speaking group of Scandinavian Romanists around Nølke and indebted to Ducrot’s linguistic approach. The Scandinavian Theory of Linguistic Polyphony deals with the plurality of points of view, abbreviated with POV in English (Nølke, 2006) communicated through an utterance. This theory is an utterance oriented, semantic, discoursive, instructional and structuralistic theory originally inspired by the Ducrotian approach (1984a; 1984b), which it aims at formalizing in order to “préciser les contraintes proprement linguistiques qui régissent l’interprétation polyphonique” [specify the strictly linguistic constraints governing the polyphonic interpretation] (Nølke, 2009, p. 15).

A certain language parochialism has likely prevented the Scandinavian Theory – not unlike its Ducrotian precedents – from expanding far beyond the French-speaking field of Romance studies. Such borders have only just begun to be removed by sporadic non-French publications: in “The semantics of polyphony (and the pragmatics of realization)” (2006), Nølke introduces the English-speaking readership to the theory, while in “Types of Discourse Entities in ScaPoLine” (Nølke, 2011, p. 58), he specialises in the “images of the ‘persons’ who are created by the speaker and the ‘persons’ who inhabit the discourse”. Dendale (2006, 2007) contributes to the propagation of the theory to the English-speaking audience by presenting and confronting it with other polyphony frameworks, namely Ducrot’s (1984a; 1984b), Bres’s (1998; 1999) and Kronning’s (1996). For other languages, one might refer to Gévaudan (2008), who explains Ducrot’s and the ScaPoLine’s approaches to polyphony on the basis of German examples. If the theoretical framework of ScaPoLine deserves a broader consideration, so does its application to different natural languages, such as Italian. Considering the relatively few non-French papers on the subject, it will not be superfluous to give a brief account of basic concepts of the ScaPoLine in this contribution as well.

The ScaPoLine theory distinguishes first of all between polyphonic structure, which deals with linguistic coding, and polyphonic configuration, which has to do with utterance meaning. From a logical perspective, the polyphonic structure precedes the configuration, since it is composed of instructions for the configuration and thus yields semantic constraints on the interpretation. But to gain insight into the structure, the starting point cannot but be the configuration (cf. Nølke, 2006, p. 145).

The polyphonic configuration is to be attributed to an entity named locutor as constructor (LOC) for its property of presenting the elements composing the polyphonic configuration. These are: LOC as a constructor itself as well as any copy of the locutor as a discourse entity, namely the locutor “as a virtual source of a point of view” (Nølke, 2006, p. 148), also called utterance locutor (cf. Nølke, 2009, p. 23); the points of view (POVs); the discourse entities (DE) and the utterance links[i] (cf. Nølke, 2013, p. 26). It is possible for the elements of the configuration to be coded in the linguistic form and therefore be part of the polyphonic structure, but this is not necessarily the case.

The POVs (cf. Nølke, 2013, pp. 32–33) are semantic units constituted by a source[ii] X, instantiated by a discourse entity, and a judgment upon a content p, which might here tentatively be qualified as of facts or actions. The POV form is expressed as

[X] (JUGE (p))

where the judgement, lacking specific indicators to the contrary, is by default one of truth. The POVs can be either simple or complex, in which case they will be either relational – as in a typical argumentative link, where a POVARG is put forward in support of a POVSTP – or hierarchical, when the judgement is made upon one or more different POVs.

The DE (cf. Nølke, 2006, pp. 147, 149-150; 2013, pp. 26-32) are semantic entities that can be held responsible for the points of view. They are constructed images of the discourse referents and relate to the LOC as string puppets to their master, to use Nølkes efficient metaphor (cf. 2009, p. 23). In ScaPoLine special attention is paid to the speaker’s role, whose images can be distinguished as the following basic DE:
* the textual locutor (L), i.e. “the source of a POV that the speaker had prior to [the] utterance act, and which” is still held (Nølke, 2006, p. 155); L can be constructed by LOC as a L at another point in time (cf. Nølke, 2013, p. 27);
* the utterance locutor (l0), i.e. the source of a POV which is held hic et nunc in the utterance[iii]; l0 exists only in the present utterance (cf. Nølke, 2013, p. 27);
* the locutor of the utterance (lt), i.e. the source of a POV held at the moment of the utterance construction and who is, in fact, an l0 at a different point in time (cf. Nølke, 2013, p. 28).

In addition, it is useful to present the represented locutor (RL), a discourse entity introduced to explain reported speech (cf. Nølke, 2013, p. 52), which is a type of what is known as external polyphony because of the presence of DE different from the locutor’s images (cf. Nølke, 2013, p. 36)[iv].

Besides the speaker’s POVs, ScaPoLine also takes into consideration POVs of the addressee (text addressee [A] and utterance addressee [at]), and of thirds. The latter can be individuals – either textual thirds [T] or utterance thirds [τt] – as well as collective entities such as the LAW or an impersonal voice named after the French indefinite pronoun ON (cf. Nølke, 2013, pp. 30-32)[v].
The utterance links (cf. Nølke, 2013, pp. 33–35) finally connect the discourse entities to the points of view. They can be of responsibility, as in an unquestioned statement, or of non-responsibility, in which case they will be either of refutation, as in a negation, or of non-refutation, as in indirect speech.
As far as the polyphonic structure is concerned, I shall confine myself to reporting two principles that apply to it (cf. Nølke, 2006, p. 152): on the one hand, the polyphonic structure necessarily contains at least one simple POV; on the other hand, the link between locutor and at least one POV is of responsibility.

2.3 Argumentative acts between Pragma-dialectics and ScaPoLine
Starting from the pragma-dialectical understanding of argumentation as a communicative and interactional complex speech act linked to the (complex) speech act of a standpoint it means to defend (cf. van Eemeren & Grootendorst, 2004; 1984), argumentative acts stricto sensu are here understood as relational POVs linking standpoint-POVs (=POVSTP) to argument-POVs (=POVARG)[vi]. The link between the two is part of LOC’s construction and is therefore to be traced back to LOC even if it can apparently be attributed to another locutor’s image: in fact, even this image is LOC’s creation. In other words, since LOC decides what elements of the polyphonic configuration to stage and in what way, it is LOC who is held responsible for the utterance and any argumentative acts occurring through it (cf. Nølke, 2013, p. 34).

3. Illustrative analysis
I now suggest an intertwinement of Pragma-dialectics and ScaPoLine by using some examples from a judgement of the Italian Constitutional Court, specifically number 440/1995, regarding the constitutional legitimacy of art. 724, clause 1, it. Poenal Code, on blasphemy (it. bestemmia)[vii]. This judgement was a milestone in the development of religious discourse in Italy, as it meant a shift in the jurisprudence and argumentation of the Court, resulting in the abolishment of the special treatment reserved for Catholicism in the punishment of blasphemy. For its intervention in the law, the aforesaid judgement is regarded in the legal community as a manipulative one (cf. Casuscelli, 2005, p. 4).

The constitutionality issue was raised by the court of Milan. In the motivation of judgement 440/1995, the final question the Constitutional Court is confronted with is: Should the norm of article 724, clause 1 it. Poenal Code (religious blasphemy) be declared unconstitutional? The outcome will of course depend on the answer to the question: Is the norm constitutionally legitimate? The answer to the latter in turn will be decided by the court’s position on two possible arguments for constitutional illegitimacy – which can be formulated as follows: Is the norm indeterminate according to art. 25 of the Italian Constitution[viii]? and Is the norm discriminatory according to art. 3 and art. 8 of the Italian Constitution[ix]? In the argumentation supporting the answer to the latter, the deciding court raises a further matter, namely: What is the object of legal protection of the norm?

Concentrating on the judgement part concerning constitutional indetermination, i.e. chronologically the first matter the court seeks to solve, I will now introduce some examples to show how, through the identification of the polyphonic configuration, it is possible to reconstruct the chosen argumentative extracts as critical discussions and identify polyphonic means specific to single natural languages – in this case, Italian.

To this end, I name the various POVs relevant for the reconstruction of the critical discussion after its four stages. Therefore, the standpoints from which the discussion starts in the confrontation stage will be POVCONF(i) and POVCONF(ii). The various speech acts that can take place in the opening stage are abstractly represented as POVOP. The arguments in support of POVCONF(i) and POVCONF(ii) shall be POVARG(i) and POVARG(ii). If an argument becomes, in turn, a standpoint, it shall be marked as such as well (POVARG/STP). Different numeration systems are given to different argumentation structures (see van Eemeren & Grootendorst, 1992, pp. 86-92): 1, 2, 3… for subordinate argumentation, i.e. when argumentation is made of various arguments supporting each other and finally the standpoint; i, ii, iii… for multiple and coordinate argumentation, i.e. respectively when multiple arguments support the standpoint independently from each other and when they function jointly. The latter joint function is signalled by an ampersand. The conclusive speech act will be named POVCONC. The integrated model proposed can be outlined as in Tab. 1:

The judgement upon the propositional content of a POV shall be given in the analytical tables of the following paragraphs through the explicitly verbalised markers found in the text. Otherwise, a judgement of truth is to be assumed, according to the ScaPoLine principles.

3.1 Confrontation Stage coded through REPORTED-DUBITATIVE SPEECH
n the first example, the confrontation stage of a critical discussion is rendered through the polyphonic use of reported-dubitative speech. On a macrolevel, example (1) stems from the confrontation stage in which the court of Milan puts forward a standpoint as to the indetermination of the contested measure.[x] This standpoint of the court of Milan is outlined here as having been said (cf. example (1): si sostiene che) by a RL: Inside this passage of reported speech, the beginning of a critical discussion is staged in the italicized utterance in example (1) around the question if the alleged indetermination stems from specific arguments:

1. Si sostiene […] che, poiché la norma impugnata sanziona […] chi pubblicamente “bestemmia […] contro la Divinità o i Simboli o le Persone venerati nella religione dello Stato”, e poiché il Protocollo addizionale dell’Accordo di modifica del Concordato lateranense […] prevede testualmente il venir meno della religione cattolica come sola religione dello Stato italiano, ne conseguirebbe […] la indeterminatezza della fattispecie penale.

[It is maintained that, since the contested norm sanctions the person who “utters blasphemy against the divinity or the Symbols or Persons revered in the State religion”, and since the Supplementary Protocol to the Modifications Agreement of the Lateran Concordat provides verbatim that the catholic religion as sole religion of the Italian State be abolished, it would follow that the legal paradigm is indeterminate.]

RL puts forward a complex standpoint, constituted by an argumentative POV and marked as such both by the fact that it is subsequently argued for and by the introductory verbum putandi sostenere (en. to maintain), while L questions it. This confrontation stage is carried – besides by the assertive verb sostenere, which concerns POVCONF(i) – by the condizionale mood in its present tense, which has here a twofold function:

* on the one hand, as condizionale riportivo, it reports a protagonist voice maintaining that certain arguments follow a certain standpoint (POVCONF(i)) and
* on the other hand, it casts doubt upon (and challenges) POVCONF(i) through an ideal antagonist (condizionale dubitativo) (cf. Patota, 2006, p. 116).

As can be clearly seen in Table 2, we have a complex polyphonic structure in which three POVs (POVSTP1, POVARG(i)(i) and (POVARG(i)(ii)) that form a coordinate compound argumentation are related and, in turn, are subject to the judgement of a different source: L responds to the POV of RL, thus setting off to become the antagonist in the subsequent argumentation stage: it is noteworthy that L does not question the arguments in defence of the first standpoint POVSTP1 (the legal paradigm is indeterminate), which arguments the protagonist RL has already given in POVARG1(i) (blasphemy against the State religion is punished) & POVARG1(ii) (the State religion was abolished). It is rather the soundness of the relation between arguments and standpoint, which constitutes a new, hierarchical, standpoint (POVCONF(i)), that is being challenged in POVCONF(ii). So in one sentence a potential critical discussion is begun, and then interrupted by another one. The difference of opinion is thus shifted from the macro-question: Is the norm indeterminate? to a sub-question: Does the indetermination of the norm follow from the given arguments (which per se constitute common starting points between protagonist and antagonist)?

These POVs can be distributed following the critical discussion scheme of Pragma-dialectics as in Table 2, in which the role of the condizionale is reported as well:

The condizionale is a typical means for LOC to stage the confrontation stage of a critical discussion in a monological text, because it implies neither a refutation nor a responsibility link, but only a non-refutation utterance link between the locutor and the POV attributed to a third, which is reported. Thus, the reader expects an argumentation either for or against a standpoint, and there is then room either for the acceptance or the refutation of said standpoint.

3.2 Confrontation & Concluding Stage through NEGATIVE + DUBITATIVE SPEECH
Example (2) gives an instance of a compound confrontation and concluding stage in the same utterance. The critical discussion revolves here around the matter expressed in the final standpoint, placed at the very beginning of the utterance and once more constituted in turn by a standpoint and arguments subordinately linked to it. For concision, we shall focus only on the marked part of the extract and operate under the assumption that the source of the POVCONF(i) and of POVCONC is L.

2. Né la censura potrebbe superarsi ritenendo che la norma denunciata continui a riguardare la religione cattolica come confessione religiosa più diffusa del Paese – mutuando l’espressione dalla sentenza n. 14 del 1973 della Corte costituzionale – poiché non verrebbe ora in discussione la ratio della norma incriminatrice, bensì la sua (sopravvenuta) incompatibilità con il principio di tassatività.

[Neither could the censure be overcome considering the contested norm as still regarding the catholic religion as the most widespread religion of the country – borrowing the expression from the judgement n. 14 of 1973 by the Constitutional Court – for it would not be the ratio of the incriminating norm that is in question, but rather its incompatibility with the taxativity principle.]

Again, we have the condizionale presente that opens the res dubia carrying two POVs (POVCONF(i) and POVCONF(ii)) and introduces to the confrontation stage. RL questions POVCONF(i) in POVCONF(ii) through the condizionale dubitativo. The argumentation stage follows, signalled by poiché[xii]. This leads to the outcome of the critical discussion, reached in the concluding stage: the refutation of POVCONF(i), which is anticipated at the very beginning of the sequence condensed in the negative particle né. It is noteworthy that negation implies a refutation, a non-responsibility link between POVCONF(i) and the discourse entity linked to it, but that the entire passage of example (2) is actually reported speech: through reported speech a non-refutation link is built between LOC’s image (which can be traced back to the extralinguistic institution of the Constitutional Court) and the whole argumentation. This supports the hypothesis that the whole critical discussions are included in the long motivation of the judgment to slowly lead the audience towards acquaintance with and acceptance of the final standpoint. The mentioned stages of the critical discussion can be reconstructed as in Table 3:

The negation is strategically placed at the beginning of the sequence and right before POVSTP1, thus orienting the audience and anticipating the outcome of the macro-discussion – in the end the censure cannot, indeed, be overcome, as the Constitutional Court will decide: the norm is (if only partly) unconstitutional. Of course, a concluding stage stricto sensu cannot take place in a monological text, but this is in fact also part of the strategic manoeuvring: the arguments are staged as if an actual discussion was taking place, where the interlocutor can explicitly accept or refute a standpoint in the end.

3.4 Argumentation Stage through QUOTED SPEECH
In example (3) the argumentation stage is conveyed by a quotation. The voice quoted is that of the Constitutional Court at another point in time:

3. A sostegno della censura, nell’ordinanza si riportano brani di precedenti pronunce di questa Corte che sono consistiti in espressi inviti al legislatore, non ancora accolti, per una revisione della disciplina in vista dell’attuazione del principio costituzionale della libertà di religione ), dal momento che “la limitazione della previsione legislativa alle offese contro la religione cattolica non può continuare a giustificarsi con l’appartenenza ad essa della quasi totalità dei cittadini italiani”.

[In support of the censure, the order reports passages of former rulings by this Court, which consisted in explicit requests addressed to the legislator to revision the discipline […], since “the limitation of the legal prevision to the offences against catholic religion can not continue to be justified with the fact that virtually the entirety of the Italian citizens is religiously affiliated to it”.]

In POVCONF(i) it is maintained by a RL that the norm must be censured. POVCONF(i) is not explicitly challenged, but reveals itself as the final standpoint of the passage at the beginning of the argumentation stage, which involves arguments for an implicit POVCONF(ii) of another RL maintaining the contrary of POVCONF(i). In POVSTP1/ARG(i)1 the argument for POVCONF(i) is that the legislator should revision the discipline; at the same time, this is a standpoint supported by a subordinate argument: the complex POV STP1/ARG(i)2, according to which the argumentative POVARG(ii)1 justifying the limitation of the legal prevision to the offences against Catholicism (POVSTP1/ARG(ii)1) with the well-known (ON) affiliation to it of the majority of Italians (POVARG(ii)2) and altogether supporting the implicit POVCONF(ii) (the norm must not be censured) is not justified.

Quoted reported speech is used here for accepting and not for refuting a standpoint, as is attested to by the fact that it is syntactically integrated in the speech of the hierarchically superior locutor, without inquit, which implies sticking to the epistemically assertive indicative. Interestingly enough, the quoted utterance is presented here as an argument for the critical discussion attributed to the Court of Milan, whose standpoint will be refuted, as seen. But it is also a decisive argument for the final declaration of unconstitutionality and it is strategically already reported at the beginning of the judgment, functioning as a material starting point. Table 4 can serve as a reconstruction of the critical discussion stage:

Direct speech, which implies the construction of the locutor’s representation with all its locutor’s properties, as a mimed LOC (cf. Nølke, 2013, p. 56), is only used for two discourse entities in the analysed judgement: the third-person-DE LAW (which in this judicial text is to be taken as a stricto sensu reference; e.g. POVSTP1/ARG(ii)1) and the images of the locutor at a given moment in the past. These voices are thus integrated in the utterance supporting the point of view of the utterance locutor. This strategy is applied throughout the judgement and it is particularly evident how it is meant to support the final decision of unconstitutionality when the DE involved is an image of the same extralinguistic subject instantiating the decision of judgement 440/1995: in fact, the new decision is staged as not so new after all, given that the voices of the same referent in the past back it up.

4. Concluding
In summary, I believe that the integration of the polyphonic approach into the pragma-dialectical can enrich the latter with an analysis apparatus that allows going beyond the universal perspective inherent to the pragma-dialectical approach. The polyphonic theory suggested here can in fact help to identify antagonistic voices coded in one and the same utterance as well as stages of an ideal critical discussion coexisting in one and the same utterance – and this while showing the specific linguistic means responsible for such phenomena, which may differ in various natural languages. Moreover, a systematic application of the ScaPoLine to a discursive tradition can highlight the patterns of strategic manoeuvring polyphony used: in the judgement taken into consideration in the present paper, polyphony could be identified as a manoeuvring by means of presentational devices, but other aspects of strategic manoeuvring are still to be taken into consideration through further research.

NOTES
i. In Nølke (2006), unlike in Nølke (2011), more confusingly for the English-speaking reader, enunciative links.
ii. The sources are variables corresponding to the utterers (énonciateurs) of Anscombre and Ducrot: cf. among others Nølke (2011, p. 64).
iii. “The utterance locutor is always responsible for the highest POV in a hierarchical point of view structure. This is why sentence adverbials, for instance are the utterance locutor’s responsibility.[…] An analogous difference between the t-locutor and the utterance locutor is that while the POVs of the latter may be shown (in Wittgenstein’s sense) […], those of the former can only be said (or narrated).” (Nolke 2006, p. 155) Following Wittgenstein’s distinction (1969, §§ 4.022 & fol.), this amounts to saying that the POVs of the lt can always be considered in terms of truth, but not the POVs that are merely shown: as such, they cannot be subject to discussion.
iv. As opposed to external polyphony, internal polyphony takes place when an utterance conveys both the POV of L and the POV of locutor0.
v. ON is rendered in Nølke (2006, p. 156) as VOX PUBLICA, in Nølke (2011, p. 66) as ONE. In my opinion, it is better left untranslated, as in Dendale (2006, p. 13), due to the useful semantic ambiguity of the French pronoun, which can be translated in English into both one and they.
vi. This is not to say that a simple standpoint cannot be endowed with argumentativity in a Ducrotian sense – but to deepen this matter here would go beyond the scope of this contribution.
vii. The norm text of art. 724, clause 1, it. Poenal Code reads: “Chiunque pubblicamente bestemmia, con invettive o parole oltraggiose, contro la divinità o i simboli o le persone venerati nella religione dello Stato, è punito con la sanzione amministrativa pecuniaria da euro 51 a euro 309.” [The person who publicly utters blasphemy against the Divinity or Symbols or Persons revered in the state religion shall be punished by a financial administrative sanction of 51 and up to 309 €.]
viii. Art. 25 co. 2 Costituzione della Repubblica italiana: “Nessuno può essere punito se non in forza di una legge che sia entrata in vigore prima del fatto commesso.” [No punishment may be inflicted except by virtue of a law in force at the time the offence was committed.]
ix. Art. 3 co. 1 Costituzione: “Tutti i cittadini hanno pari dignità sociale e sono eguali davanti alla legge, senza distinzione di sesso, di razza, di lingua, di religione, di opinioni politiche, di condizioni sociali e personali.” [All citizens have equal social dignity and are equal before the law, without distinction of sex, race, language, religion, political opinion, personal and social conditions.]
Art. 8 co. 1 Costituzione: “Tutte le confessioni religiose sono egualmente libere davanti alla legge.” [All religious denominations are equally free before the law.]
x. It is to note that, although the standpoint the norm is indeterminate will ultimately be refuted by the Constitutional Court, the argumentation in its support is given in great detail. Since the LOC holds the power to present the referents by constructing the images at will, there must be a reason why LOC indulges in this long construction of a complex critical discussion even though in the end it doesn’t revolve around the question that will be decisive in the constitutionality matter (i.e. its discriminatory nature). One of the reasons can be found in a long discursive tradition of judicial texts according to which it is first the arguments and standpoints that will not be accepted that are put forward, only eventually followed by the voice that “set things right”. The staging of critical discussions between POVs through reported speech (which means a partial, selective construction of other voices than the speaker’s) entails also the possibility to attribute wrong argumentation to other sources than the image of the LOC’s self. On the other hand, the detailed argumentation attributed to different sources in defence of the standpoint the norm is indeterminate, which is theoretically accepted as argument for the standpoint the norm is unconstitutional, means that the reader has the time to get used to the final standpoint of the unconstitutionality of the norm.
xi. In this operative translation, I have kept as close as possible to the Italian original – even at the cost of the English grammaticality, if it helped to render the polyphonic means used in Italian.
xii. It is noteworthy that the internal argumentation of POVCONF1, namely POVARG1,1/STP2 and POVARG1,2, are implicitly accepted by the parties as arguments for POVSTP1. Insofar this acceptance is part of the negotiation of the common starting points in the opening stage.

References
Brès, J. (1998). Entendre des voix: De quelques marqueurs dialogiques en français. In J. Brès, R. Delamotte-Legrand, F. Madray-Lesigne & P. Siblot (Eds.), L’autre en discours (pp. 191-212). Montpellier: Université Paul-Valéry.
Brès, J. (1999). Vous les entendez? Analyse du discours et dialogisme. Modèles linguistiques, 20, 71-86.
Casuscelli, G. (2005). L’evoluzione della giurisprudenza costituzionale in materia di vilipendio della religione. Quaderni di Diritto e Politica Ecclesiastica, 3, 1119–1142.
Constitution of the Italian Republic. Retrieved from http://www.senato.it/documenti/repository/istituzione/costituzione_inglese.pdf
Costituzione della Repubblica italiana. Retrieved from http://www.quirinale.it/qrnw/statico/costituzione/pdf/costituzione.pdf
Dendale, P. (2006). Three linguistic theories of polyphony / dialogism: An external point of view and comparison. In N. Møller Andersen, H. Nølke & R. Therkelsen (Eds.), Sproglig polyfoni (pp. 3-32). Roskilde: Institut for Sprog og Kultur, Roskilde Universitetscenter.
Dendale, P. (2007). A critical survey and comparison of French and Scandinavian frameworks for the description of linguistic polyphony and dialogism. In R. Therkelsen, N. Møller Andersen & H. Nølke (Eds.), Sproglig polyfoni. Tekster om Bachtin og ScaPoLine (pp. 109–143). Århus: Aarhus Universitetsforlag.
Ducrot, O. (1984a). Le dire et le dit. Propositions. Paris: Minuit.
Ducrot, O. (1984b). Polyphonie. Lalies. Langue et littérature, 4, 3-30.
Eemeren, F. H. van, & Grootendorst, R. (1992). Argumentation, communication, and fallacies: A pragma-dialectical perspective. Hillsdale, N.J: L. Erlbaum.
Eemeren, F. H. van, & Grootendorst, R. (2004). A systematic theory of argumentation: The pragma-dialectical approach. New York: Cambridge University Press.
Eemeren, F. H. van. (2010). Strategic maneuvering in argumentative discourse: Extending the pragma-dialectical theory of argumentation. Amsterdam, Philadelphia: John Benjamins Pub.
Eemeren, Frans H. van, & Grootendorst, R. (1984). Speech acts in argumentative discussions. Dordrecht: Foris.
Gévaudan, P. (2008). Das kleine Einmaleins der linguistischen Polyphonie. PhiN. Philologie im Netz, 43, 1-10. Retrieved from http://web.fu-berlin.de/phin/phin43/p43t1.htm
Kronning, H. (1996). Modalité, cognition et polysémie: Sémantique du verbe modal devoir. Uppsala: [Uppsala Universitet].
Nølke, H. (2006). The semantics of polyphony (and the pragmatics of realization). Acta Linguistica Hafniensia, 38, 137-160.
Nølke, H. (2009). La polyphonie de la ScaPoLine 2008. In A. Kratschmer & R. T. Birkelund (Eds.), Romanistik (Berlin, Germany): Bd. 3. La polyphonie. Outil heuristique linguistique, litteraire et culturel (pp. 11-40). Berlin: Frank & Timme.
Nølke, H. (2011). Types of Discourse Entities in ScaPoLine. Lingusitic Polyphony, 197, 58-77.
Nølke, H. (2013). La polyphonie linguistique. Lalies. Langue et littérature, 33, 7-76.
Nølke, H., Fløttum, K., & Norén, C. (2004). Scapoline: La théorie scandinave de la polyphonie linguistique. Paris: Kimé.
Patota, G. (2006). Grammatica di riferimento dell’italiano contemporaneo. Novara: Garzanti linguistica.
Wittgenstein, L. (1969). Tractatus logico-philosophicus. Frankfurt am Main: Suhrkamp.