Context and Appropriateness: Micro meets macro (Pragmatics & Beyond New Series)

Context and Appropriateness

Pragmatics & Beyond New Series (P&BNS) Pragmatics & Beyond New Series is a continuation of Pragmatics & Beyond and its Companion Series. The New Series offers a selection of high quality work covering the full richness of Pragmatics as an interdisciplinary field, within language sciences.

Editor Andreas H. Jucker

University of Zurich, English Department Plattenstrasse 47, CH-8032 Zurich, Switzerland e-mail: [email protected]

Associate Editors Jacob L. Mey

University of Southern Denmark

Herman Parret

Jef Verschueren

Susan C. Herring

Emanuel A. Schegloff

Belgian National Science Foundation, Universities of Louvain and Antwerp

Belgian National Science Foundation, University of Antwerp

Editorial Board Shoshana Blum-Kulka Hebrew University of Jerusalem

Jean Caron

Université de Poitiers

Indiana University

Masako K. Hiraga

St.Paul’s (Rikkyo) University

University of California at Los Angeles

Deborah Schiffrin

David Holdcroft

Georgetown University

Sachiko Ide

Kobe City University of Foreign Studies

Sandra A. Thompson

Thorstein Fretheim

Catherine KerbratOrecchioni

John C. Heritage

Claudia de Lemos

Teun A. van Dijk

Marina Sbisà

Richard J. Watts

Robyn Carston

University College London

Bruce Fraser

Boston University University of Trondheim University of California at Los Angeles

University of Leeds Japan Women’s University

University of Lyon 2 University of Campinas, Brazil University of Trieste

Volume 162 Context and Appropriateness. Micro meets macro Anita Fetzer (ed.)

Paul Osamu Takahara

University of California at Santa Barbara Pompeu Fabra, Barcelona University of Berne

Context and Appropriateness Micro meets macro

Edited by

Anita Fetzer Lueneburg University

John Benjamins Publishing Company Amsterdam / Philadelphia

8

TM

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

Library of Congress Cataloging-in-Publication Data Context and appropriateness : micro meets macro / edited by Anita Fetzer. p. cm. -- (Pragmatics & beyond new series, issn 0922-842X ; v. 162) Includes bibliographical references and index. 1. Discourse analysis. 2. Context (Linguistics) 3. Speech acts (Linguistics) 4. Acceptability (Linguistics) 5. Pragmatics. I. Fetzer, Anita, 19582007 P302.C6225 306.44--dc22 isbn 978 90 272 5406 1 (Hb; alk. paper)

2007015637

© 2007 – 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 Context, contexts and appropriateness Anita Fetzer

3

Part I. Bridging problems between context and appropriateness Similar situations Varol Akman

31

Appropriateness and felicity conditions: A theoretical issue Etsuko Oishi

55

Appropriateness: An adaptive view Thanh Nyan

79

Party II. Bridging problems between communicative action and appropriateness If I may say so: Indexing appropriateness in dialogue Anita Fetzer

115

The appropriateness of questions Annette Becker

147

Cooperative conflict and evasive language: The case of the 9–11 commission hearings Lawrence N. Berlin

167

vi

Table of contents

Part III. Bridging problems between micro and macro The attenuating conditional: Context, appropriateness and interaction Johanna Miecznikowski and Carla Bazzanella

203

Collaborative use of contrastive markers: Contextual and co-textual implications Francesca Carota

235

Index

261

Introduction

Context, contexts and appropriateness Anita Fetzer

University of Lueneburg, Germany

Communication is both context-creating and context-dependent

(Bateson 1972)

Introductory comments The papers collected in this edited volume on ‘Context and Appropriateness’ result from a panel organized on this topic at the 9th IPrA Conference held in Riva del Garda in the year 2005.1 As was the case with the presentations invited for the panel, this volume addresses context and appropriateness from a variety of theoretical and applied perspectives. Thus, the papers range from the research paradigms of philosophy of language, speech act theory, sociopragmatics, cognitive pragmatics and critical discourse analysis. The variety of topics, which the individual contributions cover, allows for a refined bottom-up and top-down analysis of the concepts of context and appropriateness as well of their multifaceted connectedness. It is the purpose of this introduction to examine and contextualize the concepts of context, contexts and appropriateness, to give an overview on how the notions are reflected in different – if not sometimes diverging – approaches to language and language use, and to locate the individual papers’ contributions in the overall statement made with this book.

1.

Contexts in context

Context has become a major field of research in information technology, engineering, science, and in the social sciences, arts and humanities, to name but a 1. The editor is deeply grateful to the participants of the panel, to the discussant Andreas Jucker and to the audience for their helpful input. Special thanks go to Isja Conen of John Benjamins and to the reviewers of the papers, for their insightful comments.

4

Anita Fetzer

few. Not only is the concept as such the object of investigation but so is its application to a multitude of domains, such as natural- and non-natural language communication, computer-mediated interaction and information technology, robotics, social-action based analyses or literary analyses. The heterogeneous nature of context and the context-dependence of the concept itself has made it almost impossible for the scientific community to agree on one commonly shared definition or one commonly shared accepted theoretical perspective, and frequently, only a minute aspect of context is described, modelled or formalized (cf. the interdisciplinary conferences on context: Akman et al. 2001; Blackburn et al. 2003; Bouquet et al. 1999). Because of its multifaceted nature and inherent complexity, context is no longer looked upon as an analytic prime but rather seen from a parts-whole perspective as an entity containing sub-entities (or sub-contexts). The multilayered outlook on context contains a number of different perspectives. First, context is conceived as a frame whose job it is to frame content by delimiting that content while at the same time being framed and delimited by less immediate adjacent frames. The nature of the connectedness between those different frames is a structured whole composed of individual interconnected frames (Goffman 1986). The gestalt-psychological figure-ground scenario prevails in psychological and psycholinguistic perspectives on context. It has also been adopted to cognitive pragmatics as is reflected in the relevance-theoretic conception of context as an onion, metaphorically speaking. Sperber and Wilson not only point out the interconnected nature of the individual layers but also stress the fact that their order of inclusion corresponds to their order of accessibility (Sperber & Wilson 1986). This is of particular importance to inferencing and to the calculation of implicatures. Second, context is seen as a dynamic construct which is interactionally organized in and through the process of communication. This view prevails in the domains of ethnomethodology (Garfinkel 1994; Goodwin & Duranti 1992; Heritage 1984; Schegloff 1992), interactional sociolinguistics (Gumperz 1996, 2003) and sociopragmatics (Fetzer 1999, 2004), where context is assigned the dual status of process and product. The dynamic outlook is based on (1) the premise of indexicality of social action, and (2) the (joint) construction of a common context. In those primarily qualitatively oriented paradigms, context is intrinsically connected with adjacency pair, conditional relevance and the turn-taking system on the micro level, and with institutional talk on the macro level, whose order is captured through context-independent and context-sensitive constraints and requirements. In institutional interviews, for instance, there is a clear-cut division of labour anchored to the turn-taking system and to the adjacency pair question/ answer regarding the right to ask questions allocated to the person represent-

Context, contexts and appropriateness

ing the institution and the obligation to answer questions allocated to the client (Fetzer 2000). Closely related to the conception of context as a dynamic construct is its relational conception which conceives it as a relational construct, relating communicative actions and their surroundings, relating communicative actions, relating individual participants and their individual surroundings, and relating the set of individual participants and their communicative actions to their surroundings. Third, context is seen as given as is reflected in the presuppositional approach to context which is also referred to as common ground or background information (Stalnaker 1999). Here, context is seen as a set of propositions which participants take for granted in interaction. This allows for two different conceptions of context: a static conception in which context is external to the utterance, and an interactive one, in which context is imported into the utterance while at the same time invoking and reconstructing context. While the former has been refuted in pragmatics, it still has a number of supporters in information science. To use Levinson’s (2003: 33) own words: the idea that utterances might carry along with them their own contexts like a snail carries its home along with it is indeed a peculiar idea if one subscribes to a definition of context that excludes message content, as for example in information theory. Context is then construed as the antecedent set of assumptions against which a message is construed. But it has long been noted in the study of pragmatics that this dichotomy between the message and context cannot be the right picture.

The context-dependence of context is thus reflected in its statuses as (1) given and external to the utterance, (2) re-constructed and negotiated in and through the process of communication, (3) indexical, and (4) never saturated. In the following, the parts-whole perspective on context is refined, and context as a whole is categorized into linguistic context, cognitive context, sociocultural context and social context.

1.1 Linguistic context Linguistic context comprises the actual language used within discourse. Language is composed of linguistic constructions (or parts) embedded in adjacent linguistic constructions composing a whole clause, sentence, utterance, turn or text. Thus, linguistic context, or co-text (de Beaugrande & Dressler 1981; Janney 2002) denotes a relational construct composed of local and global adjacency relations. In the stance adopted here, the nature of the connectedness between a linguistic construction (or a part) and other linguistic constructions constituting a text (or

5

6

Anita Fetzer

a whole) is looked upon analogously to Searle’s conceptions of regulative rules and constitutive rules (Searle 1969). That is to say, the rule-governed realization of linguistic constructions in context constitutes an utterance act thus counting as a move within the game of producing and interpreting utterance acts. At the same time, the utterance act counts as a move within the game of producing and interpreting speech acts. While the rule-governed realization of linguistic constructions is constrained by the rules of grammar, the production and interpretation of speech acts are constrained by felicity conditions, as is shown by Oishi’s contribution which connects illocutionary acts with context thus examining the ways in which utterances become appropriate or inappropriate, felicitous or infelicitous. The production and interpretation of an utterance act is anchored to language’s constitutive parts of syntax, morphology, phonology, semantics and pragmatics. While syntax is composed of structural units, for instance constituents in traditional grammar, phrases in functional grammar and generative grammar, groups in systemic functional grammar or constructions in construction grammar, it is the linear ordering of the individual parts within a sequence which constitutes their grammatical function. The adverb really, for instance, realizes the grammatical function of a sentence adverbial in the utterance Really, Sue is smart while it is assigned the grammatical function of the adverbial of a subjunct in Sue is really smart. Or, the proper noun Peter can realize the grammatical function of an object in Sue met Peter and it can realize the grammatical function of a subject in Peter met Sue. Thus, it is not the linguistic construction as such which is assigned a particular grammatical function. Rather, it is the positioning of a linguistic construction within a linguistic sequence which assigns the linguistic construction a grammatical function. In the framework of constitutive rules, the positioning of a linguistic construction within a sequence counts as a particular move which assigns the construction a particular grammatical function. The relational nature of linguistic context is also reflected in a sentence’s topological units of pre-field, middle-field and post-field, and their respective subfields, which are also conceived of in relational terms thus counting as further constitutive parts in the construction of a whole utterance act. For instance, a change in the canonical word order SVO in English in the utterance I met Peter to the non-canonical word order OSV Peter I met does not change the propositional meaning of the utterance. From a discursive viewpoint, however, the fronting of the object signifies a contrastive set. Here, the speaker intends the message that she met Peter while at the same time implicating that she did not meet other not-named, but presupposed members of the contrastive set. The investigation of syntax and syntactic structure from a context-anchored parts-whole perspective has demonstrated that the whole, viz. the whole utterance in context, is more than

Context, contexts and appropriateness

the sum of its parts, as the ordering of the individual parts constitutes additional discursive meaning. An investigation of morphology from a context-based perspective sheds further insights into the morphological processes of inflection, derivation and compounding. In that scenario, inflection, such as the inflectional morpheme s in English in the word form [[run][s]], is looked upon as an indicator for the connectedness between individual words in the context of a sentence by making explicit the status of the word as a pluralized noun in the utterance he did five runs or as a verbal form indicating simple present tense in the utterance Peter runs faster than Tom, for instance. Derivation and compounding are analysed as making explicit the connectedness between morphemes in the context of a lexical expression, such as the prepositional verb [[take] [over]] and the compound [[over][take]], or the derivational morpheme and free morpheme [ism] in [[minimal][ism] and in there’s too much ism nowadays. As has been shown for the domain of syntax, an analysis of morphology from a context-anchored partswhole perspective leads to more refined results in word formation with respect to inflection, derivation and compounding as the ordering of the individual parts as well as their position within the sequence constitutes their status and function. An explicit accommodation of context from a parts-whole perspective in phonology also leads to stimulating new insights. Here, assimilation is looked upon as the adaptation of a part to its phonological context (or the whole). The sequence ten pencils consists of two parts, namely [ten] and [penslz]. When realized as a whole, the alveolar nasal [n] is adapted to its local phonological context [p] and realized as another bilabial sound, viz. [m], in the phonological sequence [tempenslz]. The context-anchored parts-whole perspective is further manifest in the realization of a phonological form as a full form or as a reduced form. For instance, the preposition to in the sequence I cycled to school can be realized as a reduced form with a schwa-sound thus stressing the place where the speaker went to, namely ‘school’, or it can be realized with the monophthong [u] as a full form thus stressing the direction of the movement. A context-anchored perspective is also of immense importance in supra-segmental phonology and here in particular in the field of intonation with its unit of investigation, the intonational phrase. In systemic-functional grammar, intonation is seen as a signalling system (Halliday 1994), and in interactional sociolinguistics an intonational phrase is assigned the status of a contextualization device (Gumperz 1996). For instance, the intonational contour of a fall in English signifies the illocutionary force of request. The intonational contour of a rise, by contrast, signifies the illocutionary force of question or offer. The one-word utterance tea realized with the intonational contour of a fall contextualizes the utterance as a request in English. When realized with a rise, however, the utterance counts as

7



Anita Fetzer

an offer. Again, a context-anchored parts-whole perspective to the investigation of phonology leads to exciting insights by showing that a holistic approach to the phonological realization of utterances does not only contextualize the communicative intention of a speaker by showing what participants do with words but also that grammar needs to be looked upon from an interaction-based viewpoint (Couper-Kuhlen 1993; Pierrehumbert & Hirschberg 1992). Semantics has been traditionally defined as the investigation of meaning while pragmatics has been traditionally defined as the investigation of contextdependent meaning. From a parts-whole perspective, truth-conditional semantics examines the meaning of a whole proposition by identifying its constitutive parts, that is reference and predication. Once identified, the meaning of the whole proposition is true in any scenario and thus context-independent. In that frame of reference, the propositions Peter does not do anything to change his life and Peter does not do nothing to change his life do not share the same truth conditions. For this reason, they are true in different worlds and therefore not identical (or different). From a pragmatics-based outlook, however, they may share the same communicative status in communication. Possible-worlds-anchored semantics restricts the investigation of a proposition’s meaning from truth conditions which are valid in any context to that of one of its subsets, a possible-world scenario. Here, the meaning of a proposition is true in a pre-specified scenario only. Discourse semantics focusses on anaphora resolution as well as on cohesion and coherence, and lexical semantics examines the semantic meaning of lexical expressions, such as large. Again, the explicit accommodation of a context-anchored parts-whole perspective may lead to more refined results, as has been pointed out by Akman and Alpaslan (1999: 10) in their examination of the meaning of the adjective large. In the utterance Stephen built a large snowman the lexical meaning of the adjective large is interdependent on the seize of the discourse identity of Stephen. If Stephen is a toddler, ‘large’ denotes a seize of about 1.20 metres, and if Stephen is an adult, ‘large’ denotes a seize of about 1.80 metres. Speech acts are composed of propositional acts and illocutionary acts which are composed of further constitutive acts, such as reference acts and utterance acts (Searle 1969), or locutionary acts, illocutionary acts and perlocutionary acts which are composed of further constitutive acts, such as phatic acts or rhetic acts (Austin 1980). All of a speech act’s constitutive acts (except the perlocutionary act) are conventional in nature, and it is their performance in an appropriate context which makes them count as a particular speech act. For instance, the utterance I hereby request you to give me 100 Euros is composed of the illocutionary act of a request, which is signalled through the explicit performative ‘I hereby request’, and the propositional act is composed of the reference act ‘you’ referring to

Context, contexts and appropriateness

the hearer and the predication ‘give speaker 100 Euros’. Following Searle (1969), the performance of the constitutive acts counts as a request if the generalized felicity conditions for a request obtain, viz. normal in- and out-put conditions (speaker and hearer speak the same language), propositional content conditions (reference to a future act), preparatory conditions (speaker has the necessary social status to utter a request and make the hearer comply with the act, and hearer has the requested sum of money), sincerity condition (speaker means what he or she utters), essential condition (the utterance counts as a request). Adopting a context-anchored framework, Sbisà (2002) assigns felicity conditions the status of context categories. As has been shown for the domains of syntax, morphology and phonology, a context-anchored parts-whole perspective to the examination of language and language use achieves news insights into their rule-governed nature and into the nature of their connectedness. This also holds for the fields of semantics and pragmatics, where the whole, that is the whole proposition, discourse and speech act, is also more than the sum of its parts. The explicit accommodation of linguistic context to the investigation of syntax, morphology, phonology and semantics leads to more refined from results regarding the grammar-interaction interface. Not only does this allow for a systematic examination of linguistic parts, such as syntactic, morphological and phonological constructions or lexical expressions, and their connectedness with wholes, such as clauses, sentences, utterances, propositions and texts, but also to a holistic outlook on grammar, which may be supplemented by social, sociocultural and cognitive perspectives. In the following, cognitive context which is a necessary condition for a cognitive-based theory of language and language use is examined.

1.2 Cognitive context Cognitive context is not only of relevance to cognitive linguistics and cognitive pragmatics, it is also of immense importance to the field of psychology, and here in particular to the psychology of communication. Bateson (1972) conceives of context along the lines of the gestalt-psychological distinction between figure and ground and the related concepts of frame and framing. Frame is seen as a delimiting device which “is (or delimits) a class or set of messages (or meaningful actions)” (Bateson 1972: 187). Because of its delimiting function, “psychological frames are exclusive, i.e. by including certain messages (or meaningful actions) within a frame, certain other messages are excluded” and they are “inclusive, i.e. by excluding certain messages certain others are included” (ibid.). The apparent contradiction is eradicated by the introduction of set theory’s differentiation be-

9

10

Anita Fetzer

tween set and nonset, which – like the gestalt-psychological concepts of figure and ground – are not symmetrically related. To use Bateson’s own words: “Perception of the ground must be positively inhibited and perception of the figure (...) must be positively enhanced” (Bateson1972: 187). Against this background, he draws the conclusion that the concept of frame is metacommunicative. This also holds for context which, analogously to frame is structured and metacommunicative, or to use Bateson’s own words: “the hypothesis depends upon the idea that this structured context also occurs within a wider context – a metacontext if you will – and that this sequence of contexts is an open, and conceivably infinite, series” (Bateson 1972: 245). Bateson thus explicitly connects set and nonset, frame and metaframe, and context and metacontext with a parts-whole perspective: “whenever this contrast appears in the realm of communication, is simply a contrast in logical typing. The whole is always in a metarelationship with its parts. As in logic the proposition can never determine the meta proposition, so also in matters of control the smaller context can never determine the larger” (Bateson 1972: 267). In his work on frame analysis, Goffman (1986) uses the concept of frame as a metaphor for context, background and setting thus referring to the relational dimension of meaning: “I am not addressing the structure of social life but the structure of experience individuals have at any moment of their social lives” (Goffman 1986: 13). The relational conception of frame is reflected in Goffman’s differentiation between primary framework, key and fabrication. A primary framework provides a way of describing the event to which it is applied. Naturally, there are different types of primary framework. Key denotes “the set of conventions by which a given activity, one already meaningful in terms of some primary framework, is transformed into something patterned on this activity but seen by the participants to be something quite else” (Goffman 1986: 44). Framework is intrinsically connected with the activity of framing, key is intrinsically connected with the activities of keying, downkeying, that is opting for a more direct manner of expression, upkeying, that is opting for a more indirect manner of expression, and rekeying, that is opting for a different manner of expression. Fabrication is distinguished with respect to self-induced and other-induced fabrications, and is defined as “the intentional effort of one or more individuals to manage activity so that a party of one or more others will be induced to have a false belief about what it is that is going on” (Goffman 1986: 83). The concept of frame is fundamental to the construction of meaning: “In general, then, the assumptions that cut an activity off from the external surround also mark the ways in which this activity is inevitably bound to the surrounding world” (Goffman 1986: 249). While the connectedness between frame and framing, and between keying, upkeying, downkeying and rekeying needs to be based on metarepresen-

Context, contexts and appropriateness

tation, framing also needs to be recursive: “Frame, however, organizes more than meaning; it also organizes involvement” (Goffman 1986: 345). Cognitive context is not only key to the psychology of communication. It is also of immense importance for language processing and the corresponding inference processes involved. Relevance theory (Sperber & Wilson 1986) differentiates between cognitive environment and cognitive context: the former refers to a set of facts, while the latter refers to a set of premises, namely, true or possibly true mental representations. Constitutive elements of cognitive context are mental representations, propositions, contextual assumptions which may vary in strength, and factual assumptions. Assumptions are read, written and deleted. In the meantime, contextual implications are raised in strength, lowered in strength or erased from memory. Since cognitive contexts are anchored to an individual but are also required for a cognitively based outlook on communication, they must contain assumptions about mutual cognitive environments. Thus, cognitive context is not only defined by representations but also by meta-representations. To describe the multilayered and multifaceted phenomenon of cognitive context, relevance theory employs the onion metaphor, and cognitive context is represented by different layers of an onion. What is of importance for language processing and for inferencing is the premise that the order of inclusion corresponds to the order of accessibility. This ensures that both processes are ordered, and that their order is based on meta-layers and meta-contexts. Along these lines, Carota’s contribution examines the connectedness between a subset of cognitive context, that is common ground, and appropriateness. She shows that contrastive markers are not only context- and domain-dependent, but also of prime importance for efficient and effective processing. In functional grammar, context also denotes a psychological construct which Givón (2005) explicates in Context as Other Minds as follows: “First, we noted that context is not an objective entity but rather a mental construct, the construed relevant ground vis-à-vis which tokens of experience achieve relatively stable mental representation as salient figures. Whatever stability mental representations possess is due, in large measure, to the classification of tokens of experience into generic categories or types” (Givón 2005: 91). What is important to the investigation of cognitive context is the differentiation between types of experience and tokens of experience. While the former are of prime relevance to language processing and inferencing, the latter are intrinsically connected with practical reasoning and abduction, in and through which tokens are classified into appropriate types. Along these lines, Nyan’s contribution shows in what way processing efficiency is connected to both context and appropriateness giving particular attention to the relevance of argumentative markers to the processes and products of internal and external argumentation.

11

12

Anita Fetzer

A similar type-token differentiation is manifest in Penco’s distinction between subjective context and individual context (Penco 1999). The former refers to a cognitive or epistemic representation of the world and thus to an individual set of beliefs which may belong to an individual or a community, and the latter refers to an individual’s set of beliefs and thus to their representation of the world. The connectedness between individual contexts and subjective contexts is examined in Akman’s contribution, which demonstrates in what ways and for what purposes situations are categorized as similar or as different. A necessary consequence of Givón’s differentiation between type of experience and token of experience, and of Penco’s differentiation between subjective context and individual context is that cognitive context is both individual (or unique) and social (or typed). This ambivalence is also reflected in van Dijk’s concept of context models (van Dijk 2006), in which context is allocated to the interface between social structure and discourse. According to van Dijk, contexts are not objective or deterministic constraints of society or culture, but subjective participants interpretations, constructions or definitions of such aspects of the social environment. Cognitive context is a structured, multilayered construct which is indispensable for language processing and inferencing. The nature of the connectedness between its constitutive layers and subsystems is meta-communicative and meta-systemic. In the following, social context and its constitutive parts of physical setting and participants, who process language and perform inferencing processes, is examined in more detail.

1.3 Social context Social context is often considered to comprise the context of a communicative exchange and is defined by deducting linguistic context and cognitive context from a holistic conception of context. Constituents of social context are, for instance, participants, the immediate concrete, physical surroundings including time and location, and the macro contextual institutional and non-institutional domains. Frequently, language use in social contexts has been allocated to the domains of communicative performance or parole which have been assigned the status of an individual and momentary product. That has been denied by Recanati (1998) and by ethnographic studies. The non-individualistic use of language is also manifest in the contextual phenomenon of deixis and its realization as deictic expressions, that is temporal deixis, local deixis, participant deixis, discourse deixis and social deixis. Unlike anaphora resolution, which requires linguistic context, discourse deixis is informed by both linguistic and social contexts.

Context, contexts and appropriateness

The categories of speaker, hearer and audience are no longer conceived of as analytic primes. Rather, they denote interactional categories and have been refined by Goffman (1981) with respect to their footing, that is their ratified and unratified status in the participation framework and their interactional roles of animator, author and principal. This has been further refined by Levinson (1988) who explicitly accommodates the reception format. In an actual speech situation, however, the interactional roles do more than simply produce and interpret communicative contributions. In a social context, they subcategorize into social roles and their gendered and ethnic identities, to name but the most prominent ones. In institutional communication, the participants’ institutional roles embody institutional power which manifests itself in their corresponding rights and obligations. This is investigated in Becker’s contribution, which examines the connectedness between context and appropriateness in the framework of media communication regarding the communicative status of questions. Social context has been further differentiated by van Dijk (1981) with respect to general social context anchored to functional pragmatic coherence, and particular social context types anchored to contextual frames, assumed purposes and intentions. The context types contain information about particular institutions, frame structures, settings, functions, properties, relations, positions and frame conventions. To account for the micro-macro interface, van Dijk distinguishes between macro actions and micro actions. The micro-macro link is also at the heart of Fetzer’s and Miecznikowski and Bazzanella’s contributions. The former examines the communicative function of implicit and explicit references to appropriateness in political interviews, and the latter examine the communicative function of the conditional in spoken discourse. In the field of corpus linguistics, Biber (1988) adapts basic social-context constituents to his examination of variation in spoken language and in written language. According to him, the components of a speech situation do not only contain information about the present situation but also about previous encounters. This is reflected in his categories of participant roles and characteristics, which are further refined by the communicative roles of speaker, addressee and audience, their personal characteristics, the group characteristics, the relations among participants, the social role relations, the personal relations, the extent of shared knowledge, the setting, topic and purpose. Biber explicitly points out that purpose comprises both conventional and personal goals thus connecting the individual domain with the social domain. The research paradigm of ethnomethodology investigates the interactional organization of society. It represents a micro sociological perspective par excellence, in which the indexicality of social action is key. Ethnomethodology focuses on the domain of intersubjectivity and it examines the questions of how separate

13

14

Anita Fetzer

individuals are able to know or act within a common world, and how members (or participants) negotiate or achieve a common context: “in an interaction’s moment-to-moment development, the parties, singly and together, select and display in their conduct which of the indefinitely many aspects of context they are making relevant, or are invoking, for the immediate moment” (Schegloff 1987: 219). Here, common context is synonymous with social context, which, like linguistic context, classifies into a local (or micro) social context and a global (or macro) social context. To refine the dichotomous micro-macro interface, social contexts further classify into a number of intermediate layers, such as meso social contexts which can denote the delimiting frame of a particular speech event or the delimiting frame of a more global institutional context (Drew & Heritage 1992; Sarangi & Slembrouck 1996). The importance of social context to communication has been demonstrated, and its relevance is spelled out by Hanks as follows: “Hence it is not that people must share a grammar, but that they must share, to a degree, ways of orienting themselves in social context. This kind of sharing – partial, orientational and socially distributed – may be attributed to the habitus, or relatively stable schemes of perception to which actors are inculcated” (Hanks 1996: 235). But is it really social context, which is at the heart of communication? In the following, the connectedness between social contexts and culture is examined more closely.

1.4 Sociocultural context Social context is frequently used synonymously with extra-linguistic context which comprises the participants of a communicative exchange, their physical and psychological dispositions and the specific knowledge or assumptions about the persons involved, the knowledge of the language and the conventions regarding appropriate use of language, the knowledge of activity-types including communicative intentions and goals, and general background knowledge. Of course, the immediate extra-linguistic context is embedded in more remote extra-linguistic contexts, such as particular organizational contexts and other socio-historically constituted contexts of institutions and (sub)cultures. The synonymous use of extra-linguistic context and social context is, however, an oversimplification as research in sociolinguistics, anthropology and cultural studies has informed us. Rather, social context subcategorizes into different types of sociocultural context which are defined by a particular perspective on social context. Against this background, social context is conceived of as an unmarked type of context or as a default context, and sociocultural context is conceived of as a marked type of context in which particular variables, such as time, location or individual, are

Context, contexts and appropriateness

interpreted in a particular mode. For instance, the distinction between monochronic time and polychronic time is based on the differentiation between a linear, tangible and divisible conception of time, where events are scheduled one at a time and where this schedule takes precedence over interpersonal relationships. Polychronic time, by contrast, is characterized by things occurring simultaneously. Here, interpersonal relationships take precedence over a task-oriented outlook on communication (Hall & Hall 1989). Thus, culture provides the members of a speech community2 with a common configuration and interpretation of both extra-linguistic parameters, such as time, space, institution, individual and their multiple roles, and linguistic variables, such as conventionalized linguistic constructions and their meanings, levels of pragmatic directness or preferred sequential organization. Metaphorically speaking, culture is a filter which allows us to interpret social context in accordance with particular sociocultural-context constraints and requirements. A culture-dependent outlook on communication has been the promoted by the ethnography of communication (Saville-Troike 1989), in particular by Hymes’ speaking grid (Hymes 1974). Here, a communicative exchange is systematized with respect to its constitutive components of situation (the physical setting and the psychological scene), participants (speaker, hearer and audience, and their statuses in the participation framework), ends (the goal and the purpose of the speech event from a sociocultural viewpoint), act sequence (how something is said with regard to message form and what is said with regard to message content), key (mock or serious), instrumentalities3 (channels, i.e. spoken, written, e-mail, multi-modal), norms of interpretation and forms of speech (vernacular, dialect, standard), and genre. Hymes’s speaking grid and his notion of communicative competence has been refined by Gumperz (1977, 1992) who explicitly connects the cognitive operation of inference with the sociocultural activity of conversation. His conception of conversational inference represents a context-bound process of interpretation in which other’s intentions are assessed, and in which self illustrates her/ his understanding and comprehension through her/his response. Gumperz assigns language usage the status of actual language practice, and he interprets evaluation 2. Culture is used synonymously to Hymes’ conception of speech community. It may comprise national culture and national subcultures, ethnicity- and gender-related cultures and ethnicity- and gender-related subcultures, or work-related cultures and work-related subcultures, to name but the most prominent ones. 3. The connectedness between context, channel and appropriateness has been examined in the research domain of interpersonal communication, where appropriateness and effectiveness are anchored to channel (cf. Canary & Spitzberg 1987; Westmyer, DiCioccio & Rubin 1998).

15

16

Anita Fetzer

as a social activity. Saville-Troike summarizes his contribution to the field of ethnography of communication as follows: “Gumperz builds on this in proposing the outline of a theory of how social knowledge is stored in the mind, retrieved from memory, and integrated with grammatical knowledge in the act of conversing” (Saville-Troike 1989: 131). Because of its cultural base, the meaning that emerges in a conversation is different for participants if they are not members of the same speech community. Gumperz’s original contribution to the field of ethnography of communication bridges the gap between linguistic context and sociocultural and social contexts on the one hand, and between linguistic, social and sociocultural contexts and cognitive context on the other hand. This is due to the fact that language is seen as a socially situated cultural form and, therefore, as a specification of the more general social and linguistic contexts. Moreover, Gumperz’s interactional sociolinguistics accommodates a micro and macro outlook on communication which is not only reflected in the participant’s communicative performance but also in their conversational inferencing, or to use Gumperz’s own words (2003: 14): It is useful to distinguish between two levels of inference in analyses of interpretive processes: (a) global inferences of what the exchange is about and what mutual rights and obligations apply, what topics can be brought up, what is wanted by way of a reply, as well as what can be put into words and what is to be implied, and (b) local inferences concerning what is intended with my one move and what is required by way of a response.

In the following, the process of contextualization is examined, which provides us with the tool to bridge the gaps between internal and external contexts, and between micro and macro contexts.

1.5 Context and contextualization Contexts in context has examined linguistic, cognitive, social and sociocultural contexts and their constitutive parts, which have been subcategorized by the introduction of micro (or local) linguistic, cognitive, social and sociocultural contexts, and macro (or global) linguistic, cognitive, social and sociocultural contexts. Furthermore, context has been classified as generalized context, which is also called default context or unmarked context, and particularized context, which is also referred to as non-default context or marked context. Not only does the micromacro distinction hold for linguistic, cognitive, social and sociocultural contexts, but so does the generalized-particularized differentiation.

Context, contexts and appropriateness

In the previous section, the research paradigm of interactional sociolinguistics surfaced as a frame of reference which may not only account for different types of context, but also for the micro-macro interface examined above. Interactional sociolinguistics views language as a socially situated form and assigns inference the status of conversational inference which is a context-dependent cognitive operation par excellence. Its unit of investigation is the speech activity, which is closely connected with Levinson’s activity type (Levinson 1979), Linell’s communicative project (Linell 1998) and Luckmann’s communicative genre (Luckmann 1995). To account for the micro-macro interface, the explicit accommodation of context is a necessary condition, as Gumperz points out: “With respect to context, psychologists, cognitive scientists, and many linguists who pay attention to context tend to define it almost entirely in extra-communicative terms. I argue that, while these factors are, of course, significant, contextual information is imported into the interpretative process primarily via indexical contextualization cues, in the form of presuppositions of what the activity is and what is communicatively intended” (Gumperz 2003: 119). Contextualization cues are metalinguistic indexicals which “serve to retrieve the contextual presuppositions conversationalists rely on making sense of what they see and hear in interactive encounters. They (..) have no propositional content. That is, (..) they signal only relationally and cannot be assigned context-free lexical meanings” (Gumperz 2003: 9). Contextualization cues “channel inferential processes that make available for interpretation knowledge of social and physical worlds” (Gumperz 1996: 383). Gumperz draws the conclusion that “since all interpretation is always context-bound and rooted in collaborative exchanges that rest on shifting contextual presuppositions, contextualization must be a universal of human communication” (Gumperz 1996: 403). So what is the nature of the connectedness between context and appropriateness, or rather between contexts, contextualization and appropriateness? In the following the concept of appropriateness is examined with respect to the nature of its connectedness with context and contextualization.

2.

Appropriateness in context

Appropriateness has frequently been compared and contrasted with acceptability, well-formedness and grammaticality. A grammatical sentence needs to be in accordance with the rules of grammar, a well-formed sentence does not only have to be grammatical but also should be easy to process, and an acceptable sentence does not necessarily have to be grammatical but needs to be comprehensible and easy to process. While grammaticality and well-formedness are the result of a primarily

17

1

Anita Fetzer

language-oriented evaluation of a sentence, acceptability and appropriateness are not. Rather, they refer to the result of the evaluation of an utterance in its linguistic and social contexts. Adopting van Dijk’s differentiation between acceptability and appropriateness (van Dijk 1981), the former is assigned the status of an empirical concept and the latter is assigned the status of a theoretical construct: “The acceptability of an utterance is evaluated with regard to the nature of the connectedness between its linguistic form, sequential position and social context, and the appropriateness of an utterance is evaluated with regard to the nature of the connectedness between a communicative action, its linguistic realization and its embeddedness in linguistic and social contexts” (Fetzer 2004: 19–20). As a consequence of this, appropriateness is intrinsically connected to communicative intention. But what is the nature of the connectedness between the cognitive-context anchored notion of intention and the sociocultural concept of appropriateness? One of the fundamental premises of pragmatics, intentionality (of communicative action), is also a constitutive part of cognitive context. Following Sperber and Wilson, intention is a mental representation capable of being realized in the form of actions. It differentiates into an informative intention, namely, the intention to make manifest or more manifest to the audience a set of assumptions, and a second-order informative intention, a communicative intention, viz. the intention to have the informative intention recognized and to make it mutually manifest to the audience and the communicator that the communicator has this informative intention (Sperber & Wilson 1986: 54–64). Yet, intention is not only seen from a cognitive perspective. Intention is also considered to be a constitutive part of social action and action theory. To use Haferkamp’s own words: “social action is marked by intentionality and relation to culture: actors give their action a meaning, but this meaning is not self-created; conversely, actors operate within the framework of the general representations of their society’s culture” (Haferkamp 1987: 179). For this reason, intention is also a constitutive part of social context. In the previous section, social context has been defined as an unmarked type of context, which is particularized through a cultural perspective regarding its constitutive parts, such as time, location, individual and their roles or institution. In interactional sociolinguistics, language is conceived of as a sociocultural construct, and that is why communicative intention is assigned the status of a constitutive part of linguistic and sociocultural contexts. Thus, it is intentionality of communicative action which connects linguistic, social, sociocultural and cognitive contexts. Against this background, social actors select and construct appropriate contexts for their social actions. Like context, appropriateness is not an analytic prime and has been examined with respect to appropriateness conditions, which are analysed in the following.

Context, contexts and appropriateness

2.1 Appropriateness conditions In a social- and socioculturally oriented examination of speech act theory, felicity conditions have been assigned the status of appropriateness conditions. This is particularly true for a macro-oriented rethinking of speech act theory focussing on the well-formedness and appropriateness of longer stretches of discourse, which encounters the following difficulties, to use Ferrara’s own words: On the one hand, constraints derived from the appropriateness conditions for a certain type of speech act help to explain why certain features appear in linguistic expressions which are meant to convey that type of speech act (...). On the other hand, the study of language is brought closer to the study of society, in the sense that insights are made possible into how social constraints (as embedded in roles and institutions) translate into constraints on linguistic form” (Ferrara 1980: 322).

Regarding the appropriateness of micro speech acts looked upon from the perspective as constitutive parts of a macro speech act (van Dijk 1981), Ferrara (1980) comes to the conclusion that “ the appropriateness of speech acts embedded in sequences cannot be evaluated without reference to broader frames of action and goals than those already implicit in the act itself ” (Ferrara 1980: 323). Adopting a parts-whole perspective this means that the whole, that is the activity type, communicative genre or communicative project, is more than the sum of the individual speech acts, of which it is composed. For instance, a question-answer sequence does not constitute an interview. There need to be further constitutive elements, such as shared purpose, dovetailed answers, to name but a few. Along those lines, and supplemented by a discourse-analytic approach to communication, Berlin’s contribution examines the connectedness between truth, context and appropriateness. He demonstrates that the appropriateness of evasiveness is not only context-dependent, but also participant-dependent. In the research paradigm of conversation analysis, contextual constraints can be seen as being functionally equivalent to appropriateness conditions. Here, an encounter or a communicative exchange is typed through particular devices, such as ‘ring’ in telephone conversations or ‘self-identification’ as a form of institutional discourse. Generally, there are additional constraints on allowable topics, and constraints on the construction of topically relevant talk, such as degree of explicitness (Lerner 2004). Appropriateness conditions need to be examined in a frame of reference which goes beyond an individual speech act, communicative contribution or turn. For this reason, appropriateness has been assigned the status of a dialogical concept. Not only is appropriateness dialogical, but it is also a macro concept

19

20

Anita Fetzer

whose definition needs to be based a larger frame of reference, such as an activity type, communicative project, communicative genre, macro speech act or macro validity claim (Fetzer 2002). In the following, the micro-macro interface and its connectedness with appropriateness is examined in more detail.

2.2 Appropriateness and the micro-macro interface Appropriateness has been examined as the product of a process of evaluating a communicative contribution with respect to its connectedness with linguistic, sociocultural and social contexts. In this frame of reference, the appropriateness of communicative action is intrinsically connected with its inappropriateness. In his examination of irony, Attaro (2000) relates appropriateness to context: “Consider the following operational definition of appropriateness: an utterance u is contextually appropriate iff all presuppositions of u are identical or compatible with all the presuppositions of the context C in which u is uttered (cf. the notion of ‘common ground’: Clark 1996), except for any feature explicitly thematized and denied in u” (Attaro 2000: 818). Building on that definition, he looks at the connectedness between irony and appropriate and inappropriate contexts, and comes to the conclusion that “irony is essentially an inappropriate utterance which is nonetheless relevant to the context” (Attaro 2000: 823). Put differently, irony may be inappropriate from a generalized-context perspective, but it may nonetheless be appropriate at that particular stage in dialogue, where it is realized. In his seminal work on frame analysis, Goffman (1986) connects appropriateness with style thus going beyond the production and interpretation of linguistic acts. Again, appropriateness is seen as intrinsically connected with both sociocultural contexts and communicative genres. To use Goffman’s own words: “Indeed, all our so-called diffuse social roles can be seen partly as styles, namely the manner of doing things that is “appropriate” to a given age, sex, class and so forth. One can think of style as keying, an open transformation of something modelled after something else (or after a transformation of something else)” (Goffman 1986: 290). By connecting role behaviour with social intelligence and in particular with anticipatory interactive planning (AIP), Goody (1995: 14, 15) takes Goffman’s argument one step further. To use her own words: Thus a definition of appropriate role behaviour includes both the proper role behaviour of ego, and the proper reciprocal responses of the role partner. In this way it makes AIP more powerful, since acting in a clearly defined role makes the behaviour of both members of the dyad more predictable. Indeed, their be-

Context, contexts and appropriateness

haviour is predictable not only to the participants, but to others (in the same society). Roles also facilitate AIP through increasing predictability on another level by implying sanctions if one or other member of the dyad does not behave in the correct, predictable way. Thus roles include a control element that facilitates the AIP cognition of all members of the society. The classifying and labelling of role behaviour has an added significance at the level of the social system because sanctions are part of the mechanism for learning and maintaining ‘social structures’ and ‘social capital’. Roles lead out from individual behaviour to the reproduction of the social structure.

Not only is appropriateness connected with predictability and thus with social (or rather: sociocultural) context, but also with learning and maintaining social order, that is with the micro-macro interface. Put differently, appropriateness is not of prime importance to the examination of isolated communicative acts. Rather, it is key not only to the investigation of the connectedness between a participant, their fellow participants and a communicative act, but also to the dyad’s communicative exchange as a means of maintaining – or dis-maintaining – social order. To examine the micro-macro interface more thoroughly, the frame in which a micro communicative act is performed is looked upon with respect to the delimiting frame of a communicative genre, communicative project and activity type. According to Thibault, “genres are types. But they are types in a rather peculiar way. Genres do not specify the lexicogrammatical resources of word, phrase, clause, and so on. Instead, they specify the typical ways in which these are combined and deployed so as to enact the typical semiotic action formations of a given community” (Thibault 2003: 44). Genre is thus “a classificatory concept, a way of sorting out conventionalized discourse forms on the basis of form, function, content, or some other factor or sets of factors” (Bauman 1992: 138). By introducing genre as a bridging notion between micro and macro, the distinction between different levels of empirical reality is captured. The interrelated levels between interpersonal structural conditions and non-interpersonal macro structural conditions, that is between individualistic phenomena and collective phenomena needs to be examined by transcending an interactionally oriented microanalysis focussing on individualistic experience and thus on token, linguistically speaking, and moving towards a mediated-structure oriented analysis focussing on type. This is captured by the intermediate layers of communicative genre, communicative project or activity type, and by the sociological frame of game which all have the function of connecting individual action with collective goals (Alexander & Giesen 1987). Giesen is even more radical by stressing the indispensability of the macro domain to the investigation of micro actions:

21

22

Anita Fetzer

Macrostructures are conceived in analogy to a common language that represents the ‘categorial prerequisites’ or the ‘constitutive rules’ for ‘individual’ speech acts. They are not the intended or unintended results of social actions but the indispensable prerequisites of these actions. In consequence the relation between action and categorial structure, between language and speech act, is viewed not as an empirical relationship of causation but as an analytical relationship of constitution. (...) Categorizing the rules of language (that is, the constitutive rules) as macrostructures and the speech act as microevent is based on a special difference between rules and actions: rules are applicable not only to particular actions or interaction situations but in principle to an infinite number of possible situations, where the temporal, the spatial, or the social scope of an action is always limited (Giesen 1987: 343).

Against this background, the edited volume examines context and appropriateness from both micro and macro perspectives paying particular attention to bridging problems between micro and macro.

3.

The contributions

The contributions of this volume fall in three parts. The first part looks at bridging problems between context and appropriateness. The contributions subsumed under this header employ a top-down perspective anchored to the research paradigms of situation theory, speech act theory and cognitive pragmatics. The second part investigates bridging problems between communicative action and appropriateness focussing the connectedness between context and appropriateness with respect to social and rational explanations of individual communicative acts. The contributions included in this part use an integrated discourse-based frame of reference feeding on theories of communicative action, sociopragmatics and critical discourse analysis. The third part examines bridging problems between micro and macro focussing on the connectedness between micro phenomena, that is the conditional and its communicative functions in context as well as contrastive markers and their communicative functions in context, and the nature of their embeddedness in context. The contributions start from a bottom-up perspective which is explicitly connected with macro constraints, such as sequentiality, genre and domain. Part I, Bridging problems between context and appropriateness, comprises three contributions. Varol Akman’s Similar situations studies the notion of similarity with reference to situations of situation theory. It shows that the commonsense

Context, contexts and appropriateness

notion of two situations resembling each other appears to be valuable in daily life, but needs appropriate delimitation in the fields of context and pragmatics. Particular reference is given to the contextual constraints on situations and to their connectedness with appropriateness. Etsuko Oishi’s Austin’s speech act theory and context gives a speech-act-theoretic explanation of the concept of appropriateness. The performance of an illocutionary act is explained as the process whereby a linguistic form becomes a linguistic artefact, and appropriateness and felicity conditions concern how such a linguistic artefact is created thus clarifying a specific relationship between an illocutionary act and context. Thanh Nyan’s Appropriateness and background knowledge: An adaptive view examines the connectedness between language processing, processing efficiency, context and appropriateness. Adopting a co-evolutionary view of language and the brain, the paper shows that using language appropriately is essentially a matter of activating skills arising from Background knowledge. Particular reference is given to perceptual strategies. Part II, Bridging problems between communicative action and appropriateness, contains three contributions. Anita Fetzer’s ‘If I may say so’: Indexing appropriateness in dialogue examines those contexts in which a contribution is assigned the status of not such as is required. They are categorized with regard to the questions of where the inappropriateness occurs, what part of the contribution is assigned an inappropriate status, who refers to inappropriateness and how this is realized linguistically. The paper shows that appropriateness conditions are anchored to the micro-macro interface of communicative contribution, genre and sociocultural context. Annette Becker’s The appropriateness of questions investigates the sociocultural construct of appropriateness in the dialogical discourse type of media interview using methods from the fields of conversation analysis, pragmatics, sociopragmatics, critical discourse analysis and appraisal theory. The paper demonstrates the connectedness between context, genre and sub-genre, and between discourse identities, face and discourse topic. Lawrence N. Berlin’s Cooperative conflict and evasive language: The case of the 9–11 commission hearings examines the connectedness between appropriateness and context using critical discourse analysis supplemented by a layered analysis of context. It shows how evasiveness is co-constructed in that particular genre thus broadening the notion of cooperation to include complicity in an expanded definition of what can be considered ‘appropriate’.

23

24

Anita Fetzer

Part III, Bridging problems between micro and macro, comprises two contributions. Johanna Miecznikowski and Carla Bazzanella’s The attenuating conditional: Context, appropriateness and interaction is based on the premise that linguistic devices are multidimensional. The paper investigates the conditional in spoken Italian and French discourse paying particular attention to modality, interaction, topic management and politeness. The micro perspective is refined by the explicit accommodation of the macro phenomena of sequentiality, socio-cultural setting and relations among coparticipants. Francesca Carota’s Collaborative use of contrastive markers – co-textual and contextual implications examines the Italian contrastive markers ma (but), invece (instead), mentre (while) and però (nevertheless) by giving particular attention to their correlation with discourse structure and context. The paper shows that their distribution is domain-specific and that they fulfill important functions in topic management, and in the administration of cognitive context and common ground.

References Akman, V. and Alpaslan, F.N. 1999. “Strawson on intended meaning and context”. In Modeling and Using Context. 2nd International and Interdisciplinary Conference, Context’99, Trento, Italy, September 1999, Proceedings, P. Bouquet, P. Brezillon and L. Serafini (eds.), 1–14. Heidelberg: Springer. Akman, V., Bouquet, P., Thomason, R. and Young, R.A.(eds.). 2001. Modeling and Using Context. Third International and Interdisciplinary Conference, CONTEXT 2001, Dundee, UK, July 2001, Proceedings. Springer: Heidelberg. Alexander, J.C., Giesen, B., Münch, R. and Smelser, N.J. (eds.). 1987. The Micro-Macro Link. Berkley: The University of California Press. Alexander, J.C. and Giesen, B. 1987. “Introduction.” In The Micro-Macro Link, J.C. Alexander, B. Giesen, R. Münch and N.J. Smelser (eds.), 1-42. Berkley: The University of California Press. Attaro, S. 2000. “Irony as relevant inappropriateness.” Journal of Pragmatics 32: 793–826. Austin, J.L.1980. How to Do Things with Words. Cambridge: Cambridge University Press. Bateson, G. 1972. Steps to an ecology of mind. New York: Chandler Publishing Company. Bauman, R. 1992. “Contextualization, tradition, and the dialogue of genres: Icelandic legends of the kraftskáld.” In Rethinking Context: Language as an Interactive Phenomenon, A. Duranti and C. Goodwin (eds.), 125–146. Cambridge: Cambridge University Press. Biber, D. 1988. Variation across Speech and Writing. Cambridge: Cambridge University Press. Blackburn, P., Ghidini, C., Turner, R.M. and Giunchiglia, F. (eds.). 2003. Modeling and Using Context. 4th International and Interdisciplinary Conference,Context 2003, Stanford, CA, USA, June 23–25, 2003. Heidelberg: Springer.

Context, contexts and appropriateness

Bouquet, P., Brezillon, P., Benerecetti, M., Castellani, F. and Serafini, L. (eds.). 1999. Modeling and Using Context. Second International and Interdisciplinary Conference on Modeling and Using Context, Context’99, Trento, Italy, September 1999, Proceedings. Heidelberg: Springer. Canary, D. and Spitzberg, B. 1987. “Appropriateness and effectiveness perceptions of conflict strategies.” Human Communication Research 14(1): 93–118. Couper-Kuhlen, E. 1993. English Speech Rhythm: Form and Function in Everyday Verbal Interaction. Amsterdam: Benjamins. De Beaugrande, R. and Dressler, W. 1981. Einführung in die Textlinguistik. Tübingen: Niemeyer. Drew, P. and Heritage, J. (eds.). 1992. Talk at Work: Interaction in Institutional Settings. Cambridge: Cambridge University Press. Ferrara, A. 1980. “Appropriateness conditions for entire sequences of speech act.” Journal of Pragmatics 4: 321–340. Fetzer, A. 1999. “Non-acceptances: Re- or un-creating context”. In Modeling and Using Context. 2nd International and Interdisciplinary Conference, Context’99, Trento, Italy, September 1999, Proceedings, P. Bouquet, P. Brezillon and L. Serafini (eds.), 133–144. Heidelberg: Springer. Fetzer, A. 2000. “Negotiating validity claims in political interviews”. Text 20(4): 1–46. Fetzer, A. 2002. “Communicative intentions in context.” In Rethinking Sequentiality: Linguistics meets Conversational Interaction, A. Fetzer and C. Meierkord (eds.), 37–69. Amsterdam: Benjamins. Fetzer, A. 2004. Recontextualizing context: Grammaticality meets Appropriateness. Amsterdam: Benjamins. Garfinkel, H. 1994. Studies in Ethnomethodology. Cambridge: Polity Press. Giesen, B. 1987. “Beyond reductionism: four models relating micro and macro levels”. In The Micro-Macro Link, J.C. Alexander, B. Giesen, R. Münch and N.J. Smelser (eds.), 337–355. Berkley: The University of California Press. Givón, T. 2005. Context as Other Minds. Amsterdam: Benjamins. Goffman, E. 1981. Forms of Talk. Oxford: Basil Blackwell. Goffman, E. 1986. Frame Analysis. Boston: Northeastern University Press. Goody, E. (ed.). 1995. Social Intelligence and Interaction: Expressions and Implications of the Social Bias in Human Intelligence. Cambridge: Cambridge University Press. Goody, E. 1995. “Introduction: Some implications of a social origin of intelligence.” In Social Intelligence and Interaction: Expressions and Implications of the Social Bias in Human Intelligence, E. Goody (ed.), 1–36. Cambridge: Cambridge University Press. Goodwin, C. and Duranti, A. (eds.). 1992. “Rethinking context: An introduction.” In Rethinking Context. Language as an Interactive Phenomenon, A. Duranti and C. Goodwin (eds.), 1–42. Cambridge: Cambridge University Press. Gumperz, J.J. 1977. “Sociocultural knowledge in conversational inference.” In Linguistics and Anthropology, M. Saville-Troike (ed.), 191–211. Washington: Georgetown University Press. Gumperz, J.J. 1992. “Contextualization and understanding.” In Rethinking Context: Language as an Interactive Phenomenon, A. Duranti and C. Goodwin (eds.), 229–252. Cambridge: Cambridge University Press.

25

26

Anita Fetzer

Gumperz, J.J. 1996. “The linguistic and cultural relativity of inference.” In Rethinking Linguistic Relativity, J.J. Gumperz and S.C. Levinson (eds.), 374–406. Cambridge: Cambridge University Press. Gumperz, J.J. 2003. “Response essay.” In Language and Interaction. Discussions with John J. Gumperz, S. Eerdmans, C. Prevignano and P. Thibault (eds.), 105–126. Amsterdam: Benjamins. Haferkamp, H. 1987. “Complexity and behaviour structure, planned associations and creation of structure.” The Micro-Macro Link, J.C. Alexander, B. Giesen, R. Münch and N.J. Smelser (eds.), 177–192. Berkley: The University of California Press. Hall, E.T. and Hall, M.R. 1989. Understanding Cultural Differences: Germans, French and Americans. Yarmouth/Maine: Intercultural Press. Halliday, M.A.K. 1994. Introduction to Functional Grammar. London: Arnold. Hanks, W.F. 1996. “Language form and communicative practices.” In Rethinking Linguistic Relativity, J.J. Gumperz and S.C. Levinson (eds.), 232–270. Cambridge: Cambridge University Press. Heritage, J. 1984. Garfinkel and Ethnomethodology. Cambridge: Polity Press. Hymes, D. 1974. Foundations in Sociolinguistics: An Ethnographic Approach. Philadelphia: University of Philadelphia Press. Janney, R.W. 2002. “Cotext as context: Vague answers in court.” Language and Communication 22(4): 457–475. Lerner, G. (ed.). 2004. Conversation Analysis. Studies from the First Generation. Amsterdam: Benjamins. Levinson, S.C. 1979. “Activity types and language.” Linguistics 17: 365–399. Levinson, S.C. 1988. “Putting linguistics on a proper footing: Explorations in Goffman’s concepts of participation.” In Erving Goffman. Exploring the Interaction Order, P. Drew and A. Wootton (eds.), 161–227. Cambridge: Cambridge University Press. Levinson, S.C. 2003. “Contextualizing ‘contextualization cues’.” In Language and Interaction. Discussions with John J. Gumperz, S. Eerdmans, C. Prevignano and P. Thibault (eds.), 31– 40. Amsterdam: Benjamins. Linell, P. 1998. Approaching Dialogue. Amsterdam: Benjamins. Luckmann, T. 1995. “Interaction planning and intersubjective adjustment of perspectives by communicative genres.” In Social Intelligence and Interaction: Expressions and Implications of the Social Bias in Human Intelligence, Goody, E. (ed.), 175–188. Cambridge: Cambridge University Press. Penco, C. 1999. “Objective and cognitive context”. In 2nd International and Interdisciplinary Conference on Modeling and Using Context (Context’99), P. Bouquet, P. Brezillon and L. Serafini (eds.), 270–283. Heidelberg: Springer. Pierrehumbert, J. and Hirschberg, J. 1992. “The meaning of intonational contours in the interpretation of discourse”. In Intentions in Communication, P.R. Cohen, J. Morgan and M.E. Pollack (eds.), 271–311. Cambridge: MIT Press. Recanati, F. 1998. “Meaning and force: Introduction.” In Pragmatics: Critical Concepts, A. Kasher (ed.), 126–154. London: Routledge. Sarangi, S. and Slembrouck, S. 1996. Language, Bureaucracy & Social Control. London: Longman. Saville-Troike, M. 1989. The Ethnography of Communication. Blackwell, Oxford. Sbisà, M. 2002. “Speech acts in context”. Language and Communication 22(4): 421–436.

Context, contexts and appropriateness

Schegloff, E.A. 1987. “Between macro and micro: Contexts and other connections.” In The Micro-Macro Link, J. Alexander, B. Giesen, R. Munch and N. Smelser (eds.), 207–234. Los Angeles: University of California Press. Schegloff, E.A. 1992. “In another context”. In Rethinking Context. Language as an Interactive Phenomenon, A. Duranti and C. Goodwin (eds.), 191–228. Cambridge: Cambridge University Press. Searle, J.R. 1969. Speech Acts. Cambridge: Cambridge University Press. Sperber, D. and Wilson, D. 1986. Relevance: Communication and Cognition. Oxford: Blackwell. Stalnaker, R. 1999. Context and Content. Oxford: Oxford University Press. Thibault, P. 2003. “Contextualization and social meaning-making practices.” In Language and Interaction. Discussions with John J. Gumperz, S. Eerdmans, C. Prevignano and P. Thibault (eds.), 41–62. Amsterdam: Benjamins. Van Dijk, T. 1981. Studies in the Pragmatics of Discourse. The Hague: Mouton. Van Dijk, T. 2006. “Discourse, context and cognition.” Discourse Studies 8(1): 159–177. Westmyer, S., DiCiocci, R. and Rubin, R. 1998. “Appropriateness and effectiveness of communication channels in competent interpersonal communication”. Journal of Communication 48(3): 27–48.

27

part i

Bridging problems between context and appropriateness

Similar situations Varol Akman Bilkent University, Turkey

This paper studies the notion of similarity with reference to situations of situation theory. While the commonsense notion of two situations resembling each other appears to be valuable in our daily life, we show that it is problematic for the same reasons researchers have been pointing out in psychological and philosophical literature. That human beings can use the notion naturally (without much effort) shows that their cognitive make-up is probably much more powerful than is commonly thought.

1.

Introduction There is nothing more basic to thought and language then our sense of similarity; our sorting of things into kinds. (W. V. Quine) How do we ever understand anything? Almost always, I think, by using one or another kind of analogy – that is, by representing each new thing as though it resembles something we already know. (M. Minsky)

“I’ve been in a similar state before,” says a ruined entrepreneur to the manager of a bank and proceeds to explain his plan of survival. “Remember, we have faced a similar aggression last month.” Thus a lieutenant cautions his soldiers before he outlines his strategy of defense. “Maybe I must start to prepare for my next round of exams in order to avoid a similar reprimand,” confesses a lazy undergraduate on academic probation. Other examples easily come to mind. In all of these, the role played by the commonsense notion of ‘similar situation’ is clear but a formal account seems to be not easy and even murky.1 An obvious question is this: When

1. There is an obvious problem lurking here. Maybe one is more interested in difference, not in similarity. In other words, let one’s attitude towards his current state be like “This is a very different state.” While there are contexts in which saying something like this would be natural, a simple trick would let us reduce such utterances to their equivalents formulated in terms of

32

Varol Akman

is it appropriate to think that two situations (faced by an intelligent agent) are similar? If I am meeting with the Prime Minister tonight, maybe I am entitled to thinking that this will be comparable to my earlier meeting with the mayor of my town, but would it be all right to say that I would be in similar situations?2 If yes, in what sense? If not, why not? Our goal in this paper is to study such questions of appropriateness3 using the basic ontology of situation theory (Barwise & Perry 1983; Devlin 1991). A caveat is needed at this point. Similarity has always been a thorny topic (Gleitman et al. 1996). To use a stale turn of phrase, a lot of ink has been spilt to figure it out in its entirety. While we indisputably know more about it now compared to say, some decades ago, we still have a long way to go to achieve a fully worked out account. To put the matter less obliquely, this paper does not pretend to demystify similarity once and for all to brush it aside. It only adds to the number of growing literature in ascertaining that similarity is a complicated notion. It is quite remarkable that we human beings have the cognitive make-up to use it with ease and assurance, and more often than not, with profit (Rosch 1978). Thus, looking at Figures 1 and 2, a grown-up person would be able tell with confidence that these are both meeting situations and hence can be regarded as similar (e.g., both have participants, most probably a chair administrating the meeting, notepads, tables, etc.). It is a goal of this paper to propose a (mildly original) explication of this feat.

similarity, to wit: “Among all the states I have been to, there is not one similar to the current one.” On a related note, Minsky (1986: 238) writes that ordinary thought relies on recognition of differences. Consider two situations S (source) and G (goal). You are in S and you want to end up at G. Denote the difference between S and G by Δ and suppose there are a number of actions that can be applied at S. A straightforward approach to achieve G might be to apply actions that remove or reduce Δ. If a certain action causes Δ to become bigger, you might want to look for another action that counteracts the former. 2. According to Tversky and Gati (1978), in saying “α is like β,” we are inclined to think of α as the ‘subject’ and β as the ‘referent.’ People tend to select the more prominent stimulus (also known as the ‘prototype’) as the referent and the less prominent stimulus (also known as the ‘variant’) as the subject. Thus, during the early years of the Reagan era, one would be entitled to say, “Poland is like the Soviet Union” (but usually, not vice versa). In other words, people attend more to the subject than the referent. 3. Our use of appropriateness here is in line with Fetzer (2004: 85): “Appropriateness is a social- and communicative-action-based construct which is calculated with regard to the connectedness between the force of the communicative action, its propositional content, its linguistic representation and their embeddedness in the immediate linguistic, sociocultural and social contexts, and their embeddedness in the remote linguistic, sociocultural and social contexts. Appropriateness is a constitutive part of practical reasoning and manifests itself in felicitous communication in which the coparticipants’ information wants and face wants are satisfied.”

Similar situations

Figure 1. A meeting situation (from http://www.phoenixexecutivegroup.com/aboutus. html)

Figure 2. Another meeting situation (from http://www.ofm.org/3/econ/ECOesc01.html)

2.

Psychological and cognitive science work

Important early work on similarity and analogy can be found in psychological and cognitive science literature. We will presently touch upon such work to offer a glimpse of their essence. There are also numerous proposals in artificial intelligence (AI) which study analogy from a computational perspective. We will omit such work in its entirety but refer the reader to Gentner and Markham (1995) and the slim but significant volume that includes it. Two concise appraisals of analogy, Gentner (1998) and Gentner (1999), are also valuable. Gick and Holyoak (1983) report the results of an experimental study concerning what is known as ‘analogical transfer.’ Subjects are presented with an assign-

33

34

Varol Akman

ment in which they must suggest a way to heal a deadly tumor without using a strong beam of radiation. (A strong beam would kill the surrounding tissue.) The superlative solution is to converge on the tumor with a number of weak beams of radiation. About 10% of the subjects propose this technique. If a prior story about some soldiers converging on a fortress is told, 30% of the subjects produce the previously mentioned solution. Finally, if the subjects are overtly told to bear in mind the soldiers’ story while thinking about the tumor problem, the rate of success reaches almost 90%. Gick and Holyoak conclude that while one might have prior experience stored in his memory, it may still be challenging to retrieve this in order to employ it in a novel (but similar) situation. One early, formal view of similarity uses a geometric approach (Shepard 1974). This boils down to mapping each object to a point in a space so that the metric distances between points correspond to the similarity of the respective objects. Challenging this approach, Tverksy (1977) came up with the simple proposal that each object α is characterized by a set of features Α. The observed similarity of two objects, α and β, is then expressed as a function of their common and distinctive features. Thus, the observed similarity is a function of three arguments (Tversky & Gati 1978): – The features shared by α and β, i.e. Α∩Β (set intersection) – The features of α that are not shared by β, i.e. Α ∕ Β (set difference) – The features of β that are not shared by α, i.e. Β ∕ Α In what is commonly called a ‘contrast model,’ the similarity of α to β is given as a linear combination of the measures of their joint and idiosyncratic features, viz. c1 f(Α∩Β) − c2 f(Α ∕ Β) – c3 f(Β ∕ Α), where c1, c2, c3 ≥ 0. In this way, the contrast model expresses similarity between two objects as the weighted difference of the above (three) feature sets (c1, c2, c3 being the weights). Here, f is a function reflecting the salience of the various features; it measures the contribution of any particular feature – be it common or distinguishing – to the similarity between objects.4 In Tverksy (1977), an interesting experiment is reported. Subjects are asked which country, Sweden or Hungary, most resembles Austria. No relevant dimension of similarity is specified; this means that the answer will depend on the (background) set of countries under consideration. The answers are as follows. If the backdrop set includes Poland, then the subjects tend to say Sweden. If the set includes Norway, they are inclined to pick Hungary. The rationalization for this

4. See Kittay (1982) for a particularly clear account of Tversky’s theory of similarity and its implications re simile and metaphor.

Similar situations

is evident: Poland and Hungary have – or had, at the time of the experiment – salient geopolitical features in common. Likewise, Sweden and Norway have such features. Tverksy concludes that judgments of similarity appeal to features having a high classificatory significance. Furthermore, he notes that while the similarity features depend on the relevant contrast set, the set itself depends on the interests of the participants (thus he is making a pragmatic point bearing upon appropriateness). Sloman (1999: 567) cites a 1983 study by Tversky and Daniel Kahneman in which subjects are first told the following story: Linda is 31 years old, single, outspoken and very bright. She majored in philosophy. As a student, she was deeply concerned with issues of discrimination and social justice and also participated in antinuclear demonstrations.

The subjects are then asked to pick the more likely statement: (i) Linda is a bank teller or (ii) She is a bank teller and is active in the feminist movement. The response turns out to be overwhelmingly (ii). This is an illustration of the so-called conjunction fallacy. Statement (ii) cannot possibly be more likely that statement (i), for it takes (i) for granted. Tversky and Kahneman attribute the subjects’ disappointing performance to the ‘representativeness’ heuristic, i.e., the likelihood of an event increases to the degree that it is similar to the category under consideration.

3.

Situations

We begin with a brief overview of situation theory (or rather, situations). The theory starts with the levelheaded assumption that what is called ‘Reality’ is one big (seemingly unbounded or at least inconceivably large) situation. Limited parts of Reality are called ‘situations’ and can be individuated by cognitive agents (say, human beings). Thus, we perceive situations, cause them to be brought about, and have all sorts of attitudes toward them. One fact remains: we are always in situations. An example might help at this point. I have recently been in a flight situation where the neighboring passenger suddenly choked on his food. By calling the attention of the flight attendant to this (and she immediately asking for help from other passengers), I had my share in saving him from a dreadful end. I felt that this was one hard-hitting situation; I really detested being in it.5 5. Notice that what I have just said could have been fiction. In other words, I could have shamelessly made this whole story up. However, and this is interesting, you would have no trouble in following it and believing it. Fictional situations are inspirational but in this paper I assume that we are always talking about real situations.

35

36

Varol Akman

Situations usually have individuals standing in relation at various (spatiotemporal6) locations. The individuals have properties. Situation theory posits that real situations are metaphysically and epistemologically prior to individuals, relations, and locations. Human beings and lower organisms display a fundamental ability to notice the similarities between situations. This is done by noting the regularities – individuals, properties, relations, or locations that endure from one situation to another. Thus, I believe that snow is slippery, that police are usually available for help, that mothers care about their children, that I will receive presents on my birthday. Barwise and Perry (1983: 10) put the matter elegantly: The situations we perceive and participate in are always limited, a small part of all that has gone on, is going on, and will go on.7 Situations overlap in time and space in complicated ways, but each situation is unique unto itself, no one quite the same as any other. And in this uniqueness resides a puzzle. For if living things are to survive, they must constantly adapt to the course of events in which they find themselves, to ever changing, fleeting, and unique situations.

As Barwise and Perry also note, this adaptation takes place as a consequence of attunement to similarities between situations. Let us agree to call such similarities ‘uniformities.’ A useful uniformity in my life has to do with the newspaper boy. Every morning (a different situation), he brings the paper at about 8 o’clock and leaves it in our doorsteps. By just being attuned to this uniformity, I contribute to my well-being. Unfortunately, there is no newspaper service on Sundays; I have to go out and buy the paper myself. Unless I am aware of this, chances are that I will be disappointed from time to time: I will open the door, will not find the paper, start wondering what happened, and then suddenly realize that this is a Sunday. 6. This clumsy adjective may be omitted in the sequel, with the understanding that a location is a space-time location. 7. Clearly not all situations are current. My dentist appointment last week was one mean situation, which, thankfully, is now over. The upcoming session scheduled to next week promises to be even more troublesome. Notice that in this example I am referring to past and future situations. Someone could have videotaped the whole visit to the dentist and that would be a good representation of the past situation. No one can really predict what the future visit will be like. (For instance, it may not materialize at all if I suddenly die this week.) It is clear that future situations are vastly different from fictional situations. I can provide you with a detailed and believable account of how my visit next week will be like. While this account will necessarily be defeasible (i.e., “rationally compelling but not deductively valid,” cf. The Stanford Encyclopedia of Philosophy http://plato.stanford.edu/entries/reasoning-defeasible), it may nonetheless be quite useful for assorted purposes.

Similar situations

Nastier examples also come to mind. Suppose one notices a small, hard bump in his arm one day. He gets curious and starts observing it on a regular basis, say every week. He notices that the bump is growing in size and it is starting to hurt too. Now he is really concerned and will in all probability pay his doctor a visit soon. Once again, being attuned to this uniformity does contribute to the welfare of this person. To quote Barwise and Perry (1983: 10): [I]t is by categorizing situations in terms of some of the uniformities that are present, and by being attuned to appropriate relations that obtain between different types of situations, that the organism manages to cope with the new situations that continually arise.

Devlin (1991) has done much to clarify what exactly situations amount to and how we ‘individuate’ them. This last term means that we can single out and treat situations as identifiable entities that can later be talked about. Thus, Devlin (1991: 31): But in what sense does an agent individuate a situation (when it does)? Not, in general, as an individual. Rather, the agent individuates a situation as a situation, that is to say, as a structured part of Reality that it (the agent) somehow manages to pick out. There are a number of ways an agent can ‘pick out’ (that is, individuate) a situation. Two obvious examples are direct perception8 of a situation, perhaps the immediate environment, or thinking about a particular situation, say last night’s dinner party.

. Direct perception of a situation is not as unproblematic as it sounds, due to ever-present, high-level interpretation. The following excerpt from Foster (2000: 212–213) explains the problem: Think of the situation of someone watching a live football match on television. In some ways, this situation is like that of [a] radar operator. Both subjects are looking at a screen. Both have visual experiences which are caused, in the normal neurophysiological way, by light which comes from this screen. […] But there is also a crucial difference. In the case of the radar operator, this information becomes available purely by inference, from his more basic information about what the screen itself displays. […] The situation with the television viewer is quite different. It is true that the viewer may – almost certainly will – make certain inferences from the information which he visually acquires. But, even in its most basic form […] this information is not about the two-dimensional pattern of colours on the screen, but about the three-dimensional colour-arrangement of the football scene itself. This is because the visual experience includes not just the visual registering of the screen-pattern […] but also the interpretation of that pattern in three-dimensional terms, making it experientially seem to the viewer as if he is watching the match from a location in the football stadium – though, of course, he will know that this is not so.

37

3

Varol Akman

It is probably redundant to remind that when an agent individuates a situation, he cannot be hold to that he gives a precise description of everything that that situation comprises. Take this case of two friends (Devlin 1991: 31–32): […] John and David are having a conversation about a particular football game, say one they have both seen. Then they are both referring to a very definite situation – namely that particular game. This is a situation that they both individuate (as a situation, not as an individual). A long, informative, and confusion-free discussion can take place. And yet neither John nor David would be able to list every single event that formed a part of that game, or every item of information that related to it in some essential way. Indeed, it is highly unlikely that what makes their conversation about the game of interest to both parties is the fact that each one acquires from the other new items of information about the game, information that the one picked up but the other did not. The fact that each person left the game with some information that the other did not does not mean that their subsequent conversation is about two different games. The situation is the same for both individuals, a highly structured part of the world, having a fixed duration in time. What differs is what each knows about that situation.

Let us call what John and David know about the game situation their ‘epistemological positions.’ It can (and does) happen that maybe they are talking about two different games, say g1 and g2. Obviously, g1 should be similar to g2 in some respects. This can happen in countless ways, e.g., they were played in the same town, they were both important games, the scorers in both games had the same names, both games were interrupted by violent demonstrations, and so forth. In this case, after a while the parties (John and David) would notice – perhaps gradually – that something is amiss, that their epistemological positions no longer overlap in significant ways. A brief investigation by one or more of the parties (“What do you mean by the second penalty? There was only a single penalty.” or “The French referee? I thought he was Italian.”) would then resolve the problem and set things straight. In this case, the similarity of situations g1 and g2 is the source of confusion. Devlin (1991: 32) says: [I]f you were to interrupt John and David in the middle of their conversation and ask them what they were talking about, they would reply “Last night’s football game.” Are we then to conclude that they were in fact talking about nothing; or that neither was really sure what it was they were discussing? Clearly not.

While we are in general unable to trim down situations to a complex of more familiar objects, we are nonetheless able to enumerate the ingredients that make up a situation. A situation is a rich (intensional) object consisting of individuals enjoying various properties and standing in a variety of relations. It is, in a sense, a small world.

Similar situations

One of the attractive features of situation theory is its insistence on an information-based account of communication (an account first championed by Fred Dretske). To this end, infons are posited as discrete items of information. An infon supplies a single piece of information. It is denoted as the (n + 2)-tuple , where R is an n-place relation,9 a1, …, an are objects appropriate for the respective argument places of R, and p is the polarity (0 or 1). If p = 1 (respectively, 0) then a1, …, an stand (respectively, do not stand) in the relation R. Take the infon . This is a fact (corresponds to the way things are in Reality). On the other hand, the infon is not a fact. (If we swap the argument values in this infon, it will start denoting a fact though.) Abstract situations are proposed to be the counterparts of situations in order to make the latter more amenable to mathematical manipulation. An abstract situation is a set (Devlin 1991). Given a situation s, the set {ι | s |= ι}, where ι is an infon, is the corresponding abstract situation. Notice that this sets collects all facts (infons that are made true by the situation). Here, s is said to ‘support’ an infon ι – denoted as s |= ι above – just in case the infon ι is true of the situation s. The opening propositions of the Tractatus can then be handily (if not in a caricatured manner) summarized as Reality = {ι | Reality |= ι}, i.e. “Die Welt ist die Gesamtheit der Tatsachen.” A ‘scheme of individuation’ – a way of carving the world into uniformities – is an essential aspect of situation theory. The notions individual, relation, and location depend upon this. In other words, the basic constituents of the theory are determined by the agent’s schema of individuation. Formal representation of these uniformities yields ‘types.’ Situation theory provides a collection of basic types for individuating or discriminating uniformities of the real world. These are some indispensable types: situation, individual, relation, temporal location, spatial location. (This is not an exhaustive list.) Parameters are generalizations over classes of non-parametric objects (e.g., individuals, spatial locations). Parameters can be associated with objects which, if they were to replace the parameters, would yield one of the objects in the class that parametric object abstracts over. Hence, allowing parameters in infons re9. Unary (1-place) relations are commonly called properties. This is probably the right place to warn the reader about the difference between situation-theoretic and mathematical relations. The latter are set-theoretic constructs whereas the former are relations of the kind recognizable by human beings. Thus, in the former account, ‘being tall’ is a property and ‘being the father of ’ is a relation. No doubt, we can (and do) use mathematical relations to model these. To this end, ‘father of ’ is rendered by the mathematical notation xFy, denoting that ‘x is the father of y.’ The relation F is a set of ordered pairs, including, among many others, the pair (see presently).

39

40

Varol Akman

sults in parametric infons. For example, <sees, ĝ, Alice, 1> and <sees, ĝ, ĥ, 1> are parametric infons where ĝ and ĥ stand for individuals. These infons are parametric on the first, and the first and second argument roles of the relation ‘see,’ respectively.10 Anchoring (binding) parameters of an infon to objects yields parameterfree infons. For example, given <sees, ĝ, Alice, 1>, if F(ĝ) = Bob (F is an anchoring function) then we obtain the parameter-free infon <sees, Bob, Alice, 1>. Suppose Alice was eating ice cream yesterday. She is eating ice cream now. Both of these situations share the same ‘constituent sequence’ (basically, an infon lacking a polarity) <eats, Alice, ice cream>. These two events, occurring at different times, have the same ‘situation type.’ Situation types can be more general. For example, a situation type in which someone is eating something at home contains the situation in which Alice is eating ice cream at home. Figures 3 and 4 show two ‘feeding’ situations. Each of these are about actual situations, one depicting a woman feeding penguins, and the other depicting four persons feeding dolphins. It is true that the number of ‘feeders’ and ‘feeded’ change. (Even the food will be different. Maybe it consists of small fish but the kinds may be different.) Still, both situations are of type ‘feeding situations.’ The constituent sequence characterizes this type conveniently. In situation theory, a network of abstract links between situation types (which are uniformities) provides information flow. Thus, the statement “Smoke means fire” expresses the law-like relation that links situations where there is smoke to situations where there is a blaze. If a is the type of smoky situations and b is the type of fire situations, then having been attuned to the constraint a▶b, an agent can pick up the information that there is a fire in a particular situation by observing that there is smoke.11 Anchoring plays a major role in the working of constraints. If the above constraint holds then it is a fact that if a is realized (i.e., there is a real situation a0 of type a), then so is b (i.e., there is a real situation b0 of type b). In order to invoke the constraint, we have to use an anchoring function which binds the location parameters to appropriate objects present in the ‘grounding situation,’ i.e., we have to first find a place and time at which there is smoke. 10. Their meaning can be rendered in English as “Someone sees Alice” and “Someone sees someone,” respectively. In order to keep things simple, we do not worry about tense in this paper but see Barwise and Perry (1983: 288ff.) and Devlin (1991: 228ff.). 11. a▶b is shorthand for the factual, parameter free infon , where ‘involves’ denotes the linkage between a and b. Devlin (1991: 91) notes that “many living creatures are aware of, or attuned to, this particular constraint, and make use of it in order to survive, though only humans have the linguistic ability to describe it with an expression such as SMOKE MEANS FIRE […].”

Similar situations

Figure 3. Feeding penguins (from http://gonewengland.about.com/library/blneaq14. htm)

Figure 4. Feeding dolphins (from http://www.pirateplanet.com/Brookfield_Zoo.html)

4.

Are situations vague objects? (Digression)

Any approach which claims to regard the situation-theoretic approach fruitful for the study of a given problem (in our case, similarity) runs a certain risk. This has to do with the ontological status of situations. While lay people do not seem to detect deep-seated problems with the situations – as a matter of fact, they view them as highly intuitive and commonsensical – philosophers, having a natural tendency to make finer distinctions, unearth points to debate. One such point concerns the spatio-temporal boundaries of situations.12 Before the emergence of situation theory, ‘ontic vagueness’ (could the world itself be vague or is it only 12. Suppose it is ambiguous whether some individual in a given situation has a given property. (Alternatively, suppose it is indeterminate whether two or more individuals in a given situation have a given relation.) These possibilities could make situations problematic entities but will not be considered here further.

41

42

Varol Akman

linguistic expressions that can be vague?) drew the attention of Evans (1978). He provided a terse slingshot argument to give a negative answer to the question “Can there be vague objects?”13 Still, a number of philosophers think that there are vague objects (Parsons & Woodruff 1996). Take Ben Nevis, the highest mountain in the United Kingdom. Keefe (2000: 15) says the following about it: [A]ny sharp spatio-temporal boundaries drawn around the mountain would be arbitrarily placed, and would not reflect a natural boundary. So it may seem that Ben Nevis has fuzzy boundaries, and so, given the common view that a vague object is an object with fuzzy, spatio-temporal boundaries, that it is a vague object.14

It is not difficult to multiply the examples. Thus, the singular term ‘Toronto’ seems to pick out a unique object (Toronto), despite the fact that that object has fuzzy spatiotemporal boundaries, cf. Keefe (2000: 159–160). Now consider the sentence “Toronto has an odd number of trees.” Keefe observes that the truth of this would depend on assorted ways of circumscribing the extent of Toronto. The sentence would sometimes be regarded as making a true claim – because the orders of Toronto have been drawn in a certain way – and sometimes a false claim – because the borders have been drawn in another way. Thus, the interpretation of ‘Toronto’ is indeterminate. Keefe also notes that the sentence “Toronto is in Canada” is true (simpliciter) because regardless of how one goes about delineating the boundaries of the city it would turn out to be true. Keefe’s approach is easily applicable to situations. Consider, once again, the football game situation. “There are an odd number of spectators watching the game” would most probably be regarded as having an indeterminate truth-value

13. Evans characterizes a vague object as one about which it is a fact that the object has fuzzy boundaries. 14. Compare this with the following view of Wiggins (2001: 166): It may be said that one candidate [for a vague object] is some mountain the ordinary individuation of which leaves over numerous questions of the form ‘Is this foothill a part of x? Is that foothill a part of x?’ But this is a strikingly poor illustration of what would be needed. For it can be perfectly determinate which mountain x is without x’s extent being determinate. A mountain is not, after all, something essentially demarcated by its extent or boundary. It is not as if there were just as many mountains to be found with x’s peak as there were rival determinations of x’s boundary. An idea like that could not even occur to one with the good fortune to be innocent of classical extensional mereology.

Similar situations

because it is not clear how one should delineate the boundaries of the game.15 On the other hand, the truth “The game was played in Istanbul” is unquestionably independent of how these boundaries are drawn. Our take on this matter is as follows. As long as the similarity of two situations does not concern problematic features (e.g., in the aforementioned example, oddness or evenness of number of spectators), we do not have to pay attention to vagueness. It remains to be seen, on the other hand, whether there is a proof (or disproof) of the proposition that situations are vague objects.

5.

Russell and Goodman on similarity

Bertrand Russell was especially interested in the ‘similarity of structure.’ He used the term ‘structure’ in several senses but in general meant something along these lines: a blueprint of relations holding among parts of a complex.16 Briefly, to reveal the structure of a complex is to spell out its parts and the ways in which these are interconnected. According to Russell, two classes are similar with respect to structure if and only if the following conditions are satisfied (McLendon 1955: 83): – A relation P relates members of the first class to one another. – A relation Q relates members of the second class to one another. – Each member of the first class corresponds to one and only one member of the second class, and vice versa. This relation of correspondence requires a one-to-one relation S which holds between the members of the first class and the members of the second class and at the same time preserves P in the first class and Q in the second class. (S is the ‘correlator’ of the two classes.) – Whenever P relates two items in the first class, the corresponding images of these in the second class are related by Q, and vice versa. The proverbial (if not boring) structural similarity consideration that we are all familiar with regards an atom analogous to a miniature solar system. Both the 15. When I was a kid, I used to watch all the games of my hometown team from a two-storey apartment building next to the stadium. The owner of the building was a distant relative and she would let me sit in one of the upper balconies. The building was literally meters away from the low walls of the stadium. Was I a part of the game situation? I think so. 16. It is fitting to observe that one of the most influential research programs in computational analogy – the so-called structure-mapping engine – is based on similarity of structure, cf. Markman and Gentner (2000) for recent work and Falkenhainer et al. (1989) for the classical paper on the subject.

43

44

Varol Akman

atom and the solar system consist of one inner object, surrounded by a number of smaller objects. The orbiting objects are attracted to the middle object by a force both in the atomic and in the solar case. If one is able to see in his mind’s eye the solar system, then this depiction may act as a model for the corresponding facets of the atom (Taber 2001). A legendary critic of similarity, Goodman (1972) noted that objects could be similar in numerous ways.17 According to him, we must specify in what respect two objects are similar; otherwise, we would be making a vacuous statement. To talk about similarity, one needs a frame of reference – just like those used by physicists studying motion. If Carol says that John is similar to David, you would have no idea until she quips that they are both football fanatics or that they are pathetic liars or that they are both bald, etc. To quote Goodman (1972: 445), “we must search for the appropriate replacement in each case; and ‘is similar to’ functions as little more than a blank to be filled.” By ‘appropriate,’ we think that Goodman presumably refers to the following trick.18 Two things are similar when certain predicates hold of them. Now, of two far and wide dissimilar actual animals (say a horse and a bee) can one say that they are similar (because after all they are both ‘real’)? Yes, but this is usually absurd; the predicate ‘real’ has little communicative significance in the context of a talk about real animals. The claim would be sensible though if it were made in the context of a chat about real and unreal (fantastic) creatures, say unicorns and elves.19 A comparable argument shows that when Carol uttered, “John is similar to David,” she made an ill-defined claim. Only when she specifies the particular ‘respects’ in which these two men are similar, does she start to make sense. Overall, Goodman’s critique can be seen as paving the way to a conception of similarity as a ternary predicate, i.e., it is only meaningful to state that A is similar to B with respect to r. That similarity seems to disappear when it is analyzed closely is best rendered in this memorable definition of Goodman (1972: 439): “[T]o say that two things are similar in having a specified property in common is to say nothing more than they have the property in common.” Medin et al. (1993: 272) draw attention to a weak spot in Goodman’s thesis: 17. Our review of Goodman is based on an account by Medin et al. (1993) and also Medin and Goldstone (1995). Unfortunately, we have not yet been able to locate the original paper (Goodman 1972). 1. Our explanation was inspired by a footnote in Recanati (2004: 150) where he refers to a passage from Tverksy (1977). 19. To wit, “A bee is similar to a horse” (respectively, “An elf is similar to a unicorn”) because they are both real (respectively, imaginary) but “A unicorn is not similar to a horse” (because one is imaginary and the other is real).

Similar situations

[C]onsider what one might know about quaggas (some hypothetical or unfamiliar entity) from the statement “Quaggas are similar to zebras.” Although we have no data bearing on this question, out intuitions are that people might be at least modestly confident that quaggas are hooved animals and not especially certain about whether or not they are striped. Hooved is (by conjecture) part of an interrelated set of properties associated with zebras, whereas striped seems more to be an isolated property.20 In any event, Goodman’s framework would find the comparison completely uninformative until the respects were specifically mentioned.

Nonetheless, Medin and Goldstone (1995: 106) concur with the general thesis and extend Goodman’s original proposal in the following way: “A is similar to B” is, according to them, shorthand for “A is similar to B in respects r according to comparison process c, relative to some standard s mapped onto judgments by some function f for some purpose p.”

6.

Lewis on similarity

In his landmark work on counterfactuals, Lewis (1973) studied counterfactual conditionals of the following form: If it were the case that A (antecedent), then it would be the case that C (consequent).

According to Lewis, this roughly means: In certain possible worlds where A holds, C also holds.

Lewis asks which A-worlds – worlds in which A is true – should be considered as candidates; his answer is that not all A-worlds will do. For one thing, those that differ greatly from our actual world should be ignored. To put it more positively, we need to consider the A-worlds most similar – overall – to our world. That it is not meaningful to consider a world where A holds but everything else is just as it really is seems to require an argument. Lewis (1986a: 5) puts the matter stylishly: Differences never come singly, but in infinite multitudes. Take, if you can, a world that differs from ours only in that Caesar did not cross the Rubicon. Are his predicaments and ambitions there just as they actually are? The regularities of his character? The psychological laws exemplified by his decision? The orders of the

20. Incidentally, our intuitions in this example re ‘striped’ differ from the authors’. This does not make their general point any less valid though.

45

46

Varol Akman

day in his camp? The preparation of the boats? The sound of splashing oars? Hold everything else fixed after making one change, and you will not have a possible world at all.

This informal but convincing line of reasoning shows that we cannot have an A-world that is otherwise just like our world. Therefore, we must search for an A-world that does not differ greatly from ours. Ideally, such a world would be in disagreement with our world only as much as it is required to while making A true. Moreover, such a world will be closer to our world in similarity than other A-worlds. Based on this notion of ‘being closer,’ Lewis (1986a) offers a preliminary analysis of counterfactual conditionals as follows.21 Let A ☐→ C denote the counterfactual conditional with antecedent A and consequent C. Then A ☐→ C is true at a possible world i if and only if C holds at the closest (accessible22) A-world to i, if there is one.

While this analysis is elegant and useful, it has a serious imperfection for people like us who want to know more about what ‘close’ means. Lewis (1986a: 6) agrees: It may be objected that [the above analysis] is founded on comparative similarity – “closeness” – of worlds, and that comparative similarity is hopelessly imprecise unless some definite respect of comparison has been specified. […] Imprecise though comparative similarity may be, we do judge the comparative similarity of complicated things like cities or people or philosophies23 – and we do it often without benefit of any definite respect of comparison stated in advance. We balance off various similarities and dissimilarities according to the importance we attach to various respects of comparison and according to the degrees of similarity in the various respects.

21. By preliminary, we imply that Lewis offers further (more fine-grained and accurate) analyses in the remainder of that paper (and several other publications). However, these are not crucial for our present purposes. 22. The idea of one possible world’s being accessible to another is not a straightforward one. The reader is referred to Hughes and Cresswell (1989: 77ff.) from which the following excerpt is taken: “[A] world, w2, is accessible to a world, w1, if w2 is conceivable by someone living in w1 […].” In the same reference, the authors exemplify this by stating that a world without telephones would be accessible to us – because we can easily conceive of it – but our world would not be accessible to it. 23. The following excerpt is again from Lewis (1986b: 42): “To what extent are the philosophical writings of Wittgenstein similar, overall, to those of Heidegger? I don’t know. But here is one respect of comparison that does not enter into it at all, not even with negligible weight: the ratio of vowels to consonants.”

Similar situations

Lewis’s general awareness of the fact that similarity judgments are affected by the respects that enter into them is apparent in the following vivid passage (1986c: 54): [C]onsider three locomotives: 2818, 4018, and 6018. 2818 and 4018 are alike in this way: they have duplicate boilers, smokeboxes, and fireboxes (to the extent that two of a kind from an early 20th century production line ever are duplicates), and various lesser fittings also are duplicated. But 2818 is a slow, small-wheeled, two-cylindered 2-8-0 coal hauler – plenty of pull, little speed – whereas 4018 is the opposite, a fast, large-wheeled, four-cylindered 4-6-0 express passenger locomotive. So is 6018; but 6018, unlike 2818, has few if any parts that duplicate the corresponding parts of 4018. (6018 is a scaled-up and modernized version of 4018.) Anyone can see the way in which 6018 is more similar to 4018 then 2818 is. But I would insist that there is another way of comparing similarity, equally deserving of that name, on which the duplicate standard parts make 2818 the stronger candidate.24

Let L2818, L4018, and L6018 denote the locomotives Lewis is talking about. The picture emerging from his description is as follows: – –

L2818 ≡ L4018, meaning they have many duplicate parts in common, larger or smaller (the former is out-of-date and the latter is up to date). L6018 ≈ L4018, meaning the former is a scaled-up version of the latter (they are both up to date cars).

Lewis states that L6018 ≈ L4018 (the approximation sign standing for similarity), for the simple reason that they look quite the same (structural similarity). However, he thinks that L2818 ≈ L4018 is also a sensible claim, for they have after all many parts in common. It is not easy to settle on which similarity judgment is the ‘right’ one, for it all depends on the context. In Lewis’s related work, ‘counterparts’ of persons – inhabitants of other worlds who bear a resemblance to him closely (or more closely, compared to other inhabitants of the same world – play a key role too. Not surprisingly, it is not possible 24. Similar examples are given in the analogical reasoning literature. Thus, the following excerpt from (Weitzenfeld 1984: 138) is relevant: Consider, for example, two cars known to come from the same assembly line at approximately the same time. They are known to be identical in structure and composition and it is known that one of them is black. Does this justify the inference that the similar car is also black? Suppose there is a third car, sharing the common properties of the other two, but known to be yellow. Does it justify the inference that the unknown car is yellow? Clearly there must be some further link between the premises of an argument by analogy and its conclusion.

47

4

Varol Akman

for a similarity theorist to find solace in this line of work either, for it is also based on an unanalyzed conception of similarity. Thus, Lewis (1983a: 28) states that the counterpart relation is a relation of resemblance and that […] it is problematic in the way all relations of similarity are: it is the resultant of similarities and dissimilarities in a multitude of respects, weighted by the importances of the various respects and by the degrees of the similarities.

Continuing on the same theme, he emphasizes the key role of ‘respects,’ a notion that carries for him essentially the same meaning that it enjoys in the cognitive science literature (Medin et al. 1993). Lewis’s following example is particularly instructive (1983b: 51–52): [C]ounterpart relations are a matter of over-all resemblance in a variety of respects. If we vary the relative importances of different respects of similarity and dissimilarity, we will get different counterpart relations. Two respects of similarity and dissimilarity among enduring things are, first, personhood and personal traits, and second, bodyhood and bodily traits. If we assign great weight to the former, we get the personal counterpart relation. Only a person, or something very like a person, can resemble a person in respect of personhood and personal traits enough to be his personal counterpart. But if we assign great weight to the latter, we get a bodily counterpart relation. Only a body, or something very like a body, can resemble a body in respect to bodyhood and bodily traits enough to be its bodily counterpart.

One is undoubtedly entitled to asking whether Lewis ever comes close to proposing a scheme to deal with similarity (to bite the bullet, so to speak). The answer is in the affirmative. His so-called ‘spheres’ are aimed at exactly this issue. Let a ‘sphere’ (around a possible world i) be a set of worlds W such that for every w in W, w is accessible from i and is closer to i than any world w´ not in W. A sphere is called A-permitting if it contains some A-world (remember that this is a world where A holds). With spheres at one’s disposal, the following rendering of counterfactual conditionals becomes possible (Lewis 1986a: 12): A ☐→ C is true at possible world i if and only if A → C (where the second arrow denotes the material implication of classical logic) holds throughout some Apermitting sphere around i, if there is such a sphere.

Lewis’s spheres are no doubt well-designed as a mathematical apparatus but they offer next to nothing for a tangible analysis of similarity.25 To be fair, it must be

25. The following three comments, all found in (Lewis 1986d: 163), make Lewis’s overall goals vis-à-vis similarity rather clear: (i) “To begin, I take as primitive a relation of comparative over-

Similar situations

noted that the spheres around a world are nested and thus they can be said to preserve or model comparative closeness in some sense. But this is taking us to the technicalities of Lewis’s theory and with all due respect for the latter’s brilliance this is not the place to do so.

7.

Situations as ‘icebergs’

In this final section, I will furnish an unfinished, highly tentative sketch for understanding similarity judgments (and their appropriateness). This sketch owes considerably to an approach first detailed in (Recanati 2004).26 Let us go back to the very beginning and ask the following question again: What is it for someone to judge that two situations are similar? In other words, what is it for someone to claim that two situations are similar (while honoring appropriateness conditions)?27 The main insight that is offered by situation theory is this. When we describe an empirical situation (say, Devlin’s football game situation), we make certain features explicit, but indefinitely many other features remain implicit. These implicit features in fact constitute a sort of hidden, amorphous background. The parallel here – and this is also inspired by Recanati (2004) – is to an iceberg28 (Figure 5). Only in toy worlds (e.g., chess) we can hope to describe situations completely without omitting anything. In other circumstances, we cannot hope to describe them in their full complexity (detail). Accordingly, we will normally omit, in deall similarity among possible worlds.” (ii) “I have not said just how to balance the respects of comparison against each other, so I have not said just what our relation of comparative similarity is to be. Not for nothing did I call it primitive.” (iii) “But the vagueness of over-all similarity will not be entirely resolved. Nor should it be.” 26. Recanati himself gives some credit to Friedrich Waismann, Hilary Putnam, and John Searle. In the present account, I will be content just citing from Recanati (2004), for his way of putting things is very instructive. 27. Recanati studies another pragmatic question, viz. what it is for someone to learn a predicate P. The idea is to observe the application of P in a particular situation s; this would let one associate P and s. When encountered with a new situation s´, we can apply P in s´, provided s´ is sufficiently similar to s. There is a danger in doing so though. The new situation can resemble the old one in a way that is not relevant for the application of P. There is only one way to avoid this danger and correct it: to enlist the help of the language community. They would usually say something along the lines “No, you cannot apply P in this situation (s´).” 2. For, according to the Wikipedia (http://en.wikipedia.org/wiki/Iceberg), “[…] around 90% of the volume of an iceberg is under water, and that portion’s shape can be difficult to surmise from looking at what is visible above the surface.”

49

50

Varol Akman

Figure 5. An iceberg (from http://www.earthfuture.com/stormyweather/gallery)

scribing a situation, those features which will not make a difference (in our view). Hence, infinitely many features of a given situation are kept implicit. Consider a situation s. Take two infons ι1 and ι2 supported by s, viz. s |= ι1 and s |= ι2. Assume that the infon ι1 has to do with a particular feature that we are interested in (to be brought to the tip of the iceberg) and that the infon ι2 has to do with another feature that we would like to suppress (to be pushed to the invisible part of the iceberg). We can then talk about ι1 and forget about ι2. The aforementioned exercise can be repeated with different features in an assortment of ways. Abusing the iceberg metaphor, we can think of the iceberg as if it is made of some elastic, malleable material (say, play dough). Depending on what collection of features we want to bring to fore (respectively, push to background), we promote them to the tip of the iceberg (respectively, demote them to the invisible part). Equivalently, those features that are at the tip of the iceberg gain prominence and those that are in the invisible part lose prominence (for the purposes of a particular similarity comparison).29

29. Similarity judgments concerning two situations have all the signs of having an ‘open texture’ (a phrase coined by Waismann) like quality. Open texture means, in our context, that two situations judged similar may turn out to be dissimilar upon further scrutiny, and vice versa (two disparate situations can suddenly become similar in some respect). To formulate a similarity comparison using a novel situation, the novel situation should resemble the ‘source situations’ – situations which are similar. Yet, since we cannot hope to examine in advance all the possible dimensions of similarity between the source situations and possible target situations, we will invariably have open texture emerging.

Similar situations

Example. I intend to make a similarity comparison with the following ingredients: my dad and President Roosevelt. (The former will be the subject and the latter the referent.) Let the context of the conversation in which I find myself embedded as a coparticipant be ‘health’ (say, severe illnesses). If I utter “My dad is like President Roosevelt,” my goal may be to draw attention to the fact that dad is also incapacitated with polio. If this happened to dad when he was 39, the similarity is more powerful. If dad further was a onetime president of Turkey, then it is even more commanding. If, on top of all these, dad fought to recover the use of his legs, predominantly through swimming, then the similarity becomes astounding. With each such additional matching feature, I might be said to be making a more appropriate similarity judgment.30 (Caveat: This last remark assumes that the coparticipants are in some way knowledgeable about the aspects of dad’s life that I have just mentioned. It is also assumed that the aforementioned facts about FDR are part of this ‘common ground’ – the knowledge, beliefs, and suppositions of the participants. If these assumptions do not hold, then it is my obligation to bring up the necessary elements in accord with the Goodmanesque recipe.) Now, assume another context, that of ‘education’ (say, universities attended). When I make the same utterance as above, I may want to draw attention to the fact that dad also went to Harvard and Columbia Law School. A more impressive similarity case would arise if dad, like Roosevelt, passed the bar exam and fulfilled the requirements for a degree but did not care to actually graduate. We conclude this paper by drawing attention to the inferential characteristics of similarity. From the viewpoint of situation theory, similarity may be due to 1. The individuals salient in the source situation, 2. The relations salient in the source situation, and 3. The locations salient in the source situation. Let us exemplify, by way of corresponding examples, what these are all about. Example 1. Suppose you have observed a certain individual on television in several occasions talking from a lectern in front of cameras, media representatives, etc. 30. Take a difficult question such as what makes a metaphor work (a.k.a. aptness of a metaphor). What I have in mind when I say ‘appropriate similarity judgment’ is along the same lines, i.e., what gives strength to a similarity judgment. In the context of my example, if many predicates are true of dad and FDR, I think that the strength of my judgment would increase proportionally. (We can measure the strength, at least in the context of this particular scenario, by the amount of nods of approval by the participants.) Note, however, that there is a singularity in this process. When dad and FDR are indistinguishable in terms of the predicates that hold of them, they have become identical. Therefore, the most perfect similarity judgment is not a similarity judgment anymore; it is an identity judgment!

51

52

Varol Akman

You come to recognize these source situations as say, the White House spokesperson’s press conferences. Now, when you catch a glimpse of the same person on some television channel the next time, you conclude that you are about to witness a press conference. Example 2. You see someone standing before a lectern with the caption ‘White House’ on it. Although you have never seen this person before, you infer that he is the White House spokesperson. Here the salient relationship is that of ‘being next to a lectern.’ (We ignore the complications posed by the somewhat bizarre scenario that this person is the President and that you do not recognize this fact.) Example 3. Whenever they show a particular landmark site (say the Trafalgar Square) on television on Fridays, you notice that people are holding demonstrations there, sometimes necessitating police involvement. Therefore, when you see today (a Friday) that there is a Breaking News broadcast where the anchor says, “We now go to our correspondent John Doe who is at the Trafalgar Square,” you anticipate seeing some demonstration and perhaps, scuffles with the police.

Acknowledgments I am indebted to the editor of this volume and Carla Bazzanella (University of Turin) for their friendly support and sympathetic understanding. The opening quotes appear in Quine31 (1969: 116) and Minsky (1986: 57), respectively.

References Barwise, J. and Perry, J. 1983. Situations and Attitudes. A Bradford Book. Cambridge, MA: MIT Press. Devlin, K. 1991. Logic and Information. New York: Cambridge University Press. Evans, G. 1978. “Can there be vague objects?” Analysis 38: 208. Falkenhainer, B., Forbus, K.D. and Gentner, D. 1989. “The structure-mapping engine: Algorithm and examples.” Artificial Intelligence 41: 1–63. Fetzer, A. 2004. Recontextualizing Context: Grammaticality Meets Appropriateness [Pragmatics & Beyond New Series 121]. Amsterdam/Philadelphia: John Benjamins. Foster, J. 2000. The Nature of Perception. New York: Oxford University Press.

31. It is probably apt to note that Quine’s paper discusses concept development and is concerned with children developing ‘similarity spaces’ in diverse domains.

Similar situations

Gentner, D. 1998. “Analogy.” In A Companion to Cognitive Science, W. Bechtel and G. Graham (eds.), 107–113. Oxford, UK: Blackwell. Gentner, D. 1999. “Analogy.” In The MIT Encyclopedia of the Cognitive Sciences, R.A. Wilson and F.C. Keil (eds.), 17–20. Cambridge, MA: MIT Press. Gentner, D. and Markham, A.B. 1995. “Similarity is like analogy: Structural alignment in comparison.” In Similarity in Language, Thought and Perception [Semiotic and Cognitive Studies 1], C. Cacciari (ed.), 111–147. Turnhout, Belgium: Brepols. Gick, M.L. and Holyoak, K.C. 1983. “Schema induction and analogical transfer.” Cognitive Psychology 15 (1): 1–38. Gleitman, L.R., Gleitman, H., Miller, C. and Ostrin. R. 1996. “Similar, and similar concepts.” Cognition 58 (3): 321–376. Goodman, N. 1972. “Seven strictures on similarity.” In his Problems and Projects, 437–447. Indianapolis, IN: Bobbs-Merrill. Hughes, G.E. and Cresswell, M.J. 1989. An Introduction to Modal Logic. London: Routledge. Keefe, R. 2000. Theories of Vagueness [Cambridge Studies in Philosophy]. Cambridge, UK: Cambridge University Press. Kittay, E.F. 1982. “The creation of similarity: A discussion of metaphor in light of Tversky’s theory of similarity.” In Proceedings of the Biennial Meeting of the Philosophy of Science Association, Volume 1, 394–405. Chicago: The University of Chicago Press. Lewis, D. 1973. Counterfactuals. Cambridge, MA: Harvard University Press. Lewis, D. 1983a. “Counterpart theory and quantified modal logic.” In his Philosophical Papers, Volume I, 26–46. New York: Oxford University Press. Lewis, D. 1983b. “Counterparts of persons and their bodies.” In his Philosophical Papers, Volume I, 47–54. New York: Oxford University Press. Lewis, D. 1986a. “Counterfactuals and comparative possibility.” In his Philosophical Papers, Volume II, 3–31. New York: Oxford University Press. Lewis, D. 1986b. “Counterfactual dependence and time’s arrow.” In his Philosophical Papers, Volume II, 32–52. New York: Oxford University Press. Lewis, D. 1986c. “Postscripts to ‘Counterfactual dependence and time’s arrow’.” In his Philosophical Papers, Volume II, 52–66. New York: Oxford University Press. Lewis, D. 1986d. “Causation.” In his Philosophical Papers, Volume II, 159–172. New York: Oxford University Press. Markman, A. B. and Gentner, D. 2000. “Structure-mapping in the comparison process.” American Journal of Psychology 113 (4): 501–538. McLendon, H.J. 1955. “Uses of similarity of structure in contemporary philosophy.” Mind 64 (253): 79–95. Medin, D.L. and Goldstone, R.L. 1995. “The predicates of similarity.” In Cacciari (1995), 83– 110. Medin, D.L., Goldstone, R.L. and Gentner, D. 1993. “Respects for similarity.” Psychological Review 100 (2): 254–278. Minsky, M. 1986. The Society of Mind. New York: Simon & Schuster. Parsons, T. and Woodruff, P. 1996. “Worldly indeterminacy of identity.” In Vagueness: A Reader, A Bradford Book, R. Keefe and P. Smith (eds.), 321–337. Cambridge, MA: MIT Press. Quine, W.V. 1969. “Natural kinds.” In his Ontological Relativity and Other Essays, 114–138. New York: Columbia University Press. Recanati, F. 2004. Literal Meaning. Cambridge, UK: Cambridge University Press.

53

54

Varol Akman

Rosch, E. 1978. “Principles of categorization.” In Cognition and Categorization. E. Rosch and B.B. Lloyd (eds.), 27–48. Hillsdale, NJ: Lawrence Erlbaum. Shepard, R.N. 1974. “Representation and structure in similarity data: Problems and prospects.” Psychometrika 39: 373–421. Sloman, S.A. 1999. “Rational versus arational models of thought.” In The Nature of Cognition, A Bradford Book, R.J. Sternberg (ed.), 557–585. Cambridge, MA: MIT Press. Taber, K.S. 2001. “When the analogy breaks down: Modelling the atom on the solar system.” Physics Education 36 (3): 222–226. Tversky, A. 1977. “Features of similarity.” Psychological Review 84: 327–352. Tversky, A. and Gati, I. 1978. “Studies of similarity.” In Cognition and Categorization, E. Rosch and B.B. Lloyd (eds.), 79–98. Hillsdale, NJ: Lawrence Erlbaum. Weitzenfeld, J.S. 1984. “Valid reasoning by analogy.” Philosophy of Science 51 (1): 137–149. Wiggins, D. 2001. Sameness and Substance Renewed. Cambridge, UK: Cambridge University Press.

Appropriateness and felicity conditions A theoretical issue Etsuko Oishi

Fuji Women’s University, Japan

The present paper gives a speech-act-theoretic explanation of the concept of appropriateness. In the speech act theory proposed in the present paper, the mechanism of performing an illocutionary act is explained as the process whereby a linguistic form, which represents a linguistic convention, becomes a linguistic artefact by the speaker’s act of uttering and the hearer’s uptake. Appropriateness and felicity conditions concern how such a linguistic artefact is created, and its analysis clarifies a specific relationship between an illocutionary act and context. Upon this interpretation, I examine the ways in which an utterance becomes appropriate or inappropriate, felicitous or infelicitous, and illustrate a construct of the internal context of performing an illocutionary act.

1.

Introduction

The present paper purports to contribute to an explanation of the mechanism of communication by exploring the concepts of appropriateness and Austin’s (1962) felicity conditions. In particular, we address the issue of how the speaker utters something to the hearer with force; and this is explained as the process whereby a linguistic form, which represents a linguistic convention, becomes a linguistic artefact by the speaker’s act of uttering and the hearer’s uptake. We assert that appropriateness and felicity conditions are concepts which concern how the linguistic artefact is created. We develop the argument in the following order. In Section 2 we discuss the concept of appropriateness, and we specify it as a relationship between a sentence/ expression and a context of use. The context of use, which is distinct from the historic situation, is to be explained through a discussion of appropriateness/inappropriateness. As a related concern, we raise a theoretical question, i.e., what it is to describe a linguistic act as appropriate or inappropriate, and how it should be incorporated into the model of communication. In Section 3 we analyze Austin’s

56

Etsuko Oishi

felicity conditions and assert that they are the conditions by which a linguistic form becomes a linguistic artefact. Three aspects of the context are explained. In Section 4 we give a model of performing an illocutionary act, and explain it in terms of the three aspects of context. We also discuss how performative utterances and non-performative utterances execute an illocutionary act differently. This is followed by a brief conclusion.

2.

Appropriateness

Appropriateness is a well-discussed concept among pragmatists, discourse analysts, sociolinguists, and anthropologists. Levinson (1983), for example, uses the concept of appropriateness in the discussion of possible definitions of pragmatics: Pragmatics is the study of the ability of language users to pair sentences with the contexts in which they would be appropriate. (Levinson 1983: 24)

Levinson (1983: 25–26) in fact rejects this definition of pragmatics for the following reasons: (1) such a definition makes the field of pragmatics identical with sociolinguistics, (2) it requires a fundamental idealization of a culturally homogenous speech community, (3) since speakers of a language do not always comport themselves in a manner recommended by the prevailing mores, such a definition would make the data of pragmatics stand in quite an abstract relation to what is actually observable in language usage, (4) since pragmatic constraints are generally defeasible, such a definition would wrongly predict conditions of usage of, e.g., presuppositions, and finally (5) it would not explain exploitation of language use; in being grossly inappropriate, one can nevertheless be supremely appropriate. Although whether or not pragmatics should be defined in terms of appropriateness is not our present concern, this definition makes it clear that appropriateness is the concept which connects sentences and contexts: a sentence is uttered appropriately in a certain context. In other words, like many other pragmatists, Levinson clarifies, by means of the concept of appropriateness, a special relationship between sentences and contexts. Mey (2001) also discusses appropriateness, in particular, pragmatic appropriateness. He analyses the Japanese data in which the customer utters sumimasen “I’m sorry” to the clerk for an unpaid service, and says, “we see how the Japanese expression sumimasen ‘I’m sorry’ appears unexpectedly at a point where we in English assume an expression of gratitude to be in order, such as ‘Thanks a lot’”(2001: 263). Mey concludes as follows:

Appropriateness and felicity conditions

What we are dealing with here is not a matter of what expressions ‘mean’, abstractly taken, or one of how a speech act such as ‘thanking’ or ‘apologizing’ can be defined, in accordance with the standard accounts …. Rather, the question is one of the pragmatic appropriateness of a particular expression in a particular context of use. The problem is that those contexts of use tend to be rather different from culture to culture, and consequently from language to language. (Mey 2001: 263)

Mey describes pragmatic appropriateness as a relationship between a particular expression and a context of use, which is culturally specific and language-specific. In both Levinson (1983) and Mey (2001), an interpretation of context is elucidated by the concept of appropriateness: the use of a sentence/expression is associated with a context, according to which a particular token of the sentence/expression is evaluated either as appropriate or inappropriate. In Fetzer (2004) the connection between a sentence/expression and the context of use is explained as embeddedness of the force of the communicative action, its propositional content, and its linguistic representation in the linguistic, sociocultural, and social contexts. And appropriateness is explained as follows: Appropriateness is a social- and communicative-action-based construct which is calculated with regard to the connectedness between the force of the communicative action, its propositional content, its linguistic representation and their embeddedness in the immediate linguistic, sociocultural and social contexts, and their embeddedness in the remote linguistic, sociocultural and social contexts. Appropriateness is a constitutive part of practical reasoning and manifests itself in felicitous communication in which the coparticipants’ information wants and face wants are satisfied. Analogously to the concepts of acceptability and wellformedness, appropriateness covers both the micro domains of a sentence, utterance or communicative action and macro domains of dialogue and discourse. (Fetzer 2004: 85–86)

Why is it important, if it is, to explain, by means of the concept of appropriateness, the connectedness of a token of a sentence/expression in a historic situation with the context of use? Adopting from Austin (1961), we use the term ‘a historic situation’ for a spatio-temporal situation where a sentence/expression is uttered/used. In his paper titled “Truth”, Austin uses the term ‘historic situations’, which contrasts with ‘the type of situation’, to describe the difference between demonstrative conventions and descriptive conventions. Austin then says, “[a] statement is said to be true when the historic state of affairs to which it is correlated by the demonstrative conventions (the one to which it ‘refers’) is of the type with which the sentence used in making it is correlated by the descriptive conventions” (Austin 1961: 122).

57

5

Etsuko Oishi

The notion of ‘a historic situation’ would seem applicable not only to straightforward statements but to other types of communication, since a communicative act occurs in a historic situation, and, as performative utterances show, can be about the historic situation in which it occurs. Why is it important to explain the embeddedness of the linguistic value and communicative force of a sentence/expression in the immediate and remote linguistic, socio-cultural, and social contexts? Is it equivalent to explaining conditions for, or elements of, the use of a particular sentence/expression? This idea presupposes there being a somewhat neutral context of use for any sentence/expression, which one may access by scrutinizing the sentence/expression and the situation where it can be used. As explained above, such an abstraction of use of context is cautioned against by Levinson (1983: 25–26): the context in such a sense might only have an abstract relation to a particular use of a sentence/expression, and does not explain exploitation of language use. In being grossly inappropriate, Levinson claims, one can nevertheless be supremely appropriate: the use of the sentence/expression in this way creates special effects rather than just violating linguistic or socio-cultural conventions. The abstraction of the context of use is also challenged by critical discourse analysts (Fairclough 1995). Furthermore, if the context is an abstract entity which is independent of a particular use, it is not clear how one can learn the context of use, and how it is possible for the context of use to change as diachronic studies show. Even if we succeed in explaining the context of use for a sentence/expression by describing its use, we still need to explain what it is for a sentence/expression to have the context of use. In other words, what is it to use a sentence/expression in general, and how is the use/usage of the sentence/expression in general related to a particular use of it in a historic situation, which can be inappropriate or abusive? Therefore, what we should do is not to construct an abstract entity of the context of use, which does not contribute to the explication of the mechanism of communication, but rather to develop a model of communication which explains what it is to use a sentence/expression, and to situate the concept of context of use within it. Dummett (1993) urges for such a model, which would present a systematic account of how language functions and, according to him, this would be distinct from the account of the points the speaker has in saying what s/he says. What it is that someone says … is determined, not by his particular intentions, but by what is involved, as such, in knowing the language, together with the words he used and the circumstances in which he used them. It is determined, that is, by what is particular to that language, and is or might be different in other languages, in other words by the conventions whose acquisition constitutes learning

Appropriateness and felicity conditions

the language. To give a theory of meaning, a description of how a language functions, is to give a systematic account of these conventions; such an account will determine the meaning of each utterance just so far as fixing what it is that he actually says. (Dummett 1993: 209–210)

We try to explicate the context of use through a description of how uttering a sentence/expression is judged appropriate or inappropriate. That is, we assume that elements in terms of which an utterance becomes appropriate or inappropriate signify the construct of the context of use. We have to develop a model of communication within which this assumption makes sense. We start with a somewhat bold hypothesis of linguistic communication: communication is an act of creating a linguistic artefact, which is an amalgamation of a linguistic convention associated with a linguistic form and a physical entity, such as sounds or letters, and, metaphorically speaking, the reality of the creation of such an artefact is felt as the context. Let us explain this model. Imagine that X uttered a linguistic string L to Y: the linguistic string L that X uttered is not just a linguistic string. For example, when X said: (1) You are a damn fool,

and Y replied by saying: (2) How dare you,

what X did was not just to say “You are a damn fool”, which might be reported as “X said ‘You are a damn fool’”, or “X said that Y was a damn fool”. This is a locutionary act. Austin subcategorizes locutionary acts into phonetic acts, phatic acts, and rhetic acts, and, accordingly, “X said ‘You are a damn fool’” and “X said that Y was a damn fool” report a phatic act and a rhetic act respectively (Austin 1962: 94–95). X’s utterance You are a damn fool is rather a linguistic event, which might be summarized later such as X insulted or criticized Y. In Austin’s terms, X’s utterance You are a damn fool is an illocutionary act, which is described as an act of insulting or criticizing. Whether X’s utterance was in fact an insult or criticism is not the issue at the moment. What is significant is that a certain utterance with a certain voice quality has become something which is neither the purely linguistic value that the linguistic string represents within the language, nor purely a physical entity of a string of sounds produced by organs of a particular person, i.e. X. It is more like a linguistic artefact, which has its own form, i.e. sounds, and a communicative value as the act of insulting or criticizing. There is another aspect of the creation of an artefact: identifying the artefact as a particular act consists of recognizing certain circumstances in the historic

59

60

Etsuko Oishi

situation. Identifying the artefact as an act of insulting, for example, involves recognizing offensive circumstances in the historic situation. This indicates that there is a special internal relationship between a certain illocutionary act and the circumstances in which it is performed. This relationship reveals itself in the process by which a linguistic artefact is created: an utterance of a particular string of sounds, which represents a linguistic value, combined with certain circumstances in the historic situation, becomes an artefact as a particular illocutionary act. We may call such circumstances the context of performing the act. Recognizing those circumstances in the historic situation is part of creating an artefact, rather than the background to it. Let me use my own anecdote. When I had to leave the city where I was doing my Ph.D. and return to my native country, my supervisor said to me: (3) I promise I’ll let you finish your Ph.D.

The reality that the utterance was a promise co-existed with the recognition of the circumstances at that time: it was extremely difficult to finish my Ph.D. without my supervisor’s help and receiving his help would become difficult due to my departure. The reality of his act of a promise was, so to speak, the reality of his commitment under the existing circumstances and circumstances in the foreseeable future: although finishing my Ph.D. would entirely depend on my own effort, which was suggested by his words “… I’ll let you finish your Ph.D.”, he guaranteed the situation where my effort would lead to finishing my Ph.D., which meant he would give me more help and support than he would have given to his supervisees in a normal situation (and he did). This shows that creating a linguistic artefact as a certain illocutionary act involves recognizing certain circumstances in the historic situation, and those circumstances become apparent when there is a reality that the artefact of the act is created. In other words, the circumstances of performing an act, i.e., the context of using a sentence/expression as performing an act, can be known through the description of circumstances in the historic situation where an artefact as a particular illocutionary act is created. Our hypothesis is that the concept of appropriateness concerns the creation of such an artefact: whether a certain utterance is appropriate or inappropriate depends, at least partially, on whether or not the artefact as a certain illocutionary act is created in the historic situation. Any utterance is the speaker’s communicative move to the hearer, and, in that sense, a communicative event. When we say that an artefact is created, however, we mean a more specific move; the move the speaker makes by uttering a sentence/expression as a certain illocutionary act, which is consented by the hearer’s uptake. If we are right about this communication model, then the ways in which the utterance becomes appropriate or inap-

Appropriateness and felicity conditions

propriate show how an artefact as a certain illocutionary act is or is not created in a historic situation. Furthermore, since the utterance becomes appropriate or inappropriate in terms of something, which is the context for creating an artefact as an illocutionary act, an analysis of the elements in terms of which the utterance becomes appropriate or inappropriate shows the construct of the context for performing the act. Specifically, inappropriateness exposes the ways in which the utterance would have been appropriate: the elements of context that should have existed in order for the utterance to be uttered appropriately. An analysis of those ways or elements will reveal the construct of the context. This is what linguists describe as a special relationship between contexts and the use of a sentence/expression to perform an illocutionary act. To analyze this is a target of research not only for speech act theorists but also for pragmatists in general. How shall we begin an analysis of appropriateness/inappropriateness? We start with Austin’s (1962) concept of felicity conditions. Austin’s felicity conditions are usually interpreted as nothing more than the conditions for performing a speech act felicitously. We, however, claim that Austin, in describing how an act becomes felicitous or infelicitous, describes the special relationship between an act performed and its circumstances, i.e. between a speech act and its internal context. Such a description illustrates what it is for an act to be performed. We discuss this in the following section.

3.

Austin’s felicity conditions

The following are Austin’s felicity conditions: (A.1) There must exist an accepted conventional procedure having a certain conventional effect, that procedure to include the uttering of certain words by certain persons in certain circumstances, and further, (A.2) the particular persons and circumstances in a given case must be appropriate for the invocation of the particular procedure invoked. (B.1) The procedure must be executed by all participants both correctly and (B.2) completely. (Γ.1) Where, as often, the procedure is designed for use by persons having certain thoughts or feelings, or for the inauguration of certain consequential conduct on the part of any participant, then a person participating in and so invoking the procedure must in fact have those thoughts or feelings, and the participants must intend so to conduct themselves, and further (Γ.2) must actually so conduct themselves subsequently. (Austin 1962: 14–15)

61

62

Etsuko Oishi

These are the conditions for a speech act to be performed. The focus on analyzing the felicity conditions tends to be on whether or not, according to these conditions, a certain act is performed, or whether or not these conditions really decide the performance of a certain act. Searle (1969) interprets these conditions as constitutive rules: they are jointly constitutive of the various illocutionary forces1. Within the theoretical framework we have proposed, and likewise the concept of appropriateness/inappropriateness, the concept of felicitousness/infelicitousness concerns how a linguistic artefact as a certain illocutionary act is created. That is, when one feels that an utterance is infelicitous, then an artefact as a certain illocutionary act has not been created for one reason or another. So how an utterance becomes infelicitous in turn implies circumstances for the felicitous performance of an act, i.e., the internal context of an act. According to this interpretation, differences between circumstances in a historic situation and those as an internal context are a source of infelicities. We argue that Austin’s felicity conditions, (A), (B), and (Γ), illustrate three different aspects of the internal context of performing an act. We describe these aspects of the context as conventionality, performativity, and personification, which are specified by the felicity conditions, (A), (B), and (Γ), respectively. Violations of conditions (A.1) and (A.2) are described as misinvocations, in which the purported act is disallowed (Austin 1962: 17–18). That is, felicity conditions (A) specify the aspect of the context where a certain act is allowed. This is the aspect of conventionality. The felicity condition (A.1) shows how the speaker and the hearer share linguistic and socio-cultural conventions according to which to utter certain words in certain circumstances by certain persons is counted as performing a certain speech act, which accompanies a certain conventional effect. So the utterance in example (4) indicates circumstances as the internal context in which the speaker and the hearer share a Muslim convention of divorce, where the speaker’s utterance, together with the hearer’s acceptance, produces a conventional result, i.e., divorce. (4) I divorce you. I divorce you. I divorce you.2

Because of this internal context for performing an act of divorce, the same sentence uttered to a husband by his wife, or to a wife by her husband, in a Christian country, when both parties are Christian rather than Muslim, would not create an

1. See Levinson (1983: 238–239) for further discussions. 2. Yasmin, 2004, directed by Kenny Glenaan.

Appropriateness and felicity conditions

artefact of divorce: according to Austin (1962: 16, 27) this case would be a misfire: the procedure invoked is not accepted. Similarly the utterance in (5) indicates circumstances as the context in which the speaker and the hearer share the linguistic convention of performing an act of reprimand: the speaker in an official capacity formally and publicly blames the hearer, who accepts the blame, and this produces a formal record. (5) I reprimand you for your negligence.

Because of this internal context of the performance of the act of reprimand, the same sentence uttered to a hearer who does not understand the significance of the act of reprimand would not create an artefact of reprimand. The felicity condition (A.2) shows another aspect of the context in which particular persons and circumstances, specified by linguistic and socio-cultural conventions, exist. So the utterance in (6) indicates the context in which the speaker, a Christian priest, welcomes the hearer, an infant, to the Christian church in a religious ceremony. Since the infant is too young to consciously accept the welcome, her/his parents show it on her/his behalf. (6) I baptize thee in the name of the Father, and of the Son, and of the Holy Spirit.

Because of this internal context of the act of baptizing, the same sentence uttered to the infant who has already been baptized would not create an artefact of baptizing. Austin (1962: 17–18, 34) classifies an infelicity of this type as a misfire, in particular, a misapplication. Similarly, the utterance in (7) indicates circumstances as a context in which the speaker, who has legitimate authority over the hearer, issues an order, with the hearer’s obedience. (7) I order you to release the prisoners.

Because of this internal context of an act of ordering, the utterance by a speaker who does not have such an authority would not create an artefact of ordering. When an illocutionary act is specified as divorce, reprimand, baptize, or order, a certain convention for each act including circumstances and persons, and the actual existence of circumstances and persons are evoked. It is not the case that a convention, circumstances and persons happen to be there when an act is performed; it is rather the case that a certain act occurs with a linguistic and social specification of the circumstances, without which the act itself is disallowed. This linguistic and socio-cultural specification is conventional: the relationship be-

63

64

Etsuko Oishi

tween an act and a certain convention, circumstances, and persons is determined linguistically and socio-culturally. Within the proposed model of communication, this might be explained as follows: an utterance as an illocutionary act, which includes the specification of a certain convention, circumstances, and persons, becomes an artefact with these specifications. In other words, to recognize an artefact of a certain illocutionary act is partially to recognize, in the historic situation, the convention, circumstances, and persons evoked by the act. We call the convention, circumstances, and persons evoked the conventional aspect of the context for performing the act. Violations of the second type of the conditions in (B.1) and (B.2) are described as misexecutions, in which a purported act is vitiated (Austin 1962: 17–18). That is, the felicity conditions (B) specify the aspect of the context where a certain act is actually executed by the speaker’s performance and the hearer’s response. We describe this as the performative aspect of the context. The felicity condition in (B.1) describes the aspect in which the speaker actually utters something to construct the event of performing a particular act. That is, in using her/his own voice, the speaker achieves the performance as an illocutionary act in an historic situation. For example, the utterance in (8) indicates the performance of an act of inaugurating, which is created by a certain voice of the speaker: (8) I inaugurate the new US embassy in Caracas.

Because of this internal context of performing an act of inaugurating, the utterance the speaker makes as a slip of tongue, I inaugurate the new US empire in Caracas, would not create an artefact of performing the act of inaugurating: the inauguration of the new US embassy does not occur in this performance. Similarly, the utterance in the historic situation where there hasn’t been an embassy would not create an act of inaugurating the embassy in question: the utterance is not really the inauguration of the embassy, because it is not the new US embassy in a strict sense. Austin (1962: 17–18, 36) describes infelicities of this kind as flaws. The felicity condition in (B.2) exposes another aspect of the context, in which the speaker’s performance is completed by the hearer’s performance as a response. Austin (1962) includes other procedures which complete the speaker’s performance: … my attempt to challenge you is abortive if I say ‘I challenge you’ but I fail to send round my seconds; my attempt ceremonially to open a library is abortive if I say ‘I open this library’ but the key snaps in the lock; conversely the christening of a ship is abortive if I kick away the chocks before I have said ‘I launch this ship’. (Austin 1962: 37)

Appropriateness and felicity conditions

We, however, focus on the hearer’s response as the procedure to complete the speaker’s performance: the speaker’s performance of a certain illocutionary act is acknowledged and guaranteed as the act by a certain response from the hearer. For example, the speaker’s performance of an illocutionary act of ordering in (7) might be completed by the hearer’s response as follows: (9) Yes, sir.

Alternatively, the speaker’s performance might be completed by the hearer’s action of releasing the prisoners. Because of this internal context of completing the act of ordering, the performance of the act of ordering would not be completed by a response from the hearer such as I don’t give a damn about it. Similarly the act of ordering would not be completed if the hearer misheard the utterance and did not respond to the act performed. Austin (1962: 17–18, 36–37) says that in these cases, the attempt to perform an illocutionary act is abortive, and describes this type of infelicity as a hitch. The felicity conditions (B) reveal the performative aspect of context. Within the proposed model of communication, we explain this as follows: the convention of performing an illocutionary act, which is substantiated by the speaker’s actual performance and the hearer’s actual response, becomes an artefact with a form; the form is shaped by the speaker’s performance in using her/his voice as a communicative move to the hearer, which is completed by the hearer’s reaction. Because of this structure of communication, to recognize an artefact of a certain illocutionary act is to partially recognize the speaker’s performance in a historic situation, which is completed by the hearer’s response. Austin’s felicity conditions (A) and (B) allude to two different but interrelated ways in which illocutionary acts succeed or fail. According to the felicity conditions (A), illocutionary acts succeed or fail because the conventional procedures for performing the acts do or do not exist, or circumstances and persons specified by those conventions do or do not exist. According to the felicity conditions (B), illocutionary acts succeed or fail because the conventional procedures for performing illocutionary acts are or are not substantiated by the speaker’s accurate performance and by the hearer’s response to complete the performance. Therefore, felicitous cases under the conditions (A) and (B) are cases where the utterance as an illocutionary act, which accompanies the specification of a convention, circumstances, and persons, occurs with the substantiation of them by the speaker’s accurate performance and the hearer’s expected response to complete the performance. That is, a certain convention and circumstances and persons are evoked by the identification of the illocutionary act being performed, and substantiated by the speaker’s performance and the hearer’s response in the historic

65

66

Etsuko Oishi

situation. As a result, there exists an artefact which represents a linguistic convention, circumstances and persons, and substantiates them with the speaker’s performance and the hearer’s response in a historic situation. Such an artefact is an amalgamation of the linguistic and socio-cultural conventions, the speaker’s performance and the hearer’s response. Therefore, to create such an artefact is, in one sense, to substantiate the abstract convention by an actual performance, and, in another sense, to mark a physical entity, i.e., a string of sounds, with a linguistic and socio-cultural value. A violation of the felicity conditions (Γ.1) and (Γ.2) is described as an abuse, in which the professed act is hollow (Austin 1962: 18). Austin distinguishes these conditions from conditions (A) and (B). While a violation of the felicity conditions from (A.1) to (B.2) results in non-performance, that is to say, a purported act is not performed, a violation of the felicity conditions (Γ.1) and (Γ.2) does not result in non-performance. Although it is a case of abuse, a purported act is performed nonetheless. Austin says: The first big distinction is between all the four rules A and B taken together, as opposed to the two rules Γ (hence the use of Roman as opposed to Greek letters). If we offend against any of the former rules (A’s or B’s) – that is if we, say, utter the formula incorrectly, or if, say, we are not in a position to do the act because we are, say, married already, or it is the purser and not the captain who is conducting the ceremony, then the act in question, e.g. marrying, is not successfully performed at all, does not come off, is not achieved. Whereas in the two Γ cases the act is achieved, although to achieve it in such circumstances, as when we are, say, insincere, is an abuse of the procedure. Thus, when I say ‘I promise’ and have no intention of keeping it, I have promised but …. (Austin 1962: 15–16)

Within the proposed model, this suggests that the felicity conditions (A) and (B) specify how a linguistic artefact is created. A linguistic artefact is created when an utterance as an illocutionary act, which accompanies the specification of a convention, circumstances, and persons, occurs with the substantiation of them by the speaker’s accurate performance and the hearer’s expected response to it. In other words, a linguistic artefact comes into existence when the linguistic and socio-cultural conventions evoked by the utterance as a particular illocutionary act are substantiated by the speaker’s performance and the hearer’s response in a historic situation. If any of these conditions (A.1) to (B.2) is violated, the utterance lacks either specification or substantiation, and, therefore, there is no linguistic artefact created. On the other hand, the violation of the felicity conditions (Γ) does not endanger the creation of an artefact itself: a linguistic artefact is created, but the creation is a case of abuse because the created artefact is hollow.

Appropriateness and felicity conditions

The felicity conditions (Γ) specify the aspect of the context where the professed act is sincere: the speaker is responsible for the creation of a linguistic artefact, and indicates a certain thought/feeling/intention, and, when applicable, commits herself/himself to conducting a certain action subsequently. The felicity condition (Γ.1) describes the aspect of the context in which the speaker is sincere in performing an illocutionary act, and indicates, as her/his own, the thought/feeling/intention specified by the illocutionary act s/he is performing. For example, the utterance in (10) indicates that the speaker is sincere in performing an act of welcoming: the feeling of delight in having the hearer in her/his company is her/his own feeling: (10) I welcome you.

Because of this internal context of performing an act of welcoming, the speaker who does not feel delighted to have the hearer in her/his company would not internalize the creation of the act of welcoming: the feeling of delight is not hers/his although s/he indicates that it is. This is a fake move in communication, which endangers the communicative base between the speaker and the hearer: the speaker does not have a feeling/thought/intention which accompanies the creation of the linguistic artefact in question, and, therefore, s/he is not responsible for the speech act s/he is performing. The speaker’s welcome in such a case is hollow. Austin (1962: 18) describes this kind of infelicities as insincerities. The felicity condition (Γ.1) suggests the process by which the speaker becomes the speaking subject: while specifying the present communicative move as a certain illocutionary act, the speaker identifies the feeling/thought/intention specified by the illocution as her/his own, and presents her/himself to the hearer as an individual who has the feeling/thought/intention. The speaking subject is not the speaker who passively follows the linguistic conventions, but the one who, using those conventions, communicates with the hearer and indicates her/his own feeling/thought/intention. This process, however, is distinct from the process by which the speaker expresses her/his intention by uttering a sentence. In standard speech act theory this is assumed to be the process of performing an illocutionary act. Sbisà (2001) summarizes this as follows: Since Searle (1969: 46–49), the illocutionary act has generally been conceived as the act a speaker successfully performs when, uttering a sentence with a certain intention in certain circumstances, he or she gets the hearer to understand his or her intention. The speaker’s communicative intention determines what illocutionary act he or she should be taken to perform and therefore what illocutionary force his or her utterance may have. (Sbisà 2001: 1795)

67

6

Etsuko Oishi

The idea of the standard speech act theory that the speaker expresses her/his intention by uttering a sentence presupposes the a priori concept of the speaker’s intention: there is a definite intention which belongs to a particular individual at a particular time, and the individual is accessible to this intention without using the language. That is, the speaker clearly knows what intention s/he has before performing a certain act. While explaining the standard conception of illocutionary force, Sbisà (2001) says: The speaker’s communicative intention must be a definite one. A speech act must have one or the other of the illocutionary points available (committing the speaker to the truth of the expressed proposition, trying to make the addressee do something, and so forth) (Searle 1976). It does not matter here whether recognition of the speaker’s communicative intention is achieved by recognizing the conventional meaning of the sentence uttered (as in Searle 1969, and in ‘direct’ speech acts according to Searle 1976) or by making inferences (as in Bach & Harnish 1979; Sperber & Wilson 1986; Harnish 1994; and in ‘indirect’ speech acts in Searle 1976). (Sbisà 2001: 1795)3

This, however, does not seem to be an accurate description of performing an illocutionary act. We sometimes realize what intention we had at the time when we say something, or after we say something, or even a long time after we say something. For example, the speaker might realize, only after her/his utterance, By the way, do you have any plans for this summer?, that s/he wanted to release tension which had mounted between her/him and the hearer in discussing a controversial issue. It is not clear that a definite intention of, say, asking a question about the hearer’s summer plans in order to change the subject existed before the utterance was made. As another example, the speaker might recognize her/his intention of making an utterance, We are not ready to introduce such a new scheme, a few months after the utterance. Let us consider the following scenario. The speaker realizes that her/his intention in making the utterance was not to express her/his reluctance to introduce the new scheme, but rather to express her/his distrust of the new management that proposed it. This intention was not clearly grasped at the time of utterance, but became apparent when the management made several disastrous mistakes, and the speaker felt that this was something s/he had expected. When we think of these cases, we have a reasonable doubt about the validity of the explanation of the standard speech act theory; the speaker expresses her/his intention by uttering a sentence.

3. Sbisà (2001) claims that some aspects of standard speech act theory are incompatible with the features of linguistic action emerging from research on mitigation/reinforcement.

Appropriateness and felicity conditions

The explanation which we give by reanalyzing the felicity condition (Γ.1) seems to be more reasonable: while specifying the present communicative move as a certain illocutionary act, the speaker identifies the feeling/thought/intention specified by the illocution as her/his own, and presents her/himself to the hearer as an individual who has the feeling/thought/intention. According to this explanation, the speaker is responsible for having the feeling/thought/intention specified by the illocution, but s/he is not required to identify beforehand the intention of performing the illocutionary act. The hearer, therefore, can reasonably assume that the speaker has the feeling/thought/intention specified by the illocution, unless s/he is not sincere and not responsible for the creation of this linguistic artefact. Grice (1989) seems to hypothesize the speaking subject in this sense. He describes a conversational participant who observes the Cooperative Principle and the conversational maxims, in particular, Maxim of Quality, for our present interest. Such a conversational participant internalizes a thought expressed by the utterance as her/his own thought, and is responsible for her/his own creation of a linguistic artefact: s/he does not say what s/he believes to be false and does not say that for which s/he lacks adequate evidence (Grice 1989: 27). The felicity condition (Γ.2) describes another aspect of the context in which the speaker indicates a commitment to conduct a subsequent action. For example, the utterance in (11) indicates circumstances as the internal context in which the speaker commits her/himself to conducting a subsequent action of supporting the hearer: the speaker indicates this commitment as her/his own: (11) I promise to support you.

Because of this internal context of performing an act of promising, the speaker who will not support the hearer when s/he needs it would not internalize as a commitment the creation of the artefact of promising: the speaker acts as if s/he has promised, but her/his promise is hollow. The process by which the speaker internalizes the creation of the artefact also seems to be the process by which the speaker becomes the speaking subject: the speaker indicates a commitment to conduct a subsequent action as her/his own, and presents her/himself to the hearer as a person who has the will to conduct a certain action specified by the illocutionary act. The speaker presents her/himself to the hearer as a person with this will not only at the time of utterance: s/he does so as long as this commitment is valid. The felicity conditions (Γ) show the personification aspect of context. The utterance as an illocutionary act is personified by the speaker’s internalization of a certain feeling/thought/intention as her/his own, and also by the speaker’s in-

69

70

Etsuko Oishi

ternalization of a certain commitment to a subsequent conduct as her/his own. By means of these internalizations, the speaker becomes the speaking subject: the speaker presents her/himself to the hearer as a person who has the feeling/ thought/intention specified, and the will to conduct the action specified subsequently. Within the proposed model of communication, this might be explained as follows: an utterance as an illocutionary act, which is personified by the speaker’s internalization of a certain feeling/thought/intention and a certain commitment, becomes an artefact with a performer’s mark; it is the artefact of the performer who expresses, as an individual person, a certain feeling/thought/intention, and a certain commitment. Because of this structure of communication, to recognize an artefact of a certain illocutionary act is partially to recognize, in a historic situation, a certain performer who expresses the feeling/thought/intention and the commitment specified by the illocutionary act. We have so far developed the concept of context by reanalyzing Austin’s felicity conditions. We started with the idea that appropriateness/inappropriateness and felicities/infelicities concern the concept of context: the elements, in terms of which the utterance as an illocutionary act becomes appropriate or inappropriate, or felicitous or infelicitous, thereby illustrate the construct of the context. By describing the felicity conditions, we have presented three aspects of context in order for an illocutionary act to be performed: we call them conventionality, performativity, and personification. That is, the elements of performing an illocutionary act, other than uttering a certain sentence, include certain conventions existing and applicable, along with circumstances and persons existing (conventionality); the speaker’s actual, accurate performance and the hearer’s actual, expected response (performativity); and a thought/feeling/intention, and a commitment personified (personification). In the next section, we develop a communication model on the basis of these analyses, while discussing theoretical implications of the elements of the context and our model of communication.

4.

The context for performing illocutionary acts

What we did in the preceding section was not to rename Austin’s felicity conditions. We rather attempted to describe elements of performing an illocutionary act, which we described as the elements of the internal context of performing an illocutionary act. In the present section, we try to explain the mechanism of performing an act: how these elements are interrelated for an illocutionary act to be performed, i.e., for a linguistic artefact to be created. To perform an illocutionary act is, firstly, the process by which a linguistically specified convention, circumstances, and persons are substantiated by the speak-

Appropriateness and felicity conditions

er’s actual performance and the hearer’s actual response in a historic situation. To utter the sentence in (12), in a historic situation, to a hearer who responds to it by uttering the sentence in (13), is to evoke a linguistic/socio-cultural convention of advising, the circumstances of advice, and persons as an adviser and an advisee. It is, at the same time, to specify the present performance in the historic situation as advising: the speaker as an adviser advises the hearer as an advisee in the circumstances of advising. (12) I advise you not to enter the room. (13) I won’t.

This is the process by which the convention, and the circumstances and persons specified by the convention are substantiated by the speaker’s actual performance and the hearer’s actual response. It is the process of situating the present performance, together with the speaker, the hearer, and the circumstances, in the frame of the act of advising. It is also the process by which the speaker’s performance and the hearer’s response evoke and substantiate a linguistic/socio-cultural convention, and circumstances and persons specified by the convention. A linguistic/ socio-cultural convention of performing an illocutionary act does not exist independently of actual performances in the way specified by the convention. This is the aspect of meaning which Wittgenstein (1953) specifies by his well-known slogan, ‘meaning is use’. The speaker’s performance and the hearer’s uptake, therefore, might create a new convention. Because of this structure of the illocutionary act of advising, there exists a linguistic artefact which is an amalgamation of the linguistic/socio-cultural convention of advising and a physical entity as a communicative move by the speaker to the hearer in the historic situation. To perform an illocutionary act is, secondly, the process in which a certain feeling/thought/intention and a certain commitment by the speaker are identified by the specification of the illocutionary act, and also indicated in a historic situation. To utter the sentence in (12) in a historic situation to the hearer who responds to it by uttering the sentence in (13) is to indicate, as the speaker’s own thought, that entering the room is disadvantageous for or unbeneficial to the hearer. This is the process by which the speaker indicates her/his own feeling/ thought/intention and commitment by identifying them as the feeling/thought/ intention and commitment for performing the act of advising. The performance of illocutionary acts reveals an interesting nature of linguistic communication. When an utterance is uttered as an illocutionary act, three different factors are combined as elements of performing the act: first, the linguistic conventions associated with the expression/sentence used, which exist in the linguistic behavior not only of the present speaker and hearer, but also of a

71

72

Etsuko Oishi

wider range of language users, such as the users of the language or the dialect; second, a performance as a communicative move by the present speaker in a historic situation, which is completed by the present hearer’s uptake of it and/or her/his response to it; and third, the present speaker’s indication of her/his own feeling/ thought/intention and commitment. These rather different factors are combined by an identification of the speaker’s performance with the convention, and with the speaker’s indication of her/his feeling/thought/intention and commitment. This structure of illocutionary acts is a source of force. Knowing conventions does not guarantee making a move in communication. Knowing an illocutionary act of ‘advising’, for example, does not guarantee the ability of performing the act of advising. We know, for example, the illocutionary act of ‘declaring a war’, and how it is different from, say, the act of ‘declaring one’s intention’, but the majority of us would not know how and when to make a move to declare a war if we were in that position. Performances as a communicative move do not have value in themselves. To use a loud tone of voice might or might not be a performance of threatening, to utter a few sentences might or might not be a performance of reading a poem, or to say a certain thing might or might not be a performance of congratulating. Therefore, to use a loud tone of voice, to utter a few sentences, or to say a certain thing in itself does not have a specific value. One’s feeling/thought/ intention and commitment are vague without being specified. When one says, “I’ll be there”, s/he might commit her/himself to be there, give a prediction that s/he might be there, or intend to be there. Even the speaker her/himself might not know it for sure. To perform an illocutionary act by means of a so-called performative sentence is to specify a particular convention, and a particular performance of the speaker and a particular response by the hearer, as well as a particular feeling/ thought/intention and a particular commitment by the speaker, and to combine them as a certain type of illocutionary act. Therefore, an artefact created as a particular illocutionary act has a communicative value, a form as a communicative move, a personification of a feeling/thought/intention and a commitment. This is why the felicitous utterance of a performative sentence has force. We have deliberately analyzed performative utterances in which, when they are felicitous, the conventionality aspect of context, the performativity aspect, and the personification aspect all coincide, and one total context is indicated. That is, a specific convention, circumstances, and persons are evoked and substantiated by a specific performance of the speaker and a specific response of the hearer, and a specific feeling/thought/intention and a specific commitment are personified. Performative utterances have a mechanism of specifying which convention is evoked, which performance as a communicative act is indicated, and which feeling/thought/intention and commitment are personified by the utterance.

Appropriateness and felicity conditions

However, non-performative utterances do not have a mechanism which explicitly specifies these elements. If so, how are illocutionary acts performed implicitly by non-performative utterances? In other words, how are a specific convention, circumstances, and persons evoked implicitly? How are a specific performance of the present speaker and a specific response of the present hearer indicated implicitly? And how are a specific feeling/thought/intention and a specific commitment personified implicitly? Performative utterances are characterized by the use of the first person singular and of the present indicative active (Austin 1962: 57), but, as Austin himself explains, the same or a similar illocutionary act can be performed by other types of utterance. Using the mechanism of performing an act, which we have already described, we will show below how those utterances can perform an illocutionary act. Austin (1962: 57–58) shows that sentences in the first person plural, such as those in (14), and sentences in the second and third person (singular or plural) in a passive voice, such as those in (15), can be used to perform an illocutionary act: (14) a. We promise that you will be compensated for your loss, b. We consent that you should marry her. (15) a.

You are hereby authorized to pay 1,350 pounds for your council tax with 10% reduction, b. Passengers are warned to cross the track by the bridge only.4

Examples (14a) and (14b) are cases in which the speaker’s performance is indicated not under the speaker’s own authority but under the authority of something to which the speaker belongs, such as an organization like a company or a government (as in the case in (14a)), or a couple or a family (as in the case in (14b)). The illocutionary act performed by this type of utterance has weaker force than that of the corresponding performative utterance. This is because the performance of a certain person, i.e. the speaker, is symbolically interpreted as the performance of the company or the family. Therefore, a certain feeling/thought/intention and a certain commitment are symbolically personified in the performance the speaker makes as the feeling/thought/intention and commitment of the company or the family. Example (15a) is a case in which the speaker’s performance is indicated by the passive action of the hearer: a person is addressed as ‘you’, i.e., as the hearer, and

4. Examples (14a), (14b), (15a) are workings of Austin’s partial sentences; ‘We promise…’, ‘We consent …’, and ‘You are hereby authorized to pay …’ respectively. Example (15b) is Austin’s original sentence.

73

74

Etsuko Oishi

is authorized to do something. This indicates that there is a person who addresses the person as ‘you’, i.e., the speaker, who authorizes her/him to do it. Similarly, in example (15b), certain persons are addressed as ‘passengers’, and warned. This indicates a person who refers to them as ‘passengers’ and performs an act of warning. It seems quite interesting that there is a difference of strength of illocutionary force between these cases. Example (15b) is characterized by indirectness of the performance: the performance of warning by the speaker to the hearer(s) is implicitly indicated through the warning to the passengers, which implies someone who warns: the performer is annihilated, and the authority or the personal responsibility by which a certain thought/intention is personified is not signified. In example (15a), on the other hand, the second person pronoun ‘you’ strongly indicates the speaker who faces a certain person as the hearer and addresses her/ him as ‘you’, and performs an act of authorizing, as a result of which the hearer is authorized. The following is an interesting case: (16) Notice is hereby given that trespassers will be prosecuted.5

It is predicted from the case of (15b) that the readers in the above case are indirectly addressed as ‘trespassers’. They are, however, directly addressed with force: readers who read this notice are identified and addressed as trespassers (unless they have proper authority), and are warned that they will be prosecuted. That is, their actual act of reading the notice identifies them as trespassers, and, therefore, as warned. It is as if, by the act of reading itself, readers are trapped into a structure in which they are addressed as a trespasser and warned they will be prosecuted. In this case, the performance of warning is indicated with force, but its performer is behind the screen. Austin (1962: 58) gives the following examples, where an illocutionary act is performed by the utterance of a sentence in imperative, deontic, or subjunctive mood: (17) a. Turn right, b. You may go, c. I should turn to the right if I were you.

The imperative mood in (17a), which shows that the speaker orders the hearer, indicates a performative aspect of performing an illocutionary act, i.e., the speaker’s performance of the act of ordering, which would be explicitly indicated by the performative utterance, I order you to turn right. That is, the imperative mood 5. This is also Austin’s original sentence (Austin 1962: 57).

Appropriateness and felicity conditions

expresses the speaker’s order to the hearer, which is interpreted as the speaker’s performance of the act of ordering to the hearer. The deontic and subjunctive mood in (17b) and (17c) respectively can replace performative utterances which indicate the speaker’s performance to the hearer. The deontic mood of the sentence in (17b) expresses that the speaker permits the hearer’s act of going, which implicitly indicates the speaker’s performance of permitting the hearer to go. This performance would be explicitly indicated by the performative utterance I permit you to go. The subjunctive mood of the sentence in (17c) expresses the speaker’s choice of the action, i.e., turning to the right, in the hearer’s circumstances: this implies that the speaker advises the hearer to perform the action, i.e., turning to the right. This performance would be explicitly performed by the performative utterance, I advise you to turn right. Using the following examples, Austin (1962: 58) shows that one can perform an illocutionary act by uttering the sentence in the past tense. (18) a. You were off-side, b. You did it.

When an action has occurred, and a circumstance for judging the action is created in the historic situation, the speaker’s utterance of the sentence in the past tense, in which the past action is identified, can perform an illocutionary act of judging the action and/or specifying the consequence. Example (18a) shows that, when a certain player has made a move in the game of football, and the judgment of the move is necessary for the game to be continued, the speaker’s identification of the past move, You were off-side, indicates implicitly the performance of specifying the past action and its consequence. This performance would be explicitly indicated by the performative utterance, I rule that you were off-side. Similarly, example (18b) shows that, when the prosecution of a person for a crime and the defense for her/him have been carried out, and the judgment of whether or not the person is guilty for the crime is required, the speaker’s identification of the past event, you did it, implicitly indicates the performance of judging. This would be explicitly indicated by the performative utterance, I find you guilty. Austin (1962: 58) also shows that one can perform an illocutionary act by uttering truncated sentences as in the following: (19) a. Done, b. Guilty, c. Out.6

6. (17a) to (19c) are all Austin’s original examples.

75

76

Etsuko Oishi

These examples show that, when the frame of possible performances of the speaker is provided by circumstances in the historic situation or the preceding utterance, a word that stands for one of those performances indicates its performance. When an illocutionary act of betting is performed, it sets a frame of possible responses:7 the act of betting might be completed by the hearer’s response of accepting the bet, or the hearer might reject it. Therefore, the one-word utterance, Done, in (19a) in such circumstances incompletely but successfully indicates the speaker’s performance of accepting the bet. Similarly because of circumstances in the historic situation, the one-word utterances, Guilty and Out, incompletely but successfully indicate the speaker’s performance of finding a person guilty and calling someone out, respectively. The illocutionary acts performed by non-performative utterances show interesting correlations between illocutionary acts and sentence types, and between illocutionary acts and circumstances of the historic situation in which the act is performed. Specifically we have observed that the performative aspect of performing an illocutionary act is created linguistically and extra-linguistically in various ways. Although we have demonstrated, in limited cases, the aspect of personification created by non-performative utterances, and have not described at all the aspect of conventionality created by non-performative utterances, further research into these aspects are warranted. Such research will explain why those utterances rather than performative ones are used in particular cases.

5.

Conclusion

We started with the idea that there is a special relationship between a sentence/ expression and the context of use. We have described this as the relationship between the utterance as an illocutionary act and its internal context, and we have analyzed different aspects of the context by examining the different ways in which an utterance becomes appropriate or inappropriate, or felicitous or infelicitous. Using Austin’s concept of felicity conditions, we have described three aspects of the context for performing an illocutionary act, i.e., conventionality, performativity, and personification. An illocutionary act is performed when a linguistic/ socio-linguistic convention, and circumstances and persons specified by the convention evoked (by the utterance) are substantiated by the speaker’s actual performance and the hearer’s actual response or/and uptake, in which a particular feeling/thought/intention and a particular commitment are personified. We have

7. The frame of possible responses, however, does not exclude other responses.

Appropriateness and felicity conditions

analyzed performative utterances as the cases in which the linguistic convention, the speaker’s performance and the hearer’s response to complete it, in conjunction with a particular feeling/thought/intention, and a particular commitment, i.e. three aspects of the context, are identified and specified explicitly by the utterance. We have also described how illocutionary acts can be performed implicitly by non-performative utterances, and how diverse the illocutionary acts performed by performative utterances and non-performative utterances may be, specifically in terms of force.

References Austin, J.L. 1950. “Truth”. Proceedings of the Aristotelian Society, Supplementary Volume 24. [Reprinted in Philosophical Papers, J.O. Urmson and G.J. Warnock (eds.), 1961, 117–133. Oxford: Oxford University Press.] Austin, J.L. 1962. How to Do Things with Words. Oxford: Clarendon Press. Bach, K. and Harnish, R.M. 1979. Linguistic Communication and Speech Acts. Cambridge, MA: MIT Press. Dummett, M. 1993. The Seas of Language. Oxford: Clarendon Press. Fairclough, N. 1995. Critical Discourse Analysis: The critical study of language. London: Longman. Fetzer, A. 2004. Recontextualizing Context: Grammaticality meets appropriateness. Amsterdam: John Benjamins. Grice, P. 1989. Studies in the Way of Words. Cambridge, MA: Harvard University Press. Harnish, R.M. 1994. “Mood, meaning, and speech acts.” In Foundations of Speech Act Theory, S.L. Tsohatzidis (ed.), 407–459. London: Routledge. Levinson, S.C. 1983. Pragmatics. Cambridge: Cambridge University Press. Mey, J.L. 1993. Pragmatics: An introduction. Second edition 2001. Oxford: Blackwell. Sbisà, M. 2001. “Illocutionary force and degrees of strength in language use.” Journal of Pragmatics 33: 1791–1814. Searle, J.R. 1969. Speech Acts: An essay in the philosophy of language. Cambridge: Cambridge University Press. Searle, J.R. 1976. “A classification of illocutionary acts.” Language and Society 5: 1–23. Sperber, D. and Wilson, D. 1986. Relevance: Communication and cognition. Second edition 1995. Oxford: Blackwell. Wittgenstein, L. 1953. Philosophical Investigations. Oxford: Blackwell.

77

Appropriateness An adaptive view Thanh Nyan

University of Manchester, UK

Building on previous work predicated on the co-evolution of language and the brain, this article takes the view that (a) appropriate behavior – linguistic or otherwise – is adaptive; (b) from the standpoint of the speaker’s processing system, it is chiefly a matter of activating skills arising from background knowledge. This background knowledge, I begin by arguing, should be seen as part of a decision-making process, as construed by Damasio (1994). Next I provide a sketch of how appropriate behavior arises from the corresponding memory system. Following this I set reasons for viewing linguistic appropriate behavior in terms of the same apparatus. In the final section I consider the implications of this adaptive perspective for the notion of context.

1.

Introduction

It is not uncommon for people to assume that their spontaneous response to a situation is appropriate. What gives them that confidence? Or, more to the point, what enabling factors need to be in place before a behavior (linguistic or otherwise) may be deemed – rightly or wrongly – appropriate by its agent? Assuming one such factor to be a store of knowledge1 (henceforth “background knowledge”), the question which this paper seeks to address concerns how background knowledge gives rise to appropriate behavior.

1. Following Damasio and Damasio (1994: 61), I take “knowledge” to refer to “records of interactions between the brain […] and entities and events external to it.”

0

Thanh Nyan

1.1 Assumptions While privileging such a level of analysis I am, nonetheless, committed to a set of assumptions concerning higher levels, which are, by and large, compatible with those of other contributors, in particular Fetzer (2004 and in this volume): 1. Appropriate behavior (linguistic or otherwise) is behavior judged to be in conformity to a set of social norms or conventions in a given situation. 2. This judgment of appropriateness is made by whoever constitutes – by agreement or default – the prevailing authority on the matter under consideration, at the relevant stage in the interactive (or communicative) process. Thus, at the production stage in a verbal interaction, the Speaker (S), as the source of the action under consideration, would be the only one in a position to make such a judgment, through her assessment of the current situation and her selection of the action being taken. Once this action is in the public domain this judgment may be seen to require validation, through implicit or explicit acceptance on the part of the Respondent (R). Consider: (1) S: Why did you tell John? R: Why not?

In asking her why-question, S is making a judgment, which, she appears to believe, is appropriate to the situation. This situation, as S construes it, is one in which R has violated a shared norm. In retorting “Why not?” R challenges the status of this norm (Kahneman & Miller 1986: 148). 3. How the norms or conventions being referred to are established, maintained and updated is through collective agreement and collective continued use (Searle 1995: 43–57). Thus, to quote Searle, “for such and such bits of paper” [X] to count as money [Y], “there has to be collective agreement, or at least acceptance, both in the imposition of that status on the stuff referred to by the term X and the function that goes with that status”. Furthermore, “we must continue to accept it as money or it will become worthless”. Alongside this default scenario proposed by Searle, another should be envisaged that involves negotiations (Fetzer 2000 and 2002). An example of negotiation (over the norm) is provided by (2), an extended version of (1), at the end of which R accepts S’s initial assumption of a shared norm: (2) S: R: S: R.

Why did you tell John? Why not? I said it must stay within these walls. Oh sorry, I forgot.

Appropriateness

4. However, unlike changes affecting common ground2 assumptions in verbal interactions, those concerning social conventions, would presumably have to be imposed from top down, at least in the first instance, as well as externally enforced (e.g., insisting that children use “Please” and “Thank you” and telling them off for failing to do so). 5. Inasmuch as continued use of social conventions is an expression of commitment to the existence of those conventions, any flouting constitutes a threat to social cohesion and puts perpetrators at risk of varying degrees of marginalization. Appropriate behavior is thus adaptive both at the collective and individual level. In certain communities the practice of arranged marriages is still very much alive, because seen as key to the social fabric. For that reason, women who fail to comply have much to fear at the hands of their own families. 6. Adaptive behavior is behavior that is advantageous to the survival and wellbeing of the animal and its kin. It relates to the need to avoid danger, to seek food, shelter and pleasure. Appropriate behavior is a category of adaptive behavior required for survival (in a broad sense) in social environments. In addition to the above I am also committed to a set of co-evolutionary assumptions, which includes: 7. Language and the brain are co-evolved (Deacon 1997). Under the traditional Chomskyan paradigm, the remarkable rate at which children learn their first language is accounted for by postulating a language organ and innate linguistic knowledge. By contrast, Deacon explains this feat in terms of co-evolution of language and the brain. In his view (Deacon 1997: 105–115), the structure of language is to the brains of children what a virus is to its host, on which it depends for its reproduction. Over time this relationship evolves into a form of symbiosis for the mutual benefit of both parties, with the emergence of two types of adaptation: human adaptations to language, which ensure that “language is successfully replicated and passed from host to host”; and “language adaptations to children […] whose purpose is to make language particularly infective as early as possible in human development.” Thus, the difference in size between the dorsal forebrain and the ventral forebrain in humans led to a shift in connectional patterns, “which is responsible for two crucial features of human adaptations, the ability to speak and the ability to learn symbolic associations”. The former can be traced back to “changes in motor projections of the midbrain and brain stem”, the latter, to “the expansion of the prefrontal

2. Fetzer (2007).

1

2

Thanh Nyan

cortical region and the pre-eminence of its projections in competition for synapses throughout the brain” (Deacon 1997: 220).This prefrontal dominance led to a predisposition in our species to favor one learning strategy over competing tendencies, namely the capacity for divergent thinking. This capacity, which already underlies goal-oriented activities (such as organizing behavior sequences with a view to reaching a certain goal), comes to mediate the shift from an indexical to a symbolic mode of reference. 8. On a co-evolutionary view, language is rooted in our neurobiology and shares processing systems with other forms of cognition. So contrary to what was once proposed by Fodor (1983), the language processing system would be subject to the principle of weak modularity, whereby subsystems share components at a finer level of analysis (Kosslyn & Koenig 1995: 45). 9. Evolution hardly ever devises new solutions, especially from scratch, but tends to rely on solutions that have proved successful in the past (Damasio 1994: 190). The shift from an indexical to a symbolic mode of reference (mentioned above), which helps alleviate the processing load due to the increasing demands of communication, would be the result of the co-option of divergent thinking by language. The types of processing problem being addressed by this shift are very similar to those underlying tasks such as planning. What divergent thinking brings to bear in both cases is the capacity to suppress a response elicited by current stimuli and select, instead, a response appropriate to a (more) distant goal. 10. Argumentative discourse, like higher-order consciousness – or the capacity to be “conscious of being conscious” (Edelman 1989: 186–192) – is goal oriented, and this goal, as in the case of higher-order consciousness, is one of ensuring survival (see also Koch 2004: 105). The possession of a higherorder consciousness enables the animal to have an internal representation that is free from the influence of ongoing events, in the sense that it allows it to review its present perceptions in terms of past experiences and plan the sequence of actions required to carry out goals related to its survival. The possession of language enables it not only to provide an external representation of its plans, but to communicate its goals to others and, through argumentative discourse, involve them in its planning (Nyan 2007). The proposed parallel between argumentative discourse and higher-order consciousness is based on Edelman’s view on higher-order consciousness and language, which includes among its relevant assumptions: – As a capacity to construct an internal picture in which past, present and future are connected to an adaptive goal, higher-order consciousness entails a long-term memory, the emergence of which requires the evolution of symbolic structures.

Appropriateness

– The emergence of both higher-order consciousness and language requires, among other things, the pre-existence of brain areas devoted to concept formation. – Speech allows the capacity for internal modeling inherent in higherorder consciousness to be even freer from the influence of current perceptions. Without a higher-order consciousness an animal replete with food is not likely to set out in search of its next meal, especially if a warm sunshine beckons. One with a higher-order consciousness will remember the consequences of past failures to plan ahead and act accordingly. Now, this same animal, if also endowed with language, can do even better: it can attempt to convince another to help it secure this next meal. In light of this latter set of assumptions, linking appropriate behavior with survival presents some distinct advantages from a methodological standpoint: a. It legitimizes investigating appropriate behavior at a lower level of analysis than previously, thereby creating a possibility for grounding in respect of certain aspects of appropriateness. b. It provides a rationale for selecting candidates for structures and processes sub-serving appropriate behavior.

1.2 Processing system as vantage point In considering appropriate behavior from a lower-level perspective, I am taking the processing system, rather than the co-participants, as a vantage point. Now a major characteristic of processing systems is that they are subject to the requirement of processing efficiency; and processing efficiency is normally achieved through a high level of automatization.3 This entails using as much as possible established shortcuts between input and output points. In the way of shortcut, there appears to be, on the part of perceptual systems, a tendency to anticipate a perceptual outcome rather than wait for all relevant evidence to become available before computing it. Failure to make a split-second decision may prove fatal in the case of predator identification. This outcome is subsequently revised if found to be at odds with the evidence. “Anticipation-

3. Automatic processing is not accessible to consciousness. This entails that from the standpoint of the processing system there are no such things as judgements of appropriateness: using language appropriately is chiefly a matter of activating the relevant skills (Searle 1995: 141– 147). Judgements of appropriateness belong to a later stage, when co-participants become consciously involved.

3

4

Thanh Nyan

and-revision-on-the-fly” (Kosslyn & Koenig 1995: 240), as this strategy is known, would also apply to language processing, inasmuch as the latter relies on perceptual systems. From the perspective of the processing system, using language appropriately would thus be a matter of activating relevant skills, at least in the first instance. Due consideration of contextual cues, which include feedback on the part of co-participants, only intervenes at a later stage, and rational decision making would not normally intervene until then.

1.3 Proposal The immediate question this strategy prompts is: what must be in place for the anticipatory phase to be carried out? In virtue of the co-evolutionary assumptions set out earlier, I propose that the answer lies – among other things – in an encoded source of knowledge similar4 to the one required for action and put forward by Damasio (1994). Such background knowledge would have to present a certain categorical5 structure and be underpinned by a recategorical memory system, which would be part of an account of how appropriate behavior – or behavior regarded as such by the agent – arises.

1.4 Organization If appropriate behavior is a type of adaptive behavior, the first question that needs to be addressed concerns the structure of the background knowledge from which this behavior arises. This structure – I argue – should not only be categorical, but consist of complex categories, as based on the model of decision making proposed by Damasio, the function of which is to facilitate the selection of adaptive action. Next I examine further aspects of this background knowledge that make it adaptive. This involves looking into category formation, in particular with regard to the criteria that influence it, category alteration, and the corresponding underlying supporting system. On the basis of this decision-making apparatus, which includes a procedural memory system, I then provide a sketch of how appropriate behavior arises. My next concern is to show the plausibility of viewing appropriate linguistic behavior (as drawn from the social domain) in terms of the same apparatus, especially with regard to background knowledge. The final section 4. Such a source cannot be identical to the repository of knowledge required for action, as linguistic conventions would need to be taken into consideration. 5. That background knowledge should have a categorical structure as such is by no means a new notion (see Rosch 1978 and Kahneman & Miller 1986, among others).

Appropriateness

summarizes what the proposed notion of appropriateness entails and considers the implications of an adaptive perspective for the notion of context.

2.

Background knowledge

In order to survive an organism requires vast repertoires of knowledge, including knowledge about how to avoid danger, find food, shelter and sex. How this knowledge is acquired and updated is through learning, which involves – among other things – an ability to sense the environment, which is what perception is about. Leaving aside the details of what perception consists of, of more immediate interest to us is the fact that perceiving is a matter of assimilating, where possible, the object of perception to a category (Kosslyn & Koenig 1995: 348–349). Thus, in seeking to identify a stimulus, the processing system, as a matter of course, gives priority to a categorization process corresponding to an attempt to subsume this stimulus under categories of similar objects encoded in memory. The process whereby the stimulus is encoded as novel memory only gets triggered if this matching process fails.

2.1 Requirements This view of perception presupposes a source of knowledge – background knowledge – with a categorical structure. However in order to be adaptive, this background knowledge must: a. present a categorical structure that is maximally effective at boosting efficiency, that is, at yielding a swift output; b. arise from a category formation process which is: – governed by innate values required for survival; – sensitive to environmental changes that alter the conditions for survival. A final, but this time, overarching requirement is that the model envisaged should be compatible with Edelman’s Theory of Neuronal Group Selection (henceforth TNGS)6 (Edelman 1989: 57, 94–146). 6. In particular with respect to processes underlying perceptual categorization and concept formation. Inasmuch as Damasio’s model of brain function is in many respects similar to Edelman’s, especially with regard to a feedback process which Edelman calls “reentry”, I will also resort to descriptions found in Damasio (1994) and Damasio and Damasio (1994), if they involve fewer details or seem clearer than those provided by Edelman. Apart from Damasio’s apparatus of decision making, which has no explicit equivalent in Edelman, the differences presented

5

6

Thanh Nyan

In a nutshell, the TNGS was prompted by two main observations: 1. The world is unlabeled until it becomes categorized by organisms endowed with adaptive criteria. 2. Even within the same species, the nervous system of animals presents a significant amount of individual variations. These observations, among others, led to the hypothesis (“neural Darwinism”) that in adapting to an uncategorized world, the brain would act “by a process of selection upon variance” (Edelman 1989: 40–41). In viewing the brain as a selective system, the TNGS constitutes a rejection of information models of interactions between the world and the nervous system. Such models take a physical input and process it through their sequence of causal relation to produce a physical output. They do not regard the level of variability of the world and of nervous systems as significant. To explain the brain’s ability for perceptual categorization, generalization and memory, the TNGS postulates three basic mechanisms: developmental selection, experiential selection and re-entrant mapping (Edelman 1989: 45–72).

2.2 Categorical structure Efficiency, as will be recalled, requires a high level of automatization on the part of the process whereby background knowledge is accessed. In terms of structure, this translates into shortcuts between input and output points – in the event, a given situation and the appropriate response to it, or rather the categories that subsume them. From the standpoint of the user, this entails that conscious decision making and rule following must give way to an ability to activate corresponding patterns of behavior on the basis of given situations. Searle’s example of the baseball player (1995: 141–147) provides an illustration of this latter point – albeit at a higher level of description: at the beginning the baseball player has to learn a set of rules, but after he gets skilled, his behavior becomes much more fluent […], more responsive to the demands of the situation. In such a case […] he is not applying

by the two models are not relevant to our purposes, and can be quickly summarized along the lines suggested by Damasio and Damasio (1994: 74): Damasio’s model, but not the TNGS, uses a “convergence-divergence architecture. The maps [of neurons] are fully and reciprocally interconnected, in both a hierarchical and heterarchical manner.” Furthermore Edelman’s reentry uses the process of “synthesis of signals” as its main means of operation, while Damasio’s framework uses a “correlative operation” instead.

Appropriateness

the rules more skillfully, he has acquired a set of dispositions or skills to respond appropriately, where the appropriateness is actually determined by the structure of the rules, strategies and principles of baseball […] One develops skills and abilities that are, so to speak, functionally equivalent to the system of rules without actually containing any representations or internalization of these rules.

The question that arises is: how is this behavior caused? This is where Searle comes up with his notion of background causation. But first, what is meant by “Background” (or “the Background”)? “Background”, as used by Searle (1979, 1983), refers to a network of enabling conditions7 that makes it possible for particular forms of intentionality – such as meaning or action – to function. In other words, it is a set of “non-representational”8 capacities (or kinds of know-how about “how things are and how to do things”), which include practices, skills, abilities and stances that human beings possess, both in virtue of their biological make-up and of sharing the same physical and socio-cultural environment. This list was later extended to include “dispositions, tendencies and causal [neurophysiological] structures generally” (Searle 1995: 129–130). Searle sees the Background as made up of layers, ranging from “Deep Background”, which corresponds to forms of know-how common to the species, to “Superficial Background”, which is socio-culturally derived. Skills arising from Deep Background include an ability to interact with tables which takes account of their solidity or an ability to handle objects which presupposes the existence of a gravitational field. Skills pertaining to Superficial Background range from the ability to use light switches to the ability to use debit cards or get a divorce.

7. Between forms of intentionality such as perception, intentional states and action (which includes performing and interpreting speech acts), which are intentional and the Background, which is pre-intentional, Searle postulates a network of intentional states, the Network. To use an example given by Searle (1983: 141), a man “cannot form the intention to run for the Presidency of the United States without having certain beliefs (e.g., the United States is a republic […] That the candidate of two parties vies for the nomination of his party […], that voters vote for him…” Furthermore, he has also to take for granted that “elections are held at or near the surface of the earth”, in the sense that he would be very surprised if that were not the case, rather than in the sense of having such a belief. The Network shades into a “bedrock of mental capacities”, which is the Background. . What is meant for those capacities to be “non-representational” is that knowing how to activate them in performing skills does not involve having them as objects of representation (or intentional objects) of a mental state such as belief. “For the hardness of tables manifests itself in the fact that I know how to sit at a table, I can write on a table […] And as I do each of these things I do not, in addition, think unconsciously to myself ‘it offers resistance to touch’ ” (Searle 1983: 142–143).

7



Thanh Nyan

The notion of background causation was introduced to explain how background abilities enable us to behave in a way that conforms to institutions (Searle 1995: 139–142). Of two general models of causation, one intentionalistic and rationalistic (involving rule following), and the other based on physical brute force, Searle opts for a version of the latter, one in sympathy with the connectionist model of cognition. Searle’s background causation offers a view of appropriate behavior that is compatible with the requirement of processing efficiency (if lacking in details with respect to the nature of the processes involved and supporting structures). Furthermore, his construal of skills – as arising from the activation of non-representational background assumptions – is also compatible with a widely accepted view of skills, according to which they arise from implicit memories. Such memories bear a clear resemblance to background assumptions: they are not available for introspection or other purposes9 and can only be activated in the performance of attendant skills (Kosslyn & Koenig 1995: 375–384). However, having a connectionist model only goes so far towards solving the problem at hand.10 While such a model provides a plausible alternative to rational decision making, which takes too much time to be adaptive, an outcome still needs to be reached, which presupposes a strategy for choosing among the multiplicity of connections involved. In short, how does the appropriate or most advantageous course of action get selected? As soon as one seeks to extend this model beyond the social domain, the number of alternatives also increases dramatically. This problem requires a theory that can provide a mechanism whereby this non-rationalistic type of decision making can be implemented. This is where Damasio’s apparatus of decision making comes in, with its specific categorical structure, and its selectional process based on somatic markers. As part of his neurobiologically rooted model of decision making, Damasio also postulates a repository of background knowledge (or “factual knowledge”11), which covers a similar range of abilities. However, there is a crucial difference 9. Those purposes include providing the basis for reasoning (Kosslyn & Koenig 1995: 374). Searle (personal communication) did not derive his conception of the Background from neurophysiology but from Wittgenstein. 10. To be fair, in his theory of Intentionality, which subsumes his theory of Background, Searle is less concerned with the process whereby skills are activated than with the countless number of background assumptions they presuppose. 11. As a broad base knowledge, this repository of “factual knowledge” includes “facts about objects, persons and situations in the external world [and] facts and mechanisms concerning the regulation of the organism as a whole” (Damasio 1994: 83–84).

Appropriateness

between this factual knowledge and Searle’s Background: in Damasio’s version the categories are composite, in the sense that they consist in pairings of categories of situations and categories of response options, which in turn are linked to categories of consequences. Thus, supposing you have come down with a heavy cold and are wondering whether to go out for your usual run; if you have encountered this situation before, you will be able to base your decision on a complex category (my expression) presenting a set of initial situations (under which the one at hand can be subsumed), a set of response options (e.g., staying indoors and keeping warm or going out for a good work-out), each of which connected to consequences you have experience of. With the above type of categorical structure, the problem presented by a nonrationalistic and connectionist model such as Searle’s becomes more manageable. However there is still some way to go: having a direct link to a subset of possible options still does not quite enable the system to zero in on a solution, only one of which must be selected.

2.3 Selecting criteria How a shortlist of options is to be derived from a set of possible options is, according to Damasio (1994: 173–198) through the intervention of somatic markers. Somatic markers may be thought of as negative or positive body states (e.g., pain or pleasure) that become attached to consequences of actions. By thus highlighting options as advantageous or deleterious they narrow down the range of alternatives requiring consideration. Somatic markers involve two alternative mechanisms, one in which the somatic states are experienced, the other, where the body is bypassed (Damasio 1994: 184). Thus, people used to jay-walking across roads with impunity develop an ability to do so without experiencing any somatic state, in the event, apprehension. The shortlist to come out of this process serves as input to the evaluation process. Contingencies that might affect the degree of advantage would be taken into consideration at this stage. To summarize: the type of background knowledge likely to sub-serve adaptive behavior is one structured in a way that facilitates a decision-making process with as high a level of automatization as possible. On Damasio’s construal, this background knowledge consists of complex categories that link situations calling for action to possible response options, which, in turn, are linked to consequences somatically marked for easy selection.

9

90

Thanh Nyan

2.4 Values As Koch (2004: 233) puts it, “In a dynamic environment populated by predators […] it is usually better to come to some conclusion rapidly and act, rather than to take too long to find the best solution”. This said, a quick response in itself may not be of much use, unless it is based on a proven set of responses, one in conformity to adaptive values. The question that arises is two-pronged: what determines these values? And how do they influence behavior?

2.4.1 Values, wired-in and acquired behavior Adaptive values, which are aimed at ensuring survival – through avoidance of pain, and satisfaction of hedonic, consummatory and appetitive needs – stem from the genome (Edelman 1989: 93). Extrapolating from Damasio’s view (1994: 179) that “the internal preference [or value] system is probably pretuned for achieving these goals in social situations’, one may hypothesize that the goals they embody are carried over to non-biological domains. In other words, the need for survival applies to all domains, whether biological, physical, personal or social. What differs from one domain (or environment) to another is the form adaptive behavior takes, which is determined by biological regulations, physical laws, norms or conventions, as the case may be. Taking adaptive action at the biological level is a matter of maintaining possible states of the “internal milieu” [or environment within an organism] and of the viscera within narrow ranges, irrespective of the external environment (Damasio 1999: 134–141). At the physical level, using the stairs to leave a building under normal circumstances is adaptive, while stepping off a window ledge is not. In the personal domain, putting an end to an abusive relationship is adaptive; attempting to take it in one’s stride is not likely to be. In the social sphere, observing the law is usually adaptive, using a human target for archery practice is not. An importance difference between biological and non-biological domains is that adaptive behavior is wired-in in the former and acquired in the latter. At the biological level, where adaptive actions include physiological reactions triggered by hunger or fear, possible states of the internal milieu are maintained within narrow ranges by means of devices whereby any internal imbalance can be sensed. These consist in a “stock of dispositions to respond accordingly to what is sensed and some means to carry out the actions selected as response” (Damasio 1999: 134–141). In more external environments, complex organisms require further stocks of dispositions, ranging from Searle’s Deep Background abilities, which enable them to interact safely with their physical environment to skills that shape their daily

Appropriateness

life, such as riding a bicycle or baking bread. How these dispositions are acquired is through experience. Koch (2004: 193) refers to this as “operant or instrumental learning”. Thus situations arise that elicit responses on the part of the animal, the consequences of which are positive or negative. Experiences such as these provide the basis for complex categories which are somatically tagged (see 2.3), and on which the animal can draw when it next encounters the same type of situation. In the case of complex skills, such as driving, knowing what type of response is required may not be enough: the animal also needs to learn how to carry out the selected response, which it can only do through practice. The mediating structures underlying this procedural learning include the “sensory-motor cortex, the striatum and related ganglia structures and the cerebellum” (Koch 2004: 194). N.B. In non-biological domains, where the principles that define adaptive behavior are not innate, one must differentiate between learning what these principles are and learning the corresponding skills. Learning what these principles are (through peer, parental or societal pressure) will give the individual a form of knowledge (know-that) which is encoded in an explicit or declarative memory and can be referred to in producing adaptive behavior. Behaviors produced under such conditions, however, involve a conscious effort and are likely to lack efficiency. One need only think of how laborious it can be to apply a newly acquired grammar rule. On the other hand, if repeated applications of these principles give rise to the corresponding skills, the individual in possession of this know-how – held in an implicit memory – will be able to act in a way that is sensitive to those norms, without having to refer to them, and to do so more skillfully (see Searle’s baseball player under 2.2). Before turning to category formation, which has a key role to play in the process whereby innate values influence acquired adaptive behavior, it would be helpful, for clarification purposes, to have an overview of what we have seen so far of this process. Innate values, as we recall, are aimed at survival. Inasmuch as survival happens to coincide with the avoidance of pain and the pursuit of pleasure (Damasio 1994: 179), the somatic states, which, in our experience, become linked with entities, events and actions, will act as signposts to inhibit or enhance certain courses of action. In their capacity as somatic markers, these states are attached to outcomes of responses that are part of complex categories. The possession of such complex categories enables the animal to act in an adaptive manner. Indeed they provide: – categories of initial situations under which situations encountered can be subsumed;

91

92

Thanh Nyan

– an automatic link to categories of responses; – the means to choose among these responses, in the form of somatic markers. To elaborate further on this last point: the mere contemplation of any given response will trigger an explicit imagery related to the outcome attached to it; this imagery, in turn, causes a pleasant or unpleasant somatic state to be re-enacted; the effect of which is to encourage or inhibit the action under consideration (Damasio 1994: 187–188). Note that once a habit has been established somatic states can be bypassed: the imagery generated will directly affect the neural circuits located in the brain core, which regulate appetitive behaviors. At the neural level, learning adaptive behavior depends on the prefrontal cortices, which receive “signals about existing and incoming factual knowledge related to the external world; about innate biological regulatory preferences; and about previous and current body state as continuously modified by that knowledge and those preferences” (Damasio 1994: 181). The skills that result from this convergence of signals are held in dispositional representations, which correspond to “potential patterns of activity in small ensembles of neurons” (Damasio 1994: 102–105). We shall see shortly how their activation leads to action and in what relation they stand to complex categories. In terms of the TNGS (Edelman 1989: 56–57 and 112), which tends to focus on a yet lower-level of analysis, learning results from “alterations of synaptic strengths” of linkages between the relevant systems.

2.4.2 Values and category formation We have just seen how innate values, via somatic states, influence the choice of adaptive responses. Our next concern is how these same values influence the categorization of the external environment. Following Edelman, I take the view that perception is not value free. If it were value free, the categories an animal internalizes for a certain predator would include salient features inherent in that predator, none of which may have any relevance to its survival. Thus, a cat afflicted with a value-free perceptual system, whose experience of dogs is confined to Dalmatians, may find herself in peril of her life for associating dogs with a white coat and black spots. According to Edelman (1989: 98–99), innate values, determine what features the animal sees as salient.12 Thus, non-accidental sensory attributes of preda12. These features may or may not coincide with those that are inherently salient in the object.

Appropriateness

tors – that is, those that “reliably signal [their] presence […] in a wide range of conditions” (Kosslyn & Koenig 1995: 73) – will be perceived as salient. These features then go on to provide criteria that define what situations (including escape means) are significant with regard to survival. Value-dominated perception Edelman aptly defines as “the adaptive discrimination of an object or event from background or other objects and events” (Edelman 1989: 49). In terms of a lower-level of analysis, values affect category formation through the operation of a special memory system “capable of relating value and perceptual category by conceptual means”13 (Edelman 1989: 93–102). “Value is generated by the homeostatic14 adaptive brain system. Perceptual categories are generated by the sensorymotor systems (mainly cortical) that are devoted to categorization through motor behavior and sampling of external stimuli.” Between these systems there is a “process of temporally ongoing parallel signaling” (or “reentry”), which enables past value-category matchings to interact with current value-free perceptual categorization, leading to an alteration of the relative saliency of external objects. This, in turn, helps the animal select actions appropriate to its goals (as determined by its value-dominated memory). In short, values influence category formation by causing the animal to see features that are relevant to its survival as salient, thereby marking them out as candidates for membership criteria. Both categories of initial situations calling for action and categories of response options, which make up complex categories, would arise from those criteria. Though imparting saliency and somatic marking are distinct ways of bringing to bear innate values on action, they both involve attention allocation to what is likely to be adaptive. In the perceptual case, attention is directed to features that otherwise would not normally be salient, unless they are inherently so. In the somatic marking case, the impulse to act is enhanced if the response being considered has advantageous consequences. By way of introducing the next point, which is about recategorization, note that the type of action the animal ends up selecting depends ultimately on what is available in the actual context that can be categorized as a possible option. Thus, if a certain type of predator appears on the scene and the optimal means of escape is not available, the animal will have to go for the next best thing, which may not be 13. “A necessary basis for the special memory repertoires involves portions of the brain capable of concept formation, the ability to distinguish objects from actions, and categories from relations” (Edelman 1989: 95–96). 14. This “biological value system” (Damasio 1994: 181) includes “the hypothalamus, pituitary, various portions of the brain stem, amygdala, hippocampus and limbic system” (Edelman 1989: 94).

93

94

Thanh Nyan

in its repertoire. In other words, it appears that the perceptual system would have to be able to identify this novel contextual feature as a possibility, which it cannot do, unless the categories involved are susceptible to change. This brings us to another sense in which complex categories would have to be adaptive: given that unpredictable environmental factors may affect survival, failure to respond to them adequately may prove fatal. The animal has to be able to develop new responses, which requires that its complex categories be susceptible to refinement or alteration.

2.4.3 Recategorization In the TNGS, categories can be altered to varying degrees under the influence of ongoing changes in the environment. We have seen how values impact on category formation by making the animal perceive certain features as more salient; and how those features to which saliency has been imparted then go on to contribute to membership criteria of categories. Now, supposing that the animal finds itself in a situation where the environmental features do not bear a close resemblance to the relevant membership criteria, repeated encounters with such a situation will tell the animal that action is called for, and learning will take place (where “learning” is taken to mean “acquired context-dependent behavioral change” (Edelman 1989: 93)). Underlying this type of learning is a capacity for generalization, or a capacity for treating within any given context “a more or less diverse collection of […] entities as equivalent” (Edelman 1989: 49). This capacity, in turn, is based on a property of neural groups known as “degeneracy”, whereby “more than one combination of neuronal groups can yield a particular output and a given single group can participate in more than one kind of signaling function” (Edelman 1989: 50). This ability to generalize enables the animal to “recognize a great number of related but novel instances” (Edelman 1989: 110). Repeated exposure to these novel instances results, at the neuronal level, in changes in synaptic efficacy (Damasio 1994: 104), which provides the basis for refinement or alteration of encoded categories. In terms of membership criteria, this would lead to a readjustment. Inasmuch as learning is a capacity for acquiring context-dependent behavior aiming to satisfy innate values, it will also be mediated by a synaptic linkage (Edelman 1989: 56 and 112) made of re-entrant connections between the valuefree perceptual system that categorizes the external world, the value-dominated memory (corresponding to background knowledge), which holds complex categories, and the homeostatic adaptive brain system. Within this mode, a personal or social type of know-how is acquired and updated in the same way, that is, through learning which type of behavior is advantageous in a given context, and which is not.

Appropriateness

To this view of categorization as a continual process corresponds a dynamic notion of memory, one that is recategorical and procedural. In sharp contrast with a notion of memory as “a strict storage of coded information” (Edelman 1989: 117) that can be replicated upon recall, a recategorical memory not only is subject to change, but can provide access to the same outputs through different paths.

2.5 From perceptual to conceptual categories Under 2.4.2 a distinction was introduced between value-free categories, based on perceptual categorization, and value-dominated categories resulting from the influence of adaptive values. For further reference, this distinction should also be expressed in terms of contents. Thus perceptual categories invariably have a high sensory content. By contrast, value-dominated categories include both categories with a high sensory content, based on features provided by the perceptual system, and categories with a low sensory content (past value-dominated categories), which involve less immediate reference to objects and events. The latter, also known as “conceptual categories” or “concepts” in the TNGS, reflect generalizations arising from “various mixtures of relations concerning the real world, memories and past behavior” (Edelman 1989: 143). To account for concept formation Edelman (1989: 143–145) hypothesizes that what is required is, in the first instance, a function that involves large parts of the brain. Such a function is carried out by “dynamic patterns of activity” – or “global mappings” – which can correspond to objects, movements or a relation between different types of objects. Furthermore, inasmuch as the patterns of activity in different kinds of global15 mappings must be recombined, compared and categorized (e.g., to connect movement categories to spatial references, or distinguish global mappings corresponding to objects from those relating to actions), “back reentrant connections” are required from “the frontal cortex to other cortical regions and to the basal ganglia and hippocampus regions.” As a result of the activity allowed by these connections the cortical areas involved can give rise to “classifications of global mappings”, which provide the basis for action. For future reference, it is important to note that in the TNGS (Edelman 1989: 173–181): 15. A global mapping is a “dynamic structure containing multiple re-entrant local maps (both motor and sensory) that interact with nonmapped regions such as those of the brain stem, basal ganglia, hippocampus and parts of the cerebellum” (Edelman 1989: 54). Maps are sheets of neurons in certain areas of the brain that are related to receptor sheets (e.g., the surface of the skin).

95

96

Thanh Nyan

a. the capacity for concept formation is prior16 to the emergence of language, which is viewed as an epigenetic phenomenon; b. the existence of a conceptual memory prior to language does not preclude the emergence of linguistic memories in which symbols for concepts (e.g., morphemes, words and sentences) can be subject to categorization and recombination without further reference to their pre-linguistic conceptual basis (Edelman 1989: 174).

3.

Memory

In attempting to understand what underlies the production of adaptive behavior, we have, so far, focused mainly on the categorical structure of acquired background knowledge and how it came to be the way it is. Thus, we have seen: a. that background knowledge is made up of complex categories, which function as templates, so to speak, for adaptive action; b. that the formation of these categories is influenced by innate values; c. in what way external environments determine the specificity of what constitutes adaptive behavior in those environments and provide the sources for alterations in existing categories; d. how these operations are mediated by a system of re-entrant connections, which enables signaling to go back and forth between the systems involved. Inherent in what precedes is the idea that these categories are held somewhere, no matter how temporarily. We have just seen how reentry gives rise to a dynamic conception of memory. As our next step towards achieving a better understanding of adaptive behavior, we need to clarify what this conception of memory entails.

3.1 Dispositional representations According to Damasio (1994: 102–105) both innate and acquired knowledge is held in dispositional representations, which correspond to “potential patterns of neuron activity in small ensembles of neurons”. Dispositional representations for acquired knowledge are distributed all over the higher-order cortices. The way they are believed to arise is through perception, which gives rise, at least in the

16. Also prior to language is a capability for ordering concepts, or “presyntax” (Edelman 1989: 147–148).

Appropriateness

case of complex sensory information, to “topographically mapped representations” in the early cortices. When activated, dispositional representations can do one of three things: a. fire other dispositional representations with which they have strong connections; b. activate a topographically mapped representation, either directly, by firing back to early sensory cortices or, indirectly, by activating other dispositional representations; c. generate a movement by activating “a motor cortex or nucleus such as the basal ganglia”. Some dispositional representations contain knowledge that is imageable, and, for that reason, can be involved in a range of activities such as the generation of movements or planning. Others contain records of procedures for the handling of these images (Damasio 1994: 105). Imageable17 knowledge is reconstituted when topographically mapped representations are activated. This also applies to knowledge of words, which, when activated, gives rise to auditory and visual imagery. These two types of dispositional representations correspond to the two types of memory mentioned earlier, explicit (or declarative memory) and implicit (or procedural memory). For a discussion of why this model of knowledge representation – which implies a relative compartimentalization of the brain – is to be preferred over the “classic” alternative, see Damasio and Damasio (1994).

3.2 Complex categories The relationship between dispositional representations and topographically mapped representations also helps address certain questions that arise in connection with complex categories: – Are categories of initial situations and categories of response options to be found in the same location? – In the event that they are not, how is the linkage between them to be construed? – In view of the fact that they are acquired via perceptual categorization, what becomes of these perceptual categories once complex categories have become established to provide the basis for attendant skills? 17. According to Damasio (1994: 106), “both words and arbitrary symbols are based on topographically organized representations and can become images. Most of the words we use in our inner speech, before speaking or writing a sentence, exist as auditory or visual images”.

97

9

Thanh Nyan

In light of what was said under 3.1, complex categories would be contained in dispositional representations with strong connections between them, that is, pertaining to the same circuit. If the category of initial situations belongs, for example, to the visual modality, activation of this category would involve firing a topographically mapped representation. When the category of response options in turn gets activated, through the firing of dispositional representations in the same circuit, either the early cortices get involved and a topographically mapped representation is generated, or they are not, as is the case when a motor response is called for. In short, both at the level of dispositional representations and at the lower level, complex categories would be, strictly speaking, distributed over various memories. Each complex category corresponds to strengthened patterns of activity within the same circuit of dispositional representations, which hold the potentiality for reconstructing types of situation and types of response. In the course of the acquisition process perceptual categories would give rise to topographically mapped representations, on their way to creating dispositional representations. Topographically mapped representations would become part of a declarative memory, while dispositional representations would provide the basis for a procedural memory. When we learn how to use a computer, we begin by encoding the various steps required to log in in a declarative memory, from where they can be consciously retrieved. Over time, practice enables us to acquire the corresponding skills, which are held in a procedural memory. Those skills cannot be consciously retrieved, but can only be activated as part of the process of logging in. This may explain why highly skilled IT people tend not to be very clear when asked to take one through a step by step procedure.

3.3 How appropriate behavior arises from procedural memory Based on what precedes, this is how appropriate behavior arises from procedural memory. The situation of reference presents certain features in virtue of which it is subsumable under a category of initial situations, which is linked to a category of response options. The process whereby the situation of reference is perceived under the features in question, or “recognized” as a member of a certain category of initial situations, involves the activation of a topographically mapped representation in the early sensory cortices. This in turn fires the relevant dispositional representations, those holding categories of initial situations. These dispositional representations, in turn, fire other dispositional representations to which they are related. This second set of dispositional representations gives rise to a topographically mapped representation corresponding to a category of response options. Alternatively, the first set of dispositional representations would cause a set of

Appropriateness

motor responses to be directly generated. The next phase, that of the evaluation and selection of the appropriate response, is mediated by somatic markers. By tagging the consequences of actions as advantageous or deleterious, these markers enhance or inhibit the actions under consideration. In place of the somatic marking apparatus, one can also find the “as if ” system, which presents an even higher level of automatization and achieves the same result without any somatic states being experienced by the subject. The same basic model would apply to the production of appropriate linguistic behavior – albeit with additional special memory systems related to words and sentences at the phonological, syntactic and semantic levels (Edelman 1989: 177– 178), all of which are recategorical, and interact in multiple ways via reentry. While a detailed picture would have to await further research, this is a sketch of what the process in question could look like. Such a process starts off with perceptual categorization, but of speech, this time. Next, topographically mapped representations are activated in the early sensory cortices (e.g., the auditory cortices). This, in turn leads to dispositional representations in higher-order cortices to be fired. Among the various memories in which these dispositional representations are held, one must now include the Broca’s and Wernicke’s areas, which cater for the “means of production and recognition of coarticulated speech sounds”. The activation of these dispositional representations and of others, to which they are connected, gives rise to complex categories. Some of these complex categories would also be linguistic, in the sense that their sets of initial situations and response options would correspond to linguistic structures. Inasmuch as speech production is the outcome of this process, this would be a case where response options require the activation of the motor and premotor cortex. A note of caution is in order: the above sequential presentation may be misleading, as the reentrant connections between the various systems allow for a fair amount of parallel activities.

4.

Attempting to bridge the gap

In an attempt to bridge the gap between non-linguistic and linguistic appropriate behavior, I propose to set out in what way the latter can be accounted for in terms of the same decision-making apparatus. Inasmuch as I will be focusing on the connection between linguistic behavior and background knowledge, it would be helpful to have the following assumptions to hand. – This background knowledge is held in a non-declarative value-dominated conceptual memory, which is recategorical, in the sense that the categories

99

100 Thanh Nyan

it holds can be updated in light of new experiences. (This memory, which is pre-linguistic to begin with, can be recategorized under the influence of the linguistic memories. For that reason, “non-linguistic conceptual memory” will be preferred to “pre-linguistic conceptual memory” in a number of instances). – The emergence of language is made possible by that of linguistic memories, which are also recategorical, with the objects of this recategorization being the productions of language itself, or rather the corresponding categories (Edelman 1989: 174, 176–182).18 – These linguistic memories constitute another source of knowledge on which appropriate linguistic behaviors can draw. – The process whereby these memories are accessed involves categorization.

4.1 Data The data I have selected consist of contributions whose appropriateness with respect to a prior contribution is marked by linguistic constraints on continuation.19 The latter arise from the occurrence of a lexical item or an argumentative marker in the prior contribution. Consider the following examples adapted from Argumentation Theory: (3) John is well-off.

In the absence of contextual assumptions to the contrary, the use of ‘well-off ’ points to (a), rather than (b), as the preferred continuation: a. He can afford anything he fancies. b. He watches every penny he spends. 1. According to Edelman (1989: 179), the development of re-entrant means of linking category formation to an emerging special system of sound production, recognition and memory undoubtedly played a major role in the evolution of the hominid brain. In this language is not required for concept formation, which is prior. Nevertheless, the specialization and development of linguistic capabilities implies the development of new means of memory for concepts as concepts as well as of an internal mode for relating that memory to other memories concerned with imagery. Meaning per se is not carried in a single module and emerges only as a result of interactions at many levels. The same is true of grammar. 19. These constraints, as well as the upcoming ones were first identified by Argumentation Theory (see Anscombre 1995).

Appropriateness

(4) This editor is meticulous. (5) This editor is obsessive. (Both (4) and (5) are judgments passed on the same editor’s attention to details.)

The attitudinal component – positive in “meticulous”, and negative in “obsessive” – constitutes a strong constraint on what will or will not work as an appropriate continuation. (Strong constraints, unlike weak constraints, as in (3), cannot be easily overruled by contextual assumptions). Thus “meticulous” requires continuations such as (a), rather than (b), while the opposite is true of “obsessive”: a. You should have him assigned to your book. b. I would steer clear of him, if I were you.

A final type of constraint (on continuation) arises from argumentative markers such as mais, maintenant and toujours.20 Thus, the use of the above type of mais – to take but one example – imposes a constraint on the type of conclusion that can be drawn from (6): (6) Les plantes ont besoin d’être arrosées, mais il va pleuvoir. The plants need watering, but it is going to rain. (Where (6) is a response to a suggestion that S should water the plants.)

Thus in response to (6), (a), but not (b), would be appropriate: Donc tu ne vas pas le faire. So you are not going to do it. b. Donc tu vas le faire. So you are going to do it. a.

Three reasons underlie this choice: –

Appropriateness in the above examples being a matter of closely associated contents, it is immediately construable in terms of the linkage inherent in complex categories. This is crucial as it gives us a direct connection to the non-linguistic conceptual memory (for other types of appropriateness see Fetzer, this volume). – The linguistic constraints under consideration are attributable to recategorization at the level of the linguistic memories, in the sense that they would have arisen from the repeated use of one type of utterance in response to

20. See Nyan (2007).

101

102 Thanh Nyan

another, the contents of which are usually already associated in virtue of their linkage at the pre-linguistic conceptual level. – The existence of such constraints lends more credibility to our hypothesis concerning the underlying system. Before proceeding to show in what way the behaviors arising from these constraints can be accounted for in terms of the above apparatus, we need to go over the reasons for wanting to do so.

4.2 Assumptions Evolution, as will be recalled, seldom devises new solutions, especially from scratch, but tends to favor proven solutions in dealing with similar problems. Such solutions include pre-existing systems, supporting structures, but also strategies and capabilities. In virtue of this assumption, if finding appropriate continuations falls into the same category of problems as selecting adaptive behavior in general, then the former, as the latter, should be a matter for decision making. As it turns out, both cases are instances of what Damasio (1994: 198–199) calls “the problem of creating order out of parallel spatial displays”,21 which problem requires criteria whereby these possibilities may be ranked prior to undergoing a selection process. In other words, the solution appears to lie in a decision-making process. Argumentative discourse presents an obvious connection with decision making. Thus, there is an obvious parallel between realizing that the dog is overdue for his annual check-up and making an appointment for him to see the vet, and attempting – as in the following – to delegate the task to the co-owner of the dog, giving as a reason that the dog is overdue for his annual check-up: (7) The dog is overdue for his annual check-up (=P), you need to make an appointment for him to see the vet (=Q).

Both cases involve decision making: the former is about reaching the appropriate decision oneself (or internal modeling), the latter,22 about inducing a similar process in the respondent (R) (or external modeling). Further parallels can also be found, to which we now turn.

21. This problem is shared by thought and movement. 22. There is, in fact, this further problem of how to get this decision making represented in R’s system, and to secure her commitment to it, but it can be set aside as it is not relevant to the argument. For further details see Nyan (2007).

Appropriateness

4.3 Categorization of the situation of reference Argumentation, like decision making, starts off with a categorization of the situation of reference (SR). In uttering P (in (7)), S may be providing a representation of the SR, but she is also providing a category judgment of it, in the sense of presenting it under those features that highlight its membership to a certain category of situations C1. Why should there be a categorical in addition to a representational aspect? On a fairly general level, the use of conventional language commits the speaker to “a kind of semantic categorical imperative”’ (Searle 2001: 159). Thus, “when I say ‘That is a man,’ I am committed to the claim that any entity exactly like that in relevant respects is also correctly described as ‘a man’”. Similarly, when S says “The dog is overdue for his annual check-up” (P), she would be committed to the claim that any situation exactly like the one described in relevant respects is also correctly described as P. Searle’s generality requirement thus already allows for the fact that P functions both as a representation of the SR and the expression of a category judgment, whereby the SR is to be regarded as a member of a pre-existing category of situations C1. Now, granted that P has a categorical aspect, why should this aspect be privileged in this particular case? From the standpoint of adaptive action, what matters is not whether P is a true representation of the SR, but whether the SR, as represented in P (i.e., under certain aspects rather than others), is subsumable under a category of situations C1, for which there is a category of proven responses, C2. Privileging the categorical aspect also makes sense from an argumentative standpoint. S’s intention (in (7)) is to bring R to the view that the action described in Q is called for on the basis of the situation in P. This intention is not one of attempting to convince R that the situation described in P is the case. Furthermore, unless the argumentation in (7) is ad-hoc, it must be based on a shared principle, one that validates drawing the conclusion under consideration from the argument in P. Such a principle is construable in terms of a pre-existing complex category, [C1-C2]. In other words, advancing an argument (in P) in support of a conclusion (in Q) amounts to providing (in P) a category judgment of the SR, or better, to asserting that the SR, as represented, falls into a category of situation C1, which is linked to a category of responses C2, under which the action described in Q is subsumable. The significance of being able to view argumentative discourse in terms of categorization is that categorization provides a procedural link to memory, and hence background knowledge.

103

104 Thanh Nyan

Further parallels between argumentative discourse and decision-making are provided by the data introduced earlier, to which we now turn.

4.4 Constraints on continuation and complex categories Consider again: (3) John is well-off.

which, in the absence of contextual assumptions to the contrary, takes (a), rather than (b), as the preferred type of continuation: a. He can afford anything he fancies. b. He watches every penny he spends.

The parallel is one between the weak constraint that links the occurrence of “welloff ” to type (a) continuations and the linkage inherent in complex categories. The question this parallel prompts is: how would the above apparatus account for the emergence of such constraints? From living in a certain social environment, one is exposed to prevailing practices and norms, which, in the first instance, are stored in a declarative (or explicit) memory. Selecting appropriate action, at an early stage, entails conscious reference to the contents of that memory. Over time, through adaptive learning, one acquires a corresponding set of skills, which allows appropriate action to be selected virtually automatically. These skills are mediated by complex categories, which incorporate an internal linkage between certain types of situations and corresponding adaptive responses. In (3), we would already have a complex category and hence a direct link between wealth and the practices it allows at the pre-linguistic conceptual level. How the linguistic constraint between “well-off ” and continuations of type (a) arises would be through repeated associations over time of strings describing situations featuring the possession of wealth with situations describing practices associated with it. In other words, while the linkage inherent in the complex category at the level of pre-linguistic conceptual memory may be what biases the association of such strings in the first instance, the emergence of the linguistic constraint would be due to a recategorization of linguistic productions (in the event, of the strings under consideration) at the level of linguistic memories.

Appropriateness

4.5 Attitudinal components and somatic markers Attitudinal components present in lexical items such as “meticulous” and “obsessive” provide an obvious parallel with somatic markers. Consider again: (5) This editor is meticulous. (6) This editor is obsessive. a. You should have him assigned to you. b. I would steer clear of him, if I were you.

As will be recalled, the positive component in “meticulous” calls for continuations such as (a) rather than (b) as appropriate, while the opposite is true of “obsessive”. Inasmuch as the occurrence of “editor” activates a network of attendant practices and stances, and the attitudinal component in “meticulous”, or “obsessive”, has the effect of narrowing down the range of available options, attitudinal components appear to have a similar function to somatic markers (or somatic states, which, in the decision-making process, serve to highlight options as advantageous or deleterious). For this parallel to be more than fortuitous, one needs to be able to show the plausibility of either (a) or (b): a. Attitudinal components derive from somatic markers. b. The apparatus under consideration appears to be able to cater for them in another way.

4.5.1 Can attitudinal components derive from somatic markers? There are serious difficulties to be considered: – Somatic markers, one recalls, arise from individual experiences. Inasmuch as such experiences are subject to variation, individuals are not likely to develop the same somatic states in response to the same negative and positive consequences of the same category of actions. Two people who have suffered the consequences of an outstanding gas bill, and have learned from the experience, may still feel very differently about the unpleasantness that has lead to a change in behavior on their part. If individuals need not experience the same somatic states in the same situations, how can there be any shared somatic states that could make their way into a shared language? – But there is a further difficulty. While somatic markers are attached to consequences of responses, attitudinal components are part of representations of initial situations. To say of an editor that he is meticulous is not only to provide a description of the SR, but also to include in it a positive judgment

105

106 Thanh Nyan

of this SR, one that biases the choice of certain types of action. Is there a plausible way of explaining how attitudinal components – assuming that they do arise from somatic markers – could have “migrated” upstream to become attached to initial situations? Concerning the first point: people may not develop identical somatic states in response to the same situations. However, from sharing the same biological makeup and the same environment they can certainly develop the same types of somatic states, especially if the nature of the consequences of certain actions are socially determined. Furthermore, the same types of somatic markers can be sufficient to inhibit or facilitate a given type of behavior. Those who learn about the new penalty for using a mobile phone while driving, need only experience the same type of somatic state for this action to be flagged as negative. Whether they are absolutely terrified at the prospect of losing their driver’s license, or merely nervous about it, the end result can be the same. Thus, if people can share generic somatic states in response to the same types of situation, the bias these states provide in their capacity as somatic markers at the non-linguistic conceptual level is likely to influence which linguistic representations of situations they associate. This, in turn, gives rise to linguistic constraints, through recategorization at the level of linguistic memories. Still, this does not tell us how somatic markers would make their way into initial situations and lexical items, hence the need for an alternative scenario: whether or not individual speakers agree with prevailing social norms, from the moment they use a common language, which reflects those norms, they will be using a language that has encoded in it socially sanctified attitudes towards what is appropriate and what is not. How these attitudes make their way into the language is through continued collective linkages of representations of initial situations and appropriate or inappropriate actions (see Searle 1995: 70–71). Through continued collective associations of these representations, a constraint arises through recategorisation at the level of linguistic memories. This scenario obviates the need for attitudinal components to derive from somatic markers. Furthermore it is accountable in terms of global mappings, as postulated by the TNGS. Note that the use of lexical items with attitudinal components can cause the non-linguistic conceptual system to be recategorized. Thus, the non deliberate use of a politically incorrect vocabulary by a child can influence her construal of a given category of persons or objects: from hearing bald males being referred to as ‘baldies’, a child is likely to acquire prejudices towards this section of the population.

Appropriateness 107

4.6 Schematized situations and the decision-making process Examples such as (6), which include a discourse marker, present a different type of parallel with the decision-making process. But to see this one must first describe the core meaning of mais in terms of a schematized situation (or an implicit argumentative sequence), as construed by Argumentation Theory (Cadiot et al. 1985: 106–124; Ducrot 1988: 10, 1995: 94). This schematized situation, which constrains interpretation, is construable in terms of interacting viewpoints. An instantiated version of it is set out in: (8) V1: The plants need watering (P), so S should do it (C). V2: P may be the case, but it does not follow that C (in other words, the agreement, which is over P in a representational capacity, entails no commitment on S’s part as to whether the situation as described in P, is conducive to C. The disagreement concerns the aspect under which the situation of reference is described in P). V2’: The reason for V2 – one arguably more pertinent to the issue under consideration – is to do with a further aspect of the situation of reference, namely that it is going to rain (=P’). V2”: In view of P’, C’ is the correct conclusion to reach (where C’ is opposed to C).

Of these, V1 is imputed to R, while V2, V2’ and V2” detail various stages of the stance taken by S. We saw earlier that argumentative discourse is about inducing a decisionmaking process in R. Now, following Nyan (2006) I propose that mais belongs to a category of argumentative markers that encapsulate strategies aimed at speeding up this process. Thus, if the schematized situation is recast in terms of decision making, what becomes immediately apparent is that the above usage of mais intervenes at the stage where the SR is categorized. Consider again (6): (6) Les plantes ont besoin d’être arrosées (P), mais il va pleuvoir (Q).

In saying P, S is presenting a prior category judgment of the SR, one which invokes a complex category CC1. CC1 consists of a pairing of initial situations featuring plants showing signs of needing water and response options involving various means of supplying them with water through human agency. Mais is construable in terms of instructions, one of which is that the SR is to be recategorized as in Q. This recategorization involves accessing another complex category CC2, which differs from CC1 in two respects: the set of initial situations includes the likelihood of impending rainfall and the set of responses includes watering methods that do not require human intervention. In other words, the way in which mais

10 Thanh Nyan

(in (6)) helps R’s decision-making process along is by identifying, among competing complex categories, the one that is advantageous to S’s argumentative intent. The appropriate contribution on R’s part is one underpinned by CC2. There remains to say a few words about how the above apparatus would account for the coding of such schematized situations, which are at the core of the strategy under consideration. While details will have to await further investigation, there are indications that certain key aspects would be compatible with this apparatus. The fact that an interactive context could have made its way into the meaning of a linguistic expression can be accounted for in terms of explicit memory encoding (Kosslyn & Koenig 1995: 317), the process whereby the information to be remembered is encoded with its context of occurrence. Inasmuch as discourse markers are a late development, historical investigation would have much to contribute to the viability of the correlation between schematized situations and typical co-texts for linguistic elements that later evolved into discourse markers. With regard to the schematic nature of schematized situations, it can be accounted for in terms the conceptual system postulated by the TNGS. One feature of this system is that is capable of relational categorization (Edelman 1989: 1140–1146), in the sense that it can abstract relational concepts from immediate perceptual categorizations and past categorizations, in this case, actual fully instantiated co-texts accompanying focal linguistic elements. The big question that remains concerns the nature of the connection between the process whereby argumentative markers acquire the above function, and the decision-making process. The linguistic data I have been considering exemplify contents based cases of appropriateness, which also happen to be marked by varying degrees of linguistic constraints. The overall picture that emerges is that appropriateness can be defined in terms of two levels of conceptual memories, one non-linguistic, the other linguistic. The difference between these cases lies in the way in which the two sets of memories interact. In (3), which linguistic representations of initial situations and of response options are to be associated is determined, in the first instance, by the non-linguistic conceptual (or n.l.c.) memory. This n.l.c. memory appears to have influenced recategorization at the level of the linguistic conceptual (or l.c.) memories. The constraint associated with “well-off ” in (3), which weakly constrains preferential appropriate continuations, would be the result of such recategorization. The cases involving attitudinal components open up the possibility for the linguistic memories to be prior: people begin by associating certain linguistic representations, rather than others, in conformity to prevailing linguistic norms. The strong linkage created by collective usage at the linguistic level, which controls linguistic appropriateness, then goes on to trigger recategorization at the level

Appropriateness 109

of the n.l.c memory, thereby causing existing categories to become linked. Obviously, if the complex categories under consideration are already present in the n.l.c. memory, then there would be mutual reinforcement of the corresponding linkages at the linguistic and non-linguistic levels. The case of mais and its schematized situation exemplifies yet another type of interaction between the two sets of memories. The l.c. memories provide the schematized situation, which arises through recategorization and corresponds to a complex category linking a set of argumentative sequences and a set of appropriate conclusions. Once instantiated by the contents available from host utterances, this schematized situation determines which complex categories are accessed from the n.l.c. memory, and which is to be used to recategorize the SR. This, in turn, ensures that the conclusion is appropriate at the level of its contents.

5.

Conclusion

By way of conclusion, I begin by summarizing how appropriate action is construed from an adaptive perspective. Following this, I propose to revisit the notion of context in relation to which appropriateness is usually defined. An adaptive perspective crucially entails an internal viewpoint, that of individuals having to learn what constitutes adaptive (or appropriate) behaviors and how to produce them efficiently. At the pre-learning stage, an individual faced with an SR (i.e., an external situation) which appears to require action, needs to make a conscious decision about what to do by referring to domain specific norms. The latter are available from the social environment, or from her declarative memory. Once learning has taken place, appropriate action can be automatically triggered. Conscious decision making thus gives way to a highly automatized process whereby the SR is categorized in terms of a complex category held in a procedural memory, and a pre-selected range of appropriate responses is delivered. At this stage, appropriate behaviors no longer require that reference be made to a set of norms, they have become sensitive to them. Thus, on this view, appropriate action: a. does not stand in a direct relationship to the norms that define it: this relationship is mediated by a learning process governed by innate values; b. is the end result of a process whereby an acquired procedural knowledge is tapped to provide a template for action, in the form of a complex category.

110 Thanh Nyan

This has clear implications for the notion of context, whether construed in terms of a set of norms or conventions (=a) with respect to which appropriateness is usually defined, or in terms of an event (=b) that precedes appropriate action. Context, in sense (a), may be seen as a key element in the acquisition of the procedural knowledge under consideration, but one that ceases to play a role in generating appropriate behavior, once the acquisition phase is over. Context, in sense (b), as exemplified by the SR, to which the individual has to respond, is an external situation. However, this situation cannot be seen to require adaptive action, unless it possesses certain features which make it subsumable under a complex category. In other words, context in this sense must also be defined in terms of individuals’ adaptive needs. The external entity as such is of no relevance to this perspective. What matters is how it is perceived and categorized. If the idea of context is to be retained, perhaps the best candidate for it is either the knowledge being accessed during the (highly automatized) decision-making process on its own, or this knowledge and some aspects, at least, of the decisionmaking process itself. Either way, we are dealing with a mental construction. The question this prompts is: how do non-linguistic and linguistic cases differ in terms of mental construction? Inasmuch as the latter are argumentative and argumentative discourse is about inducing a certain decision-making process in R’s working memory, the expectation is that the logistics involved in context construction cannot be quite the same. In self-determined decision making (where someone tries to select a response), the process starts off with the use of a complex category to arrive at a category judgment of the SR. In the case of induced decision making, we see a reversal of this process: S provides a categorization of the SR by means of a linguistic representation, and this has the effect of triggering the corresponding complex category in R’s working memory to the exclusion of other categories. Getting R to access the right category finds a simpler parallel in the problem facing an animal which needs to respond to a situation, but finds itself confronted with more than one object, each of which requiring a different response. Such an animal cannot select a course of action unless it possesses the capacity to direct action to one object to the exclusion of the others. This corresponds to a functional view of attention (Allport 1987: 405–415; and Neumann 1987: 373–382) according to which attention is adaptive in that it enables the animal to select actions appropriate to a predetermined goal. In short, linguistic representation is a way of directing R’s attention to categories that are advantageous to S’s goal, while excluding competing categories. In all three linguistic cases, linguistic representations of the SR would thus be responsible for triggering a certain mental context in R. The third case, which involves an argumentative marker and the attendant schematized situation, pres-

Appropriateness

ents a more complex level of interference with R’s decision-making process. Basically, this is carried out by means of two sets of constraints: a. one concerning the relation in which R’s and S’s categorizations of the SR are to stand; b. the other, the way in which these categorizations are evaluated. Assuming that the schematized situation is imposed in a top-down mode,23 the implication is that the constraints involved determine which contents are to be used to instantiate R’s and S’s views in the schematized situation and the conclusion. In terms of the construction of R’s mental context, the hypothesis I am putting forward is that linguistic representations (corresponding to category judgments of the SR) impact on the nature of the contents of the complex categories to be accessed from the n.l.c. memory (This impact is direct or indirect, depending on whether the contents represented are explicit or implicit). By contrast, argumentative markers govern the type of complex categories being accessed in terms of their relation, the way in which these competing complex categories are evaluated, and the order in which they are accessed.

Acknowledgements I wish to thank the British Academy for the aid I received in the preparation of this article. I am also grateful to my anonymous reviewers for their helpful comments.

References Allport, D.A. 1987. “Selection for action: Some behavioral and neurophysiological considerations of attention and action.” In Perspectives on Perception and Action, H. Heuer and A.F. Sanders (eds.), 395–419. Hillsdale, NJ: Lawrence Erlbaum. Anscombre, J.-C. (ed.). 1995. La Théorie des Topoï. Paris : Kimé. Cadiot, A., Ducrot, O., Nguyen, T. and Vicher, A. 1985. “Sous un mot une controverse: Les emplois pragmatiques de toujours.” Modèles Linguistiques 7(2): 106–124. Damasio, A. 1994. Descartes’Error. New York: The Free Press. Damasio, A. 1999. The Feelings of what Happens. San Diego: Harcourt.

23. Sequences that are not fully instantiated (e.g., “les plantes ont besoin d’être arrosées, mais …”) are unambiguous as to what type of conclusion they require.

111

112

Thanh Nyan

Damasio, A. and Damasio, H. 1994. “Cortical systems for retrieval of concrete knowledge.” In Large-scale Neuronal Theories of the Brain, C. Koch and J.L. Davis (eds.), 61–74. Cambridge MA: The MIT Pres. Deacon, W.T. 1997. The Symbolic Species. New York: Norton. Ducrot, O. 1988. “Topoï et formes topiques.” Bulletin d’Etude de Linguistique Française 22:1– 14. Ducrot, O. 1995. “Topoï et formes topiques.” In La Théorie des Topoï, J.-C. Anscombre (ed.), 85–99. Paris: Kimé. Edelman, D.G. 1989. The Remembered Present. New York: Basic Books. Fetzer, A. 2000. “Negotiating validity claims in political interviews.” Text 20 (4): 415–59. Fetzer, A. 2002. “Micro situations: Natural language communication and context.” Foundations of Science 7: 255–291. Fetzer, A. 2004. Recontextualizing Context. Grammaticality meets Appropriateness. Amsterdam: John Benjamin. Fetzer, A. 2007. “Reformulation and common grounds.” In Lexical Markers of Common Grounds, A. Fetzer and K. Fischer (eds.), 159–182. Amsterdam: Elsevier. Fetzer, A. This volume. “If I may say so: Indexing appropriateness in dialogue.” Fodor, J.A. 1983. The Modularity of Mind. Cambridge, MA: The MIT Press. Kahneman, D. and Miller, D.T. 1986. “Norm theory: Comparing reality to its alternatives”. Psychological Review 93 (2): 136–153. Kosslyn, S.M. and Koenig, O. 1995. Wet Mind. New York: The Free Press. Koch. C. 2004. The Quest for Consciousness: a neurobiological approach. Englewood, Colorado: Roberts and Company Publishings. Neumann, O. 1987. “Beyond capacity: A functional view of attention.” In Perspectives on Perception and Action, H. Heuer and A.F. Sanders (eds.), 361–391. Hillsdale, NJ: Lawrence Erlbaum. Nyan, T. 2006. “Discourse markers and the decision-making process.” Sprache und Datenverarbeitung 30 (1): 109–118. Nyan, T. 2007. “Common ground, categorization and decision making”. In Lexical Markers of Common Grounds, A. Fetzer and K. Fischer (eds.), 195–210. Amsterdam: Elsevier. Rosch, E. 1978. “Principles of categorization.” In Cognition and Categorization, E. Rosch and B. Lloyd (eds.), 28–48. Hillsdale, NJ: Laurence Erlbaum. Searle, R.J. 1979. Expression and Meaning. Cambridge, MA: CUP. Searle, R.J. 1983. Intentionality. Cambridge, MA: CUP. Searle, R.J. 1995. The Construction of Social Reality. New York: The Free Press. Searle, R.J. 2001. Rationality in Action. Cambridge, MA: The MIT Press.

part ii

Bridging problems between communicative action and appropriateness

If I may say so Indexing appropriateness in dialogue Anita Fetzer

University of Lueneburg, Germany

This contribution examines those contexts in which a contribution is assigned the status of not such as is required (Grice 1975). The first part analyses the theoretical construct of appropriateness in an integrated framework based on Habermas’s theory of communicative action, Gice’s logic and conversation and Sbisà’s approach to speech act theory. The second part operationalizes the theoretical construct as appropriateness conditions. The third part presents a data analysis in which the form and function of explicit and implicit references to appropriateness are analysed in the genre of a political interview. Particular attention is given to the questions of where the inappropriateness is manifest, what constitutive part of the contribution is assigned the status of being inappropriate, who refers to the inappropriateness and how the inappropriateness is realized linguistically.

1.

Introduction

Natural-language communication is an extremely complex endeavour, which has been investigated from a number of different perspectives. Critical discourse analysis examines the linguistic representation of social actors and social actions and their connectedness with context (e.g., Fairclough 1995; van Dijk 1998). Conversation analysis identifies recurring patterns and sequences in context to account for intersubjectivity and shared understanding (e.g., Garfinkel 1994; Heritage 1984; Prevignano & Thibault 2003). While both research paradigms are essentially anchored to surface data and look upon language and language use from a sociosemiotic perspective thus backgrounding the relevance of coparticipants’ communicative intentions and inferencing processes, a sociopragmatic framework accounts for the fundamental pragmatic premises of rationality, intentionality and cooperation (e.g., Grice 1975; Habermas 1987; Searle 1969) on the one hand, and for the context-dependence and sequential organization of

116 Anita Fetzer

communication on the other hand (Fetzer & Meierkord 2002; Fetzer 2004; Sarangi & Slembrouck 1996). Implicit in all the approaches is the assumption that natural-language communication is a dialogic endeavour. To be felicitous, it requires minimally two coparticipants who share a common linguistic code, common linguistic practices and a common context.1 In sociopragmatics, coparticipants anchor their communicative contributions to context and base their inferencing on context. In order to produce and interpret communicative contributions in a rule-governed manner, however, they need to be conceived of as rational agents who perform communicative acts intentionally and strategically. While the more traditional notion of rationality tends to be restricted to the dichotomy of truth and falsehood, socioculturally based rationality goes beyond that dualistic frame of reference. In Brown and Levinson’s interpersonal-oriented framework, rationality covers a model person’s information wants and face wants (Brown & Levinson 1987). In Brandom’s philosophy-of-language anchored work making it explicit (Brandom 1994), rationality is expressed in discourse and manifests itself in the communicators’ acts which they perform in accordance with their conception of rules. Habermas’s theory of communicative action (Habermas 1987) is based on a sociocultural conception of rationality. In that frame of reference, rationality is reflected in the coparticipant’s performance and interpretation of communicative action in accordance with the sociocultural values and norms of a speech community. The rules and values are defined in a tripartite system of truth represented by the objective world, normative rightness represented by the social world, and subjective truthfulness represented by the subjective world. As regards the concept of action, Habermas differentiates between the one-world based concepts of teleological action and strategic action, and the concepts of norm-regulating action and communicative action which are attributed to the objective and social worlds. The dialogic framework, on which this analysis is based, is informed by Grice’s logic and conversation (Grice 1975) and Habermas’s conception of communication as postulation and ratification of validity claims through an acceptance, rejection or neutral stance (Habermas 1987). However, it goes beyond Habermas’s top-down approach by anchoring communicative action to the micro domain of a face-to-face interaction thus accommodating its interpersonal, interactional and textual dimensions. The change of perspective from macro to micro allows for an

1. Common linguistic code, common linguistic practices and common context are represented by scalar concepts, and can therefore never refer to identical individual representations. Rather, coparticipants presuppose a commonly shared representation with fuzzy boundaries.

Indexing appropriateness

interpersonal outlook on dialogue in which appropriateness is no longer a macro concept and theoretical construct, but rather is done in communication. The goal of this chapter is to examine those environments in which a communicative contribution is assigned the status of not appropriate; this may refer to the contribution as a whole, or to one or more of its constitutive parts. Assigning a contribution, or some of its parts, an inappropriate status is taken to be functionally equivalent with the negated Gricean constraint of not “such as is required, at the stage at which it occurs, by the accepted purpose or direction of the talk exchange” (Grice 1975: 45). The contexts in which the constraint is referred to are classified with regard to the questions of where the reference to inappropriateness occurs, what constitutive part(s) of the contribution are assigned an inappropriate status, who refers to the inappropriateness and how the inappropriateness is realized linguistically. The following section, dialogue and dialogues, examines the theoretical construct of dialogue and its instantiation in context in a integrated sociopragmatic framework. The third section, appropriateness, analyses the theoretical construct by examining its default status and its instantiation as appropriateness conditions. The fourth section, appropriate, inappropriate, or more and less appropriate?, looks at the form and function of explicit and implicit references to appropriateness in the genre of a political interview, and the final section summarizes the results obtained and concludes.

2.

Dialogue and dialogues

Dialogic theories of natural-language communication have focussed on the cognitive (or internal) and social2 (or external) domains of dialogue and on their impact on language use. While the former tend to employ polyphonous (e.g., Bakthin 1981; Ducrot 1984; Johansson 2002) and heteroglossic frames of reference (e.g., White 2003), the latter generally use conversation-analytic (e.g., Atkinson & Heritage 1984; Boden & Zimmerman 1993) or critical-discourse-analytic frameworks (e.g., Fairclough 1995; van Dijk 1998). Integrated approaches to dialogue have adopted a holistic stance (e.g., Fetzer 2004; Linell 1998; Vygotsky1986), in which perception and internal reasoning are connected with the contextual constraints and requirements of an external face-to-face interaction in sociocultural contexts. In the philosophy-of-language paradigm, dialogue has been conceptualized as a form of language game or activity type in which coparticipants perform 2. In spite of the fact that ‘sociocultural’ is a hyponym of ‘social’, the two terms are used as a functional synonyms in this contribution.

117

11

Anita Fetzer

communicative acts by exchanging communicative contributions in context (e.g., Austin 1980; Brandom 1994; Grice 1975; Levinson 1979). Despite the fact that all of the paradigms have approached dialogue from different, but not mutually exclusive angles, their common goal is to account for intersubjectivity, reciprocity and collective action, as is reflected in the conversation-analytic premises of collaboration and joint production (Schegloff 1991), in the philosophy-of-language premises of cooperation and collective intentionality (Searle 1991) and in the accommodation of I-thou perspectives (Brandom 1994). From a parts-whole perspective, dialogue has been described as a macro sequence which itself is composed of a (finite) number of micro sequences (Fetzer & Meierkord 2002). Depending on the research paradigms, the micro sequence is called dialogue act (Bunt 1999; Fetzer 2004), discourse representation structure (Kamp & Reyle 1993), sentence and proposition (de Beaugrande & Dressler 1981), proposition and speech act (Sbisà 2002a; Searle 1991) or utterance and turn (Schegloff 1991). Accordingly, it is defined with regard to frame-of-investigation specific constraints and requirements, and with regard to frame-of-investigation-specific purposes. Thus, a micro sequence expresses semantic content in discourse representation theory, communicative function in conversation analysis and critical discourse analysis, or both semantic content and communicative function in speech act theory, dynamic interpretation theory and sociopragmatics. Furthermore, a micro sequence changes the context by adding propositional content (Stalnaker 1999), and it reconstructs a communicative genre (Luckmann 1995). In text linguistics, text and dialogue are composed of sentences and propositions, and they have been investigated with regard to cohesion and coherence (Halliday & Hasan 2000). In conversation analysis, dialogue is composed of turnconstructional units, turns and topical sections (Schegloff 1996), and in sociopragmatics, dialogue is composed of utterances which are employed in an intentional and strategic manner in order to perform communicative acts (Fetzer 2000; Mey 1996; Sarangi & Slembrouck 1996). While all of the frameworks share the premise that dialogue as a whole is more than the sum of its parts, their methodology differs in significant ways. The application of sentence-grammar methodology in text linguistics and of speechact-theoretic methodology in dialogue analysis (Sbisà 2002a, 2002b) is based on discrete units, such as a sentence and its constituents, and a speech act and its constitutive acts, which are connected to compose the dialogue as a whole. By looking upon utterances and turns as jointly produced in context, conversation analysis no longer employs discrete units. Its dynamic frame of reference considers communication as both process and product.

Indexing appropriateness

2.1 The dynamic nature of dialogue Dialogue is composed of a finite number of micro sequences, whose sequential organization adheres to a particular linear order,3 to construct a particular macro sequence. The order of the macro sequence is strongly influenced by the constraints (or appropriateness conditions) of genre, for instance lecture, interview or narrative. Looked upon from a top-down perspective, a macro sequence subclassifies into a finite number of meso sequences whose sequential organization is also constrained by genre. For instance, a lecture generally commences with an opening section, in which title, content and structure are made explicit. After the opening section there are a number of topical sections which are semantically connected with the lecture’s overall theme. It is at that stage, where the sequential organization of the meso themes may vary. Depending on sociocultural preferences, there can be inductively- or deductively based sequential organizations, and the individual sequences may be connected with explicitly realized cohesive links. A closing section generally sums up the content by putting it in another context. Against this background, dialogue is both process and product. It is process insofar as the coparticipants’ exchange of micro sequences can be conceptualized as an exchange of building blocks which are connected in the process of communication by joining adjacent blocks in order to construct a meso sequence (or an intermediate whole), only to be further connected to reach its ultimate goal: the dialogue as a whole. Metaphorically speaking, dialogue can be conceptualized as a building site or as dialogue-under-construction, in which individual parts of the whole building are assigned the status of meaningful parts, such as a lounge, dining room or kitchen. The kitchen stands for the opening section, the dining room stands for a topical section and lounge stands for the closing section, for instance. Dialogue represents a product insofar as one type of dialogue, such as a political interview, can be composed of very differently constructed building blocks (or micro blocks), such as briefly or elaborately formulated agreements, which require different seconds (or continuation blocks). Brief or elaborate disagreements, by contrast, tend to require follow-up seconds, in which the disagreement is accounted for. Thus, the sequential organization of dialogue may result in an almost infinite number of differently constructed wholes. To accommodate a dialogue’s dual status as process and product, bottom-up and top-down perspectives are required. The former can account for the construc3. This holds for traditional texts without any footnote or endnotes. The more recent phenomenon of hypertext exploits the constraint of strict linearity and allows for multiple text construction.

119

120 Anita Fetzer

tion of micro sequences and thus for the construction of parts, such as a communicative act on the level of turn and the negotiation of intersubjective meaning on the level of exchange, while the latter can account for the delimiting phenomenon of genre and its contextual constraints and requirements, for instance the genrespecific turn-taking system of an interview or the genre-specific discourse topics of a political interview, job interview or medical interview. Of course, both perspectives are necessary for a theory of dialogue, and they can be reconciled by the explicit accommodation of context.

2.2 Dialogue and context Context is one of those concepts which are frequently referred to in theories of linguistics, discourse and dialogue, but hardly ever made explicit. In more recent approaches to context, its relational and interactive nature has been stressed (cf. Duranti & Goodwin 1992; Edmonds & Akman 2002; Fetzer & Akman 2002), and context has been categorized into cognitive context, linguistic context, sociocultural context and social context (Fetzer 2004). Cognitive context comprises the local and global cognitive surroundings of a proposition, a mental representation or an assumption. Linguistic context comprises the local and global linguistic surroundings of a linguistic unit, such as a phoneme, morpheme, lexical expression, sentence, speech act, communicative act, communicative contribution or validity claim as well as their language-use based equivalents of turn-constructional unit, utterance and turn. If the surroundings are of an extra-linguistic nature, that is non-cognitive and non-linguistic material, they are called local and global social context. The former comprises the speech situation including time, space and coparticipants, and the latter consists of society and institution. Sociocultural context comprises a culture-specific interpretation of social context. Here, social-context information passes through one particular filter assigning social-context information a culture-specific reading, for instance the interpretation of a social agent as an individual who performs their acts autonomously. Employing another filter, the identical social agent might be assigned a completely different reading as a collective-anchored agent whose acts are a constitutive part of collective action. Another example is the distinction between monochronic time and polychronic time which is based on the differentiation between a linear, tangible and divisible conception of time, where events are scheduled one at a time and where this schedule takes precedence over interpersonal relationships. Polychronic time, by contrast, is characterized by things occurring simultaneously. Here, interpersonal relationships take precedence over a task-oriented outlook on communication (Hall & Hall 1989). In other words,

Indexing appropriateness

culture provides us with a filter mechanism which allows us to interpret social context in accordance with particular sociocultural-context constraints and requirements. Social, sociocultural, linguistic and cognitive context are looked upon as relational and interactive constructs which are constantly updated in and through the process of communication. In a relevance-theoretic frame of reference, cognitive context is conceptualized as an onion, metaphorically speaking, with different layers which are ordered in line with the principle of inclusion. Naturally, they can only be accessed in accordance with the principle of inclusion (Sperber & Wilson 1996). Put differently, the individual layers of an onion are connected in a particular order to construct a whole, and that particular order constitutes the background for the calculation of intersubjective meaning and the corresponding processes of inferencing. If the cognitive framework is adapted to the construction of dialogue, a micro dialogue sequence is the local context for its adjacent dialogue sequences, and the whole macro sequence, of which the micro dialogue sequence is a constitutive part, frames the micro parts by delimiting them from surrounding context (or onion layers). Following Akman and Alpaslan (1999), the language produced (formulated) and interpreted (decoded) by coparticipants is assigned a dual function. On the one hand, it invokes linguistic context by constructing it itself; on the other hand, its sole production and interpretation provide the context for subsequent talk and recovery of intended meaning. Accordingly, the acts of speaking and interpreting build contexts and at the same time constrain the building of contexts. To employ Heritage’s terminology, “the production of talk is doubly contextual” (Heritage 1984: 242). An utterance relies upon the existing context for its production and interpretation, and it is, in its own right, an event that shapes a new context for the action that will follow. Context can be examined from a bottom-up viewpoint thus referring to the local cognitive, linguistic, sociocultural and social contexts, it can be examined from a meso top-down perspective thus referring to the delimiting frame of a cognitive script, text-type, genre or activity type, and it can be investigated from a macro top-down perspective framing the meso setting thus referring to the contextual constraints and requirements anchored to culture, institution and society (cf. Gumperz & Levinson 1996). In a political interview, this means that an interviewer’s communicative contribution CC1 represents the local context for the adjacent communicative contributions CC2 and CC3, and at the same time CC1 is framed by the meso constraint of the genre of a political interview. Because of that, the communicative meaning of CC1 is interpreted in accordance with a political interview’s contextual constraints and requirements and counts as an instance of public and political information. For instance, the interviewer’s question “What did you discuss on Saturday?” in the genre of a political interview does

121

122 Anita Fetzer

not count as a question about the interviewee’s conversations with their children or partners but rather as a question about the contents of a conversation with another politician which had taken place on that particular day. So, context and contexts are omnipresent in dialogue from the analyst’s viewpoint, but are they also of immediate relevance to the coparticipants? In ordinary dialogue, the embeddedness of a communicative contribution in cognitive, linguistic, sociocultural and social contexts and its connectedness with context are generally presupposed and not mentioned. Only in critical situations, disruptive settings (Fetzer 2005) or other types of metatalk, where the felicity of a communicative contribution is at stake, references to context are made explicit. In those negotiation-of-validity sequences, the dynamic nature of communication comes to a halt and references to its dialogic nature, that is to the postulation, interpretation and ratification of a validity claim, are made explicit to be negotiated in order to re-construct common ground. Only then felicitous communication may proceed. In the following, a frame of reference is introduced which accounts for the dynamic nature and context-dependence of dialogue by accommodating the embeddedness of a communicative contribution in local and global contexts.

2.3 Dialogue: An integrated approach The dialogue act of a plus/minus-validity claim is based on an integrated approach to communication. Its foundations are Grice’s cooperative principle (Grice 1975), Habermas’s theory of communicative action (Habermas 1987) and Sbisà’s research on speech act theory (Sbisà 2002a, 2002b), which are introduced below.

2.3.1 ‘Logic and conversation’ in dialogue Grice’s logic and conversation (Grice 1975) is firmly anchored to the pragmatic premises of cooperation and rationality, which are defined in the cooperative principle, the maxims of quality, quantity, relation and manner, and in the conversational implicature. This is reflected in the fact that whenever coparticipants communicate, the minimal presupposition comprises the assumption that their fellow coparticipants’ contributions are in accordance with the cooperative principle, which states the following constraint: “Make your conversational contribution such as is required, at the stage at which it occurs, by the accepted purpose or direction of the talk exchange in which you are engaged” (Grice 1975: 45). The maximal presupposition goes beyond the CP and states that their coparticipants’ contributions are in accordance with both the cooperative principle and the maxims of quantity (“1. Make your contribution as informative as is required

Indexing appropriateness

(for the current purpose of the exchange). 2. Do not make your contribution more informative than is required”), the maxim of quality (with the supermaxim “Try to make your contribution one that is true”, and two more specific maxims: “1. Do not say what you believe to be false. 2. Do not say that for which you lack adequate evidence”), the maxim of relation (“Be relevant”) and the maxim of manner (with the supermaxim “Be perspicuous” and various maxims such as: “1. Avoid obscurity of expression. 2. Avoid ambiguity. 3. Be brief (avoid unnecessary prolixity). 4. Be orderly” (Grice 1975: 45, 46)). If the maximum presupposition does not obtain, the medium presupposition, which entails the minimal presupposition, obtains. In that scenario, one or more maxims are exploited. This means that a coparticipant gets in a conversational implicature whose communicative meaning needs to be calculated by the coparticipants in context. The conversational implicature can be a generalized implicature whose calculation consists of an almost automatized process, or it can be a particularized implicature whose calculation requires a more elaborate process in which additional contextual information needs to be processed. The Gricean framework thus explicitly accounts for the connectedness between communicative contribution, coparticipants, and linguistic and social contexts. Because of the implicature, which is detachable, non-conventional, calculable and defeasible, it can also account for the dialogue’s dual status of process and product. It implicitly accounts for the sequential organization of dialogue in its reference to “the stage at which it occurs” (Grice 1975: 45), but does not refer to dialogue’s constitutive micro and macro components.

2.3.2 Validity claims in dialogue An integrated approach to dialogue feeds on different, but not diverging frameworks. In order to account for the dynamics of dialogue, the Gricean concept of contribution is refined and dynamized through the Habermasean concept of a validity claim defined within his theory of communicative action (Habermas 1987). The theoretical construct of a validity claim is a dialogic concept par excellence. Not only is the original concept explicitly connected to speaker, hearer and to the set of speaker and hearer. Taken one move further, it is also both process and product. That is to say, a validity claim denotes a process insofar as it requires to be both postulated by the speaker and ratified by a hearer to be felicitous. And it denotes a product insofar as a validity claim is the result of a hearer’s ratification regarding its status as accepted (or plus-), rejected (or minus-) or neutral-stance validity claim. The Habermasean conception of a validity claim does not explicitly account for the validity claim’s embeddedness in context and for its connectedness with context, nor for the sequential organization of dialogue and its parts-whole con-

123

124 Anita Fetzer

figuration. To accommodate context, sequentiality and the parts-whole configuration, the theoretical construct of a validity claim needs to be adapted to the contextual constraints and requirements of the micro domain of dialogue. The result, the dialogue act of a plus/minus-validity claim (Fetzer 2002, 2004), is examined in the following.

2.3.3 The dialogue act of a plus-minus-validity claim Habermas’s conception of validity claim is anchored to the subjective, objective and social worlds and its foundations of truthfulness, truth and normative rightness. In the dialogue act of a plus/minus-validity claim, the subjective-world construct of truthfulness is seen as functionally equivalent to speech act theory’s sincerity condition, which is assigned the status of a constitutive rule (Searle 1969). Sincerity manifests itself in the coparticipants’ communicative intentions meant as uttered and interpreted as meant. The objective-world construct of truth is seen as functionally equivalent to speech act theory’s concept of proposition (Searle 1969). Truth manifests itself in the felicitous acts of reference and predication. The social-world construct of normative rightness is seen as functionally equivalent to the sociopragmatic concept of appropriateness. Appropriateness manifests itself in the coparticipants’ communicative actions which are evaluated as being in accordance with the contextual constraints and requirements of a particular communicative act. Not only is the context, in which communicative action is performed, refined by the accommodation of the tripartite system of subjective, social and objective worlds, but so are its necessary and sufficient conditions. The dialogue act of a plus/minus-validity claim is anchored to the over-arching cooperative principle (Grice 1975), which has been explicated above. A plus/ minus-validity claim is defined within a tripartite system of an objective world and its pillar of truth, a subjective world and its pillar of sincerity and a social world and its pillar of appropriateness. The social world subcategorizes into another tripartite configuration of a textual system based on the linguistic realization of a validity claim in accordance with the Gricean maxims and implicature, an interactional system based on adjacency and turn-taking, and an interpersonal system based on the coparticipants’ face wants and status in the participation framework. Every validity claim is composed of references to the three systems and to its presuppositions, which can be realized explicitly and implicitly. The former uses the propositional and explicit-performative formats, and the latter uses social-world based indexical expressions and conversational implicatures. For a validity claim to be felicitous, all of the references to the three worlds and to its presuppositions need to be ratified in dialogue. In the case of ratification through an acceptance, the postulated validity claim is assigned the status of a plus-validity claim, and in

Indexing appropriateness

the case of ratification through a rejection, the validity claim is assigned the status of a minus-validity claim. Both the contexts of a dialogue act and the dialogue acts themselves are defined by a tripartite configuration: the former is composed of subjective, social and objective worlds, and the latter is composed of the acts of production, interpretation and ratification. But how do plus- and minus-validity claims manifest themselves in dialogue? Adapting Sbisà’s claim (2002b) that the felicity of a speech act obtains by default unless indicated otherwise to the dialogue act setting, the felicity of the dialogue act of a plus/minus-validity claim manifests itself in the ratification of all of its references to the three worlds, that is truth, sincerity and appropriateness, unless indicated otherwise. Put differently, if there are no explicitly or implicitly realized rejections of the references of a validity claim4 to the three worlds and/or to its three systems, then the validity claim is ratified through an acceptance and assigned the status of a plus-validity claim. In this scenario, dialogue proceeds in the coparticipant-intended manner. If there are implicitly or explicitly realized rejections, dialogue comes to a halt and the non-accepted validity claims need to be negotiated to come to a shared agreement about their statuses as plus- or minus-validity claims. Only then, felicitous communication may proceed. To summarize and conclude, in dialogue coparticipants negotiate the validity of their communicative contributions which comprises their truth, sincerity and appropriateness. A contribution’s truth is captured by its truth conditions anchored to the objective world, which is true in any context. A contribution’s sincerity is captured by its sincerity condition anchored to the cognitive-context concept of subjective world, and a contribution’s appropriateness is captured by its appropriateness conditions anchored to the social-context concept of appropriateness, which is examined more closely below.

3.

Appropriateness

Acceptability and appropriateness are differentiated by van Dijk (1981) along the following lines: the former denotes an empirical concept which refers to linguistic form, and the latter denotes a theoretical construct. Crystal (1997) refines van Dijk’s definition by explicitly accounting for the impact of situational context on linguistic variation. He looks upon appropriateness as a variety of form which is considered suitable or possible in a given situation. Kasher (1998) defines ap4. A validity claim is a constitutive part of a communicative contribution. It can be realized through verbal signs and non-verbal signs, such as gesture, gaze or posture.

125

126 Anita Fetzer

propriateness in a formal frame of reference. Unlike van Dijk and Crystal, he does not attribute appropriateness to a communicative contribution (or linguistic form), but to context: Holding that the spirit of any linguistic use is a linguistic institution, I would suggest the following criterion of adequacy of pragmatical theories: (CA) For every context of utterance C, every sentence S of a natural language L, and every ideal speaker Alpha of that language, the following biconditional should be a true theorem: Context C is linguistically appropriate for speaker Alpha to utter in it sentence S of language L, if, and only if, there is a linguistic institution of L which grants Alpha an institutional role which enable him (or her) to achieve a literal purpose he (or she) entertains in C, by uttering in it sentence S. I shall call this adequacy condition “Criterion A” of appropriateness. (Kasher 1998: 89)

The stance taken in this contribution neither attributes appropriateness to context nor to linguistic form. Rather, appropriateness is conceived of as a relational construct which connects coparticipants and their contributions with context on the local domain, and which connects the local domain of dialogue with the macro category of genre. Appropriateness thus supplements and refines the notion of context-dependent pragmatic meaning by the explicit accommodation of sociocultural context. In an appropriateness-based scenario, pragmatic meaning is not only inferred with regard to illocutionary goal and force but also with respect to the additional features of coparticipant and their social, interactional and discursive roles, and the constraints and requirements of a communicative setting. As a necessary consequence of the connectedness between context, coparticipants and communicative action, appropriateness is anchored to the dyad of (minimally) a speaker and a hearer seen from both I-we (Searle 1995) and I-thou perspectives (Brandom 1994). Or, to adopt Linell’s (1998) distinction between monologue and dialogue, appropriateness represents a dialogical concept par excellence. It is a social concept, whose frame of reference goes beyond the individual contribution. This is due to the fact that the appropriateness of dialogue cannot be calculated in the framework of individual contributions only. It rather requires a collectively oriented framework, as is explicated by Vanderveken and Kubo: “Conversations are language games that several speakers play by performing together a joint action rather than several distinct individual activities” (Vanderveken & Kubo 2002: 16).

Indexing appropriateness

3.1 Common-sense notion and theoretical construct The concept of appropriateness has been differentiated with respect to the domains of common sense and theoretical construct (Fetzer 2004). In a common-sense scenario, appropriateness denotes the product of the evaluation of a communicative act which is seen as produced in accordance with a speech community’s norms for that particular communicative act. For instance, the performance of the speech act of a request through a reference to the hearer (you) and the preparatory conditions (ability) with “can you do X” is seen as appropriate in the Anglo-American context, while a self-reference to the negated preparatory conditions (not ability) with “I can not do X” does not necessarily count as an appropriate request in similar contexts (cf. also Akman, this volume). Thus, commonsense appropriateness refers to the product of a process of evaluation based on the nature of the connectedness between coparticipants, communicative act and its linguistic realization in linguistic and sociocultural contexts. If the performance of a communicative act is seen as produced and interpreted in accordance with a speech community’s ethnographic norms and strategies, it is assigned the status of an appropriate communicative act. Should it violate one or more of the norms and strategies, it is assigned the status of an inappropriate communicative act. As a theoretical construct, appropriateness is conceived of as a relational concept which is far more complex. It is informed by the contextual constraints and requirements of (1) coparticipants and their social, interactional and discursive roles, (2) communicative action, (3) genre, and (4) ethnographic norms and strategies of a speech community. For instance, in the genre of a therapeutic interview, the request “can you do X” examined above produced by the therapist does not generally express a request to do X but rather a yes/no-question about the client’s physical condition. If uttered by the identical speaker to the identical client in the genre of a therapeutic session, it counts as an appropriate request for a particular (non-verbal) action. However, if the request is directed to the therapist’s assistant, it counts as an appropriate request to perform the action X. Naturally, the conceptualization of appropriate coparticipant role, appropriate communicative action and appropriate genre are interdependent on other sociocultural norms, conventions and strategies, which themselves are interdependent on the sociocultural norms, conventions and strategies of subsocieties and subcultures. This is reflected in politically correct and politically incorrect conceptualizations of the linguistic code and its use in context (or social practice), such as the more or less appropriate references “chairman”, “chairwoman”, “chairperson” and “chair”, whose appropriateness depends on context. The former two tend to be appropriate in a conservatively oriented setting, while the latter two are appropriate in a wider domain of reference. Or, requesting the hearer to get X for-

127

12 Anita Fetzer

mulated through “get X, will ya” is appropriate only in informal settings while the standard version “could you get X” is appropriate in a wider domain of reference. So, there are subset-specific norms, conventions and strategies for the production and interpretation of a communicative act.

3.2 Social-world construct Appropriateness has been defined as the pillar of the social world which is one of the three constitutive worlds of the dialogue act of a plus/minus-validity claim. It manifests itself in the performance of a communicative act with respect to the formulation (or linguistic representation) of a communicative contribution and with respect to the interpretation (or decoding and inferencing) of a communicative contribution. Here, the linguistic surface of the contribution is de-composed with respect to its references to sociocultural contexts which can be realized explicitly or implicitly. The latter triggers a process of inferencing to calculate the implicature. For instance, the evaluation of a contribution as inappropriate can be realized explicitly by saying “what you’ve just said is inappropriate”, and implicitly by saying “well” or “is that so”. The latter two variants signify that the prior contribution has not been “such as is required” (Grice 1975: 45). Analogously to the tripartite configuration of the dialogue act of a plus/minus-validity claim which categorizes into social, subjective and objective worlds, the social world categorizes into another tripartite configuration of textual, interpersonal and interactional systems. The interpersonal system is based on the social-psychology concept of face (Brown & Levinson 1987) and Goffman’s notions of footing and participation framework (Goffman 1974; Levinson 1988). The interactional system is based on the Levinson’s interpretation of the conversation-analytic concepts of adjacency and conditional relevance (Levinson 1983), and the textual system is based on the Gricean maxims and implicature (Grice 1975). From a natural-language communication perspective, appropriateness is not only a social-world construct but rather a sociocultural-world construct. This is because the textual, interpersonal and interactional systems of the idealized social world obtain a culture-specific interpretation. For instance, the interpersonal-system constraint of face-needs and face-wants is quite different in Anglo-American and Scandinavian contexts, for instance, and validity claims referring to that system would result in different interpretations. In the following, the contextual constraints and requirements of the social world and of its constitutive systems are examined with respect to the question of whether they can be assigned the status of appropriateness conditions.

Indexing appropriateness

3.3 Appropriateness conditions The evaluation of a communicative contribution as appropriate or inappropriate is carried out with respect to its connectedness with the social world and its systems. As has been explicated above, the over-arching Gricean cooperative principle is inherited to the social world, where it is supplemented with the four maxims and the implicature. For a contribution to be appropriate, it means that it is such as is required, at the stage at which it occurs, by the accepted purpose of direction of the talk exchange (Grice 1975: 45). The constraint formulated in the cooperative principle does not only hold for the micro domain of exchange but also for the macro domain of genre, as has been demonstrated elsewhere (Fetzer 2002, 2004). To be more precise, such as is required refers to the contribution’s textual system which determines the contribution’s formulation regarding quantity, quality, relation and manner, it refers to the contribution’s interpersonal system which determines the coparticipants’ face wants and their status in the participation framework, and it refers to the interactional system at the level of exchange with respect to adjacency pair, adjacency position, adjacency relation and conditional relevance. For instance, the contribution “well done” said by A to B in a context where B had just received a 1st class university degree is produced in accordance with the maxim of quality (A has evidence that B has just received a 1st class degree), it is produced in accordance with the maxims of quantity (it is as informative as required), relation (it is relevant as it is addressed to the university graduate) and manner (it is brief). It is produced in accordance with the contextual constraints and requirements of the interpersonal system (in a face-to-face encounter as well as in a telephone conversation, an explicitly realized other-reference can be omitted, and the degree of positive politeness outweighs the possible intrusion anchored to negative politeness). It is produced in accordance with the contextual constraints and requirements of the interactional system, where the contribution “well done” is a directly adjacent response to the news about the results. Had there been exploitations of one or more of the Gricean maxims, a process of inferencing would have been triggered to calculate the conversational implicature. In that medium-presupposition scenario, the process of inferencing would have used additional interpersonal- and interactional-system specific information. A negotiation of meaning on the micro level of exchange is not only performed locally. It is also connected with the macro category of genre which is reflected in Grice’s reference to the domain of the stage, which represents another relational concept. This is because the concept of stage is always connected with a particular position in a sequence, and it is at that particular stage in a sequence anchored to the macro category of genre, where particularized values for the maxims of quality, quantity, relation and manner, for face-wants and footing, and for adjacency

129

130 Anita Fetzer

can become significant. That is to say, in a default dialogue the generalized values set for a genre are generally inherited to the micro domain where they obtain. The generalized values may, however, be exploited thus getting in a conversational implicature. It is those particularized values, in accordance with which the conversational implicature is calculated. These particularized values of the local domain of stage are then explicitly connected with the generalized values of the macro domain, that is the genre where the generalized macro values of the textual, interpersonal and interactional systems are set. This is implicit in Grice’s reference to the accepted purpose of the direction of the talk exchange. Thus, appropriateness is a relational concept anchored to coparticipants, communicative contribution, genre and context, and is, for this reason, both micro and macro. Analogously to speech act theory’s felicity conditions, dialogue analysis is anchored to appropriateness conditions. These entail a speech act’s felicity conditions which are adapted to the contextual constraints and requirements of a dialogue act’s linguistic realization, as is pointed out by Saville-Troike: “The choice of appropriate language forms is not only dependent on static categories, but on what precedes and follows in the communicative sequence, and on information which emerges within the event which may alter the relationship of participants” (Saville-Troike 1989: 53–54). Appropriateness conditions do not only accommodate the constitutive systems of a particular type of speech act. They also accommodate the systems required for its linguistic realization, viz. the syntactic, morphological, phonological and semantic systems. They further accommodate the systems required for its expression and interpretation, that is the interpersonal system, the interactional system, and the textual system. Thus, appropriateness and its operationalization in the framework of appropriateness conditions is a relational construct which feeds on an individual contribution and on the contribution’s connectedness on the local level of exchange, and on the connectedness with the macro category of genre and the genre’s embeddedness in social and sociocultural contexts. In the following, the question of how appropriateness is done in dialogue is examined in the genre of a political interview.

4.

Appropriate, inappropriate, or more and less appropriate?

Appropriateness has been defined in the framework of the dialogue act of a plus/ minus-validity claim as the pillar of the social world, which categorizes into another tripartite configuration of interpersonal, interactional and textual systems. The constraints of the social world and of its systems have been assigned the status of appropriateness conditions, which categorize into generalized appropriateness

Indexing appropriateness

conditions, which obtain in a more general sociocultural context, and particularized appropriateness conditions, which obtain in a particular context, such as the context of a genre.5 Regarding its semantics, appropriateness represents a relational construct, and the relational nature is inherited to its constitutive interactional, interpersonal and textual systems, which are also relational in nature. Because of the multi-layered configuration and the interactive systems, appropriateness denotes a scalar concept with fuzzy boundaries. Thus, there are communicative contributions which are in full accordance with all of the constraints of the social world and its systems and, for this reason, are appropriate contributions par excellence. And there are contributions which infringe on one or more of the social-world constraints and requirements and, for this reason, are more and less appropriate. Depending on the quantity and quality of the infringement, contributions can also be inappropriate, if not fully inappropriate. In the previous section, appropriateness conditions have been assigned a presuppositional status. For this reason, they are generally not made explicit. As a consequence of that, a reference to a contribution’s appropriateness conditions signifies that its appropriateness is at stake. In natural-language communication this can be expressed in the explicit and implicit modes through – – – – – –

explicit references to the social-world construct of appropriateness explicit references to generalized appropriateness conditions explicit references to particularized appropriateness conditions implicit references to the social-world construct of appropriateness implicit references to generalized appropriateness conditions implicit references to particularized appropriateness conditions

As dialogue is a dynamic context-dependent endeavour, it is significant to identify (1) the appropriateness conditions which have been infringed on, and (2) the coparticipant who has referred to the infringement, namely – – –

self other mediated other

Furthermore, the distribution of the explicitly and implicitly realized references to appropriateness conditions as well as their referential domains need to be taken into consideration in the analysis of their communicative functions.

5. In the particular context of a genre, there can be further particularized values which may infringe on the genre-specific constraints.

131

132 Anita Fetzer

In the following, the appropriateness conditions of the genre of a political interview are analysed in more detail.

4.1 Doing appropriateness in the genre of a political interview The dialogue act of a plus/minus-validity claim allows for an investigation of the micro domain of communication regarding the linguistic realization of a communicative intention, and it allows for the analysis of the macro domain of communication regarding the context-sensitive specification of its constraints and requirements anchored to the objective, subjective and social worlds (Fetzer 2002). In the previous section, the constraints of the social world and its interpersonal, interactional and textual systems have been assigned the status of generalized appropriateness conditions, while their context-sensitive specifications in the genre of a political interview are assigned the status of particularized appropriateness conditions. In the Anglo-American context, political interviews have been defined by the constraint of neutralism manifest in the genre’s strict regulations of the turn-taking system (Clayman 1992). The constraint of neutrality has been extended to the coparticipants’ language use, that is to their production, interpretation and ratification of communicative contributions and thus assigned the status of a leitmotif (Fetzer 1999, 2000). The appropriateness conditions of a political interview subcategorize into macro conditions and micro conditions. The former specify the institutional and media-specific constraints and requirements, such as prespecified length and agenda. The latter are defined within the dialogue act of a plus/minus-validity claim, and here within the social world and its constitutive systems: –

–

–

The interactional system is based on the premises of adjacency and turn-taking. – For a political interview to be appropriate, adjacency is particularized to the adjacency pair of question/answer, and the turn-taking system is particularized to interviewer selection. There is the additional particularization that the interviewer asks questions, which the interviewee answers. The interpersonal system is based on the premises of face-wants and participation framework. – For a political interview to be appropriate, face-wants are particularized to neutral face-wants, and the participation framework is particularized to all of the coparticipants having a ratified status. The textual system is based on the Gricean maxims of quality, quantity, relation and manner, and the implicature.

Indexing appropriateness

–

For a political interview to be appropriate, the ratified coparticipants produce and interpret their communicative contributions in accordance with the maxims and implicature constrained by the particularized value of neutrality.

In the following, explicitly realized references to the sociocultural construct of appropriateness are examined in a corpus of 50 full-length political interviews recorded between 1990 and 2004 in Britain (40 interviews) and the United States (10 interviews).6

4.2 Explicit references to appropriateness In the political interviews, all of the explicitly realized references to appropriateness are found in the context of a disruptive setting which is signified by crosstalk, follow-up questions, hesitation markers and attenuation devices. In the data, explicitly realized references are rare. In all of the 11 instances identified, the construct is never referred to in the nominal form but always in the adjectival form appropriate (7 instances) and its negated counterpart inappropriate (4 instances). In all of the scenarios, the adjective modifies a sociopolitical entity, which is referred to by generic nouns and other semantically indeterminate expressions. While the inherently negative adjective inappropriate expresses a strong negative evaluation of the constituent it modifies, e.g., ‘tax people’, ‘language’, ‘thing’, its affirmative counterpart appropriate does not express that strong type of negative evaluation. Because of its embeddedness in negative contexts, the speaker gets in a conversational implicature which signifies that the constituent modified by the adjective appropriate is not quite appropriate, e.g., ‘electoral’, ‘adjustment’, ‘thing’, ‘spend’, ‘stick on’, ‘step’, ‘system’. Since an implicated challenge is not determinate, it can be cancelled, should the communicative need arise. The British data contain only 5 explicitly realized references to appropriateness. All references collocate with sociopolitical-context references, such as ‘electorally erm appropriate’, ‘to make what adjustments may be appropriate’, ‘the appropriate things’, ‘to tax people at the level of ... is inappropriate’ and ‘spend more than was appropriate’. In the framework of the dialogue act of a plus/minus-validity claim, these references are assigned the status of a validity claim indexing the textual system. The US data contain more explicitly realized references to appropriateness, relatively speaking. The majority of the references index the sociopolitical con6. I would like to thank Peter Bull (York, UK) and Gerda Lauerbach (Frankfurt, FRG) for sharing their data with me.

133

134 Anita Fetzer

text, such as ‘we just pick up the next piece and stick it on wherever it’s appropriate’, ‘that was an appropriate step’, ‘is the system appropriate’, ‘that the language (..) was inappropriate’, ‘to be strong an inappropriate language’ and ‘that’s completely erm inappropriate’. This has also been the case in the British data. What is peculiar, however, is the fact that there is one explicitly realized reference which indexes the appropriateness of a contribution as a whole, namely ‘this is an inappropriate question’. In the British political interviews, explicitly realized references to appropriateness are used by all coparticipants, that is by the professional interviewer, by members of the audience acting as interviewers and by the interviewee, as is the case with the following extracts:7 (1) IR .. I understand why it is that it is electorally erm appropriate or necessary for you to adopt all the sort Tory erm targets on the economy ... (2) IR It’s very disillusioning to be told you’re going to get millions of pounds for cleaning, and then you can’t use it as hospitals feel appropriate. (3) IR This is an appropriate question. Do you two think you can work together in the next Congress? (4) IR Well, what they say what they say is that erm the language that you expressed, that House Majority Whip Tom Delay expressed erm was inappropriate .. (5) IR Well, I’ll give you the specif the specific language they’re referring to after that first Florida Supreme Court erm result was erm released, when you said erm the following: “It is not fair to change erm the election laws of Florida by judicial fiat.” That that that they considered that to be strong an inappropriate language. (6) IE ... and it reflects the very proper conservative or conservative party to make what adjustments may be appropriate ... now I think that there are problems ... (7) IE ... it will do so to satisfy itself that it has done the best and the appropriate things it may have quite different ideas to me .... (8) IE ... and we therefore judge that to tax people at that level at fifty thousand is inappropriate and wrong and we set that tax level for earnings ... (9) IE ... it would be simply irresponsible for a government to spend more than was appropriate with a burden of debt ... 7. The goal of the micro analysis is to identify explicitly and implicitly realized references to appropriateness. For this reason, and to improve readability, the transcription follows orthographic standards. IR denotes interviewer, and IE denotes interviewee.

Indexing appropriateness

(10) IE ... erm and that’s what erm but that’s really not what the issue is here Wolf. The issue is, is the system appropriate and proper and constitutional... (11) IE .. that somehow that’s erm a completely erm inappropriate, or unfair, or illegal thing to do. If they think they can swing some of them ...

Extracts (3), (8) and (10) do not display explicitly realized attenuation devices: (3) indexes the appropriateness of the communicative action as a whole thus implicating that the interviewee has the obligation to answer the question. The explicit reference in (8) is a constitutive part of an argumentative sequence introduced by the argumentative markers therefore and we judge; and the conclusion extends the negative evaluation of appropriateness (inappropriate) to the negative ethical norm wrong. Extract (10) is very similar to (8): here the explicitly realized reference to appropriateness is also a constitutive part of an argumentative sequence. The other instances of explicitly realized references to appropriateness co-occur with attenuation devices, such as hesitation markers in (1), (4) and (11), modality markers in (1), (6) and (9), subjectification markers (2), (5) and (8). What is peculiar is the fact that the source of the negative evaluation ‘not appropriate’ is not always clear: in (3) is the speaker the source, and in (2), (4), (5), (6) and (7), the sources are mediated, which reduce the force of the implicit challenge.

4.3 Implicit references to appropriateness Implicit references to appropriateness index the social world’s interactional, interpersonal and textual systems. Like all instances of implicit language use, such as indexical expressions and indirect speech acts, their communicative meaning is a context-dependent and can be cancelled by the introduction of arguments, should the communicative need arise. In spite of this difference, the contexts of implicitly realized references to appropriateness share very similar configurations, namely cross-talk, hesitation markers and attenuation devices. Their referential domain comprises one or more validity claims and their interactional, interpersonal and textual presuppositions, the contribution’s force or its communicative status. While explicitly realized references to appropriateness do not generally invite follow-up question, this seems to be the case with implicitly realized references.

4.3.1 References to the interactional system References to the interactional system index the generalized value of the adjacency pair question/answer and the interactional roles allocated to it. In the data investigated, references to the appropriateness conditions of the interactional system

135

136 Anita Fetzer

are used by the interviewee only, and they co-occur with other contextualization devices, such as the discourse marker well and the interpersonal marker let me. In (12), the IE infringes on the appropriateness condition by formulating a question and directing it at the IR thus signifying that he intends to exploit the particularized value explicated above. The IR ratifies the infringement and the deviation through metatalk, which is functionally equivalent to taking a neutral stance. As a response, the IE supports his attempted introduction of a local deviation at that stage of the interview with the booster ‘that’s very critical’. The deviation is then ratified by the IR through an acceptance. In (13) the IE indexes the appropriateness condition of the interactional system with the metapragmatic comment ‘let me answer your question by another question’ thus postulating the deviation and the employment of the question format, which the IR ratifies through an acceptance: (12) IE Well, let me probe you a little bit further. Are you saying that in the relation to the second car company they will have indicated .... IR Well that’s. Then I’ll answer that question because it’s in terms of another question IE But that’s very critical IR What you’re saying is, what you’re saying is then by way of clarification of what probably means is .... (13) IE Let me answer your question by another question, ... (14) IE One moment. Can I ask you one question? IR Well of course IE Was the story in the Mail on Sunday that you censored a pro Tory poll last week, is that true?

In (14), the appropriateness conditions of the interactional system are indexed twice. The first reference ‘one moment’ signifies a halt in the flow of discourse thus making manifest the particularized value of the question-answer sequences and indexing the introduction of a deviation, while the second reference ‘can I ask you one question’ makes explicit the infringement by spelling out the deviation, which the IR ratifies through an acceptance. Indexing the appropriateness conditions of the interactional system is neither very frequent in the British data, nor in the US data. Far more frequent are references to the interpersonal system in both sets of data, as is examined in the following.

Indexing appropriateness

4.3.2 References to the interpersonal system References to the interpersonal system index the particularized values of neutral face wants and of a ratified participation status. They tend to co-occur with other contextualization devices, such as the discourse markers now or but. Interpersonal-system references are used by IR and IE, and the conventionalized device I’m sorry is used by both of them. References to the appropriateness conditions indexing the coparticipants’ neutral face wants subcategorize into references to the coparticipants’ positive face-wants and the coparticipants’ negative face-wants. While other-references using the first name as in extracts (10) and (25) or other means of creating solidarity as the device ‘you can correct me if I’m wrong’ in extract (26) have the function of signifying positive politeness, both self-references expressing a subjective stance and other-references signifying respect have the function of indicating negative politeness. Prototypical examples for the expression of negative politeness are ‘I’m sorry, I think, in my view, let me, I may say or with all due respect’. All of these devices have the communicative function of expressing subjectification. In (15), the IE infringes on the appropriateness condition of neutral facewants by intruding into other’s territory through the expression of a strong disagreement intensified by the booster just (‘the point you are making there is just wrong’). In (16)–(24) the IR and the IE infringe on the appropriateness condition of neutral face-wants by intruding into other’s territory through the initiation of a follow-up move, which is a question in the case of the IR and an argument in the case of the IE: (15) IE ... with all due respect I really think the point you are making there is just wrong (16) IR .. Now let me take the other four examples ... (17) IE

... if I might if I might just make ....

(18) IE ... and you’ve also through the ballot got to ensure that the workforce has the ability to have a say in whether it should come out or not now I may say that with that ballot provision there ... (19) IR but but forgive me when you put that question to me ... (20) IR ...I’m sorry to keep pressing this one this is fundamentally important ... (21) IE ... it should be up to the workforce to decide whether it takes action irrespective of the circumstances now in my view you have to strike a balance here (22) IE ..... the situation that I think you’re talking about the situation that was raised in your film ...

137

13 Anita Fetzer

(23) IR .... what I do say to you that this will be interpreted as saying in erm my humble judgement that if a car worker ... (24) IR ... but the only the only necessary condition that you’ve given at the moment is as far as I can see is that erm .... (25) IE I’ll tell you why because erm I’m sorry to bring politics into this George but I don’t actually trust the labour party on defence (26) IR ... and all they have to say to themselves you can correct me if I’m wrong (27) IR ... can I make a second point about that in a moment if I may but just stick to erm this particular ...

In (25), there are two implicitly realized references to the appropriateness conditions of the interpersonal system, a reference to the positive face-wants with the first name ‘George’ and a reference to the negative face-wants ‘I’m sorry’ signifying a more severe infringement. While the former infringes on neutral solidarity by signifying closeness, the latter infringes on respect by indicating intrusion, i.e. that the Labour Party is not to be trusted on defence. In (26), there is an implicitly realized reference to the appropriateness condition of neutral positive face-wants signifying non-neutral closeness, and in (27) there are implicitly realized references to the appropriateness conditions of the interpersonal system’s negative face-wants (‘if I may’, ‘can I’) signifying intrusion. The latter is supported by the implicitly realized references to the textual system ‘make a second point about that in a moment’ and ‘but just stick to erm this particular’. Unlike references to the interactional and interpersonal systems, which tend to have wide scope, references to the textual system generally have narrow scope. They are examined in the following.

4.3.3 References to the textual system References to the textual system index the generalized value of the appropriateness condition of the ratified coparticipants producing and interpreting their communicative contributions in accordance with the maxims and implicatures constrained by the leitmotif of neutrality. Since conversational implicatures are context-dependent, the linguistic and social contexts need to be taken into consideration, and the contribution’s sequential status also needs to be accounted for explicitly. This is not to say that there are no linguistic devices which signify a possibly infringement on the textual system’s appropriateness conditions. Prototypical devices are, for instance, hedges (Lakoff 1975), whose job it is to make things more fuzzy and less fuzzy. Hedges have been categorized with respect to their more-fuzzy function and their less-fuzzy function: more-fuzzy hedges are,

Indexing appropriateness

for example, sort of, kind of or something like, while less-fuzzy hedges are, for example, technically/strictly speaking, actually or in real terms (Fetzer 1994). Both more-fuzzy hedges and less-fuzzy hedges represent implicitly realized references to the textual system’s appropriateness conditions. Analogously to references to the interactional and interpersonal systems, hedges co-occur with contextualization devices, such as the discourse markers well, and and now, or the interpersonal markers I believe or I think. More-fuzzy and less-fuzzy hedges are used by both IR and IE. In (28), the IR uses the less-fuzzy hedges ‘in fact’, ‘very clear on this particular point’, ‘precisely’, ‘explicitly’ and ‘clearly’. The less-fuzzy hedge ‘in fact’ indexes the Gricean maxim of quality thus getting in a conversational implicature which reinforces the IR’s argument and his implicit request to the IE to provide a clear and precise answer. In his direct response the IE uses the less-fuzzy hedge ‘obviously’ thus getting in a conversational implicature which signifies that he has answered the IR’s question in an appropriate manner. The other less-fuzzy hedges ‘very clear on this particular point’, ‘precisely’, ‘explicitly’ and ‘clearly’ are used by the IR and all of them index the Gricean maxim of manner thus getting in another conversational implicature which requests the IE to be clear and precise in his answer thereby implicating that the IE’s previous response IE1 has not been appropriate: (28) IR1 I I I would remind you that the chancellor of the exchequer said after the budget to me in fact in this programme, interest rates will be materially lower at the beginning of next year IE1 Well we obviously are forecasting that inflation erm will come down next year IR2 And our interest rates will be materially lower at the beginning IE2 And I believe that, that of next year will happen when inflation erm comes down interest rates will be able to come down as well. But at this moment when inflation has just hit 9.4 % the key thing is to emphasise our determination to pursue our anti inflation policy and to pursue it until inflation comes down IR3 I understand that. But you will understand too why people want to be very clear on this particular point as to whether the chancellor`s prediction after the budget still stands. Let me put it to you precisely. Interest rates will be materially lower at the beginning of next year. Is it still the case that interest rates will be materially lower? Can you explicitly and clearly confirm that? IE3 We’re not in the business and I`m not going to be tempted to forecast

Unlike less-fuzzy hedges, which in the genre of a political interview fulfil an anaphoric function by signifying that a previous contribution has not been in ac-

139

140 Anita Fetzer

cordance with the textual system’s generalized appropriateness conditions, morefuzzy hedges have a cataphoric function and indicate that a constituent, validity claim or contribution may infringe on the textual system’s appropriateness conditions. Like less-fuzzy hedges, more-fuzzy hedges are used by both IR and IE, and they fulfil similar functions. In (29), (30) and (31) the more-fuzzy hedges ‘sort of ’ and ‘something like’ index the textual system’s appropriateness conditions and signify a possible infringement. In general, more-fuzzy hedges have narrow scope and indicate that the constituent over which they have scope may be not appropriate. In (29) the more-fuzzy hedge ‘sort of’ indicates the possibly inappropriate formulation ‘spun against’, and in (30) and (31) the more-fuzzy hedge ‘something like’ indexes a possibly inappropriate figure, that is ‘five point seven billion’ in (30) and ‘fifty percent’ in (31): (29) IE Well, I- I sometimes think we’re more, sort of, spun against than spinning, in the sense that people have this huge thing about it all. But in the end, why not just judge us on what we do? I mean, judge us on what we do ... (30) IR Of course, some taxation went up because of growth in the economy, but you will concede also that, er, if the IFS figures are right and you don’t reject them, that something like five point seven billion of the increase in the tax take came from the increases in indirect taxes, that’s about the equivalent of two pence, erm, in the pound. Do you concede that those taxes went up? (31) IE .... But what I can tell you is, that, for example, as we launched in the manifesto today, there’s an extra billion pounds going into cardiac and cancer services. That will mean, for example, by I think 2004 we will have something like a fifty per cent increase in cardiac surgeons, we’ll have a third increase in cancer surgeons ....

Not only can the local domain of a micro contribution and its constitutive validity claims be looked upon as not appropriate, but so can the macro domain of genre, which is examined in the following.

4.3.4 References to the macro validity claim References to the macro validity claim of the genre of a political interview are a constitutive part of the opening and closing sections. In the topical section, they are extremely rare (Fetzer 2000, 2006). If there are explicitly realized references to genre in that section, they signify that the macro validity claim can no longer be ratified through an acceptance, and that the felicitous performance of the genre

Indexing appropriateness

of a political interview is at stake. This is generally due to multiple infringements on the social world’s appropriateness conditions and to multiply negotiated values of the textual, interpersonal and interactional systems. These multiple and severe local violations reflect on the macro validity claim’s appropriateness conditions, which are at stake. An explicitly realized reference to the macro validity claim at that stage of the interview functions as a regulative device requesting all of the ratified coparticipants to terminate the negotiation of the particularized values and to return to the status quo, as is the case with the following extract: (32) IE No but you and now you are going to get the last word in which you leave an impression in front of your viewers IR No, I have the right to say this is an interview and I’m sure you will recognize it because you understand the nature of political interviews, Mister Heseltine as well as I do

Extract (32) is embedded in a long negotiation-of-validity sequence in which the appropriateness of the IR’s questions is at stake. The IE has already referred to the appropriateness conditions of the social world and its constitutive textual, interpersonal and interactional systems. The IR’s explicitly realized reference to the genre of a political interview is a direct response to the IE’s challenge that the IR has exploited its particularized values by not having adhered to the constraint of neutrality. This is implicated through the IE’s reference to the interpersonal system’s constraint of ratified coparticipants, namely the audience (‘viewers’), the reference to the IR’s right to ask questions only which stands in a fierce contradiction to getting ‘the last word’ and by the reference to the perlocutionary effects of the IR’s communicative contributions, that is ‘leave an impression’. The IE’s challenge has a very strong force and is functionally equivalent to accusing the IR of being unprofessional. This is countered by the IR’s explicit reference to the macro validity claim, that is ‘the nature of political interviews’, thus implicitly requesting the IE to perform his communicative contributions accordingly. In dialogue, implicitly and explicitly realized references to the appropriateness conditions of the social world and to its constitutive interactional, interpersonal and textual systems tend to co-occur thus allowing the coparticipants to express their challenges in a fine-graded manner. Not only can they do face-work when performing a challenge but also specify what particularized values are accepted and looked upon as appropriate, and what particularized values are not accepted and looked upon as inappropriate. Regarding the values of the macro validity claim of genre, these values are generally not open for negotiation, unless a different type of genre is being interactionally organized.

141

142 Anita Fetzer

5.

Conclusions

This contribution has investigated the theory and practice of dialogue and appropriateness. Dialogue has been conceived of as a context-dependent endeavour in which coparticipants negotiate the communicative status of their contributions through the postulation, ratification and acceptance or rejection of validity claims. The theoretical frame of reference is an integrated one: it is anchored to Habermas’s conception of validity claim and its references to the social, subjective and social worlds. The original macro-oriented configuration has been adapted to the contextual constraints and requirements of the micro domain, as is reflected in the ethnomethodological and Gricean approaches to dialogue. In the refined frame of reference, appropriateness is defined as the pillar of the social world and its tripartite configuration of interpersonal, interactional and textual systems. Appropriateness is ‘done’ in dialogue and therefore is both process and product. Doing appropriateness in dialogue means acting in accordance with the contextual constraints and requirements of the social world and its generalized and particularized values of the textual, interpersonal and interactional systems. Against this background, inappropriateness manifests itself in infringements on and deviations from these constraints. The micro analysis examines how appropriateness is done in the genre of a political interview. Particular attention is given to the questions of what functions explicitly and implicitly realized references to the social-world construct of appropriateness fulfil and in what contexts they occur. Explicitly and implicitly realized references to the social-world construct of appropriateness and to its appropriateness conditions tend to co-occur thus intensifying the force of the validity claim. Implicitly realized references to the appropriateness conditions are categorized with respect to their references to the textual, interpersonal and interactional systems where they signify an infringement on the constraints thus getting in a conversational implicature about a possible inappropriateness. References to the appropriateness conditions of micro validity claims tend to attenuate its pragmatic force, while references to the appropriateness conditions of the macro validity claim tend to function as a regulative device. From a linguistic-context perspective, explicit references to appropriateness are realized in attributive constructions, such as the appropriate thing, the appropriate people or an appropriate moment, and they generally co-occur with explicit and implicit negatives, such as not appropriate, inappropriate or doubt + appropriate, with expressions of epistemic modality, such as would be more appropriate or may be appropriate, with expressions of hypothetical meaning, such as if it is appropriate. Implicitly realized references to appropriateness generally co-occur

Indexing appropriateness

with pragmatic and metapragmatic devices, such as the hedges kind of, something like and precisely or clearly, the discourse markers well, but or now, and the subjectification markers if I may say, if I’m not mistaken, I think, if I may say or if I may express my opinion. From a social-context perspective, explicitly and implicitly realized references to appropriateness occur in disruptive settings, negotiation-of-validity and negotiation-of-meaning sequences.

References Akman, V. This volume. “Similar situations.” Akman, V. and Alpaslan, F. 1999. “Strawson on intended meaning and context”. In Modeling and Using Context. 2nd International and Interdisciplinary Conference, Context’99, Trento, Italy, September 1999, Proceedings, P. Bouquet, P. Brezillon, L. Serafini (eds.), 1–14. Heidelberg: Springer. Atkinson, J. M. and Heritage, J. 1984. Structures of Social Action: Studies in Conversation Analysis. Cambridge: Cambridge University Press. Austin, J.L. 1980. How to Do Things with Words. Cambridge: Cambridge University Press. Bakthin, M.M. 1981. The Dialogic Imagination: Four Essays. Ed. by Michael Holquist. Austin, Tex: University of Texas Press Boden, D. and Zimmerman, D. (eds.). 1993. Talk and Social Structure: Studies in Ethnomethodology and Conversation Analysis. Cambridge: Polity Press. Brandom, R.B. 1994. Making it Explicit: Reasoning, Representing, and Discursive Commitment. Cambridge, Massachusetts: Harvard University Press. Brown, P. and Levinson, S. 1987. Politeness: Some Universals in Language Usage. Cambridge: Cambridge University Press. Bunt, H. 1999. “Contextual representation for dialogue management”. In Modeling and Using Context. 2nd International and Interdisciplinary Conference, Context’99, Trento, Italy, September 1999, Proceedings, P. Bouquet, P. Brezillon, L. Serafini (eds.), 77–90. Heidelberg: Springer. Clayman, S. 1992. “Footing and achievement of neutrality: the case of news interview discourse”. In Talk at Work, P. Drew and J. Heritage (eds.), 163–198. Cambridge University Press: Cambridge. Crystal, D. 1997. Dictionary of Linguistics and Phonetics. Oxford: Blackwell. De Beaugrande, R. and Dressler, W. 1981. Einführung in die Textlinguistik. Tübingen: Niemeyer. Ducrot, O. 1984. Le dire et le dit. Minuit. Duranti, A. and Goodwin, C. (eds.). 1992. Rethinking Context. Language as an Interactive Phenomenon. Cambridge: Cambridge University Press. Edmonds, B. and Akman, V. (eds.). 2002. Context in Context. A Special Issue of Foundations of Science 7(3). Fairclough, N. 1995. Critical Discourse Analysis. London: Longman. Fetzer, A. 1994. Negative Interaktionen: kommunikative Strategien im britischen Englisch und interkulturelle Inferenzen. Frankfurt/Main: Peter Lang.

143

144 Anita Fetzer

Fetzer, A. 1999. “Challenging the unspoken: exploiting the ideology in and of political interviews”. In Language and Ideology: Selected Papers from the 6th International Pragmatics Conference. Vol. 1, J. Verschueren (ed.), 98–113. Antwerp: International Pragmatics Association. Fetzer, A. 2000. “Negotiating validity claims in political interviews”. Text 20(4): 1–46. Fetzer, A. 2002. “Communicative intentions in context”. In Rethinking Sequentiality: Linguistics meets Conversational Interaction, A. Fetzer and C. Meierkord (eds.), 37–69. Amsterdam: Benjamins. Fetzer, A. 2004. Recontextualizing Context: Grammaticality meets Appropriateness. Amsterdam: Benjamins. Fetzer, A. 2005. “Negative theme zones in political interviews: a contrastive analysis of German and English turn-initial positions”. In Pressetextsorten im Vergleich. Contrasting Text Types in the Press, A. Chesterman and H. Lenk (eds.), 283–301. Hildesheim: Olms. Fetzer, A. 2006. “‘Minister, we will see how the public judges you”. Media references in political interviews.” Journal of Pragmatics 38(2): 180–195. Fetzer, A. and Meierkord, C. (eds.). 2002. Rethinking Sequentiality: Linguistics meets Conversational Interaction. Amsterdam: Benjamins. Fetzer, A. and Akman, V. (eds.). 2002. Contexts of Social Actions. A Special Issue of Language & Communication 22 (4). Garfinkel, H. 1994. Studies in Ethnomethodology. Polity: Cambridge. Goffman, E. 1974. Frame Analysis. Boston: North Eastern University Press. Grice, H.P. 1975. “Logic and conversation”. In Syntax and Semantics. Vol. III, M. Cole and J.L. Morgan (eds.), 41–58. New York: Academic Press. Gumperz, J. and Levinson, S. (eds.). 1996. Rethinking Linguistic Relativity. Cambridge: Cambridge University Press. Habermas, J. 1987. Theorie des kommunikativen Handelns. Frankfurt/Main: Suhrkamp. Hall, E. and Hall, M.R. 1989. Understanding Cultural Differences. Yarmouth/Maine: Intercultural Press. Halliday, M.A.K. and Hasan, R. 2000. Cohesion in English. London: Longman. Heritage, J. 1984. Garfinkel and Ethnomethodology. Cambridge: Polity Press. Johansson, M. 2002. “Sequential positioning of represented discourse in institutional media interaction”. In Rethinking Sequentiality: Linguistics Meets Conversational Interaction, A. Fetzer and C. Meierkord (eds.), 249–271. Amsterdam: John Benjamins. Kamp, H. and Reyle, U. 1993. From Discourse to Logic: Introduction to Modeltheoretic Semantics of Natural Language, Formal Logic and Discourse Representation Theory. Dordrecht: Kluwer. Kasher, A. 1998. “What is a theory of use?” In Pragmatics: Critical Concepts, A. Kasher (ed.), 85–103. London: Routledge. Lakoff, G. 1975. “Hedges: A study in meaning criteria and the logic of fuzzy concepts”. In Contemporary Research in Philosophical Logic and Linguistic Semantics. Hockney, D. et al. (eds.), 221–271. Dodrecht: Foris. Levinson, S. 1979. “Activity types and language”. Linguistics 17: 365–399. Levinson, S. 1983. Pragmatics. Cambridge: Cambridge University Press. Levinson, S. 1988. “Putting linguistics on a proper footing: explorations in Goffman’s concepts of participation”. In Erving Goffman. Exploring the Interaction Order, P. Drew and A. Wootton (eds.), 161–227. Cambridge University Press: Cambridge. Linell, P. 1998. Approaching Dialogue. Amsterdam: Benjamins.

Indexing appropriateness

Luckmann, T.1995. “Interaction planning and intersubjective adjustment of perspectives by communicative genres”. In Social Intelligence and Interaction: Expressions and Implications of the Social Bias in Human Intelligence, E. Goody (ed.), 175–188. Cambridge University Press: Cambridge. Mey, J. 1996. Pragmatics: an Introduction. Oxford: Blackwell. Prevignano, C. and Thibault, P. (eds.). 2003. Discussing Conversation Analysis. Amsterdam: Benjamins. Sarangi, S. and Slembrouck, S. 1996. Language, Bureaucracy & Social Control. London: Longman. Saville-Troike, M. 1989. The Ethnography of Communication. Blackwell, Oxford. Sbisà, M. 2002a. “Cognition and narrativity in speech act sequences”. In Rethinking Sequentiality: Linguistics meets Conversational Interaction, A. Fetzer and C. Meierkord (eds.), 71–97. Amsterdam: Benjamins. Sbisà, M. 2002b. “Speech acts in context”. Language and Communication 22(4): 421–436. Schegloff, E.A. 1991. “To Searle on conversation: a note in return”. In (On) Searle on Conversation, J.R. Searle et. al. (eds.), 133–128. Amsterdam: Benjamins. Schegloff, E.A. 1996. “Turn organization: one intersection of grammar and interaction”. In Interaction and Grammar, E. Ochs, E.A. Schegloff and S.A. Thompson (eds.), 52–133. Cambridge: Cambridge University Press. Searle, J.R. 1969. Speech Acts. Cambridge: Cambridge University Press. Searle, J.R. 1991. “Collective intentions and actions”. In (On) Searle on Conversation, J.R. Searle et al. (eds.), 401–415. Amsterdam: Benjamins. Searle, J.R. 1995. The Construction of Social Reality. New York: The Free Press. Sperber, D. and Wilson, D. 1996. Relevance: Communication and Cognition. Blackwell: Oxford. Stalnaker, R. 1999. Context and Content. Oxford: Oxford University Press. Vanderveken, D. and Kubo, S. (eds.). 2002. Essays in Speech Act Theory. Amsterdam: Benjamins. Van Dijk, T. 1981. Studies in the Pragmatics of Discourse. The Hague: Mouton. Van Dijk, T. 1998. Ideology: A Multidisciplinary Approach. London: Sage. Vygotsky, L. 1986. Thought and Language. Cambridge, MA: MIT Press. White, P. 2003. “Beyond modality and hedging: a dialogic view of the language of intersubjective stance”. Text 23 (2): 259–284.

145

The appropriateness of questions Annette Becker

University of Frankfurt, Germany

When is a question appropriate? In Recontextualizing Context, Fetzer (2004) distinguishes between “grammaticality” and “appropriateness” of sentences and utterances. Appropriateness, other than grammaticality, is a socio-cultural construct. Whereas the grammaticality of sentences may be judged without consideration of their textual, interpersonal, or interactional context, the appropriateness of utterances is highly context-dependent on all these levels. This is especially apparent in dialogical discourse types like media interviews. This contribution assumes as a working hypothesis that the appropriateness of questions can be approached through an analysis of the subsequent answers. Data is taken from videotaped interviews television journalists conducted with politicians and experts during British election night coverages. Their analysis is based on the pragmatic framework developed by Harris (1991) for the analysis of politicians’ evasiveness, and the multidisciplinary frameworks developed by Lauerbach (2003, 2004) and Becker (2005, 2007) for the analysis of interviewing practices. Comparison of interviews with politicians and interviews with experts reveals clear differences as to what answerers interpret as an appropriate question within the global context of the election night coverage and with respect to the question’s local textual, interpersonal, or interactional context.

1.

Introduction

When is a question appropriate? This paper examines criteria according to which the appropriateness of questions can be judged. Let us recall the prominent example of the election night coverage of the British General Election 2005, where a remarkable interview, which was much discussed on the BBC, made it into the headlines of the British media. It even led to the BBC publicly defending their interviewer. This was the famous – or, as many thought, rather infamous – interview between Jeremy Paxman, influential political interviewer of the BBC, also known by the nicknames “The Grand Inquisitor” or “Mr Kebab” for his relentless questioning style, and George Galloway, leader and founder of the Respect

14 Annette Becker

Party, who had just won the seat in the constituency of Bethnal Green and Bow in London, and defeated Oona King from the Labour Party. Paxman started the interview with the question: “Mr Galloway, are you proud of having got rid of one of the very few black women in Parliament?” To which George Galloway replied: “What a preposterous question. I know it’s very late in the night, but wouldn’t you be better starting by congratulating me for one of the most sensational election results in modern history?“ Paxman repeated his question, Galloway refused to answer it again, and the interview went on in a highly antagonistic way. In the end, Galloway removed his microphone and left the studio. Obviously, Paxman and Galloway had different views as to what was appropriate in the given sociocultural context. And the British audience also had their opinions as to what was appropriate in that context. After the interview, they flooded the BBC with comments in which they either praised or condemned Paxman for his line of questioning. More than a hundred e-mail comments by viewers (and voters) were published on the BBC website, and in the end, the BBC’s Head of Political Programmes, Sue Inglish, publicly defended Paxman against his critics. All in all, both within the interview itself and in its aftermath, the appropriateness of the interviewer’s questions in relation to the sociocultural contexts of the election campaign in general and the BBC’s election night coverage in particular was hotly discussed. So what is it, then, that makes a question appropriate?

2.

Some basic remarks on appropriateness in political media interviews

Within pragmatics, appropriateness has often been explicitly or implicitly equated with felicity conditions of speech acts (van Dijk 1997; Crystal 2003). Research from sociolinguistics as well as research from the ethnography of speaking and also from Critical Discourse Analysis often treats appropriateness as a synonym for acceptability (Gumperz 1982a; Fairclough 1992). Occasionally, appropriateness has also been equated with politeness (Meier 1995), or with linguistic etiquette (Kasper 1997) in general. However, Fetzer (2004) has suggested that appropriateness should be regarded as a superordinate concept, encompassing both felicity conditions (Searle 1969) and face (Brown and Levinson 1987). This view has proven to be more helpful for the analysis of questions and appropriateness, especially in political interviews. But what about the appropriateness of impoliteness? Studies on the systematic use of impoliteness suggest that there are contexts and genres where facethreatening acts seem to be the rule rather than the exception (Culpeper 1996, 2005; Kotthoff 1996; Culpeper, Bousfield and Wichmann 2003). This has also been examined for political media interviews and the different types of face in-

The appropriateness of questions 149

volved here. Jucker (1986) applies Brown and Levinson’s face model to broadcast news interviews, postulating that “what is primarily at issue in news interviews is the interviewee’s positive face” (Jucker 1986: 71). Others, such as Bull, Elliott, Palmer and Walker (1996) and Bull and Elliott (1998) have expanded Jucker’s view, arguing that in media interviews not only the individual face of individual politicians is at stake. Instead, politicians are concerned with three faces: “their own individual face, the face of significant others and the face of the party which they represent” (Bull, Elliott, Palmer and Walker 1996: 271). These three types of face correspond closely to what Weizman (2006) observes regarding roles, interactional roles, membership categorizations, identities and performed identities in news interviews. Within the frequently antagonistic and confrontational genre of the political media interview (Bell and van Leeuwen 1994; Lauerbach 2003, 2004), interviewers are entitled to ask questions that are potentially threatening to their interviewees’ positive or negative face. Not surprisingly, they do not always succeed in obtaining answers to their questions. For instance, several studies have demonstrated that the average amount of non-replies in political interviews is around 30 percent (Harris 1991; Bull 1994). This shows a certain discrepancy as to what is believed to be appropriate by both sides, as well as a genre-specific appropriateness profile for political media interviews, which very often are antagonistic and confrontational, as mentioned above, or even adversarial (Bell and van Leeuwen 1994), as opposed to more cooperative interview genres (for a description of different interview genres see Bell and van Leeuwen 1994). Generally, the appropriateness of questions is highly dependent on their sociocultural context, and especially on the genre or activity type (Levinson 1979) they occur in. Specific rules apply within institutional contexts and genres, such as the different types of media interviews. At the same time, the genre itself may show a wide range of variation, depending on further contextual constraints, such as channel identities, or culture-specific expectations that emerge in cross-cultural analysis (Lauerbach 2003, 2004; Becker 2005, 2007), and last but not least, the sociocultural identities of the interview participants, for instance winners, losers and neutral experts in election night coverages. But even within the same genre, and even regarding the same speech event, opinions of participants and overhearers may differ strongly as to what is to be regarded as an appropriate question, as the Paxman-Galloway interview and the divergent reactions to it show. One of the most interesting questions regarding appropriateness in general and the appropriateness of questions in particular is the seemingly trivial question: “Says who?” Who is to judge whether a question is appropriate? This question is not only relevant because critical linguists and critical discourse analysts, such as Fairclough (1995), have alerted us to always watch out

150 Annette Becker

Type of context Physical Body language Verbal Cultural Group Personal Imagined

What referred to Objects, sensations Gestures, facial expressions, posture, social distance, paralinguistic features Verbal utterances Norms, values, expectations, assumptions of the culture at large Norms, values, expectations, assumptions of the current group Each individual’s private views, attitudes, prejudices, predilections Playacted social scenes, frames, or roles

Channel Sensory Sensory/social Linguistic Cultural Group-dynamic Psychological Imaginative

Figure 1. Types of context (Robinson 2006: 115)

for the agent. It is also crucial because any attempt at an answer reveals much about the dialogic nature of appropriateness. Just as the concept of context, the concept of appropriateness has been notoriously difficult to define without acknowledging its dialogicality. Therefore, a few words about context are in order now.

3.

Some basic remarks on context in political media interviews

When talking about context, researchers occasionally despair at its “maddening elasticity” (Givón 2005: 1). This is mirrored by the fact that the recent new edition of the Encyclopedia of Language and Linguistics (Brown 2006) devotes no less than four different entries to this topic (Clark 2006; Hanks 2006; Jackman 2006; Stainton 2006). An overview is provided by Robinson (2006), who distinguishes between seven different types of context (Figure 1). What is missing even in Robinson’s detailed survey is the institutional dimension which is central for a discussion of context in political media interviews. Here, the conversation analytical (CA) perspective on institutional discourse yields important insights.1 The conversation analytical concept of the dynamic nature of institutions emerged in close connection with the development of the ethnomethodological concept of context. CA research tries to demonstrate how institutions are realized on the micro-level through “talk-in-interaction” (Schegloff 1982). This term has come to replace the term “conversation”, as CA gradually moved away from mundane conversation as its prime object of research and took up the analysis of institutional talk as well when referring to its object of study 1. The following passages are partly based on Becker (1997: 17–20).

The appropriateness of questions

(Drew and Heritage 1992). Up to the seventies, sociolinguistic research focused mainly on the social attributes speakers bring to talk (e.g. age, class, ethnicity or gender). “Personal” context was seen as having an important impact on fundamental linguistic phenomena, which it certainly does, but not to the extent many of those early studies claimed. The social attributes of speakers were seen as invariant, static features and thus the role they played within discourse was rather overestimated. Goffman (1964) was one of the first to consider the influence of “interactive” or “textual” context on the linguistic behavior of participants. He found that the relevance of the attributes mentioned above depends upon and varies with the particular setting in which the talk occurs (Goffman 1964). Generally, it is important to speak of “participant orientation” rather than of “contextual influence”, because context is dynamic, not static. It is always coconstructed by the participants and always collaboratively achieved. The use and interpretation of “contextualization cues” (Gumperz 1982, 1992a/b) is a dynamic process which demands the active participation of all parties. All aspects of verbal or nonverbal behavior, be it lexical choice, prosody, phonology, syntax, choice of register, or dialectal variant, may function as such “contextualization cues” i.e. signals which “enable interactants to make inferences about one another’s communicative intentions and goals” (Drew and Heritage 1992: 8). Drew and Heritage (1992) see strong similarities between Gumperz’ linguistic concept of contextualization cue and Goffman’s sociological concepts of “frame” and “footing” (Goffman 1974, 1979). These concepts also focus on the step-by-step negotiation of situations and relationships between interactants, this negotiation also being almost never conducted explicitly but by subtle and minute changes of conduct. “In combination, they have done much to advance and develop a more complex and dynamic concept of the ‘context’ of interaction” (Drew and Heritage 1992: 9). Utterances derive their sense as actions to quite a large extent from context (Schegloff 1984, 1987), context being not only constituted by the speech event as a whole (macro-level) but also by the immediate environment of an utterance in particular (micro-level). These levels correspond closely to what Brown and Yule (1983) termed “context” and “co-text”. On the micro-level, utterances are interpreted “in terms of whether, or to what extent, they conform to or depart from the expectations that are attached to the ‘slot’ in which they occur” (Drew and Heritage 1992: 12), because the sequential environment of talk provides highly relevant contextual information for participants’ understanding, appreciation and use of what is being said, meant and, most importantly, done in and through the talk (Schegloff and Sacks 1973). In other words, utterances are measured against the preferences operating at a particular slot. A second type of expectations arises from the more general context constituted by the speech event as a whole. For instance, there are many restrictions

151

152 Annette Becker

concerning participants’ range of options of participants with respect to highly formal institutional talk, such as legal discourse (Atkinson and Drew 1979) or news interviews (Clayman and Heritage 2002). Within these settings, turns tend to be pre-allocated: the “turn-taking machinery” (Sacks, Schegloff and Jefferson 1974) of ordinary conversation is suspended. Usually, the member of the institution is in control of the interaction, and the expectations established by turns of the member of the institution are more normative. The degree to which these expectations are met or not met gives rise to activity type specific inferences (Levinson 1979). Initially, CA mainly considered expectations of the first type. But within these “local” preferences, the second type of expectations has been shown to play an important role in the domain of institutional talk (Boden and Zimmermann 1991; Drew and Heritage 1992) In other words, the impact of the setting on the preferences operating is quite considerable. The preferences within highly formal institutional talk may even be diametrically opposed to the preferences of ordinary conversation. It would be one-sided to consider only this type of connection between social actions and context, though. As stated above, context is interactively achieved by the participants. Therefore, utterances and their underlying social actions are connected to context in two ways: they are not only context shaped but also context renewing (Heritage 1984), each utterance contributing to the context for the next utterance. Turn design is not only matched to the local context, i.e. to the speech event and the previous utterance, but, vice versa, procedurally consequential itself, both on the local level and occasionally even with respect to the speech event as a whole, for instance when an interview turns into a confrontation, as it was the case in an interview between former U.S. president George Bush and CBS news anchor Dan Rather in the context of the Iran-Contra affair (Schegloff 1989).

4.

Appropriateness and dialogue

Just like context, appropriateness is a sociopragmatic phenomenon that is, by its very nature, dialogical, as opposed to grammatical correctness, which is a monological concept (Fetzer 2004). The dialogic nature of appropriateness is especially apparent in question-answer exchanges, and even more so in mediated questionanswer exchanges like television interviews. Within conversational question-answer exchanges, it is the individual addressees of questions who ratify questions as appropriate. They do so by measuring their textual, interactional, and interpersonal properties against the appropriateness conditions of the sociocultural context of the speech event, and design their answers accordingly.

The appropriateness of questions

As King and Sereno put it, “(s)ituational demands may restrict or expand the range of utterances deemed appropriate” (King and Sereno 1984: 266). Additionally, conversational maxims as defined by Grice (1975) may play an important role regarding appropriateness within the context of interpersonal relationships: “The conversational imperative of relational appropriateness can be derived from two of Grice’s maxims, Relation and Manner” (King and Sereno 1984: 267). King and Sereno concede “(t)hat relationships are fluid presents no challenge to the process described here, for appropriateness is a judgement comparing an utterance to a range of utterances that the existing relationship and situational definitions will allow” (King and Sereno 1984: 267) Both “any number of linguistic realizations” (King and Sereno 1984: 268) and the content of utterances may serve as the basis of judging utterances as either relationally appropriate or inappropriate. For instance, the initiation of very intimate topics may violate current relationship definitions. The correlation between relationship definition and content can be summed up as follows: Relationship definition is the ground upon which the figure of content is constituted and that content is interpreted against the ground of relationship. If the figure, content, does not fit the interpretative ground of relationship, a new ground (relationship) is inferred to fit the figure. Accordingly, as both a generative and interpretive aspect of communication, content analogues relationship. The central assumption of this view is that communicators mutually assume end expect that the content (figure) will be appropriate to (fit) the relationship (ground). Maintenance of that assumption in interaction accounts for both relational implicature and, thereby, relational change (King and Sereno 1984: 272).2

In mass media interviews, interviewees judge the appropriateness of their interviewers’ questions before they decide whether they are going to answer them at all, directly or indirectly, or whether they are going to evade them or to contest them openly. Other than in private conversations, this is done publicly, with an overhearing mass media audience watching. This audience will also judge the appropriateness of what is being asked (and, of course, what is being answered) but will not necessarily come to the same results, depending on the appropriateness conditions they take as their basis for their evaluation. For instance, interviewers and onlookers who are convinced that political interviews are an antagonistic genre in which it is the task of interviewers to act as “honest brokers” (Bell and van Leeuwen 1994) will generally approve of controversial interviews with potentially face-threatening presuppositions which are difficult to contest.

2. Bracketing corresponds to the original.

153

154 Annette Becker

First Parts Question Request Offer/Invitation Assessment Blame

Second Parts Preferred expected answer acceptance acceptance agreement denial

Dispreferred unexpected answer or non-answer refusal refusal disagreement admission

Figure 2. Preference in adjacency pairs (Levinson 1983: 272) First Parts Question

Second Parts Preferred direct answer

Dispreferred indirect answer rejection of one or more presuppositions of the question, or of its pragmatic force

Figure 3. Preference in question-response pairs

The following section examines to which extent interviewees treat their interviewers’ questions as appropriate or inappropriate, and how this is connected to the sociocultural norms of the genre interview. For this purpose, a coding practice was developed on the basis of some general assumptions and their modifications. These assumptions were developed on the basis of conversation analysis, pragmatics, and cross-cultural media analysis, starting from Levinson’s (1983) summary of the basic preference correlations in adjacency pairs (Figure 2). Preferred second parts are usually delivered immediately and with little communicative effort, whereas dispreferred second parts are usually delivered with dispreference markers, such as pauses, hesitations, items like er, well, you know, actually and more communicative effort. Dispreferred second parts usually trigger activity-type specific inferences, as Levinson (1979) suggests. One potential inference triggered by dispreferred formats could, of course, also be the assumption that the addressee of a first part simply does not believe this first part to be appropriate in the given sociocultural context, for reasons to be specified on the basis of the appropriateness conditions of that context. For questions in interviews, the preference framework for question-response adjacency pairs can be tentatively modified as follows: Direct answers are preferred responses, while indirect answers or the rejection of one or more presuppositions of the question, or of its pragmatic force are usually dispreferred (Figure 3). Inferences as to the assumed appropriateness of a particular question are triggered accordingly, depending on the preferred, or dispreferred format of the re-

The appropriateness of questions

sponse. Although it may be problematic to hypothesize about what people believe on the basis of what they do, one may generally assume (1) that interviewees (IEs) who answer a question directly believe that it is appropriate; (2) that IEs who answer a question indirectly are ambivalent concerning the appropriateness of the question; and (3) that IEs who explicitly reject of one or more presuppositions of an interviewer (IR) question, or its pragmatic force, believe the question to be inappropriate (for a discussion of non-acceptances in context see Fetzer 2006).

Of course, it is possible (4) that IEs who answer questions directly do not believe that it is appropriate but answer it because they do not wish to depart from their role; (5) that IEs who answer a question indirectly are not necessarily ambivalent concerning the appropriateness of the question but either do not know the answer, or, more likely, do not intend to answer it; (6) that IEs who explicitly reject of one or more presuppositions of a IR question, or its pragmatic force, do not believe the question to be inappropriate but either do not know the answer, or, more likely, do not intend to answer it.

Therefore, we need to expand what was said in (1), (2) and (3), assuming (7) that IEs who answer a question directly believe that it is appropriate (or do not wish to depart from their role); (8) that IEs who answer a question indirectly are ambivalent concerning the appropriateness of the question (or are either unable, unwilling, or unable and unwilling to provide the answer); and (9) that IEs who explicitly reject of one or more presuppositions of an IR question, or its pragmatic force, believe the question to be inappropriate (or are either unable, unwilling, or unable and unwilling to provide the answer).

Ratifying an interviewer’s question as appropriate, treating it ambivalently, or treating it as inappropriate corresponds closely with the formal and functional features observed by Harris (1991) in her pragmatic analysis of evasiveness in political interviews, where she distinguishes between direct answers, indirect answers, and responses that contest one or more of the presuppositions of a question, or even its pragmatic force (for further discussion see Bull 1994). Modifications of Harris’ definitions for those three categories were used to distinguish the following types of IE treatment of IR questions:

155

156 Annette Becker

a1 question ratified as appropriate by IE (= direct answer, or an indirect answer from which a “yes”, “no”, or a value for a missing variable in a wh-question can be easily inferred) (10) IR Dr Mawhinney, what do you put this extraordinary defeat down to? IE The British people decided that it was time for a change. (…)

a2 question treated ambivalently by IE (= indirect answer from which a “yes”, “no”, or a value for a missing variable in a wh-question cannot be inferred) (11) IR Who’s your favorite candidate for that leadership contest? IE I think it depends who’s going to come forward. Who’s prepared to put himself or herself into the ring.

a3 question treated as inappropriate by IE (= rejection of one or more presuppositions of IR question, or of its pragmatic force) (12) IR Michael Portillo, are you going to miss the ministerial limo? IE As Brian Mawhinney said, we will wait for the real results.

A corpus of British interviews from the BBC’s coverage of the 1997 General Election was examined to explore the relationship between the properties of IR questions and their ratification by IEs as appropriate or as inappropriate. For the 250 interview questions directed at politicians, the following results emerged (Figure 4). The data show that in interviews with politicians a relatively high percentage of questions were treated either ambivalently or as inappropriate. This differs from the interviews with experts conducted in the course of the very same election night broadcast, where most questions were treated as appropriate. There is

50 40 30 20 10 0

Politicians' reactions to IR questions (n = 250)

a1

44,4

a2

19,2

a3

24,3

combinations

12

Figure 4. Politicians’ reactions to IR questions [%]

The appropriateness of questions

good reason to assume that this difference is due to the fact that interviews with experts and interviews with politicians serve different purposes. Whereas experts are there to co-operatively co-construct the news story, politicians are usually interviewed about topics closely related to their own or their party’s success or failure. At this point, the notion of face (Brown & Levinson 1987), in spite of all the criticism it has been attracting (Fraser 1990; Eelen 2001), plays an important role. The face-threatening potential of questions that immediately concern an interviewee’s public self and public image is much higher than the face-threatening potential of the questions that experts are commonly asked during election night coverages (for discussion of face threats in news interviews see Jucker 1986; Bull, Elliott, Palmer and Walker 1996; Bull and Elliott 1998). Additionally, questions concerning an election result that is not yet clear at the time of the interview, are frequently treated as inappropriate because both politicians and experts try to avoid speculation about unwarranted facts for a variety of reasons. If they belong to the losing party, they try to avoid admitting their defeat. If they belong to the winning party they try to avoid counting their chickens before they are hatched (as one British politician put it in an interview after the 1997 election). And if they are experts they simply try to avoid being too disastrously wrong. However, even within the genre of interviews with politicians, appropriateness judgments differ along the lines of various contextual parameters such as identities and topics.

5.

IE identities and appropriateness

Generally, in the BBC election night coverage, 250 IR questioning turns were directed at politicians of the different parties. 67,6 percent were directed at members of the losing Conservative party, the party that had been in power for 18 years, 22 percent were directed at members of the winning Labour party, 5,6 percent were directed at members of the Liberal Democrats, and 4,8 percent were directed at members of other parties (Figure 5). For the purpose of the present analysis, only the reactions of members of the two major parties were considered and compared with the reactions of neutral experts, who were mostly political scientists, such as Prof. Anthony King. During the election night, 23 percent of all IR questions were directed at members of the winning Labour party, 70 were directed at members of the losing Conservative party and 7 percent at the experts (Figure 6). The surprisingly low percentage of interview questions to experts has to do with the fact that only sequences of talk with at least two question-answer exchanges were counted as interviews, with the result that single questions directed at co-present experts did not qualify.

157

15 Annette Becker

80 60 40 20 0

Politicians' identities (n = 250) 67,6

Cons. Labour

22

Lib. Dem.

5,6

Other

4,8

Figure 5. Politicians’ identities [in %]

80 60 40 20 0

IE identities (n = 242)

Labour Party (Winner)

23

Cons. Party (Loser)

70

Experts

7

Figure 6. IE identities [in %]

60 40 20 0

Reactions of winners, losers and experts (n = 242)

a1

48

a2

19

a3

22

Combinations

11

Figure 7. Reactions of winners, losers and experts [in %]

The appropriateness of questions

100

50

0

Labour (n = 55)

Cons (n = 169)

Experts (n = 18)

a1

54

40

100

a2

15

22

0

a3

15

27

0

Combinations

16

11

0

Figure 8. IE identities and appropriateness [in %]

Obviously, the IRs found 100% of their 242 questions appropriate, within the given sociocultural context, and with respect to their addressees. Otherwise they would not have asked them. However, quantitative analysis reveals that their IEs did not quite agree. Their judgment of the questions’ appropriateness, as inferred from their direct and direct answers, their evasions and explicit rejections, was as follows (Figure 7). As a whole, members of the winning Labour party treated a higher proportion of questions as appropriate than members of the losing Conservative party, whereas the experts treated 100 percent as appropriate (Figure 8). This distribution shows an interesting correspondence to the questions’ topics and subtopics and their face-threatening potential.

6.

Topics and appropriateness

The superordinate topic of all interviewer questions in the data is, expectably, the election result. Within this superordinate topic, most questions are about one of three main topics: the “figures” in general, the “reasons” and the “consequences”. Generally, the distribution of these three main topics within the questions directed at politicians and experts is as follows (Figure 9). How did the interviewees react to these topics? We will leave the experts out for the time being, as they treated all questions as appropriate, and concentrate on the 224 questions directed at members of the winning Labour party (n = 55) and members of the losing Conservative party (n = 169). As to the main topic “figures”, politicians of both parties accepted most questions (Figure 10).

159

160 Annette Becker

80 60 40 20 0

Labour (n = 55) Cons (n = 169) Experts (n = 18)

reasons

37

39

0

consequences

33

38

22

figures

30

23

67

other

0

0

11

Figure 9. IE identities and question topics [in%]

60 40 20 0

L-fig (n = 18)

C-fig (n = 40)

a1

50

55

a2

11,1

12,5

a3

5,6

20

Combinations

33,3

12,5

Figure 10. Main topic “figures” and appropriateness [in %]

80 60 40 20 0

L-reas (n = 20)

C-reas (n = 65)

a1

70

36,9

a2

10

23,1

a3

10

24,6

Combinations

10

15,4

Figure 11. Main topic “reasons” and appropriateness [in %]

The appropriateness of questions

60 40 20 0

L-cons (n = 17)

C-cons (n = 64)

a1

41,2

34,4

a2

23,5

26,6

a3

29,4

32,8

Combinations

5,9

6,2

Figure 12. Main topic “consequences” and appropriateness [in %]

As to the main topic “reasons”, Labour interviewees responded much more willingly to questions regarding the reasons for the results than Conservative interviewees did (Figure 11). And finally, as to the main topic “consequences”, Labour IEs also treated more questions as appropriate (Figure 12). A closer look at the subtopics within the three main topics may serve to shed more light on this. Within the main topic “figures”, the two main subtopics within IR questions were “acknowledgement” and “expectations/reactions”. “Within the main topic “reasons”, the main subtopics of IR questions were “party unity” and “responsibility”. Within the main topic “consequences”, the main topics of IR questions were “party leadership” and “future government”. It is apparent that the threats to the three types of political face are quite different for winners and losers. While topics such as the lack of party unity as a potential reason for a massive loss in voters’ voices, or the replacement of an unsuccessful party leader are highly face-threatening to members of the losing party, winners are quite willing to speculate about the reasons for their success, as soon as the result is confirmed. But why is it, then, that George Galloway, one of the winners of the 2005 U.K. election, contests his interviewers’ question regarding his positive result so harshly? One of Jeremy Paxman’s favorite questioning strategies is spicing his questions with unfavorable presuppositions. In this case, his question “Mr Galloway, are you proud of having got rid of one of the very few black women in Parliament?” suggests racist motivations, not altogether unfounded, because Galloway’s election campaign was partially conducted along such lines. But this is not what a winner likes to hear. Generally, both politicians and experts are careful about discussing results that have not been finally confirmed. But whereas politicians generally avoid comments on early results, experts usually give comments. Occasionally, their

161

162 Annette Becker

replies are hedged by formulations such as “If this is correct …”, but nevertheless they may count as direct or indirect answers. One might argue that the BBC’s first expert interview within the 1997 election night coverage occurs much later than the first politician interview, namely 37 minutes after the beginning, whereas the first politician is interviewed after less than six minutes. However, even at that comparatively late time, no single constituency has declared its result yet. And still, the interviewee, Prof. Anthony King (AK), is willing if not enthusiastic to answer anchor David Dimbleby’s (DD) question regarding the result. Moreover, he even upgrades Dimbleby’s presentation: (13) DD So, there we are, we were still waiting for the first result to come in, it’s ehm three minutes if Sunderland are [going] to keep their word, I haven’t heard any news yet whether they’re going to be any later than that, but uh: (.) in three minutes’ time we should have the first result. Now, Tony King, if our (.) exit poll is correct (.) this is a sensational night (.) we face. AK Oh absolutely, landslide is much too weak a word, this is a // DD // Well what stronger word than landslide do you have? AK I offer you eh the following metaphor: It’s an asteroid hitting the planet and destroying practically all life on earth, (.) it is an extraordinary result there’s been nothing like it, if this is what’s going to happen, since the second World War, there’s really been nothing like it in this century, most notably the total collapse it would appear of the conservative vote. Labour is doing very well, but the Conservatives are doing horrendously badly, as you’ve said a few minutes ago, it looks like being their worst night possibly since the great Reform Act of eighteen thirty-two. DD I should say we’re still talking exit poll talk, we haven’t had any real results in, you have to stay with this programme until we get real results in, and then we can say whether (.) our exit poll is telling the truth. Now let’s go round to some of the (.) interesting (.) uh contests that have been going on ...

In this sequence, it is the interviewer himself who appeals to caution, because the BBC might lose its public face by jumping to premature conclusions.

7.

Conclusion

Of course, this paper tries to avoid premature conclusions, too. However, it may be safely said that the analysis of interviewees’ answers may contribute valuable insights as to the appropriateness of interviewer questions in political interviews in the media. This has been demonstrated on the basis of both theoretical discussion

The appropriateness of questions

and empirical analysis along the lines of an interdisciplinary framework. In a next step, this framework needs to be expanded to take account of further factors such as evaluative linguistic resources, or the different discursive functions of questions, because these factors may also play a role for the face-threatening potential of questions in political interviews, and, accordingly, on the interpretation of questions as either appropriate or inappropriate. Additionally, the appropriateness of answers might be discussed as well. Another step might be the examination of the correlation between the expectations created by different question types such as yes-no questions, wh-questions or alternative questions (Quirk, Greenbaum, Leech and Svartvik 1985) and the frequency in which topically similar questions of these types are answered, evaded or rejected. At the same time, such empirical analysis might lead to a re-evaluation and, finally, re-writing of the felicity conditions for questions formulated by Searle (1969: 66–69), who only mentions “real” questions and “exam” questions but not questions in other institutional contexts such as political interviews.

Acknowledgments I wish to thank Anita Fetzer, Gerda Lauerbach and the anonymous reviewer at Benjamins for their helpful comments on an earlier version of this paper, as well as the participants of the panel “Context and Appropriateness” at the IPrA 2005 for their inspiring feedback and questions.

References Atkinson, J.M. and Drew, P. 1979. Order in Court: The Organisation of Verbal Interaction in Judicial Settings. London: Macmillan. Becker, A. 1997. Pragmatic Aspects of Practitioner-Patient Interaction as a Subtype of Medical Discourse. M.A. Thesis, Frankfurt am Main: Johann Wolfgang Goethe-University. Becker, A. 2005. “Interviews in TV election night broadcasts: A framework for cross-cultural analysis.” In Dialogue Analysis IX: Dialogue in Literature and the Media: Selected Papers from the 9th IADA Conference, Salzburg 2003, A. Betten and M. Dannerer (eds.), 65–76. Tübingen: Niemeyer. Becker, A. 2007. “‘Are you saying ...?’ A cross-cultural analysis of interviewing practices in TV election night broadcasts.” In Political Discourse in the Media: Cross-Cultural Perspectives, A. Fetzer and G. Lauerbach (eds.), Amsterdam: Benjamins. Bell, P. and van Leeuwen, T. 1994. “Political interviews: The adversarial genre.” In The Media interview: Confession, Contest, Conversation, P. Bell and T. van Leeuwen (eds.), 124–177. Marrickville NSW: Southwood Press.

163

164 Annette Becker

Boden, D. and Zimmermann, D.H. (eds.) 1991. Talk and Social Structure: Studies in Ethnometholology and Conversation Analysis. Cambridge: Cambridge University Press. Brown, G. and Yule, G. 1983. Discourse Analysis. Cambridge: Cambridge University Press. Brown, K. (ed.). 2006. Encyclopedia of Language and Linguistics. Second Edition. Amsterdam: Elsevier. Brown, P. and Levinson, S.C. 1987. Politeness: Some Universals in Language Usage. Cambridge: Cambridge University Press. Bull, P. 1994. “On identifying questions, replies, and non-replies in political interviews.” Journal of Language and Social Psychology 13: 115–131. Bull, P. and Elliott, J. 1998. “Level of threat: A means of assessing interviewer toughness and neutrality.” Journal of Language and Social Psychology 17: 220–244. Bull, P., Elliott, J., Palmer, D. and Walker, L. 1996. “Why politicians are three-faced: the face model of political interviews.” British Journal of Social Psychology 35: 267–284. Clark, H.H. 2006. “Context and common ground.” In Encyclopedia of Language and Linguistics. Second Edition, K. Brown (ed.), 105–108. Amsterdam: Elsevier. Clayman, S. and Heritage, J. 2002. The News Interview: Journalists and Public Figures on the Air. Cambridge: Cambridge University Press. Crystal, D. 2003. A Dictionary of Linguistics and Phonetics. Malden, MA: Blackwell. Culpeper, J. 1996. “Towards an anatomy of impoliteness.” Journal of Pragmatics 25: 349–367. Culpeper, J. 2005. “Impoliteness and entertainment in the television quiz show: ‘The Weakest Link’.” Journal of Politeness Research 1: 35–72. Culpeper, J., Bousfield, D. and Wichmann, A. 2003. “Impoliteness revisited: With special reference to dynamic and prosodic aspects.” Journal of Pragmatics 35: 1545–1580. Drew, P. and Heritage, J. (eds.). 1992. Talk at Work: Social Interaction in Institutional Settings. Cambridge: Cambridge University Press. Eelen, G. 2001. A Critique of Politeness Theories. Manchester: St: Jerome. Fairclough, N. 1992. “The appropriacy of ‘appropriateness’”. In Critical Language Awareness, N. Fairclough (ed.), 233–252. London: Longman. Fairclough, N. 1995. Critical Discourse Analysis: The Critical Study of Language. London/New York: Longman. Fetzer, A. 2004. Recontextualizing Context: Grammaticality meets Appropriateness. Amsterdam: Benjamins. Fetzer, A. 2006. Non-acceptances in Context. Essen: LAUD. Fraser, B. 1990. “Perspectives on Politeness.” Journal of Pragmatics 14: 219–236. Givón, T. 2005. Context as Other Minds: The Pragmatics of Sociality, Cognition and Communication. Amsterdam: Benjamins. Goffman, E. 1964. “The neglected situation.” American Anthropologist 66: 133–136. Goffman, E. 1974. Frame Analysis: An Essay on the Organisation of Experience. Boston: Northeastern University Press. Goffman, E. 1979. “Footing.” Semiotica 25: 1–29. Grice, H.P. 1975. “Logic and conversation.” In Syntax and Semantics 3: Speech Acts, P. Cole and Jerry Morgan (eds.), 41–58. New York: Academic Press. Gumperz, J.J. 1982a. Discourse Strategies. Cambridge: Cambridge University Press. Gumperz, J.J. 1982b. “Contextualization conventions.” In Discourse Strategies, J.J. Gumperz (ed.), 130–152. New York: CUP. Gumperz, J.J. 1992a. “Contextualization revisited.” In The Contextualization of Language, P. Auer and A. Di Luzio (eds.), 39–53. Amsterdam, Philadelphia: Benjamins.

The appropriateness of questions

Gumperz, J.J. 1992b. “Contextualization and understanding.” In Rethinking Context: Language as an Interactive Phenomenon, A. Duranti and C. Goodwin (eds.), 229–252. New York: Cambridge University Press. Hanks, W.F. 2006. “Communicative context.” In Encyclopedia of Language and Linguistics. Second Edition, K. Brown (ed.), 115–128. Amsterdam: Elsevier. Harris, S. 1991. “Evasive action: How politicians respond to questions in political interviews.” In Broadcast Talk, P. Scannell (ed.), 76–99. London: Sage. Heritage, J. 1984. Garfinkel and Ethnomethodology. Cambridge: Polity Press. Jackman, H. 2006. “Contextualism in epistemology.” In Encyclopedia of Language and Linguistics. Second Edition, K. Brown (ed.), 128–130. Amsterdam: Elsevier. Jucker, A. 1986. News Interviews: A Pragmalinguistic Analysis. Amsterdam: Benjamins. Kasper, G. 1997. “Linguistic etiquette.” In The Handbook of Sociolinguistics, F. Coulmas (ed.), 374–385. Oxford: Blackwell. King, S.W. and Sereno, K. 1984. “Conversational appropriateness as a conversational imperative.” Quarterly Journal of Speech 70: 264–273. Kotthoff, H. 1996. “Impoliteness and conversational joking.” Folia linguistica 30: 299–325. Lauerbach, G. 2003. “Opting out of the media-politics contract: Discourse practices in confrontational political interviews.” In Dialogue Analysis 2000: Selected Papers from the 10th IADA Anniversary Conference, Bologna 2000, M. Bondi and S. Stati (eds.), 283–294. Tübingen: Niemeyer. Lauerbach, G. 2004. “Political interviews as hybrid genre.” Text 24: 353–398. Levinson, S.C. 1979. “Activity types and language.” Linguistics 17: 365–399. Levinson, S.C. 1983. Pragmatics. Cambridge: Cambridge University Press. Meier, A.J. 1995. “Defining politeness: Universality in appropriateness.” Language Sciences 17: 345–356. Quirk, R., Greenbaum, S., Leech, G. and Svartvik, J. 1985. A Comprehensive Grammar of the English Language. London: Longman. Robinson, D. 2006. “Creating context.” In Introducing Performative Pragmatics, D. Robinson, 103–128. New York: Routledge. Sacks, H., Schegloff, E.A. and Jefferson, G. 1974. “A simplest systematics for the organisation of turn-taking for conversation.” Language 50: 696–735. Schegloff, E.A. 1982. “Discourse as an interactional achievement: Some uses of ‘uh huh’ and other things that come between sentences.” In Analysing Discourse: Text and Talk. Georgetown University Roundtable on Languages and Linguistics 1981, D. Tannen (ed.), 71–93. Washington D.C.: Georgetown University Press. Schegloff, E.A. 1984. “On some questions and ambiguities in conversation.” In Structures of Social Action: Studies in Conversation Analysis, J.M. Atkinson and J. Heritage (eds.), 28–52. Cambridge: Cambridge University Press. Schegloff, E.A. 1987. “Between macro and micro: Contexts and other connections.” In The Micro-Macro Link, J.C. Alexander, B. Giesen, R. Münch and N.J. Smelser (eds.), 207–33. Berkeley: University of California Press. Schegloff, E.A. 1989. “From interview to confrontation: Observations of the Bush/Rather encounter.” Research on Language and Social Interaction 22: 215–240. Schegloff, E.A. and Sacks, H. 1973. “Opening up closings.” Semiotica 8: 289–327. Searle, J.R. 1969. Speech Acts. Cambridge: Cambridge University Press. Stainton, R.J. 2006. “Context principle.” In Encyclopedia of Language and Linguistics. Second Edition, K. Brown (ed.), 108–115. Amsterdam: Elsevier.

165

166 Annette Becker

van Dijk, T.A. 1997. “Context models and text processing.” In Language Structure, Discourse and the Access to Consciousness, M.I. Stamenov (ed.), 189–226. Amsterdam: Benjamins. Weizman, E. 2006. “Roles and identities in news interviews: The Israeli context.” Journal of Pragmatics 38, 154–179.

Cooperative conflict and evasive language The case of the 9–11 commission hearings Lawrence N. Berlin

Northeastern Illinois University, USA

In the aftermath of September 11, 2001, concerns surrounding the efficacy of United States intelligence and security agencies led to the formation of the 9–11 Commission. The Commission was charged to conduct a series of hearings to investigate “the truth” and possible culpability for the security failure. During the hearings, many high-ranking officials of the United States Government – past and present administrations – were called to give testimony. Exchanges between some of those officials (Secretaries of State) and their interlocutor-interrogators can best be described as “evasive”, representing a series of discursive tactics which produce nonlinear patterning between adjacency pairs. Yet, despite evidence that use of indirect, evasive language is not uncommon in courtroom discourse, the construction of evasion within these exchanges, especially along partisan lines, appears to be co-constructed, broadening the notion of cooperation to include complicity and suggest an expanded definition of what can be considered “appropriate”. Thus, in an order of discourse fraught with conflicting agents – interrogators and witnesses (and in this particular case, Republicans and Democrats), witnesses maintain the appearance of cooperation within the conflict by answering questions through changing the direction of the question, mitigating the force of the imposition, and deflecting responsibility from an action that could prove politically embarrassing or damaging. This chapter uses a critical discourse analysis framework to examine the mediated political discourse of these interactions, illustrating how cooperative conflict and its concomitant evasive language necessitate a layered analysis of context.

1.

Theoretical framework

Delimiting context has been both a preoccupation and confounding feature of much work in pragmatics and discourse analysis (Goodwin & Duranti 1992; Halliday 2002b; Widdowson 2004). In attempting to resolve one of the issues involved, Sbisà (2002), examining the situating of speech acts, poses three ques-

16 Lawrence N. Berlin

Figure 1. Critical discourse analysis, within and between layers of context

tions about context. In turn, these questions relate to whether context is “given” or “constructed”, “unlimited” or “limited”, and “objective” or “subjective”. These dichotomies can be instructive when superimposed on a general framework depicted in Janney (2002) that incorporates multiple levels of context. Thus, context in a broad sense can be subdivided into the linguistic context (or cotext), the interactional context, the situational context, and the extrasituational context. I will discuss each of these in turn as separate, yet interacting levels. Then, I will define appropriateness as an intermediary which emerges in the interface between the various levels of context. In my discussion, I will interweave these terms within an approach to critical discourse analysis proffered by Fairclough (1995a, 1995b, 1998; Chouliaraki & Fairclough 1999) (see Figure 1).

1.1 Linguistic context (or cotext) Linguistic context, or cotext, can be defined in Halliday’s (1978, 1984; cf. Halliday 2002b) sense of “language as code”. This level of context (i.e., the system) relates directly to discreet, identifiable constituents in the language and their immediate environments from the perspective of traditional grammatical analysis (Halliday 1984, 1994, 2002a). In Halliday’s own terms, these constituents fall into one of two categories: units (e.g., word, phrase, clause, sentence,…) or classes (e.g., word-class: verbal, nominal, adverbial). Consequently, by definition, linguistic context is limited, given, and objective using Sbisà’s (2002) dichotomous framework. To specify my classification, herein I would add that this level of context relates more to the perspective of the analyst than to the conscious knowledge of participants in the interaction; that is, the analyst chooses a relevant unit of analysis and proceeds to

Cooperative conflict and evasive language 169

analyze the text accordingly. Moreover, taking the stance that the choice of a unit of analysis is relative to the question under investigation (Lincoln & Guba 1985), the examination of linguistic context can take on manifestations beyond the units and classes proposed by Halliday. Indeed, studies abound in the examination of features covering every level of linguistic analysis from intonational contours to discourse markers, from the repetition of keywords and phrases to speech acts. As I am referring in this chapter to interaction, however, it is especially germane to move from a limited, one-dimensional notion of linguistic context (i.e., represented by constituents within a single individual’s talk) to a multidimensional one. Flanders (1970) posits an initiative and a response as basic to interactive dialogue. While Flanders’ suggestion coincides with a more traditional idea of the speaker-hearer relationship in communication, subsequent studies have found it necessary to redefine communicative interaction into a variety of structures and levels of analysis – adjacency pairs, moves, turns, transactions, exchanges, sequences, etc. (see Coulthard & Montgomery 1981; Sinclair & Coulthard 1975). While any of these features could, individually or combined, serve as units of analysis, the ultimate decision of what counts in the linguistic context depends on the scope of the research, thus rendering the defined unit as limited, given, and objective. In this section, I am only referring to interaction in terms of units of analysis; in the following section I will elaborate on the interactional context as interpretive rather than merely structural.

1.2 Interactional context Beyond the linguistic context, an understanding of context in broader terms as socially constituted (Goodwin & Duranti 1992) must take into consideration the scope of language use. “In fact it must burst the bonds of mere linguistics and be carried over into the analysis of the general conditions under which a language is spoken” (Malinowski 1923: 306). In Halliday’s work, this level of context corresponds with his notion of “language as behavior” wherein “code […] has been determined in the course of linguistic evolution by the patterns of its use” (1984: 7). Accordingly, the interactional context is defined by the choices of a speaker to produce a particular meaning and/or effect through the choices of words. I would further subdivide this level of context into two categories: what is indexed (i.e., behind the language or ideational in Halliday’s sense) and what is accomplished (i.e., by the language or interpersonal in Halliday’s sense); the distinction might otherwise be specified as internal, speaker-oriented interactional context and external, hearer-oriented interactional context.

170 Lawrence N. Berlin

In the former, internal context, speaker-oriented choices are what participants carry with them upon entering into a communicative interaction (though the possibility of making new choices cannot be precluded and emerges in reaction to the degree of success a speaker achieves in the interaction). Internal interactional context, then, includes intent, presupposition, entailment, implicature, etc. and differs from its external counterpart in that it is not readily apparent in the text, but interpreted through the pragmatic situation (cf. Janney 2002). In the external interactional context, conversely, hearer-oriented choices are apparent and emerge from the successful or unsuccessful accomplishment of language aims. For example, a speaker may choose to employ certain strategies or tactics (e.g., circumlocution, employing degrees of evidentiality) when believing that his or her original aims are not successful. The ability to engage in what Wittgenstein (1958) has called “language games” demonstrates that speakers have an implicit knowledge of how language works and that language itself has the capacity to accomplish things in the world (Austin 1962; Searle 1969). Speech acts are especially informative at this level of context. In the previous section, I mentioned speech acts as a potential unit of analysis in the linguistic context. However, speech acts were originally conceived not only as extending beyond a mere representation of text, but also as having an actual force. Relative to the discussion of interactional context, speech acts can be said to possess both internal (speaker-oriented) and external (hearer-oriented) elements. The former are represented in the illocutionary force of an utterance; the latter are evidenced in the perlocutionary effect. That is, if I say, “It’s hot in here,” the illocutionary force of the utterance may be either declarative or directive (Levinson 1983; Searle 1969) (illocutionary force). If, in fact, my interlocutor takes my meaning as directive and chooses to open the window, my utterance and all the concomitant requirements determine that my speech act has been successful due to the production of the desired effect through implicature (perlocutionary effect: opening the window). Sbisà (2002) suggests that speech acts are context-changing social actions; as such, they are both indexical of the type and nature of the interactional event and the assumed roles of the interlocutors, and performative1 in the sense that their accomplishment can produce changes in the interaction itself and in the external world. Interactional context, then, is limited in that the functions one can perform 1. Here I intend “performative” to mean producing a change in the external interactional context rather than in Austin’s (1962) notion of performative as a type of speech act. I make this distinction since I am using a broader definition that could include not only Austin’s performatives, but also constatives, directives, etc. in that they all include some future action which can be context-changing (cf. Levinson 1983).

Cooperative conflict and evasive language

with language have been defined. Yet while language as a system is finite, relatively speaking, it is creative and always changing, continuing to allow speakers to find unique ways to present ideas and information within its delimited functionality; therefore, interactional context can also be deemed constructed. The determination of the type of speech act, then, is only interpretable through the reaction obtained. For instance, returning to the above initiative “It’s hot in here,” my interlocutor may respond by saying, “That’s funny; I feel cold.” It is at this juncture that work initiated by Pomerantz (1978) in the response type classification becomes useful, further emphasizing a need to analyze units in terms of all interlocutors within the interaction for a full understanding of human communication, rather than simply attending to one speaker’s turn. Section 2.2 presents a taxonomy of response types as a means of constructing cooperative conflict through evasion. The final dichotomous question Sbisà poses, however, must be split within this level of context between the internal interactional context and its concomitant speaker-oriented choices as subjective (and ultimately only suggestively interpretable, relative to the level of explicitness), and the external interactional context and its concomitant hearer-oriented choices as objective. Within a critical discourse analysis (CDA), the analysis of the discourse occurs at the nexus between the linguistic context and the interactional context (Figure 1). That is, the interface between the structural analysis and the interactional analysis (evocative of Halliday’s language as code and language as behavior) allows for a richer analysis in which the way the language and interaction define each other reciprocally is realized (Chouliaraki & Fairclough 1999; Fairclough 1992, 1995b).

1.3 Situational context Beyond the linguistic context and the interactional context as defined above, there is a physical environment (Malinowski’s “context of situation”) which has substantial import for the communicative interaction. Malinowski proposes that “a statement, spoken in real life, is never detached from the situation in which it has been uttered,” (1923: 307). The physical context of the situation enables (or disables) certain kinds of interaction to take place. In the analysis of language as communication from various disciplinary perspectives, situational context has been a relevant feature in rendering language data comprehensible (e.g., Hymes 1972; Lynch et al. 1983; Mey 1993). In Participant Observation, Spradley (1980) outlines the social situation in terms of a domain which is delimited, minimally, by actors, activities, and places. Though a minimal framework for defining a situational context, Spradley’s elements that define a domain correspond with the prevalent view in many disciplines and methodologies (e.g., CDA) where a dis-

171

172 Lawrence N. Berlin

course is defined by its actors, the activities they are engaged in, and the place wherein the interaction occurs (Chouliaraki & Fairclough 1999). Following the definition forwarded here, then, situational context emerges as limited, objective, and given (Sbisà 2002). While considerations of a broader understanding of context are basic to ethnographic research and to discourse analysis in general, traditional linguistic research often considers them peripheral considerations, hoping to find all the answers in the language alone. Nonetheless, in some circumstances, situational context can dictate the roles of participants, the direction of talk and its sequences, and the outcomes of interactions. In the current chapter, for example, the right of 9–11 Commissioners to ask high-ranking government officials, such as the current Secretary of State, to provide testimony is inherent in the fact that they are respectively functioning as authorized interrogators and summoned witnesses in a legal hearing (NCTA 2004).2 Moreover, the hearing proceeds in the format of a trial and results in the expected question-answer patterning indicative of the situational context (e.g., witnesses don’t ask interrogators to answer probing questions). Another level of analysis in CDA is the analysis of the particular practice or practices. This level of analysis enables the researcher to identify and locate the particular discourse at the nexus of the interactional context and the situational context. Therein, practices are situated in places and among specific co-participants, the roles they assume, and their interaction. Chouliaraki and Fairclough present four main moments of social practice for consideration at this level: “material activity (specifically non-semiotic, in that semiosis also has a material aspect, for example voices or marks on paper); social relations and processes (social relations, power, institutions); mental phenomena (beliefs, values, desires); and discourse” (1999: 61). Corresponding to these, I would associate material activity with elements in the situational context, social relations and processes with elements in the external interactional context, and mental phenomena with elements in the internal interactional context. Discourse, as an entity, however, emerges as both code and behavior and coincides with elements in both the linguistic context and the interactional context, labeled (appropriately) within CDA as analysis of the discourse.

2. Despite the authorization to function as interrogators and witnesses, however, the participants in the present study maintain multiple roles and act accordingly.

Cooperative conflict and evasive language

1.4 Extrasituational context Extrasituational context is inclusive and pervasive, incorporating all other levels of context (see figure 1) and running through them at the same time. To qualify my meaning of inclusive and pervasive, I offer Kumaravadivelu’s (1999) 3-level hierarchical structure of a critical classroom discourse analysis wherein he posits a (socio)linguistic level, a sociocultural level, and a sociopolitical level. I have defined this elsewhere (Berlin 2005) as respectively relating to considerations in the discourse itself, along with its synchronic and local (i.e., sociocultural), and diachronic and global (i.e., sociopolitical) influences. While the three levels can be viewed separately, they can be conceived of as multiple levels within the same construct (not unlike the multiple levels of context discussed here), all influencing and interacting upon each other simultaneously. The first interface (cf. Janney 2002), between the (socio)linguistic and the sociocultural, is exemplified in the participants’ ability to recognize an interaction as an example of a particular type of discourse; furthermore, a single discourse as an example of a particular order of discourse depends upon synchronic and local influences and, thus, concurrent expectations within a society and culture – a cultural domain (cf. Fairclough 1998). Following post-structuralist practice, Foucault calls a cultural domain of knowledge a “discourse.” Discourse, here, should not be confused with its usage in other analytic traditions in which it means simply the flow of conversation, or a text longer than a sentence. Rather, for Foucault, a discourse is a cultural complex of signs and practices that regulates how we live socially. As such, Foucault’s “discourse” has many similarities with Bourdieu’s “habitus” – mostly unthought but still learned ways of thinking, feeling, and acting. (Goodwin & Duranti 1992: 30)

A cultural domain is demonstrable in everyday practices wherein discourse is located. Take, for instance, the discourse associated with the activity of getting on a bus. I have seen foreign tourists in New York baffled by the concept of “exact change” when attempting to gain entry to the vehicle. Though the concept is considered given in terms of local knowledge (Geertz 1983), or the local situational context, further attested to by the print on the side of the bus itself, it is not universal in any sense and can be just as confounding for an Italian tourist in New York as local expectations would be for an American tourist in Rome (e.g., I found it utterly incomprehensible on my first visit to Rome that bus fare could not be paid on the bus and even more confounding that tickets should be purchased at a tobacconist).

173

174 Lawrence N. Berlin

Cultural domains – social practices and their respective discourses – in addition, have historicity (Bakhtin 1981); that is, they come into being and are recognized as iterations of social practices through their replication over time among specific actors engaged in particular activities in a given place (Spradley 1980; Urban 1991). Consequently, the sociopolitical level (Berlin 2005; Kumaravadivelu 1999) – which simultaneously takes on global, ideological, and possibly hegemonic (Fairclough 1998) influences – informs the local. As such, analysis of language and its use should not be conceived of in terms of examining “isolated monologic utterances” (Vološinov), but as discursive practices which “acquire life and evolve historically […] in concrete verbal communication, and not in the abstract linguistic system of language forms, not in the individual psyche of speakers” (1973: 95). This is not to say, however, that understanding of the linguistic and interactional contexts are superfluous, but that they are not completely interpretable in isolation. CDA links the multiple layers of context through the analysis of the conjuncture. Establishing boundaries – albeit fuzzy – for any consideration of the extrasituational context, the analysis of the conjuncture emerges as a specification of the configuration of practices which the discourse in focus is located within. […] Such a conjuncture represents a particular path through the network of social practices which constitutes the social structure. Conjunctures can be more or less complex in terms of the number and range of practices they link together, more or less extended in time and in social space. (Chouliaraki & Fairclough 1999: 61)

Finally, returning for the last time to the productive dichotomies forwarded by Sbisà, extrasituational context presents itself as an anomaly. For the first question – given versus constructed – it is both: given because it is recognized by participants who employ it, but only because it was and continues to be co-constructed by those very participants. Further, it is unlimited in scope, though a single analysis will attempt to limit the context as it situates the discourse as an iteration of a particular type within a particular order. Finally, extrasituational context is objective and subjective in that it is “determined, not by content of the participants’ intentional states, but by relevant states of affairs occurring in the world, of which participants might not even be aware, [as well as comprising] a set of propositional attitudes” (2002: 428).

Cooperative conflict and evasive language

1.5 Appropriateness Appropriateness is relative to context or, more specifically, realized in the interface between the various levels of context. At the interface between the linguistic and the interactional contexts, for example, “participants make contributions, but they do so as required by the accepted purpose or direction of talk. So what they say is designed to fit into some agreed pattern of interaction which is accepted as appropriate for the occasion,” (Widdowson 2004: 52–53). Grice’s (1975) Cooperative Principle can be extremely informative for delimiting this notion of what is appropriate. In the case of courtroom interaction, then, imagine the following exchange: (1) The Crown Prosecutor, Mr. Myers (M), interrogates his witness, Christine Helm (H) (Small 1957) M: Your name, in fact, is Christine Helm? H: Yes, Christine Helm. M: And you have been living as the wife of the prisoner, Leonard Vole? H: Yes. M: Are you actually his wife? H: No. I went through a marriage ceremony with him in Hamburg, but I already had a husband. He’s still alive. […] M: Mrs. Helm, is this a certificate of marriage between yourself and one Otto Ludwig Helm, the ceremony having taken place in Breslau on the 18th of April, 1942? H: Yes. […] M: Mrs. Helm, are you willing to give evidence against the man you have been calling your husband? H: Yes. M: You stated to the police that on the night that Mrs. French was murdered, Leonard Vole left the house at 7:30 and returned at 25 minutes past nine. Did he in fact return at 25 past nine? H: No. He returned at 10 minutes past 10.

In the example above of direct testimony, the interrogator provides sufficient cotext to keep the answers directed on a specific course of inquiry that she intends to pursue. As a result, the contributions of the witness can be said to adhere to the Cooperative Principle by meeting the maxims associated with the categories Quantity, Relation, Manner, and Quality (presuming that the witness is being truthful). This type of question-answer sequence is typical in cases where the wit-

175

176 Lawrence N. Berlin

ness is called by the particular interrogator, with the understanding that answers are rehearsed (i.e., lawyers typically prepare their own witnesses) and expected. Proceeding from the elaboration of context given in the previous sections, then, appropriateness in verbal behavior can be defined by the participants engaging in the interaction and interpreted by those outsiders who may hear, observe, and/or read the transcripts of the interaction. Thus, co-participants use their knowledge of extrasituational context to recognize and make use of the situational context and its concomitant discourse to proceed. They possess their own intentions (i.e., aims they wish to achieve) in the internal interactional context; they select roles relative to their interlocutors (external interactional context) and words (i.e., manifestations of those intentions through words) in the immediate linguistic context. Outsiders3 use their knowledge of extrasituational context, their knowledge and/or perceptions of the situational, interactional, and linguistic contexts to interpret the discourse. As such, what may be deemed appropriate or inappropriate performance in interaction is defined locally by pre-established and reified (albeit arbitrary) norms of linguistic and sociocultural behavior. Following Sbisà (2002), the evaluation of what is appropriate (or inappropriate) must be separated from the notion of whether an utterance is truthful or not (cf. Fetzer 2005). The distinction is exemplified in Janney (2002) – a study relevant to this chapter – in his analysis of the O. J. Simpson trial. Janney contends that the nature of a trial forces participants to interact in specific ways; as such, the interpretation by interrogators that defendants or witnesses are being evasive or non-responsive is directly related to their individual intentions in the communication – interrogators want to get answers to their questions (typically corresponding to the way the questions are framed) and witnesses, especially when being cross-examined, are compelled to answer, but want to avoid self-incrimination. Thus, a witness will have to carefully weigh what she says and how she says it, as in example (2).

3. Outsiders as used here can include casual observers, overhearers (Bell 1984), researchers, etc. While I will not attempt an elaborate scientific distinction here, casual observers may be thought of as those who infer something from an interaction, but don’t necessarily hear anything (e.g., Those two people look like they’re having an argument). Overhearers are defined by Bell as fulfilling many roles of varying degrees of involvedness, such as those immediately present and directly engaged in and by the interaction (e.g., a child at a parent-teacher conference, jurors at a trial), those present but not directly engaged in the interaction (e.g., an audience at a performance, eavesdroppers), or those not present (e.g., people listening or watching some type of broadcast, live or pre-recorded). Researchers, then, are those of us who attempt to analyze, interpret, and/or somehow situate interactions for a variety of purposes, usually associated with attempting to extend knowledge.

Cooperative conflict and evasive language

(2) Sir Wilfrid Robarts (R), Barrister for the Defendant, cross-examines Christine Helm (H) (Small 1957) W: However, Frau Helm, it will appear, when you first met the prisoner in Hamburg, you lied to him about your marital status? H: I wanted to get out of Germany, so= W: You lied, did you not? Just yes or no, please. H: Yes. W: Thank you. And subsequently in arranging the marriage, you lied to the authorities? H: I, um, did not tell the truth to the authorities. W: You lied to them? H: Yes. W: And in the ceremony of marriage itself, when you swore to love and to honor and to cherish your husband, that too was a lie? H: Yes. W: And when the police questioned you about this wretched man who believed himself married and loved, you told them… H: I told them what Leonard wanted me to say. W: You told them that he was at home with you at 25 minutes past nine and now you say that was a lie? H: Yes, a lie. W: And when you said that he had accidentally cut his wrist, again you lied? H: Yes. W: And now today you’ve told us a new story entirely. The question is, Frau Helm, were you lying then, are you lying now, or are you a chronic and habitual LIAR?

It is at this juncture that the Cooperative Principle runs somewhat afoul. The same witness in (1) is being cross-examined by opposing counsel. Personal aims (e.g., the desire to avoid self-incrimination) may cause the witness to violate or flout a maxim. As a result, the analysis of the practice (i.e., the interface between the interactional and situational contexts) would determine that the participants’ contributions, driven by personal aims and awareness of the levels of context and surpassing any attempts to cooperate in a completely forthright and/or truthful way, can still be considered appropriate. While Janney refers to the typically “unequal balance of power in favor of the questioner in courtroom examinations” (2002: 459), the current study is not equivalent as the multiple levels of context dictate. In the case of the 9–11 Commission Hearings (9–11 Commission 2004; NCTA 2004), not only do the participants maintain their various statuses in performance of the interaction, but

177

17 Lawrence N. Berlin

the public portion was televised, adding layers to the participants’ contributions and simultaneous monitoring of their performance as their overhearers included a very attentive audience. To that end, the claim I am making that witnesses engaged in the use of evasive language relates directly to the interface of the many levels of context and causes them to perform in ways that were not completely open, direct, or truthful. For example, in the analysis of the practice (i.e., between the situational and interactional contexts), the interrogators do not have a position of dominance by default (as they would in a traditional courtroom situation) since the witnesses share the further distinction of being Secretaries of State. In their dual roles, they are both attempting to avoid implicating themselves or the administrations they represent as possessing any degree of culpability, yet are being treated with a relatively high degree of deference as due someone in their position.4 In my analysis, then, I label the interactions as evasive and interpret that evasion within CDA as co-constructed.

2.

Analysis of the discourse

The text examined for the current study consists of transcripts from the 8th and 9th Public Hearings of the 9–11 Commission, specifically the hearing on counterterrorism policy on March 23, 2004, and the hearing on national security on April 8, 2004. In the former, two previous U.S. Secretaries of State, Madeleine Albright and Colin Powell, were questioned by the commissioners on matters of diplomacy. In the latter – and the one which I focus on primarily – former National Security Advisor to the President and the current Secretary of State, Condoleezza Rice, was questioned on matters of national security surrounding the 9–11 attacks on the United States.

2.1 Hearing structure Since the 9–11 Commission Hearings represent a government investigation, they tend to follow typical patterning of Congressional Hearings. With few exceptions (e.g., the personal statement), the patterning within the interaction between the 4. The dual statuses of the Secretaries of State-witnesses can be detected throughout the transcripts as many interrogators – regardless of political affiliations – may exhibit bouts of (1) cooperation and obsequiousness toward the witnesses; (2) facilitation in the production and completion of question-answer sequences; (3) acceptance of responses that directly answer or completely avoid their questions; and (4) grandstanding for the audience and cameras within the same interaction.

Cooperative conflict and evasive language

• Administration º Introduction º Instruction • Oath • Statement • Interrogation º Opening º Question {Orientation} – Answer/Non-Answer º {Admonition – Rebuttal} º Closing

Figure 2. 9–11 Commission hearing structure

interrogator and the witness also flows similarly to most courtroom interactions (see Figure 2). Since I’m primarily concerned with the structure and nature of cooperative and co-constructed evasive talk, I focus my attention on the interrogation sequences, specifically the question-response adjacency pairs. Moreover, I present a taxonomy of response tactics that form the basis for evasion and may extend the notion of what may be considered appropriate within communicative interaction, given the particular type and order of discourse.

2.2 Response tactics Responses to initiatives can initially be divided into two broad categories (cf. Pomerantz 1978). In the first case, the respondent provides information upon request; these responses can loosely be termed answers. In the second case, the respondent clearly does not provide the information sought; I refer to this category as non-answers. In the construction of evasion, though, non-answers cannot be equated with non-responsive behavior (Janney 2002); in fact, it is the very nature of evasion that presumes the witness to be attempting to act responsively while not providing the information requested. Likewise, answers as presented herein (with the exception of direct answers) can functionally aid in the co-construction of evasion, assuming the cooperation between interrogator and witness, through opting out, flouting, and/or violating the maxims of the Cooperative Principle (Grice 1975) (see Table 1). Both, answers and non-answers, can be deemed appropriate in the analysis of the practice where witnesses may choose a course of action relative to their own needs in the interaction.

179

10 Lawrence N. Berlin

Table 1. Response tactics RESPONSE TYPE Answer • Direct Answer • Disavowed Knowledge • Elaboration • Mitigation Non-Answer • Assertion of Platform • Reassignment of Culpability or Responsibility • Restructure • Clarification • Partial Answer • Repetition

CODE

MAXIM FUNCTION

DIR DIS ELA MIT

Opt Out Quality Flout Quantity/Clash Manner Violate Manner

APL REA STR CLR PTA REP

Flout Quantity/Flout Manner Flout Relation/Clash Quality Flout Relation/Opt Out Quality Flout Relation Flout Quantity Violate Manner

2.2.1 Answers The first and most basic answer type is the direct answer (DIR). It represents full adherence to the Cooperative Principle where interlocutors are presumably working together toward the same goal. Direct answers frequently emerge in testimony where the interrogator provides sufficient linguistic context for the witness to supply a simple “yes” or “no”, as in example (1); nonetheless, they can also appear in a cross-examination, as in example (2). In the case of the 9–11 Commission Hearings (9–11 Commission 2004; NCTA 2004), direct answers were most often observed in interactions between members of the same political party (i.e., Democrat interrogator-Democrat witness or Republican interrogator-Republican witness) as can be seen in examples (3) and (4). (3) Gorelick (Democrat) Interrogation of Albright (Democrat) [Q 3] MS. GORELICK: […] You issued a demarche or a warning to the Taliban before the Cole, saying that you would hold, or the U.S. government would hold the Taliban responsible for any harm to Americans, is that correct? [A 3/DIR] MS. ALBRIGHT: We did, yes. [Q 4] MS. GORELICK: And -- and after the Cole, you -- you, in answer to a question from -- from Secretary Lehman said -- or maybe it was Congressman Roemer, you said, well, we didn’t know, by the time we left office, you didn’t know that the attack on the Cole was the responsibility of Bin Ladin. Is that correct? [A 4/DIR] MS. ALBRIGHT: That is correct.

Cooperative conflict and evasive language

(4) Lehman (Republican) Interrogation of Rice (Republican) [Q 2] MR. LEHMAN: […] First, during the short or long transition, were you told before the summer that there were functioning al Qaeda cells in the United States? [A/MIT 2] MS. RICE: In the memorandum that Dick Clarke sent me on January 25th, he mentions sleeper cells. There is no mention or recommendation of anything that needs to be done about them. And the FBI was pursuing them. And usually when things come to me it’s because I’m supposed to do something about it, and there was no indication that the FBI was not adequately pursuing the sleeper cells. [Q 3] MR. LEHMAN: Were you told that there were numerous young Arab males in flight training, had taken flight training, were in flight training? [A/DIR 3] MS. RICE: I was not. And I’m not sure that that was known at the center. [Q 4] MR. LEHMAN: Were you told that the U.S. marshal program had been changed to drop any U.S. marshals on domestic flights? [A/DIR 4] MS. RICE: I was not told that. [Q 5] MR. LEHMAN: Were you told that the red team in FAA, the red teams, for 10 years had reported their hard data that the U.S. airport security system never got higher than 20 percent effective and was usually down around 10 percent for 10 straight years? [A/DIR 5] MS. RICE: To the best of my recollection, I was not told that. [Q 6] MR. LEHMAN: Were you aware that INS had been lobbying for years to get the airlines to drop the Transit Without Visa loop hole that enabled terrorists and illegals to simply buy a ticket through the Transit Without Visa waiver and pay the airlines extra money and come in? [A/DIR 6] MS. RICE: I learned about that after September 11th. [Q 7] MR. LEHMAN: Were you aware that the INS had quietly, internally halved its internal security enforcement budget? [A/DIR 7] MS. RICE: I was not made aware of that -- I don’t remember being made aware of that, no. [Q 8] MR. LEHMAN: Were you aware that it was the U.S. government established policy not to question or oppose the sanctuary policies of New York, Los Angeles, Houston, Chicago, San Diego, for political reasons, which policy in those cities prohibited the local police from cooperating at all with federal immigration authorities?

11

12 Lawrence N. Berlin

[A/DIS 8] [Q 9]

MS. RICE: I do not believe I was aware of that. MR. LEHMAN: Were you aware -- to shift a little bit to Saudi Arabia, were you aware of the program, that was well-established, that allowed Saudi citizens to get visas without interviews?

[A/DIR 9] MS. RICE: I learned of that after 9/11.

In example (3), Jamie S. Gorelick, a Democrat, is questioning Madeleine Albright. In example (4), John F. Lehman, a Republican, is questioning Condoleeza Rice. These same-party exchanges, then, operate cooperatively to the extent that both interlocutors adhere to the Cooperative Principle and appear to have common goals. Especially in the case of example (4), the exchange tends to allow for more questions where the interrogator provides sufficient context to enable the witness to provide direct and rather short answers, adhering closely to the maxims of the Cooperative Principle. The second answer type is disavowed knowledge (DIS). In this type, the respondent claims to be incapable of providing the requested information by negating the perceptual verb (i.e., a reported verb, such as tell or a cognitive verb, such as know) in the initiative question, also in (4). Within the cooperative framework, disavowed knowledge allows for the possibility of opting out of the maxim of quality in that it is unclear whether or not the respondent is being completely truthful, merely that the identity of trustworthiness is established for the respondent through the willingness of the interrogator to accept the answers. This particular tactic is very familiar in political discourse where politicians often make use of disclaimers, such as “I am unaware of that” or “I have no knowledge of that” to avoid liability in a given situation. In the situational context of the 9–11 Hearings, the tactic may be further encouraged by the presence of an audience. Disavowing knowledge, then, may be said to be doubly warranted as a response choice in the analysis of the discourse (i.e., both political and forensic). As such, use of the tactic can be claimed appropriate and cooperative, despite the potential flouting situation. What is especially interesting in this exchange is the way Lehman poses the questions. His questions provide a good deal of orientation, leading the interrogation in a way that allows Rice to give short answers which maximize the appearance of cooperation. The continuity and fluidity in the exchange evince an acceptance on the part of the interrogator with the responses he receives, even when his interlocutor denies possessing any knowledge of the information he is seeking. It may also be worth noting that this type of exchange is most typical between participants from the same political party. Additionally, another feature that distinguishes same party from opposing party interrogative discourse is the

Cooperative conflict and evasive language

total number of questions: interrogators from the same political party as the respondent tend to average more questions overall, given the same time constraints as their counterparts from the opposing party. The next answer type where the question is clearly being answered is mitigation (MIT). Through mitigation, the respondent violates manner by interjecting a degree of ambiguity suggesting either that the force of the question is being given too much weight and/or that the culpability in the answer should not be given too much weight. In example (4), for instance, Rice answers Lehman’s question [Q2] “were you told…” by first lessening the impact of the reporting verb “tell” by changing it to “mention”; she then goes on to mitigate the force even further by adding that there was no accompanying call to action (“There is no mention or recommendation of anything that needs to be done…”). The final answer type is elaboration (ELA). When a respondent engages in elaboration, he or she flouts the maxim of quantity by providing more information than is necessary. The respondent also runs the risk of clashing with the maxim of manner through prolixity (i.e., long-windedness). Though the respondent does in fact answer the question, he or she goes on to extend the response; consequently, considering the high political position of the interlocutors, elaboration may be an attempt to take control of the text and its direction in order to establish or maintain it as “the dominant one in a given social domain, and therefore [establish or maintain] certain ideological assumptions as commonsensical” (Fairclough 1989: 90). An example of this type of strategy is found in (5) where Rice responds to Kean’s question, but extends her turn by taking the discourse in a direction of her choosing, bringing in private information to continue to advance the perceived trustworthiness of her testimony. (5) Kean (Republican) Interrogation of Rice (Republican) [Q 2] MR. KEAN: I’ve got a question now I’d like to ask you. It was given me by a number of members of the families. Did you ever see or hear from the FBI, from the CIA, from any other intelligence agency any memos or discussions or anything else between the time you got into office and 9/11 that talked about using planes as bombs? [A/ELA 2] MS. RICE: Let me address this question because it has been on the table. I think that concern about what I might have known or we might have known was provoked by some statements that I made in a press conference. I was in a press conference to try and describe the August 6th memo, which I’ve talked about here in my opening remarks and which I talked about with you in the private session. And I said at one point that this was a historical memo, that it was not

13

14 Lawrence N. Berlin

[DIS 2]

[MIT 2] [REA 2]

based on new threat information, and I said no one could have imagined them taking a plane, slamming it into the Pentagon -I’m paraphrasing now -- into the World Trade Center, using planes as a missile. As I said to you in the private session, I probably should have said “I” could not have imagined, because within two days, people started to come to me and say, “Oh, but there were these reports in 1998 and 1999, the intelligence community did look at information about this. “ To the best of my knowledge, Mr. Chairman, this kind of analysis about the use of airplanes as weapons actually was never briefed to us. I cannot tell you that there might not have been a report here or a report there that reached somebody in our midst. Part of the problem is -- I think Sandy Berger made this point when he was asked the same question -that you have thousands of pieces of information, […]

Using several answer strategies, evasion is achieved through the cooperation of both interrogator and witness in the interactional context. The analysis of the discourse and the analysis of the practice evince a copasetic relationship analogous to a defense attorney and defendant or a prosecutor and plaintiff – a partnership that reifies political party lines. In (5), Thomas H. Kean, the Chair of the 9–11 Commission, first allows Rice to go on at length, countering a particular accusation made against the Bush administration, thereby usurping the dialogue and forwarding a particular point of view. Subsequently, however, she disavows all knowledge of the claim – another way to assuage the face threat of culpability in the 9–11 attacks. She then proceeds with this portion of the exchange through mitigation, suggesting that an inordinate amount of information made it impossible to predict the attacks and thus inconceivable that anyone should hold such an expectation.

2.2.2 Non-answers Non-answers may occur for a variety of reasons. The witness may find the question undesirable, unanswerable, or irrelevant from his or her perspective. In all circumstances, therefore, the non-answer response does not provide the complete information sought by the interrogator. By offering non-answers, then, witnesses epitomize the multi-directionality of interaction, complicating the traditional speaker-hearer model where focus is typically given to the initiative and exemplifying the free will of the respondent to utilize a full range of possibilities in the response. Non-answers also seem to highlight the conflictive nature of the adversarial process within the situational context. That is not to say, though, that

Cooperative conflict and evasive language

the interaction itself is necessarily uncooperative or inappropriate in a theoretical sense. Indeed, especially in same party interactions, the interrogators readily accept non-answers and move on to another question or close the interrogation, creating the appearance that both interlocutors share common aims in the interaction. The first type of non-answer, the assertion of platform (APL) is similar to elaboration in that the respondent flouts the maxim of quantity. In example (6), the prolixity of this lengthy turn, however, differs from mere elaboration in two ways. First, it does not respond to the question asked; rather, in extending the linguistic context, it completely avoids providing the information requested. Second, it flouts the maxim of manner through aggrandizement. Indeed, Madeleine Albright, the former Secretary of State under the Clinton Administration, goes on at length to show her own dedication and tireless efforts while not touching on the question at all. She takes the floor from her fellow Democrat, Timothy J. Roemer, by detailing the many tasks she performed while in office with regard to handling potential terrorist threats. It is not until prompted once again (Q 3’) that she finds an answer to the question she previously framed as unanswerable, adding the caveat that “it was hard to quantify.” (6) Roemer (Democrat) Interrogation of Albright (Democrat) [Q 2] MR. ROEMER: […] Isn’t it enough at this point to say al Qaeda did it, and respond in that kind of way, either in December, or certainly in the months that come after your administration? [N-A/STR 2] MS. ALBRIGHT: Well, I think the real question is to try to figure out what really did happen, and when we left office we did not have all the answers to it, and as you point out, there were numerous investigations. I myself called the President of Yemen to help us in this issue and to press for additional investigations. [N-A/REA 2] I think the results came after we were out of office and I would have hoped that action could have been taken. But there was no definitive action of any kind at the time that we left office. [Q 3] MR. ROEMER: In terms of the time that you spent as secretary of State on terrorism -- we’ll have Secretary Powell follow you -- what percent of your time, if you can give us a rough estimation, did you spend? You had Middle East peace, you certainly were one of the driving forces in being a hawk with respect to Kosovo and

15

16 Lawrence N. Berlin

[N-A/APL 3]

[Q 3’]5 [A 3’]

using our military there. What percent of your time can you best estimate that you spent on counterterrorism policy? MS. ALBRIGHT: It’s very hard, Congressman, to give you an exact estimate. But I can tell you what I did, which is every morning, when I came into my office, I obviously read the intelligence, but I also met with the assistant secretary for security. I had changed the standard practice and named a law enforcement officer to that job, David Carpenter, who was a retired Secret Service agent, and so I had a real expert dealing with it. We spent whatever time was necessary in the morning, in order to go over the threats. Then either I or Ambassador Pickering, depending upon who was in town, went to the small meetings that took place on counterterrorism issues. […] MR. ROEMER: Can you guess at all? Twenty percent? Fifty percent? MS. ALBRIGHT: I would probably say somewhere about 35 percent, because it was something that was constant, and it was very hard to quantify.

In another example of assertion of platform (7), Colin Powell, the Secretary of State during the first George W. Bush administration (i.e., 2001–2004), also responds to his interrogator without answering while going into some detail. In this particular case, Powell’s aggrandizement is not self-oriented, but other-oriented in that he depicts President Bush’s 9–11 response team as rational men who were simply considering all options in response to the attacks. In the example shown, his interrogator, Richard Ben-Veniste, indicates through various devices in the linguistic context that he finds Powell’s response evasive and uncooperative (e.g., repetition of the question in Q 3’ and Q 3’’), but he ultimately accepts that the interrogation is over by deferring to the time restraints and the call by the Chairman to conclude (“I’m out of time and I’m going to listen to my chairman.”). (7) Ben-Veniste (Democrat) Interrogation of Powell (Republican) [Q 2] MR. BEN-VENISTE: = Secretary Wolfowitz’s position as whether or not we ought to attack Iraq. Is it not the case that he advocated for an attack on Iraq? [A/DIS 2] SEC. POWELL: He presented the case for Iraq, and whether or not it should be considered along with Afghanistan at this

5. Prime markers (’, ’’, ’’’, etc.) following coding in the transcripts indicate uptake of either the same question by the interrogator or the same answer by the witness.

Cooperative conflict and evasive language

[Q 3]

[N-A/REA 3]

[Q 3’] [N-A/STR 3’] [Q 3’’] [N-A/REP 3’’] [N-A/APL 3’’]

time. I can’t recall whether he said instead of Afghanistan. We all knew that Afghanistan was where al Qaeda was. MR. BEN-VENISTE: But was there any concrete basis upon which that recommendation was founded in your view to attack Iraq for 9/11? SEC. POWELL: Secretary Wolfowitz was deeply concerned about Iraq being a source of terrorist activity. You will have a chance to talk to him directly about =. MR. BEN-VENISTE: I’ve asked for your view, with all due respect, Secretary Powell. SEC. POWELL: With all due respect, I don’t think I should characterize what Mr. Wolfowitz’s views were. MR. BEN-VENISTE: No, I asked for your view. In your view =. SEC. POWELL: My view =. MR. BEN-VENISTE: = was there a basis? SEC. POWELL: My view was that we listen to all the arguments at Camp David that day. And Mr. Wolfowitz felt that Iraq should be considered as part of this problem having to do with program, and he considered -- he wanted us to consider whether or not it should be part of any military action that we were getting ready to take. We all heard the argument fully. We asked questions back and forth. And where the President came down was that Afghanistan was the place that we had to attack, because the world and the American people would not understand if we didn’t go after the source of the 9/11 terrorists. MR. BEN-VENISTE: I’m out of time and I’m going to listen to my chairman.

The second type of non-answer is the reassignment of culpability or responsibility (REA). Again, in the interrogation examples with Albright and Powell – (6) and (7), respectively – the attempt to redirect the interaction is evident. Reassignment, though, tends to flout the maxim of relation while clashing with the maxim of quality. In Albright’s case, her interrogator – another Democrat – provides an opportunity for evasion by offering the choice of “in December, or certainly in the months that come after your administration”. Albright takes this opportunity for giving the outward cooperation in the interaction and puts the blame for inaction on the Bush administration by stating that “the results came after we were out of office.” The reason her contribution is classified as a non-answer (and uncooperative in a Gricean definition), however, is that she has already restructured the origi-

17

1 Lawrence N. Berlin

nal question (the third type of non-answer). Her choice to seize an opportunity provided by the interrogator, then, does not provide an adequate response to his question, but to her own. A difference that emerges along party lines, however, lies in whether the interrogator chooses to accept or reject the response. Interestingly, the exchange between same party interlocutors leads to acceptance as the interrogator simply proceeds to the next question as we see with Roemer in (6). By contrast, the exchange between opposing party interlocutors, as in (7), is not immediately accepted and it takes Powell several turns, each attempting a different evasive tactic, before Ben-Veniste decides to break off the interrogation due to time constraints. Returning to further discussion of the restructure (STR), it is obvious that the witness-respondent flouts the maxim of relation, but does not do so in order to seemingly preserve quality (the definition of a clash); therefore, the restructuring of the question (or answer) opts out by providing a truthful answer, just not the one that was sought by the interrogator. In opposing party interaction, such as in (7), the restructure can lead to direct confrontation as when Powell attempts to use this tactic by stating “I don’t think I should characterize what Mr. Wolfowitz’s views were.” Ben-Veniste does not accept this response and repeats his original query after correcting the witness: “No, I asked for your view.” When deftly handled, however, the interrogator may accept the non-answer as can be seen in (8) where the same interrogator ultimately allows the restructure given by Rice (“Now, the question is,…”) as evidenced in his first “Thank you” in the linguistic context of the interaction. His acceptance only occurs, however, after she employs a variety of response strategies before finally disavowing knowledge. Thus, while he allows her to go on at length, she does, in fact, respond to his repeated question (Q 1’’’) about whether she told the President, albeit unsatisfactorily in terms of obtaining any truth value (i.e., she states “I really don’t remember…”). (8) Ben-Veniste (Democrat) Interrogation of Rice (Republican) [Q 1] Did you tell the President at any time prior to August 6 of the existence of al Qaeda cells in the United States? [N-A/CLR 1] MS. RICE: First, let me just make certain =. [Q 1’] MR. BEN-VENISTE: If you could just answer that question =. [N-A/CLR 1’] MS. RICE: Well, first =. MR. BEN-VENISTE: = because I only have a very limited -[N-A/CLR 1’] MS. RICE: I understand, Commissioner, but it’s important =.

Cooperative conflict and evasive language 19

MR. BEN-VENISTE: Did you tell the President? (APPLAUSE). [N-A/CLR 1’’] MS. RICE: It’s important that I also address -- it’s also important, Commissioner, that I address the other issues that you have raised. So I will do it quickly, but if you’ll just give me a moment. [Q 1’’’] MR. BEN-VENISTE: Well, my only question to you is whether you told the President. [N-A/CLR 1’’’] MS. RICE: I understand, Commissioner, but I will -- if you will just give me a moment, I will address fully the questions that you’ve asked. First of all, yes, the August 6th PDB was in response to questions of the President. [REA 1’’’] In that sense he asked that this be done. It was not a particular threat report. And there was historical information in there about various aspects of al Qaeda’s operations. Dick Clarke had told me, I think in a memorandum -- I remember it as being only a line or two -- that there were al Qaeda cells in the United States. [STR 1’’’] Now, the question is, what did we need to do about that? And I also understood that that was what the FBI was doing, that the FBI was pursuing these al Qaeda cells. I believe in the August 6th memorandum it says that there were 70 full-field investigations under way of these cells. And so there was no recommendation that we do something about this, but the FBI was pursuing it. [A/DIS 1] I really don’t remember, Commissioner, whether I discussed this with the President. MR. BEN-VENISTE: Thank you. [A/ELA 1] MS. RICE: I remember very well that the President was aware that there were issues inside the United States. He talked to people about this. But I don’t remember the al Qaeda cells as being something that we were told we needed to do something about. [Q 2a/b] MR. BEN-VENISTE: Isn’t it a fact, Dr. Rice, that the August 6th PDB warned against possible attacks in this country? And I ask you whether you recall the title of that PDB. [A 2b] MS. RICE: I believe the title was “Bin Laden Determined to Attack Inside the United States.” Now, the PDB =. [Q 1’’]

MR. BEN-VENISTE: Thank you.

190 Lawrence N. Berlin

Clarification, the fourth type of non-answer, presents another evasive tactic that resembles elaboration. In a clarification, though, the respondent flouts relation by forwarding information that is purportedly (by the respondent) relevant, but is clearly unrelated to the question posed. Again in example (8), Rice insists on sharing what she claims to be essential information without responding to the question: “I understand, Commissioner, but I will -- if you will just give me a moment, I will address fully the questions that you’ve asked.” In the analysis of the discourse, it is clear that what follows does not contribute anything to the linguistic context beyond further prolixity as she ultimately disavows recalling her discussions with the President after clarifying information about a question that Ben-Veniste never asked. The final type of non-answer, the partial answer (PTA), is similar to the platform assertion in that it also flouts quantity. Conversely, however, it is the first non-answer type we see where the contribution tends to provide too little rather than too much detail. In example (9), Rice’s answer (N-A/PTA 1’) to Bob Kerrey, a Democrat, lacks precision and requires him to pose a follow-up question (Q 2) with more direction (Q: “Did you talk…?” → A: “We talked -- → Q: “did you instruct…?”). Initially, she evades going on record with a direct answer and gives a partial answer which does not respond to the prompt. It is only after he continues with the subsequent question (Q 2) that Rice repeats a portion of his question, violating manner by suggesting that there was somehow something unclear in his interrogation. (9) Kerrey (Democrat) Interrogation of Rice (Republican) KERREY: […] Let me ask a question that -- well, actually, let me say -- I can’t pass this up. I know it will take into my 10-minute time. But as somebody who supported the war in Iraq, I’m not going to get the national security advisor 30 feet away from me very often over the next 90 days. (LAUGHTER). And I’ve got to tell you, I believe a number of things. I believe, first of all, that we underestimate that this war on terrorism is really a war against radical Islam. Terrorism is a tactic; it’s not a war itself. Secondly, let me say that I don’t think we understand what the -- how the Muslim world views this, and I’m terribly worried that the miliary tactics in Iraq are going to do a number of things, and they’re all bad. One is the -(APPLAUSE). (to the audience) No, please don’t. Please do not do that. Do not applaud.

Cooperative conflict and evasive language

[Q 1]

[A/ELA 1]

[Q 1’] [N-A/PTA 1’]

[Q 2] [N-A/REP 2] [Q 2’] [N-A/STR 2’]

[…] Let me ask you, first of all, a question that’s been a concern for me from the first day I came onto the Commission, and that is the relationship of our executive director to you. Let me just ask you directly, and you can just give me -- keep it relatively short, but I wanted to get it on the record. Since he was an expert on terrorism, did you ask Philip Zelikow any questions about terrorism during transition, since he was the second person carded in the National Security Office and had considerable expertise? MS. RICE: Philip and I had numerous conversations about the issues that we were facing. Philip was, in fact, as you know, had worked in the campaign and helped with the transition plans. So, yes. MR. KERREY: Yes, you did talk to him about terrorism? MS. RICE: We talked -- Philip and I, over a period of -- you know, we had worked closely together as academics, of course talked about -MR. KERREY: During the transition, did you instruct him to do anything on terrorism? MS. RICE: Oh, to do anything on terrorism? MR. KERREY: Yes. MS. RICE: To help us think about the structure of the terrorism -- Dick Clarke’s operations, yes.

Consequently, a form of non-answer may frequently be the most appropriate response type in the interface between the linguistic and interactional contexts, especially considering the participant roles. Still, the expectation and subsequent performance of appropriate behavior need not automatically be cooperative to be appropriate (e.g., in a cross-examination); for instance, participants may be appropriately uncooperative (or blatantly conflictive) when their internal aims do not coincide. At such junctures, conflict may appear outwardly combative as the examples show.

3.

Analysis of the conjuncture – themes

One of the primary advantages of CDA is its ability to go beyond a purely superficial analysis, responding to the concerns of various scholars to contextualize findings within their environment. In the analysis of the conjuncture, for example, it is possible to identify themes within the discourse through a variety of means. In

191

192 Lawrence N. Berlin

themselves, themes may be considered a unit of analysis (cf. Shuy 1982, for a discussion of topic). However, in this particular investigation, I was concerned with the emergent themes as they were marked in text (i.e., the linguistic context) and indexed an explanation with regard to the broader sociocultural and sociopolitical levels of influence in the situational and extrasituational contexts.

3.1 “Pre-/post-9/11” (time-related) In just the three interrogations I examined (only a small fraction of the 9–11 Commission Hearings), the shift in thinking as a result of the airplane attacks of September 11, 2001 seems to stand out as the end of one period and the beginning of another like perhaps no other in recorded history. As a recurrent theme, it demarcates an epoch in the United States (and perhaps other parts of the world) that the participants in the hearings reference to forward their perspectives or mitigate their responsibility. Thus, “pre-9/11” becomes a way to claim that the United States could not have anticipated the events of that fateful day, a veritable Age of Innocence. Conversely, both within and outside the United States, certain individuals have suggested that the U.S. was engaged in an Age of Arrogance more than anything else, believing itself impervious to an attack on its own soil, either because its defenses were so good or other countries (or entities) would never be so presumptuous to dare attack the world’s last remaining superpower. While I will not go so far as to take too strong a stance here for one position or the other (finding that there are always two or three or more sides to every argument), I will simply refer to 9/11 as the end of the Age of Ignore-ance to cover either perspective. That is, as a result of September 11, the United States could no longer afford to ignore the lives and viewpoints of other peoples in the world; it could no longer see its perspective as the only subject position, objectifying all others. Indeed, it has since become incumbent on the government to try to understand the subject positions of other polities – the extent to which it is succeeding, I leave to you to decide. In any event, the date as reference point is critical to the discourse and admission of responsibility; it is mentioned nearly 70 times in roughly 90 pages of transcript, specifically as a point of demarcation (i.e., in its “pre-/post,” “before/after” form).

3.2 “233 days” (time-related) Another time-related theme that is mentioned repeatedly is “233 days” – the amount of time between when the Bush administration had been in office and September 11, 2001. This particular reference is made exclusively by Condo-

Cooperative conflict and evasive language

leezza Rice, usually as a means to lessening the culpability of the government. An interesting difference is that reference to this topic occurs elsewhere in the data, referred to by both various interrogators and respondents, but in each occurrence by Rice, the number of days is always expressed precisely as “233 days,” rather than “7-and-a-half months” or “eight or so months”; I would argue that this precision emerges as a striking contrast to other parts of her testimony that can be deemed as evasive, such as her extensive use of evidentiality, her insistence on using clarification as a non-answer form, especially in example (8) with Ben-Veniste, and her multiple use of disavowing knowledge in example (4) with Lehman. It may also be worth noting that, in actuality, 233 days represents a little less than 1/6 of the entire time an administration will be in office, assuming a one-term presidency.

3.3 CIA-FBI disconnect (tactic-related) A final recurrent theme that emerges as grounded in the data is the disconnect between the agencies charged with security. Though this theme is referred to over 50 times in a number of different ways, Rice, in half of the citations, refers to this disconnect as a structure, ostensibly one that the Bush administration inherited along with its shortcomings. Signified differentially as a “structural problem” or “structural impediment” to mitigate, clarify, or reassign responsibility, and as “structural reform” or “structural changes” to assert the positive advances the Republicans have made in terms of the U.S. Patriot Act6 and the establishment of an Office of Homeland Security7 (i.e., asserting platform), Rice and others advance this theme as a key issue in understanding and justifying the security breach surrounding September 11.

6. The U.S. Patriot Act was enacted after September 11, 2001, to enable the Executive Branch of the government (i.e., the President) wider powers to act when he deems appropriate under the guise of rapid response to a terrorist threat. The law has been controversial – despite its overwhelming approval when originally voted on in both houses of Congress – especially as regards potential violations of the U.S. Constitution with its system of checks and balances and the assurances of personal civil liberties against illegal search and seizure. 7. The establishment of the Office of Homeland Security emerged in direct response to the recommendations of the 9–11 Commission to bring all security and information agencies under one office in order to assure the sharing of relevant information regarding potential threats to the nation.

193

194 Lawrence N. Berlin

4.

Analysis of the practice – motivation

As has often been claimed, participants want to protect their own self-interests; this, in turn, can lead to a less than completely truthful or accurate representation within the context of court, in other words, evasive language. Shuy distinguishes between the pursuit of self-interest (“intentionality”) from ambiguity, stating that the reasons for engaging in ambiguous talk may vary greatly, not necessarily always due to personal volition. It may simply be that the interrogator and the witness are speaking from two different places, causing an unintentional ambiguity, or “they simply may be verbally sloppy” (2001: 446). In the case of the 9–11 Hearings, however, I would argue that this is not the case since the contributions by respondents appear to be carefully constructed for the greatest effect, such as with Rice’s repeated references (cf. Tannen 1989) to “233 days” as a device to suggest that there was not enough time to reorganize the security agencies and become aware of the threat of attack.

4.1 Time positioning Within the immediate context of the hearings, several synchronic features were particularly relevant. For instance, the hearings took place in two phases: closed and open (i.e., live) sessions. The portions of the proceedings that were closed presumably dealt with matters of security that were ongoing and could not be released to the public. The live sessions had both an audience in attendance and were televised, adding a heightened level of awareness which participants (interrogators and witnesses) incorporated into their verbal performance (cf. Bell 1984). The closed-door sessions were referred to several times within the frame of the interactional context as a means of gaining support from the live audience (in the situational context) to get certain memoranda declassified and/or to appear more cooperative and sincere. Another synchronic feature that was important was the timing of the hearings (cf. Bell 1998). They took place well after September 11 (about two and a half years). But the positioning seems especially poignant, given that the presidential elections were to be held later that same year and the U.S. invasions of Afghanistan and Iraq had already taken place. Additionally, several books had been published and were widely known as they had condemned the Bush administration’s policies, most notably a book by Richard Clarke (2004), former counter-terrorism expert who had been employed since the time of the Reagan administration. Clarke’s book is also referred to numerous times in the transcripts. The combination of synchronic factors made it absolutely necessary for the Bush affiliates to

Cooperative conflict and evasive language

appear in control while simultaneously denying any culpability of the events that led to the attacks during their administration. In referring to making a case for invading Iraq on the grounds of weapons of mass destruction, Rudd states that, “Certitude sells. […] If politicians wish to win elections, they must market themselves. They must portray certitude when in fact they have no objective evidence” (2004: 517). In the case of the 9–11 Commission Hearings, however, I would argue that it is equally important to express certitude even in the face of refuting objective evidence as this could prove disastrous for them in forthcoming elections.

4.2 Political positioning As previously indicated, the 9–11 Commission was composed of both Republicans and Democrats. Considering the timing – shortly before the 2004 Presidential election (sociocultural influence in the situational context) – it would seem the hearings presented an ideal opportunity for the Democrats to position themselves more favorably in the public eye by making the Republicans, especially the Bush administration, look bad. However, any maneuvering by the Democrats was especially difficult given the outcome of the 2002 midterm elections where the country, still in the grip of fear, showed general support for those who supported the protectionist policies of Bush (i.e., the Republicans). In the midterm elections, Republican candidates prevailed in virtually every election, making them the majority party in both houses of the Congress. Indeed, “legislators who did not show proper support for the president’s plan of action […] were portrayed in a most unfavorable light” (Rudd 2004: 512). As a consequence, several decisions were taken in the Congress by Republicans and Democrats alike under the Patriot Act, such as expanding the executive power of the President to take action, expanding the surveillance capability of security agencies (previously deemed illegal) to the extent of violating personal freedom, and expanding the ability of law enforcement agencies to hold someone without an arrest warrant or trial, all in the name of protecting the people of the United States from suspected terrorist activities. Along with the several justifications for complicity among government officials around the elections and despite the many negative reactions to the Bush doctrine both from within the country and abroad, it still appears that the primary job of politicians is to keep their job. Thus, at such a high level of political interaction, it may be particularly appropriate for witnesses to engage in evasive linguistic behavior and for their interrogators to allow them to get away with it. Considering Brown and Levinson’s (1987) measurement of face-threatening acts, more attention is given to the face needs of co-participants – regardless of the potential conflict between them – when the social distance is the lowest and the is-

195

196 Lawrence N. Berlin

sue is of great importance (Holmes & Marra 2004). The individual aims and needs of the interlocutors may account for the high degree of apparent cooperation (despite the violations and flouts of the Cooperative Principle) since all actors in the hearings were high-level (i.e., high power) individuals in the government, ostensibly working together to uncover the causes of the most horrific attack on U.S. soil. Indeed, considering the various levels of context analyzed through a CDA, the usage of non-answers as response forms, while not always entirely cooperative, is to be expected and may even be interpreted as appropriate.

5.

Conclusion

According to Wilson, “It is not simply manipulation that is at issue in the case of political language; it is the goal of such manipulation which is seen as problematic. Politicians seem to want to hide the negative within particular formulations such that the population may not see the truth or the horror before them” (2001: 400). Recognizing that the public is always watching on some level, politicians’ manipulation of or attention to linguistic forms – in this case, through evasive language – creates a public discourse (Berlin 2005; Caldas-Coulthard 2003; Lakoff 1990; Weigand 1999). 1. By means of description, evasive language during a hearing is comprised of a variety of evasive response tactics – both answers and non-answers – given to questions posed, often coupled with varying degrees of evidentiality to remove the level of claimed culpability of the witness. 2. In terms of interpretation of multiple levels of context and their interface, the conflictive nature of the hearings notwithstanding, the appearance of cooperation is achieved through strict adherence to the rules of the general hearing and specific interrogation process, and a tacit agreement that opting out, violating, clashing, or even flouting maxims of the Cooperative Principle (i.e., uncooperative behavior) can be deemed appropriate language usage. 3. Participants have knowledge of and make full use of the various levels of context to manipulate language. That is, they skillfully use language in the linguistic context, relative to the interactional (internal and external), situational, and extrasituational contexts, as well as the various influences, both local and global, with which the interlocutors and overhearers are aware in varying degrees and will use to interpret meaning. In turn, effective evasion in use can construct an effective, persuasive, and appropriate discourse.

Cooperative conflict and evasive language

References 9–11 Commission. “Official website of the 9–11 Commission.” http://www.9-11commission. gov (accessed July 20, 2004). Austin, J. 1962. How to Do Things with Words. Oxford: Oxford University Press. Bakhtin, M.M. 1981. The Dialogic Imagination. M. Holquist (ed.). Austin: University of Texas Press. Bell, A. 1984. “Language style as audience design.” Language in Society 13: 145–204. Bell, A. 1998. “The discourse structure of news stories.” In Approaches to Media Discourse, A. Bell and P. Garrett (eds.), 64–104. Oxford: Blackwell. Berlin, L.N. 2005. “Media manipulation.” In Dialogue Analysis IX: Dialogue in Literature and the Media: Selected Papers from the 9th IADA Conference Salzburg 2003 (Part 2: Media), A. Betten and M. Dannerer (eds.), 173–182. Tübingen: Max Niemeyer Verlag. Brown, P., and Levinson, S.C. 1987. Politeness: Some Universals in Language Usage (2nd ed.). Cambridge: Cambridge University Press. Caldas-Coulthard, C. 2003. “Cross-cultural representation of ‘otherness’ in media discourse.” In Critical Discourse Analysis: Theory and Interdisciplinarity, G. Weiss and R. Wodak (eds.), 272–296. Hampshire: Palgrave Macmillan Ltd. Chouliaraki, L., and Fairclough, N. 1999. Discourse in Late Modernity: Rethinking Critical Discourse Analysis. Edinburgh: Edinburgh University Press. Clarke, R. 2004. Against All Enemies: Inside America’s War on Terror. New York: Free Press. Coulthard, M., and Montgomery, M. (eds.). 1981. Studies in Discourse Analysis. London: Routledge & Kegan Paul, Ltd. Fairclough, N. 1989. Language and Power. London and New York: Longman. Fairclough, N. 1992. Discourse and Social Change. Cambridge: Polity Press. Fairclough, N. 1995a. Media Discourse. London: Edward Arnold. Fairclough, N. 1995b. Critical Discourse Analysis: The Critical Study of Language. London and New York: Longman. Fairclough, N. 1998. “Political discourse in the media: An analytical framework.” In Approaches to Media Discourse, A. Bell and P. Garrett (eds.), 142–162. Oxford: Blackwell. Fetzer, A. 2005. “Non-acceptances in political interviews: British styles and German styles in conflict?” In Dialogue Analysis IX: Dialogue in Literature and the Media: Selected Papers from the 9th IADA Conference Salzburg 2003 (Part 2: Media), A. Betten and M. Dannerer (eds.), 87–96. Tübingen: Max Niemeyer Verlag. Flanders, N.A. 1970. Analysing Teaching Behaviour. Reading, MA: Addison-Wesley. Geertz, C. 1983. Local Knowledge (3rd ed.). San Francisco: Basic Books. Goodwin, C., and Duranti, A. 1992. “Rethinking context: An introduction.” In Rethinking Context: Language as an Interactive Phenomenon, A. Duranti and C. Goodwin (eds.), 1–42. Cambridge: Cambridge University Press. Grice, H.P. 1975. “Logic and conversation.” In Speech Acts, P. Cole and J.L. Morgan (eds.), 41– 58. New York: Academic Press. Halliday, M.A.K. 1978. Language as Social Semiotic: The Social Interpretation of Language and Meaning. London: Edward Arnold.

197

19 Lawrence N. Berlin

Halliday, M.A.K. 1984. “Language as code and language as behaviour: A systemic-functional interpretation of the nature and ontogenesis of dialogue.” In The Semiotics of Culture and Language: Volume 1: Language as Social Semiotic, R.P. Fawcett, M.A.K. Halliday, S.M. Lamb, and A. Makkai (eds.), 3–35. London and Wolfeboro, NH: Frances Pinter. Halliday, M.A.K. 1994. Introduction to Functional Grammar (2nd ed.). London: Edward Arnold. Halliday, M.A.K. 2002a. On Grammar, Vol. 1 in the Collected Works of M.A.K. Halliday, J.J. Webster (ed.). London: Continuum. Halliday, M.A.K. 2002b. Linguistic Studies of Text and Discourse, Vol. 2 in the Collected Works of M.A.K. Halliday, J.J. Webster (ed.). London: Continuum. Holmes, J., and Marra, M. 2004. “Leadership and managing conflict in meetings.” Pragmatics 14 (4): 439–462. Hymes, D. 1972. “Models of the interaction of language and social life.” In Directions in Sociolinguistics: The Ethnography of Communication, J.J. Gumperz and D. Hymes (eds.), 35–71. New York: Holt, Rinehart and Winston. Janney, R.W. 2002. “Cotext as context: Vague answers in court.” Language & Communication 22: 457–475. Kumaravadivelu, B. 1999. “Critical classroom discourse analysis.” TESOL Quarterly 33 (3), 453–484. Lakoff, R.T. 1990. Talking Power: The Politics of Language. San Francisco: Basic Books. Levinson, S. 1983. Pragmatics. Cambridge: Cambridge University Press. Lincoln, Y.S., and Guba, E.G. 1985. Naturalistic Inquiry. Newbury Park, CA: Sage. Lynch, M.E., Livingston, E., and Garfinkel, H. 1983. “Temporal order in laboratory work.” In Science Observed: Perspectives on the Social Study of Science, K. Knorr-Cetina and M. Mulkay (eds.), 205–238. Beverly Hills: Sage. Malinowski, B. 1923. “The problem of meaning in primitive languages.” In The Meaning of Meaning, C.K. Ogden and I.A. Richards (eds.), 296–336. New York: Harcourt, Brace and World, Inc. Mey, J. 1993. Pragmatics: An Introduction. Oxford: Blackwell. NCTA. 2004. The 9/11 Commission Report: Final Report of the National Commission on Terrorist Attacks upon the United States. New York/London: W.W. Norton. Pomerantz, A. 1978. “Compliment responses: Notes on the co-operation of multiple constraints.” In Studies in the Organization of Conversational Interaction, J. Schenkein (ed.), 79–112. New York: Academic Press. Rudd, P.W. 2004. “Weapons of mass destruction: The unshared referents of Bush’s rhetoric.” Pragmatics 14 (4): 499–525. Sbisà, M. 2002. “Speech acts in context.” Language & Communication 22: 421–436. Searle, J.R. 1969. Speech Acts. Cambridge: Cambridge University Press. Shuy, R.W. 1982. “Topic as the unit of analysis in a criminal law case. In Analyzing Discourse: Text and Talk, D. Tannen (ed), 113–126. Washington: Georgetown University Press. Shuy, R.W. 2001. “Discourse analysis in the legal context.” In The Handbook of Discourse Analysis, D. Schiffrin, D. Tannen, and H.E. Hamilton (eds.), 437–452. Malden, MA: Blackwell Publishing. Sinclair, J., and Coulthard, M. 1975. Towards an Analysis of Discourse. London: Oxford University Press. Small, E. Witness for the Prosecution. Prod. by Arthur Hornblow, Dir. by Billy Wilder. 116 min. MGM/UA, 1957. Videocassette.

Cooperative conflict and evasive language 199

Spradley, J.P. 1980. Participant Observation. New York: Holt, Rinehart and Winston. Tannen, D. 1989. Talking Voices: Repetition, Dialogue, and Imagery in Conversational Discourse. Cambridge: Cambridge University Press. Urban, G. 1991. A Discourse-Centered Approach to Culture: Native South American Myths and Rituals. Austin: University of Texas Press. Vološinov, V.N. 1973. Marxism and the Philosophy of Language, L. Matejka and I.R. Titunik (trans). New York: Seminar Press. Weigand, E. 1999. “Dialogue in the grip of media.” In Dialogue Analysis and the Mass Media, B. Naumann (ed.), 35–54. Widdowson, H.G. 2004. Text, Context, Pretext: Critical Issues in Discourse Analysis. Malden, MA: Blackwell Publishing. Wilson, J. 2001. “Political discourse.” In The Handbook of Discourse Analysis, D. Schiffrin, D. Tannen, and H.E. Hamilton (eds.), 398–415. Malden, MA: Blackwell Publishing. Wittgenstein, L. 1958. Philosophical Investigations (2nd ed.), G.E.M. Anscombe and R. Rhees (eds.), G.E.M. Anscombe (trans). Oxford: Blackwell.

part iii

Bridging problems between micro and macro

The attenuating conditional Context, appropriateness and interaction Johanna Miecznikowski and Carla Bazzanella* University of Turin, Italy

The attenuative reading of the conditional form (condatt) in Italian and French illustrates the multifunctionality of linguistic structures and the necessity of analyzing them in their synergies with relevant contextual parameters. The condatt allows the speaker to allude to the existence of pragmatic prerequisites for his/her utterance without presenting them as taken for granted. An empirical study of spoken data reveals that at least in French this property has specific effects on various levels of interaction: the form expresses negotiability of the speaker’s acts; it preferredly encodes new information and contributes to structure the thematic progression of discourse; it tends to encode initiative and/or dispreferred communicative acts. These results lead us to formulate some hypotheses as to the condatt’s appropriateness with regard to global context. In particular, given that the form signals negotiability and is better compatible with a reduced rather than with a rich common ground, we expect it to be most appropriate in dialogical genres characterized by a low degree of acquaintance, a high degree of social distance between co-participants, and symmetry of social roles.

* Carla Bazzanella has written the section “Context, appropriateness and interaction: Some general remarks”; Johanna Miecznikowski has written the remaining sections. All sections have been planned and revised jointly by both authors.

204 Johanna Miecznikowski and Carla Bazzanella

1.

Context, appropriateness and interaction: Some general remarks

1.1 Grammar and interaction In recent years, a perspective in which language is seen as designed for and shaped by interaction in context1 has gained considerable importance in linguistics. In the 70’s, the pragmatic turn had brought to our attention the relevance and crucial significance of context, in its multifarious aspects and relationships, for linguistic analysis. But the investigation of the complex relationship between grammar and interaction is still in progress, thanks also to the fruitful interchange between linguists and other scholars, such as sociologists and anthropologists, doing research on language use and communicative action. When one considers grammar “as a part of a broader range of resources – organizations of practices, if you will – which underlie the organization of social life, and in particular the way in which languages figures in everyday interaction and cognition” Schegloff, Ochs, Thompson (1996: 2), one cannot avoid to take “a variety of telling linkages between interaction and grammar” (id.: 3) into account. In a nutshell, the matrix of ordinary language philosophy (cf. Austin and Searle’s theories of speech acts, Wittgenstein’s language game, Grice’s cooperative principle), together with functional linguistics, conversation/discourse analysis,2 and, more recently, cognitive linguistics, has strongly influenced subsequent linguistic research, in the direction of a strong involvement both of contextual components and of the interactive framework where the language system is deeply embedded in the social world and cannot be viewed as an autonomous module. Language use and communicative action have become the starting point of a wide range of interdisciplinary work, and, more specifically, the relations between grammar and interaction have been focused upon.3 As a consequence, linguistic meaning has been conceived of as both deriving from and helping to constitute social practices and activities.

1. Cf. inter alia Ochs, Schegloff and Thompson (1996), Selting and Couper-Kuhlen (2000), Kerbrat-Orecchioni (2001, 2005) and Mondada (2001). 2. If we focus on the object of study, Levinson’s classical distinction between Conversational Analysis and Discourse Analysis weakens, as both Tannen (1989: 6) and Kerbrat-Orecchioni (2005: 14) point out. 3. “Grammar is not only a resource for interaction and not only an outcome of interaction, it is a part of the essence of interaction itself. […] Grammar is viewed as lived behavior, whose form and meaning unfold in experienced interactional and historical time.” Schegloff, Ochs and Thompson (1996: 38).

The attenuating conditional 205

Within this framework, starting from the canonical situation of face-to-face conversation,4 the various forms of discourse (be they spoken, written, or computer-mediated) are to be studied in their specific interactive context, as situated and joint actions, “carried out by an ensemble of people acting in coordination with each other” (Clark 1996: 3). While certain linguistic phenomena such as discourse markers or personal pronouns inevitably and evidently require a treatment which encompasses the interaction level and a wide range of contextual components, the relevance of such parameters for the analysis of other categories such as tense and mood is less commonly recognized. In this paper, we will focus on one specific modal morpheme, the conditional form in Italian and French in its so-called attenuating use (condatt), arguing that its functions and its appropriateness conditions can be understood only with reference to situated interaction. Before doing so, we will make some general remarks on context and appropriateness as a theoretical outline.

1.2 Local and global context Context is a key problem of scientific and philosophical inquiry in general,5 and of the analysis of language in particular.6 At any moment of an interaction, different aspects of context constitute a complex pragmatic configuration (cf. Bazzanella 2004, 2005) of interrelated features, perceived holistically and analyzed only occasionally (and for all practical purposes) by participants, especially when a communicative problem of some kind is encountered or anticipated (see, e.g., misunderstandings, repair, and accounting practices). This complexity7 can be captured analytically be distinguishing various types, levels, aspects or dimensions of context (cf. Note 6). Making a choice among other possible terms, such as micro and macro, to delimit its properties/features, we 4. Cf., e.g., Kerbrat-Orecchioni (2005: 14); for a typology of forms of dialogic interaction, taking face-to-face interaction as a starting point, cf. Bazzanella (2002). 5. Cf., inter alia, Kaplan (1977), Barwise and Perry (1983), Sperber and Wilson (1986), Perry (1993), Light and Butterworth (1993), Bouquet et al. (1999, 2003), Stalnaker (1999), Akman et al. (2001), Penco (2002), Andler (2003) and Riehle (2003). 6. Cf. inter alia Giglioli (1972/1973), Lyons (1981), Duranti and Goodwin (1992), Auer and Di Luzio (1992), Givón (1989, 2005), Edmonds and Akman (2002), Linell and Thunqvist (2003), Akman and Bazzanella (2003), Fetzer (2004, Introduction this volume). 7. “[…] context is no longer looked upon as an analytic prime but rather seen from a partswhole perspective as an entity containing sub-entities (or sub-contexts).” Fetzer (this volume).

206 Johanna Miecznikowski and Carla Bazzanella

will speak of local and global aspects of context (cf. Akman & Bazzanella 2003; Bazzanella 1998, 2004). Local features of context are features that are intrinsically related to, or activated by (Sperber & Wilson 1986/1993), the immediate structural environment; in verbal interaction, this means that they depend on the on-going interaction. So the local context of an utterance in face-to-face interaction prototypically includes the linguistic co-text, the sequential development of interaction (or dialogue common ground, cf. Fetzer 2004: 202–203, Fetzer 2007), interpersonal relations as they are shaped through interaction, and objects/spaces integrated into interaction by gesture and bodily movement in general. Global features of context are features that do not depend intrinsically on on-going interaction, e.g. sociolinguistic parameters like age and status, global time and space localization, relatively stable and (more or less) shared knowledge and beliefs (cf. Clark 1996). Speakers usually treat global context as pre-existing to interaction, which does not exclude that they make selected aspects of it explicit, e.g. through contextualization cues (Gumperz 1977) and categorization devices (Sacks 1992; Bonu, Mondada & Relieu 1994). Under certain circumstances, specific aspects of background knowledge may be negotiated; this is the case e.g. of lexical meaning, which speakers normally – for all practical purposes – treat as pre-existing, but which in certain contexts such as plurilingual settings (cf. Lüdi 1991) or scientific discussions (cf. Miecznikowski 2005), happen to be topicalized and redefined. Note that the local or global status of certain contextual features may evolve. An important example are discourse genres or activity types (Levinson 1992), i.e. the conversational tasks, goals and roles related to them: though treated by participants as providing a conventional and relatively stable (global) setting, they may change and be implicitly or explicitly negotiated during the interaction (i.e. locally; cf. Weigand & Dascal 2001), by switching to another topic or to a given sub-genre.

1.3 Context and appropriateness in linguistic analysis The relationship between language and context is dynamic, language being both context-dependent and context-changing. Speakers and hearers use local and global context to interpret and negotiate communicative actions; simultaneously, by performing these communicative actions, they modify local context and enact (in an ethnomethodological sense) global context. Thanks to grammaticalization and lexicalization processes, this bidirectional relationship holds also between linguistic forms or constructions and their con-

The attenuating conditional 207

text. So speakers and hearers use context in a routinized and conventionalized8 manner to disambiguate and interpret linguistic items. For example, in order to interpret a polysemous verb form like the Italian imperfetto, assigning it temporal, aspectual, textual, modal and/or pragmatic meanings (cf. Bazzanella 1990, 2000), hearers take into account a series of parameters of local context, such as the Aktionsart of the verb and definiteness/number marking of subject and/or object (relevant for the choice of an iterative vs. non iterative interpretation), clause type and clause combining (relevant e.g. for the foregrounding of simultaneity marking as a temporal meaning component, or the foregrounding of conditionality as a modal meaning component), temporal reference (indicators of future reference such as domani ‘tomorrow’ e.g. induce the foregrounding of modal and attenuative functions), or, finally, discourse genre, activity type (the textual function of the imperfetto as a backgrounding device e.g. is used especially in narrative genres; the imperfetto ludico, used to create an imaginary world, is particularly related to children’s play), and the specific relationship between co-participants. Conversely, speakers and hearers use linguistic items in multiple ways as conventionalized resources in interaction,9 thus triggering simultaneous processes: e.g., via discourse markers or topicalization procedures, they can modify local context; via specific lexical items, status related address terms, and other grammatical forms which function as contextualization cues, they can index elements of global context. If appropriateness is seen as a context dependent notion, the considerations on context made above become immediately relevant to the way we define appropriateness, and especially to the appropriateness of specific linguistic items, with which we are concerned in this paper. It appears quite clearly, in particular, that the context dependent appropriateness of linguistic choices should not be conceptualized in terms of a simple matching of these choices with a pre-established and rigid set of contextual parameters. Rather, appropriateness is to be seen as an intersubjective category, a “dialogical concept par excellence” (Fetzer 2004: 90), which is interactively construed and essentially dynamic. In particular, the appropriateness of a linguistic choice (as an utterance act, cf. Introduction, this volume) depends on the sequentially situated action the speaker is performing, on the context changing potential of the linguistic item itself, and on subsequent conversational events confirming or redefining the relevant local and global contextual parameters. Like other aspects of meaning assignment, the status of a linguistic . Of course conventionalisation encompasses both diachronic changes (cf. e.g. Traugott & Dasher 2002) and cross-cultural differences which are selected by different cultures (cf. e.g. Wierzbicka 1996). 9. Let us think about phenomena such as politeness; see 3.1 with specific regard to condatt.

20 Johanna Miecznikowski and Carla Bazzanella

entity as being more or less appropriate emerges in the sequential unfolding of interaction. The ways a linguistic form interacts with its relevant context, at the interface between grammar and interaction, are non-obvious and should be investigated empirically on the basis of naturally occurring data. The condatt, which we will turn to now, is a good example. If all existing descriptions of the Romance conditional distinguish an attenuating reading of this form with predominantly pragmatic functions, the focus on intuition and written data has left unexplored a large part of the functional spectrum of this reading as well as the contextual constraints on its use. In a more interactionist perspective, we will propose an analysis of the condatt in which context, in particular sequential context, is given key importance. After a first characterization of the condatt in French and Italian, we will explore its local context-changing functions on the interpersonal, thematic and sequential level and discuss the implications of this analysis for the understanding of the form’s appropriateness conditions, taking also into account further, more global, contextual parameters such as discourse genre.

2.

The attenuating conditional in context: A proposal of analysis

2.1 Overview and data The Italian and French conditional is morphologically composed of the -r- morpheme that is present also in the infinitive and in the future tense, and of a past tense ending – an imparfait ending in French, and a passato remoto ending in Italian. Like the imperfetto mentioned in the previous section, it is a highly polysemous verb form. In order to delimitate our unit of analysis, i.e. the condatt in Italian and French, we will have a brief look at the range of uses of the conditional in these two languages, insisting particularly on the specifying (in language production) and disambiguating (in language comprehension) role of the immediate lexical and syntactical co-text. Section 2 will give a first impression of the form’s multifunctionality, i.e. of the fact that the functional scopes of its various readings extend over different levels and goals of discourse. Our investigation draws on an on-going corpus-based study making use of both qualitative analysis and quantitative analysis of co-occurrence patterns (cf. also Miecznikowski forth. a). The data that have been used for the semantic, pragmatic and sequential analyses proposed in this chapter are:

The attenuating conditional 209

(a) A plurilingual corpus of scientific conferences and group meetings. In a rough approximation, the French part of this corpus counts about 400’000 words, the Italian part about 15’000 words; it contains 1908 tokens of the French conditional form and 32 Italian tokens).10 (b) The interactions of type C included in the LIP corpus of spoken Italian (cf. De Mauro et al. 1993 and, for an on-line queryable version, http://languageserver.uni-graz.at/badip/). Text type C is defined as “scambio comunicativo bidirezionale con presa di parola non libera faccia a faccia” (bidirectional communicative exchange, face-to-face, with unconstrained turn-taking) and consists of meetings, interrogatories, interviews and exams of various types. It contains 350 tokens of the Italian conditional. (c) the transcriptions of six meetings of a parent’s committee held in Italian, published in Franceschini (1998)11 (80 tokens of the conditional).

2.2 Modal uses of the conditional form in Italian and French The core meaning of the conditional may be seen as expressing a proposition that is not actually asserted, in the deictic centrum of ego-hic-nunc, but is presented by the speaker as asserted or assertable at some other point of reference, perceived as distant from the origo in one way or the other (cf. Dendale 2001: 16, who speaks of a “déplacement de coordonnées”, i.e. a shift of coordinates). The conditional’s semantics contains an instruction to look for that other point of reference in the linguistic and/or extralinguistic context, and, depending on the type of point of reference, the specific meaning of the conditional is further refined.12 It is useful to conceptualize the reference point as a ground with respect to which the proposition in the conditional emerges as a figure (for a detailed semantic analysis cf. Miecznikowski (forth. b):

10. These materials have been gathered within a research project on the interactional construction of scientific discourse, realized at Basel University from 1997 to 2001 under the direction of Lorenza Mondada (cf. Mondada 2005). We thank the project leader and the Swiss National Foundation, which has funded the project (no 1214-051022.97), for allowing further use of these data. 11. We wish to thank Rita Franceschini for having provided us the electronic version of this corpus. 12. For overviews over different uses of the conditional in French cf. Martin (21992), Riegel (1996), Tasmowski and Dendale (2001), Haillet (2002). For Italian, cf. Schwarze (1983), Renzi et al. (1991: VIII; XIII 2.3. / 2.4.), Berretta (1992) and Bazzanella (1994, chap. 5).

210 Johanna Miecznikowski and Carla Bazzanella

The conditional’s basic meaning is to express a relationship of (temporal, causal, cognitive or pragmatic) antecedence between a contextually given ground, conceptualized as being distant from the modal-deictic centre, and an entity p containing the conditional form (a state of affairs, a proposition or a speech act).

We will not treat here the uses in which the reference point is located in the past, a case in which the conditional expresses posteriority; we will concentrate on the conditional’s modal uses instead. The conditional’s modal ground may be at least of the following types: –

– –

a condition, inferable from the context or expressed explicitly in the congiuntivo (Italian) / imparfait (French) as part of a potentialis or irrealis hypothetical construction; in this context, the conditional form strongly suggests a causal link between antecedent and consequent, favoring a predictive reading of the construction (Dancygier & Sweetser 2005: 31–35), a reading which cross-linguistically is prototypical, according to Comrie 1986 (cf. also Traugott et al. 1986; Couper-Kuhlen & Kortmann 2000); a universe of discourse (as in the form’s quotative reading); indirect evidence on the basis of which the speaker makes a weak inference. This inferential use is quite restricted, however: in French, it is limited to interrogative co-texts, in Italian it is ungrammatical (cf. Squartini 2001: 315– 317).

As to the condatt, it is seen by some authors (e.g. Martin 21992) as referring to implicit conditions and therefore as tightly related to the hypothetical conditional in potentialis constructions. Other authors prefer an explanation in terms of speaker distantiation, underlining affinities with the quotative conditional (so for Haillet 2002: 88–94 the condatt signals a “dédoublement du locuteur”, i.e. a duplication of the speaker). We suggest an explanation of the condatt that is closer to the first type of explanation than to the second, although we do not postulate systematic paraphrasability by a hypothetical construction, let alone ellipsis of implicit underlying antecedents (cf. also our discussion of example (3) below): we argue that its modal ground is given by the presuppositions s of p – or rather by what would have been s if p had been formulated in the indicative present. As we will see, in this type of use, the conditional construes s as modally distant in the sense that the speaker questions its status as shared knowledge. This kind of presupposition weakening bears a family resemblance to presupposition cancellation or downgrading (Leonardi forth.) in general, discussed in the literature with regard to presupposition projection from simple clauses to complex clauses and discourses (cf. also Van der Sandt 1989). In particular, certain explicit if-clauses in hypothetical constructions may function in a similar way as the con-

The attenuating conditional

ditional form does. However, in the case examined here, the idea of presupposition weakening does not refer to the effects of adding explicit (or implicit, cf. the precedent note) linguistic material on the syntagmatic axis. Rather, it refers to a paradigmatic contrast between verb forms, comparable in certain respects to the paradigmatic opposition, in the nominal domain, between the presuppositional behavior of different determiners. Note that a selection of the presuppositional ground s as point of reference is possible –

–

mainly when s is in some way related to the verb bearing the conditional ending, the range of the conditional’s possible scopes as a presupposition modalizer being more extended in French than in Italian (this comparative issue will be subject to further investigation in the current study); when the factuality of s is not clearly given by virtue of the co- and context – an aspect we will consider in more detail in Section 4.

We will discuss this type of use in some detail, focusing, in particular, on the conditional with different kinds of modal predicates and with verbs of saying. We will concentrate on declarative contexts, leaving aside the interesting question of how the conditional’s semantic and pragmatic properties interact with those of questions.

2.3 The conditional with modal verbs: The example of ‘volere’/‘vouloir’ With modal predicates, the virtualization of presuppositions may be situated either on the level of propositional content (de re) or on the interactional level. The latter is the case when the possibility of calling s into question is related to the fact that s depends on the on-going interaction and can therefore be treated by participants as being still negotiable. We will illustrate this by discussing the example of volere/vouloir. An utterance containing volere/vouloir in the indicative present + infinitive clause presupposes, we claim, that the agent/experiencer considers the embedded proposition q possible and in his/her range of influence (cf. Miecznikowski forth. c). This presupposition s is activated by the lexical semantics of volere/vouloir: these verbs focus on an agent’s volition with regard to objects or options that are available to him/her,13 contrasting in this respect, by the way, with other verbs 13. In modern Italian and French, the combination of these semantic properties with the expression of perfective aspect in the past triggers the conventionalized implicature that the agent’s desire has been fulfilled (cf. Squartini forth. for Old Italian, which behaves differently).

211

212 Johanna Miecznikowski and Carla Bazzanella

in the sphere of volitional modality such as sperare/espérer. The conditional of volere/vouloir evokes s, but presents it as questionable. It is therefore compatible with an eventual impossibility of q. In our first example, this operation on the utterance’s presuppositions is de re: the speaker expresses explicitly, in the immediately subsequent co-text, a counterfactuality of q due to factors that are independent of the on-going interaction: (1) (lip mc 10)14 tant’e’ che ci sono molte molte_ richieste eh a Milano ma anche altrove still there are many many_ demands erm in Milano but also elsewhere di_ di appunto superfici superiori ma non vengono accolte il che significa for_ for more space as I said but they are not successful which means che l’ambulante che vorrebbe essere in regola con la legge_ e’ obbligato that the trader that wantscond to conform to the law_ is obliged a essere contro la legge to be against the law

In other cases, and more frequently, the factors that could make the desired q possible or impossible are related to the on-going interaction. In the following metacommunicative announcement, for instance, q is an intended communicative action; the conditional signals that the speaker does not take the possibility that this communicative action be realized for granted: (2) (franceschini_5)15 in the introducing section of a longer turn (cf. also ex. 6, l. 12): io vorrei premettere una cosa I’d like to say something in advance

14. The transcription conventions of each of the examined corpora have been maintained for this presentation. In this example extracted from the LIP corpus, the underscore _ means word final vowel lengthening. 15. The following transcription conventions used in Franceschini (1998) are relevant for the transcripts cited here: short, medium, long pause = rapid continuation -, --, --aa, ll vowel / consonant lengthening , slightly rising intonation typical interrogative intonation ,, high rising intonation ? <, > increasing, decreasing volume . falling intonation breathing (x) incomprehensible passages (:h) ( ), ( ; ) tentative transcription, transcr. (( )) comments alternatives XX anonymized names

The attenuating conditional

Note that speakers, when using volere/vouloir in the conditional, make sometimes use of if-clauses to make explicit the interactional factors that could impede the realization of q: (3) (franceschini_5) (= ex. 6, l. 3–4) 1 Q va=be’ al secondo punto all’ordine del giorno c’è la -- la lettera-o.k. the second point on the agenda is the -- the letter -2 di dimissione da parte del - del presidente, - attuale del XX. of resignation from the - the actual president, - of the XX. 3 D no se - permettete se sembro io il primo destinatario, vorrei no if – allow me if I seem to be the primary addressee, I wantcond 4 - se non c’è niente in contrario parlare > (xx). to speak - if there are no objections16 > (xx).

This particular use of if-clauses confirms that the condatt is related to the use of the conditional in hypothetical constructions, but that it cannot be reduced to the latter. The tendency towards a causal and predictive interpretation, typical of hypothetical uses (cf. 2.1), is not observed when the condatt is accompanied by an if-clause (so in example (3) the speaker’s desire to speak does not depend on the absence of objections), and this semantic difference corresponds to formal peculiarities, above all to the use of the present tense in the protasis and a certain prosodic and syntactical independence between if-clause and main clause (cf. e.g. the parenthetical position in example (3)). These constructions are in fact best described as a kind of pragmatic conditional (Haegeman 1984; Athanasiadou & Dirven 2000), and more exactly as a specific type of discourse conditional (Athanasiadou & Dirven 2000: 13–17) or speech act conditional (Dancygier & Sweetser 2005: 113–117).

2.4 The conditional with verbs expressing communicative acts Also with a large class of verbs of saying, the conditional may be used to modalize underlying presuppositions. This is only possible, however, when these verbs are used in the first person to refer to the ongoing verbal interaction. In these (quasi-)performative uses, the conditional refers to pragmatic presuppositions s of the expressed communicative acts, signaling that the speaker does not take them for granted. S can be described, in speech act theoretical terms, as a subset of the act’s felicity conditions, i.e. those conditions that do not depend directly on the

16. Lit. “I wantcond – if there are no objections – to speak”.

213

214 Johanna Miecznikowski and Carla Bazzanella

speaker him/herself – implying an approach to speech acts that does not reduce the latter to the realization of speaker-intentions (cf. Oishi in this volume). So in the following example, the speaker defends a view and, using the conditional, evokes the possibility that the interlocutors might not take up this communicative act as a relevant and well-founded contribution to the on-going discussion: (4) (franceschini_2) non mi sembra inopportuno che noi,-proprio a chiusura d’anno adesso it doesn’t seem inappropriate to me that we, - right now at the end of the year mandiamo una specie di riassunto di quello che si è fatto.-io personalmente send a sort of summary of what has been done.- I personally dato che questa bozza c’è - riterrei se c’è da limarla la si lima given the fact that there is this outline, considercond that if it’s necessary to revise it we’ll revise it la si completa per i mesi che non ci sono anche s’è un po’ lunga.we’ll complete it for the lacking months, even if it’s a bit long

2.5 The conditional with predicates expressing a positive or negative evaluation A third class of predicates with which the conditional combines regularly to modify pragmatic presuppositions are predicates used to positively or negatively evaluate a state of affairs, the latter being expressed by a subject NP, an if-clause, an infinitive clause, or a complement clause, introduced, in French, by que + subjonctif and in Italian by che + congiuntivo presente/imperfetto or ∅ + congiuntivo imperfetto. (5) is a typical example: (5) (overheard) Sarebbe meglio tenerli al guinzaglio, però. (said to a lady going for a walk with her dogs) it would be better to keep them on a leash, you know

This type of use of the conditional is characterized by the fact that it potentially has a double modal ground. On the one hand, the embedded clause functions as an if-clause. So example (5) may be paraphrased as (5’): (5’) Sarebbe meglio se Lei li tenesse al guinzaglio, però. it would be better if you kept them on a leash, you know

However, the resulting hypothetical construction is not a prototypical conditional (Comrie 1986): the positive evaluation expressed by the speaker does not de-

The attenuating conditional

pend causally or logically on the verification of the state of affairs denoted by the embedded proposition. This semantic peculiarity favors, in constructions with evaluative predicates, a conventionalized implicature that makes the conditional’s interpretation shift towards the deictic center. The implicature is that the speaker would not only evaluate positively or negatively a state of affairs if it was to produce itself, but that he or she is evaluating, here and now, a possible state of affairs as a conceptual entity (cf. also the distinction between states of affairs and possible facts made in functional grammar, e.g. by Dik 1989: 248–250). When this implicature becomes prominent, the modal ground of the conditional shifts towards the pragmatic presuppositions of the utterance, making these constructions similar to constructions with modal verbs expressing volitional or deontic modality. So in example (5), the positive evaluation expresses a wish, which, however, is not to be understood as the speaker’s personal desire, but refers deontically to a specific rule of the shared community. An alternative paraphrase is therefore (5’’): (5’’) Bisognerebbe tenerli al guinzaglio, però. one should keep them on a leash, you know

This phenomenon has interesting formal counterparts and can be captured in terms of grammaticalization. One important formal counterpart is the anteposition of the evaluation, which reflects the fact that the framing function within the construction is taken over by this part of the construction rather than by the embedded proposition. Other formal characteristics regard the syntactic coding of the embedded proposition. So markings such as que + subjonctif or che + congiuntivo formally reflect the conceptualization of the embedded proposition as a possible fact rather than as a state of affairs, according to the distinction made above, and syntactically reflect semantic integration into the frame set by the evaluation. In the Italian case, the choice of the congiuntivo presente, which superficially violates a sequence of tense rule,17 is evidence for the attenuating conditional’s shift towards the deictic center.

17. The choice of congiuntivo presente or imperfetto in a complement clause follows a sequence of tenses rule: a main clause in the past requires congiuntivo imperfetto in the subordinate clause, whereas a main clause in a non-past tense requires congiuntivo presente. A hypothetical conditional in the main clause counts as a past tense and requires therefore congiuntivo imperfetto.

215

216 Johanna Miecznikowski and Carla Bazzanella

3.

The attenuating conditional as an interactional resource

In the preceding section, we have described the condatt as a form which operates on an element of its immediate co-text, i.e. the presuppositions of the utterance that it is part of. In this section, we intend to show that the condatt constitutes a linguistic resource for the management of interaction on multiple levels.

3.1 Attenuation and management of interpersonal relations According to the analysis of the condatt in traditional grammar and in pragmatics, this form is essentially a softener. It commutes in many cases with the indicative present and expresses attenuation – in speech act theoretical terms, attenuation of illocutionary force – compared to the latter. This function is relevant in the interpersonal dimension of interaction with respect to politeness and face-management (cf. Brown, Levinson 1978/1987; Roulet 1993; Kerbrat-Orecchioni 2001, 2004). The semantic and pragmatic analysis we’ve proposed above can explain this attenuating effect: questioning pragmatic presuppositions means speaking with caveats and may be a means to avoid imposing oneself (cf. Lakoff 1973) and to reduce persuasiveness (cf. Sbisà 1999). However, it suggests that, when the presuppositions in question are related to the on-going interaction, effects on the interpersonal level are neither necessarily a consequence of attenuation nor necessarily limited to face management. We’d rather insist in the conditional’s expressing negotiability of the speaker’s acts, thereby establishing the relation between speaker and hearer(s) as being one of cooperation and co-construction. Our brief discussion of volere/vouloir, verbs of saying and evaluative predicates shows that the condatt can be used to index the intersubjective component of the speaker’s desires, axiological judgments and communicative actions, underlining their interactive negotiability. Further examples of these predicate types are to be found in example (6): (6) (continuation of ex. 3, meeting of a parents’ committee) (franceschini_5) 1 Q va=be’ al secondo punto all’ordine del giorno c’è la -- la lettera-o.k. the second point on the agenda is the -- the letter -2 di dimissione da parte del - del presidente, - attuale del XX. of resignation from the - the actual president, - of the XX. 3 D no se - permettete se sembro io il primo destinatario, vorrei no if – allow me if I seem to be the primary addressee, I wantcond 4 - se non c’è niente in contrario parlare > (xx). to speak - if there are no objections > (xx).

The attenuating conditional

5 6 7 8 9 10

11 12

13 14

15

Q vorrei ch` chiedere se l’hanno ricevuto tutti questa lettera. wantcond to a’ < I wantcond to> ask if everybody has got that letter. ((4 sec. of murmur)) B allora il XXX se ne può anchee andare, (in fondo). so in fact the XXX can leave, (after all). Q no. sarebbee positivoo il comitato XX rimanesse anche qui no. it would be positive Ø the XX committee stayed here too ((4 sec. of murmur, several instances of “yes” are to be heard)) D ce l’hanno=ce l’hanno tutti la lettera?---All(:h)ora visto che sono does everybody have=does everybody have the letter? --- So(:h) since I am il primo destinatario mi permetto di essere il primo a parlare,the primary addressee I take the liberty of talking first, allora. io direi,-- che eem- vorrei premettere una cosa. che - il o.k. I’d say, -- that erm – I wantcond to say something in advance. that signor nardini ((2 lines omitted)) mi ha dato delle prove, - veramente Mr. Nardini ((2 lines omitted)) has given me, - really particolari di <>coraggio - di <>intelligenza - di <>attaccamento - e strong proofs of <>courage - of <>intelligence - of <>attachment - and soprattutto, - quello che mi ha fatto molto piacere di serietà. -above all, - what I have appreciated very much of correctness. --

As mentioned earlier (2.2), vorrei with verbs of saying (ex. 6, l. 3, 5, 12) alludes to possible obstacles against the realization of the intended conversational act. It is therefore easily interpreted as a request for permission (cf. also its combination with “permettete” at l. 3), attributing decisional power to the interlocutor and modeling the speaker-hearer relation accordingly. Direi, on the other hand, focuses on the hearer’s reception as a prerequisite for a successful act of saying; this makes it well-suited to attenuate statements like the one in ex. (6), l., 12–15, in which the speaker’s subjective stance is expressed very clearly. The evaluative construction sarebbe positivo + Ø + congiuntivo imperfetto at l. 8, then, expresses a recommendation whose committing character is mitigated in at least two ways: (a) the speaker has chosen an impersonal construction; (b) the hypothetical reading of the construction and the choice of the congiuntivo imperfetto suggest that the committee’s further participation is a mere hypothesis. According to the analysis of the condatt proposed here, the conditional’s “attenuating” reading, however, has an additional effect. Impersonal essere positivo conveys the idea of intrinsic or at least obvious positiveness and presupposes, or better postulates, shared criteria of evaluation. The condatt may be interpreted as weakening this presupposition,

217

21 Johanna Miecznikowski and Carla Bazzanella

alluding to a possible divergence of perspectives between the speaker and the hearer(s). The analysis of the condatt proposed here may be further extended to deontic statements. Consider example (7), l. 5: (7) (discussion during a meeting of a group of historians, after a series of presentations made by members of the group) (HR 07117 a3)18 ((the discussion has been going on for a certain time)) 1 GAU xxx je peux intervenir un peu sur euh . sur les deux xxx I can say a few words with regard to er . to the two 2 dernières euhm . communications que j’ai trouvées tout à last er . papers that I found really 3 fait in- tout à fait intéressantes/ .. <et importantes in- really interesting/ .. . pour souligner une: . distinction que nous ((rapid))> . underlining a: . distinction 5 aurions PEUT-être intérêt à faire\ maybe it would be in our interest to make ((turn continues: 48 lines omitted)) 6 c’est c’est c’est c- ce ce ce mécanisme xxx faut pas le perdre it’s it’s it’s th- we shouldn’t loose sight of this mechanism/ 7 de vue/=hein c’est un mécanisme euh: xxx (constant)\ =you see that’s a mechanism euh: xxx (constant) 8 BRU faut peut-être ajouter que le grand homme c’est c’est maybe one should add that the great man is 9 [ x] dont on cite le nom et tout le monde& [x] whose name is cited and everybody& 10 X [((sneeze))]

1. The following transcription conventions have been used for example (7): [ ] begin/end of overlap . .. ... pauses (< 2 sec.) (2s) pauses in seconds (> 2 sec.) xxx incomprehensible segment / \ rising/falling intonation : stretched vowel par- truncation & continuation of turn across line break = rapid continuation (il va) tentative transcription (h) inbreath wIrklich vowel in cap.s: high pitch exTRA dynamic accent < > delimitation of phenomena indicated between (( )) ((laughs)) phenomena not transcribed, comments by the transcriber N.N. name

The attenuating conditional 219

11 BRU 12 GAU 13 GAU 14 BRU 15 16 17 DUM 18 19 20 21 GAU

& sait[=euh . de qui il s’agit &knows [=who you’re talking about [oui [yes m[hm [pour Regulus on a également d- des allusions qui sont [in the case of Regulus we also have a- allusions that are vraiment xx dans une SEquence xxx où on cite que son nom\ really xx in a sequence xxx where it’s just his name that is cited\ (3 s.) je crois qu’cette distinction est effectivement très très I think that this distinction is very very important indeed/ importante/ mais . on pourrait peut-être l’enrichir aussi but . we could maybe further enrich it taking into de la considération du CONtexte dans lequel est énoncé consideration the CONtext in which le[: ]le personnage ou la vertu correspondant/ the[:] the corresponding figure or vertue is mentioned/ [hm] ((turn continues for another 10 lines))

The lexical semantics of the complex predicate avoir intérêt à (l. 5) allows the speaker to express a deontic statement motivated by an evaluative judgment about what is in the group’s interest, the group being referred to by inclusive nous (cf. Bazzanella 2002). The conditional, like sarebbe positivo in example (6) discussed above, allows the speaker not to postulate shared criteria of evaluation. On the other hand, it indicates that the speaker does not presuppose the group’s readiness to comply with the obligation expressed. The presupposition that the agent is able and disposed to behave in accordance with a given obligation imposed on him/her is in fact associated with various deontic predicates, including the modals dovere/devoir, when used in the indicative (cf. Miecznikowski forth. c). The use of the conditional cancels that presupposition (de re or pragmatically, with regard to agents that participate in the on-going interaction). With these predicates, it is therefore compatible, much better so than the present indicative, with an eventual or factual behavior of the agent that is in contradiction with the expressed obligation. Speakers typically choose devoir/dovere + condatt, for instance, when comparing actual behavior to expected and desired behavior. Consider e.g. the following Italian example, where a substitution of the conditional (dovrebbero, l. 2) by the present indicative would be semantically odd (de re):

220 Johanna Miecznikowski and Carla Bazzanella

(8) (lip rc 8) 1 secondo le persone che sono state intervistate eh consigli comunali according to the people that have been interviewed the city councils 2 le amministrazioni locali non si preoccupano_ come dovrebbero the local administrations don’t engage themselves as they should.

We’ll conclude this section by discussing one more modal verb with which the condatt is used frequently, i.e. potere/pouvoir expressing agent-oriented (Bybee et al. 1994) possibility (cf. ex. 7, l. 18). The agent-oriented use of potere/pouvoir in the indicative asserts the fact that external circumstances make a state of affairs q possible (participant external possibility, cf. Bybee et al. 1994), presupposing that a specific agent is taking q into account as a valuable option of action for him/herself. By means of the conditional, the speaker cancels that presupposition, indicating that he/she cannot take the agent’s readiness of performing q for granted. As example (7) illustrates (l. 18), this operation may have effects on the interpersonal level. Pouvoir is used here to express an option of action of the copresent participants (referred to by inclusive on); the choice of the condatt underlines the negotiability of the suggestion made, alluding to the possibility that the interlocutors may not consider q a relevant option of action and/or may not be willing to realize it.

3.2 Thematic and argumentative development within turns First results of our study, which at the moment is more advanced for French than for Italian (for French, cf. the quantitative results as to the distribution of vouloir presented in Miecznikowski forth. a), suggest that the interpersonal dimension of interaction is not the only dimension of local context that is relevant for the use of the conditional when it functions as a modalizer of presuppositions. We will argue that the condatt contributes also to the thematic and argumentative organization of turns and to the sequential organization of conversation. First of all, turn-internally, the condatt is frequently combined with structuring devices signaling thematic boundaries and change of perspective. More precisely, it is used to encode what follows immediately these markers, i.e. an information that is particularly new or different in some respect from what precedes. This function appears e.g. –

in example (6), l. 12, where “io direi” is preceded by allora + falling intonation and contributes to mark the passage from preliminaries concerning the management of the conversational floor to the first point the speaker wants to make;

The attenuating conditional

–

–

in example (7), l. 5, where the conditional accompanies the transition from a part of the turn that reacts to preceding turns to – again – the first point the speaker wants to make; or in example (7), l. 18, where “on pourrait”, preceded by mais, has a similar function; it introduces a slightly contrasting opinion and a new topic, “CONtexte” (l. 19), a thematic and argumentative shift that the interlocutor, GAU, acknowledges by producing the back-channel “hm” (l. 19).

In an important number of cases, moreover, the attenuating conditional is part of metacommunicative utterances that announce subsequent talk and therefore have structuring functions themselves. Such utterances include performative uses of verbs of saying in the conditional, e.g. “io direi” (ex. 6, l. 12) and modal verbs in the conditional + verb of saying, e.g. “vorrei premettere una cosa” (ibid.). In contrast, speakers prefer modals in the indicative present when developing/reformulating their discourse, e.g. “il faut pas le perdre de vue” in ex. (7), l. 6– 7. This goes for metacommunicative utterances, too: when these contain modals in the indicative present, they tend to underline continuity rather than change. An example is parenthetical je veux dire (cf. Miecznikowski forth. a) / voglio dire; it would be interesting, moreover, to analyze in this perspective other expressions such as je dis / dico or je dois dire que / devo dire che / devo dire una cosa. Consider, for instance, “faut peut-être ajouter que” in example (7), l. 8, in which the choice of the indicative contributes to mark thematic and argumentative continuity across a turn boundary – an effect that is reinforced, in this case, thanks to modal and lexical priming (cf. Bazzanella 1994) by the immediately preceding “il faut pas le perdre de vue” (l. 6–7). The condatt at thematic and argumentative boundaries may be seen as part of sequence-initial focusing activities (Kallmeyer 1978): it contributes to the participants’ moving their focus of attention from one topic to the next or from one argumentative position to the next. This solidarity with focusing activities may partly be explained by the interpersonal functions we have underlined in the preceding section. One might hypothesize that the condatt underlines interactive negotiability at points in discourse where the hearer’s reception and/or consensus is particularly important for the speaker. On the other hand, the condatt’s function is more specific than that, since it seems to be specialized to signal thematic and argumentative change and does not accompany any discursive operation that can reasonably be assumed to be strongly recipient-designed (e.g. paraphrastic reformulations). In any case, if the condatt’s interpersonal functions are relevant for sequence-initial (re)focusing, they are so in a way that is largely unrelated to attenuation: what is typical for fo-

221

222 Johanna Miecznikowski and Carla Bazzanella

cusing activities is up-grading (Relevanzhochstufung in Kallmeyer’s 1978 terms) rather than down-grading (Relevanzrückstufung). To understand the condatt’s specificity at thematic and argumentative boundaries, it is helpful to take into account the anaphorical dimension of presuppositions. Presuppositions are a crucial element linking an utterance to the discourse common ground (cf. Fetzer & Fischer 2007), among others to preceding co-text as an important element of the common ground. Van der Sandt (1989), for instance, argues convincingly that a tight relationship holds between presupposition induction and anaphor, not only for pronouns and definite descriptions, but also in other less well explored cases: Those lexical and syntactic constructions that are traditionally called presupposition inducers can be conceived of as anaphors, only differing from normal denoting pronouns in that they do have descriptive content. This property enables them to create a discourse marker at some level of representation in case the context does not provide an accessible marker to establish an anaphoric link. With respect to definite descriptions this view is found in various authors. It can moreover easily be generalized to other types of presupposition inducers as cleft constructions, factives, aspectual verbs and presuppositional adverbs by introducing reference markers for states and events. (Van der Sandt 1989: 287)

In our view, Van der Sandt’s general claim may be usefully applied also to the presupposition triggering uses of modal predicates and the pragmatically presupposing performative use of speech act verbs and the like. The presupposition of shared criteria of evaluation, for example, in the case of deontic falloir used in the indicative present in example (7) (“ce méchanisme xxx faut pas le perdre de vue”, l. 6–7) can be seen as containing an instruction to look for indications in the preceding context – which is shared in the sense that it is textually given – that are suited to specify which evaluative criteria the speaker is using, and to adopt these also for the current utterance. Il faut thus establishes an anaphoric link to preceding context, in this case to a series of arguments showing why the mechanism (“méchanisme”) referred to is important. Conversely, doubting or avoiding presuppositions means not establishing any anaphorical link to argumentative and pragmatic frames that were valid for preceding utterances. In the same example, the condatt used in the noun phrase “une: . distinction que nous aurions PEUT-être intérêt à faire” (ex. 7, l. 5) indicates that the criteria according to which the speaker makes his deontic statement are not shared in the sense that they are not shared yet. The speaker does not refer back to arguments given in the presentation to which he is reacting, but invites the hearer to construct a new thematic and argumentative frame, which will confer intelligibility and plausibility to the deontic statement made.

The attenuating conditional 223

3.3 Interactional sequences The condatt’s “reset” function observed with regard to thematic development within turns is relevant also for another aspect of interaction, i.e. the level of adjacency pairs and preference structure. In the corpus examined, we often find the condatt at sequence beginnings (in first pair parts) and in dispreferred moves. We hypothesize that by virtue of its semantic and pragmatic properties it contributes in fact to signal the non-reactive nature of conversational moves, accentuating sequential discontinuity and prospective orientation. In the examples (6) and (7), this function is illustrated by the already mentioned tokens of the conditional marking the transition to the first point a speaker wants to make. Such “first points” in discussions can be considered as highly initiative, in contrast to the reactive turn parts that precede them. In other cases, the condatt occurs in turns that occupy the position of a sequentially implicated second pair part, but are in fact first pair parts, thus deceiving the expectations created by a previous speaker. Instances of this kind of dispreferred sequential progression are the turns at l. 3-4 (“vorrei {…} parlare”) and l. 5 (“vorrei chiedere”) in example (6). Furthermore, we find the condatt in dispreferred second pair parts, e.g. “no. sarebbe positivoo il comitato XX rimanesse anche qui” in example (6) (l. 8), which refuses a suggestion made by the previous speaker. In contrast, the condatt is less often used to express preferred reactions or to signal continuation of a previous activity. It is probable that in linguistic cotexts and sequential contexts that strongly favor an interpretation of a turn in that direction interlocutors must make additional efforts to interpret instances of the condatt. An examination of corpus data shows, in fact, that in these contexts speakers regularly recur to modals or verbs of saying in the indicative present rather than in the conditional. A first example illustrating this pattern is the metacommunicative announcement “je peux intervenir” in ex. (7), l. 1. This announcement is preceded by a 5 second pause (not reported in the transcript), before which the discussion has been conducted exclusively in German. It is highly probable that during that pause the chairperson makes a non-verbal move to encourage Gaurdard to speak. Not only do chairpersons intervene regularly after long pauses to make discussions advance. In the multilingual corpus examined here, they intervene, in particular, to encourage language change after long stretches of discussion in one language (cf. Miecznikowski, Mondada, Müller & Pieth 1999). Since Gaudard has a particularly high status among the French speaking participants, it is probable, moreover, that an invitation to speak first (in French) be addressed to him specifically. In this context, his metacommunicative announcement is to be analyzed as a reactive utterance.

224 Johanna Miecznikowski and Carla Bazzanella

As a second example, we will reconsider BRU’s metacommunicative announcement “il faut {…} ajouter” in ex. (7), l. 8. We have observed earlier that it establishes thematic continuity. On the sequential level, this corresponds to the fact that the speaker introduces a subordinate side-sequence, which expands GAU’s previous turn (l. 1–7). This status of BRU’s turn emerges progressively thanks to the contributions of BRU’s interlocutors at l. 12, 13 and 17. In particular, DUM’s utterance in l. 17 shows that the side-sequence maintains the expectance of an answer to GAU’s turn preceding the side-sequence: The expression “cette distinction” refers back to the distinction introduced by GAU at l. 4–5, underlining discourse cohesion by lexical repetition19 and by the choice of the determiner. Moreover, it retrospectively confirms the initiative character of GAU’s “first point” made by means of an utterance in the condatt. A typical case, finally, in which we find modals or verbs of saying in the indicative present is that of repeated metacommunicative announcements. An example is “mi permetto di essere il primo a parlare” (ex. 6, l. 11), which echoes the previous announcement “vorrei – se non c’è niente in contrario parlare” (ex. 6, l. 3–4). By choosing the indicative mode, the speaker underlines that he/she is not actually beginning a new sequence, but is merely taking up an opening procedure initiated earlier, and acknowledging that his/her right to speak has been negotiated in the meantime.

4.

Discussion: Understanding the contextual constraints on the use of condatt

In the preceding section, we have underlined –

–

–

the value of the condatt as a marker of negotiability, which alludes to the hearer’s possibility of having diverging attitudes and of impeding an act intended by the speaker; the condatt’s local context changing potential as an element of sequence-initial focusing activities (within turns and in the management of complex sequences); its local context changing potential as an element contributing to mark the non-reactiveness of conversational moves.

The interaction-related functions we have individuated are crucial both for understanding the “synergies” (Bazzanella, Caffi & Sbisà 1991) of relevant local pa-

19. Cf. Bazzanella (1996) for an introduction to the functions of repetition in dialogue.

The attenuating conditional 225

rameters with the uses of the condatt and for making hypotheses about the form’s solidarity with parameters of global context. On the local level, these functions correspond directly to a prototypical ordering of context types, from highly typical and frequent contexts to less typical contexts in which the use of the condatt is less frequent and may give rise to inferences, to contexts in which the condatt’s is not attested or occurs only marginally, and is incoherent or hardly interpretable. Prototypical local contexts are contexts in which several components commonly associated with the condatt co-occur: reference to certain types of acts (future acts by interaction participants, acts that are part of the on-going interaction), certain sequential positionings (placement after a second pair part, placement after a preliminary/reactive part of the turn) as well as certain types of thematic and argumentative developments (introduction of new topic, topic change or contrast/disagreement signaled by lexico-grammatical means and/or discourse markers). These local contexts can be assumed to have played a crucial role in the condatt’s diachronic development, the condatt taking over contextually expressed functions by ways of metonymical semantic/pragmatic change (Hopper & Traugott 1993). Less prototypical contexts are those in which only some components are present, whereas others are unmarked or clearly absent. An example is second pair part position, a case in which the condatt’s non-reactive function may be re-interpreted as signaling dispreference, and the function of expressing contrast and disagreement may be reinforced. Highly untypical contexts, finally, are those in which the condatt’s functional potential enters in contradiction with several contextual components because the presuppositions it alludes to are contextually given and cannot easily be construed as questionable. Examples are co-texts that explicitly mark the expressed proposition as being part of the common ground, e.g. conjunctions such as French puisque ‘since’ or Italian dato che / visto che ‘given that’, with which the condatt occurs only once in our corpora. On the sequential level, our corpus data suggest that positioning in second pair parts is highly untypical when these parts cannot be interpreted as dispreferred. Given the way the condatt operates on local context, we may formulate hypotheses, furthermore, as to its “embeddedness” (Fetzer 2004: 234) in different types of global context – passing from a micro level of analysis towards a macro level of analysis where different kinds of sociolinguistic variation play a role. A first hypothesis could be to assume that, since the condatt is an only weakly presupposing form, it is better compatible with a reduced common ground than with a rich one, and that, as a consequence, its use is sensible to parameters such

226 Johanna Miecznikowski and Carla Bazzanella

as the degree of acquaintance and the degree of social distance between co-participants. A second relevant aspect is the condatt’s function of signaling co-construction and negotiability of acts, of attitudes and of the organization of turns and sequences, which accentuates the nature of conversation as “collaborated action” (Fetzer 2004: 175; cf. also Duranti 1986). We might expect it to occur typically, by virtue of this macro-function, in text genres and interaction types in which coconstruction plays an important role, in particular: –

–

– –

in dialogue rather than in monologue: for instance in face-to-face interaction, written correspondence (on paper, by e-mail or on internet forums), in direct reported speech in fiction; within dialogic genres, in activity types characterized by symmetrical roles, rather than (institutionalized) activity types with clearly asymmetrical roles such as doctor-patient interaction, military training etc.20 in situations in which decision-making is an important communicative goal; in situations in which the sequential development itself tends to be negotiated explicitly (e.g. complex multi-party participation structures with the necessity of managing long turns, rather than dyadic interactions with rapid turn taking).

Evidence in favor of the hypotheses formulated above might be gathered for example by frequency measurements in corpus-based genre analysis. Observed and hypothesized associations of the condatt with local and global contextual parameters may be conceptualized as pattern recurrences which correspond to degrees of markedness: the more prototypical the context in which the condatt occurs, the more unmarked is its choice. Appropriateness, on the other hand, is a qualitatively different notion: it is not gradual, but has to do with the choice of interactants or analysts to accept or not to accept communicative acts. As outlined in the introduction, we consider appropriateness an interactively accomplished property of communicative acts. This means, among others, that we take the interactants’ perspective on appropriateness to be a primordial phenomenon, which the analyst will build on when 20. It must be added that the extent to which language users consider that a text or activity type involves the hearer’s active cooperation is culture-specific in the largest sense. For example, didactic discourse may be more directive or more co-constructive depending on underlying educational models; oral narratives may be more or less interactive depending on the cultural background of the co-participants. The appropriateness of a form such as the condatt may be expected to vary accordingly, and conversely might function as a contextualization cue with regard to cultural background.

The attenuating conditional 227

modeling appropriateness conditions as a “second order construct” (Schütz 1971 [1953]). The analyses in this chapter have shown that the interlocutors’ reactions to utterances in the condatt retrospectively display their interpretation of these utterances and can therefore usefully be taken into account by the analyst when assigning determinate functions to this form – confirming a key tenet of interactional linguistics and conversation analysis. With regard to appropriateness, the interlocutors’ reactions can be seen as displaying not only specific interpretations of an act, but also, more generically, its up-take and its acceptance as being appropriate. By default, as it were, acceptance concerns all aspects of the act, its linguistic form included. Acceptance contrasts with the absence of any projected reaction, the most generic signal of non-acceptance, and with various kinds of repairs and recontextualizations (cf. below), which may signal inappropriateness in a more specific way. In the interactive accomplishment of appropriateness, the degree of markedness of a linguistic item with regard to various local and global contexts becomes relevant as an element of participants’ communicative competence, allowing them to detect and interpret potentially inappropriate items and to signal inappropriateness in a recipient-designed way. We will discuss these aspects briefly in what follows. What interactants know about the markedness of a linguistic item in a given context (and about the markedness, with regard to that context, of eventual subsequent actions it projects) allows them to assess the necessity of evaluating that item’s appropriateness in the first place. The question of an item’s appropriateness arises in fact as a problem for interactants only when its global context and/or the preceding local context are in some respect atypical. The rejection of a form as inappropriate (be it by interlocutors or by the speaker, via self-correction) is one of several possible outcomes of inferencing triggered by its use in an atypical context. In the process of evaluating a linguistic item’s appropriateness, especially in the case of polysemous and polyfunctional forms like the conditional, a crucial problem is disambiguation. This problem arises, of course, above all on the hearer’s side: any appropriateness judgment presupposes a reduction of the to-bejudged utterance’s range of possible interpretations and an – at least tentative – attribution of speaker intentions to it. Such a reduction is possible only on the basis of knowledge on recurrent associations between meanings and context types (cf. e.g. the prototypical contexts of the conditional’s attenuative reading listed above, which contrast with those of the temporal, quotative, hypothetical and inferential readings), and between a given meaning’s pragmatic functions and context types (the specific functions fulfilled by the condatt tend to differ, e.g., depending on

22 Johanna Miecznikowski and Carla Bazzanella

whether its host utterance constitutes a turn of its own or is placed in the middle of a long turn). Finally, in the (rare) case a participant decides to manifest a problem regarding a particular linguistic choice, knowledge about prototypical contexts plays a role in the recipient design of the problem manifestation. Typical examples are recontextualizations, i.e. evident and abrupt changes of the contextual frame within which to interpret an already produced utterance, and explicit categorizations in terms of speaker roles and social categories. These devices can be used by participants to signal what kind of context a problematic linguistic item typically evokes for them, positing that the form-context association is part of a (at least partially) shared norm. The item’s inappropriateness is signaled by evaluating the evoked context negatively and/or by foregrounding its incompatibility with the actual context. If it is the hearer to signal inappropriateness, the construction of conflicting contextualizations can function as a hetero-initiation of repair, making explicit the type of problem posed by the problematic item and orienting the hearer towards possible solutions. *** To conclude, in our description of the attenuating conditional’s discourse functions as well as in our final discussion, we have underlined the significance of the complex interplay between language system/language use/communicative action and the high degree of interrelatedness of different levels of analysis, mainly with regard to the following points: –

–

–

The form’s use as an interactional resource is closely related to its semantics as a polysemous linguistic sign, in particular to its capacity of selecting an utterance’s presuppositions as an implicit modal ground. The condatt functions as an operator on local context in culturally specific (cf. footnote 20) social practices. Correlations between the condatt as a linguistic form and components of local and global context must be seen as mediated by its interpersonal, thematic and sequential functions. Appropriateness/inappropriateness emerges in situ, and requires interactants to use socially construed shared knowledge on pattern recurrence and formcontext associations.

In a nutshell, meaning assignment, inferencing and the calculus of appropriateness must be seen as context-dependent both in the sense that they are embedded in the process of socially situated interaction and in the sense that they rely on a communicative competence allowing speakers to construct relevant contexts for conversational events; as Fetzer (2004: 235) states, “The analyses of well-formedness, acceptability and appropriateness, require a frame of investigation which goes be-

The attenuating conditional 229

yond the constraints and requirements of a sentence grammar […] they require a context-sensitive approach to the investigation of language and language use”.

References Akman, V., Bouquet P., Thomason, R. and Young, R.A. (eds.). 2001. Modeling and Using Context. Berlin: Springer–Verlag. Akman, V. and Bazzanella, C. 2003. “The complexity of context”. In Akman, V. and Bazzanella, C. (eds.), On Context, Journal of Pragmatics, special issue 35: 321–329. Andler, D. 2003. “Context: The case for a principled epistemic particularism”. In Akman, V. & Bazzanella, C. (eds.), On Context, Journal of Pragmatics, special issue 35: 349–371. Athanasiadou, A. and Dirven, R. 2000. “Pragmatic conditionals”. In Foolen, A. and van der Leek, F. (eds.), Constructions in Cognitive Linguistics. Selected Papers from the Fifth International Cognitive Linguistics Conference, Amsterdam, 1997, 1–26. Amsterdam/Philadelphia: John Benjamins. Auer, P. and Di Luzio, A. (eds.). 1992. The Contextualization of Language. Amsterdam/Philadelphia: Benjamins. Banca dati dell’italiano parlato. Consulted in November 2005 at http://languageserver.uni-graz. at/badip/. Barwise, J. and Perry, J. 1983. Situations and Attitudes. A Bradford Book. Cambridge, MA: MIT Press. Bazzanella, C. 1990. “’Modal’ uses of the Italian indicativo imperfetto in a pragmatic perspective”. Journal of Pragmatics 14 (3): 237–255. Bazzanella, C. 1994, 21996, 31997. Le facce del parlare. Firenze: La Nuova Italia. Bazzanella, C. 1996 “Introduction”. In Bazzanella, C. (ed.), Repetition in Dialogue, vii–xvii. Tübingen: Niemeyer. Bazzanella, C. 1998. “On context and dialogue”. In S. Čmejrková, J. Hoffmannová, O. Müllerová, J. Světlá (eds.), Dialogue in the Heart of Europe, 407–416. Tübingen: Niemeyer. Bazzanella, C. 2000. “Tense and meaning”. In Marconi, D. (ed.), Knowledge and Meaning – Topics in Analytic Philosophy, 177–194. Vercelli: Mercurio. Bazzanella, C. 2002. “The significance of context in comprehension: The ‘we case’”. Foundations of Science 7: 239–254. Bazzanella, C. (ed.) 2002 Sul dialogo. Contesti e forme di interazione verbale. Milano: Guerini e associati. Bazzanella, C. 2004. “Atténuation et intensification dans l’Italien: dimensions et configuration pragmatique”. In Maria Helena Araùjo Carreira (ed.), Plus ou Moins!? L’Atténuation et l’Intensification dans les Langues Romanes (Travaux et documents 5), 173–200. Paris: Université Paris 8 Vincennes Saint-Denis. Bazzanella, C. 2005. Linguistica e Pragmatica del Linguaggio. Roma-Bari: Laterza. Bazzanella, C., Caffi C. and Sbisà, M. 1991. “Scalar dimensions of illocutionary force”. In I.Z. Zagar (ed.), Speech Acts. Fiction or Reality?, 63–76. Ljubljana: IPRA distribution Center for Jugoslavia. Berretta, M. 1992. “Sul sistema di tempo, aspetto e modalità nell’italiano contemporaneo”. In Moretti, B. et al. (eds.), Linee di Tendenza dell’Italiano Contemporaneo, 135–153. Roma: Bulzoni.

230 Johanna Miecznikowski and Carla Bazzanella

Bonu, B., Mondada, L. and Relieu, M. 1994. “Catégorisation: l’approche de H. Sacks”. In Fradin, B., Quéré, L. and Widmer, J. (eds.), L’Enquête sur les Catégories. Raisons Pratiques, 5, 129–148. Paris: Editions de l’EHESS. Bouquet, P. et al. 2003. “Theories and uses of context in knowledge representation and reasoning”. In Akman, V. and Bazzanella, C. (eds.), On Context, Journal of Pragmatics, special issue 35: 455–484. Bouquet, P., Serafini, L., Brézillon, P, Benerecetti, M. and Castellani, F. (eds.) 1999. Proceedings of the Second International and Interdisciplinary Conference on Modelling and Using Context. Berlin: Springer. Brown, P. and Levinson, S. 1978/1987. Universals in language usage: politeness phenomena. In Goody, E.N. (ed.), Questions and Politeness. Strategies in Social Interaction, 56–248. Cambridge, Cambridge University Press; republished 1987: Politeness. Cambridge University Press: Cambridge. Bybee, J., Perkins, R. and Pagliuca, W. 1994. The Evolution of Grammar. Tense, Aspect and Modality in the Languages of the World. Chicago: The University of Chicago Press. Clark H.H. 1996. Using Language, Cambridge, Cambridge University Press. Comrie, B. 1986. “Conditionals: A typology”. In Traugott, E., ter Meulen, A., Snitzer Reilly, J. and Ferguson, C.A. (eds.), On Conditionals, 77–99. Cambridge: Cambridge University Press. Couper-Kuhlen, E. and Kortmann, B. (eds.). 2000. Cause, Condition, Concession, Contrast. Berlin: Mouton de Gruyter. Dancygier, B. and Sweetser, E. 2005. Mental Spaces in Grammar. Conditional Constructions. New York: Cambridge University Press. De Mauro, T., Mancini F., Vedovelli M. and Voghera M. 1993. Lessico di Frequenza dell’Italiano Parlato-LIP, Milano: EtasLibri. Dendale, P. 2001. “Les problèmes linguistiques du conditionnel en français”. In Tasmowski, J. and Dendale, P. (eds.). Le Conditionnel en Français, 7–20. Metz: Université de Metz. Dik, S. 1989. Functional Grammar. Part I: The Structure of the Clause. Dordrecht: Foris publications. Duranti, A. 1986. “The audience as co–author: An introduction”. Text 6: 239–247. Duranti, A. and Goodwin, C. (eds.). 1992. Rethinking Context: Language as an Interactive Phenomenon. Cambridge: Cambridge University Press. Edmonds, B. and Akman, V. (eds.). 2002. Context in Context, Foundations of Science, special issue 7: 233–364. Fetzer, A. 2004. Recontextualizing Context. Grammaticality Meets Appropriateness. Amsterdam/ Philadelphia: John Benjamins. Fetzer, A. 2007. “Reformulation and common grounds”. In Fetzer, A. and Fischer, K. (eds.), Lexical Markers of Common Grounds. 159–182. Amsterdam: Elsevier. Fetzer, A. “Introduction”. This volume. Fetzer, A. and Fischer, K. (eds.). 2007. Lexical Markers of Common Grounds. Amsterdam: Elsevier. Franceschini, R. 1998. Riflettere sull’Interazione. Milano: FrancoAngeli. Giglioli, P.P. (ed.). 1972/1973. Language and Social Context. Harmondsworth: Penguin. Givón, T. 1989. Mind, Code and Context. Essays in Pragmatics. Hillsdale: Lawrence Erlbaum. Givón, T. 2005. Context as Other Minds. The Pragmatics of Sociality, Cognition and Communication. Amsterdam hiladelphia: John Benjamins. Grice, H.P. 1989/1993. Studies in the Ways of Words. Cambridge: Cambridge University Press.

The attenuating conditional

Gumperz, J. 1977. “Sociocultural knowledge in conversational inference”. In Saville-Troike, M. (ed.), Linguistics and Anthropology, 191–211. Washington: Georgetown University Press. Haegeman, L. 1984. “Pragmatic conditionals in English”. Folia Linguistica, 18, 485–502. Haillet, P. 2002. Le Conditionnel en Français: une Approche Polyphonique. Paris: Ophrys. Hopper, P. and Traugott, E.C. 1993. Grammaticalization. Cambridge, Cambridge University Press. Kallmeyer, W. 1978. “Fokuswechsel und Fokussierungen als Aktivitäten der Gesprächskonstitution”. In Meyer-Hermann, R. (ed.), Sprechen – Handeln – Interaktion, 191–241. Tübingen: Max Niemeyer. Kaplan, D. 1977. “Demonstratives. An essay on the semantics, logic, metaphysics, and epistemology of demonstratives and other indexicals”. In Almog J., Perry J. and Wettstein H. (eds.), Themes from Kaplan, 481–565. Oxford: Oxford University Press. Kerbrat-Orecchioni, C. 2001. Les Actes de Langage dans le Discours. Théorie et Fonctionnement. Paris, Nathan. Kerbrat-Orecchioni, C. 2004. “L’adjectif “petit” comme procédé d’atténuation en français”. In Araùjo Carreira, M.H. (ed.), Plus ou Moins!? L’Atténuation et l’Intensification dans les Langues Romanes (Travaux et Documents 5), 153–175. Paris: Université Paris 8 Vincennes Saint-Denis. Kerbrat-Orecchioni, C. 2005. Le Discours en Interaction. Paris: Armand Colin. Lakoff, R. 1973. “The logic of politeness; or minding your p’s and q’s”. In Papers from the Ninth Regional Meeting of the Chicago Linguistic Society: 292–305. Chicago: Chicago Linguistic Society. Leonardi, P. (forth.). “Presupposizioni e implicature”. Preprint consulted in December 2005 at http://www.utenti.dsc.unibo.it/~leonardi/pl/presupposizionieimplicature.doc. Levinson S.C. 1992. “Activity types and language”. In Drew, P. and Heritage, J. (eds.), Talk at Work: Interaction in institutional settings, 66–100. Cambridge: Cambridge University Press. Light, P. and Butterworth, G. (eds.). 1993. Context and Cognition. Ways of Learning and Knowing. Hillsdale: Lawrence Erlbaum. Linell, P. and Thunqvist, D.P. 2003. “Moving in and out of framings: Activity contexts in talks with young unemployed people within a training project”. In Akman, V. and Bazzanella, C. (eds.), On Context, Journal of Pragmatics, special issue 35: 409–434. Lüdi, G. 1991. “Construire ensemble les mots pour le dire. A propos de l’origine discursive des connaissances lexicales”. In Dausendschön-Gay, U., Gülich, E. and Krafft U. (eds.), Linguistische Interaktionsanalysen, 193–224. Tübingen: Max Niemeyer. Lyons, J. 1981. Language, Meaning and Context. London: William Collins & Sons. Martin, R. 21992. Pour une Logique du Sens. Paris: Presses Universitaires de France. Miecznikowski, J., Mondada, L., Müller, K. & Pieth, C. 1999. “Gestion des asymétries et effets de minorisation dans des discussions scientifiques plurilingues”. In Actes du Colloque “Langues Minoritaires en Contexte”, Coire 21–23.9.98, Bulletin Suisse de Linguistique Appliquée, 69 (2): 167–187. Miecznikowski, J. 2005. Le Traitement de Problèmes Lexicaux lors de Discussions Scientifiques en Situation Plurilingue. Procédés Interactionnels et Effets sur le Développement du Savoir. Bern: Peter Lang. Miecznikowski, J. (forth. a). “Modality and conversational structure in French”. In Proceedings of the international IADA conference “Theoretical approaches to dialogue analysis”, Chicago, March 30 – April 3, 2004.

231

232 Johanna Miecznikowski and Carla Bazzanella

Miecznikowski, J. (forth. b). Gli usi del condizionale nel parlato italiano e francese. In Atti del convegno ‘La Comunicazione Parlata’, Napoli, 23.–25.2.2006. Miecznikowski, J. (forth. c). I verbi modali ‘volere’, ‘potere’ e ‘dovere’ come attivatori presupposizionali. In: Atti del IX convegno SILFI, Firenze, 15–17 giugno 2006. Mondada, L. 2001. “Pour une linguistique interactionnelle”. Marges Linguistiques, 1: 1–21. Mondada, L. 2005. Chercheurs en Interaction. Comment émergent les Savoirs. Lausanne: Presses Polytechniques et Universitaires Romandes. Ochs, E., Schegloff, E.A. and Thompson, S.A. 1996. Interaction and Grammar. Cambridge: Cambridge University Press. Oishi, E., 2007. “Austin’s speech act theory and the speech situation”. This volume. Penco C. (ed.). 2002. La Svolta Contestuale. Milano: McGraw–Hill. Perry, J. 1993. The Problem of the Essential Indexical and Other Essays. New York: Oxford University Press. Renzi, L., Salvi, G. and Cardinaletti, A. (eds.). 1991. Grande Grammatica Italiana di Consultazione, vol. 3. Bologna: Il Mulino. Riegel, M. 1996. Grammaire Méthodique du Français. Paris: Presses Universitaires de France. Riehle, A. 2003. “Neural correlates of context–related behaviour.” Journal of Pragmatics 35(3): 485–504. Roulet, E. 1993. “Des formes et des emplois des modalisateurs de proposition dans l’interaction verbale”. In Dittmar, N. and Reich, A. (ed.), Modality in Language Acquisition, 27–39. Berlin: De Gruyter. Sacks, H. 1992. Lectures on Conversation. Oxford: Basil Blackwell. Sbisà, M. 1999. “Ideology and the persuasive use of presupposition”. In Verschueren, J. (ed.), Language and Ideology. Selected Papers from the 6th International Pragmatics Conference, Vol. 1, 492–509. Antwerp: International Pragmatics Association. Schegloff, E.A. Ochs, E., and Thompson, S.A. 1996. “Introduction.” In Ochs, E., Schegloff, E.A. and Thompson, S.A. (eds.), Interaction and Grammar, 1–41. Cambridge: Cambridge University Press. Schütz, A. 1971 [1953]. “Wissenschaftliche Interpretation und Alltagsverständnis menschlichen Handelns”. In Gesammelte Aufsätze I. Den Haag, Martinus Nijhoff. Schwarze, Ch. 1983. “Modus und Modalität”. In Ch. Schwarze (ed.), Bausteine für eine italienische Grammatik, vol. I, 265–302. Tübingen: Gunter Narr. Selting, M. and Couper-Kuhlen, E. 2000. “Argumente für die Entwicklung einer ‘interaktionalen Linguistik’”. Gesprächsforschung, 1: 76–95. Source: www.gespraechsforschug-ozs.de (consulted in April 2003). Sperber, D. and Wilson, D. 1986/1993. Relevance. Cambridge (Mass.): Harvard University Press. Squartini, M. 2001. “The internal structure of evidentiality in Romance”. Studies in Language, 25 (2): 297–334. Squartini, M. (forth.). “L’espressione della modalità”. In Salvi, G. and Renzi, L. (eds.), Grammatica dell’Italiano Antico. Preprint consulted in November 2005 at http://ludens.elte.hu/ ~gps/konyv/modalita.doc. Stalnaker, R. 1999. Context and Content. Oxford: Oxford University Press. Tannen, D. 1989. Talking Voices. Repetition ,Dialogue, and Imagery in Conversational Discourse. Cambridge: Cambridge University Press. Tasmowski, J. and Dendale, P. (eds.). 2001. Le Conditionnel en Français. Metz: Université de Metz.

The attenuating conditional

Traugott, E. C., ter Meulen, A., Snitzer Reilly, J. and Ferguson, C. A. (eds.). 1986. On Conditionals. Cambridge (UK): Cambridge University Press. Traugott, E. and Dasher, R. 2002. Regularity in Semantic Change. Cambridge, Cambridge University Press. Van der Sandt, R. 1989. “Presuppositions and Discourse Structure”. In Bartsch, R., van Benthem, J. and van Emde, B. (eds.), Semantics and Contextual Expression, 267–294. Dordrecht: Foris. Weigand, E., and Dascal, M. 2001. Negotiation and Power in Dialogic Interaction. Amsterdam/ Philadelphia: Benjamins. Wierzbicka, A. 1996. Understanding Cultures through their Key Words. New York: Oxford University Press.

233

Collaborative use of contrastive markers Contextual and co-textual implications Francesca Carota

Institut des Sciences Cognitives-CNRS, Lyon

The study presented in this paper examines the context-dependence and dialogue functions of the contrastive markers of Italian ma (but), invece (instead), mentre (while) and però (nevertheless) within task-oriented dialogues. Corpus data evidence their sensitivity to a acognitive interpersonal context, conceived as a common ground. Such a cognitive state – shared by co-participants through the coordinative process of grounding – interacts with the global dialogue structure, which is cognitively shaped by “meta-negotiating” and grounding the dialogue topic. Locally, the relation between the current dialogue structural units and the global dialogue topic is said to be specified by information structure, in particular intra-utterance themes. It is argued that contrastive markers re-orient the co-participants’ cognitive states towards grounding ungrounded topical aspects to be meta-negotiated. They offer a collaborative context-updating strategy, tracking the status of common ground during dialogue topic management.

1.

Introduction

Corpus investigation has recently evidenced that contrastive markers of Italian – such as ma (but), però (but, however), mentre (while/whereas) and invece (instead) – are used pervasively in dialogue (Carota 2004, 2005). Nevertheless, the pragmatic function of these linguistic elements has been analyzed rather fragmentarily concentrating on isolated items. For example, various roles have been described in a fine-grained way for ma, ranging from a textual topic shifter (Berretta 1984; Mortara-Garavelli 1993)1 to a turn-taking device used to interrupt the current interlocutor’s turn in spoken interaction (Bazzanella 1994). On the other 1. The marking of topic-shifting can be due to a hierarchical interpretation of the adversative core meaning of ma (Bertinetto & Marconi 1984).

236 Francesca Carota

hand, it is still unclear whether any common unifying features underlie such multifunctionality, and then the contrastive markers form a homogeneous functional class with their own pragmatic peculiarities and dialogue appropriateness. This has motivated the attempt to account systematically for the roles and appropriateness conditions of these markers, which is the focus of the present discussion. What turns out from previous work concerned with English is that the socaled contrastive markers form a sub-set of the ampler functional category of discourse connectives (Fraser 1988, 1998). Within this ampler functional category of “sequentially dependent elements which bracket units of talk” (Schiffrin 1987: 31), their interpretation is usually based on the way how the “contrast” they convey is configured and assigned to them as a distinctive property. For instance, it is largely acknowledged that, when spelling out a contrast relation between two contiguous discourse units, a connective like but denies a proposition, which in the speaker’s mind follows from a prior discourse segment, such as the intra-turn U1 and the inter-turn A, in the examples (1) and (2) drawn on Fraser and Malamud-Makowski (1996: 866). (1) U1. I can go at 4 p.m. tomorrow. U2. But I can’t go if you insist on coming along. (I can go at 4 p.m. tomorrow). (2) A: Do you love me? B: Yes. But I won’t take you out of the garbage. (If you love me, you will take me out of the garbage)

However, even when specified as a denial of the interlocutor’s expectation, for instance, that being tall means being good at basketball, as is the case in the Lakoff ’s example in (3) (Lakoff 1971), it is questionable whether the notion of denial captures de facto a pragmatic aspect intrinsic in to the function of these explicit contrast indicators, especially on the light of their ambigous uses in dialogue.2 (3) John is tall, but he’s not good at basketball.

From an a priori unrelated perspective, the emphasis on the contribution of but to the interlocutor’s optimally relevant interpretation of the speaker’s message (Blakemore 2000) might illuminate relevance-oriented effects of the connective function from a processing viewpoint. At the same time, however, it leaves un-

2. Note, for example, that the type of but-contrast usually referred to in terms of denial might have overlapping boundaries with concession: “although John is tall, he is not good at basketball”.

Collaborative use of contrastive markers 237

determined the speaker’s preferences in favor of a an explicit linking device well adapted to the contextually distributed information. A more linguistically motivated criterion to enlighten such choices of a coparticipant seems to come from those approaches to the connectives that invoke the notions of information structure and focus. Against this background, the butcontrast becomes dependent on the focus of the sentential portion it introduces, like in the context of the question-answer pair shown in example (4) taken from Umbach (2001). Here, the first part of the question remains confirmed by the answer (John cleaned up his room), while but is used to deny the second part of it (i.e. John washed the DISHES), so that its interpretation is driven in situ by the focalized element introduced in the host, i.e. DISHES in opposition with ROOM. (4) Adam: Did John clean up his room and wash the dishes? Ben: ‘yes’. John cleaned up his ROOM, but ‘no’ he didn’t wash the DISHES.

The focus structure of the host utterance has been shown to represent a pertinent contextual constraint for the function of other contrastive markers, for instance otherwise, which acquires in English an information-structurally-based contextupdating potential (Kruijff-Korbayovà & Webber 2001). The information structural framework offers a starting point for those analyses of the contrastive markers that aim at relating the intra-utterance, local level of information “packaging” to more global dialogue dimensions. Following this direction, the present paper discusses ample corpus data from naturally occurring dialogues, suggesting that the interaction of specific contextual dimensions and linguistic, cotextual features of dialogue organization offer relevant parameters for the disambiguation of the behavior of the contrastive markers in the various co(n)textual surroundings of their occurrence. For instance, let’s consider the occurrences of contrastive markers in some exploratory examples, such as the map-task dialogue in (5), and the information-seeking dialogue in (6) and (7). Although a shade of denial of expectation may be implicit in (5d), the interpretive keys based on the marking of a contrast relation, and of the correlations with topic shifts and with turn-taking find a rather sporadic validity. (Instruction) Devi passare oltre lo steccato, e poi trovi le barche. ‘You have to overcome the fence, and then you find the boats’. b. Si. Ma qual è la direzione? ‘Yes. But which is the direction?’ c. Devi andare a sinistra. ‘You have to go to the left’.

(5) a.

23 Francesca Carota

d. Peró prima dello steccato io devo andare sulla collina. ‘But before the fence I have to reach the hill’. La partenza quindi, c’è un volo di mattina… ‘For the departure then, there is a flight in the morning’. b. Mentre con l’altra compagnia c’è un volo alle 15.00. ‘While with the other company there is flight at 3 PM’.

(6) a.

Allora per l’andata le prenoto questo. ‘So, for the departure I will book this one’. b. Invece per il ritorno vorrei partire verso le 18.00. ‘While for the return, I would like to leave around 6.00 AM’.

(7) a.

At this point, for a pars construens to be profiled, the question why the contrast conveyed by these elements arises co(n)textually becomes of interest. Answering this question would enlighten the contrastive marker inherent features which are of impact on dialogue organization. Coming back to the examples just sketched, for instance, ma, situated in turn medial position (see (5b)), seems to attenuate the turn-initial acknowledgment si, a positive feedback for (5a). The contrast marker ma would thus introduce a request for clarification of the information given in (5a). Similarly, in turn-initial position peró (see (5d)) opens a clarification of the information provided in the previous turn (5c). As for mentre and invece (see (6b) and (7b) respectively), they seem to correlate with the co-participants’ negotiation of somehow competing pieces of already discussed and current information. Such empirical evidence gives rise to the intuition that the contrastive markers correlate significantly with the psycho-cognitive process of “grounding” (Clark 1996), whereby the co-participants’ achieve a “common ground”, i.e. a cognitive-interpersonal, or social type of dialogue context, on which the information management is based. Assuming that grounding is mirrored by the surface, i.e. the linguistic structure of the dialogue organization, or cotext, by means of several cues indicating its current status at a certain time of the dialogue course, it is argued that it interacts, locally, with the online intra-utterance packaging of information, and, globally, with the management of dialogue topic. Accordingly, it is claimed that the contrastive markers evoke different dialogue components, namely at the intersection between the common ground and the dialogue topicinformation structure interface. Thus the interpretation of these elements turns out to be sensitive to the interplay of both local and global aspects of information packaging and dialogue topic management. Consequently, it is highlighted that the contrastive markers signal the appropriate way in which the local, or microstructural, linguistic co-text relates to the local cognitive context of the common ground, and in which the local co(n)textual level relates to the global dimensions

Collaborative use of contrastive markers 239

of a vertically extended dialogue structure (i.e. the concatenation of more turns into ampler dialogue units). Along these lines, the purpose of the paper is to specify the pragmatic import of the contrastive markers based on the context-sensitivity of their dialogue functions, and to motivate a criterion to determine appropriateness conditions of their uses depending on the co(n)textually driven co-participants’ preferences for a given contrastive marker. Tangentially, the question is tackled to what extent contrast marking strategies – and their levels of appropriateness – depend on a specific conversational genre or domain. The organization of the paper is as follows. Section two sets up the theoretical premises on which the work is based; section three presents the notion of common ground as it is employed here; in section four the concept of dialogue topic is introduced; section five discusses the theoretical construct of pragmatic appropriateness; section six undertakes a corpus investigation of Italian contrastive dialogue markers. Lastly, section seven summarizes the work presented by stressing some conclusive remarks.

2.

Prelude to an alternative framework

Inter-subjectivity and co-negotiation are peculiar features of task-oriented dialogues. For the communicative interaction to fulfill effectively the current task-related conversational needs, a co-operative mechanism should exist which allows the co-participants to agree upon a common conversational project, as well as to share the information required for accomplishing it. In a concerted joint activity (Clark 1996), they have then to establish collaboratively a common path of navigation over the information flow (Chafe 2001). For instance, when exchanging information for organizing a travel (see examples (8) and (9)), the co-participants determine on line a common path at any time of the communicative process by co-selecting the information relevant for the current task-related purpose. The co-selection may involve two distinct types of information co-negotiation. More specifically, in the present paper, a conceptual distinction is introduced between the co-negotiation concerning the task and the one concerning the cognitive-ideational dialogue meta-scheme, i.e. the dialogue topic. The former coincides with the reciprocal discussion of some alternative solutions for the same task-related issue (Larsson 2002), like the alternative ‘trainplane’ in example (8). (8) A: Preferisce viaggiare in treno o in aereo? ‘Do you prefer to travel by train or by plane?’

240 Francesca Carota

The latter can be configured, on the other hand, in terms of “meta-negotiation” (Carota 2004). Consider example (9): (9) B.a.: B.b.:

L’andata va bene. Poi vorrei vedere gli orari del ritorno. ‘The depature is ok. Then I would like to see the return time’. Allora, per il ritorno, vediamo…ci sarebbe un treno alle 8.30, troppo presto? ‘So, for the return, let’s see…there would be a train at 8.30 am., is it too early?’

While l’andata (the departure) represents a topical element that has been already meta-negotiated and closed up, the return is a new topical element to be meta-negotiated as offering the ideational frame for the next co-participant’s contribution in (b). Such an interpersonal meta-communicative strategy allows the co-participants to co-opt for the ideational coordinates to which their incoming contributions are anchored and to integrate them appropriately into the cognitive-social dimension intrinsic in the communicative context. In this line, the meta-negotiation strategy contributes to the dialogue topic management while interacting with the grounding process, i.e. the psycho-cognitive, interpersonal process whereby the co-participants coordinate reciprocally their private mental states, for instance states of attention, intention, belief, attitude, at every time and level of communication, in order to achieve a common ground on the task-related, topical information under meta-negotiation (Carota, forthcoming). Dialogue topic and common ground are assumed to be the co(n)textual dimensions that found the process of structuring a dialogue. It will be argued here that they have an impact on the determination of the appropriateness conditions for contrastive connectives.

3.

Common ground

The notion of Common Ground refers here to a particular type of cognitive context, consisting of the cognitive state that becomes shared by the co-participants’ private cognitive states through the coordination activity aimed at grounding a piece of information introduced in the dialogue. In this sense, the notion of common ground integrates both the cognitive and the social aspects inherent in the communicative space. In its original formulation, the grounding process has been schematically described as including two phases, i.e. a presentation phase, in which an ungrounded parameter x is presented to the addressee’s cognitive state to form a common

Collaborative use of contrastive markers 241

ground on it, and an acceptance phase, in which the presented parameter is potentially accepted by actually creating a common ground on it (Clark 1996). Such a “contribution model” has been criticized in the literature since it admits the possibility of a “graded” evidence for grounding, which causes an acceptance to represent itself a presentation to be accepted and so on, along a regression fed potentially ad infinitum. Since regression is unnatural with respect to real conversation, it has been suggested to postulate two degrees of linguistic evidence for grounding (Traum 1999). On the one hand, a positive degree (+grounded) corresponds to an acceptance of the contributed informative material which is signaled linguistically by positive feedback such as the acknowledgments si (yes), ok (ok), va bene (ok), capisco (I understand, I see). On the other hand, a negative degree (-grounded) implies the beginning of a new presentation which is signaled linguistically by negative feedback like no, the hesitations mmh, non so (I don’t know), and so on. As for the above mentioned ungrounded parameter x, it indexes a piece of information associated to one or more communicative levels taking part in dialogue management (Traum 1999), like the ones from 1 to 4 described below: 1. Perception of and access to communicative actions. 2. Understanding of communicative actions and their meaning. 3. Local-global contextualization of communicative actions with respect to dialogue topic and domain specific levels such as task-related goals and plans.3 4. Agreement about communicative actions and task-related goals and plans. The grounding is thus a co-ordination activity, during which the co-participants work in order to align – to use the terminology of Pickering and Garrod (2004) – their private cognitive states (attitudes, attention, beliefs) with respect to the above-mentioned dialogue communicative levels. This implicates that for complex communicative actions, multiple degrees of common ground can be discriminated in context. More specifically, the grounding process gets articulated in more consecutive steps before a temporary state of common ground for the levels from 2 to 4 is established around a communicative action, and a complete dialogue unit of common ground is formed. In example (10), for instance, five distinct contributions are needed before the co-participants gather a full ratification of the common ground. For instance, the information in focus dal lato sinistro introduced in G067 is not grounded yet: the request for confirmation in F068 shows that the co-participant accessed the previous dialogue act and perceived

3. The third point is a reformulation of the corresponding level in Traum (1999), which has been made more consistent with the present theoretical perspective.

242 Francesca Carota

it as an instruction, but she still needs to get assured of her access, perception, understanding before reaching a common ground state concerning the meta-negotiated information: (10) G067:

F068: G069: F070: G071: F072: G073: F074: G075:

Eh...passi dal lato...eeh SINISTRO della casa del bigne”. ‘Eh…you pass on the left side...eh…at the left of the “house of bigné”’. Dal lato sinistro ? ‘On the left side ?’. Si”. ‘Yes’. Si”. ‘Yes’. Lato #sinistro## ‘Side … left side’. # Si” # ‘Yes’. E arrivi fino a sopra la casa del bigne”. ‘And you arrive until the “house of bigné’. Sopra. ‘Above’. # Si”# non molto sopra <eh> ‘Yes not too much eh’.

The current state of common ground needs thus to be continuously re-established and updated with respect to the information negotiated by the co-participants: after gathering a common ground about a given informative, topical parameter, the grounding is centered on the next informative parameter relevant for a new (sub)-task. Moreover, it requires the constant and reciprocal evidence for the current common ground state. Thus, the common ground corresponds to a multilayered and dynamic notion of cognitive context. A further feature of common ground is, from the perspective of social cognition, the inter-subjective, inter-personal axe along which the co-participants’ private cognitive states move from a distal space – when they are far from getting coordinated or aligned with respect to an ungrounded informative parameter – to a proximal one – when they get co-ordinated with respect to a grounded informative parameter. In conclusion, the view of common ground just introduced differs substantially from its philosophical conception in terms of knowledge about the world mutually shared by the speaker and the hearer, as well as from the theories that consider it as including the set of propositions and associated presuppositions

Collaborative use of contrastive markers 243

that the participants assume to be taken for granted and no subject to further discussion at that particular time of a conversation (inter al., Stalnaker 1974; Heim 1982). In this view, common ground describes a context set consisting of the set of worlds, such as the actual world, in which all the propositions in the common ground are true, the set of worlds being currently assumed to be taken for granted. The common ground is adopted here as a major contextual analytic dimension. The divergence between the two accounts of the common ground has a decisive impact on the way to conceive pragmatic appropriateness, as discussed in section five.

4.

Dialogue topic

Dialogue topic is the co-participants’ dialogue ideational meta-scheme, often considered in terms of aboutness, or what the discourse is about, that is the discourse-semantic content. The notion has been informally formulated in the following way: A coherent aggregate of thoughts introduced by some participant in a conversation, developed either by that participant or by several participants jointly, and then either explicitly closed or allowed to peter out. (Chafe 2001: 674)

The linguistic realization of a dialogue topic is assumed to be due to the interface between both the micro-, or local, and the macro-, or global levels of the multilayered linguistic structure (see also Gómez-González 2001). The semantic/pragmatic notion of dialogue topic can then be captured through the dialogue-linguistic texture, or co-text, on the base of a number of linguistic cues, such as the linguistic encoding by means of definite/indefinite articles and pronouns of the referential status of dialogue entities (i.e. given, or previously mentioned in discourse, and new, or just introduced into the discourse), where the dialogue entities refer to semantic objects denoting things, states or events. I propose to model the association of the dialogue entities with the cognitive context in terms of degrees of co-activation in the co-participants’ common ground (i.e. co-activated, not co-activated), rather than in terms of the given /new dichotomy.4 Co-textual indexes of the dialogue topic are information structural phenomena which are also associated with the packaging of information. In particular, it is argued that information structure plays a role in orienting the co-participants’ cognitive states towards the entities which are to be co-activated in the common 4. For a taxonomy of the referential status of discourse entity based either on giveness/newness see Prince (1981), or on the activate, semi-active, inactive distinction see Chafe (2001).

244 Francesca Carota

ground. Information structure is intended to refer to the internal articulation of the utterance in terms of informational units, such as theme and rheme, which are marked in Italian by interfacing syntax and intonation (Lombardi-Vallauri 2002). Specifically, according to a two-dimensional approach to information structure, information structural phenomena such as theme/rheme and focus will be explained as follows. Theme and rheme are the patterns which package dialogue entities (Chafe 1976). Themes are optional utterance elements, and are defined as setting an explicit topical coordinate, which specifies the way in which the host local utterance (and the predication of its rheme) relates to the global topic in the co-participants’ common ground. Themes are not necessarily associated to grounded information: they can have a certain update potential with respect to the dialogue topic, when introducing a dialogue entity which is inactive in the common ground. Rhemes are utterance-obligatory elements which perform the speaker’s communicative goal by accomplishing the dialogue act which the utterance realizes. The term ‘dialogue act’ denotes the communicative purpose behind the utterance, related to the surrounding dialogue context, and can correspond to the term of communicative act. They present to the common ground those dialogue entities which need to be co-activated and already attached to the topical coordinate currently co-activated. Focus is a prosodic prominence functioning as a semantic/pragmatic operator orthogonal to either theme or rheme (or to both of them) (Steedman 2000). It is crucial to highlight a dialogue entity to be submitted to the common ground. When evoking a contextually available, closed set of alternatives for this entity, it receives a contrastive interpretation (see also Molnàr 2001). The global interplay of these co-textual parameters with the surface linguistic indexes of common ground determines the dialogue structure (Carota 2004). Globally, the dialogue structure is cognitively shaped by means of the grounding process. It can be refined as the organization of dialogue in dialogue units, namely sequences of contributions developed around the same topic while trying to gather a common ground on its sub-topical coordinates, i.e. the informative aspects related to that topic. These common ground units are successfully formed and closed when the task-related information has been attributed to the common ground. Locally, the dialogue structure results from the interconnection of local utterance themes and the topic globally distributed in ample dialogue units (Carota 2004, 2005, forthcoming). The relationship between co-text and context can be represented schematically as follows: Co-text: Context:

Feedbacks, {information structural phenomena, dialogue entities} ↑ ↓ ↑ ↓ common ground status, Dialogue Topic status in the common ground

Collaborative use of contrastive markers 245

Thus, the dialogue structure is as an inherently dynamic, multilayered object expanding on-line during dialogue management. It is assumed here as being the co-textual analytic dimension of the study. The next section relates the contextual and the co-textual components of dialogue analysis to the notion of appropriateness, while enlightening its relevance for the analysis of contrastive markers.

5.

Appropriateness

To what extent do the theoretical and empirical parameters examined above profile an operational criterion for delineating the co(n)textual appropriateness of the contrastive markers? To my knowledge, the notion of appropriateness has not been specifically accounted for from the perspective of the psycho-cognitive theory of grounding to which the present work is anchored. In the classical view, the conditions on appropriateness of Stalnaker’s matrix (Stalnaker 1974) predict that an utterance is contextually appropriate if its illocutionary force contributes to the updating of the context set of possible worlds, as mentioned in Section 3. For instance, an assertive utterance is meant to be appropriate if it produces an update of the common ground by providing new information. When the sentence is accepted by the interlocutors, the proposition expressed by it is added to the common ground, and the context set is updated by removing the worlds in which the proposition is false and maintaining the ones in which the proposition is true. The truth of the proposition expressed by the sentence is part of the common ground and thus mutually assumed, taken for granted and not subject to further discussion. In the case of utterances inducing pragmatic presuppositions, by referring to world knowledge presented as being already part of the common ground, the presupposition has to be included in the common ground. In other words, the context set has to include the world in which the presupposition is taken for granted and does not require to be discussed.5 Two aspects of this model are of relevance for the present discussion, namely the illocutionary import of the host utterance, and the common ground updating. However, these notions are not seen here from the perspective of a theory of presuppositions (and their accommodations), and, in particular, the working notion 5. Similarly, in File Change Semantics (or Context Change Model) proposed by Heim (1982), which elaborates Stalnaker’s work, it is assumed that speakers produce a sentence in order to update the information state – roughly equivalent to Stalnaker’s term of context set –, assuming that the presuppositions of that sentence are satisfied in their current information state.

246 Francesca Carota

of common ground does not rely on the set of assumptions about the world knowledge that are mutually shared by speaker and hearer, as in the philosophical view. The illocutionary dimension and the common ground updating will be adopted as being two interdependent criteria of appropriateness driven by the local and global co(n)text. The appropriateness will be then delineated here consistently with the interpersonal-cognitive framework proposed here. Based on its illocutionary configuration, each utterance can be conceived as a pragmatic whole, which performs a complete, autonomous communicative act or dialogue act. It has already been said above (Section 4) that the contrastive markers – just as discourse markers – are constitutive semantic-pragmatic components of utterances and lack illocutionary autonomy, as they do not perform a pragmatically independent communicative act or dialogue act, but rather depend on the illocutionary force of the host utterances and thus are sensitive to the internal modulation of this force, which determines their information structure. Locally, the evaluation of the appropriateness of the contrastive markers is based on this criterion. More globally, it can be speculated that the contrastive markers are appropriate in those conversational circumstances in which the co-participants consider the task-related information to be not (fully) co-activated in the common ground, and the grounding process to be kept open or to be re-opened in order to integrate it. Once a common ground is gathered at the communicative levels from 1 to 4 (see Section 3), the co-participants’ minds (or cognitive states) get balanced with respect to a common state of information. In order to maintain this balance, a co(n)textually appropriate signal of cognitive-interpersonal discrepancy has to be emitted by the speaker to orient the co-participants towards the updating of the topical information in the dialogue context. Accordingly, the contrastive markers turn out to be appropriate when the reciprocal cognitive context, or the common ground, is compromised or needs to be achieved. In the cognitive-pragmatic view proposed here, thus, the appropriateness of the contrastive markers consists of their co-phasing with both the structuring of information within the host utterances and with the grounding process during dialogue topic management. Being directly related to the cognitive environment of the common ground, the appropriateness of a connective can be said to be the property by which the connectives are congruent with respect to the interplay of co-textual and contextual parameters evidenced so far, i.e. locally with respect to the micro-cotext of the information structural encoding of the topical coordinates, and globally with respect to the specific status of the grounding of the meta-negotiated topic.

Collaborative use of contrastive markers 247

It seems worth to note additionally that the notion of appropriateness is usually seen as being dependent on the category of genre, and on the stylistic connotation of a discourse element. Along this line, … the appropriateness of an utterance is evaluated with regard to the nature of the connectedness between a communicative action, its linguistic realization and its embeddedness in linguistic and social contexts. (Fetzer 2004: 20)

In this perspective, it has been related to the diastratic variation on the sociolinguistic level due to the sociocultural status of the interlocutors. Nevertheless, the present corpus study does not focus on the sociolinguistic aspect. Rather, it highlights the impact of another parameter regarding the appropriateness of contrastive markers, namely the specific typology of dialogue: the use of the contrastive connectives invece and mentre, abundant in the information-seeking dialogues, seems to be inappropriate in the case of the map-task, as shown in Section 6.3. This would mean that that pragmatic property is to some extent domain-specific.

6.

Corpus analysis

The corpus investigation of the contrastive markers presented here is based on data acquired from both information-seeking dialogue within the travel domain (the Adam corpus: Soria et al. 2001), and map-task dialogues (the API corpus: Crocco et al. 2003).

6.1 Ma Moving from a co-textual perspective, it can be observed that the connective ma tends to be positioned at the beginning of an utterance which follows a positive feedback produced within the same turn (e.g. si (yes), d’accordo/ok (ok), like the example (11) illustrates.6 The host utterance can be, like in the present case, a question, which can be interpreted as a dialogue act of request for clarification about a new piece of inactive information under focus, i.e. the dialogue entity orario. (11) A1:

Il ritorno, cosa voleva? ‘The return, what did you need?’.

6. Note that the connective does not occupy the initial position of a turn in the examples provided with the present section, attenuating the interpretation of its function in terms of interruption of the interlocutor’s current turn mentioned in Section 1.

24 Francesca Carota

B2: 3:

→ A3:

Il ritorno, le avevo detto domenica, domenica POMERIGGIO sul presto. ‘As I said, the return will be on Monday, Monday early in the afternoon’. Si. ma come ORARIO? ‘Yes. But which is the hour?’.

From the point of view of dialogue topic management, the host interrogative utterance inherits from the prior co-text the topical coordinate, which is packaged in the previous utterance theme il ritorno (outlined both in A1 and in B2). As a result, a topic change can be excluded here. The ma function in A3 is rather configured on the base of the partial acceptance, in the common ground, of the information provided locally in B, as if it was incomplete for the co-participant’s communicative needs and incongruent with respect to her current cognitive state. The use of the connective appears to be related to a contextual strategy of the communicative exchange, in which the grounding process cannot be closed by means of an acceptance of the information previously contributed. Thus, the occurrence of ma correlates significantly with a circumstance in which the grounding needs to be re-opened at a local level of dialogue structure, through the meta-negotiation of the topical coordinate instantiated by the dialogue entity orario. Although such an entity has been implicitly requested in the elliptical question A1, it cannot be recovered explicitly from the immediate prior co-text of the contribution in B. In this sense, ma turns out to be sensitive, co-textually and locally, to the focused part of its host utterance. This element under focus is then the scope of the connective function. More exactly, ma cues that the focused information is not grounded into the common ground. At the same time, it limits the scope of the acceptance into the common ground introduced by the previous positive feedback si at the communicative levels 1-2-3 seen so far, with exclusion of the level 4, or the co-participants’ agreement. In example (12), ma also follows a positive feedback signaling the acceptance at the communicative levels from 1-2-3, but not at the level 4. (12) F: Allora io i mobili di Elena ce l’ho ... ce l’ho sulla SINISTRA rispetto alla roulotte, hai capito? ‘So, me, I have the Elena’s furniture …I have them on the left with respect to the caravan, do you understand?’ → G: Si” , ho capito ma .. rispetto alla STRADA, i mobili di Elena...dove li trovi? ‘Yes, I see, but with respect to the street, the Elena’s furniture where do you find them?’

Collaborative use of contrastive markers 249

In this case, however, the request for clarification is about the information i mobili di Elena, presented in the utterance theme in F and repeated in G, thus already co-activated in the common ground. This request aims at checking the common ground status with regard to the negotiated information, in a situation of discrepancy between the instructions provided by the giver about the map (i.e. the map-task game instructor), and their application on the map of the follower (i.e. the game player who has to find the goal on the map based on the giver’s indications). More specifically, ma is employed to introduce a question about a topical coordinate which has not been grounded yet by the speaker. This view is supported by the evidence provided by ample corpus examples. For instance, in the examples 13–14 ma arises systematically when the common ground about the information lastly negotiated has to be temporarily suspended because it conflicts with an ungrounded topical or informative parameter, which is the one in focus within the host utterance. (13) G: F:

(14) G019:

F020: G021: F022: G023: → F024: G025: F026: G027: F028:

Poi sali su. ‘Then you go up’. Ma alla scritta BABBUINI te c’hai BANANO sotto o sopra? ‘But at the sign “baboon” do you have “banana-tree” below or above?’. Poi sal+ sali un pochino su fino a FIUME ce l’hai quella figurina? ‘Then you go up a bit until the river, do have this figure?’. Fiume FIUME si” #G021 si”# ‘River... river yes yes’. # <eh># li”, presente ? ‘Eh there, clear?’. Si” # si” , ho capito# ‘Yes yes, I see’. Poi ecco arrivi a # fiume # ‘Then, here, you arrive to the river’. Ma a fiume e” necessario usare le BARCHE ? ‘But at the river , do I need to use the boats?’. Si”, si” usale usale , vacci. ‘Yes yes use them use them go towards them’. <Mhmh> ‘Mhm’. Vai alle BARCHE , quindi sei li”, no ? ‘Go towards the boats, then you are there, aren’t you?’. # Si”# ‘Yes’.

250 Francesca Carota

G029: F030: G031: F032: G033: → F034: G035: F036: G037: F038: G039: → F040:

G041:

Poi gira un pochino a SINISTRA, fai una specie di curva ma piccola. ‘Then turn on the the left, but not too much’. <eh> !# Si”# ‘Eh! Yes’. Fatta la # curva ?# ‘Did you turn?’. # Vai # # vai# ‘Go <noise> go’. E poi giri a DESTRA. ‘And then turn on the right’. Ma sono sempre sulla BARCA, io ? ‘But am I always on the boat?’. Si”. ‘Yes’. <Mhmh> ‘<Mhm>‘. Sempre LI” giri a destra. ‘Always there turn on the right’. Si” ‘Yes’. Ah e<ee> la valle limpida te la trovi a destra. ‘ the limpid valley, you will find it on the right’. Si” # si” , ho capito, ma a questo punto io la barca non...non mi serve piu”?# ‘Yes yes, I see, but at this point, the boat, I don’t… I don’t need it anymore’. Okay, ci sei

no ‘un ‘dialettale’. ti serve piu” no ‘Ok, there you are, no, you don’t need it anymore’.

In a slightly different setting (example 15), ma opens a statement for clarification following the positive feedback si, which acknowledges all the communicative levels of the request for confirmation risen in F084 about co-activated information. However, the host utterance of ma aims at preventing potential misunderstandings in a grounding phase in which the co-participant is still establishing a common ground about the giver’s instructions. (15) F084: → G085:

Quindi sto risalendo e sulla mia destra trovo piazza della musica. ‘I’m going up then and on the right I find the square of the music’. Si” ma non torni indietro <eh!> non torni indietro : dopo la rotonda scendi. ‘Yes, but you don’t go back, eh, you don’t go back: after the round square, you go down’.

Collaborative use of contrastive markers

In the light of the previous considerations it can be stressed that, in the task-oriented dialogues considered here, the use of ma exhibits a large compatibility with host utterances that perform some specific types of dialogue acts, such as requests for clarification or for confirmation, and statements for clarification. These dialogue acts are linked to a still opened grounding process around an informative parameter. It can be concluded that ma is used to signal that the information provided by the co-participant in the previous co-text cannot be fully grounded by the speaker because a new parameter – under focus in the host – is missing or a given parameter is not grounded yet, but still needs to be co-activated in the speaker’s common ground. Generalizing on all the examples just considered, the role of ma can be characterized functionally as a cognitive-interpersonal operation whereby the flow of the meta-negotiated topical information is re-orientated in order to become shared by the participants’ cognitive states, in other terms, to get appropriate with respect to the current phase of grounding finalized towards a temporarily shared cognitive context on that information, or a common ground.

6.2 Però A positional behavior close to the one just presented for ma pertains to però, occurring after a positive feedback as well, when a new task-related topic is being presented to the common ground. From the point of view of the dialogue structure, like ma, però can open a new unit of common ground centred on an inactive topical parameter. This common ground unit can be formed at different levels of structural granularity. More specifically, però seems to cue a deeper dialogue boundary with respect to ma, a boundary in which both a new-topic unit (with its corresponding dialogue sub-task), and a new unit of common ground start. An example (16) is provided with the following fragment. (16) A1: → B2: A3:

Allora l’andata alle 15.30 da Roma…le prenoto un posto? ‘So, the departure at 3.30 from Rome may I reserve you a place?’. Si. però vorrei prenotare anche il RITORNO. ‘Yes, but I would like to book the return as well’. D’accordo. ‘Ok’.

Furthermore, like in (17), però can occur -just as ma- in a statement of clarification that follows a confirmation (the positive feedback si in G251) about information already co-activated in the common ground. The clarification statement provides some additional inactive information (“Sali un poco”, go up slightly),

251

252 Francesca Carota

which serves to guide the grounding towards the appropriate assessment of the topical information under meta-negotiation in F250 in a common ground whose achievement is still in fieri. (17) F250: → G251: F252: G253: F254:

E arrivo al quartiere piccolo che ce l’ho sulla sinistra. ‘And I arrive at the small quarter that is on my left’. Si” però “ dopo questa piccola deviazione, SALI un poco. ‘Yes but after this small deviation, go up slightly’. Si” dopo il # quartiere# ‘Yes after the quarter’. # Come se# ci fosse un DOSSO ‘As if there was a hump’. Dopo il quarti+ ? si” ho capito ho capito. ‘Right after the quart…yes I’ve understood I’ve understood!’

The inactive information introduced into the utterance hosting però can correct the information grounded until now or a part of it (see 18). (18) 1 C 1: 2: 3: 2 O 1: 3: 3 C 1: 4 O 1: →

2: 3: 4:

Va bene. ‘Ok’. Il rientro dovrebbe essere il MERCOLEDI. ‘The return should be on the Wednesday’. Vorrei arrivare vorrei partire scusi prima delle delle SEI di sera. ‘I would like to arrive, to leave sorry, before6 PM’. Partire prima delle sei di sera dunque vediamo subito dunque. ‘To leave before 6 PM, so, let see soon’. Le va bene partire alle DICIASSETTE e trentatre? ‘It is ok to leave at 5.33 PM’. Perfetto. ‘Perfect’. Diciassette e trentatré. ‘17.33’. In questo caso … però… deve CAMBIARE a Padova. ‘In this case, however, you have to change in Padua’. E l’ arrivo previsto a roma e` alle VENTITRE e zero cinque. ‘And the arrival at Rome will be at 11.05 PM’. Dunque, anche in questo caso, si tratta di due intercity. ‘So, also in this case there will be due INTERCITY trains’.

Però is in accordance with a request dialogue act which opens a new unit of common ground, in which the co-participants aim at aligning their reciprocal mental states with respect to a piece of information co-textually encoded, to make them

Collaborative use of contrastive markers

converge towards a shared cognitive state, the common ground. Otherwise, it correlates with a clarification/correction dialogue act, in which a topical parameter is clarified because it still needs to be assessed in the shared cognitive context of the common ground, which is being achieved by the co-participants. It emerges thus that both però and ma signal that, before being grounded into the cognitive context of the common ground, a piece of information already in the common ground needs to be either corrected or clarified according to a topical coordinate currently highlighted by focus. Both connectives cue then the initial boundary of a dialogue sub-unit of common ground, in which a new presentation to the common ground itself requires the explicit clarification of some co-textually missing, or co(n)textually underspecified informative parameter. The linguistic elements under investigation play the crucial function of reorienting the grounding process towards a new perspective to take for the achievement of a common ground. In this sense, però and ma display a distal attitude of the speaker, whereby she takes distance from the co-participant’s prior contribution in relation to the cognitive status of the topical information in the common ground. The behavior of the connectives just examined emerges to be interestingly interrelated with the positive or negative status of the meta-negotiated topical information in the common ground. In conclusion, their use appears to be appropriate to the grounding process, when an inactive sub-topical coordinate, which has to be accepted as being the current topic under meta-negotiation, is presented to the cognitive context of the common ground. From the viewpoint of the cotext, they are appropriate in utterances which can be characterized functionally in terms of clarification/correction statements, or requests for clarification. At the co-textually more fine-grained, micro-structural level of information structure, these connectives are appropriate when an inactive topical information, which is instantiated by a certain dialogue entity, receives a focus signaling that this is the informative element to be integrated within the cognitive context, or, more precisely, co-activated into the co-participants’ common ground.

6.3 Invece and mentre In the corpus data examined here, the connective invece occurs systematically in a request for new information about alternative solutions to the same task-related topic. For instance, in the example (19), it opens a topically elliptical open question and follows a positive feedback. The ellipsis can be resolved, by inferring the topic from the previous context, specifically from the turn A1, i.e. i voli (flights). (19) A1:

Vediamo un po’ quali sono i VOLI. ‘Let’s have a look at the flights’.

253

254 Francesca Carota

B2 a: b: A3 a: →

b:

Si. ‘Yes’. Allora, con l’ Alpi Eagles ci sarebbe un volo alle DIECI di mattina. ‘So, with Alpi Eagles there is a flight leaving from Rome at 10 AM’. Ah, va bene. ‘Ah, OK’. Invece con la MERIDIANA? ‘Instead with the Meridiana (scilicet: company)?’.

The alternative information has to be co-activated, and usually represents the topical coordinate to be discussed, marked by a contrastive focus which opposes that coordinate to the polar one packaged into the first theme of the previous co-participant’s contribution. The connective host utterance asks for new information (i.e. i voli) related to a piece of given information, i.e. the dialogue entity Meridiana, which has to be established as being the current task-related topical coordinate. The co(n)text just depicted profiles a prototypical negotiation, in which an alternative solution to the same task-related issue is discussed. In the present communicative setting, invece operates a change of perspective within the ongoing negotiation, by introducing the topical coordinate to be discussed as an alternative to the previous one. It is worth highlighting that this topical coordinate, although given, or previously mentioned in the co-text, has to be now newly co-activated contextually, in the co-participants’ common ground. Furthermore, although the dialogue unit related to the alternative in the immediately preceding co-text has been grounded through the acknowledgement ok, the grounding on the main dialogue topic i voli is still open. Consequently, invece is implicated in the speaker’s strategy of re-orienting the ongoing negotiation, by revising a parameter of the previous alternative solution to the current task-related issue. The connective has thus a role in maintaining the grounding process opened through the sequential presentation of alternatives to the common ground. Similarly, in the example (20), invece follows a positive feedback and re-orients the information flow with respect to the cognitive context of the common ground. However, in the present fragment, invece does not introduce any alternative solution to the same issue in the sense considered so far. First, the connective figures in a declarative utterance that represents a new presentation to the common ground. Secondly, the host statement is internally, micro-co(n)textually articulated according to the following information structural pattern: the theme presents an inactive topical coordinate alternative to the one co-activated until now, while the rheme presents the inactive information to be co-activated with respect to the new topic. Furthermore, a left-detached contrastive theme packages the dialogue entity ritorno, which co-refers to a previously mentioned given

Collaborative use of contrastive markers

information, that is per rientrare, but needs to be co-activated currently as being opposed to the previously meta-negotiated entity andata (departure), which has to be substituted in the common ground. (20) 1 O: 2 C: 3 O:

4 C:

Venerdì c’è un treno alle 8.35. ‘On Friday, there is a train at 8.35 AM’. Benissimo. ‘Perfect’. Ok. Per il ritorno invece c’ e` un treno alle 15 da Roma. Puo` andar bene? ‘Ok. As for the return instead there is a flight from Rome at 3 PM. Is it Ok?’. Alle 15 potrebbe andar bene. ‘At 3 PM., it could be ok’.

The effect of the contrast conveyed by invece is appropriate when the speaker intends to provide together with the host utterance an explicit instruction to the co-participant: she aims at replacing into the cognitive context of the common ground the alternative sub-topical coordinate recoverable from the immediate co-text, i.e. the departure, with the one which is indicated locally by the contrastive theme in the host utterance. The connective mentre (whereas) exhibits a close functional proximity with invece, occurring in statements aiming at presenting alternatives solutions to a same issue, as when comparing the prices of trains leaving at different hours like in (21). (21) 1 O 1:

3: 2 C 1: → 3 O 1:

2: 4 C 1:

In seconda classe dunque questo delle dodici e trentacinque e` un pendolino fino a bologna e poi deve prendere un interregionale. ‘In the second class, well this one of 12.35 AM is a pendolino until Bologna and after an interregional’. In totale sono in seconda classe sessantottomila lire. ‘It is in total -in the second class- 68.000’. Ho capito. ‘I have understood’. Mentre il pendolino delle quattordici e quarantacinque sono settantamila lire. ‘Whereas the train of 2.45 PM is 70.000’. Quindi duemila lire in piu` . ‘2.000 more then’. Si` si` non e` niente. ‘Yes yes it’s nothing’.

255

256 Francesca Carota

2: 5 O 1:

→

2:

E l’ arrivo? ‘And the arrival?’. L’ arrivo...allora… -quello delle dodici e trentacinque- e` previsto per le diciassette e sedici. ‘The arrival well the one of 12.35 AM is at 5.16 PM’. Mentre per quello delle quattordici e quarantacinque l’ arrivo e` previsto alle diciotto e quarantadue. ‘Whereas the one of 2.45 PM the arrival will be at 6.45 PM’.

Mentre can also be significantly implicated in a meta-negotiation of the topical coordinates in the common ground, as the example (22) confirms. (22) 1 O 1: 2:

4:

2 C 1: → 3 O 1:

Per il rientro quindi mi diceva in serata tardo pomeriggio… ‘As for the return, then, for the evening late in the afternoon’. Con la meridiana si potrebbe fare diciamo o alle sedici eeh… partenza. Alle sedici e trentacinque e arrivo alle diciassette e trentacinque. ‘With meridiana there is, let say, either at 4 ehm departure at 4.35 PM and arrival at 5.35 PM’. Oppure alle diciannove e trentacinque e arrivo alle venti e trentacinque ‘Otherwise at 7.35 PM and arrival at 8.35 PM’. Ho capito. ‘I understand’. Mentre l’ alpi eagles ce ne ha uno… un˜ alle diciotto e quaranta, che arriva alle diciannove e quaranta. ‘Whereas l’alpi eagles there is one, one at 6.40 PM, which arrives at 7.40 PM’.

To conclude, invece correlates with requests for new information, as well as with statements, in which case it seems interchangeable with mentre. The connectives are appropriate in those communicative settings in which the given information is still insufficient to complete the negotiation because of a missing inactive parameter or of an ungrounded given parameter, and cannot be fully grounded by the speaker using the connective. A further condition of appropriateness for the connectives mentre and invece is offered by those statements in which the topical coordinate to be co-activated in the co-participants common ground is being updated though the co-textual introduction of an inactive one. The effect of the contrast conveyed by invece and mentre serves to signal interpersonally that, before grounding the presented information, the same topic has to be discussed about an equivalent dialogue entity, which is polar, or opposite to the topical coordinate currently under (meta-)negotiation. After grounding the

Collaborative use of contrastive markers 257

(meta-)negotiated information, a new unit of common ground can start to presents new information about a topical coordinate alternative to the one until now co-activated (which is substituted and not persistent anymore in the common ground). These contrastive markers do not occur with this function in the map-task corpus, where they are rare and have the function of adversative conjunctions. This circumstance can be due to the different tasks and social roles of the co-participants in the two domains. In the map-tasks, the co-participants’ social roles are hierarchically organized, the follower’s moves depending on the giver’s instructions, and there is not a choice between alternatives, which turns out to be a peculiarity of the information-seeking dialogue. Mentre and invece turn out to be appropriate as contrastive markers of alternatives within a dialogue unit built to gather a co-participants’ common ground around the options inherent into a same issue under discussion, or into a same task-related topic meta-negotiation.

7.

Conclusion

In this contribution, a cognitive approach to the contrastive markers of Italian has been proposed, which results in a uniform interpretation of their dialogue function in the setting of task-oriented dialogues. The account is compatible with previous hypotheses about the functions of contrastive markers, but seems to be empirically more adequate in motivating their context sensitivity in this particular dialogue typology. Furthermore, an innovative criterion to determine their appropriateness is provided with the results of the corpus analysis. The appropriateness of the contrastive markers is modeled in terms of their efficiency in signaling the lack of alignment between the co-participants’ cognitive states, in those contexts in which an informative topical parameter is missing in the common ground and has to get included in it. The contrast conveyed by the contrastive markers turns out to be a powerful pragmatic strategy operating from the micro- to the macro-dimensions of the dialogue structure for updating the co-text according to the grounding process. Different types of update can be suggested on the basis of the different types of contrast that the contrastive markers have been shown to convey based on corpus data. As discussed so far, the results highlight that a contrast inducing a dialogue update by means of a clarification request dialogue act is cued by ma (see examples (15)–(17)), and arises when, within the same topic and common ground unit, a currently inactivated informative parameter – usually marked by focus in the host – is not accepted in the common ground. As a result, the grounding process,

25 Francesca Carota

which is still in fieri on the information under presentation, is temporarily suspended and re-opened around that parameter. Però (examples (20)–(22)), as well as ma (examples (19)), is mostly used in the update of preventive/corrective type, in which the contrast aims at the local re-organization of the common ground, where a piece of information, which is currently co-activated and has already been grounded, needs to be corrected or clarified (as indicated by a clarification dialogue act). An oppositional type of contrast is due to invece and mentre, which have been associated with requests and statements about new factual information. This occurs within the co-negotiation of a particular topic, where a polar alternative solution to a same task-related issue is at stake. The alternatives may be allocated to the same level of topicality or they may be subordinated to a shared main topic. In this circumstance, in which the grounding process is still under negotiation, and in fieri, the connectives indicate that a new local presentation is opened and that the inactive information has to be attributed to the common ground. The contrast of mentre and invece can also convey a substitutive update within a meta-negotiation, to re-open the grounding process because of a task-related topic shift. The interpretation of the contrastive markers used in the task-oriented dialogue examined here turns out to be sensitive to the interplay of grounding and to the dialogue topic-information packaging interface. The contrastive markers emerge to be a homogeneous sub-set of discourse markers, which signal different polarities established between the information introduced in the host utterances and some piece of information in the preceding co(n)-text. For this reason, the contrastive markers correlate with different boundaries of the dialogue structure, conveying the speaker’s attitude towards the information state in the cognitive context of the common ground. They function as crucial re-orientation devices, employed by co-participants during their coordination activity, in order to reciprocally regulate their cognitive states of activation with respect to the information exchanged while co-constructing the conversation. Besides their significant role in anchoring the local host utterance to the global dialogue structure and contributing to discourse coherence, the contrastive markers help to constantly monitor the status of the information in the common ground. These change-of-perspective devices are appropriate in the conversational circumstances requiring the grounding process to be re-directed towards the achievement of a common ground. The contrast they convey is a building-block of a fully collaborative strategy.

Collaborative use of contrastive markers 259

References Bazzanella, C. 1994. Le facce del parlare. Un approccio pragmatico all’italiano parlato. Firenze/ Roma: La Nuova Italia. Bertinetto, P. M. and Marconi, D. 1984. “Ma in italiano. Parte prima: Semantica e pragmatica”. Lingua e Stile 19: 223–258. Berretta, M. 1984. “Connettivi testuali in italiano e pianificazione del discorso”. In Linguistica testuale, L. Coveri (ed.), 237–254. Roma: Bulzoni. Blakemore, D. 2000. “Indicators and Procedures: Nevertheless and but”. Journal of Linguistics 36: 463–486. Carota, F. 2004. “On some effects of lexical contrast in information-seeking dialogues”. In Proceedings of the Eighth Workshop on the Semantics and Pragmatics of Dialogue, Ginzburg, J. and Enric Vallduví (eds.), 148–149. Barcelona: Universitat Pompeu Fabra. Carota, F. 2005. Fondo cognitivo comune e negoziazione dell’informazione: funzionalità dei marcatori discorsivi dell’Italiano. PhD. Thesis. Pisa: Università degli Studi di Pisa. Carota, F. forthcoming. “Gathering common ground on the negotiated topic: The role of the contrastive markers of Italian. In Theoretical Approaches to Dialogue, L. Berlin (ed). Tübingen: Niemeyer. Chafe, W. 1976. “Givenness, contrastiveness, definiteness, subjects, topics, and point of View”. In Subject and Topic, Charles N. Li (eds.), 25–55. New York: Academic Press. Chafe, W. 2001. “The analysis of discourse flow”. In The Handbook of Discourse Analysis, D. Schiffrin, D. Tannen, and H. E. Hamilton (eds.), 673–687. Oxford: Blackwell. Clark, H.H. 1996. Using Language. Cambridge: Cambridge University Press. Crocco, C., Savy, R. and Cutugno, F. 2003. Api (Archivio di Italiano parlato) in dvd. Napoli, CIRASS-Università degli Studi di Napoli “Federico II”. http://www.cirass.unina.it. Fetzer A. 2004. Recontextualizing Context: Grammaticality Meets Appropriateness. Amsterdam/ Philadelphia: John Benjamins. Fraser, B. 1988. “Types of English discourse markers”. Acta Linguistica Hungarica 38: 19–33. Fraser, B. and Malamud-Makowski, M. 1996. “English and Spanish contrastive discourse markers”. Language Sciences 18: 863–881. Fraser, B. 1998. “Contrastive discourse markers in English”. In Discourse Markers. Descriptions and theory, A.H. Jucker and Y. Ziv (eds.), 1–13. Amsterdam/Philadelphia: John Benjamins. Gómez-González, M.A. 2001. The Theme-Topic Interface. Amsterdam/Philadelphia: John Benjamins. Heim, I. 1982. “On the semantics of definite and Indefinite Noun Phrases”. Ph.D. Thesis, University of Massachusetts at Amherst. Kruijff-Korbayovà, I. and Webber, B. 2001. “Information structure and the interpretation of ‘otherwise’”. In ESSLLI 2001 Workshop on Information Structure, Discourse Structure, and Discourse Semantics, I. Kruijff-Korbayova and M. Steedman (eds.), 61–78. Helsinki: University of Helsinki. Larsson, S. 2002. “Issues under negotiation”. In Proceedings of the Third SIGdial Workshop on Discourse and Dialogue, K. Jokinen and McRoy (eds.), 103–112. Philadephia: USA. Lakoff, R. 1971. “If ’s, and’s and but’s about conjunction”. In Studies in Linguistic Semantics, Fillmore, Langendoen (eds.), 114–149. Mouton de Gruyter: Berlin. Lombardi-Vallauri, E. 2002. La struttura informativa dell’ enunciato. Firenze: La Nuova Italia.

260 Francesca Carota

Molnár, V. 2001. “Contrast from a contrastive perspective”. In Proceedings of the ESSLLI’01 Workshop on Information Structure, Discourse Structure and Discourse Semantics, I. Krujiff-Korbayovà and M. Steedman (eds.), 99–114. Helsinki: University of Helsinki. Mortara-Garavelli, B. 1993. “Strutture testuali e retoriche”. In Introduzione all’italiano contemporaneo. Le strutture, A.S. Sobrero (ed.), 371–402. Roma-Bari: Laterza. Pickering, M.J. and Garrod, S. 2004. “Toward a mechanistic psychology of dialogue”. Behavioral and Brain Sciences 27: 169–225. Prince, E. 1981. “Toward a taxonomy of given-new information”. In Radical Pragmatics, P. Cole (eds.), 223–255. New York: Academic Press. Schiffrin, D. 1987. Discourse Markers. Cambridge: Cambridge University Press. Soria, C., Cattoni, R., and Danieli, M. 2000. “ADAM: An architecture for XML-based Dialogue Annotation on Multiple Levels”. In Proceedings of the First SIGDIAL Workshop on Discourse and Dialogue, Hong-Kong, 9–18. Stalnaker, R.C. 1974. “Pragmatic presuppositions”. In Semantics and Philosophy, Munitz, M.K. and Unger, P.K. (eds.), 197–214. New York University Press. Steedman, M. 2000. “Information structure and the syntax-phonology interface”. Linguistics Inquiry 31(4): 649–689. Traum, D. 1999. “Computational models of grounding in collaborative systems”. In Working notes of AAAI Fall Symposium on Psychological Models of Communication: 124–131. Umbach, C. 2001. “Contrast and contrastive topic”. In Proceedings of the ESSLLI 2001 Workshop on Information Structure, Discourse Structure and Discourse Semantics, Kruijff-Korbayova, I. and Steedman, M. (eds.), 175–188. Helsinki: University of Helsinki.

Index

A acceptability 17, 18, 57, 148, 228 acceptance 62, 80, 116, 124, 125, 136, 140, 142, 154, 178, 182, 188, 227, 241, 248 act/action collective 118, 120 communicative 21, 58, 72, 120, 124, 127, 128, 214, 244, 246 face-threatening 148, 195 illocutionary 6, 8, 65, 70–73, 76, 77 locutionary 8, 59 macro speech 19, 20 micro speech 19 phatic 8, 59 phonetic 59 rhetic 8, 59 social 18, 22, 115, 152, 170 speech 3, 8, 9, 19, 20, 22, 23, 25, 55, 57, 61, 62, 67, 68, 115, 118, 120, 122, 124, 125, 127, 130, 145, 170, 171, 210, 213, 216, 222, 232 utterance 6, 207 activation 88, 92, 98, 99, 243, 258 activity type 17, 19–21, 117, 121, 149, 152, 207, 226 adjacency pair 4, 129, 132, 135 position 129 relation 5, 129 Akman 8, 121, 206 alternatives 88, 89, 244, 254, 255, 258

amalgamation 59, 66, 71 ambiguity 123, 183, 194 analogy 33, 47 analogical transfer 33, 53 anchoring 40, 116, 258 answer direct 154, 180 non- 179, 184, 185 partial 190 appropriateness condition 19, 49, 117, 119, 125, 128, 130–132, 135–142, 152–154, 205, 208, 227, 236, 239, 240 pragmatic 56, 57, 239, 243 artefact linguistic 56, 59, 60, 62, 66, 67, 69–71 atom 43, 44 attenuation 133, 135, 216, 221 Austin 8, 57, 59, 61–67, 70, 73–77, 118, 170, 204 automatization 83, 86, 89, 99

B background knowledge 14, 84–86, 88, 89, 94, 96, 99, 103, 206 Barwise 32, 36, 37 Bateson 3, 9, 10 Bell 149, 153, 176, 194 bottom-up 3, 22, 119, 121 Brandom 116, 118, 126 Brown and Levinson 116, 149, 195

C categorization 85, 86, 92, 93, 95–97, 99, 100, 103, 108, 110, 206 category alteration 84, 93, 94 complex 89, 98, 103, 104, 107, 109, 110 conceptual 95 perceptual 95, 97, 98 value-dominated 95 Chafe 239, 243, 244 clarification request 91, 136, 190, 193, 238, 247, 249–253 Clark 20, 150, 205, 206, 238, 239, 241 co-construction 179, 216, 226 co-text 5, 151, 206, 208, 212, 216, 222, 238, 243, 244, 248, 251, 253–255 cognitive science 33, 48 coherence 8, 13, 118 commitment 60, 69–73, 76, 81, 102, 107 common ground 5, 11, 20, 51, 81, 122, 206, 222, 225, 238–246, 248–259 communicative genre 17, 19–21, 118 move 60, 65, 67, 69, 71, 72 project 17, 19–21 concept formation 83, 85, 93, 95, 96, 100 condition essential 9

262 Index

felicity 6, 9, 19, 56, 61–67, 69, 70, 76, 130, 148, 213 macro 132 micro 132 preparatory 9, 127 propositional content 9 sincerity 9, 124, 125 conditional attenuating 208, 215, 216, 221 form 205, 209, 210 configuration common 15 parts-whole 123, 124 tripartite 124, 125, 128, 130 connective 236, 246–248, 254, 255, 257 consciousness higher-order 82, 83 context cognitive 4, 5, 9, 11, 12, 14, 16, 18, 116, 120, 121, 238, 240, 242, 243, 246, 251, 253–255 construct of 59, 61, 70 default 14, 16 external 16 extrasituational 168, 174, 176 generalized 16 global 122, 227 individual 12 institutional 14 interactional 147, 168–172, 176, 184, 194 internal 55, 61–65, 67, 69, 70, 170 linguistic 5, 6, 9, 12, 14, 16, 120, 121, 168–172, 176, 180, 185, 186, 188, 190, 192 local 121, 152, 206–208, 220, 224, 225, 227, 228 marked 16 organizational 14 particularized 16 prototypical 225, 227, 228 sequential 208 situational 125, 168, 171–173, 176, 182, 184, 194, 195

social 12, 14, 16, 18, 32, 57, 58, 121–123, 138, 247 sociocultural 16, 18, 20, 117, 127, 128, 130, 148 subjective 12 type 13, 225, 227 unmarked 16 contextualization cue 17, 151, 206, 207, 226 device 7, 136, 137, 139 recontextualization 227, 228 contrast 34, 35, 81, 95, 236–239, 255 conventionality 62, 70, 72 cooperation 118, 122, 178, 179, 182, 184, 187, 216, 226 cooperative principle 122, 124, 129, 204 coordination 205, 240, 258 counterfactual 45, 46, 48 cross-examination 180, 191 culture 12, 14, 15, 18, 57, 120, 121, 128, 149, 150, 173, 226 D Damasio 82, 84–86, 88–94, 96, 97, 102 Deacon 81, 82 default 14, 16, 80, 117, 125, 130, 178, 227 Devlin 32, 37–40, 49 dialogue act 118, 122, 124, 125, 128, 130, 132, 133, 241, 244, 246, 247, 253 entity 244, 248, 253–255, 257 structure 235, 238, 244, 245, 248, 251 task-oriented 258 topic 238–241, 243, 244, 246, 248, 254 unit 238, 244 disambiguation 227, 237 discourse analysis 143, 148, 204 argumentative 82, 102–104, 107, 110 conditional 213 institutional 19, 150

marker 107, 108, 136, 137, 139, 169, 207, 222, 225, 246 structure 197 dispreference 154, 225 Dummett 58, 59 dyad 20, 21, 126 E Edelman 82, 85, 86, 90, 92–96, 99, 100, 108 elaboration 176, 183, 185, 190 environment social 12, 104, 109 ethnography 15, 16, 148 ethnomethodology 4, 13 evasive language 178, 194, 196 talk 179 F Fairclough 58, 115, 117, 148, 149, 168, 171, 172, 173, 174, 183 felicity condition 6, 9, 19, 55, 56, 61, 62, 64–67, 69, 70, 76, 130, 148, 213 Ferrara 19, 25 Fetzer 13, 18, 20, 32, 57, 80, 81, 101, 118, 120, 122, 124, 127, 129, 132, 139, 140, 147, 148, 152, 155, 176, 205–207, 222, 225, 226, 228, 247 flout 177, 187 footing 13, 26, 128, 129, 144, 151 force 7, 32, 44, 55, 57, 58, 67, 68, 72–74, 88, 126, 135, 141, 154–156, 167, 170, 183, 216, 245, 246 frame 4, 8–10, 13, 14, 17, 19–22, 44, 71, 116, 118, 121, 122, 126, 151, 194, 215, 222, 228, 240 G game 6, 21, 38, 42, 43, 49, 75, 117, 204 Garfinkel 4, 115, 144 Goodman 43–45 Goodwin and Duranti 4, 167, 169, 173

Index 263

Goody 20 Grammar functional 6, 7, 11, 215 grammaticality 17, 147 Grice 69, 115–118, 122–124, 128–130, 144, 153, 175, 179, 197, 204 ground/grounding common 5, 11, 20, 51, 81, 122, 203, 206, 222, 225, 235, 238, 239–246, 248–257 Gumperz 4, 7, 15–17, 121, 144, 148, 151, 206 H Habermas 115, 116, 122–124 habitus 14, 173 Halliday 7, 118, 167–169, 171 hedge less-fuzzy 139, 140 more-fuzzy 138–140 Heritage 4, 14, 115, 117, 121, 151, 152 I imperfetto 207, 208, 214, 215, 217, 229 implicature conversational 122, 123, 129, 130, 133, 139 generalized 123 particularized 123 inference/inferencing conversational 15–17, 231 process 11, 12, 115 infon 39, 40, 50 information flow 40, 239, 254 seeking 237, 247, 257 structure 237, 238, 243, 244, 246, 253 intention communicative 8, 14, 18, 67, 68, 115, 124, 132, 151 informative 18 intentionality 18, 87, 115, 118, 194 collective 118

internalization 69, 70, 87 interrogation 179, 182, 185–188, 190 interview institutional 4 media 147–150, 153 political 13, 132–134, 141, 148, 149, 153, 155 with experts 147, 156, 157 interviewee 122, 132, 134–136, 149, 157, 162 interviewer 121, 132, 134, 147, 148, 155, 159, 162 invece 238, 247, 254–257 J Janney 5, 168, 170, 173, 176, 177, 179 K know-how 87, 91, 94 L Lakoff 138, 216, 236 language game 117, 126, 170, 204 processing 11, 12, 23, 82, 84 use 3, 9, 12, 56, 58, 115, 117, 132, 135, 169, 204, 228 learning 21, 58, 82, 85, 91, 92, 94, 104, 109 Levinson 5, 13, 17, 56–58, 62, 116, 118, 121, 128, 148, 149, 152, 154, 157, 170, 195, 204, 206, 216 Lewis 45–49 Linell 17, 117, 126, 205 M ma 24, 212, 235, 238, 247–253, 258 Malinowski 169, 171, 198 marker argumentative 100, 110 contrastive 238, 239 discourse 107, 136, 222 maxim of manner 123, 139, 183, 185 of quality 123, 129, 139, 182, 187

of quantity 183, 185 of relation 123, 187, 188 memory declarative 91, 97, 98, 109 explicit 108 implicit 91 procedural 84, 97, 98, 109 recategorical 84, 95 value-dominated 93, 94 mentre 238, 247, 254–257 Mey 56, 57, 118, 171 Minsky 31, 32 misapplication 63 misexecution 64 misfire 63 misinvocation 62 mitigation 68, 183, 184 modality 98, 135, 212, 215 multifunctionality 208 N negotiation co- 239, 258 normative rightness

116, 124

O object vague 41–43 opt out 180 P part dispreferred second pair 223 first pair 223 second pair 223, 225 participation 13, 15, 124, 128, 129, 132, 137, 151, 217, 226 parts-whole 4–10, 19, 118, 123, 124, 205 performative 8, 56, 58, 64, 65, 72–74, 124, 170, 213, 221, 222 performativity 62, 70, 72 però 214, 215, 235, 251–253 Perry 32, 36, 37, 40, 52, 205 personification 62, 69, 70, 72 politeness 129, 137, 148, 207, 216 polysemy 207, 208, 227, 228

264 Index

practice interviewing 147, 163 social 127, 172, 174, 204, 228 predicate evaluative 215, 216 preference 90, 154, 223 presupposition downgrading 210 maximal 122 medium 123 minimal 122, 123 processing efficiency 83, 88 language 11, 12, 82, 84 prolixity 123, 183, 185, 190 prominence 50, 244 Q question question-response pair 154, 179 Quine 31, 52 R ratification 116, 122–125, 132, 156, 241 rationality 115, 116, 122 reality 21, 59, 60, 112 reasoning practical 11, 32, 57 Recanati 12, 44, 49 reentry 85, 86, 93, 96, 99 reinforcement 68, 109 relevance conditional 4, 128, 129 theory 11 repetition 169, 186, 224 representation dispositional 92, 96–99 resemblance 47, 48, 88, 94, 210 response option 89, 93, 97–99, 107, 108 tactic 179, 196 type 171, 191 rheme 244, 255 rule constitutive 6, 22, 62 regulative 6

S salience 34 Saville-Troike 15, 16, 130 Sbisà 9, 67, 68, 115, 118, 122, 125, 167, 168, 170–172, 174, 176, 216, 224 Schegloff 4, 14, 118, 150–152, 204 Searle 6, 8, 9, 49, 62, 67, 68, 80, 83, 86–91, 103, 106, 115, 118, 124, 126, 148, 170, 204 sequence argumentative 107, 109, 135 macro 118, 119, 121 micro 118–120 negotiation-of-meaning 143 negotiation-of-validity 122, 141 question-answer 136, 178 sequentiality 22, 24, 124 Shuy 192, 194 similarity comparative 46, 49 structural 43, 47 sincerity 9, 124, 125 situation abstract 39 fictional 36 future 36 game 38, 42, 43, 49 historic 55, 57–62, 64–66, 70–72, 75 initial 89, 91, 93, 97–99, 105–108 of reference 98, 103, 107 past 36 schematized 107, 109–110 similar 31, 32 theory 31, 32, 35, 39–41, 49, 51 speech act 3, 8, 9, 19, 20, 25, 55, 57, 61, 62, 67, 68, 115, 118, 120, 122, 124, 125, 127, 130, 170, 171, 210, 213, 216, 222 act theory 3, 19, 55, 67, 68, 115, 118, 122, 124, 130 community 15, 16, 56, 116, 127

event 14, 15, 149, 151, 152 Sperber and Wilson 4, 11, 18, 68, 121, 206 substantiation 65, 66 system interactional 124, 128–130, 132, 135, 136 interpersonal 124, 128–130, 132, 136–138, 141 textual 124, 128–130, 132, 133, 138–140 T theme 48, 119, 144, 192, 193, 244, 248, 249, 254, 255 token 11, 209, 223 top-down 116, 119, 121 topic management 238, 240, 246, 248 truth condition 8, 125 turn constructional unit 120 taking 226 Tversky 32, 34, 35 type 12, 40, 57, 207 U uptake 55, 60, 71, 72, 86 utterance non-performative 56, 73 performative 56, 58, 72, 73, 75 V vagueness ontic 41 validity claim macro 20, 140–141 micro 142 minus 122, 124, 125, 128, 130, 132, 133 plus 122, 124, 125, 128, 130, 132, 133 value adaptive 90, 95 van Dijk 12, 13, 18, 19, 115, 117, 125, 126, 148

Index 265

verb cognitive 182 modal 211, 215, 221 of saying 221 perceptual 182 reported 182

W well-formedness 17, 19, 57, 228 Widdowson 167, 175 Wittgenstein 46, 71, 88, 170, 204

world objective 116, 124, 125 possible 45, 49, 245 social 116, 124, 128–132, 135, 141, 204 subjective 116, 124, 125

Pragmatics & Beyond New Series A complete list of titles in this series can be found on the publishers’ website, www.benjamins.com 166 NakaNe, Ikuko: Silence in intercultural communication. Perceptions and performance. Expected October 2007 165 BuBlItz, Wolfram and axel HüBler (eds.): Metapragmatics in Use. Expected October 2007 164 eNgleBretsoN, robert (ed.): Stancetaking in Discourse. Subjectivity, Evaluation, Interaction. Expected October 2007 163 lytra, Vally: Play Frames and Social Identities. Contact encounters in a Greek primary school. xi, 291 pp. + index. Expected September 2007 162 Fetzer, anita (ed.): Context and Appropriateness. Micro meets macro. 2007. vi, 265 pp. 161 Celle, agnès and ruth Huart (eds.): Connectives as Discourse Landmarks. 2007. viii, 212 pp. 160 Fetzer, anita and gerda eva lauerBaCH (eds.): Political Discourse in the Media. Cross-cultural perspectives. 2007. vii, 380 pp. 159 MayNard, senko k.: Linguistic Creativity in Japanese Discourse. Exploring the multiplicity of self, perspective, and voice. xv, 356 pp. Expected July 2007 158 Walker, terry: Thou and You in Early Modern English Dialogues. Trials, Depositions, and Drama Comedy. 2007. xx, 339 pp. 157 CraWFord CaMICIottolI, Belinda: The Language of Business Studies Lectures. A corpus-assisted analysis. 2007. xvi, 236 pp. 156 Vega MoreNo, rosa e.: Creativity and Convention. The pragmatics of everyday figurative speech. xii, 243 pp. + index. Expected August 2007 155 HedBerg, Nancy and ron zaCHarskI (eds.): The Grammar–Pragmatics Interface. Essays in honor of Jeanette K. Gundel. 2007. viii, 345 pp. 154 HüBler, axel: The Nonverbal Shift in Early Modern English Conversation. 2007. x, 281 pp. 153 arNoVICk, leslie k.: Written Reliquaries. The resonance of orality in medieval English texts. 2006. xii, 292 pp. 152 WarreN, Martin: Features of Naturalness in Conversation. 2006. x, 272 pp. 151 suzukI, satoko (ed.): Emotive Communication in Japanese. 2006. x, 234 pp. 150 Busse, Beatrix: Vocative Constructions in the Language of Shakespeare. 2006. xviii, 525 pp. 149 loCHer, Miriam a.: Advice Online. Advice-giving in an American Internet health column. 2006. xvi, 277 pp. 148 FløttuM, kjersti, trine daHl and torodd kINN: Academic Voices. Across languages and disciplines. 2006. x, 309 pp. 147 HINrICHs, lars: Codeswitching on the Web. English and Jamaican Creole in e-mail communication. 2006. x, 302 pp. 146 taNskaNeN, sanna-kaisa: Collaborating towards Coherence. Lexical cohesion in English discourse. 2006. ix, 192 pp. 145 kurHIla, salla: Second Language Interaction. 2006. vii, 257 pp. 144 BüHrIg, kristin and Jan d. ten tHIJe (eds.): Beyond Misunderstanding. Linguistic analyses of intercultural communication. 2006. vi, 339 pp. 143 Baker, Carolyn, Michael eMMIsoN and alan FIrtH (eds.): Calling for Help. Language and social interaction in telephone helplines. 2005. xviii, 352 pp. 142 sIdNell, Jack: Talk and Practical Epistemology. The social life of knowledge in a Caribbean community. 2005. xvi, 255 pp. 141 zHu, yunxia: Written Communication across Cultures. A sociocognitive perspective on business genres. 2005. xviii, 216 pp. 140 Butler, Christopher s., María de los Ángeles góMez-goNzÁlez and susana M. doVal-suÁrez (eds.): The Dynamics of Language Use. Functional and contrastive perspectives. 2005. xvi, 413 pp. 139 lakoFF, robin t. and sachiko Ide (eds.): Broadening the Horizon of Linguistic Politeness. 2005. xii, 342 pp. 138 Müller, simone: Discourse Markers in Native and Non-native English Discourse. 2005. xviii, 290 pp. 137 MorIta, emi: Negotiation of Contingent Talk. The Japanese interactional particles ne and sa. 2005. xvi, 240 pp.