Spaces of Polyphony [1 ed.] 2012016822, 9789027210326, 9789027273581

Citation preview

Spaces of Polyphony

Dialogue Studies (DS) Dialogue Studies takes the notion of dialogicity as central; it encompasses every type of language use, workaday, institutional and literary. By covering the whole range of language use, the growing field of dialogue studies comes close to pragmatics and studies in discourse or conversation. The concept of dialogicity, however, provides a clear methodological profile. The series aims to cross disciplinary boundaries and considers a genuinely inter-disciplinary approach necessary for addressing the complex phenomenon of dialogic language use. This peer reviewed series will include monographs, thematic collections of articles, and textbooks in the relevant areas. For an overview of all books published in this series, please see http://benjamins.com/catalog/ds

Editor

Assistant Editor

Edda Weigand

Sebastian Feller

University of Münster

A*STAR - Institute of High Performance Computing, Singapore

Editorial Advisory Board Adelino Cattani

Marion Grein

Masayoshi Shibatani

Kenneth N. Cissna

Fritjof Haft

Talbot J. Taylor

Světla Čmejrková

John E. Joseph

Wolfgang Teubert

François Cooren

Werner Kallmeyer

Linda R. Waugh

Robert T. Craig

Catherine Kerbrat-Orecchioni

Elda Weizman

Stefanie Molthagen-Schnöring

Yorick Wilks

Università di Padova University of South Florida Czech Language Institute Université de Montréal University of Colorado at Boulder

Marcelo Dascal

University of Mainz University of Tübingen University of Edinburgh University of Mannheim Université Lyon 2

Tel Aviv University

Hochschule für Technik und Wirtschaft Berlin

Valeri Demiankov

Geoffrey Sampson

Russian Academy of Sciences

Rice University

College of William and Mary University of Birmingham University of Arizona Bar Ilan University University of Sheffield

University of Sussex

Volume 15 Spaces of Polyphony Edited by Clara-Ubaldina Lorda and Patrick Zabalbeascoa

Spaces of Polyphony Edited by

Clara-Ubaldina Lorda Patrick Zabalbeascoa Universitat Pompeu Fabra

John Benjamins Publishing Company Amsterdam / Philadelphia

8

TM

The paper used in this publication meets the minimum requirements of the American National Standard for Information Sciences – Permanence of Paper for Printed Library Materials, ansi z39.48-1984.

Library of Congress Cataloging-in-Publication Data Spaces of polyphony / edited by Clara-Ubaldina Lorda, Patrick Zabalbeascoa. p. cm. (Dialogue Studies, issn 1875-1792 ; v. 15) Includes bibliographical references and index. 1. Dialogue analysis. 2. Dialogism (Literary analysis) 3. Diglossia (Linguistics) 4. Intercultural communication. I. Lorda Mur, Clara Ubaldina, 1947- II. Zabalbeascoa Terran, Patrick. P40.5.D53S73   2012 306.44--dc23 2012016822 isbn 978 90 272 1032 6 (Hb ; alk. paper) isbn 978 90 272 7358 1 (Eb)

© 2012 – John Benjamins B.V. No part of this book may be reproduced in any form, by print, photoprint, microfilm, or any other means, without written permission from the publisher. John Benjamins Publishing Co. · P.O. Box 36224 · 1020 me Amsterdam · The Netherlands John Benjamins North America · P.O. Box 27519 · Philadelphia pa 19118-0519 · usa

Table of contents Introduction Clara-Ubaldina Lorda and Patrick Zabalbeascoa

1

part 1.  Strategies in daily conversations chapter 1 Strategy and creativity in dialogue Robert E. Sanders

11

chapter 2 Conversational irony: Evaluating complaints Maria Christodoulidou

25

chapter 3 Speaking through other voices: Conversational humour as a polyphonic phenomenon Béatrice Priego-Valverde

43

part 2.  Plural identities and viewpoints in acquisition and language learning chapter 4 The self as other: Self words and pronominal reversals in language acquisition Aliyah Morgenstern chapter 5 The function of formulations in polyphonic dialogues Claudio Baraldi chapter 6 Observing the paradox: Interrogative-negative questions as cues for a monophonic promotion of polyphony in educational practices Federico Farini

57

73

87



Spaces of Polyphony

chapter 7 Co-construction of identity in the Spanish heritage language classroom Rachel Showstack

101

part 3.  The play of voices in mass media and politics chapter 8 Polyphonic strategies used in polemical dialogue Daciana Vlad chapter 9 Metacommunication and intertextuality in British and Russian parliamentary answers Maria Sivenkova

117

129

chapter 10 The role of prosody in a Czech talk-show Martin Havlík

143

chapter 11 Intertextuality as a means of positioning in a talk-show Světla Čmejrková and Jana Hoffmannová

161

part 4.  Social and cultural polyphony and intertextuality chapter 12 Rumour in the present Romanian press: Aspects of knowledge sources and their linguistic markers Margareta Manu Magda

175

chapter 13 Peritextual dialogue in the dynamics of poetry translatability Ana Ene

189

chapter 14 Voices through time in Meso-American textiles Yosi Anaya

205



Table of contents 

part 5.  Dialogism in literary discourse chapter 15 “Finn Mac Cool in his mind was wrestling with his people”: Polyphonic dialogues in Flann O’Brien’s comic writing Flore Coulouma

225

chapter 16 Dialogization, ontology, metadiscourse Emilia Parpală

237

chapter 17 Ironic palimpsests in the Romanian poetry of the nineties Carmen Popescu

251

chapter 18 Polyphony in interior monologues Ilaria Riccioni and Andrzej Zuczkowski

265

General references Index

279 297

Introduction Clara-Ubaldina Lorda and Patrick Zabalbeascoa

This collective volume focuses on dialogism in dialogue, a topic which is, according to Bakhtin’s formulations from the 1930s, the basic reality of language, conceived as actual use, in which the other is the condition of discourse. About forty years later, Benveniste (1974: 79–88) showed how individuals turn language into discourse through the use of deictic elements and other enunciation marks, and stated that as soon as a speaker uses language, and turns to such formal elements, s/he establishes an interlocutor and underlines a relationship with a partner. Dialogue is, indeed, the universal and everlasting way in which language is used. Furthermore, beyond the alternative exchange of interlocutors in a face to face conversation, dialogism designates the complex and multifarious interrelations among words, utterances and points of view in discourse. Language exists on that creative borderzone or boundary between human consciousnesses, between self and other that constitutes the capacity of language to produce new meaning. (Morris 1994: 5)

The term polyphony is used here metaphorically, as in most linguistic approaches to the phenomenon we are studying. It was first proposed by Bakhtin in his study of Dostoyevsky’s novels (1984), to refer to different types of language, which share the same discursive space and where there is no hierarchy: i.e. narrators and heroes’ discourses interact as equals. Such was the meaning of the musical term used in the XVII and XVIII centuries: several voices are heard, they are independent, they imitate each other but they also answer each other. For Bakhtin, a particular case of polyphony is heteroglossia, which should refer to the clash of antagonistic forces through different languages, one whose manifestations is diglossia, or polyglossia; i.e. the simultaneous use of two or more national languages in a given society. Beyond these three or four terms, the chapters of this volume, exploring ‘spaces of polyphony’, aim to describe different forms of dialogism in various discourse genres from several scientific approaches and an array of points of view. Fittingly, then, this is a polyphonic book.



Clara-Ubaldina Lorda and Patrick Zabalbeascoa

Let’s not forget the multilayered meaning of the term ‘dialogism’. First of all, Bakhtin’s insight tells us that each word sounds and resounds with the various aspects of the multilingual conscience that surround it and each utterance is always a complex response to other utterances. These phenomena underlie the notion of interdiscourse, theorised by various scholars and particularly described by Authier­Revuz as interdiscursive dialogism. The non-coincidence of the discourse with itself (or its constitutive discursive heterogeneity) refers, as constitutive, to the fact that all speech cannot, incessantly, avoid being embedded, determined, divided and so forth by the reality of the external discursive space where it is produced, received, sent. (Authier-Revuz 2012: 23)1

This French scholar (1982: 112) recalls Bakhtin’s metaphor: only the mythical Adam, approaching a freshly created world, could have avoided the dialogical relationship with the already said and the words of the others. Secondly, dialogism also applies to dialogues, the ones that correspond to conversational genres where the speakers take turns, as studied by Conversational Analysis (CA). These dialogues are regulated by social norms, and depend on genre (Charaudeau 2009). In the face-to-face exchange, a range of different polyphonic phenomena can be observed. Beyond turn-taking exchanges, the utterances of one of the participants can be repeated, rejected, or discussed by the other participant as both interlocutors shape their speeches in anticipation of their addressee’s response. From a dialogical point of view, comprehension is (more or less) inter-­ comprehension, and meaning is the result of a strategic and asymmetric co-­ construction of the participants. This is why Authier-Revuz (1982: 119) brings up Lacan: “The style is.... the addressee.” Furthermore, a speaker can point out someone else’s words in her/his utterance by the formal means of reported speech, which can vary depending on the amount of shifting introduced by the speaker who uses these words. Today these means are categorised with different linguistic labels (quotation, indirect discourse, free direct/­ indirect discourse, represented or evoked discourse). Bakhtin, under the pen name of Volosinov (1986), had already divided these different ways of reporting the words of others into two main expressive categories according to style: linear and pictorial. Reporting the words means to assign a place to the other, and, in doing so, following Authier-Revuz, the speaker subject tries to create the illusion of being the creator of her/his own words. Embedded in the impenetrable strangeness of one’s own words, when marking explicitly the heterogeneities in one’s discourse, the speaking subject circumscribes the other in those words. In this way, s/he affirms that the other is not everywhere. (1982: 145)2 1.

Our translation

2. Our translation



Introduction

These two main manifestations of dialogism are connected. Double dialogism implies at least a double parameter in the analysis: the dialogue between the addresser and the addressee, plus the dialogue with what has already been uttered. Finally, dialogism also reflects the internal division of the subject. Like Bakhtin, Benveniste states that a “monologue” is an interiorised dialogue, formulated as “inner language” between an “I who speaks and an I who listens.” (1974: 85, our translation) The division of the subject is, in fact, a matter of degree, ranging from the inner dialogue all the way to an unconscious fragmentation of the self, as described by Lacan (1966), in the slipstream of Freud’s reflections. In between these extremes, a host of possibilities exist in language for indicating the diversity of points of view, the nuances and the contradictions in the same speaking subject. Different linguistic approaches, mainly French (for instance, Ducrot 1984), have described these phenomena, starting from the distinction between the person actually speaking and the enunciator(s), not always one and the same, which explains how it is possible to express several points of view in a single utterance. The forms of dialogism that we have listed above appear in different ways, depending on the range of discourse genres, where they are materialised more or less consciously and more or less strategically. We have divided this volume into five Parts, covering five different genre areas. The first part of this volume is dedicated to polyphonic phenomena in everyday or casual conversation. This is the discursive area where polyphonic phenomena are less consciously applied; they are generated by the conversational competence that the speaking subjects acquire over time. Sanders, in Chapter one, shows that this competence is acquired at a very early age. He proposes an overview of the strategic mechanism regulating verbal interactions, illustrated by children’s conversations. Sanders reminds us of the anticipatory character of the utterances, assessed by Bakhtin again and again, as we have said above. The question, for Sanders, is how to reconcile the individual notion of ‘strategy’ with co-construction of polyphonic meaning in dialogue. In this respect, French discourse analysis has theorised that conversation implies an external social space where (at least) two participants exchange utterances while evaluating each other. Discourse developed by enunciators is the result of such an evaluation, namely the participants’ use of what freedom they have to develop their own strategies (Charaudeau 2009). Sanders suggests, through several samples from children’s conversations, that this natural property of discourse is developed very early in life. Chapters two and three in this part study specific polyphonic phenomena in casual conversation. Maria Christodoulidou analyses polyphony as a diversity of points of view transported by irony. She shows, on the one hand, that Goffmann’s frame could be doubled, giving rise to a new interpretation of the utterances and,





Clara-Ubaldina Lorda and Patrick Zabalbeascoa

on the other hand, how reported speech can be used for obtaining polyphonic effects in complaint story-telling. Béatrice Priego-Valverde’s object of study is humour in friendly and family conversations. Her focus is two-tiered; she looks at generalized interdiscourse, in which all the interactions take place, and also at the division of points of view, which can be emphasised in humoristic exchanges. Priego-Valverde, also takes into consideration other approaches to humour, and in doing so underlines its complexity, its ambivalence and the phenomenon of double-voicing. The context of the exchanges she analyses, involving people who are close, allows her to show a mixture of affective attitudes conveyed by humour. Among other issues, her study sheds light on linguistic markers which involve connecting speech turns by repeating a word, completely or partially, previously used by another participant or by allusion to other texts. The item may be repeated with a different meaning, producing something akin to a pun. Part 2 deals with language learning. Chapter four is devoted to the evolution of children and self-designation, from 18 to 36 months. Aliyah Morgenstern shows the slow setting-up of the different notions corresponding to the subject in young children and how they shift among several designations. Through her study of the pronominal reversals produced by two French children, and the functions associated to each form, Morgenstern shows their identity variations. Namely, following the situation of the speeches, they talk as another: they use third person for narratives; they use second person when they simulate an adult identity and, in doing so, they echo the words addressed to them; in this way, by trial and error, they progress towards the stabilisation of the I (je). The other three chapters in Part 2 explore three different educational environments. Chapter five by Claudio Baraldi, studies intercultural educational communities. The focus is on a specific aspect of conversation, the use of formulations in intercultural dialogues. He demonstrates the educational value of this strategy in promoting the joint construction of meanings; its functions vary according to the way in which the formulation interacts with previous turns or others yet to come; this can also appear as part of a series of formulations or be coordinated with speech acts. The use of formulations, concludes Baraldi, might contribute to harmonising values, obtaining shared knowledge and fulfilling affective expectations. In contrast, Chapter six reveals an educational misunderstanding. Federico Farini has developed his research within the context of several primary schools in Northern Italy. From the standpoint of the current trends in educational programs, he examines the use of an educator’s interrogative-negative questions. Farini observes that this kind of question involves certain constraints on the addressees, who are, at some point, obliged to answer affirmatively and, thereby accept the educator’s proposal. As Farini shows in his analysis, the pupils do not



Introduction

always respond accordingly and try to present their own point of view, which is in turn not always accepted by the teacher. Unfortunately, the attempt to change the actor’s roles in the educational premise sometimes becomes a mere simulation of polyphonic dialogue. In Chapter seven, Rachel Showstack also takes into consideration a teaching context, namely in US schools where phenomena of English-Spanish diglossia occur. She aims to present, from a critical approach, the negotiation of ideologies and identities, and the processes of legitimization and delegitimization through discourse in classrooms. In particular, Showstack explores how students use metalinguistic strategies which show different attitudes in “heritage speakers” regarding their position as bilinguals. English appears to have a symbolic power; whereas speaking (good) Spanish and having a hybrid identity might add value as well. In any case, bilinguals often challenge and contest any essentialisation of these different linguistic conditions. The contributors of Part 3 analyse characteristic television genres and/or political discourse. They study polyphonic strategies of three genres (electoral debate, parliamentary question-time sessions and talk-show), where the external dialogue (question-and-answer; turn taking) is intrinsically overlapped with interdiscursive context; as a result, other discourses and others’ discourses criss-cross with the ongoing interaction. Daciana Vlad studies an aspect of the televised debate in which Ségolène Royal and Nicolas Sarkozy debated their proposals before the 2007 presidential election. Vlad establishes two kinds of gradations with regard to polyphonic strategies: firstly, polyphony can be more or less dialogic; secondly, polyphony can be more or less conflictive. In the case of the debate under her analysis, polyphonic strategies are, indeed, dialogic and strongly conflictive. This character is determined by the contract of the genre (Charaudeau 2005). Both participants, who are supposedly to be on equal footing, debate their proposals and try to enhance their individual ethos by disparaging each other. Vlad shows that these disparaging attempts are developed through polyphonic strategies, mainly the echoing of words used by the opponent. These words are discussed, contradicted, ironically quoted or even manipulated, in order to obtain a victory over their rival, precisely by using the other’s own words. Their aim is to obtain the general public’s confidence and, ultimately, their votes. Maria Sivenkova, Chapter nine, explores political discourse as well, but in this case it is not mediated by television or other mass media. Sivenkova examines a sub-genre of parliamentary debates, from a comparative standpoint, in British and Russian parliaments. The overt goals of parliamentary discourses are to help legislate or to check the government. Sivenkova studies question-time control sessions, specifically,





Clara-Ubaldina Lorda and Patrick Zabalbeascoa

criticisms manifested by metadiscursive utterances. Her typology demonstrates the variety of forms and functions of such strategies, hereby MPs regulate their negotiations and their conflicts; their remarks refer not only to the ongoing interaction, but they also rely on previous sessions and other sub-genres. This contributor also outlines certain trends that distinguish the parliaments under consideration; whereas British MPs seem to criticise their opponent’s statements’s Russian MPs would rather point a blaming finger at their behaviour. The other two chapters in this part dissect a talk-show broadcast on Czech television. Martin Havlik, and then Světla Čmejrková and Jana Hoffmannová, breaking away from a straightforward description of the genre, show the polyphonic strategies which allow the host to fulfil his purposes. In contrast with presidential debates, the aim of a talk show is to entertain the audience. There is a hierarchical distance between the participants, since in “Uvolněte se, prosím” (“Relax, please”), for the purpose of providing fun, the host takes advantage of his guests by playing on their words or on what third parties have said about them. In Chapter ten, Havlik mostly studies prosodic and other non-verbal markers which produce ironic effects. In Chapter 11, Čmejrková and Hoffmannová, also explore rhetorical wrestling between host and guest where the words of the latter are used along with other discourses, for instance embarrassing rumours. Part 4 contains three chapters which deal with polyphonic social and cultural aspects of intertextuality. Margareta Manu (Chapter twelve) describes the phenomenon of rumour in written journalism in Romania. In doing so, she distinguishes two aspects; firstly, the source(s) and the spreading agents of rumour; secondly, the textual marks of rumour, such as introductory formulæ, the use of evidentials or the splitting of enunciative instances. Manu shows the ultimate heteroglossia of the texts, as being stylistically hybrid, and the strategic use of rumour by journalists, who legitimize their gossiping statements as coming from “public opinion”. In Chapter thirteen Ana Ene leads us towards a more extended dialogism. She treats the issue of translation from the standpoint of a dialogue between texts and cultures. Ene defends the dialogical nature of literature and its endeavour towards universality, which can only be totally attempted through translation. This contributor then raises the issue of the difficulty involved in translating. Ene advocates for an understanding of translation as proposed by theoreticians such as Berman or Meschonnic, who appeal for an ethics of translation, an ethics which implies a better dialogue among languages and cultures. To this end, they propose several means and procedures. One of them is to compare translations. In line with this idea, Ene presents an intertextual contrastive analysis of several translations of the same poem.



Introduction

She sets up the rhetorical analysis, as an initial step, which is followed by an examination of the poem’s translations. Ene concludes by proposing a peritextual dialogue in order to improve the quality of the translations. Finally, Chapter fourteen in Part 4 broadens the scope for studying dialogism to encompass an artistic, cultural and social perspective. Yosi Anaya invites the reader to turn to American textiles and look at them as almost the last refuges for ancient texts. Anaya reminds us of Bakhtin’s broad view of heterogolossia, which can be applied not only to words but also to forms. Therefore, textiles created by Meso-American indigenous women, notwithstanding a hegemonic tendency to erase them, constitute a space of dignity and cultural resistance. Anaya describes the discourses crossing these layered textiles, paradoxically through silent voices. The various motifs and models of huipil allow us to reconstruct different hues of individual creativity and a community’s cultural and spiritual life. The last part of this book concerns literary discourse. Bakhtin had already pointed out that literary language becomes a dialogue of languages where social variations are reflected. In the case of the novel, the very principle of its discourse is the presence of alternating voices (basically narrator and characters) which express different points of view of the world. Through artistic elaboration a work of literature aims to achieve a more relevant reference (Ricœur 1975: 289), which Jakobson had noted when he spoke of a “doubled reference” (1963: 238). Thus, dialogism and dialogue in literary texts manage to reveal something meaningful about the world. Moreover, since writers often explore the behaviour and the language of human beings, dialogues in novels teach us a great deal about conversational dynamics. Chapter fifteen delves into the world of the Irish author Flann O’Brien. Flore Coulouma studies dialogism in two of his novels on three levels. The first level is the interlocutive. Coulouma analyses, in the light of the Conversation Analysis, how the characters subvert cooperative rules to reach a symbolic significance.The novel also shows the diglossia of language itself, by the language play involved in dissonant voices. Thirdly, the article explores the complex relationship between Flann O’Brien and the literary traditions that preceded him. The novel is not the only genre veined by a diversity of voices and discourses. Emilia Parpala (Chapter sixteen) and Carmen Popescu (Chapter seventeen) study the polyphonic strategies used by several Romanian poets at the end of the 20th century. Parpala focuses her work on the poetry of the eighties. She discusses the restricted place Bakhtin has attributed to poetry, from a polyphonic approach. The Russian thinker, indeed, distinguished the popular novel from highbrow monologic literature, whose main point of reference would be (presumably highbrow) poetry. Responding to poetological constraints, Parpala shows, in her analysis, the poets’ use of different enunciation markers in order to stage the division of the self,





Clara-Ubaldina Lorda and Patrick Zabalbeascoa

to include other speaking subjects (even imaginary or fake ones) in their texts, and how, finally, these poets succeeded in furtively alluding to an oppressive regime in which they had to live. Popescu devotes her work to the evolution of Romanian poetry in the nineteen-nineties, whose ironic stylization and parodies exploit the use of such techniques as the inclusion of fictional conversations, the hybridisation of languages or the general presence of interdiscourse in their poems. Through these strategies, and multiple readings, the poem also accomplishes its function of revealing the harm of an oppressive situation. Finally, Chapter eighteen returns to the novel, but not to any novel in particular. As psychologists, Ilaria Riccioni and Andrzej Zuczkowski know the manifestations of the inner world in therapy. Nevertheless, because of the inaccessibility of the inner world, they also turn to several novels in order to show, through the analysis of verbal elements, how the essential division of the subject is manifested in discourse. In this way, by exploring shifters, conversational forms, semantic contradictions and pragmatic markers, they stage the disidentity of the individuals produced in inner discourse, as a natural result of the different parts composing each I. In this volume we bring together interweaving strands of thinking from Veracruz to Belarus, combining models of communication and culture based on the concepts of interdiscourse, dialogism and polyphony. The contributors found spaces of polyphony in discourses of young and old people, in private conversation and in the media, in politics and in poetry, in education and in jokes, in the press and in translation, and even in non-verbal discourse literally woven into clothing as an alternative voice to Meso-American women. The book is also a shared space for looking at and beyond the relevance of influential scholars, especially Bakhtin, and others like Authier-Revuz, Benveniste and Ducrot. It is a space, too, for making progress in discourse analysis and giving a voice to researchers who are breaking new ground in studies of polyphony. The book itself is accordingly polyphonic, and we hope the reader will also find it harmonious and choral. The chorus of the contributors’ voices is a positive metaphor for scientific research, where different voices must take each other into account. We are sure the reader will find the dialogue established between the various Chapters and Parts of the book both enlightening and enriching.

part 1

Strategies in daily conversations

chapter 1

Strategy and creativity in dialogue Robert E. Sanders

University at Albany, State University of New York Bakhtin’s idea that any utterance has built into its meaning its connectedness to other utterances gives rise to his idea of the “anticipatoriness” of utterances – that utterances are made with the “expectation of a response” (Bakhtin 1986: 69). If so, then each utterance delimits the utterances that can meaningfully come next, that would be responsive. The upshot is that Bakhtin’s ideas about the meaningfulness and interrelatedness of utterances implicate that utterances have an inescapably consequence-oriented, strategic aspect. Yet the idea of speaking strategically seems monologic rather than dialogic, antithetical to the premise of co-construction. To reconcile what seem to be incompatible concerns, a dialogical account is developed of two key underpinnings of strategic speaking − that responses to utterances are somewhat predictable and not open-ended, and that speakers form and act on intentions about what their utterances will bring about.

1. Strategy and creativity from a dialogical perspective Little attention has been given either to interactional strategy and creativity from a dialogical perspective, or their connection, perhaps because on the surface they seem to be at odds with dialogism. However, as I will show, rather than running contrary to these phenomena, the dialogical perspective sheds light on them. Considering that discursive strategy and creativity are empirically real, this is a good thing both for our understanding of these phenomena and for dialogism. Interactional strategy and discursive creativity are arguably basic human capabilities. They are exhibited early in a child’s development, at least by the age of 5 and probably much earlier (Kidwell & Zimmerman 2006; Sanders 2007). Consider, for example, the following segment from an interaction between two girls of about 6½ years, Rachel and Lola. They had been given a Lego building block set and asked to collaborate in making something. They were making a house. At one point Lola found a window among the assorted pieces in the set and positioned it on the wall of the house. Rachel almost importantly moved the window to a



Robert E. Sanders

different place from where Lola had positioned it (See the Appendix for a glossary of transcription symbols). (1)   1 Lola: (°Look!/There!°) (.) >The:re’s another window: dow:n   2 here:<   3 (1.7) ((L installs the window))   4 Rachel: We nee:d that later. (.) ((shifts gaze)) We put, this like   5 this. ((R is arranging pieces on base)) (.) A:nd (0.2)   6 ↑Whoa:: (.) (We nee:d) (0.2) >a window to go on the   7 other si:de.<   8 (0.7) ((R removes window from where L   9 installed it)) 10 Rachel: So we can see what’s going on:. 11 (2.7) ((R installs window further back from 12 where L had it)) 13 Lola: >Oh but that < Speech within “more than”/”less than” inequality signs is markedly fast. < > Speech within “less than”/”more than” inequality signs is markedly slow. (word) Single parentheses around speaker text indicates transcriber uncertainty. ( ) Single parentheses around blank space indicates talk unintelligible to the transcriber. (( )) Text within double parentheses is a transcriber comment.

chapter 2

Conversational irony: Evaluating complaints Maria Christodoulidou

Frederick University Cyprus This chapter is concerned with ironic assessments in spontaneous CypriotGreek conversations. The phenomenon is studied here by a close examination of ironic assessments in the specific context of naturally occurring complaint sequences where the speaker is complaining about a non-present party’s misbehaviour or reported words. The main focus is on the sequential and interactional role of the placement of ironic utterances in a structural place of the organisation of storytelling that is upon the culmination of a complaint, where usually an evaluation is expected from the recipient. Ironic evaluations are investigated by exploring the whole construction of the complaint and the relationships established among the participants up to the slot which is filled with the ironic element. This is why I draw from selected conversational fragments to show that in the data there is a pattern which is revealed by combining irony with an expression of evaluation. This type of evaluation is very effective in producing a stronger claim than the one reported and proffered with the punchline. This moves the topic to closure. Examples of ironic evaluations are effectively expressed through non-literal means, such as extreme case formulations, impossible descriptions, self-contradictory assessments, and rhetorical questions.

1. Introduction In recent years, there has been widespread interest in verbal irony, but the main object of investigation has been the ironic sentence, regarding it as self-contained irony in isolation from its context (cf. Sperber and Wilson 1992), or at the most in the context of a constructed text (Jorgensen et al. 1984, Gibbs and O’Brien 1991, Dews et al. 1995, Giora 1995). By focusing on elicited or invented data, these theoretical models fail to illuminate what irony does in conversation. An adequate characterization of verbal irony manifested in different forms and an exploration of its functions is most clearly realised by investigating naturally occurring



Maria Christodoulidou

conversational talk. To this end, major contributions to the discussion of irony are Clift’s (1999) and Kotthoff ’s (1998, 2003) studies on irony in conversation. This chapter examines ironic assessments in complaint-story sequences in Cypriot-Greek conversations.1 In particular, I analyze the process by which tellers invite affiliation and recipients respond affiliatively in their complaints about the misbehaviour of third non-present parties. My main focus is the sequential position of ironic assessment utterances in complaint sequences and the role of their placement in establishing meaning. Following Clift (1999), I adopt Goffman’s (1979, 1981) concept of framing in my interpretation of irony, paying particular attention to shifts in footing. Goffman describes footing as “the alignment we take up to ourselves and the others present as expressed in the way we manage the production or reception of an utterance” (1981: 128). Changes in alignment produce new frames in which the utterance is interpreted. The notion of frame is used to refer to the way people define the situation they are in and organize events (Goffman 1974: 155). Although Goffman is not concerned with irony, his reference to a speaker who employs conventional inverted commas to indicate that s/he does not mean what s/he says or that s/he is repeating the words of someone else (1974: 512) makes the notion of framing applicable to irony, because, in being ironical a speaker indicates lack of responsibility for the truth-value of what s/he says. The use of such cues allows a speaker to shift her or his footing into a frame of irony, indicating how the framed words should be interpreted. Here, I look at examples of irony in complaint sequences but I have a broader aim: to illustrate the importance of conversation analysis in shedding light on what irony is through an account of what it can be used to do. 2. Data and methodology The study of ironic responses investigated in this work is based on recordings of informal, spontaneous, face-to-face conversations among close friends or relatives, exclusively conducted in Cypriot Greek. The extracts included in this chapter 1. Cyprus is an independent island republic in the Eastern Mediterranean. Data from the 2001 census of population showed that on 1st October 2001 the total population of the Cyprus Republic was 689,565 composed of 89.7% Greek Cypriots, 0.2% Armenian, 0.5% Maronites, 0.04% Cypriots of European origin (called Latins) and 0.05% Turkish Cypriots; 0.1% did not declare their ethnic religious group (Census of Population 2001); the remainder being foreigners from Europe and Asia. The Greek speech community in Cyprus is defined as diglossic. Diglossia in Cyprus refers to the simultaneous use of the dialect (Cypriot Greek dialect, CD) and the demotic Greek (Modern Greek, MG).



Chapter 2.  Conversational irony: Evaluating complaints 

comprise transcriptions of approximately three hours of tape-recorded, naturally produced conversations in a variety of social encounters (e.g. while dining or over a cup of coffee). The method adopted in the data analysis is Conversation Analysis (CA), which has its origins in the pioneering work in the sixties by the sociologist Harvey Sacks (1992a, 1992b).2 The transcription symbols used in this study (Appendix I) are based on the transcription conventions developed by Jefferson for the analysis of conversational turns in English conversation. (cf. Sacks, Schegloff and Jefferson 1974) 3. The complaint sequence Storytelling sequences are generally composed of three serially ordered and adjacently placed types of sequences: the preface, the telling, and the response (Sacks 1974: 337). Complaint sequences emerged as a subcategory of storytelling in my data. CA shows that complaints are well organized social phenomena, produced methodically in order to reveal the participants’ stance. In the examples analyzed in this chapter, a speaker tells a story in order to criticize or complain about what a third non-present party has done or said to her and a listener responds to the story. At this point, I draw on selected conversational fragments to point out some of the features of complaint-storytelling sequences through which speakers report the behaviour − or words being criticised − and how the listeners respond. My main interest is in the listener’s slot upon story completion, which is a common place for the occurrence of evaluation (Schegloff 1984). What interests me is how this slot is filled with ironic assessments in the extracts of my data. A good example of the use of an ironic assessment after a narrative is shown in example (1).

(1) (L = Litsa; C = Christiana. Other women are present in the conversation. This is a second story in talking about Fortune-tellers.)   1 L guys we- £3we went once to one4 for fun£ I don’t remember   2 when, long time ag- and she said to me-- I used to go::, I used 2. Conversation analysis has focused its analytical attention on recorded, naturally occurring talk-in-interaction (Hutchby and Wooffitt 1998). These recordings of actual speech are transcribed using a system which is intended to capture in detail the characteristics of the sequencing of turns, including gaps, pauses and overlaps; and the element of speech delivery such as audible breath and laughter, stress, enunciation, intonation and pitch. (Hutchby and Drew 1995: 182) 3.

See Appendix II for Key to transcription symbols.

4. Fortune-teller.



Maria Christodoulidou

  3 to go:: often, I was like ha ha ha5 I was laughing, yes I’ll   4 believe now::, and she was turning ca::rds and she was turning   5 thi::ngs. I don’t even remember, I don’t remember what she said   6 to me. one thing I kept and can you believe I was stressed? she   7 told me that, supposedly, when I am twenty seven, around the   8 age of twenty seven, you’ll have a serious illness she said to me   9 but you’ll pass it. can you belie-? I was stresse::d. 10 C → thank you very much. 11 L what kind of illness will I pass? I don’t know::, I say to myse::lf, 12 is it for my fathe::r’s health and for any of his illnesses? After Litsa’s narration of her experience, Christiana (1: 10), the interlocutor, takes the side of the speaker by ironically saying “thank you very much” to the fortuneteller. This occurrence of ironic evaluation is the focus of the following analysis. Following, preface, telling, and response sequences of examples (2)–(4) are examined to show how the speaker initiates a complaint and describes someone else’s transgression, and how the interlocutor(s) respond(s) to it with ironic evaluation. Table 1 below is a diagram of the storytelling structure in the extracts included in this chapter, as presented in Sections 3.1. to 3.3. 3.1

The preface sequence: initiation of complaint/criticism

In a seminal article, Drew (1998) showed that complaint sequences are bound sequences, the beginning and the end of which are easily identifiable. In the examples that follow, complaints are developed as stories and begin with a preface. A story preface is defined by Sacks as “an utterance that asks for the right to produce extended talk” (1970: 226). It may be two turns long, the first one involving talk by the speaker and the second by an interlocutor. In this case, the speaker is producing an utterance that contains sequentially relevant components, including: (1) an offer or a request to tell the story; (2) a preliminary characterization of it and (3) some reference to the time of occurrence of the story events. (Sacks 1974: 337–341) Table 1.  Storytelling/Complaining Teller: Preface → Initiation of Complaint/Criticism Telling → Description of Transgression Recipient: Response → Ironic Evaluation

5.

“ha” here is used to indicate description of laughter.



Chapter 2.  Conversational irony: Evaluating complaints 

If the first addressee expresses a request for the story or an acceptance of the offer, then the preface sequence can take a minimal length of two turns and after that the narrative can be undertaken, the story-teller holding the floor for that activity (ibid.). The prefaces of examples (2)–(4) illustrate the initiation of a complaint/criticism of a third non-present party.

(2) (T = Thekla; M = Maria; D = Demetra. Maria, Thekla and Anna, a nonpresent participant are classmates. Earlier that day Anna got to class late. The following exchange is about her delay.)   1 M well as I was goi::ng to class and I was in the booksho::p   2 (.) I see ↑Anna re:: far o::ff! an- walking going ↑UP6   3 [and I sa- and I tho   4 T [why was she twenty minutes late?   5 M exactly::.   6 T where:: wa::s she?   7 M e[xactly::.   8 T [let me tell you because-- let me tell you because I forgot to tell   9 you this, 10 M so:: go:: on. In example (2) two participants cooperate in telling the story. Thus, in (2), the story-teller and a knowing addressee seem to share some knowledge about the story as shown by the knowing addressee’s contribution (2: 4) to the story-telling initiated by M (2: 1–3). The narrator (2: 5, 7) endorses the contributions and when the knowing addressee (2: 8–9) offers to tell the story the first narrator (2: 10) responds to that with a go-ahead (cf. Schegloff 2007). To proffer a contribution and its endorsement is to declare a shared stance. Now, let’s consider example (3).

(3) (C = Christiana; M = Maria; A = Angelina; P = Petra; E = Eleana. Before the following exchange, Christiana was saying that the previous night a young guy was flirting with her in the club before Andie showed up. The conversation is about Andie, who is not present during the conversation.)   1 C ((to E)) ((do you)) reme::MBE::R! [Andie’s   2 Μ [hu   3 C syndrome::!   4 M what is her syndrome?   5 C whenever someone is talking to us she starts talking to hi::m?   6 Ε n::,7

6. Away from where the class was. 7.

no



Maria Christodoulidou

  7 C   8   9 Α 10 C 11 Ε 12 C 13 Ε 14 C 15 16 Ε

since th-- don’t you remember that day that those guys [LIZA’s friends from Limassol? [don’t say that agai::n re Christiana::. came? that she was going to introduce to us actually. yes. those. and when someone was talking to us Andie was coming and pulling him one side like that and was talking to ↑hi::m! so? (.) did she do that last ni::ght too?

In (3: 1–16) two participants share some knowledge about the story’s events. Thus, the story-teller (3: 1, 3) solicits reminiscence recognition (cf. Lerner 1992: 255) from E, the listener who knows about the principal character’s (cf. Goodwin 1984) behaviour and foreshows a possible story. By characterizing Andie’s behaviour as a “syndrome”, not only does the narrator (3: 3) foreshow a negative criticism of Andie and establishes her stance towards the upcoming narration but she also invites the knowing recipient to confirm what it assesses and express a similar stance.8 The recognition by the knowing recipient comes in (3: 11) and the request to proceed with the story, the go ahead, in (3: 16).

(4) (C = Christiana, P = Petra. This is part of a longer extract that occurs after other tellers’ complaints on the same topic. Now, Christiana initiates a presentation of her own perspective). 37 C =[shall I tell you something? 38 P [(------) 39 C what got to me the most? The complaint sequence of fragment (4) begins with a preface in the form of an interest arouser (Sacks 1968: 10, 18; 1970: 226), which announces the initiation of a complaint/criticism, worded as a question, which in 4: 37, 39 is audible as a request to proceed to the telling. The prefaces of the extracts presented above show the story-tellers’ attention in displaying their stance towards what they are reporting. This is in accordance with Drew’s observation (1998) that at some point in the narrative the complainant expresses indignation about what the other has done by reporting how this 8. The addressee, however, withholds a response, while M (3: 4), another participant, initiates a next-turn repair. The teller (3: 5) makes a clarification which is also audible as a second solicitation of reminiscence recognition (3: 5) addressed to E, the knowing recipient. E (3: 6) responds negatively to the solicitation and this is in disagreement with the expectations of the solicitation. The teller initiates another request for reminiscence recognition (3: 7–8).



Chapter 2.  Conversational irony: Evaluating complaints

made her feel. This way the teller elicits recipients’ affiliation from the very beginning. As shown, recipients’ affiliation is expressed in many ways which give the “go-ahead” to the teller to proceed to the telling. 3.2

The telling sequence: Description of transgression

When storytelling has been properly prefaced, its narrator should proceed to tell it to its completion. Unlike the organization of the preface sequence, place for talk by recipients within the telling sequence need not be given by the teller and the telling can then take a minimal length of one teller turn. (Sacks 1974: 344) The prefaces presented in 3.1. foreshadow an upcoming telling regarding the non-present party’s behaviour. Hence, the following examples represent the telling sequence of the examples collected, which in (5) and (6) consists of a description of the non-present party’s misbehaviour, whereas in (7) culminates in a specific direct reported utterance, attributed to the person complained about. (5) (T = Thekla; M = Maria)   8 T [let me tell you because-- let me tell you because I forgot to tell   9 you this, 10 M so:: go:: on. 11 T be::cause when I gave her the copies that she returned to me they 12 were:: (.) 13 M staple::d. 14 T yeah. 15 M so? and she went to staple them, 16 T and she brought the::m today when she brought back one of the 17 articles, It didn’t even have staple ma::rks (.) which means that 18 she went and made a copy. 19 D and she gave you the co::py::?= 20 T =>and she gave me the copy yeah. okay it is just as well that she 21 considered to copy it after one and a half monthsyes, she liked it so much, that’s why she wanted to switch it with 51 Maria’s.< 52 M how come I liked it so much? 53 C she said something about cutting off the slee::ves.

Example (7) develops differently; the pre-announcement which initiates a complaint sequence is followed by the teller’s direct reporting of the third person’s words she is complaining about. Thus, the third person’s impoliteness is represented specifically through what she said and this is what the teller is complaining about. Through her direct reported speech “the blouse will do” (7: 41, 47–48) the third person is reported as expressing a compromise in accepting the gift. This is in accordance with Drew’s observation (1998) that in complaint stories where reported speech plays a part, reported speech attributed to the person complained about constitutes the climax of the narrative. By referring to these words as said “in the end” (7: 41, 47) the teller presents them as the exemplification of her offensive behaviour and the climax of the telling. Thus, this assessment is offensive based on



Maria Christodoulidou

the social norms of politeness that when receiving gifts people are supposed to express gratitude rather than a lukewarm, grudging acknowledgment. Hence, in the telling sequences of the examples examined above the teller is complaining about a third party’s misbehaviour or about the direct reported utterance of the non-present party’s speech. This utterance is placed at the climax of the telling. 3.3

The response sequence: Ironic evaluation

Sacks (1973, 1974) considers a story recipient’s slot after story completion a structural place in conversation and, according to Schegloff, one reason for this is that turn-taking operates on an utterance-to-utterance basis, so any subsequent utterance completion point is where the next speaker may seek to effect transition. (1984: 43–44) A sequential problem for storytellers, inasmuch as stories take more than a single sentence to produce, is how to get prospective speakers, who may start talking at the first possible completion point, not to start talking. Extended utterance completion will thus have to be identified by recipients by identifying story completion. Given that, as long as the story is still being told, other speakers properly refrain, story completion is fundamental to the sequential structure of the conversation and the story recipient’s slot after story completion is an especially marked place because that is where recipients must show appreciation or understanding of story completion (Sacks 1984). In the examples analyzed, speakers resort to ironic turns in order to perform their evaluations. In examples (8)–(12) the story recipients’ slots are indicated with arrows. (8)

26 [instead of copying it from the very beginning she 27 was keeping mine:: and she didn’t even bring me the other 28 one. 29 (3) 30 M → ˚I see (2) clever Anna˚. 31 T she was coming down from the library?

In (8: 29) the three-second delay is attributable to M because the teller comes to a strong completion in (8: 28). However, it should be noted that most climaxes get immediate uptake. We cannot dismiss the possibility that something else is going on, and video recording could be very useful. The recipient’s assessment “˚I see (2) clever Anna˚” occurs in a position where an evaluation is expectable. This description might be interpreted as an impossible description (Torode 1996) of the actual event where an extraordinary reality is momentarily acknowledged and shared and must be treated rhetorically or figuratively (ibid.). What makes it impossible is



Chapter 2.  Conversational irony: Evaluating complaints 

the fact that is juxtaposed with Anna’s obvious lack of cleverness. The shift of footing achieved with Anna’s impossible description frames the utterance as ironic. Moreover, the sequential organization of the extract also frames the recipient’s assessment as ironic evaluation since it occurs upon story completion. The softer articulation of the whole turn and the two-second pause after the evaluation-­ reminiscent “I see” followed by the impossible description “clever Anna” suggest a certain deliberation, effectively framing and marking off the irony. With this evaluative summary the recipient shows agreement and affiliation with the teller. The irony here serves as a criticism of Anna, but at the same time it is affiliative towards Thekla, enhancing common ground. So, although co-operation and agreement was established much earlier, this assessment serves as upgraded ironic agreement not only of what is uttered by the previous speaker but also of what is uttered in the whole extract. With this evaluative summary of the whole extract, M makes an offer to close the topic. (9)

26 27 28 29 30 31 32 33 34 35

C and he was [dancing Ε [what, C and Andie [[here within his, Ε [[a nerve! C legs and they were dancing stuck like glue like tha::t, Μ ↑oh really re::! Α seriously::? P → so [does she spend all her time with the priest?9 Α [and what did Liza say to her? C ↑nothi::ng.

In (9) there is more than one listener to the story and most of them, that is, M (9: 31), A (9: 32) and P (9: 33) make an evaluation upon the story-completion one after the other. Thus M (9: 31) and A (9: 32) both display assertions of ritualized disbelief, conveying their amazement and disbelief (Wilkinson & Kitzinger 2006: 34). One of the recipients, P (9: 33), produces a rhetorical question, identified as such because it does not beckon a response since it involves common knowledge. This makes it interpretable as an ironic evaluation of the third person. The question is framed as ironic evaluation, based on the fact that is not sequentially linked to the previous description. In addition, the extreme description10 “all her time” adds to the ironic interpretation. The ironic evaluation conveyed is also 9. Lit: and is she chasing the priest all day? 10. Extreme case formulations (ECFs) are descriptions or assessments that deploy extreme expressions such as every, all, none, best, least, as good as it gets, always, perfectly, brand new, and absolutely (Pomertantz 1986)



Maria Christodoulidou

recognized as it is based on the shared knowledge that Andie often visits a priest and consults with him. With this assertion, P (9: 33) offers another argument for Andie’s behaviour being reprehensible by ironically evaluating her incompatible actions. Her behaviour as described by the teller contradicts the fact that she is known to spend a great deal of time with the priest. (10)

36 M ((to P)) would you like some milk?= 37 C =[shall I tell you something? 38 P [(------) 39 C what got to me the mo::st? 40 M [shall I get you some? 41 C [that in the end she said [well, the blouse will do too. 42 [((to M)) just a little. 43 M what blouse? 44 C that we gave her= 45 G =as a gi::ft 46 M I didn’t hear that 47 C in the e::nd that she said-- that she said-- she says-- she says to 48 you well okay re Aleka, the blouse [will do too. 49 [will do too. 50 P → >yes, she liked it so much, that’s why she wanted to switch it with 51 Maria’s.< 52 M how come I liked it so much? 53 C she said something about cutting off the slee::ves.

The storytelling in example (10) culminates in an extended turn where the teller is reporting only the non-present party’s words. Given that it is at the climax of the story that affiliation is sought, the use of reported speech at this point may be seen as a means of eliciting affiliation without being seen explicitly to solicit it (Clift 2000). Specifically, in (10: 47–49) upon the culmination of the story, one of the recipients, P (10: 50, 51) ironically agrees in order to disagree with the complained-about reported words. The ironic agreement is followed by an impossible description which consists of an extreme ironic self-contradictory assessment. Thus, the shift of footing achieved with the fact that the first part of this description is an overbuilt assessment “she liked it so much which” is in contrast with the second “that’s why she wanted to switch it with Maria’s” frames the assessment as ironic. This extraordinary shared reality makes the ironic evaluation discernible. The second part of the evaluation “that’s why she wanted to switch it with Maria’s” consists of another argument against the third person added by the recipient, which reinforces that this evaluation is against the third person. This way the recipient attempts “to extrapolate a possible description to the point of unreality.



Chapter 2.  Conversational irony: Evaluating complaints 

This reductio ad absurdum seeks to refute the actual prior description” (Torode 1996: 33). 4. Conclusion Using the analytical tools of Conversation Analysis, this chapter has examined the use of irony in talk-in-interaction, giving us a means of exploring what irony is through an account of what it can be used for. One of the first issues that was dealt with was where irony ‘“lives’ in order to find out its ‘habits”’. Data from Cypriot-Greek conversations revealed the occurrence of irony in the context of complaints where the teller describes another’s transgression towards her or reports what the other said in the form of direct reported speech. In exploring the sequential position of ironic assessments it was revealed that a regular place for their occurrence is in the recipient’s slot upon story completion. The occurrence of irony in that marked place where appreciation is expected required a careful investigation In order to elucidate the role of the placement of ironic assessments in complaints, the whole construction of complaint sequences up to the slot where irony occurs was analysed. As was shown in the extracts analysed, the complainant expresses the speaker’s stance towards what she is reporting and invites the recipients to collaborate and express a similar stance. A common locus in the extracts under study is that collaboration between the teller and the recipients is established very early on in criticizing a third party. However, as the story continues, place for response by addressees within the story-telling sequence need not be given. The recipient’s slot where is supposed to express evaluation is upon story completion. The deployment of irony in that slot is very important if one takes into consideration that the slot before the occurrence of the evaluation includes the climax of the story consisting of the punchline of the description of the other’s transgression or the complained about reported words. Bearing in mind that complainants seek affiliation and expression of similar stance, this is the place where recipient/recipients is/are supposed to express their strongest agreement with the teller and evaluation of the transgression. With ironic evaluations recipients pretend to agree in order to disagree with the opponent and that is how they support/defend the attitude/viewpoint of the storyteller. The combination of irony with an expression of evaluation is very effective in producing a stronger claim than the claim reported as proffered with the punch line and is interpretable as an argumentative summary that defends the teller and moves the topic to closure as shown by the fact that irony is followed by agreement.



Maria Christodoulidou

What makes irony discernible in the extracts presented here is a shift of footing that frames the utterance as ironic each time. In irony there is a gap between what is said and what is meant; the teller says something and at the same time indicates a lack of responsibility surrounding what she said. In the examples presented here, ironic evaluations are effectively expressed with non-literal means, such as extreme case formulations, impossible descriptions, self-contradictory assessment, and rhetorical questions. This chapter presents ironic evaluations in a single context of their occurrence. However, a huge amount of work remains to be done in describing the full extent of irony and its resources in everyday talk. Appendix I: Original examples (1)   1 L peθca epi- £epiamen ja χazin se m\an emis m\a fora pote epiamen   2 £ en iθθimume prin χro\a poll- tw’ ipe mmu tuti emenan-- eγo   3 epienna:: epienna:: polla, milo su oti χa χa χa mes sto χaŋχano,   4 nne pou na pistepso tora::, twe jirizen mu χarca:: twe jirizen mu   5 p’ afta::. en θimume kan, ute θimume indam bu mu ipe. ena   6 prama mu eminen monon, twe kseris to oti aŋχoθika? ipen mu   7 miwimu oti sta kosiefta:: mu, jiro sta kosiefta:: mu::, enna perasis   8 m\a sovarin aroscan ipem mu, alla enna tin perasis. piste-?   9 aŋχoθika::. 10 C → efχaristo para poli. 11 L inda aroscan enna peraso? ksero γo::, lalo twe γo::, e ja tin ijia tu 12 papa:: mu twe ja opcaδipote tu aroscan? (2)   1 M   2   3   4 T   5 Μ   6 T   7 Μ   8 T   9 10 Μ

lipon opos epienna:: tw’ imun mesa sto vivliopolio::n (.) vlepo tin ↑Anna re:: pu makria::! tw- na perpata n’ anevenni ↑PANO [twe lal- twe sce[jati arcisen ikosi lepta::? ↑a bravo::. pu:: ita::n? ↑a [bravo::. [na su po jati-- na su po eγo jati ekseχasa na su to po tuto, e a::te::.

(3)   1   2   3   4

((to E)) θima::SE::! [to [hu sinδromo tis Andi::ς! ti sinδromon ewi?

C Μ C M



Chapter 2.  Conversational irony: Evaluating complaints 

  5 C   6 Ε   7 C   8   9 Α 10 C 11 Ε 12 C 13 Ε 14 C 15 16 Ε

opcos mas mila pai twe pcanni ton twe mila ↑tu::! o::, pu twin din-- en di θimase twin din imera pu rt- twin da peθca ta:: [i fili tis i Lemewani::? [mem mu to ksanapi::s re Xristiana::. tiz LIZA::S? pu tan na mas proksenepsi:: telospanton. ne twinus. tw’ opcos ercetun twe milam mas ercetun tw’ epcanen ton etsi i Andi:: tw ’ epienne tw’ emilan ↑tu::! e? (.) ekamen do twe pse::s?

(4) 37 C 38 P 39 C

=[na su po kati? [(------) emenan di me eknevrise parapano::?

(5)   8 T   9 10 Μ 11 T 12 13 Μ 14 T 15 Μ 16 T 17 18 19 D 20 T 21 22 Μ 23 T 24 Μ 25 26 T 27 28

[na su po jati-- na su po eγo jati ekseχasa na su to po tuto, e a::te::. δio::ti ta filla::δia pu mu feren, eγo pu tis ta ’δoka ita::n (.) karfitsomena::. nne. e? twe pien na su ta karfitsosi, twe eferem mu ta:: simmera pu mu to efere do ena to articol, δen iχaŋ gan tripu::e::s (.) pu simeni epie tw’efkale to kkopy. tw’ efere su to kko::py::?= => tw’ eferem mu to kkopy nne. daksi pale kala pu to sceftice na to fkali kkopi meta pu anamisi minanne tosom bolla tis arese j’ afton iθele na tin allaksi me twini 51 tis Marias.< 52 Μ emena jati mu arese para polla? 53 C kati eleen oti iθele na kopsi ta manitwa::.

(8) 26 T [andi na to 27 fkali pu tin arçin ekratusen do δiko mu:: twe to allon emmu to fereŋ 28 gan. 29 (3) 30 Μ → ˚malista (2) eksipni Anna˚. 31 T pu ti vivlioθici pu katevenne tes skales?



Chapter 2.  Conversational irony: Evaluating complaints 

(9)

36 37 38 39 40 41 42 43 44 45

C twe χorefce [twinos Ε [inda, C tw’ i Andi [[δame mes ta Ε [[pco θarros! C poθca tu twe χorefkan kolliti etsi::. Μ ↑ate re::? Α ma sovaromila::s? P → tw’ [ulli mera vura tom bater pu piso:: twini::? Α [tw’ i Liza ti tis ipen? C ↑tipoTE::.

Appendix II: Transcription system [ Separate left square brackets, one above the other on two successive lines [ with utterances by different speakers, indicates a point of overlap onset, whether at the start of an utterance or later. [[ Double separate left square brackets, distinguish pairs of overlapped utterances. [[ = Equal signs ordinarily come in pairs – one at the end of a line and another at the start of a next line. If the two lines connected by the equal signs are by the same speaker, then there was a single, continuous utterance with no break or pause, which was broken up in order to accommodate the placement of overlapping talk. If the lines connected by two equal signs are by different speakers, then the second followed the first with no discernible silence between them. (2) Numbers in parenthesis indicate silence. (.) A dot in parentheses indicates a micropause. . The period indicates a falling or final, intonation contour, not necessarily the end of a sentence. ? A question mark indicates rising intonation, not necessarily a question. , A comma indicates continuing intonation, not necessarily a clause boundary. :: Colons are used to indicate the prolongation or stretching of the sound just preceding them. The more colons the longer the stretching. - A hyphen after a word or part of a word indicates a cut-off or self interruption, often done with a glottal or dental stop.



Maria Christodoulidou

word Underlining is used to indicate stress or emphasis. word Capital letters indicate louder than the rest talk. ˚ ˚ Two degree signs indicate that the talk between them is markedly softer than the talk around it. ↑ The up arrow indicate a segment starting on sharper rise. > < The combination of “more than” and “less than” symbols indicates that the talk between them is compressed or rushed. .hhh The dot followed by “h’s” indicates inbreath (h) The letter “h” in parentheses inside the boundaries of a word indicates laughter. (( )) Double parentheses are used to mark transcriber’s descriptions of events, e.g. ((telephone rings)), ((sniff)) etc. (word) When all or a part of an utterance is in parentheses, this indicates uncertainty on the transcriber’s part, but represents a likely possibility. £word£ Word or Words enclosed by pound sterling signs indicate the word is articulated through a hearably smiling voice. ( ) Empty parentheses indicate that something is being said, but no hearing can be achieved. → An arrow marks significant turns.

chapter 3

Speaking through other voices Conversational humour as a polyphonic phenomenon Béatrice Priego-Valverde Université de Provence

Conversational humour is a complex phenomenon for a number of reasons. It is ambivalent (both aggressive and benevolent), and it is contextualized and based on shared knowledge (hence difficult to grasp fully by someone outside the group). In this study I will explore another reason for this complexity, and that is humour as a polyphonic phenomenon, a heterogeneous discourse produced, of course, by the speaker her/himself but at the same time by many other voices (Ducrot 1984) which cross and converge with the speaker’s discourse. I will show that what makes conversational humour more complex is not only the fact that it is a polyphonic phenomenon but, rather, that the speaker plays hide-and-seek with the various voices s/he invokes.

1. Introduction Conversational humour is a complex phenomenon for a number of reasons. It is ambivalent (both aggressive and benevolent), it is contextualized and based on shared knowledge (hence difficult to perceive by someone outside the group). In this study, I will explore another reason for this complexity. Humour can be seen as a polyphonic phenomenon, i.e. a heterogeneous discourse produced, of course, by the speaker her/himself but, at the same time, by many other voices (Ducrot 1984) which cross and converge with this same discourse. In this paper, I will apply the double voicing theory (Bakhtin 1929) to humorous sentences in order to reveal the game of hide-and-seek the speaker plays with the other voice(s) s/he uses in her/his discourse.



Béatrice Priego-Valverde

2. The data 2.1

The corpus

The corpus is made up of various casual conversations among friends and relatives recorded during evenings spent together. The participants were naturally very well acquainted with each other. They were all between twenty-five and thirty years old. The recordings were made with a visible microphone although none of the participants knew why they were being recorded, and, with the exception of the first few minutes when some participants asked about the reasons for the recording, they took little notice of it. The relationships among the friends were so close and their encounters so frequent that the microphone was quickly forgotten. As such, the present corpus can be considered a sample of spontaneous and natural speech behaviour. 2.2

Nature of the interactive setting

We usually assume that spontaneous conversation obeys certain norms (a–e). a. Symmetric positions between the participants. Theoretically (Kerbrat-Orechioni 1996), they all have the same rights and the same duties, especially as concerns the alternate role of speaker and hearer. b. A degree of cooperation (Grice 1975) that overrides whatever competition may exist. There is competition nonetheless, and we can reasonably assume that potential loss of face is present in conversation. c. An “inward goal, centred on contact” (Vion 1992), the maintenance of the relationship, the cohesion of the group where the only goal acknowledged is the pleasure of each other’s company and conversation. d. A mood of conviviality, which is a consequence of the previous conditions (a–c). e. An apparent informality which is sustained as much in the discourse (to speak about anything and everything, spontaneously, without a particular goal in mind) as in the interaction itself. As demonstrated by Sacks, Schegloff and Jefferson (1974), there are no explicit rules concerning the order or duration of speech turns in conversation, as turntaking is determined progressively. Thus, conversations are so auspicious that it is natural for them to provide a preferential space for humour. It is all the more natural in our corpus, as the conversations are colloquial and between people who know each other very well. Following Traverso, conversations are the space of “pre-eminence of relationship



Chapter 3.  Speaking through other voices 

and of the complicity” (1996: 13), thus underlining the importance of shared knowledge and experience. 2.3

The kind of humour evidenced in the data

In theory, an interaction such as a familiar conversation triggers a kind of humour which can appear at any time (due to the informality of the conversation), about anything (as a result of both the informality and the conviviality), very frequently (internal goal, the pleasure of being together and laughing) and finally, quite a friendly kind of humour, because of the close relationship between the participants. In fact, although conversational humour is potentially benevolent, it may also be quite aggressive precisely because of the close relationship between participants. In conversation, the facework constraints (Goffman 1967) are less important than in other interactions. It is cooperative but it may be competitive (to elicit laughter is one stake of the interaction). In sum, conversational humour is an ambivalent phenomenon, as it is friendly and aggressive, cooperative and uncooperative, it can be a bonding force in a group or exclude a member who does not appreciate or understand it. Conversation is also highly ambiguous because it is polyphonic. This ambivalence is one of the reasons which explains why, in humour studies, we encounter terms as various as irony, teasing, sarcasm and so on. 2.4

Questions of methodology

2.4.1 Humour as a generic term My study was carried out in the linguistic field of pragmatics, within a framework of interactionist and enunciative trends. I consider humour as it actually occurs in everyday conversations. I studied it to explain how it works and how it influences actual interactions and relationships between participants. Various forms of humour emerged, related to irony, mockery, and joking. These forms may be evaluated as being questionable in their taste, for instance, or more on the witty side. The aim of this paper is not to produce a graded scale of humorous enunciations, nor to classify them according to their types. This is why I shall – like the majority of researchers studying humour in interaction – adopt the term ‘humour’ in the generic sense. 2.4.2 Being a “participant observer” For the analyst to be both observer and participant (Labov 1976) might be considered as methodologically questionable. How is one to know if the analyst has not interfered with the interaction while recording? How is one to know if s/he uses too much of her/his knowledge about the participants to interpret the data rather



Béatrice Priego-Valverde

than just describe it? In other words, how can one measure the potential bias of this methodology? In this study, I play the role of a participant observer and I justify this by arguing the nature of conversational humour. Conversational humour is so contextual, so anchored in conversational history based on shared knowledge, that it is often quite impossible for an external observer not only to understand the humour produced, but indeed to identify it. 3. Theoretical framework 3.1

The double voicing theory

For Bakhtin, interaction is the central and founding element of language. The actual reality of language-speech is not the abstract system of linguistic forms, not the isolated utterance, and not the psychological act of its implementation, but the social event of verbal interaction implemented in an utterance or utterances. Thus, verbal interaction is the basic reality of language. [...] Any utterance, no matter how weighty and complete in and of itself, is only a moment in the continuous process of verbal communication. (1929 [1986]: 94–95)

Thus, any discourse must be considered in a sort of inter-discourse: each discourse is built on previous ones – what we can call intertextuality (Kristeva 1969) – and in its turn, the basis of future discourse – the conversational history (Golopentja 1988). This is the essence of the notion of dialogism as defined by Bakhtin. In 1984, Ducrot wrote a study called “A polyphonic theory of the enunciation” where, following Bakhtin’s work, he discussed the uniqueness of the subject. For Ducrot, the subject is heterogeneous and divided into two components: (i) S is the speaker her/himself; and (ii) E is the enunciator, the voice(s) speaking through the speaker. In other words, when the speaker is speaking, s/he is not alone, another/ other voice(s) is/are speaking, too. The double-voicing theory tries to identify, whenever possible, this/these other voice(s). The other voice(s) may belong to the listener (diaphony, Roulet 1985), or to a third person (present or absent). They may be identifiable or not and they may be real or fictitious (very frequent in humour). The double voicing theory of discourse leads to a double voicing approach towards humour (Kotthoff 2007; Priego-Valverde 2003, 2009). When a humorous speaker is speaking, another voice is -or other voices are -speaking as well. Considering humour as a double voicing phenomenon does not supersede any of the current theories of humour. Rather, it completes them and allows the analyst, for instance, to attribute a kind of responsibility to each different meaning of a humorous utterance, in addition to the speaker her/himself. Thus, this



Chapter 3.  Speaking through other voices 

approach emphasizes both the distance the speaker maintains from her or his own discourse and the consequences of this attitude. From this, it is possible to make the following claims (1–4). 1. The speaker-as-humorist can create a distance from the seriousness of language (play with words), from the hearer, from the situation, from her/himself (self disparaging humour), and in general, s/he can create a distance from serious reality (Bange 1986). 2. This double voicing is the materialization of a contrast between two modes of communication – one serious and the other playful – which creates humoristic incongruity. 3. Double voicing, in this case, refers to a double-coded discourse. It involves, on the producer’s side, a speaker’s ambiguous intention and an ambivalent enunciation. On the side of reception, this double-coded discourse forces a twofold interpretation which is not possible without a minimum involvement, both on the affective plane (accepting an absurd, illogical or indecent enunciation) and on the cognitive plane. 4. The double voicing is necessarily playful to some degree. It is part of what ensures the benevolent nature of humour. Connected with the distance (“what I am saying is not serious and maybe not even true”), it reduces – or indeed cancels out – all of the possible aggressive, vexing, subversive or indecent literal meaning in a humorous utterance. 4. Data analysis Let’s take a closer look at the first example: (1) Four friends are talking about their personalities and trying to define themselves. They form 2 couples (M1/F1 & M2/F2) and they have to say what they like about their boy/girlfriend. F1 is saying that what she likes about M1 is the fact that he is unpretentious. F1:  no but it’s your simplicity M2:  you’re simple-minded1 The others:  (laughter) M1:  (singing) I am called simple-minded the village simpleton (laugh) The others:  (laugh) M2:  oh god F1:  oh dear:::

1. In French, the word used is ‘simple’. It is a polysemous word which means both ‘simple minded’ and ‘non snobbish’.



Béatrice Priego-Valverde

Concerning the structure of this humorous sequence, the word ‘simplicity’, used by F1 refers to what she describes as M1’s main character trait. But in French this word has two meanings, ‘unpretentious’ (what F1 wants to say) and ‘simple-­ minded’. In other words, ‘simplicity’ is not humorous per se. It is the connecteur (Greimas 1966) of the humorous sequence, the potentially polysemic word on which M2 will rebound to produce his humorous sentence. As a polysemic word, ‘simplicity’ has two levels of meaning. One is serious and expected and one is nonserious, unexpected, and incongruous in the given context, i.e. humorous. M2 rebounds on the word ‘simplicity’ by using a disjoncteur (Morin 1966) (‘simple-minded’), a word which actualizes the second and unexpected meaning and which diverts the communication from a bona fide to a non bona fide one (Raskin 1985). While the connecteur is polysemic, the disjoncteur is polyphonic. Three voices are present in M2’s utterance. E1, the first enunciator, responds to F1 as if she said that M1 were simple-minded. It is a case of diaphony i.e. a case where M2 takes up the speaker’s purpose. Of course, it is a false diaphony because F1 never meant to say that M1 was simple-minded. In other words, E1 is a fictitious enunciator. E2 corresponds to a false M2, one who may believe that F1 said that her boyfriend is simple-minded. In this case, too, E2 is fictitious. And finally, E3 is present. He is a facetious M2 who is laughing about the situation. Thus, if three voices – two fictitious and one facetious – are present in M2’s discourse, who is really speaking? Actually it is impossible to determine if M2 is hiding behind the voices he invokes or what he really thinks. Is it just a play on words or a true case of mocking M1? Whatever the answer, M1 accepts being teased and he shows it by overbidding M2 in singing an old French song. Singing this song, M1 produces an intertextual sentence in order to produce self-disparagement humour where he presents himself as a simple-minded. Finally, everybody is laughing and M1’s face is saved. Below the second example: (2) F2 is talking about the dog she offered to her parents. Such a cute puppy became ugly as it grew. F3 (M3’s girlfriend) is pregnant. The first “that” refers to the dog. The second one refers to the future baby. F2:  but we didn’t know how that would turn out M3:  same thing for us + we are expecting the worst F3:  here we are F2  (laughing) oh yes (laugh) M3:  The others:  (laugh) M3:  that will be horrible



Chapter 3.  Speaking through other voices 

F2’s first sentence refers to the dog she and her sister gave their parents for Christmas. The pronoun ‘that’ refers to the dog and she wants to say that they did not know how the dog would turn out when older. M3 rebounds on F2’s sentence in order to make a comparison between the situation F2 is talking about and his own situation: he and his pregnant girlfriend are expecting a baby. The linguistic process used is very well known in humour studies and this comparison is a conciliation des inconciliables2 (Aubouin, 1948: 65), i.e. an incongruous comparison between a human being and an animal. We can consider that M3 is producing selfdeprecating humour here. Do we reasonably think that M3 truly believes what he is saying? F3’s brief comment and the others’ laughter show that M3’s humorous sentence is not taken seriously. In other words, the other participants do not believe in M3’s responsibility of his purpose. So, who is really speaking through M3? We can consider that we have two different voices here. A facetious M3 who is laughing at the situation and a fictitious M3 who is pretending he believed what he was saying. However, as he has the assurance that his friends are taking what he is saying as humour (they are laughing), he can even overbid (“that will be horrible”). We continue the analysis with (3): (3) The participants are talking about an internet class the 2 women have taken. F1 & F2 are close friends. F2 & M2 are a couple. F2 is Jewish. F1: yeah but it’s interesting F2: what is interesting ↑ M2: interesting (+) qualifying adjective meaning to have interest for F1: interest (+) masculine noun M2: interest (+) for the Jewish (+) bait for benefit F1: (roar of laughter) M2: (laugh) F2: (laugh + gesture of denial)

This excerpt is very interesting because it shows the importance of the relationship not only to produce humour but also to feel authorized to produce it. The excerpt begins with a sentence by F1 saying that she was thinking that the class she and her friend F2 took was interesting. At that point, F2 asks “what is interesting?” which is, incidentally, a strange question if interpreted as a request for a definition (because it is just impossible to think that F2 does not know the meaning of this term). F1 and M2 do not seem to know which meaning to give to such a question and they seem surprised. M2 pauses, looks at F1 and begins to perform a recitation 2. the conciliation of unsympathetic elements (my translation)



Béatrice Priego-Valverde

of the dictionary to tell F2 the meaning of the word ‘interesting’. F1 overbids and goes into the game with M2. At that point, F1 and M2 pretend that F2 did not know the word’s meaning, as if she were too ignorant or uneducated to understand. First, they create a false image of F2 (as a stupid person) and then, they start speaking to this false F2. So, what they are saying cannot be taken seriously and cannot be attributed to them. Actually, F1 and M2 are invoking various enunciators as they speak: E1, a common facetious enunciator who corresponds to the feigned speakers, E2, a common fictitious enunciator who corresponds to an F1 and an M2 who believe that F2 is stupid and E3 a fictitious, racist enunciator who utters cultural stereotypes against Jews. Given the relationship between F2 and M2 nobody – not even F2 – can believe that E3 is real and corresponds to M2; the fact that they all laugh is evidence of this. Here, the hide-and-seek game between the speakers and the other voices they invoke is perceptible thanks to two elements. One is the image the speakers create of F2 – a stupid woman – which cannot be taken seriously (nobody can reasonably think that F1 and M2 really believe that F2 is so gormless). The other element is the nature of the relationship between F2 and M2: as they are a couple, it is impossible to think that M2 can have anti-Semitic prejudices. Let’s take a look at an additional example below: (4) The six friends are playing to a mime game. One player has to mime a word taken at random and the others have to guess what the mime means. In the next example, F2 is miming and it is her boyfriend who finds the solution. M2:  to pierce F2:  yes:::: M3:  goddammit M1:  goddammit M3:  they understand each other very well F2:  (laughter) well I don’t know M1:  they are complementary F2:  7 years of experience will always make the difference (laughter) The others:  (laughter)

According to the various reactions (especially M1 and M3’s), all the participants seem to be surprised by M2’s success. To produce her humorous sentence, F2 rebounds on M3’s utterance which describes the osmosis between F2 and M2. This osmosis allows them to find the right answers. F2 explains their understanding of each other: they have been living together for seven years. But she says it humorously, using a commercial slogan for a chocolate spread3. She reuses a very famous 3.

The original one is «Thirty years will always make the difference».



Chapter 3.  Speaking through other voices

old slogan, adapting it to make it appropriate to the fact that she has lived with her boyfriend for seven years. This process is a mark of intertextuality, the integration of a discourse into another discourse. Two voices are thus speaking through F2: her own voice and a voice which is not really identified but which more or less corresponds to the first producer of the slogan and all those who can use it. Let us take a closer look at (5): (5) The participants are watching a television documentary about dogs. F2, who knows a lot about dogs, explains that if the dogs slobber, it is because their owners feed them between regular meals. F2: well + it seems that dogs that slobber + it’s because you feed them uh:: + between meals you see + you give them the habit when you eat to give them things + so it accentuates the slobbering + and if you don’t give them the habit + they don’t slobber + so it’s sure that if the dog does that there is slobber all around // M2: ah that’s the reason why:: M1: it’s the reason why you slobber M2 M2: it’s the reason why you don’t want me to eat between the meals (Everybody laughs) M2: I slobber too much (laugh + the others laugh)

In this excerpt, F2 explains why dogs slobber too much. M2, her boyfriend, interrupts her and, in overlapping, begins a humorous sentence (“it’s the reason why”). M1, who has probably guessed M2’s humorous intentions, consisting in comparing himself with the dogs, anticipates (in overlapping) the utterance and compares M2 to the dogs (“it’s the reason why you slobber M2”). Here, M1 proceeds by a conciliation d’inconciliables comparing animal with human behaviour. This completely incongruous comparison is possible and acceptable only if it is taken humorously, with a minimum of distance. This distance is materialized by a double voicing, i.e. by the using of two different voices: E1 which corresponds to a facetious M1 who is teasing M2, and E2, a fictitious enunciator who believes that M2 is slobbering and who gives the reason for such behaviour. In the next utterance (“It’s the reason why you don’t want me to eat between the meals”), M2 follows his own previous utterance interrupted by M1 and finishes it, also (as M1) comparing his own behaviour with an animal’s but without making it too explicit. With his self-deprecating humour (he shows a disgusting image of himself) he gives a false reason why his girlfriend does not want him to eat between meals. Indeed, if it is true that F2 does not want him to eat between regular meals, it is because of his weight and not because he slobbers. By giving such an incongruous reason, he creates various enunciators speaking through him: E1, the still facetious one who laughs at the situation, E2, a fictitious F2 who





Béatrice Priego-Valverde

forbids M2 to eat because he slobbers, and E3, who corresponds to a false M3 who is slobbering. 5. Conclusion In this study, I have shown that conversational humour is a highly complex phenomenon. The two main reasons are: (i) the fact that conversational humour is highly contextualized making it sometimes difficult for external observers to perceive and (ii) it is a polyphonic phenomenon. Indeed, when a speaker produces a humorous utterance, we can say that s/he is not alone: other voices are speaking through her/him. Here, I have found up to three different enunciators or voices. The first one, which is identifiable, always corresponds to the speaker and it is a facetious enunciator laughing at the situation s/he creates. The other(s) is (are) hard to identify because they are often fictitious and correspond(s) to a false speaker or hearer, i.e. the image of a speaker or hearer specifically created to make people laugh. What makes conversational humour more complex is not only the fact that it is a polyphonic phenomenon, but, rather, that the speaker plays hide-and-seek with the various voices s/he invokes. As if s/he is playing around in order to cloud the issue. As if s/he does not want the hearer know what s/he really thinks. It is as if the speaker does not want to show whether what s/he is saying is serious or not. Appendix I: Data in French (1)

F1:  non mais c’est ta simplicité d’esprit M2:  t’iés simple les autres:  (rires) M1:  (il chante) on m’appelle:: simplet l’innocent du village:: (rires) les autres:  (rires) M2:  (rires) oh fan F1:  oh:: là là:::↑

(2) F2:  mais on savait pas comment ça allait devenir M3:  ben nous c’est pareil hein (+) on s’attend au pire F3:  ben voilà



Chapter 3.  Speaking through other voices 



F2:  (en riant) eh oui (rires) M3:  les autres:  (rires) M3:  ça va être horrible

(3)

F1:  ouais mais c’est intéressant F2:  c’est quoi intéressant ↑

F1:  intéressant // M2:  intéressant (+) adjectif qualificatif que signifie avoir de l’intérêt pour F1:  intérêt (+) nom masculin M2:  intérêt (+) chez les Juifs (+) appât du gain F1:  (éclat de rire) M2:  (rires) F1:  (rires + signe de dénégation) (M2 et F1 continuent à rire)

(4)

M2:  transpercer F2:  oui::: M3:  putain M1:  con= M3:  i se comprènent bien tous les deux F2:  (rires) ben:: “chepa” // M1:  i sont complémentaires F2:  sept ans d’expérience feront toujours la différence (rires) (rires des autres)

(5) F2:  ben (+) i paraît que les chiens qui bavent (+) c’est pace que::: tu leur en/donne à bouffer heu:: (+) entre les repas tu vois (+) tu les habitues quand tu bouffes à leur filer des trucs (+) alors ça accentue la bave (+) et si tu les:: habitues pas (+) i bavent pas (+) alors c’est sûr que si i fait ça et qu’ien a partout // M2:  ah c’est pour ça que:: M1:  c’est pour ça que tu baves M2 M2:  c’est pour ça que tu veux plus que je bouffe entre les repas (rires de tout le monde) M2:  je bave trop (rires + rires des autres) F1:  parce que quand tu tu baves: (rires + rires des autres)





Béatrice Priego-Valverde

Appendix II: Conventions of transcription F/M Female/male with same number indicating same couple (F1, M1), (F2, M2) : Vocalic lengthening. The number of: symbols is proportional to the duration / Self-interruption of the discourse // Interruption by another speaker (+) Pause. Number of + symbols is proportional to the duration ↑ High intonation after the syllable concerned ↓ Low intonation after the syllable concerned = Fast speech after the word or syllable concerned ( ) Into brackets: description of behaviour (in italic) Observer’s commentary or interpretation Doubts about the interpretation Hesitation between two possible words Inaudible word or sequence NON, BONjour Increased word or syllable pas-du-tout To speak haltingly Underlined words: overlaps

part 2

Plural identities and viewpoints in acquisition and language learning

chapter 4

The self as other: Self words and pronominal reversals in language acquisition Aliyah Morgenstern

Université Sorbonne Nouvelle – Paris 3 The study of reversals may shed some light on the problems children encounter when they try to use the formal, linguistic marks to convey the distinction between the self and the non-self. In two longitudinal case studies of French speaking boys age 1;08 to 3;03, I investigate how they use self-reference. I first focus on a particular use of pronominal reversal in Léonard’s data: the little boy’s use of the third person to designate himself. In a second longitudinal case study, we investigate how Guillaume uses the form tu (‘you’) instead of the first person pronoun. They both speak of themselves with the others’ voices in contexts in which they have either done a misdeed (third person) or an exploit (second person). In those two typical developing children, pronominal reversals seem to occur when they begin to assimilate and internalize the representations the parents formulate regarding their child’s experiences.

1. Introduction Very early on in their lives, children can assimilate representations of themselves as verbally proposed to them by adults and eventually use them in their own discourse. It is particularly interesting to study self-reference as a “window to the child’s emerging self-concept” (Bates 1990, 165) constructed in and thanks to social interaction. In this chapter, I focus on subject self-words and pronouns with their shifting reference (Jespersen 1921, Jakobson 1963, Benveniste 1966). When we use the term ‘subject’, we can refer to four distinctive notions: (1) the grammatical subject, (2) the semantic subject, (the agent for example), (3) the topic, (4) the speaker. Referring to the self in the first person seems to be very complex at the theoretical level. The idea has founded an entire linguistic tradition ever since Emile Benveniste’s famous writings about the ego (1966). We can therefore wonder how young children are able to join the four levels in a single marker, which in French is je (‘I’). However, this acquisition process is not salient, or catastrophic in Thom’s sense (1977). Parents



Aliyah Morgenstern

do not remember the first day their child used I, although they will remember the first use of mummy, daddy, or certain linguistic phenomena which they find striking. However, as we were doing research focused on French children’s self-words, we noticed that for a short period some children used tu (‘you’) and of il (‘he’) instead of je (‘I’) (Morgenstern and Brigaudiot 2004). This phenomenon is referred to in the literature as pronominal reversal. It is a rare phenomenon, but it does seem to play a significant role in the mastery of the pronominal system. It can help us understand how children deal with shifting reference and come to understand the problems they encounter when they try to use formal linguistic marks to convey the distinction between the self and the non-self. Pronominal reversal is also frequently observed in language disorders, especially in autism (Bettelheim 1967). The study of reversals as a normal development, albeit infrequent, may therefore also contribute to the understanding of pathological language development. Previous hypotheses on children’s use of self-words will be discussed followed by theories about pronominal reversals. I will then present analyses of the data of two French little boys. Léonard occasionally uses the third person instead of the first person for a short period. Guillaume (around the same age) sometimes uses the second person instead of the first person. 2. Children’s self words Between the ages of 18 and 30 months, various markers are used by French- and English-speaking children to refer to themselves. (Brigaudiot, Nicolas, Morgenstern, 1994, Budwig 1995, Morgenstern 1995, 2006) – Ø form (zero form), or absence of form. Speaking of oneself is implicit. The speaker relies on a shared situation and background with the interlocutor. In such a context, it would also be interesting to study sign languages because there is no need to mark the grammatical subject when it is the speaker and when there is no contrastive agency. The signer’s body is the grammatical subject, thus when the signer and the grammatical subject coincide, there is no need for a grammatical marker. – The use of the child’s name, which is not correct in adult language and has been described as referring to the “social self ” (Bain 1936, Cooley 1908). – The accusative in French (moi), but also the genitive in English (my) (Budwig 1995, Morgenstern 2003a). – The second or third person (Morgenstern 2003b). Between two and a half and three, the use of je/I becomes stabilized and the other markers tend to disappear in the subject function. At the same time, children stop



Chapter 4.  The self as other: Self words and pronominal reversals in language acquisition 

resorting to a strictly deictic use of language and manipulate different tenses, aspects, modalities (Nelson 1989). The following questions could therefore be raised. – How could we account for the simultaneity of these forms for a given period of time (in our data from 2;3 to 2;8)? – Are these forms used randomly or do they have different values according to the context they are used in? In previous work (Morgenstern 1995, 2006), I assigned two main values to selfwords produced in discourse by children between 18 months and 3 years old. 1. The referential value, focusing on who performs the action (there is a contrastive agency). For example, a child says, “moi fait la photo”/[I’m the one who films (takes the picture)], as he is trying to take the camera from my hands. The self-word is used in context with a contrastive value. The child uses the forms name and moi to express this value (similar to Budwig’s “high transitivity” value 1989). 2. The modal/expressive value. The child under analysis expressed his desires, his will, his projects, his points of view as in “veux zouer à la pâte à modeler” [I want to play with clay]. At first, the child used bare predicates, then preverbal vowels (fillers) and little by little, the filler [ø] became [yø] then [zø] and finally, around three, je. These self-words can be seen as subjectivity indicators. The child considers the situation he is in, does not find it satisfactory and uses language to express what he would like to substitute for the reality he has before him (Danon-Boileau 1994). He therefore produces utterances such as: “j’entends pas maman, tu peux mettre plus fort” (I can’t hear Mum, could you make it louder?). But he will also give his opinion “je trouve qu’elle est jolie cette musique” [I find this music melty]. I also found a difference between the way self-words were used in dialogic interaction and in narratives. Léonard used the third person ‘i/il’ for a short period to refer to himself in narratives. In order to tell a narrative, children must verbalize internal representations consisting in images, words and a theatrical development. They mentally view the scenes in which they have played a major role and therefore see themselves as a character among others. That might be a reason for their use of the third person just before two and a half years old. The language addressed to the two children in our study is particularly relevant since Léonard’s mother uses the third person in contexts in which she makes up stories about him (she happens to be a writer). Guillaume’s mother never uses the third person when she speaks to him. Léonard makes fairly frequent use of third person pronouns during a few months to refer to himself, while Guillaume uses the second person quite frequently in this function. We will examine which

 Aliyah Morgenstern

contexts the two children use these forms in; at first they seem quite deviant. I will try to show how by resorting to alterity (second or third person instead of first person), the child represents himself differently and asserts his own identity through other standpoints, other perspectives, other “voices” (Bakhtin 1986) in dialogic contexts. 3. Pronominal reversal Pronominal reversal is rare in children whose linguistic behaviour is typical, but it is quite frequently mentioned in the literature. Studying the errors children make in the output may be a way to raise “specific questions whose answers may shed light on the mechanism of pronoun acquisition” (Chiat 1989: 383). Besides, pronoun reversals are frequently observed in language disorders, especially in autism, since it is part of Kanner’s original description as well as Asperger’s. “Personal pronouns are repeated just as heard, with no change to suit the altered situation (...). Not only the words but also the intonation is retained” (Kanner 1943: 244). The initial attempt to explain the phenomenon was made in terms of echolalia and of autistic children’s difficulties in recognizing people as centres of subjectivity and as the occupants of reciprocal roles in discourse given that the use of personalreference terms varies depending on the speech role one occupies in the conversation. The study of reversals as a normal development (even if infrequent) might therefore contribute to the understanding of pathological language development. Many hypotheses have been proposed to explain reversals. These include (a–c). a. The name hypothesis: the idea that pronouns are used as names (Clark 1978). b. The person hypothesis (Charney 1980); a given pronoun is taken to refer to a constant person or set of people, leading to pronoun reversals. c. The risk-taking hypothesis: precocious children take the risk of using pronouns and fail to perform a deictic shift (Dale & Crain-Thoreson 1993). The above hypotheses involve the following factors, which can account for reversals. – Lack of semantic knowledge. Not knowing which words are pronouns (Bellugi & Klima 1982). – Simple imitation of the speech heard. “I’ll help you” uttered by a child may contain two pronominal reversals and the child may be asking for help which would be a consequence of reliance on imitative, holistic strategies of language learning. – Not understanding perspective shifting (Loveland 1984).



Chapter 4.  The self as other: Self words and pronominal reversals in language acquisition 

– The nature of the input. It could be impoverished (Oshima-Takane 1992) and the child needs to observe speech directed to others containing personal pronouns. All these interpretations stem from the fact that children do not use pronouns the way adults do, corresponding to speech roles. In my data pronominal reversal is not consistent, and always occurs along with adequate usage. So we found the “perspective-shifting hypothesis” proposed by Chiat (1981) highly interesting. It claims that such “errors” would be a deliberate use of reversals to shift mental perspective. In her study of Matthew’s reversals, Chiat argues that this perspective-shifting hypothesis cannot be mistaken for the mere imitation interpretation, since reversals occur in novel constructions. She shows that “many of the reversals in both directions occur in contexts where the addressee’s point of view is being expressed (...) reversed ‘you’ and ‘your’ occur in references to himself that Matthew would hear from other people: in utterances of threat, warning, accusation, prediction, or permission” (Clark 1974: 377). This hypothesis is in contradiction with authors who say that children cannot take the other’s perspective and therefore do not comprehend the reversible nature of pronouns. According to Pettito (1987), up to a certain age, children can only refer to themselves since their language is egocentric. But Piaget (1962) did answer Vygotsky’s criticism (1934) of the notion of egocentric language and proposed the word ‘centrism’, meaning that the child cannot differentiate his perspective and the perspective of others. Oshima-Takane (1992) explains that children understand the relation between pronouns and speech roles when they see two speakers talking to each other and become aware that the second person pronoun refers to the addressee. She therefore emphasizes the importance of vision (also see Oshima-Takane, Cole and Yaremko 1993). A few studies on the acquisition of pronouns by blind children conclude that there is some delay in the use of the first person pronoun and a lot of reversals (Fraiberg and Adelson 1973, Sampaio 1989). Loveland (1984) also insists on the spatial aspects of pronoun acquisition and the importance of seeing different speakers in the activity of dialogue in order to understand perspective shifting. However, Perez-Pereira’s thorough study does not support the claim that blind children by and large use personal-reference terms comparatively late and with a great deal of reversal errors. Pérez-Péreira does agree with Oshima-Takane on the implication of the “failure to observe pronouns in speech addressed to another person” and also mentions the impact of the “large proportion of directives and requests used by mothers” (1999: 677), which may prevent some blind children from using pronouns correctly. We can conclude from those controversial studies that blind children are probably able to compensate their lack of vision by the fact that they can hear and distinguish speech which is not addressed at



Aliyah Morgenstern

them. They must develop the quality of internally locating speech partners in space from what they hear. But that capacity depends on a child’s environment and development; some blind children are hindered in their pronoun acquisition while others are not. For Chiat (1982) pronouns are plurifunctional, and this has to do with the possibility of adopting another’s point of view. Adults occasionally do this explicitly by saying, “I wouldn’t do that if I were you”. She illustrates her perspective-shifting hypothesis with examples such as: C – When you’re (speaker) a big boy you (speaker) can go to play tennis (p. 372). Not only is the child using ‘you’ to refer to himself, expressing the addressee’s point of view and shifting perspectives, but he is speaking for the adult. The child might be uttering this mixture of prediction and permission with an evaluation of her/his future capacities just because that is the kind of utterance s/he usually hears in the same type of situation. The child says what he expects to hear or what should be said in the present situation. If this is the case, we might not consider the use of the pronoun ‘you’ as a reversal and as being the incorrect form, but think of the whole sentence as being uttered by the wrong speaker. In fact, the utterance could be what the child expects – or would like – to hear in the present situation. A longitudinal study of Léonard and Guillaume’s use of the second and third person pronouns for self-reference, will help to analyze this possibility. 3. The third person The data I used for this study consists of the video recordings and lined transcriptions of Léonard, age 20 months to 39 months, which are part of the Paris Corpus on the CHILDES database (Mac Whinney 2000, the data was analyzed in Morgenstern 2006). Each self-word used in subject position by Léonard during this period of time was counted and analyzed. Three periods could be distinguished in the data: I non adult uses 1;7 to 2;2; II All self-words 2;3 to 2;7; III Adult uses mainly 2;8 to 3;3. Table 1.  Léonard’s self-words Period 1 1;7 to 2;3 No adult form  Ø, filler, name, moi, il

Period 2 2;2 to 2;7 All forms  Ø, filler, name, moi, il, moi il, je, moi je

Period 3 2;7 to 3;3 Over 80% adult forms   je, moi je



Chapter 4.  The self as other: Self words and pronominal reversals in language acquisition 

Table 2.  Percentage of Léonard’s self-words AGE Period 1 1;8–2;2 Period II 2;3–2;7 Period III 2;7–3;3

Ø

filler

moi

il

name

name+ il

je

moi je

32

46

10

5.2

3.6

2

1.2

0

26.75

29.25

5.5

21.5

0

0

11.75

5.25

11.25

2.75

4.5

2

0

0

59.5

18.5

The percentage of the forms used in the three periods is shown in Table 2. During the second period, Léonard uses all the forms, in particular ‘il’. He talks about himself in the third person, marking the discourse object without identifying it to the speaker. That amounts to over 20% of his self-words during this period, so it is quite an important phenomenon. Of course, only a detailed analysis of the data can allow us to understand whether it is used randomly or if it has a specific function. The analysis of a sample from the data will enable us to understand Léonard’s use of self-words and in particular his use of the third person in autobiographical narratives. Léonard’s father asks him a question about what he did in nursery school. This question shows the child that his parents need information that only he knows; they did not witness the events that took place at school. The child is therefore drawn into the narrative mode. See the example below: (1) P: L:

Qu’est-ce que t’as fait à la crèche aujourd’hui? [What did you do in nursery school today?] atata lakewlafelafe/lapatamodle (axx la crèche, l’a fait, l’a fait, la pâte à modeler) [In nursery school I did, I did modelling clay]

Fifteen minutes later, as he is getting ready for his bath, Léonard is naked in his mother’s arms and is a little aggressive with her: she kisses his arms and he wipes away the kisses with his hand and says “no kiss on my arms”. See (2) below: (2) L: ladipã adavib (l’a dit pan à David)/[he said pan to David] M: T’as dit pan à .../[you said pan to...] L: ladip ladipã adavib (l’a dit p... l’a di pan à David)/[he said pan to David]



Aliyah Morgenstern



P enters the bathroom. P: T’as dit pan à David?/[you said pan to David?] L: wi/yes M: David de la crèche? [David from nursery school] L: leonaila/leonaadip ladipã adadib (Léonard il a, Léonard a dit p... l’a dit pan a David)/[Leonard, Leonard said pan to David] M: Pourquoi?/[why?] L: wi/[yes] M: Pourquoi t’as dit pan à David? David il t’avait dit pan aussi?/ [Why did you say pan to David? Did he say pan to you, too?] L: pabodadibpaboadadib (pas beau David, pas beau à David)/[not nice David, not nice] The child is going to combine the narrative mode and his aggressive mood by reinitiating his entry into the narrative from a visual and gestural association. He gives the impression he is re-living the scene and punctuates his speech with violent gestures. He tells the story by embodying the scene, helping his parents to visualize what actually happened. This distance between the child who is about to take a bath at home and the child in kindergarten is marked by the use of the third person. Besides, since the little boy was filmed, we were able to see that when he produces his utterance, he does not look at his interlocutor. His gaze is vague as if he were concentrating on the scene he is visualizing in his mind. That reminds us of Cuxac’s analysis of narrative in French Sign Language (Cuxac 2000): when the signer is engaged in a narrative, he does not look at the interlocutor, he is not embodying his own self, but a character in a story, as opposed to discourse, in Benveniste’s terms (1966). Then there is a clarification of the identity of the two protagonists thanks to the mother’s question: “David de la crèche?” [David from kindergarten]. Léonard does not answer directly but localizes another actor and uses a name, his own name. L.: leonaila/leonaadipaadadib. [Léonard, (he) said pan to David] Léonard uses his name to talk about the agent once he needs to mark the various places and roles of the characters in the scene. As a narrator, he shares his parents’ point of view and relates the scene as if he were a mere observer; he names himself with his name as if someone else were narrating the scene. This identity splitting enables him to distinguish between the world of kindergarten and the world of home. As he narrates, he also mimes the scene, which creates a link between Léonard as speaker and Léonard as actor. The body of Léonard the speaker



Chapter 4.  The self as other: Self words and pronominal reversals in language acquisition 

(including his arm hitting in the air) enables him to show his mother the character Léonard in his narrative. This is similar to what happens in narratives in Sign language; the signer’s body becomes the character of the story, which is called personal transfer. I systematically studied all of Léonard’s uses of the third person to refer to himself and I observed that he only uses il when he is producing some kind of narrative which implies the wording of a cinematographic projection of the events which unwind in his head. The images he sees arouse the same feeling of otherness as the image of oneself in a photo or in a mirror. And we know that during the same period, children do not refer to themselves by saying je or moi when they talk about their image but by their name or the third person (Zazzo 1993). During this period, Léonard uses the third person instead of the first person to construct a character, an Other, who is elsewhere and is naughty; he says pan to David, he tears the book to pieces, he eats up all the cakes. Guillaume, the second little boy presented in this study, happened to use the second person for a few months to refer to himself. The following part focuses on his productions of pronominal reversals. 4. The second person Guillaume1 was recorded every two weeks from his birth to the age of six. The transcription was made by his mother, immediately after each recording so as to have as much information as possible, and was checked by a second researcher. His mother also kept a diary (Brigaudiot & Nicolas 1990). For this study we selected only part of the data from 1;72 to 2;10, from the time both the forms elided t’ and tu appear in formulaic expressions, as in “t’entends” [you hear?] or “tu vois” [you see] to the stabilization of the use of tu to refer to the addressee. We counted and analyzed each self-word used in subject position by Guillaume during this period of time. We also distinguished three periods in the data: I non adult uses 1;7 to 2;2; II All self-words 2;3 to 2;7; III Adult uses mainly 2;8 to 2;10. Table 3 presents the percentages of self-words according to the period.

1.

The author thanks Mireille Brigaudiot for her collaboration in this study.

2. The notation for ages is number of years; number of months. Example: 1;7 = 1 year and 7 months.



Aliyah Morgenstern

Table 3.  Percentages of Guillaume’s self-words AGE Period 1 1;8–2;2 Period II 2;3–2;7 Period III 2;7–3;3

Ø

filler

moi

il

name

name + il

je

moi je

32

46

10

  5.2

3.6

2

  1.2

0

26.75

29.25

5.5

21.5

0

0

  11.75

  5.25

11.25

  2.75

4.5

2

0

0

59.5

18.5

The second period is the one I am interested in for that is when the child uses tu (you). It accounts for 12% of the uses of self-words in the data for that period. Example 3 is an extract in which various subject self-words are used in the same sequence. I’m gonna mix the flour. Done, (3) Guillaume 1: Ze vais tourner la put everywhere. You mixed, you farine. Ça y est, mis partout. T’as mixed the flour! tourné! T’as tourné l’œuf!! Mother 1: Bien! Good! Guillaume 2: Ah ça colle. Ze veux Oh it’s sticky. I wanna mix some encore tourner un œuf. Encore un. more egg. Another one. Mother 2: Tourne amour, tourne. Mix honey, mix. G. looks at his mother. Guillaume 3: Ai tourné. I’ve turned. Mother 3: Maintenant il faut encore Now we need another egg. un œuf. His mother is about to break the egg. Guillaume holds out his hand. Guillaume 4: Guillaume i fait casser un Guillaume he does it/breaks an œuf!!! egg All occurrences of tu (you) in the data were coded and analyzed whomever they referred to. As shown in Graph 1 there are two peaks, one at 2;4 and then after a decline, the use of tu rises again and becomes more or less stable. We categorized those occurrences according to whether they were adult uses (tu meaning the addressee) versus non adult uses (tu meaning the child) as shown in Graph 2.



Chapter 4.  The self as other: Self words and pronominal reversals in language acquisition  18 16

Percentage

14 12 10 8 6 4 2 0

1;7

1;8

1;9 1;10 1;11 2;0

2;1

2;2

2;3

2;4

2;5

2;6

2;7

2;8

2;9 2;10

Age

Graph 1.  Percentage of tu (over number of utterances) in Guillaume’s data

25 20 ?

15

tu = speaker tu = other

10 5 0

1;7 1;8 1;9 1;10 1;11 2;0 2;1 2;2 2;3 2;4 2;5 2;6 2;7 2;8 2;9 2;10

Graph 2.  Guillaume’s uses of tu

At this point we can make two important claims. 1. Reversals only occur once the second person form is already part of the child’s productions and used adequately but in formulaic expressions. 2. Reversals might be a necessary stage for that particular child before the use of adult like tu (referring to the second person) in non-formulaic utterances. In the following we will examine a few extracts in detail: (4) Guillaume 2;3 G. is making the inventory of all the shoes in the house and he makes comments. G: C’est à papa/[It’s Daddy’s] M: C’est les souliers de papa/[It’s Daddy’s shoes] G: Peux, veux à mettre, veux./[Can, want put, want] M: Tu veux les mettre?/[You want to put them on?]



Aliyah Morgenstern

G: Oui he puts on his father’s shoes comme ça marche/[Yes. Like that walk] M: Hein?/[What?] G: he stands up with the shoes on Marcher comme ça/[Walk like this] M: Tu veux marcher comme ça?/[You want to walk like this?] G: Ouais he takes a few steps bravo tu marches!!/[Yea. Congratulations! You’re walking] G. is making the inventory of all the shoes in the house. He tries on his father’s shoes, takes a few steps and says, “bravo tu marches” [congratulations you’re walking!]. Once he has started to accomplish this, we find tu as the marked form. Does this tu mean ‘I’? Is the child really the one who is supposed to congratulate himself? Is he the right speaker for that utterance? Shouldn’t his mother be saying it? We can consider the situation as being an accomplishment (from the child’s perspective at least). This utterance is very similar to previous utterances that have been addressed at him in comparable situations. The child could be expressing his own pride by using an utterance usually produced by his mother in similar situations, showing he has internalized previous dialogic situations and adopting a form of discourse in which he expresses another’s perspective on himself (Fernyhough 2008). That enables him to speak about himself as if another person were addressing the utterance to him. We can therefore consider this tu as a device to mark identity and otherness. It is very interesting to notice that the mother continues the dialogue without asking anything about this use of tu. She understands the tu as referring to the child. (5) Guillaume (2;05) has eaten a peanut G: t’as avalé encore!/[You swallowed again!] M: non, une seule cacahuète, pas tout/[No, only one peanut, not all of them] Here the child is misbehaving, and what he is doing could actually be considered dangerous. The tu is the same one the child uses when he is running on the pavement and stops just before crossing the road, and says “tu traverses pas” [don’t cross]. It can also remind us of when the little child says ‘no’ or shakes his head as he is about to touch a dangerous object. We can see that the child is producing an utterance out of a fixed scenario. He uses his auditory memory with a situation associated to a sort of quotation. He therefore applies the formula ‘tu + predicate’ to a situation. It is as if he were snatching the mother’s utterance out of her mouth, as if he were borrowing her role, her place in the dialogue, as if the most important was not who the speaker is but that the utterance should be spoken. The script exists. He does not create an utterance, but uses it because it applies to the present situation.



Chapter 4.  The self as other: Self words and pronominal reversals in language acquisition 

Here again, his mother has continued the dialogue without asking who this tu was. But there comes a day when the clarification of the reference becomes an issue. And this moment coincides with metalinguistic play. Let us consider the next example: (6) Guillaume 2;7 (last ‘tu for je’ in the data) G and M talk about absent toys. M: y’en a beaucoup! i sont de quelle couleur?/ [There are many of them! What color are they?] G: bleu/[blue] M: oh c’est pas vrai/[ah that’s not true] G rit: t’es menteur!/[you’re a liar!] M: t’es un menteur, ah ah ah, t’es un menteur! i sont pas bleus i sont rouges!/ [you’re a liar, ah ah ah, you’re a liar! They’re not blue, they’re red!] G: i sont rouges, menteur!/[they are red, liar!] M: petit menteur, tu joues hein ma puce/[little liar, you’re playing, aren’t you darling] G: t’es un petit menteur/[you’re a little liar] M: t’es un petit menteur/[you’re a little liar] G: maman t’es un menteur/[Mummy you’re a liar] M: ah moi j’suis pas menteuse non. (...)/[oh, I’m not a liar, I’m not] G: non, papa i dit p’tit menteur papa, après idit p’tit con/[no, Daddy says little liar and after he says little asshole] The quotation “t’es un petit menteur” [you’re a little liar] is playful. The mother has just told the child he is not telling the truth. The expression “t’es un petit menteur” is teasingly manipulated as the mother points out “tu joues” [you’re playing]. The child even compares the mother’s speech with the father’s. He now has enough distance with reported speech to play with roles and perspectives. Our analysis points to two levels of pronominal reversal in Guillaume’s data: a. During the first period, this special tu is used in an utterance that is a sort of echo of speech previously addressed to him. The situation the child is in reminds him of a previous experience (or of several previous experiences) during which a similar utterance was produced. Therefore, he makes a comment using the second person pronoun and expressing his own pride or shame as if he were his mother congratulating or scolding him. We can call this an assimilation process; the echo is applied to the current event with no distance at first. But that process is fundamental in the construction of a theory of mind, it is a first step. It reminds us of what autistic children do when they reverse pronouns.



Aliyah Morgenstern

b. There is a second level in which the echo is transformed into metalinguistic play. It becomes similar to reported speech. The mother plays this metalinguistic game with him because she intuitively knows he has a theory of mind. There is the necessary distance. We see Guillaume use a kind of false speech; he is joking, he is playing with his mother and deliberately manipulating an utterance she must have addressed to him. After that stage, all uses of tu for ‘I’ in older children and adults will be metalinguistic and deliberate in one way or another. It may be the case that autistic and psychotic children do not reach that second level, and stay at the level of assimilation in which their speech sounds like an echo of speech addressed to them, in a seemingly similar situation, without their having fully appropriated the content of what they say about themselves. But even if they do not assert themselves as authors of the content, their use of the other’s perspective is a first step towards internalization of another’s speech. After that period of pronominal reversal, in our data, Guillaume always uses je to refer to himself in dialogue in subject position. 5. Conclusion Léonard and Guillaume use the second and the third person instead of the first person when they need to separate the agent and the speaker. They give the impression that another person is speaking and it enables them to present themselves with another voice at certain moments, extraordinary moments, when they have been either villains or heroes and to unveil other facets of their selves. This could be glossed by saying, “I’m talking about myself as if it weren’t me talking”. The self considered as other, seen through the eyes of another. This all happens at a nonconscious level and it is all solidly anchored in reality since the child is talking about feats or naughty behavior he has actually carried out. This hypothesis emphasizes the role of addressed speech (Perez-Pereira 1999) and of the care-giver as “self regulating other” (Stern 1985) throughout language acquisition. But there is a sequel to this splitting up of their self. A few months later, the children in this study invented an alter ego to which they gave another name in order to be able to speak about another self. Around the age of 2;05, Léonard abandons the third person to refer to himself but he makes up the character of Jean-Patou and refers to him as the author of all his misbehaviour, “Jean-Patou broke the little bed because he jumped on it so much, he tore up the book because he didn’t like the story...” Guillaume invents the character Rouda who is his hero and does all the great things he would like to do. So the children talk



Chapter 4.  The self as other: Self words and pronominal reversals in language acquisition 

about another individual who is their other self, the good or the bad self. They create a fictional character, an alter ego who can be either naughty or extraordinary. This phenomenon has already been mentioned in the literature and is often reported by parents (Astington 1993). The two children kept their imaginary friend for a few months (Léonard), and over a year (Guillaume), and then gave them up. At a certain time in the genesis of the pronominal system, pronominal reversals vanish from the data of typical developing children. These children merge their role as speaker and as agent, topic, grammatical subject, into one form. They conceive of themselves with a certain permanency, linked to Piaget’s object permanency (1923) and can keep their integrity, their identity in time. As Ricoeur writes in Soi-même comme un autre: Il ne s’agit pas de s’assurer qu’on parle de la même chose, mais qu’on peut l’identifier comme étant la même chose dans la multiplicité de ses occurrences. Or cela ne se fait que par repérage spacio-temporel: la chose reste la même en des lieux et des temps différents. (1990: 45) [The issue is not to make sure you talk about the same thing, but that you can identify it as being the same thing in the multiplicity of its occurrences. That can only be done in the spatio-temporal frame: the thing stays the same in various places and times]

At the end of the acquisition process, at around 3 years old, children join the grammatical subject, the semantic subject, the subject of conversation in a single form, the first person pronoun, but they have gone through a transition period during which they could separate these various levels and produce various markers, breaking away from regular adult forms. Children then understand that these various facets of themselves marked in the different forms they use (zero form, filler syllables, name, third person, accusative form, genitive form) do not alter the unity of their identity. At the end of the process, they have become speakers, or in enunciative terms, enunciators. They can speak of themselves and judge their own selves through what they have been and have done in the past, what they are and do in the present, what they will do and be in the future or would like to be and do. They therefore succeed in breaking away from the other, from the other’s speech and have built their own place in dialogue, on the intersubjective scene.

chapter 5

The function of formulations in polyphonic dialogues Claudio Baraldi

University of Modena and Reggio Emilia Formulation can be defined as a specific type of action, in interaction, which has the function of summarising, commenting or developing the gist of previous turns. Furthermore, formulations can elicit a choice of interlocutors’ next turns, in a way that interlocutors provide their own perspectives on the meaning of the interaction, and promote polyphonic dialogues. This chapter concerns video-recorded data collected in intercultural educational communities in English Lingua Franca. It looks at the structured sequences of actions including formulations. Complementarily, it focuses on the ways in which different cultural presuppositions, giving form to expectations, contextualise formulations. In this sense, formulations may be distinguished as referring to different forms of expectations. This study looks at various consequences of these forms of expectations for polyphonic dialogue.

1. The concept of formulation The term formulation (Heritage 1985: 104) indicates a type of action taking place in institutional interactions; questioners assuming institutional roles (e.g. news interviewers, lawyers in courts) may utter a formulation as a third-turn following a question-answer dyad. Through formulations, questioners avoid taking the third-turn receipt and permit overhearers (e.g. news audience, judges/juries) to be the primary recipients in the interaction. Providing a formulation consists in “summarising, glossing, or developing the gist of an informant’s earlier statement” (Heritage 1985: 100). Formulations “advance the prior report by finding a point in the prior utterance and thus shifting its focus, redeveloping its gist, making something explicit that was previously implicit in the prior utterance, or by making inferences about its presuppositions or implications.” (Heritage 1985: 104) Moreover, formulations may project a direction for subsequent turns by inviting new responses from answerers. This use of formulations has been explored by



Claudio Baraldi

Ian Hutchby (2005, 2007) in the context of psychological counselling with children. Hutchby’s analysis focuses on answerers as direct addressees. Following Heritage, Hutchby conceives formulation as the third turn in a triplet (QAF: question-answer-formulation); however, questioners’ (psychologists’) formulations are produced here in order to advance answerers’ (children’s) prior reports. Formulations project (Goodwin and Heritage 1990) interlocutors’ statements about their feelings or thoughts in order to move the interaction forward. In this approach, formulations are observed in interactions in which sequences of alternate actions are produced without discontinuities; they may be seen as turns which both develop prior utterances (“advance” them, in Heritage’s terms) and project further utterances. Insofar as formulation is produced with the specific purpose of eliciting interlocutors’ perspectives on the meaning of the interaction, it can promote a joint construction of meanings, and thus polyphonic dialogue. Two questions arise concerning the use of formulations in polyphonic dialogue. In which ways do formulations align with interlocutors’ previous turns? And, in which ways do formulations project interlocutors’ following turns? While I take the concept of formulation from Heritage and Hutchby, here I look at a different part of the sequence in which formulation is included. I focus on formulation as the second turn in a triplet with the following structure: (1) turn formulated, (2) formulation, (3) turn projected by formulation. By looking at formulation in the second-turn of a triplet, I concentrate on the relationship that a formulation creates with both the previous and the following turn; this means that formulation can be seen as both response and initiation. Formulations may therefore enhance a polyphonic dialogue, promoting both understanding and active participation among the interlocutors. Moreover, while Heritage and Hutchby look at dyadic interactions, my analysis concerns non-dyadic interactions (e.g. group interactions, dialogue interpreting), where polyphonic dialogue is particularly evident, because of the number of participants and their positioning in the interaction; in non-dyadic interactions, the primary addressees of formulations in turn 3 may be different from the speakers in turn 1 and may react to formulations. The next sections will deal with research data (§2) and analysis on the use of formulations in dialogue (§3), meanings of structured sequences of actions in which formulations are included (§4), meanings of cultural presuppositions of formulations (§5), and function of formulations in polyphonic dialogue (§6).



Chapter 5.  The function of formulations in polyphonic dialogues 

2. Research data An exploratory study of formulations in community interpreting (Baraldi & Gavioli 2007) shows that interpreters produce formulations in translations for institutional representatives and migrant people. In this section I shall show that in educational settings educators may produce formulations in promoting students’ learning. In educational settings, analyses have been made of triplets like questionanswer-comment sequences (McHoul 1978), or initiation-response-evaluation (IRE) sequences (Mehan 1979). Formulations can be understood as different third turns in substitution of either comments or evaluations. However, using Mehan’s terminology, formulations can also be observed as responses and initiations in the triplet, with important consequences for educational interactions. The corpus analysed for the purposes of this paper includes video-recordings of 200 interactions involving educators and small groups of 14–19-year-old adolescents in two types of international annual summer camps. Videotaping took place at camps in Italy, with participants from 38 different countries, i.e. highly intercultural settings. The first type of camp was organised by CISV (Children’s International Summer Villages), an international charity based in the UK and active in more than seventy countries all over the world. In these camps, eleven national delegations of four adolescents meet for three weeks. The second type of camp was organised by the Peace School of Monte Sole, near Bologna (Italy), a place where the Nazis killed hundreds of people in 1944. In these camps, four national delegations of ten adolescents met for two weeks. In both types of camps, delegations were guided by adults, called ‘leaders’ in CISV camps and ‘teamers’ in camps organised by the Peace School of Monte Sole. English was used as a lingua franca. The extracts which are used in the following sections were chosen for their clarity, and are in line with the prevalent use of formulations by leaders and teamers throughout the corpus. They can thus be considered fully representative of the kinds of formulations observed in the collection of data. The extracts have been transcribed according to Conversation Analysis (CA) conventions (e.g. Button and Lee 1987); formulations are highlighted in italics. 3. Analysis: Use of formulation in dialogue In my analysis, I have observed different uses of formulations with specific functions in educational settings, all of which aim at developing adolescents’ prior utterances and promoting the production of new meanings in the interaction. Here,



Claudio Baraldi

I shall describe the three most frequent uses of formulations and the specific functions they perform, as they emerged from the analysis. a. Formulation as implicit comment, following a sequence of turns and aligned with the previous turn The function of this type of formulation is that of producing new, potentially conclusive meanings, in an educational interaction, as it is shown in Extract 1.

(1) 1 LFhgk: so: ple:ase raise your hands if you have something to say about it, ok? We saw that group two won, m:: so: why do you think group two won? 2 Fbra: because there was (.) was: Paulo he was good in pone = put the water in the cup 3 LFhgk: Bjorn? 4 Mswe: m: they take only a little water so they wasted less water than us 5 LFusa: Inga? 6 Fpol: e: they are good 7 LFhgk: Brad? 8 Musa: they help each other so: they were so: effective 9 LFhgk: good (.) so: cooperating is very important In Extract 1, the female leader (LF) from Hong Kong (LFhgk) and the female leader from the USA (LFusa) are encouraging the adolescents to participate in the interaction (turns 1, 3, 5, 7). In turn 9, LFhgk summarises (formulates) the results of the adolescents’ contributions, in particular Musa’s turn 8, interpreting their meaning as a positive assessment of cooperation. In turn 9, LFhgk’s formulation is introduced by a positive assessment (“good”) and the prefatory discourse marker “so”, already observed by Ian Hutchby (2005, 2007) as a typical way of introducing formulations. b. Series of coordinated formulations This use of formulations can have two different functions. The first possible function is that of checking the formulating agent’s understanding of interlocutors’ statements, as shown in Extracts 2 and 3.

(2)   1 Fita: yeah but we can do the: (.) they (2) we can ahm no they can: use another, another sense, because the eyes are, must use   5 LFgre: mm mm   6 Fita: you must use   7 LFgre: because it’s night you mean? We must use them?   8 Fita: yeah, we must use (both the eyes) (?) you understand?   9 Farg: no



Chapter 5.  The function of formulations in polyphonic dialogues 



10 11 12 13 14

Fita: but LFgre: they cannot see Fita: yeah LFgre: so they must use another sense Fita: ((nodding)) another sense

In Extract 2, turns 4, 8 and 10, the leader from Greece (LFgre) tries to clarify the Italian adolescent’s attempts to explain her proposal (turns 1, 3, 5, 7) using formulations. In turn 4, the first formulation has an interrogative form aiming to check the meaning of Fita’s previous turns; in turn 8, the second formulation is an interpretation of turn 5. This interpretation is confirmed by Fita, thus allowing LFGre to conclude her interpretation through a third formulation, introduced by the prefatory “so”. This discourse marker is used to introduce a conclusion about previous turns, which is confirmed by Fita (turn 11). The series of formulations therefore combines (1) a clarification question, (2) a tentative interpretation and (3) a negotiated and shared conclusive interpretation.

(3)   1 Farg: ah for example (..) ah:m he can’t see ah we put like an activity that he, he, he (?) can’t do, only he can do and I need to (.) to: (.) to cross ahm one place without ahm falling with some objects and the other ahm, the other participants in that group have to tell him what [where he have to go   2 Mdan: [where he can go   3 LFgre: mm mmm   4 Farg: so that he doesn’t fall and in that way they are helping him   5 LFgre: yea   6 Farg: ahm although she can’t see   7 Fita: ok   8 Farg: do you understand?   9 LFgre: yea, I understood 10 Mdan: I also [have 11 LFgre: [but there, there are gonna be six groups and the six groups are gonna do something at the same time? 12 (0.2) 13 Farg: no, no! they are gonna be in different places 14 LFgre: yea, so they are gonna be in different places at the same time doing different things? 15 Farg: yes and then they are going to (.) change, like to go around 16 LFgre: so every group has to go to every 17 Farg: every situation yea



Claudio Baraldi

In Extract 3, LFgre initially uses continuers (Gardner 2001) to encourage the adolescents’ participation (turns 3, 5, 9). In turn 11, she uses a formulation in order to check her understanding (as in Extract 2, turn 4). This formulation is introduced by a self-disclaiming “but”, which is used to lower the adults’ authority (Baraldi 2009). Since the adolescent from Argentina refuses LFgre’s interpretation (turn 13), LFgre changes it in turn 14. This second interpretation is initiated by a confirmation (“yea”) and prefaced by the discourse marker “so”, as in Extract 1, turn 9. This time, however, it has an interrogative form: the leader does not seem to be sure about her interpretation. Only in turn 16, after Farg’s confirmation, LFgre shows confidence in her interpretation, prefacing her formulation with the discourse marker “so”. As in Extract 2, this formulation is finally confirmed by the adolescent (turn 17). Once again, the sequence of formulations instantiates clarification, tentative interpretation (here punctuated by the interrogative form) and negotiated and shared conclusive interpretation. The second possible function of this use of a series of formulations is that of summarising interlocutors’ statements in order to reach a conclusion about the meanings produced in the interaction and change the interlocutors’ discursive positioning, as it is shown in Extracts 4 and 5.

(4) 1 TF1ita: and how did you fulfil the gaps (..) that you don’t have? For example (..) if I imagine to be an illegal immigrant from Mali (..) I mean for example (..) M1alb (..) How did you fulfil the gaps you don’t have about a Chinese boy (..) a Chinese son of a Chinese father that hold a successful restaurant? How were you able to fulfil the gaps (..) so to imagine? What did you use in your mind to fulfil? What you didn’t know about it? 2 M1alb: I tried to imagine like that because I did myself in that position because one of my cousin is in Sweden and he is Albanian and he owns a middle successful restaurant in Sweden and he told me a little bit what they can do abroad (..) especially in Sweden and I thought his position, and I compared him (..) his position with the Chinese boy (..) that was the key. 3 TF1ita: so you had something direct (..) but how you compared the situation, Chinese and Albanian in Sweden for example? What has to do? It has to do something? 4 M1alb: example Albanians are black and in Sweden as he said me they saw you that you don’t have (..) not the majority (..) they saw that you don’t have for example 5 M2alb: white skin. 6 M1alb: like this (?) there was some directions I used to fulfil the gaps.



Chapter 5.  The function of formulations in polyphonic dialogues 

7 TF1ita: so it was the same (..) that you are different in a foreign country (..) you are a people that is different in a country and you realize that you are different from the other and (..) you can do something but not all the things that the other do = 8 M1alb: = not everything (..) not all the things that the others do in their country (..) you have some things. In Extract 4, the Italian team leader initiates the interaction (turn 1) by inviting the adolescents to reflect upon migration processes. In turns 3 and 7, she summarizes the Albanian adolescents’ statements in order to reach a generalized conclusion starting from their personal stories. In both cases, TF1ita uses the prefatory “so” to underline the conclusive nature of her formulations. Here, unlike Extracts 2 and 3, the series of formulations works as reinforcement of the formulating agent’s position, as well as invitation to change the adolescents’ positions, without negotiation and sharing. In Extract 5, the two functions of checking understanding and reaching a conclusion are combined. LFdut: ok. So you think it’s not working?((addressing Feng)) Feng: I’ve just put all together LFdut: but you are not (sure) that it will work((addressing Mdut)) Mcos: she’s just saying that it’s not stable Mdut: it’s stable, you can put it Mcos: no, no Mdut: if we do more tape we can also can do this Mcos: we can do LFdut: (?) but maybe you can work on (.) on make it (..) I think you don’t like it because you don’t like the tape, right? You don’t like [the tape 10 Mdut: [it’s just ugly

(5)   1   2   3   4   5   6   7   8   9

In turn 1, the Dutch leader asks a question about the English adolescent’s previous wailing about the activity in the group. This formulation is similar to turn 14 in Extract 3: after a confirming acknowledgement token (“ok”) (Gardner 2001), LFdut produces a formulation using questioning intonation and prefacing it with the discourse marker “so”, showing her dubious interpretation. In turn 3, the initial self-disclaimer (“but”) has an analogous function in the doubtful formulation of the Dutch adolescent’s position. In turn 9, however, the initial “but” has a different function: it is another-disclaimer in that it introduces doubts about Mdut’s position in the interaction. After two formulations to check understanding, this final formulation has the twofold function of reaching a conclusion about the meanings



Claudio Baraldi

produced by the adolescents (“you don’t like the tape”), and changing their discursive positioning without negotiating (“maybe you can work on (.) on make it”). Mdut confirms this final formulation in turn 10. c. Formulations coordinated with other kinds of action (e.g. continuers, acknowledgment tokens, echoes, direct invitations to participate), produced by the same (formulating) agent The function of this type of formulations is that of promoting the interlocutors’ (adolescents) active participation, as it is shown in Extracts 6, 7 and 8. (6) 1 Mger: religion shouldn’t take decision, political decision 2 TF2ita: ah ok, so you didn’t agree with the (4) with the putting in of the religious theme into this text for all European citizens 3 Mger: yeah 4 TF2ita: were you all agree? 5 Adolescents: yes 6 Ffra: yeah, but we had to talk a bit because (3) ((Ffra asks something to Mfra)) 7 Mfra: and we take the topic united in diversity. 8 TF2ita: ah ah 9 Mfra: because it’s the (?) of the European union and also the motto for the European union, united in diversity, so it’s really many nations but we are united, we’re together (3) go on somebody In Extract 6, the Italian teamer introduces a formulation in turn 2, initiating it with a token of acknowledgement (“ah ok”) and a prefatory “so”. TF2ita’s formulation summarises the previous turn strengthening it and projecting a confirmation in turn 3. However, the teamer shows her interest in promoting further conversation on this topic, by first asking for further perspectives (turn 4), and then encouraging the discussion by means of a continuer (turn 8). This is how the initial function of the formulation is completed through a promotional question and a continuer, which support it in promoting the adolescents’ active participation.

(7)   1 Mger: when homosexual people educate a small child, this child would be with this, with this ill people, with this are =   2 TF1ita: = yes, so, if I got well, it’s (..) ok, it’s that if they are educated, if a child is educated in a family of homosexual people they get the same illness?   3 Mger: yes   4 TF1ita: and =   5 Mita: = I think the opposite



Chapter 5.  The function of formulations in polyphonic dialogues 

  6   7   8   9

TF1ita: eh? Mita: I think the opposite TF1ita: please, say your opinion Mita: if a child is grow up by maybe two men, two people that are looked different by the society, this two people may (..) ((in Italian))insegnare è (teaching is) (.) teach to this child to accept all the different people from you and have no prejudice, maybe it’s better grow up with two homosexual parents than two normal parents 10 TF1ita: did you understand what Mita said, everybody understood? (?) you understood as well? ok, and what do you think about it (..) and then I’m very curious as well (..) yes ((TF1ita refers to Fger who has asked to speak))

In Extract 7, turn 2, an analogous initiation (“yes so”) can be seen, as well as a formulation checking understanding, which is stressed by an introductory “if I got well” expressing doubts, and aiming to clarify the meaning of the previous utterance. In the following turns, the Italian teamer promotes the adolescents’ active participation through a continuer (turn 4), an explicit encouragement (turn 8), and a promotional question (turn 10). (8) 1 TF1ita: so you could never answer yes (..) or quite never (..) to the question 2 Fita: no, never. 3 TF1ita: What about you Katrine? 4 Fger: I also like her, because I couldn’t move (..) because I can’t say yes to the question. 5 TF1ita: you can’t say yes to the question

In Extract 8, turn 17, TF1ita formulates the previous turns (not shown), receiving Fita’s confirmation (turn 2). In turn 3, she encourages Fger’s participation directly and in turn 4 her echo “reflects” (Mearns & Thorne 1999) Fger’s position, showing her understanding and support. To sum up, we have observed: (i) formulations as implicit comments producing new, potentially conclusive meanings of the interaction (Extract 1); (ii) series of coordinated formulations either checking agents’ understanding (Extracts 2, 3 and 5) or summarising interlocutors’ turns (Extracts 4 and 5); (iii) formulations coordinated with other kinds of action by the same agent promoting interlocutors’ active participation (Extracts 6, 7 and 8).



Claudio Baraldi

4. Structured sequences of actions including formulations The analysis conducted shows that formulations: (i) may be preceded by complex sequences of turns, as in Extract 1; (ii) may be produced in a series of turn (series of coordinated formulations), as in Extracts 2, 3, 4, and 5; (iii) may work in cooperation with other kinds of actions (promotional questions, continuers, acknowledgment tokens, echoes, direct invitations to participate), which follow them in the interaction, as in Extracts 6, 7, and 8. Therefore, the triplets with formulations, analysed in the sequences, are frequently expanded in the first part (as plurality of turns preceding formulations), in the second part (as series of formulations), and in the third part (as plurality of actions following formulations). This kind of expansion is typical of group interactions, which enhances more complex sequences of actions, especially by virtue of the adults’ coordinating roles. It is worth dwelling here on the difference between the two general ways of using formulations in the interaction: advancing prior turns and projecting further turns. It is interesting to observe that the development of prior turns may be initiated through: (i) confirming acknowledgment tokens (“ok”, “ah”, “yea”, etc.); (ii) the discourse marker “so” (see also Hutchby 2005, 2007), which announces inference or interpretation; (iii) the disclaiming “but”, which can introduce both doubtful checking and conclusive interpretations. Acknowledgment tokens, the discourse marker “so”, and the disclaimer “but” are cues for the different uses of formulations and their functions in the interaction. While formulation may work as advancement of previous sequences of turns, it may also be followed by sequences of further formulations which expand its meanings and by other types of actions, such as promotional questions, continuers, acknowledgment tokens, echoes, direct invitations to participate. In particular, additional types of actions integrate formulation in the function of promoting polyphonic dialogue in the interaction. 5. Cultural presuppositions of formulations In the previous sections we have seen that formulations may have the following functions: to produce new meanings which are potentially conclusive for the interaction (Extract 1); to check agents’ understanding (Extracts 2, 3, 5); to provide summaries that change interlocutors’ discursive positioning (Extract 4, 5); to promote interlocutors’ active participation (Extracts 6, 7, 8). The difference in these functions cannot be explained simply in terms of the structure of the sequences of actions (§4). Rather, it is correlated to a difference in cultural presuppositions (Gumperz 1982, 1992): it is to these presuppositions that



Chapter 5.  The function of formulations in polyphonic dialogues 

the participants “must rely on to maintain conversational involvement and assess what is intended” (Gumperz 1992: 230). Cultural presuppositions shape expectations in the interaction. For each important social system in society, we can observe expectations concerning communication processes (Luhmann 1984), specifically concerning the values which guide these processes, role performances and personal opportunities of expression which can be included in these processes, and their results. Therefore, cultural presuppositions can be regarded as forms of expectations that guide the participants’ conversational involvement, i.e. their actions (Baker 2006, Luhmann 1984). By investigating interactions we can detect the cues for these presuppositions, or contextualization cues (Gumperz 1982, 1992), in the usage of language. In particular, the usage of formulations seems to be correlated to three different cultural presuppositions, i.e. three forms of expectations which guide their production and function in the interaction. 1. Normative expectations, concerning regulation of behaviours and respect of rules (Extracts 1, 4). In these cases, formulations fix norms for adolescents. The function of formulation is either production of conclusive meanings (Extract 1), or summary aiming to change adolescents’ discursive positioning (Extract 4). 2. Cognitive expectations, concerning information processing and production of knowledge (Extract 5). In these cases, formulations participate in the construction of knowledge considered useful for adolescents. The function of the formulation is summary, aiming to change adolescents’ discursive positioning. 3. Affective expectations, concerning personal self-disclosure (Extracts 2, 3, 6, 7, 8). In these cases, formulations promote adolescents’ expressions of thoughts and feelings. The function of formulation is either checking adults’ understanding (Extract 2, 3) or promoting adolescents’ active participation (Extracts 6, 7, 8). The functions of checking understanding and above all promoting active participation through formulations seem to be based on affective expectations, as demonstrated by Baraldi and Gavioli (2007). In educational settings, affective expectations seem to be correlated with the most complex chains of adults’ actions and to their most doubtful and reflective use of language (e.g. self-disclaiming “but”, acknowledgement tokens). There are not, however, linear correlations between forms of expectations and particular uses of language, as the case of the prefatory “so” shows.



Claudio Baraldi

6. Consequences for polyphonic dialogue Different forms of expectations may have different consequences for polyphonic dialogue. In polyphonic dialogues, formulations based on normative expectations promote harmonizing behaviours, formulations based on cognitive expectations promote shared knowledge, and formulations based on affective expectations promote personal expression of thoughts and feelings. Harmonisation, shared knowledge, and personal expression may be seen as different possible features of polyphony, with different cultural presuppositions of formulations. Harmonisation requires formulations which have a conclusive meaning and are able to change interlocutors’ positions. Shared knowledge also requires formulations which are able to change interlocutors’ positions. By contrast, self-disclosure through personal expression requires formulations which check understanding and promote active participation. Polyphony based on personal expressions can be considered the most effective in enriching the quantity and quality of voices included in the interaction. The use of formulations in educational settings is particularly interesting. The typical triplets observed in education (question-answer-comment and initiationresponse-evaluation) can be substituted by triplets including formulations as both initiations and responses. In particular, the use of formulations based on affective expectations is associated with changes in education which are in line with some recent ideas on the importance of children’s active participation and the limits of traditional education in promoting it (e.g. Baraldi 2008; James & James 2004; Prout 2003). These formulations may introduce effective polyphonic dialogue in educational settings. In particular, expectations may have consequences for polyphonic dialogue in situations of intercultural communication (e.g. Samovar and Porter 1997; Ting-Toomey 1999). Not surprisingly, formulations based on normative and cognitive expectations promote cultural standards and suppress polyphony which is based on cultural differences in the interaction, as they are either conclusive or aim to promote change of positions. This type of formulations does not produce intercultural polyphonies. More surprisingly, affective expectations do not necessarily promote intercultural polyphony. In fact, formulations based on affective expectations promote interpersonal polyphonies, as they either check or promote personalised positions. Overall, the use of formulations does not seem to promote intercultural polyphonic dialogue, as the expectations guiding them do not allow this promotion.



Chapter 5.  The function of formulations in polyphonic dialogues 

7. Conclusions The study of formulations is interesting for the analysis of polyphonic dialogue. Formulations may be analysed as both included in structured sequences of actions and guided by cultural presuppositions (i.e. forms of expectations). The analysis enhances understanding of the various functions of formulations in the interaction: producing new meanings, checking understanding, summarising, promoting active participation. An analysis of this kind may even delve into the complexity of interactions which include formulations and grasp the function of acknowledgment tokens, discourse markers, and disclaimers as cues for the advancement of prior turns through formulations. Finally, such an analysis shows that formulations based on affective expectations can promote changes in children’s social positioning, giving them voice. In this way, they can re-contextualise the interaction (Baker 2006), through a different distribution of active participation, demonstration of sensitivity for interlocutors, and empowerment of interlocutors (Baraldi 2009); however they do not promote intercultural polyphonies. Transcription conventions (?) audio untranscribable (.) shortest pause (less than 0.6 seconds) (..) short pause (between 0.6 and 1 second) (n) long pause (n = length in seconds) = latched to the preceding/following turn in the transcript [text spoken in overlap with aligned [text text emphases te:xt lengthening of previous sound or syllable (number of colons indicates extent of lengthening) text - tone group interrupted (text) translation of Italian turns ((text)) transcribers’ comments ,.?! rough guide to intonation

chapter 6

Observing the paradox: Interrogativenegative questions as cues for a monophonic promotion of polyphony in educational practices Federico Farini

University of Modena and Reggio Emilia Many pedagogical publications (Hengst & Zeiher 2005, Walsh 2011) offer prescriptive resources to empower children’s voices in educational practices, placing particular emphasis on children’s self-realization and agency. However, none of them discusses the results of the empirical application of their theoretical prescriptions. In this chapter, I analyze interactions taken from an educational project addressed to empower children’s skills in managing social contexts, particularly with respect to conflict management. I focus on a specific sequence of actions, “educators’ interrogative-negative questions/children’s answers.” This sequence is a contextualization cue working at various discourse levels, including intertextuality, for a specific culture of education. Specifically, it foregrounds the paradox of a monophonic promotion of polyphony where, besides promotional intentions, education selects meanings, criteria and values, binding children’s actions and choices to them.

1. A new representation of education James (1983) conceptualized education as an intersubjective relation where children actually play an active role in influencing the outcomes of their education, rather than being regarded as empty boxes to be filled with knowledge. Uncertainty which is associated with the outcomes of education has since been a concern for education scientists and practitioners. Pedagogy has devoted many efforts to the design of curricular and behavioural rules, as well as structures, which incorporate the cultural presuppositions of standardised role performances and cognitive expectations. However, the concept of education as a development of personality controlled by educators using educational techniques has become



Federico Farini

more and more controversial. Against this backdrop, unsatisfactory reforms were put in place, one after another, until a picture of failure in the education system spread and became popular among education scientists, sociologists and politics (Luhmann & Schorr 1979). Arendt (1993) accepts crisis as a permanent condition of education, whose appearance in the public debate derives from the transfer to the political agenda of a structural limit of education, that is, its incapacity to control the development of children’s personality. Arendt highlights a twofold paradox that is inherent to the idea of education as a developmental process controlled by educators: (1) the development of personality brings about the problem of trying to know a mind that resists being known, (2) teachers have to take responsibility for children, while children are inescapably free in their construction of meanings of education. Since the 1980s, following a generalized lack of trust in education, the culture of childhood development has placed particular emphasis on children’s self-realization and agency (Vanderbroeck & Bouverne-De Bie 2006), and has looked at the inevitable autonomy of children as social agents as a resource for education. This representation leads to the promotion of children’s sense of responsibility and skills in planning, designing, monitoring and managing social contexts, thus showing an “understanding of their own competencies.” (Matthews 2003: 274) The success of this new vision of children as social agents has promoted a new role for the educator as an “organiser of learning” (Holdsworth 2004: 149), who gives up pre-planned activities, and is inventive and able to understand that children can and must tackle important issues on their own. Today, many publications in the field of pedagogy offer prescriptive resources to empower children’s voices in educational practices, for example through teachers’ active listening and consideration for children’s creativity (Gordon 1974; Rogers 1951). However, none of these publications discusses the results of the application of its theoretical prescriptions. This chapter discusses naturally occurring interactions recorded in the context of an educational project aimed at promoting children’s active participation in educational practices in the classroom and children’s ability in dialogic conflict management (see Section 2). I focus on the failures and crises of the promotional intentions of the project, in which educational practices suffocate the voices of children, thus creating a monophonic educational discourse. In particular, I focus on the educators’ use of questions having negative interrogative frames as their first component to induce children to align with their educational agendas. I will argue that the use of interrogative-negative questions is an educational practice which reflects a culturally specific professional practice of ‘‘being educators’’, whereby educators elicit children’s standardized cognitive performance through monologues, rather than empowering their voices.



Chapter 6.  Observing the paradox 

2. The relevance of intertextuality in educational discourse In any social encounter, participants rely on repertoires of cultural presuppositions to foreground the expectations of others, therefore being able to choose how to act and react (re-act) to the actions of their co-participants. (Gumperz 1992; Luhmann 1984) Educational discourse, both the discourse on education (pedagogy) and discourses in education (educational interaction), is permeated by cultural presuppositions concerning role performances and the interrelation of educators’ actions and children’s actions in educational interaction. While cultural presuppositions provide resources to select meanings and support interaction, the linguistic structures of interaction provide resources for understanding and accepting these presupposed meanings, thus enhancing understanding and generalized participation. The cultural presuppositions of interaction may be empirically observed by looking at contextualization cues, which highlight, foreground or make them salient in the interaction. According to Gumperz (1992: 230–231), contextualization cues are verbal and non-verbal signs used by interlocutors “to relate what is said at any one time and in any one place to knowledge acquired through past experience in order to retrieve the presuppositions they must rely on to maintain conversational involvement and assess what is intended.” Contextualization cues work at various levels of discourse: prosody (intonation, pitch shift, etc.), paralinguistic signs (tempo, pausing and hesitation, latching or overlapping of speaking turns), code choice (style, language, etc.), and choice of lexical forms or formulaic expressions. (Gumperz 1992) This paper focuses on paralinguistic signs, code choice and choice of lexical expressions. Moreover, contextualization cues which are not included in Gumperz’s repertoire are analyzed. These pertain to the discourse level of intertextuality. I will use the term intertextuality, originally developed by Kristeva (1980) in the field of semiotics, to refer to interactions which are constructed as a mosaic of references to prior talk, common knowledge and shared expectations. Intertextual references are used to perform specific social activities. In my data, intertextual references convey expected answers after interrogative-negative questions, making it possible for educators to steer children’s actions within monophonic educational discourses. I argue (see Section 4) that since interrogative-negative questions use intertextual references, they may be intended as contextualization cues for education as monophonic discourse, which, paraphrasing Bakhtin’s (1984a) criticism toward nineteenth-century realism in literature, simulates a dialogue between educators

 Federico Farini

and children but in fact simply provides a medium for the educator’s own discourse to be more prominent. 3. Method and data This section provides the reader with a description of the dataset and some background information regarding methodological issues, before moving to the analysis of the data. The educational project under investigation involved 250 children (aged 10–12) in eleven primary schools in Northern Italy (April-May 2005). The study was financed by the Italian Ministry of Education, University and Research (for an account of the whole research project see Iervese 2006). The project consisted of a number of interventions in schools, whereby the classes involved were divided into groups of five children, and each group was asked to create a fantasy story; to accomplish their task, children had to face decision-making processes which could engender conflicts. When conflicts arose in one group, educators with experience in the promotion of children’s active participation would join the group in order to promote children’s reflection on conflict and conflict management. All of the educators had been trained to support children’s self-expression and self-reflexivity, so that they would not play the role of experts addressing a passive audience, transmitting knowledge about the correct ways to manage conflicts. The methodology of the interventions took into account the most recent cultural presuppositions of children’s empowerment, observing children’s participation as a form of involvement in decision-making, which make them feel influential (Lawy and Biesta 2006). Empowerment means acknowledging that when children are able to determine the issues that they consider important the outcome cannot be known in advance (Britzman 2007). Thus, empowering children’s voices entails certain risks; first, the emergence of contradictions and refusals among children is more likely within a form of educational communication where educators give up their potential role of child controllers. This risk is related to a form of communication where educators become facilitators rather than technicians, and both adults and children are considered as co-constructors of knowledge and expertise (Murray & Hallett 2000). With regard to the connection between children’s active participation and conflict management, previous research indicates that the promotion of positive conflict management should be seen as a complement to the empowerment of children’s voices (Holdsworth 2004; Baraldi 2009). As the project provided for three two-hour interventions in each of the eleven primary-school classes involved, I had the opportunity to work on 66 hours of videotaped interactions. The cases shown here are representative of my collection,



Chapter 6.  Observing the paradox 

and were selected according to which ones would best illustrate the range of operations performed by interrogative-negative questions as clear examples of the phenomenon. My purpose is to highlight a range of possible meaningful communicative situations and problems related to the use of interrogative-negative questions. Data were transcribed using Jefferson’s transcription system (see Annex 1). All personal details in the talk were altered in the transcription to protect the participants’ right to anonymity. Italian originals of the Examples are provided in Annex 2. We use Conversation Analysis to study talk-in-interaction. The objective of Conversation Analysis is to discover how participants understand and respond to one another in their turns at talk, with a central focus on how sequences of actions are generated. The issue for conversation analysis is how participants understand, and make use of, any given utterance. (Ten Have 2007; Hutchby & Wooffitt 1998) People’s understanding of each other’s action actually unfolds as sequences themselves unfold, and can be analysed using the next turn proof procedure: any next turn in a sequence displays its producer’s understanding of the prior turn, and if that understanding happens to be incorrect, that in itself can be displayed in the following turn in the sequence (Clift, Drew & Hutchby 2006; Schegloff 2007). The second turn in the sequence makes the interlocutors’ understanding evident. (Heritage 2006; Mazeland 2006) Here, I discuss a specific sequence of actions-in-interaction, namely educators’ interrogative-negative questions followed by children’s answers. I argue that this sequence represents a contextualization cue for a specific culture of education. I aim to demonstrate that interrogative-negative questions foreground a culture-­specific concept of education, whereby educators must guide children’s socialization by shaping their choices and actions according to shared criteria and values, and also by selecting correct meanings and binding children to them (Baraldi 2008). Finally, my analysis of the sequence (educators’ interrogative-negative question followed by child’s answer) aims to bring the issue of the structural limits of educational communication to the fore, thus meeting James’ claim for an evolution of educational practices toward polyphony. 4. Observing the paradox, a monophonic approach to the promotion of polyphony The way interrogative-negative questions work relies on the sequential properties of questions as social actions (Heritage 2002). Asking a question establishes significant constraints on what the recipient does next, and therefore places the questioner in an interactionally powerful position (Heritage and Raymond 2003;



Federico Farini

Stivers and Makoto 2010). When a person asks someone else a question, s/he establishes constraints not only in terms of the type of action the recipient should produce next, but also in terms of action design (Robinson and Heritage 2005). For instance, while a polar yes/no question restricts the expected answer to either yes or no, an alternative-type question limits the possible range of expected answers to one of the ‘alternatives’ provided (Raymond 2003). Questions open a sequence whereby the action that follows is interpreted as an answer. Question-answer sequences are instances of adjacency pairs (Schegloff and Sacks 1973). In adjacency pairs, a given first-pair part projects a range of seconds.. Given the initial condition of a first pair part being uttered, the second part of that pair is then relevant. Consequently, the absence of such a second is a noticeable absence, and the speaker of the first may infer a reason for that absence. Adjacency pairs are two-action sequences in which participants’ mutual perception may be observed at its basic level. As interaction exists when communication takes participants’ mutual perception into account, and adjacency pairs are the basic form of interaction. The relation between first- and second-pair parts is not accidental. Among alternative second parts, there are preferred and dispreferred second parts (Pomerantz 1984). Preferred second parts are normal, seen but unnoticed. They are expected actions that require no account. On the contrary, dispreferred second parts are noticeable and therefore have to be accounted for. In line with Bolinger’s idea (1957), later developed by Boyle (2000) and Koshik (2002), my data show that the preference for interrogative-negative questions may be conveyed by intertextual references. For instance, a question such as “Isn’t it a beautiful sunny day?” in a setting where the questioner is out of doors in full sunlight conveys the questioner’s stance on the weather and expresses a marked preference for the recipient’s agreement. The act of questioning implies claiming that the questioner lacks certain information, while at the same time there is an assumption that the addressee has such information. Thus, the addressee is projected in a knowledgeable position. However, different question designs can adjust, or even reverse, the knowledge gap between questioner and respondent (Heritage and Raymond 2010). Interrogative-negative questions are instances of questions designed to claim a knowledgeable position for the questioner, and are used to seek confirmation for information that is already in play. If the object of an interrogative-negative question is a piece of shared knowledge, the question conveys the questioner’s stance toward the matter under discussion, and expresses a marked preference for the recipient’s agreement.



Chapter 6.  Observing the paradox 

4.1

Resisting the course of action: Non conforming-answers

In example 1, two groups are arguing after group 1 (G1) has asked a member of group 2 (G2) to perform in the role-play of their fantasy story. Pia and Lucia are members of G1, Raggi is a member of G2. (1) ((Castel S. Pietro, Bologna. Primary school. Children aged 10)) 1 Edu: excuse me but: (.) let’s avoid confusion (0.7) wasn’t the decision that Raggi would perform for both teams made this morning in class? 2 Pia(G1): but we: 3 Edu: if you had something to say (0.3) you should have said that this morning; hh didn’t you listen to yourselves making the decision? 4 Lucia(G1): °no, it’s tha:t° 5 Edu: this is your problem hh, don’t you think? 6 Raggi(G2): yes, >but then we said [also-]< 7 Ed: [I’m not] interested in that, it is your problem, and that has wasted ten minutes (.) stop raising your hand ’cause in a quarter of an hour we haven’t been able to find a solution to a banal proble:m The interrogative-negative question in turn 1 (example 1) is the first part of an adjacency pair and makes an answer relevant as the following action. Two features of this interrogative-negative question concur to establish the educator’s evaluative position as pre-established and, as a corollary to that, to assert his primary rights to assess the children’s action: (1) being a polar question, it calls for either yes or no, and thereby asserts command of the terms to be used by the recipient in the assessment of the referent (Heritage and Raymond 2006); (2) as its referent is part of the shared knowledge between the parties, the question is to be understood as a reverse polarity question, which conveys a strong preference for a yes answer (Koshik 2002; Raymond 2003). The reverse polarity is established by the context. The children are arguing, so it is obvious that they are not working well. By means of the interrogative-negative question in turn 1, the educator asserts his epistemic authority in establishing the rules which regulate the activity, epistemic authority concerning the rights to identity-bound knowledge in self-other relations. (Heritage and Raymond 2005; Clift 2006) In example 1, the interrogative-negative questions in turns 1, 3 and 5 are designed to display the educator’s critical stance on children’s behaviour, and display a marked preference for a yes answer which aligns with that stance. Being questions with known answers, the interrogative-negative questions are understood by



Federico Farini

children as a way to force them to acknowledge the inconsistency between their behaviour and an activity rule established earlier in the morning. Children are aware that, by providing the preferred yes answer, they will perform the relevant action after each of the interrogative-negative questions, thus granting the educator the right to regain the status of speaker. Indeed, from the educator’s perspective, the (preferred) yes answer is all that should be said after each of the interrogativenegative questions, and coincides with a transition relevance place. (Sacks, Schegloff & Jefferson 1974) Instead of providing the preferred yes answer, children provide non-conforming answers (Raymond 2003) in turns 2, 4 and 6 (example 1). Children use the status of non-conforming answers as dispreferred second pair parts as a resource to make room for their voices in the interaction. Being dispreferred second pair parts, nonconfirming answers destabilize the expected course of action, allowing children to make room for an account of their conflict. Non-conforming answers are cues for children’s misalignment with the trajectory projected by interrogative-negative questions. The interrogative-negative questions in turns 1, 3 and 5 (example 1) reduce the range of options for children. The only relevant action in the next turn is to provide an answer, and the only relevant and preferred next action is to answer yes. Interrogative-negative questions deprive children of the opportunity to explain their interest in the conflict. By offering non-conforming answers, children refuse to disown their behaviour; however, the educator is not interested in hearing their accounts. His interrogative-negative questions are addressed to assist children’s performance, and help them to correct inappropriate behaviour. Children’s misalignment with the interrogative-negative questions’ preference is seen as an inappropriately defensive behaviour. The educator treats children’s misalignment with the course of action projected by the interrogative-negative question as behaviour “to be corrected”. He does not allow children to complete any of the accounts they start until, in turn 7 (example 1), he sanctions non-conforming answers with a strong negative assessment, accusing children of showing a lack of competence in relationship management. In example 1, the children refuse to align with the course of action projected by the interrogative-negative question with reverse polarity, avoiding providing a yes/no answer. Episodes where the children systematically refuse to produce the expected second parts of question-answer adjacency pairs are rare in my data, and may be understood as cues for either total disagreement with the presupposition embedded in the question (i.e. the children’s behaviour is inappropriate and has to be corrected); or the children’s awareness that providing the preferred second pair parts would imply agreeing with this presupposition.



Chapter 6.  Observing the paradox 

In example 1, the educator treats children’s misalignment to the course of action projected by the interrogative-negative question as accountable and punishable behaviour. By doing so, he reactivates the social asymmetry of the educational relation between an epistemic authority who has something to teach and the children, who, being incompetent social actors, must passively experience the educator’s action and competences. 4.2

Playing with intertextuality. The failure of a rhetorical device

In example 2, two children, Luca and Sara, are arguing about the script of their team’s story. The educator’s intervention is structured as follows. First, a positivequestion with reverse polarity, preferring a no answer (Koshik 2002), is produced in order to force children to acknowledge the inconsistency between their common goal to produce a good story and their behaviour (turn 1). Once, Sara, one of the children, has provided the preferred no answer (turn 2), the educator produces an interrogative-negative question to promote children’s reflection on alternative ways of managing conflict. (turn 3) (2) ((S. Martino, Reggio E. Primary school. Children aged 10)) 1 Edu: but: does your way of doing things help your work? 2 Sara: no 3 Edu: and: can’t you find a solution? 4 Luca: as far as I can see no 5 Edu: mh? Do you think it’s impossible to find a solution? isn’t it possible for you to do anything together? 6 Luca: you’re right (.) it is not possible 7 Edu: ah, we:ll (.) I’ll be back later Even though the interrogative-negative question (example 2, turn 3) is apparently aimed at finding out if children are able to find an alternative way to manage their conflict, it is a question with a known answer, in that it has a piece of common knowledge as its referent. The educator expects children to be able to manage the conflict by avoiding non-constructive quarrelling, and children know that. They understand the interrogative-negative question as an indirect speech act urging them to suspend their conflict by relying on intertextual references to the educator’s expectations. Like all types of reverse polarity questions, interrogative-negative questions do not display their preference for a specific answer only through their design. Questions which are similar in design may be interpreted as either reversed polarity questions or as questions which prefer answers of the same polarity, depending on the displayed knowledge state or epistemic strength with which the question is



Federico Farini

asked. Example 2 is taken from the last intervention in the project, where children are expected to be able to manage conflicts without quarrelling, and they know what is expected of them. This expectation represents the intertextual reference which works as a cue for the function of the interrogative-negative question, i.e. expressing a negative evaluation of children’s behaviour, urging them to find an alternative way to manage the conflict. The interrogative-negative question (example 2, turn 3) relies on the sequential properties of the question/answer adjacency pair. Children have to stop arguing to produce the relevant action after a question, that is, to provide an answer. Moreover, the interrogative-negative question offers a candidate answer, giving children the sense of what the anticipated answer might be. According to Pomerantz (1988), offering a candidate answer is a resource for the speaker to guide the recipients towards giving particular information. Candidate answer questions are a common social object in many settings (Atkinson & Drew 1979; Arminen 2005; Heritage and Clayman, 2010), but in educational interaction they have a special relevance. (Margutti 2006) Providing or not providing a candidate answer in educational settings is an accountable action, which is subject to evaluation. Although in my data children (the question recipients) typically abide by the constraints that interrogative-negative questions impose on them, they can and sometimes do resist these constraints. In example 2, the preference structure of the interrogative-negative question (example 2, turn 3) relies on intertextual references to shared expectations toward children’s ability in conflict management. Unlike the interrogative-negative question in example 1, turn 1, the intertextual reference remains implicit. The ambiguity of the intertextual reference provides a child, Luca, with the opportunity to subvert the educator’s agenda. In example 2, turn 4, Luca plays with intertextuality. As the intertextual reference of the interrogative-negative question in example 2, turn 3, remains implicit, he has a good hand in ignoring the preference it conveys. Luca gives the educator the piece of information she is asking for, treating the interrogative-negative question as if it were a real request for information. Luca’s answer is an appropriate second part of the question-answer adjacency pair initiated by the interrogative-negative question in example 2, turn 3, and is therefore not accountable per se. Luca defers the answer until the end of the turn in the interest of emphasizing his agency. By deferring the appropriate yes/no answer, Luca forces the educator to monitor his turn to search for the preferred yes. The educator has to refrain from intervening, even though the delay in proffering the yes/no answer foregrounds a dispreferred response. Clearly, the educator may well sanction Luca’s action. However, by sanctioning Luca’s answer she would reveal that the interrogative-negative question in



Chapter 6.  Observing the paradox 

example 2, turn 3, was not a request for information, but an attempt at trying to force children to disown their behaviour. By sanctioning Luca’s answer, the educator would explicitly show that she is doing education, rather than empowering children’s voices. Luca disaffiliates with the educator’s intentions without misaligning with the course of action projected by the interrogative-negative question. A sanction against such relevant second pair part would represent a cue for asymmetry in the right to assess one’s own position, and this that would be inconsistent with the presuppositions of the intervention. After example 2, turn 4, the educator has no alternative but to reiterate the interrogative-negative question (example 2, turn 5), and again Luca treats the interrogative-negative question as if it were a genuine request for information (example 2, turn 6). When it becomes clear that her strategy is ineffective, the educator leaves the interaction without having reached her educational goals (example 2, turn 7). The educator’s interrogative-negative questions involve assumptions, or cultural presuppositions, that must be true for the question to be valid. In example 2, interrogative-negative questions involve a culture-specific concept of education, whereby education must guide children’s socialization. From the educator’s perspective, by arguing about the story’s script children show their need for education. Thus, they have to learn from the educator better ways to manage their conflicts. Our data are in line with Stivers and Makoto’s (2010) analysis of question-answer adjacency pairs in everyday conversation. In particular, they show that interrogativenegative questions are a resource for speakers to build a case for points of view in educational contexts as well as in casual conversation. In my data, educators use interrogative-negative questions as a resource to advance, albeit implicitly, the claim for a knowledgeable status with regard to the correct way to manage conflicts. Hence, interrogative-negative questions are cues for the link between the local distribution of rights and responsibilities regarding what educators and children can be expected to know and whether they have the right to articulate it, and on what terms, on the one hand, and the identities of educators and children, on the other. Luca’s answers to interrogative-negative questions are a denial of both the latter’s presuppositions and the claim for a knowledgeable status with regard to the correct way to manage conflicts advanced by the educators. They resist the agenda of the questions, i.e. what the questioner is doing with those questions. (Bolden 2009; Stivers and Makoto 2010) Despite some pedagogical concerns, example 2 provides evidence that children are competent communicators. For instance, they are able to understand the educators’ use of intertextual references as cues for their hidden agenda which foreground that they are doing teaching, rather than providing empowerment of children’s voices. In this way, children acquire the ability to break free from the



Federico Farini

constraints of interrogative-negative questions, and exert agency in proposing alternative agendas and challenging cultural presuppositions of the interactions. Not surprisingly, when children understand their educator’s objectives and the motivations behind interrogative-negative questions as cues for a representation of them as incompetent social actors in need of education, they will tend to mistrust the opportunity of active participation in the interaction. 5. Conclusions. On the limits of educating towards autonomy In the interventions analyzed, educators resort to various strategies, such as interrogative-negative questions, to reduce the possibility of surprises and risks for educational activities. In the contexts where the limitation of risks is a priority, educators systematically try to actively orient children’s participation, giving education the form of a monologue. Educators’ efforts at controlling children’s participation are cues for a culture of education that looks at children as incomplete persons, who can achieve a better understanding of conflict management through a passive experience of educators’ values and competence. As noticed by Boyd and Heritage (2006) with regard to the relation between doctors and patients during medical interviews, educators, in order to establish a relationship with children (as doctors do with patients), cannot avoid communicating assumptions and expectations about themselves, about children, and about their mutual relationship. Interrogative-negative questions are cues for these assumptions and expectations, which represent the cultural presuppositions of the educational encounter from the educators’ perspective. Even in primary educational settings, interaction is not completely under the educator’s control. If children understand what education expects them to do and learn, they can avoid it, or even subvert it. (Luhmann & Schorr 1979; Vanderstraeten & Biesta 2001; Vanderstraeten 2003) My data show that children can break free from the constraints of interrogative-negative questions challenging their presuppositions. Like all types of social activities (Peräkilä 2006), interrogative-negative question-answer adjacency pairs in educational settings are co-constructions, which necessarily involve complementary actions on the part of both educators and children. In this study, I have focused on the limits of the empowerment of children’s voice as is done in interaction, an often neglected aspect of research in education. I have focused on the paradox of a monophonic promotion of polyphony, where asymmetries of power between educators and children end up excluding children from active participation.



Chapter 6.  Observing the paradox 

Throughout the chapter, I have pointed out several questions. What is the importance of learning and doing in a way that meets the educator’s expectations? To what extent do the cues for an educational relationship (where adults are the competent persons who educate not-yet-complete persons) influence children’s expectations and their attitude toward the interaction? Can the paradox of a polyphonic dialogue emerging from monophonic educational discourse be solved? To sum up, the basic question concerns the absolute importance of role performances within educational interactions, and its relation to the empowerment of children’s voices. Good intentions and theories are just a part of the picture. I believe that further research must be conducted at the intersection of linguistic and social structures of educational practices in order to transform educational practices from monophonic discourses to polyphonic dialogues. Annex 1: Italian originals of examples 1 & 2 (Example 1) 1 Edu: scusate ma: (.) non facciamo confusione (0.7) la decisione che Raggi partecipa ad uno e all’altro gruppo è stata presa questa mattina in aula, no? 2 Pia(G1): però noi: 3 Edu: se dovevate dire: (0.3) dovevate dirlo al mattino; hh non vi siete sentiti prendere la decisione? 4 Lucia(G1): °no è che:° 5 Edu: questo è un problema hh vostro, non credete? 6 Raggi(G2): sì, >ma poi l’abbiamo detto [anche-]< 7 Edu: [non] mi interessa, questo è un problema vostro, che sta portando via dieci minuti (.) è inutile che alzate la mano perché in un quarto d’ora non siamo riusciti a trovare una soluzione a un problema: banale (Example 2) 1 Edu: ma: questo vostro modo di fare aiuta a fare il lavoro? 2 Sara: no 3 Edu: e: non potete trovare una soluzione? 4 Luca: secondo il mio punto di vista no 5 Edu: mh? secondo te è impossibile trovare una soluzione? non è possibile per voi fare niente assieme? 6 Luca: è vero (.) non è possibile 7 Edu. ah, o:k (.) passo dopo

 Federico Farini

Annex 2: Transcription conventions []

Brackets mark the start and end of overlapping speech

Text

Underlining indicates emphasis; the extent of underlining within individual words locates emphasis

0text0

Degree signs enclose hearably quieter speech

(0.4)

Numbers in round brackets measure pauses in tenths of a second

(.)

A micropause, hearable but too short to measure

Te:xt

Colons show degrees of elongation of the prior sound

h:

Aspiration (out-breaths); proportionally as for colons

.h:

Inspiration (in-breaths); proportionally as for colons

Tex-

Hyphens mark a cut-off of the preceding sound

>text
Y’all didn’t have a baile M2: folklórico?