Catherine Kerbrat-Orecchioni39
9Conversation and interaction
Abstract: In an interactionist perspective, communication is no longer conceived of as a transfer of information from an encoding subject to a decoding recipient but as a process of co-construction implying constant re-adjustments between participants. Initially anchored in sociology, this approach envisages the various types of discourse as socially situated practices however without overlooking its cognitive aspects. In this chapter, we first give an overview of the trends of research that inspired the analysis of verbal interactions (Goffmanian sociology, ethnography of communication, ethnomethodology and conversation analysis, interactionist psychology and Relevance theory). Then we turn to a number of methodological issues raised by this approach, such as the nature of data, the way the context should be handled and how the relevant units are to be separated from each other. Two types of communicative interactions illustrate these problems in this chapter: exchanges in small independant shops in France and TV-broadcasted political debates.
Keywords: context, TV debates, communicative interactions
1Introduction
Whereas one may well follow Gumperz (1996: 374) in assuming that “research on verbal communication since the 1960s has made fundamental contributions to our understanding of how language works in everyday encounters”, the reverse can be held as well: research on how language works in everyday encounters has made fundamental contributions to our understanding of verbal communication. Since decades, under the influence of Erving Goffman and Harvey Sacks, the study of communication has merged to a large extent with that of real exchanges in everyday situations, that is, with the study of verbal interactions (their most ‘ordinary’ type being conversations).
In a very general way, the concept of interaction refers to the action that two (or more) objects or phenomena perform on one another. It is a somehow ‘nomadic’ concept, which appeared initially in natural and life sciences and was adopted later in the humanities, starting from the second half of the twentieth century, in order to describe communicative interactions. Yet regardless of whether it’s about physical particles or human beings, we are talking about a system of mutual influences. In the case of verbal interactions, these influences arise by means of language. Hence, the word ‘interaction’, which was originally about processes of actions and reactions, came to refer to the communicative events themselves through a metonymic lexical change, communicative events being by excellence the scene where these mutual influences occur:
“For the purpose of this report, interaction (that is, face-to-face interaction) may be roughly defined as the reciprocal influence of individuals upon one another’s actions when in one another’s immediate physical presence. An interaction may be defined as all the interaction which occurs throughout any one occasion when a given set of individuals are in one another’s continuous presence; the term ‘an encounter’ would do as well.” (Goffman 1959: 15)
When conceived of as interaction, communication bears the following properties:
(i)It is a reciprocal process: not only do the participants intervene each at their turn, but at every time the one holding the position of hearer has an obligation to ensure the constant maintenance of the communication circuit through the production of verbal and non verbal signals of listening (‘back channelling’). More broadly, the various participants have to coordinate their behaviour along the verbal exchange in order to build up in common this particular object, an interactional sample, which is then the result of a collaborative work (an ‘interactional achievement’, Schegloff 1982).
(ii)This process aims not only at exchanging information and thus at modifying the cognitive system of the participants but also at orienting their behaviour. In a classroom or in a meeting room, in a shop or in a workshop, the circulation of information is inextricably related to the coordination of the actions performed by each participant. This pragmatic, or ‘praxeological’, conception of communication finds its roots in the theory of Speech Acts founded by John Austin (1962), which is reassessed in an interactional perspective.
(iii)Communication so envisaged develops upon a set of knowledge and know-hows that the participants can mobilize in the situation where they stand. This set forms their global communicative competence (see 2.1.2 below), which involves not only the knowledge of the language in which the exchange is made (together with the kinetic units that supplement the verbal signs, since oral interactions are ‘multimodal’) but also the knowledge of all sorts of principles and rules at play at the various levels at which interactions work, such as the alternation of turns, the sequence of replies, the management of topics, the management of interpersonal relations (itself in turn implying the command of the system of politeness and of ritual constraints), etc. Scholars studying interaction generally start form the assumption that human beings are significantly different from one another and that they each have a different personal cognitive equipment; therefore, admittedly, these competences are partly ‘malleable’, that is, they can be adapted to the infinite variety of interlocutors and of situations of interlocution. The co-construction of interaction can only happen at the cost of constant ‘negotiations’, and all components of interaction can be negotiated in the course of its occurrence (cf. Kerbrat-Orecchioni 2005).
(iv)In contrast with the purely codic conception of communication (i.e. a sender and a recipient exchange information via a common language which permits its coding and decoding), the interactionist will grant a primordial importance to the context, both narrow and large, in which the utterances exchanged inscribe themselves, and insist on their property of being socially situated. It is not surprising, then, that this new way of looking at communicative phenomena was mostly adopted, initially, by scholars in the field of sociology.
2Social aspects
2.1Historical overview
Being interdisciplinary since its beginning, the interactionist approach to communication initially found its theoretical methodological consistency within American sociology, in line with the works of the Chicago school on symbolic interactionism and of Alfred Schütz in phenomenological sociology. This vast field or research is made of numerous trends, among which the following main perspectives deserve specific mention.
2.1.1Erwing Goffman
Goffman is a solitary and independent-minded scholar and he doesn’t really belong to a specific research trend. Nonetheless, he dramatically influenced the way scholars think of everyday interactions. He conceived of them as kinds of theatrical performances, in which the participants seek above all to save and manage their mutual faces (this view is further developed by Penelope Brown and Stephen Levinson, who propose a full-fledged theory of politeness seen as ‘face-work’ Brown & Levinson 1978, 1987). Goffman’s import about the ‘presentation of the self’, ‘interaction rituals’ and the various modes of participation and commitment in encounters is considerable (Goffman 1959, 1967). According to Duncan & Fiske (1985: 68), “No other investigator has so compellingly illuminated the subtlety and complexity of interactional processes”.
2.1.2Ethnography of communication
Reacting against the Chomskyan conception of the ‘ideal speaker’, Hymes (1962) publishes a paper titled The Ethnography of Speaking where he claims that speech is, above all, a communicative process to be studied in its context as ethnographers do. Together with Gumperz and other colleagues, Hymes establishes the foundations of a new discipline, ‘ethnography of communication’, whose objective is to uncover the set of norms which underlie the way in which the various types of interactions seen in human societies work. The main characteristics of this trend are:
(i)an extension of the concept of ‘competence’ to ‘communicative competence’, linguistic competence being only one component, among others, of communicative competence;
(ii)the importance granted to the context, both physical and socio-cultural: communicative norms and discursive practices are always to be envisaged in relation to the ‘frame’ and the ‘site’ where they take place (Hymes’ SPEAKING model inventories the various components of the context: ‘setting’, ‘participants’, ‘ends’, ‘acts’, ‘key’, ‘instrumentalities’, ‘norms’ and ‘genre’);
(iii)The commitment to an inductive, empirical and naturalistic approach. Speech events are in his view to be observed in their natural environment and the analyses must be based on patient fieldwork.
(iv)A special attention is given to phenomena of variation, not only between speech events but also between various discursive communities (in a cross-cultural view) and between groups within these communities (sub-cultural variations).
2.1.3Ethnomethodology and Conversation analysis
For Harold Garfinkel, the founder of ethnomethodology and inventor of the term, ethnomethodology pursues the objective of describing the ‘methods’ (i.e. procedures, types of knowledge and know-hows) that are used by members of a given society in order to adequately manage all the communicative problems that they need to solve in their everyday life (Garfinkel 1967). In this perspective, the norms that underlie social behaviour are constantly refreshed and regenerated in everyday practice within the endless movement of the interactive construction of social order; life in society appears thus as a ‘continuous achievement. The ethnomethodological approach applies in principle to all domains of social life. But conversations appear as a privileged place to observe this phenomenon because they appear as a sort of micro social system on their own. With the impulse of Sacks’ ‘lectures on conversation’ (delivered from 1964 to 1972 but published only posthumously in 1992 under the direction of Gail Jefferson), a school of ethnomethodology focused on talk-in-interaction and became an autonomous field of research, namely ‘conversation analysis’ (CA). Their goal is to show that ordinary conversations are organized, and to detail very precisely how they are so, showing how participants make use of specific techniques in order to accomplish in common the various tasks they have to perform: ensure turn-taking with minimizing gaps and overlaps (Sacks, Schegloff & Jefferson 1974), repair the occasional problems in the exchange, achieve the negotiation of topics and of conversations openings and closings, lead together a narration or a description, etc. The objective is thus to show the ‘technology’ of talk-in-interaction through the cautious observation of carefully recorded and transcribed samples.
2.2Interaction and social context
These various perspectives have in common that they view discourses as socially situated practices and that they consider the relation between discourse and context as a dynamic one. Of course, every discursive event involves certain ‘ways of speaking’. From the point of view of its production, the speaker mobilizes a load of knowledge that allows him to “communicate effectively in culturally different settings” (Gumperz & Hymes 1972: vi); from the point of view of reception, the interlocutor will make use of the same array of knowledge in order to interpret the utterances submitted to him. Yet he may also extract from the very text of the interaction those relevant contextual assumptions which he didn’t previously hold, thanks to what Gumperz (1982a) calls contextualisation cues. The context must therefore not be understood as a fixed frame within which the discourse would unfold, but is on the contrary constantly remodelled by the talk-in-interaction. In this view, the discourse is an activity both conditioned by the context and transformative of it; existing prior to the interaction, the situation is permanently redefined by the conversational events: “Context shapes language and language shapes context. [...] Context is not simply a constraint on language, but also a product of language use.” (Duranti & Goodwin 1992: 30–1).
The idea that discourse fashions the context as much as the context fashions discourse is a leitmotiv in today’s interactionist literature. It can be illustrated by the example of the selection of the pronoun tu or vous (casual and formal forms for you) when addressing someone in French: the choice is dictated by the state of the relationship between the interactants when the exchange takes place and by various situational factors. However as the rules determining the use of tu and vous are somewhat unclear in some cases, the speaker can occasionally try to modify the relashionship, for example in tempting to come closer to the partner by switching from vous to tu. In most cases, nonetheless, the discourse’s latitude to remodel the context is relatively small and the choice of the pronoun is very strongly constrained by the situation of communication. Thus, the possibilities of remodelling discourse are in fact relative and their amplitude varies from one type of discourse to the other, and from one kind of linguistic phenomenon to the other. Everything is not equally ‘negotiable’ in interaction.
Let us add that the particular importance given to the context in this framework however resulted in a renewed interest for issues such as:
(i)the question of interaction types which can be described according to a number of oppositions (the nature of the participative framework (Goffman 1981), the exchange’s objectives, its degree of formality, etc.); it is assumed that each type unfolds according to a ‘script’, more or less rigid or flexible, which corresponds to the succession of the activities that compose the communicative event, in relation with the ‘roles’ involved:
“A script is a structure that describes appropriate sequences of events in a particular context. […] Thus, a script is a predetermined stereotyped sequence of actions that defines a wellknown situation. […] Every script has associated with it a number of roles.” (Schank & Abelson 1977: 41)
If at the beginning Conversational Analysis (henceforth CA) widely privileged ordinary conversations, which are considered the basic form of verbal interactions in general, it directed later its attention towards more formal and institutional forms of talk-in-interaction. The literature about, for example, interactions in classrooms, commercial and service settings, medical contexts, clinical and therapeutic exchanges, judiciary and parliamentary contexts, or within the realm of media interaction, is today wide.
(ii)the question of the participants’ identities and their socio-affective relation (notably their relation of power). These identities and relations, as all other components of interaction, are partly determined by the context and partly (re)constructed and negotiated along the flow of the exchange (Gumperz 1982b; Thornborrow 2002).
3Cognitive aspects
The co-construction of interaction by the participants forces them to take into account the narrow context (the situation of communication) and the wide context (socio-cultural context); but it also implies a cognitive work in order to achieve the operations of online production and interpretation of the semiotic material that gets exchanged in the course of the interaction. This cognitive work is achieved by mobilizing the participants’ ‘competences’, that is, procedural knowledge which allows them to perform these operations successfully. Some approaches focus specifically on these cognitive processes which underlie verbal and non verbal activities conducted by the individuals engaged in such collective tasks. We take here the example of two such approaches which emerged and were developed within very different contexts and seek to achieve different aims.
3.1The Mental Research Institute of Palo Alto
Following the legacy of Gregory Bateson, a group of scholars in California developed a theory of communication mostly intended to solve some problems of pathological communication, with the assumption that these impairments result from a circular causality involving a malfunction of the global relational system within which the individual is inscribed. Thus in this view, a therapy must act on this very system rather than on the individual. However, regardless of the therapeutic goals, some of the concepts that were designed in this framework (see Watzlawick et al. 1967) are easily transposable from pathological communication to normal communication and hence are useful for the description of ordinary, everyday, interaction. In particular, these scholars distinguish between ‘symmetrical’ and ‘complementary’ communication, between the level of ‘contents’ and that of ‘relation’, and they bring about the notion of ‘double bind’, originally associated to the genesis of certain kinds of schizophrenia, into the study of social life. In fact, arguably, we are constantly facing situations involving double binds and subject to contradictory imperatives, for example, between negative politeness (do not disturb) and positive politeness (to attend to others’ interests, needs and wants) or between politeness (do not hurt other people) and sincerity (say straightforwardly what you really think). But in normal communication, fortunately, we only need to deal with ‘soft’ double binds, which can be solved by more or less original or routinized means.
3.2From conversational maxims to Relevance theory
According to Herbert Paul Grice (1975), conversation rests on, so to say, a tacit contract to which the participants are committed. This contract consists in observing the general ‘cooperation principle’ which deploys itself across four ‘conversational maxims’: maxim of quantity, maxim of quality, maxim of relevance (or of ‘relation’) and maxim of manner. It often happens that one or the other maxim is violated, but for most times this will result in the communication of a ‘conversational implicature’ which in return normalizes the utterance. An example will help us illustrate how these maxims work and the issues brought by their application. In the newspaper Le Monde, a journalist (named Thomas Ferenczi) replies to the criticisms raised by some readers about how an incident was reported in the journal.
Au cours des dernières semaines, Le Monde a reçu plusieurs lettres dont les signataires nous reprochent d’avoir procédé à un « travestissement de la vérité » et à un « mensonge par omission » en rendant compte de l’agression commise il y a deux mois dans le RER contre une femme policier. Ils notent que nous n’avons pas fait mention de l’origine ethnique présumée des agresseurs (lesquels, à ce jour, n’ont été ni identifiés ni arrêtés) et ils s’en indignent. Nous indiquions seulement, en effet, que la victime avait été agressée « par cinq jeunes gens ». « Il ne s’agit pas de ‹ jeunes ›, nous répliquent nos interlocuteurs, mais bel et bien de quatre jeunes maghrébins et d’un jeune noir. » Une telle précision était absente de notre article. Le Monde en effet, à la différence de certains de ses confrères, ne jugeait pas utile de préciser l’origine ethnique présumée des agresseurs tant qu’il n’était pas établi que celle-ci avait un lien avec l’acte commis. […] Pour éviter toute dérive raciste il faut en effet faire attention à ne pas suggérer, sans avoir mené une enquête sérieuse, que l’appartenance d’un individu à tel ou tel groupe contribue à expliquer son comportement. » (Le Monde du 29-12-1996)
Along the recent weeks, Le Monde received several letters of readers reproaching us to have “altered the truth” and “lied by omission” when we reported about the assault of a policewoman in the Parisian metro. They notice with irritation that we did not mention the presumed ethnic origin of the assaulters (who, up to this day, have not yet been arrested nor even identified). It is true that we only mentioned that “five young men” assaulted the victim. “They are not ‘young men’, our readers tell us, but actually four North African and one black young men”. Such a precision was absent from our article. In effect, Le Monde, contrarily to some other journals, did not consider it useful to detail the ethnic origin of the assaulters as long as it is not clear that it has a connection with the event (...). In order to avoid the risk of a racist drift, it is necessary to be careful not to suggest, without having performed a thorough investigation, that the ethnic origin of an individual contributes to explaining his behaviour (Le Monde, Sept. 29, 1996, our translation).
In Gricean terms:
(i)some readers accuse the journalist of violating the first maxim of quantity: “Make your contribution as informative as required”; that is, they accuse the journalist of having retained pieces of information when they reported the events (cf. “lying by omission”, “we only mentioned ...”). Certainly, “five young men” is less precise than “four North African and one black young men”.
(ii)The journalist’s justification is based on both the maxim of quality: “Do not say that for which you lack adequate evidence” (cf. “presumed ethnic origin”) and the maxim of relevance, to which the other maxims are subordinated: one has to provide as much as possible relevant information, that is, information which is to the point (because if not, the utterance breaks the second maxim of quantity: “do not make your contribution more informative than is required”). Yet for the journalist, making a precision about the ethnic origin of the assaulters would be irrelevant (cf. “not useful”) “as long as it is not clear that it has a connection with the event”.
The last sentence perfectly illustrates the way the Gricean maxims are put at work: if one declares that “some North Africans have committed an assault in the metro”, the recipient will automatically make the utterance relevant by construing an implicature about the link between being from a certain origin and committing assaults (an implicature that the journalist seeks precisely to avoid, cf. “suggest”); such an implicature cannot be taken as true, “without having performed a serious investigation”. Furthermore the implicature potentially leads to racist drift and thus making the point would not only be superfluous but harmful.
This example shows at the same time the descriptive and theoretical usefulness of these maxims (since discourses are produced and interpreted by applying such principles), the fact that they can enter in contradiction with one another (forcing us to tack among them), and the fact that their application is locally negotiable in the course of interaction according to the conceptions and objectives of each participant.
The suggestions by Grice have been taken further by Dan Sperber and Deirdre Wilson (1986, 1995) who elaborated a global theory of communication based on the only maxim of relevance and more precisely on a principle of ‘optimal relevance’: Within a given ‘cognitive environment’ (the set of facts that an individual is capable of representing at a moment t), the speaker will try to produce the highest possible cognitive effects at the lowest possible cost. On his side, the interlocutor, who recognizes this intention on the part of the speaker, will calculate on this basis the interpretation which will fit best (see this volume, chapters 6 and 24).
Sperber and Wilson (1986, 1995), as well as Grice, assume that the basic goal of communication is to ensure and improve the transmission of information among individuals through their cognitive systems, i.e. their apparatus dedicated to information processing. A slightly different approach is to consider the cognitive processes deployed by groups of individuals who engage into a common activity, and to observe in close detail the processes of coordination and the scenarios of decision-making. In such a perspective, cognition is seen no more as an ‘internal’ process but as a ‘distributed’ one, with two dimensions: on the one hand, within the groups of actors, and, on the other hand, in the material ecology of the environment (spatial layout, manipulated objects, artefacts, cf. Suchman 1987; Clark 1997).
From what precedes, one gathers that verbal interaction occurs between individuals, who each have their own competences and resources and who dedicate themselves to a cognitive working out of the utterances meanings. However, firstly, this work is carried out in constant interaction with the environment and the other participants to the communicative event, and secondly, the cognitive process, which is not restricted to pure logical tasks but involves also affects and emotions (the recent works on the topic have a strong focus on the emotional component), must be set visible so that it leaves traces in the interaction. These traces are what the analyst looks for and interprets in adopting a decidedly empirical approach.
4Methodological issues
Whatever the diversity of approaches and the focus of those who state their belonging to the interactionist paradigm, they all share a consensus on this fundamental fact: given that conversations and other forms of talk-in-interaction are sorts of collective improvisations, the work of describing them involves trying to identify the mechanisms and procedures which ground this co-construction activity. Such a task can be achieved only through a detailed observation of specific interactions but which are admittedly typical of the type of encounter they belong to. The analysis seeks indeed to uncover some regularities and general ways of operation: given that even the apparently most ordinary and chaotic exchanges are “systematically and strongly organized”, it is the scholar’s burden to put to light this “order” obscured by the “apparent disordeliness of natural speech” (Goodwin 1981: 55–59).
Nonetheless, these commonly admitted principles can be handled in quite different ways and give rise to divergent practices and options, notably about the following aspects:
(i)The data: if everyone agrees that the essential material of the analysis is provided by samples of discourses which actually happened, which are carefully recorded and transcribed (and thus not by simple notes or answers to questionnaires), this principle can be applied more or less stricly. Purists will only accept ‘naturally occurring’ data, whereas other scholars will also take in consideration some kinds of elicited data (that is, data triggered by the scholar himself). There are also recurring debates about whether it is legitimate or not to enrich the analysis by studying various types of data which enables to better understand the naturally occurring interaction, such as fiction (in particular movie dialogues) or ‘follow-up interviews’ with the speaker.
(ii)The context: Gumperz (in Eerdmans et al. 2002: 22), among others, argues that the analyst “always needs a priori analysis of context” and has to begin by collecting as much ethnographic information as possible about the site being investigated. However, promoters of CA claim on the contrary that external data has to be kept aside; for them, the context must be entirely considered “something endogenously generated within the talk of the participants and, indeed, as something created in and through that talk.” (Heritage 1984: 283)
In such a perspective, one has to take into account the features of the context only when the conversational behaviour of the participants inscribes them within the very text of the interaction itself.
(iii)The focus of analysis: the scholar feeling close to CA will mostly focus on the ‘conversational machinery’ and on micro-level phenomena such as turn-taking, adjacency pairs and the sequential organization of talk, repairs management, joint productions, etc. But one may wish to focus on larger units and try to identify the script that underlies the whole of the interaction, wonder about some aspects of contents, conflicts that may occur within it, the interpersonal relationships that form in it, argumentative processes, the circulation of knowledge ...
(iv)In the end, the essential question raised by the study of communication within discursive productions in the context of conversational exchanges is the following: what does it mean exactly to describe an interaction, if describing always amount to interpreting? We will get back to this after having looked at two very distinct types of interaction: exchanges in small independent shops, and political debates broadcasted on TV.
5Case studies
Let us recall that:
(i)The researcher never operates on the event itself (raw data) but on recordings (primary data), which only provide a partial grasping of the event (even more so if it’s only an audio recording). She needs to stabilize the object of study by transcribing it (secondary data), and it’s mostly starting from this artefact that the analysis is pursued, with constant checks of the original recording.
(ii)The aspects on which the analysis focuses can vary according to the scholar’s interests, the tools at his disposal, and also to the properties of the considered interaction – as we shall see in the following two examples.
5.1Small independent shops
We will now say a few words of a research about interactions in various types of small independent shops (bakery, butchery, florist, tobacco shop, shoe shop, open market, garage, etc.) conducted in Lyon (France) but also in several Arabic-speaking countries and in Vietnam, which allows for cross-cultural comparisons (see Kerbrat-Orecchioni & Traverso 2008).
As we are dealing with a type of interaction where the ‘setting’ is a fundamental component, we focussed first on the spatial layout and on the physical characteristics of the place where the exchanges occurred. Then we proceeded to analysing the corpus, distinguishing between the following levels of analysis.
(i)The organisational level, where the ‘script’ of these interactions is revealed (the unfolding of both verbal and non-verbal activities accomplished by the individuals bearing the two interactional roles of seller and buyer) with a particular emphasis on the management of openings and closings of the encounters.
(ii)The transactional level, which was studied more specifically from the angle of how the request and the delivery of the product are realized (with various possible negotiations) and how the request and realization of the payment are done (with bargaining in some cultures).
(iii)The relational level, which concerns the relation between the two interactional roles. In the case of a seller and a client engaged in a ‘complementary’ interaction, this relation is complex, since the seller is to some extent an ‘expert’ (thus occupying a higher hierarchical position), but at the ‘service’ of the client and in this respect occupying a lower position.
We examined how the participants constructed their identities of seller and buyer, what was their mutual relation (considering in particular how that relation evolved through several episodes that diverge slightly from the routine of the business exchange: small talk, moments of humour, of tension or even of conflict), and the manifestations of politeness, which can take various forms – ritual acts of greeting, thanking, sometimes apology, but also indirect formulations of requests with the conditional or with the use of a minimizing item (“je voudrais un petit bifteck” [lit. “I would like a little beefsteak”], which does not imply anything about the physical dimensions of the steak).40 Forms of politeness are always somehow pervasive in French businesses: on the average, more than half of the semiotic material exchanged in such sites has only the function of facilitating the interaction, whose objective however, in principle, is only utilitarian (Kerbrat-Orecchioni 2005b).
5.2Political debates on TV broadcasts
Political debates is a communicative event than can have particularly variable forms and we won’t seek here to find out their general properties. We examine a sample of a debate of a particular subtype: the confrontation of François Mitterrand, then President of France, with his challenger for the presidential elections in 1988, Jacques Chirac, who also bears the function of Prime Minister (in the context of ‘cohabitation’ where the parliament’s majority is not that of the President).
As these debates are broadcasted (media) interactions, they are inscribed in a complex participative format with two communication circuits. The first one is situated on the TV stage, involving four persons corresponding to two different interactional roles (the two debaters forming a ‘conflicting dyad’ and the two presenters forming a ‘solidary dyad’). Additionally, these exchanges are taken in an embedding circuit between the stage and the public; this communication is not reciprocal but unilateral: the public, despite being the ultimate addressee, lacks the capacity of intervening online.
In this example, as it happens, François Mitterrand de facto occupies the highest position, which he attempts to mark by choosing to use a specific term to address his opponent. The rules in France impose that the debaters address each other with the formulation ‘monsieur (i.e. mr.) + name’. This egalitarian and symmetrical wording categorizes them as mere candidates: this is, actually, their relevant contextual identity. It is in fact the only expression debaters have ever used in presidential debates (six to this day), banning titles, even in the mouth of Mitterrand who addressed a minister in 1974 and a President in 1981. It is also what Jacques Chirac does in this very debate. But François Mitterrand introduces a dissymmetry in calling Jacques Chirac “Monsieur le premier ministre” (‘Mr. Prime Minister’) which of course lowers the status of Chirac in the debate by underlying his subordinate function. This takes Chirac to protest and remind Mitterrand that this evening (the relevant identities are indeed ‘situated’), Mitterrand is a mere candidate, equally to himself. He concludes by an utterance which is involuntarily ambiguous, but anyhow accompanied by a little smile of satisfaction (premature, judging by what follows):
| Jacques Chirac: | permettez-moi juste de vous dire que ce soir/ (.) je ne suis pas/ le Premier ministre (.) et vous n’êtes pas/ le président de la République (.) nous sommes (.) deux candidats/ (.) à égalité/ (.) et qui se soumettent au jugement des Français (.) le seul qui compte (.) vous me permettrez donc de vous appeler monsieur Mitterrand |
| François Mitterrand: | mais vous avez tout à fait raison/ monsieur le Premier ministre |
| Jacques Chirac: | Allow me simply to tell you that tonight, I am not the Prime Minister, and you are not the President of the Republic. We are two equal candidates, who submit themselves to the judgement of the Frenchs, the only one that counts. You will thus allow that I call you Mr. Mitterrand. |
| François Mitterrand: | But you are perfectly right, Mr. Prime Minister. |
After this reaction, François Mitterrand sits up straight and gives a little mischievous smile. The presenter (Michèle Cotta) can’t help but smile, while looking around while all remain silent for 4 seconds, until the other animator intervenes and introduces the next topic of the debate without any explicit reference at the incident.
Commentators noticed the contradiction embedded in the reply by Mitterrand, but upon closer scrutiny the reply can be interpreted in two different ways depending on the ‘scope’ granted to “you are perfectly right”. Right about what exactly?
The most immediate interpretation is that the evaluation scopes over Chirac’s preceding utterances and mostly over its nucleus, “we are two equal candidates”. According to this interpretation, “you are perfectly right” means: “you are right in saying that we are two equal candidates, which imply that we both need to use the expression monsieur + name to call each other”. In this interpretation there is a contradiction between the contents of the utterance and the term actually used, a joke which triggers smile or laughter.
But at the same time François Mitterrand protects himself and finds the possibility to get away with the discursive scandal of a contradiction by suggesting another interpretation. “You are perfectly right” can in fact well be interpreted as anchoring only to the end of the preceding turn, i.e. “you will thus allow that I call you Mr. Mitterrand” which would then mean something like “I will go on talking and with your permission I will go on calling you Mr. Mitterrand” (without hinting at what the other is supposed to do). This other interpretation is made possible by the fact that Jacques Chirac omitted to explicitly bring about the inference that his reasoning entails: “you will allow that I call you Mr. Mitterrand [and I will ask you to please call me Mr. Chirac]” (it’s more like an indirect request than a simple demand of permission). The inference is quite obvious but unfortunately for Jacques Chirac, one can always be deaf to inferences. Mitterrand takes advantage of this killing two birds with one stone: he constructs an image of himself as a polite debater (calling someone by a title indicates in principle the respectful attitude of the speaker) and an open-minded one, who thinks that everyone is allowed to express himself as he likes (conveying something like “I don’t see it a problem that you call me with my name, and therefore let me call you as it pleases me, that is, with your title”). The true reason of this choice, of course, is that he marks and repeats all along the debate the difference of status between Chirac and himself. But at the same time and most of all (since this remained in the people’s memories), he amuses the audience with this joke and plays a good trick to the opponent, victim of the clumsiness of his expression.
6Conclusion
In order to describe a sample communicative event, one needs first to observe with close scrutiny the verbal and non verbal material which composes it (see also this volume, chapters 3 and 17). In the case above, for example, Mitterrand’s discreet smile, the more blooming smile of the presenter, and the 4-second long silence that follows the candidate’s joke are certainly meaningful, but of what?
According to Schegloff (1997: 183–184), ‘talk-in-interaction’ offers to interpretation a ‘leverage’ in the fact that the meaning of exchanged utterances is “embodied and displayed” in and by the behaviour of the interactants. In the perspective of CA, thus, to describe a sample of interaction amounts to gathering these facts which are made ‘available’ and ‘accountable’ by and for the participants of the interaction, but also, simultaneously, for the external observers attempting at describing it.
Yet what is available are signifiers, which are not transparent. It’s not enough to scrutinize even very carefully the collected data in order to see their meaning miraculously emerge. In fact, the principle of ‘accountability’ merely reformulates the semiotic principle: every participant delivers signifiers (whatever they may be called: ‘cues’, ‘markers’, ‘indicators’ or ‘features’) to others, who would then simply have to interpret them. Interpretation, however, amounts to an indeed complex activity which in fact mobilizes all sorts of knowledge (about the language, the functioning of communication, the specific rules of the interactional genre, the context, both narrow and wide, etc.) and is certainly not of the straightforward kind sometimes assumed. The analyst’s task is to review the work that was performed online by the participants, a task which seems only possible to achieve if he has the same interpretive resources at his disposal. However, while it is plausible that all participants share more or less the same linguistic and communicative competence, their ‘encyclopaedias’ vary, in particular in the case of media communication where the audience is numerous and heterogeneous: depending on their degree of expertise in the political domain, they will be able or not to decipher the multiple allusions conveyed by the debaters’ utterances, not to mention their ideological preferences which will inevitably influence their interpretations.
The analyst needs to acquire the means of predicting these more or less diverging interpretations. He is a super-interpret who needs to ‘adopt the point of view of members’, that is, of all ratified participants, be they emitters and recipients or confined to the role of reception. This implies that he has to take into account not only what happens in the interaction but also the normative expectations associated with it, since the interactional effects of a given span of interaction results at the same time from what the participants do and what they are expected to do given their status, their role, their identity and all the contextual parameters preexisting to the event itself, which does not arise ex nihilo.
It would be absurd to suggest that the context, the language or the rules of the exchange are entirely created at every step of the discourse. However, observations on authentic data and about the functioning of communication grasped in situ have the advantage to shed light on the flexible character of communicative norms, on the adaptability of communicating subjects and the vast unpredictability of the communicated objects:
“Conversation is like playing tennis with a ball made of Krazy Putty that keeps coming back over the net in a different shape.” (David Lodge, Small World, Penguin Books, 1985: 25)
References
Austin, John L. 1962. How to do Things with Words. Oxford: Clarendon.
Brown, Penelope & Stephen C. Levinson. 1978. Universals in language use: Politenesse phenomena. In Eesther Goody (ed.), Questions and Politeness. Strategies in Social Interaction, 56–289. Cambridge: Cambridge University Press.
Brown, Penelope & Stephen C. Levinson. 1987. Politeness. Some Universals in Language Use, Cambridge: Cambridge University Press.
Clark, Herbert H. 1996. Using Language. Cambridge: Cambridge University Press.
Duncan, Starkey & Donald Winslow Fisk. 1985. Interaction Structure and Strategy. Cambridge: Cambridge University Press.
Duranti, Alessandro & Charles Goodwin (eds.). 1992. Rethinking Context. Language as an Interactive Phenomenon. Cambridge: Cambridge University Press.
Eerdmans, Susan L., Carlo L. Prevignano & Paul J. Thibault (eds.). 2002. Language and Interaction. Discussions with John Gumperz. Amsterdam &Philadelphia: John Benjamins.
Garfinkel, Harold. 1967. Studies in Ethnomethdology. Englewood Cliffs, N.J.: Prentice Hall.
Goffman, Erving. 1959. The Presentation of Self in Everyday Life. New York: Doubleday Anchor Books.
Goffman, Erving. 1967. Interaction Ritual: Essays on Face-to-Face Behavior. Chicago: Aldine Publishing Company.
Goffman, Erving. 1981. Forms of Talk. Oxford: Blackwell.
Goodwin, Charles. 1981. Conversational Organization: Interaction between Speakers and Hearers. New York: Academic Press.
Grice, H. Paul. 1975. Logic and Conversation. In Peter Cole & Jerry Morgan (eds.), Syntax and Semantics, Speech Acts, Vol. 3, 41–58. New York: Academic Press.
Gumperz, John. 1982a. Discourse Strategies. Cambridge: CUP.
Gumperz, John (ed.). 1982b. Language and Social Identity. Cambridge: Cambridge University Press.
Gumperz, John. 1996. The linguistic and cultural relativity of conversational inference. In John Gumperz & Stephen Levinson (eds.), Rethinking Linguistic Relativity, 374–406. Cambridge: Cambridge University Press.
Gumperz, John J. & Dell Hymes (eds.). 1972. Directions in Sociolinguistics. The Ethnography of Communication. New York: Holt, Rinehart and Winston.
Heritage, John. 1984. Garfinkel and Ethnomethodology. Cambridge: Polity Press.
Hymes, Dell. 1962. The Ethnography of Speaking. In Thomas Gladwin & William C. Sturtevant (eds.), Anthropology and Human Behavior, 13–53, Washington, D.C.: The Anthropological Society of Washington.
Kerbrat-Orecchioni, Catherine. 1990 [1994]. Les interactions verbales. Paris: A. Colin.
Kerbrat-Orecchioni, Catherine. 2005a. Le discours en interaction. Paris: A. Colin.
Kerbrat-Orecchioni, Catherine. 2005b. Politeness in France: How to buy bread politely. In Leo Hickey & Miranda Stewart (eds.), Politeness in Europ., 29–44. Clevedon: Multilingual Matters.
Kerbrat-Orecchioni, Catherine & Véronique Traverso (eds.). 2008. Les interactions en site commercial: invariants et variations. Lyon: ENS Éditions.
Sacks, Harvey. 1992. Lectures on Conversation, 2 Vol. Oxford: Blackwell.
Sacks, Harvey, Emanuel A. Schegloff & Gail Jefferson. 1974. A simplest systematics for the organization of turn-taking for conversation. Language 50. 696–735.
Schank, Roger C. & Robert P. Abelson. 1977. Scripts, Plans, Goals and Understanding. Hillsdale, NJ: Lawrence Erlbaum.
Schegloff, Emanuel. 1982. Discourse as an interactional achievement: some uses of ‘uh huh’ and other things that come between sentences. In Deborah Tannen (ed.), Analyzing discourse: Text and Talk, 71–93. Washington, DC: Georgetown University Press.
Schegloff, Emanuel. 1997. Whose text? Whose context?. Discourse & Society 8(2). 165–187.
Sperber, Dan & Dierdre Wilson. 1986. Relevance: Communication and Cognition. Oxford: Blackwell.
Suchman, Lucy A. 1987. Plans and situated actions: the problem of human-machine communication. New York: Cambridge University Press.
Thornborrow, Joanna. 2002. Power Talk: Language and Interaction in Institutional Discourse. Upper Saddle River: Prentice Hall.
Watzlawick, Paul, Janet H. Beavin & Don D. Jackson. 1967. Pragmatics of Human Communication. Palo Alto: Mental Research Institute.