Anita Fetzer - Context and Appropriateness PDF

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: ahjucker@es.uzh.ch
Associate Editors
Jacob L. Mey Herman Parret Jef Verschueren
University of Southern Belgian National Science Belgian National Science
Denmark Foundation, Universities of Foundation,
Louvain and Antwerp University of Antwerp
Editorial Board
Shoshana Blum-Kulka Susan C. Herring Emanuel A. Schegloff
Hebrew University of Indiana University University of California at Los
Jerusalem Angeles
Masako K. Hiraga
Jean Caron St.Pauls (Rikkyo) University Deborah Schiffrin
Universit de Poitiers Georgetown University
David Holdcroft
Robyn Carston University of Leeds Paul Osamu Takahara
University College London Kobe City University of
Sachiko Ide
Foreign Studies
Bruce Fraser Japan Womens University
Boston University
Catherine Kerbrat- Sandra A. Thompson
University of California at
Thorstein Fretheim Orecchioni
Santa Barbara
University of Trondheim University of Lyon 2
John C. Heritage Claudia de Lemos Teun A. van Dijk
Pompeu Fabra, Barcelona
University of California at Los University of Campinas, Brazil
Angeles
Marina Sbis Richard J. Watts
University of Berne
University of Trieste
Volume 162
Context and Appropriateness. Micro meets macro
Anita Fetzer (ed.)
Context and Appropriateness
Micro meets macro
Edited by
Anita Fetzer
Lueneburg University
John Benjamins Publishing Company

Amsterdam / Philadelphia
TM The paper used in this publication meets the minimum requirements of
8
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, 1958-
P302.C6225 2007
306.44--dc22 2007015637
isbn 978 90 272 5406 1 (Hb; alk. paper)
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 3

Anita Fetzer
Part I. Bridging problems between context and appropriateness
Similar situations 31
Varol Akman
Appropriateness and felicity conditions: A theoretical issue 55
Etsuko Oishi
Appropriateness: An adaptive view 79
Thanh Nyan
Party II. Bridging problems between communicative action

and appropriateness
If I may say so: Indexing appropriateness in dialogue 115

Anita Fetzer
The appropriateness of questions 147
Annette Becker
Cooperative conflict and evasive language:
The case of the 911 commission hearings 167
Lawrence N. Berlin
vi Table of contents
Part III. Bridging problems between micro and macro
The attenuating conditional: Context, appropriateness and interaction 203

Johanna Miecznikowski and Carla Bazzanella
Collaborative use of contrastive markers: Contextual
and co-textual implications 235
Francesca Carota
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 re-
sult 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 theo-
retical and applied perspectives. Thus, the papers range from the research para-
digms of philosophy of language, speech act theory, sociopragmatics, cognitive
pragmatics and critical discourse analysis. The variety of topics, which the indi-
vidual contributions cover, allows for a refined bottom-up and top-down analysis
of the concepts of context and appropriateness as well of their multifaceted con-
nectedness. 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, engi-

neering, 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 Ben-
jamins 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 ap-
plication to a multitude of domains, such as natural- and non-natural language
communication, computer-mediated interaction and information technology, ro-
botics, 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 defini-
tion or one commonly shared accepted theoretical perspective, and frequently,
only a minute aspect of context is described, modelled or formalized (cf. the in-
terdisciplinary 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 lon-
ger looked upon as an analytic prime but rather seen from a parts-whole per-
spective 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 con-
tent while at the same time being framed and delimited by less immediate ad-
jacent 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 cogni-
tive 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 or-
ganized in and through the process of communication. This view prevails in the
domains of ethnomethodology (Garfinkel 1994; Goodwin & Duranti 1992; Heri-
tage 1984; Schegloff 1992), interactional sociolinguistics (Gumperz 1996, 2003)
and sociopragmatics (Fetzer 1999, 2004), where context is assigned the dual sta-
tus 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 con-
text. 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-
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 re-
lational conception which conceives it as a relational construct, relating commu-
nicative 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 informa-
tion (Stalnaker 1999). Here, context is seen as a set of propositions which partici-
pants 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
Levinsons (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 infor-
mation 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 con-
text as a whole is categorized into linguistic context, cognitive context, sociocul-
tural 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) de-
notes 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
6 Anita Fetzer
a whole) is looked upon analogously to Searles 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 inter-
preting speech acts. While the rule-governed realization of linguistic construc-
tions is constrained by the rules of grammar, the production and interpretation of
speech acts are constrained by felicity conditions, as is shown by Oishis contribu-
tion 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 languages
constitutive parts of syntax, morphology, phonology, semantics and pragmatics.
While syntax is composed of structural units, for instance constituents in tradi-
tional 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 gram-
matical 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 re-
ally 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 con-
struction 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 sentences to-
pological units of pre-field, middle-field and post-field, and their respective sub-
fields, 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
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 mak-
ing 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 Pe-
ter 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 com-
pound [[over][take]], or the derivational morpheme and free morpheme [ism]
in [[minimal][ism] and in theres too much ism nowadays. As has been shown for
the domain of syntax, an analysis of morphology from a context-anchored parts-
whole 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 real-
ized 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-seg-
mental phonology and here in particular in the field of intonation with its unit of
investigation, the intonational phrase. In systemic-functional grammar, intona-
tion is seen as a signalling system (Halliday 1994), and in interactional sociolin-
guistics an intonational phrase is assigned the status of a contextualization device
(Gumperz 1996). For instance, the intonational contour of a fall in English signi-
fies the illocutionary force of request. The intonational contour of a rise, by con-
trast, 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
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 commu-
nicative 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 context-
dependent meaning. From a parts-whole perspective, truth-conditional seman-
tics 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 proposi-
tions 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 ac-
commodation 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 acts constitutive acts (except the perlocutionary
act) are conventional in nature, and it is their performance in an appropriate con-
text which makes them count as a particular speech act. For instance, the utter-
ance 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 re-
quest, and the propositional act is composed of the reference act you referring to
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 so-
cial 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, Sbis (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 na-
ture and into the nature of their connectedness. This also holds for the fields of se-
mantics 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 syn-
tax, morphology, phonology and semantics leads to more refined from results
regarding the grammar-interaction interface. Not only does this allow for a sys-
tematic 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 cogni-
tive-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 de-
limiting 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 theorys differentiation be-
10 Anita Fetzer
tween set and nonset, which like the gestalt-psychological concepts of figure
and ground are not symmetrically related. To use Batesons own words: Per-
ception 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 metacom-
municative, or to use Batesons 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 proposi-
tion 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 relation-
al 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 Goffmans
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 conven-
tions 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 in-
trinsically 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 expres-
sion. 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 as-
sumptions 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-
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) differ-
entiates 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 pos-
sibly 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 repre-
sented by different layers of an onion. What is of importance for language pro-
cessing 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, Carotas
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
Givn (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 rel-
evant ground vis--vis which tokens of experience achieve relatively stable mental
representation as salient figures. Whatever stability mental representations pos-
sess is due, in large measure, to the classification of tokens of experience into ge-
neric categories or types (Givn 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, Nyans 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.
12 Anita Fetzer
A similar type-token differentiation is manifest in Pencos distinction be-

tween 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 individuals set of beliefs and thus to their representation of the world.
The connectedness between individual contexts and subjective contexts is exam-
ined in Akmans contribution, which demonstrates in what ways and for what
purposes situations are categorized as similar or as different.
A necessary consequence of Givns differentiation between type of expe-
rience and token of experience, and of Pencos differentiation between subjec-
tive context and individual context is that cognitive context is both individual
(or unique) and social (or typed). This ambivalence is also reflected in van Dijks
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 subjec-
tive 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-communi-
cative 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 ex-
amined in more detail.
1.3 Social context
Social context is often considered to comprise the context of a communicative ex-

change 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.
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 com-
municative 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 institu-
tional power which manifests itself in their corresponding rights and obligations.
This is investigated in Beckers 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 par-
ticular 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 Fetzers and Miecznikowski and Bazzanellas contributions. The former
examines the communicative function of implicit and explicit references to ap-
propriateness 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 con-
stituents to his examination of variation in spoken language and in written lan-
guage. According to him, the components of a speech situation do not only con-
tain 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 audi-
ence, 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 excel-
lence, in which the indexicality of social action is key. Ethnomethodology focuses
on the domain of intersubjectivity and it examines the questions of how separate
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 interactions mo-
ment-to-moment development, the parties, singly and together, select and display
in their conduct which of the indefinitely many aspects of context they are mak-
ing 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 delimit-
ing 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 so-
cially 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 con-
nectedness 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 regard-
ing appropriate use of language, the knowledge of activity-types including com-
municative intentions and goals, and general background knowledge. Of course,
the immediate extra-linguistic context is embedded in more remote extra-linguis-
tic contexts, such as particular organizational contexts and other socio-histori-
cally 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 con-
text 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
interpreted in a particular mode. For instance, the distinction between mono-

chronic time and polychronic time is based on the differentiation between a lin-
ear, tangible and divisible conception of time, where events are scheduled one
at a time and where this schedule takes precedence over interpersonal relation-
ships. Polychronic time, by contrast, is characterized by things occurring simul-
taneously. Here, interpersonal relationships take precedence over a task-oriented
outlook on communication (Hall & Hall 1989). Thus, culture provides the mem-
bers 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 se-
quential 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 con-
tent), 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.
Hymess speaking grid and his notion of communicative competence has
been refined by Gumperz (1977, 1992) who explicitly connects the cognitive op-
eration of inference with the sociocultural activity of conversation. His concep-
tion of conversational inference represents a context-bound process of interpreta-
tion in which others 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 com-

prise national culture and national subcultures, ethnicity- and gender-related cultures and eth-
nicity- 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).
16 Anita Fetzer
as a social activity. Saville-Troike summarizes his contribution to the field of eth-

nography 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.
Gumperzs 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, Gumperzs interactional socio-
linguistics accommodates a micro and macro outlook on communication which
is not only reflected in the participants communicative performance but also in
their conversational inferencing, or to use Gumperzs own words (2003: 14):
It is useful to distinguish between two levels of inference in analyses of interpre-
tive processes: (a) global inferences of what the exchange is about and what mu-
tual 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 intro-
duction of micro (or local) linguistic, cognitive, social and sociocultural contexts,
and macro (or global) linguistic, cognitive, social and sociocultural contexts. Fur-
thermore, 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 micro-
macro distinction hold for linguistic, cognitive, social and sociocultural contexts,
but so does the generalized-particularized differentiation.
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 so-
ciolinguistics views language as a socially situated form and assigns inference the
status of conversational inference which is a context-dependent cognitive opera-
tion par excellence. Its unit of investigation is the speech activity, which is closely
connected with Levinsons activity type (Levinson 1979), Linells communicative
project (Linell 1998) and Luckmanns 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, psycholo-
gists, 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 in-
terpretative 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 con-
tent. 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 physi-
cal 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 ac-
cordance 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
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 Dijks 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 accept-
ability of an utterance is evaluated with regard to the nature of the connectedness
between its linguistic form, sequential position and social context, and the appro-
priateness of an utterance is evaluated with regard to the nature of the connected-
ness between a communicative action, its linguistic realization and its embedded-
ness in linguistic and social contexts (Fetzer 2004: 1920). 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 communi-
cative 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 inten-
tion 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: 5464). 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 Haferkamps 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 societys culture (Hafer-
kamp 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 institu-
tion. 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 intentional-
ity 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.
2.1 Appropriateness conditions
In a social- and socioculturally oriented examination of speech act theory, felic-

ity conditions have been assigned the status of appropriateness conditions. This
is particularly true for a macro-oriented rethinking of speech act theory focus-
sing on the well-formedness and appropriateness of longer stretches of discourse,
which encounters the following difficulties, to use Ferraras 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 linguis-
tic 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 per-
spective 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). Adopt-
ing 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 in-
dividual 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 communica-
tion, Berlins 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 institu-
tional 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 dialogi-
cal concept. Not only is appropriateness dialogical, but it is also a macro concept
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 ap-
propriateness 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: Con-
sider 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 con-
nectedness 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 appropriate-
ness with style thus going beyond the production and interpretation of linguistic
acts. Again, appropriateness is seen as intrinsically connected with both sociocul-
tural contexts and communicative genres. To use Goffmans own words: Indeed,
all our so-called diffuse social roles can be seen partly as styles, namely the man-
ner 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 Goffmans ar-
gument 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-
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 or-
der, 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 dyads commu-
nicative 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 delim-
iting 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 com-
bined 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, func-
tion, content, or some other factor or sets of factors (Bauman 1992: 138).
By introducing genre as a bridging notion between micro and macro, the dis-
tinction 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, linguis-
tically speaking, and moving towards a mediated-structure oriented analysis fo-
cussing 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:
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 indis-
pensable prerequisites of these actions. In consequence the relation between ac-
tion and categorial structure, between language and speech act, is viewed not as
an empirical relationship of causation but as an analytical relationship of consti-
tution. (...)
Categorizing the rules of language (that is, the constitutive rules) as mac-
rostructures 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 appropri-
ateness 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 bridg-
ing problems between context and appropriateness. The contributions subsumed
under this header employ a top-down perspective anchored to the research para-
digms of situation theory, speech act theory and cognitive pragmatics. The sec-
ond part investigates bridging problems between communicative action and ap-
propriateness 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 con-
trastive markers and their communicative functions in context, and the nature of
their embeddedness in context. The contributions start from a bottom-up per-
spective which is explicitly connected with macro constraints, such as sequential-
ity, genre and domain.
Part I, Bridging problems between context and appropriateness, comprises three

contributions. Varol Akmans Similar situations studies the notion of similarity
with reference to situations of situation theory. It shows that the commonsense
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. Par-
ticular reference is given to the contextual constraints on situations and to their
connectedness with appropriateness.
Etsuko Oishis Austins speech act theory and context gives a speech-act-theo-
retic explanation of the concept of appropriateness. The performance of an il-
locutionary 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 Nyans 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 mat-
ter 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 Fetzers If I may say so: Indexing appropriate-
ness 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 sociocul-
tural context.
Annette Beckers The appropriateness of questions investigates the sociocultur-
al construct of appropriateness in the dialogical discourse type of media interview
using methods from the fields of conversation analysis, pragmatics, socioprag-
matics, 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. Berlins Cooperative conflict and evasive language: The case of the
911 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.
24 Anita Fetzer
Part III, Bridging problems between micro and macro, comprises two contribu-
tions. Johanna Miecznikowski and Carla Bazzanellas 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 Carotas Collaborative use of contrastive markers co-textual and
contextual implications examines the Italian contrastive markers ma (but), invece
(instead), mentre (while) and per (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, Context99, Trento,
Italy, September 1999, Proceedings, P. Bouquet, P. Brezillon and L. Serafini (eds.), 114.
Heidelberg: Springer.
Akman, V., Bouquet, P., Thomason, R. and Young, R.A.(eds.). 2001. Modeling and Using Con-
text. Third International and Interdisciplinary Conference, CONTEXT 2001, Dundee, UK,
July 2001, Proceedings. Springer: Heidelberg.
Alexander, J.C., Giesen, B., Mnch, 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. Mnch and N.J. Smelser (eds.), 1-42. Berkley: The University of California
Press.
Attaro, S. 2000. Irony as relevant inappropriateness. Journal of Pragmatics 32: 793826.
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 kraftskld. In Rethinking Context: Language as an Interactive Phenomenon, A. Duranti
and C. Goodwin (eds.), 125146. 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 2325, 2003. Heidelberg: Springer.
Bouquet, P., Brezillon, P., Benerecetti, M., Castellani, F. and Serafini, L. (eds.). 1999. Model-
ing and Using Context. Second International and Interdisciplinary Conference on Model-
ing and Using Context, Context99, 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): 93118.
Couper-Kuhlen, E. 1993. English Speech Rhythm: Form and Function in Everyday Verbal Inter-
action. Amsterdam: Benjamins.
De Beaugrande, R. and Dressler, W. 1981. Einfhrung in die Textlinguistik. Tbingen: Niemey-
er.
Drew, P. and Heritage, J. (eds.). 1992. Talk at Work: Interaction in Institutional Settings. Cam-
bridge: Cambridge University Press.
Ferrara, A. 1980. Appropriateness conditions for entire sequences of speech act. Journal of
Pragmatics 4: 321340.
Fetzer, A. 1999. Non-acceptances: Re- or un-creating context. In Modeling and Using Con-
text. 2nd International and Interdisciplinary Conference, Context99, Trento, Italy, Septem-
ber 1999, Proceedings, P. Bouquet, P. Brezillon and L. Serafini (eds.), 133144. Heidelberg:
Springer.
Fetzer, A. 2000. Negotiating validity claims in political interviews. Text 20(4): 146.
Fetzer, A. 2002. Communicative intentions in context. In Rethinking Sequentiality: Linguistics
meets Conversational Interaction, A. Fetzer and C. Meierkord (eds.), 3769. 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. Mnch and N.J. Smelser (eds.), 337355.
Berkley: The University of California Press.
Givn, 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 Intel-
ligence, E. Goody (ed.), 136. 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.), 142.
Cambridge: Cambridge University Press.
Gumperz, J.J. 1977. Sociocultural knowledge in conversational inference. In Linguistics and
Anthropology, M. Saville-Troike (ed.), 191211. 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.), 229252. Cambridge:
Cambridge University Press.
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.), 374406. Cambridge: Cambridge Uni-
versity Press.
Gumperz, J.J. 2003. Response essay. In Language and Interaction. Discussions with John J.
Gumperz, S. Eerdmans, C. Prevignano and P. Thibault (eds.), 105126. Amsterdam: Ben-
jamins.
Haferkamp, H. 1987. Complexity and behaviour structure, planned associations and creation
of structure. The Micro-Macro Link, J.C. Alexander, B. Giesen, R. Mnch and N.J. Smelser
(eds.), 177192. Berkley: The University of California Press.
Hall, E.T. and Hall, M.R. 1989. Understanding Cultural Differences: Germans, French and Amer-
icans. 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 Rel-
ativity, J.J. Gumperz and S.C. Levinson (eds.), 232270. Cambridge: Cambridge University
Press.
Heritage, J. 1984. Garfinkel and Ethnomethodology. Cambridge: Polity Press.
Hymes, D. 1974. Foundations in Sociolinguistics: An Ethnographic Approach. Philadelphia: Uni-
versity of Philadelphia Press.
Janney, R.W. 2002. Cotext as context: Vague answers in court. Language and Communication
22(4): 457475.
Lerner, G. (ed.). 2004. Conversation Analysis. Studies from the First Generation. Amsterdam:
Benjamins.
Levinson, S.C. 1979. Activity types and language. Linguistics 17: 365399.
Levinson, S.C. 1988. Putting linguistics on a proper footing: Explorations in Goffmans con-
cepts of participation. In Erving Goffman. Exploring the Interaction Order, P. Drew and A.
Wootton (eds.), 161227. 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.), 175188. Cambridge: Cambridge
University Press.
Penco, C. 1999. Objective and cognitive context. In 2nd International and Interdisciplinary
Conference on Modeling and Using Context (Context99), P. Bouquet, P. Brezillon and L.
Serafini (eds.), 270283. Heidelberg: Springer.
Pierrehumbert, J. and Hirschberg, J. 1992. The meaning of intonational contours in the inter-
pretation of discourse. In Intentions in Communication, P.R. Cohen, J. Morgan and M.E.
Pollack (eds.), 271311. Cambridge: MIT Press.
Recanati, F. 1998. Meaning and force: Introduction. In Pragmatics: Critical Concepts, A. Kash-
er (ed.), 126154. London: Routledge.
Sarangi, S. and Slembrouck, S. 1996. Language, Bureaucracy & Social Control. London: Long-
man.
Saville-Troike, M. 1989. The Ethnography of Communication. Blackwell, Oxford.
Sbis, M. 2002. Speech acts in context. Language and Communication 22(4): 421436.
Schegloff, E.A. 1987. Between macro and micro: Contexts and other connections. In The Mi-
cro-Macro Link, J. Alexander, B. Giesen, R. Munch and N. Smelser (eds.), 207234. 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.), 191228. Cambridge: Cambridge Uni-
versity Press.
Searle, J.R. 1969. Speech Acts. Cambridge: Cambridge University Press.
Sperber, D. and Wilson, D. 1986. Relevance: Communication and Cognition. Oxford: Black-
well.
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.), 4162. 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): 159177.
Westmyer, S., DiCiocci, R. and Rubin, R. 1998. Appropriateness and effectiveness of commu-
nication channels in competent interpersonal communication. Journal of Communication
48(3): 2748.
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 situa-
tion 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 phil-
osophical literature. That human beings can use the notion naturally (without
much effort) shows that their cognitive make-up is probably much more power-
ful 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)
Ive 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 ones 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 (Glei-
tman 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 remark-
able that we human beings have the cognitive make-up to use it with ease and as-
surance, 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 partici-
pants, 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 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 . If a certain action causes to become bigger, you might want to look for
another action that counteracts the former.
2. According to Tversky and Gati (1978), in saying is like , we are inclined to think of
as the subject and 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 so-
cial- and communicative-action-based construct which is calculated with regard to the connect-
edness 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.
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 intel-
ligence (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 concern-
ing what is known as analogical transfer. Subjects are presented with an assign-
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 suc-
cess 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 pro-
posal that each object is characterized by a set of features . The observed simi-
larity of two objects, and , 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 and , i.e. (set intersection)

The features of that are not shared by , i.e. (set difference)
The features of that are not shared by , i.e.
In what is commonly called a contrast model, the similarity of to is given as a

linear combination of the measures of their joint and idiosyncratic features, viz.
c1 f() c2 f( ) c3 f( ), 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 di-
mension 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 Tverskys theory of similarity and its
implications re simile and metaphor.
is evident: Poland and Hungary have or had, at the time of the experiment sa-
lient 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 appropri-
ateness).
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 philoso-
phy. 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 re-
sponse 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 disap-
pointing 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 situa-
tions. An example might help at this point. I have recently been in a flight situa-
tion 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.
36 Varol Akman
Situations usually have individuals standing in relation at various (spatiotem-

poral6) 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 indi-
viduals, properties, relations, or locations that endure from one situation to an-
other. 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 oclock 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 situ-
ation, 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 Encyclope-
dia of Philosophy http://plato.stanford.edu/entries/reasoning-defeasible), it may nonetheless be
quite useful for assorted purposes.
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 differ-
ent 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 situa-
tions 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, indi-
viduate) a situation. Two obvious examples are direct perception8 of a situation,
perhaps the immediate environment, or thinking about a particular situation, say
last nights 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: 212213) explains the prob-
lem:
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 neuro-
physiological 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 avail-
able 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 informa-
tion which he visually acquires. But, even in its most basic form [] this informa-
tion 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.
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 situa-
tion comprises. Take this case of two friends (Devlin 1991: 3132):
[] 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 situa-
tion 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 dis-
cussion 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, informa-
tion 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 dura-
tion in time. What differs is what each knows about that situation.
Let us call what John and David know about the game situation their epistemo-
logical 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 re-
spects. 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 sig-
nificant 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 confu-
sion. 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 nights 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.
One of the attractive features of situation theory is its insistence on an infor-

mation-based account of communication (an account first championed by Fred
Dretske). To this end, infons are posited as discrete items of information. An in-
fon supplies a single piece of information. It is denoted as the (n + 2)-tuple <R,
a1, , an, p>, 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 (re-
spectively, 0) then a1, , an stand (respectively, do not stand) in the relation R.
Take the infon <father of, George Herbert Bush, George Walker Bush, 1>. This is
a fact (corresponds to the way things are in Reality). On the other hand, the infon
<father of, George Herbert Bush, Prescott Bush, 1> 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 { | s |= }, where 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
denoted as s |= above just in case the infon 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 = { | Reality |= }, 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 loca-
tion depend upon this. In other words, the basic constituents of the theory are de-
termined by the agents 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 lo-
cation. (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 re-
9. 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 <GHB, GWB> (see
presently).
40 Varol Akman
sults in parametric infons. For example, <sees, , Alice, 1> and <sees, , , 1> are
parametric infons where and stand for individuals. These infons are paramet-
ric on the first, and the first and second argument roles of the relation see, respec-
tively.10 Anchoring (binding) parameters of an infon to objects yields parameter-
free infons. For example, given <sees, , Alice, 1>, if F() = 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 differ-
ent 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 ac-
tual 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 <feeds, feeder agents, feeded animals, food> 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 ab, an agent
can pick up the information that there is a fire in a particular situation by observ-
ing that there is smoke.11 Anchoring plays a major role in the working of con-
straints. 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. ab is shorthand for the factual, parameter free infon <involves, a, b, 1>, 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 [].
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.
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 ob-
ject.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: 159160). 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 To-
ronto 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.
Keefes 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 xs
extent being determinate. A mountain is not, after all, something essentially demar-
cated by its extent or boundary. It is not as if there were just as many mountains to
be found with xs peak as there were rival determinations of xs boundary. An idea
like that could not even occur to one with the good fortune to be innocent of classical
extensional mereology.
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 situa-
tions 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 re-
veal 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 re-
spect 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 computation-
al 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.
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 minds 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 re-
spect two objects are similar; otherwise, we would be making a vacuous state-
ment. 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, Goodmans 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 simi-
lar 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 Goodmans 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 (Good-
man 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).
[C]onsider what one might know about quaggas (some hypothetical or unfamil-
iar 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 inter-
related set of properties associated with zebras, whereas striped seems more to
be an isolated property.20 In any event, Goodmans framework would find the
comparison completely uninformative until the respects were specifically men-
tioned.
Nonetheless, Medin and Goldstone (1995: 106) concur with the general thesis and
extend Goodmans 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 (conse-
quent).
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 pre-
dicaments 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.
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 prelimi-
nary analysis of counterfactual conditionals as follows.21
Let A C denote the counterfactual conditional with antecedent A and con-
sequent 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 peo-
ple 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 similar-
ity closeness of worlds, and that comparative similarity is hopelessly impre-
cise 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 bal-
ance off various similarities and dissimilarities according to the importance we
attach to various respects of comparison and according to the degrees of similar-
ity in the various respects.
21. By preliminary, we imply that Lewis offers further (more fine-grained and accurate) analy-
ses 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 worlds 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 tele-
phones 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 philosophi-
cal writings of Wittgenstein similar, overall, to those of Heidegger? I dont 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.
Lewiss 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 lo-
comotive. 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). How-
ever, 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 Lewiss related work, counterparts of persons inhabitants of other worlds
who bear a resemblance to him closely (or more closely, compared to other inhab-
itants 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 ex-
cerpt from (Weitzenfeld 1984: 138) is relevant:
Consider, for example, two cars known to come from the same assembly line at ap-
proximately the same time. They are known to be identical in structure and composi-
tion 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 un-
known car is yellow? Clearly there must be some further link between the premises
of an argument by analogy and its conclusion.
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 impor-
tances 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). Lewiss following example is particularly
instructive (1983b: 5152):
[C]ounterpart relations are a matter of over-all resemblance in a variety of re-
spects. If we vary the relative importances of different respects of similarity and
dissimilarity, we will get different counterpart relations. Two respects of similar-
ity 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 propos-
ing 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 ones disposal, the following rendering of coun-
terfactual 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 A-
permitting sphere around i, if there is such a sphere.
Lewiss 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 Lewiss overall goals
vis--vis similarity rather clear: (i) To begin, I take as primitive a relation of comparative over-
noted that the spheres around a world are nested and thus they can be said to pre-
serve or model comparative closeness in some sense. But this is taking us to the
technicalities of Lewiss theory and with all due respect for the latters 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 un-
derstanding 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, Devlins football game situation), we make certain fea-
tures 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 de-
all 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 similar-
ity 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 as-
sociate 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 portions shape can be difficult to surmise
from looking at what is visible above the surface.
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 1 and 2 supported by s, viz. s |= 1 and s |= 2. Assume
that the infon 1 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 2 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 1 and forget about 2.
The aforementioned exercise can be repeated with different features in an as-
sortment 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 back-
ground), 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 tex-
ture (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 simi-
larity comparison using a novel situation, the novel situation should resemble the source situ-
ations 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.
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 embed-
ded 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 co-
participants are in some way knowledgeable about the aspects of dads 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 sever-

al 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 meta-
phor). 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!
52 Varol Akman
You come to recognize these source situations as say, the White House spokes-
persons 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 sce-
nario 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 demonstra-
tions 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), re-
spectively.
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: Algo-
rithm and examples. Artificial Intelligence 41: 163.
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 Quines paper discusses concept development and is con-
cerned with children developing similarity spaces in diverse domains.
Gentner, D. 1998. Analogy. In A Companion to Cognitive Science, W. Bechtel and G. Graham

(eds.), 107113. Oxford, UK: Blackwell.
Gentner, D. 1999. Analogy. In The MIT Encyclopedia of the Cognitive Sciences, R.A. Wilson
and F.C. Keil (eds.), 1720. Cambridge, MA: MIT Press.
Gentner, D. and Markham, A.B. 1995. Similarity is like analogy: Structural alignment in com-
parison. In Similarity in Language, Thought and Perception [Semiotic and Cognitive Stud-
ies 1], C. Cacciari (ed.), 111147. Turnhout, Belgium: Brepols.
Gick, M.L. and Holyoak, K.C. 1983. Schema induction and analogical transfer. Cognitive Psy-
chology 15 (1): 138.
Gleitman, L.R., Gleitman, H., Miller, C. and Ostrin. R. 1996. Similar, and similar concepts.
Cognition 58 (3): 321376.
Goodman, N. 1972. Seven strictures on similarity. In his Problems and Projects, 437447.
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:
Kittay, E.F. 1982. The creation of similarity: A discussion of metaphor in light of Tverskys
theory of similarity. In Proceedings of the Biennial Meeting of the Philosophy of Science As-
sociation, Volume 1, 394405. 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, 2646. New York: Oxford University Press.
Lewis, D. 1983b. Counterparts of persons and their bodies. In his Philosophical Papers, Vol-
ume I, 4754. New York: Oxford University Press.
Lewis, D. 1986a. Counterfactuals and comparative possibility. In his Philosophical Papers, Vol-
ume II, 331. New York: Oxford University Press.
Lewis, D. 1986b. Counterfactual dependence and times arrow. In his Philosophical Papers,
Volume II, 3252. New York: Oxford University Press.
Lewis, D. 1986c. Postscripts to Counterfactual dependence and times arrow. In his Philo-
sophical Papers, Volume II, 5266. New York: Oxford University Press.
Lewis, D. 1986d. Causation. In his Philosophical Papers, Volume II, 159172. New York: Ox-
ford University Press.
Markman, A. B. and Gentner, D. 2000. Structure-mapping in the comparison process. Ameri-
can Journal of Psychology 113 (4): 501538.
McLendon, H.J. 1955. Uses of similarity of structure in contemporary philosophy. Mind 64
(253): 7995.
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 Re-
view 100 (2): 254278.
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.), 321337. Cambridge, MA: MIT Press.
Quine, W.V. 1969. Natural kinds. In his Ontological Relativity and Other Essays, 114138. New
York: Columbia University Press.
Recanati, F. 2004. Literal Meaning. Cambridge, UK: Cambridge University Press.
54 Varol Akman
Rosch, E. 1978. Principles of categorization. In Cognition and Categorization. E. Rosch and

B.B. Lloyd (eds.), 2748. Hillsdale, NJ: Lawrence Erlbaum.
Shepard, R.N. 1974. Representation and structure in similarity data: Problems and prospects.
Psychometrika 39: 373421.
Sloman, S.A. 1999. Rational versus arational models of thought. In The Nature of Cognition, A
Bradford Book, R.J. Sternberg (ed.), 557585. Cambridge, MA: MIT Press.
Taber, K.S. 2001. When the analogy breaks down: Modelling the atom on the solar system.
Physics Education 36 (3): 222226.
Tversky, A. 1977. Features of similarity. Psychological Review 84: 327352.
Tversky, A. and Gati, I. 1978. Studies of similarity. In Cognition and Categorization, E. Rosch
and B.B. Lloyd (eds.), 7998. Hillsdale, NJ: Lawrence Erlbaum.
Weitzenfeld, J.S. 1984. Valid reasoning by analogy. Philosophy of Science 51 (1): 137149.
Wiggins, D. 2001. Sameness and Substance Renewed. Cambridge, UK: Cambridge University
Press.
Appropriateness and felicity conditions
A theoretical issue
Etsuko Oishi
Fuji Womens 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 speakers act of uttering and the hearers uptake. Appro-
priateness and felicity conditions concern how such a linguistic artefact is cre-
ated, 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 Austins (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 speakers act of uttering and the hearers uptake. We assert that
appropriateness and felicity conditions are concepts which concern how the lin-
guistic 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/in-
appropriateness. 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 Austins
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 utter-
ances 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 ana-

lysts, 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: 2526) in fact rejects this definition of pragmatics for the fol-
lowing reasons: (1) such a definition makes the field of pragmatics identical with
sociolinguistics, (2) it requires a fundamental idealization of a culturally homog-
enous 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 gener-
ally 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 appropri-
ateness is not our present concern, this definition makes it clear that appropriate-
ness 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 relation-
ship between sentences and contexts.
Mey (2001) also discusses appropriateness, in particular, pragmatic appropri-
ateness. He analyses the Japanese data in which the customer utters sumimasen
Im sorry to the clerk for an unpaid service, and says, we see how the Japa-
nese expression sumimasen Im 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 57
What we are dealing with here is not a matter of what expressions mean, ab-
stractly 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 differ-
ent 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 elu-
cidated by the concept of appropriateness: the use of a sentence/expression is as-
sociated with a context, according to which a particular token of the sentence/ex-
pression 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 con-
texts. 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 commu-
nicative 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 well-
formedness, appropriateness covers both the micro domains of a sentence, ut-
terance or communicative action and macro domains of dialogue and discourse.
(Fetzer 2004: 8586)
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 spa-
tio-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).
5 Etsuko Oishi
The notion of a historic situation would seem applicable not only to straightfor-
ward 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 lin-
guistic, socio-cultural, and social contexts? Is it equivalent to explaining condi-
tions 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/ex-
pression, 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: 2526): 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 con-
text 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 lan-
guages, in other words by the conventions whose acquisition constitutes learning
the language. To give a theory of meaning, a description of how a language func-

tions, 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: 209210)
We try to explicate the context of use through a description of how uttering a sen-
tence/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 com-
munication within which this assumption makes sense.
We start with a somewhat bold hypothesis of linguistic communication: com-
munication 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: 9495). Xs utterance You are a damn fool is rather a linguistic event, which
might be summarized later such as X insulted or criticized Y. In Austins terms, Xs
utterance You are a damn fool is an illocutionary act, which is described as an act
of insulting or criticizing. Whether Xs 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 physi-
cal 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 com-
municative 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
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 pro-
cess 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 Ill 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 supervisors 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 foresee-
able future: although finishing my Ph.D. would entirely depend on my own effort,
which was suggested by his words Ill 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 re-
ality 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 speakers communica-
tive 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 hearers uptake. If we are right about this communica-
tion model, then the ways in which the utterance becomes appropriate or inap-
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 utter-
ance 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/ex-
pression 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 Austins (1962) concept of felicity conditions. Austins felicity condi-
tions 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. Austins felicity conditions
The following are Austins 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.
(.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
(.2) must actually so conduct themselves subsequently.
(Austin 1962: 1415)
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 condi-
tions, 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 con-
stitutive rules: they are jointly constitutive of the various illocutionary forces1.
Within the theoretical framework we have proposed, and likewise the con-
cept of appropriateness/inappropriateness, the concept of felicitousness/infelici-
tousness concerns how a linguistic artefact as a certain illocutionary act is cre-
ated. 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 inter-
pretation, differences between circumstances in a historic situation and those as
an internal context are a source of infelicities.
We argue that Austins felicity conditions, (A), (B), and (), illustrate three dif-
ferent aspects of the internal context of performing an act. We describe these as-
pects of the context as conventionality, performativity, and personification, which
are specified by the felicity conditions, (A), (B), and (), respectively.
Violations of conditions (A.1) and (A.2) are described as misinvocations, in
which the purported act is disallowed (Austin 1962: 1718). That is, felicity condi-
tions (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 lin-
guistic 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 speakers utter-
ance, together with the hearers 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 sen-
tence 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: 238239) for further discussions.

2. Yasmin, 2004, directed by Kenny Glenaan.
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 par-
ticular persons and circumstances, specified by linguistic and socio-cultural con-
ventions, 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 re-
ligious 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 bap-
tizing. Austin (1962: 1718, 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
hearers 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 per-
formed; 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-
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, circumstanc-
es, 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: 1718). That
is, the felicity conditions (B) specify the aspect of the context where a certain act
is actually executed by the speakers performance and the hearers response. We
describe this as the performative aspect of the context.
The felicity condition in (B.1) describes the aspect in which the speaker actu-
ally 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 illocu-
tionary 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 utter-

ance the speaker makes as a slip of tongue, I inaugurate the new US empire in Ca-
racas, would not create an artefact of performing the act of inaugurating: the in-
auguration of the new US embassy does not occur in this performance. Similarly,
the utterance in the historic situation where there hasnt 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: 1718, 36) describes infelicities of this kind as flaws.
The felicity condition in (B.2) exposes another aspect of the context, in which
the speakers performance is completed by the hearers performance as a response.
Austin (1962) includes other procedures which complete the speakers perfor-
mance:
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)
We, however, focus on the hearers response as the procedure to complete the
speakers performance: the speakers performance of a certain illocutionary act is
acknowledged and guaranteed as the act by a certain response from the hearer.
For example, the speakers performance of an illocutionary act of ordering in (7)
might be completed by the hearers response as follows:
(9) Yes, sir.
Alternatively, the speakers performance might be completed by the hearers ac-

tion 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 dont 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: 1718, 3637) 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 conven-
tion of performing an illocutionary act, which is substantiated by the speakers
actual performance and the hearers actual response, becomes an artefact with a
form; the form is shaped by the speakers performance in using her/his voice as a
communicative move to the hearer, which is completed by the hearers reaction.
Because of this structure of communication, to recognize an artefact of a certain
illocutionary act is to partially recognize the speakers performance in a historic
situation, which is completed by the hearers response.
Austins felicity conditions (A) and (B) allude to two different but interrelated
ways in which illocutionary acts succeed or fail. According to the felicity condi-
tions (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 per-
forming illocutionary acts are or are not substantiated by the speakers accurate
performance and by the hearers response to complete the performance. There-
fore, felicitous cases under the conditions (A) and (B) are cases where the utter-
ance as an illocutionary act, which accompanies the specification of a conven-
tion, circumstances, and persons, occurs with the substantiation of them by the
speakers accurate performance and the hearers 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 sub-
stantiated by the speakers performance and the hearers response in the historic
66 Etsuko Oishi
situation. As a result, there exists an artefact which represents a linguistic con-

vention, circumstances and persons, and substantiates them with the speakers
performance and the hearers response in a historic situation. Such an artefact is
an amalgamation of the linguistic and socio-cultural conventions, the speakers
performance and the hearers 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 (.1) and (.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 condi-
tions 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 (.1) and (.2) does not
result in non-performance. Although it is a case of abuse, a purported act is per-
formed nonetheless. Austin says:
The first big distinction is between all the four rules A and B taken together, as
opposed to the two rules (hence the use of Roman as opposed to Greek letters).
If we offend against any of the former rules (As or Bs) 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 con-
ducting 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 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: 1516)
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 speakers accurate performance and the hearers 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 speakers performance and the hearers 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 ()
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.
The felicity conditions () specify the aspect of the context where the pro-
fessed act is sincere: the speaker is responsible for the creation of a linguistic ar-
tefact, and indicates a certain thought/feeling/intention, and, when applicable,
commits herself/himself to conducting a certain action subsequently.
The felicity condition (.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 per-
forming. 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 in-
ternalize 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 en-
dangers 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 speakers welcome in such a case is hollow.
Austin (1962: 18) describes this kind of infelicities as insincerities.
The felicity condition (.1) suggests the process by which the speaker be-
comes 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 ex-
presses her/his intention by uttering a sentence. In standard speech act theory
this is assumed to be the process of performing an illocutionary act. Sbis (2001)
summarizes this as follows:
Since Searle (1969: 4649), 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 speakers communicative intention determines what illocu-
tionary act he or she should be taken to perform and therefore what illocutionary
force his or her utterance may have. (Sbis 2001: 1795)
6 Etsuko Oishi
The idea of the standard speech act theory that the speaker expresses her/his in-
tention by uttering a sentence presupposes the a priori concept of the speakers
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 illocution-
ary force, Sbis (2001) says:
The speakers 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 rec-
ognition of the speakers 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 & Har-
nish 1979; Sperber & Wilson 1986; Harnish 1994; and in indirect speech acts in
Searle 1976). (Sbis 2001: 1795)3
This, however, does not seem to be an accurate description of performing an il-

locutionary 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 some-
thing. 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
hearers 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 expect-
ed. 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. Sbis (2001) claims that some aspects of standard speech act theory are incompatible with
the features of linguistic action emerging from research on mitigation/reinforcement.
The explanation which we give by reanalyzing the felicity condition (.1)

seems to be more reasonable: while specifying the present communicative move
as a certain illocutionary act, the speaker identifies the feeling/thought/inten-
tion 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 reason-
ably 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 de-
scribes a conversational participant who observes the Cooperative Principle and
the conversational maxims, in particular, Maxim of Quality, for our present in-
terest. 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 (.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 () show the personification aspect of context. The ut-
terance as an illocutionary act is personified by the speakers internalization of
a certain feeling/thought/intention as her/his own, and also by the speakers in-
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 subse-
quently. Within the proposed model of communication, this might be explained as
follows: an utterance as an illocutionary act, which is personified by the speakers
internalization of a certain feeling/thought/intention and a certain commitment,
becomes an artefact with a performers 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 situ-
ation, 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 Austins felic-
ity 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 de-
scribing the felicity conditions, we have presented three aspects of context in or-
der for an illocutionary act to be performed: we call them conventionality, perfor-
mativity, 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 (conventional-
ity); the speakers actual, accurate performance and the hearers 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 Austins felicity condi-
tions. 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 per-
forming 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-
ers actual performance and the hearers 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 cir-
cumstances of advising.
(12) I advise you not to enter the room.
(13) I wont.
This is the process by which the convention, and the circumstances and persons
specified by the convention are substantiated by the speakers actual performance
and the hearers actual response. It is the process of situating the present perfor-
mance, 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 speakers performance
and the hearers response evoke and substantiate a linguistic/socio-cultural con-
vention, and circumstances and persons specified by the convention. A linguistic/
socio-cultural convention of performing an illocutionary act does not exist in-
dependently 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 speakers performance and the hearers uptake, there-
fore, 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 com-
municative 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 identi-
fied 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 speakers 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 linguis-
tic communication. When an utterance is uttered as an illocutionary act, three
different factors are combined as elements of performing the act: first, the lin-
guistic 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
72 Etsuko Oishi
wider range of language users, such as the users of the language or the dialect; sec-
ond, a performance as a communicative move by the present speaker in a historic
situation, which is completed by the present hearers uptake of it and/or her/his
response to it; and third, the present speakers indication of her/his own feeling/
thought/intention and commitment. These rather different factors are combined
by an identification of the speakers performance with the convention, and with
the speakers 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 ones intention, but the major-
ity 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. Ones feeling/thought/
intention and commitment are vague without being specified. When one says,
Ill 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 sen-
tence 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 par-
ticular 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 personi-
fied. 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.
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 spe-
cific commitment personified implicitly?
Performative utterances are characterized by the use of the first person sin-
gular 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: 5758) 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 speakers performance is indi-
cated not under the speakers own authority but under the authority of something
to which the speaker belongs, such as an organization like a company or a govern-
ment (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 speakers 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 Austins partial sentences; We promise, We
consent , and You are hereby authorized to pay respectively. Example (15b) is Austins
original sentence.
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 warn-
ing. It seems quite interesting that there is a difference of strength of illocution-
ary 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 some-
one 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 indi-
rectly 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, there-
fore, 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, in-
dicates a performative aspect of performing an illocutionary act, i.e., the speakers
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 Austins original sentence (Austin 1962: 57).

expresses the speakers order to the hearer, which is interpreted as the speakers
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 speakers performance to the hearer.
The deontic mood of the sentence in (17b) expresses that the speaker permits
the hearers act of going, which implicitly indicates the speakers 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 speakers choice of the action, i.e., turning to the right, in
the hearers circumstances: this implies that the speaker advises the hearer to per-
form 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 cre-
ated in the historic situation, the speakers 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 speakers identification
of the past move, You were off-side, indicates implicitly the performance of speci-
fying 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 speakers 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 ut-
tering truncated sentences as in the following:
(19) a. Done,
b. Guilty,
c. Out.6
6. (17a) to (19c) are all Austins original examples.

76 Etsuko Oishi
These examples show that, when the frame of possible performances of the speak-
er is provided by circumstances in the historic situation or the preceding utter-
ance, 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 re-
sponses:7 the act of betting might be completed by the hearers response of ac-
cepting the bet, or the hearer might reject it. Therefore, the one-word utterance,
Done, in (19a) in such circumstances incompletely but successfully indicates the
speakers 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 speakers performance of finding a person guilty and
calling someone out, respectively.
The illocutionary acts performed by non-performative utterances show in-
teresting 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 per-
forming an illocutionary act is created linguistically and extra-linguistically in
various ways. Although we have demonstrated, in limited cases, the aspect of per-
sonification 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 be-
tween 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 Austins concept of felicity conditions, we have described three aspects of
the context for performing an illocutionary act, i.e., conventionality, performa-
tivity, and personification. An illocutionary act is performed when a linguistic/
socio-linguistic convention, and circumstances and persons specified by the con-
vention evoked (by the utterance) are substantiated by the speakers actual per-
formance and the hearers 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.
analyzed performative utterances as the cases in which the linguistic convention,

the speakers performance and the hearers response to complete it, in conjunc-
tion with a particular feeling/thought/intention, and a particular commitment,
i.e. three aspects of the context, are identified and specified explicitly by the ut-
terance. We have also described how illocutionary acts can be performed im-
plicitly 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, 117133.
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: Long-
man.
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.), 407459. London: Routledge.
Levinson, S.C. 1983. Pragmatics. Cambridge: Cambridge University Press.
Mey, J.L. 1993. Pragmatics: An introduction. Second edition 2001. Oxford: Blackwell.
Sbis, M. 2001. Illocutionary force and degrees of strength in language use. Journal of Prag-
matics 33: 17911814.
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: 123.
Sperber, D. and Wilson, D. 1986. Relevance: Communication and cognition. Second edition
1995. Oxford: Blackwell.
Wittgenstein, L. 1953. Philosophical Investigations. Oxford: Blackwell.
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 speakers processing
system, it is chiefly a matter of activating skills arising from background knowl-
edge. 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 other-
wise) may be deemed rightly or wrongly appropriate by its agent? Assum-
ing one such factor to be a store of knowledge1 (henceforth background knowl-
edge), 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 in-
teractions 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 agree-
ment 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: 4357). 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 envis-
aged that involves negotiations (Fetzer 2000 and 2002). An example of nego-
tiation (over the norm) is provided by (2), an extended version of (1), at the
end of which R accepts Ss initial assumption of a shared norm:
(2) S: Why did you tell John?
R: Why not?
S: I said it must stay within these walls.
R. Oh sorry, I forgot.
Appropriateness 1
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 tell-
ing them off for failing to do so).
5. Inasmuch as continued use of social conventions is an expression of commit-
ment to the existence of those conventions, any flouting constitutes a threat
to social cohesion and puts perpetrators at risk of varying degrees of mar-
ginalization. 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 fami-
lies.
6. Adaptive behavior is behavior that is advantageous to the survival and well-
being 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 be-
havior 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 assump-

tions, 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 linguis-
tic knowledge. By contrast, Deacon explains this feat in terms of co-evolution
of language and the brain. In his view (Deacon 1997: 105115), 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 lan-
guage 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).
2 Thanh Nyan
cortical region and the pre-eminence of its projections in competition for

synapses throughout the brain (Deacon 1997: 220).This prefrontal domi-
nance 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 com-
ponents 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 (men-
tioned above), which helps alleviate the processing load due to the increasing
demands of communication, would be the result of the co-option of diver-
gent 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 capac-
ity to be conscious of being conscious (Edelman 1989: 186192) 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 higher-
order 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 representa-
tion of its plans, but to communicate its goals to others and, through argu-
mentative discourse, involve them in its planning (Nyan 2007). The proposed
parallel between argumentative discourse and higher-order consciousness is
based on Edelmans 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 evol-
ution of symbolic structures.
Appropriateness 3
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 higher-
order consciousness to be even freer from the influence of current per-
ceptions. 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 sur-
vival 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 cer-
tain 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-sec-
ond 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 stand-
point 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 con-
sciously involved.
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 per-
ceptual 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 encod-
ed 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 cat-
egorical5 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 adap-
tive. This involves looking into category formation, in particular with regard to
the criteria that influence it, category alteration, and the corresponding underly-
ing 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 appropri-
ate 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 5
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, includ-

ing 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: 348349). 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 pro-
cess 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 knowl-

edge with a categorical structure. However in order to be adaptive, this back-
ground knowledge must:
a. present a categorical structure that is maximally effective at boosting efficien-

cy, 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 Edelmans Theory of Neuronal Group Selection (henceforth
TNGS)6 (Edelman 1989: 57, 94146).
6. In particular with respect to processes underlying perceptual categorization and concept

formation. Inasmuch as Damasios model of brain function is in many respects similar to Edel-
mans, especially with regard to a feedback process which Edelman calls reentry, I will also re-
sort 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 Damasios apparatus
of decision making, which has no explicit equivalent in Edelman, the differences presented
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 sig-
nificant 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: 4041). In viewing the brain as a selec-
tive system, the TNGS constitutes a rejection of information models of interac-
tions 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 physi-
cal output. They do not regard the level of variability of the world and of nervous
systems as significant. To explain the brains ability for perceptual categorization,
generalization and memory, the TNGS postulates three basic mechanisms: de-
velopmental selection, experiential selection and re-entrant mapping (Edelman
1989: 4572).
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 deci-
sion making and rule following must give way to an ability to activate correspond-
ing patterns of behavior on the basis of given situations.
Searles example of the baseball player (1995: 141147) provides an illustra-
tion 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 re-
sponsive 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): Damasios 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 Edelmans re-
entry uses the process of synthesis of signals as its main means of operation, while Damasios
framework uses a correlative operation instead.
Appropriateness 7
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 with-
out 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 Back-
ground (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-representational8 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 [neurophysiologi-
cal] structures generally (Searle 1995: 129130).
Searle sees the Background as made up of layers, ranging from Deep Back-
ground, 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 Presi-
dency 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 capaci-
ties, 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: 142143).
Thanh Nyan
The notion of background causation was introduced to explain how back-

ground abilities enable us to behave in a way that conforms to institutions (Searle
1995: 139142). Of two general models of causation, one intentionalistic and ra-
tionalistic (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.
Searles 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-repre-
sentational background assumptions is also compatible with a widely accepted
view of skills, according to which they arise from implicit memories. Such memo-
ries 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: 375384).
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 ra-
tional 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 dra-
matically.
This problem requires a theory that can provide a mechanism whereby this
non-rationalistic type of decision making can be implemented. This is where
Damasios 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 knowledge11),
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 neu-
rophysiology 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: 8384).
Appropriateness 9
between this factual knowledge and Searles Background: in Damasios 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 catego-
ries 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 non-
rationalistic and connectionist model such as Searles 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, accord-
ing to Damasio (1994: 173198) through the intervention of somatic markers. So-
matic 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 alterna-
tives requiring consideration. Somatic markers involve two alternative mecha-
nisms, 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 ad-
vantage would be taken into consideration at this stage.
To summarize: the type of background knowledge likely to sub-serve adap-
tive behavior is one structured in a way that facilitates a decision-making process
with as high a level of automatization as possible. On Damasios 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.
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 Damasios 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 dif-
fers from one domain (or environment) to another is the form adaptive behav-
ior 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: 134141). 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 ones 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: 134141).
In more external environments, complex organisms require further stocks of
dispositions, ranging from Searles Deep Background abilities, which enable them
to interact safely with their physical environment to skills that shape their daily
Appropriateness 91
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 struc-
tures 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 prin-
ciples 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 effi-
ciency. 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 Searles
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 hap-
pens 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 enti-
ties, 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 out-
comes 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;
92 Thanh Nyan
an automatic link to categories of responses;

the means to choose among these responses, in the form of somatic mark-
ers.
To elaborate further on this last point: the mere contemplation of any given re-
sponse 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-enact-
ed; the effect of which is to encourage or inhibit the action under consideration
(Damasio 1994: 187188).
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 knowl-
edge 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 en-
sembles of neurons (Damasio 1994: 102105). 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: 5657 and 112), which tends to fo-
cus 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 cat-
egorization 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 in-
clude salient features inherent in that predator, none of which may have any rel-
evance 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: 9899), innate values, determine what features
the animal sees as salient.12 Thus, non-accidental sensory attributes of preda-
12. These features may or may not coincide with those that are inherently salient in the ob-
ject.
Appropriateness 93
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 fea-
tures then go on to provide criteria that define what situations (including escape
means) are significant with regard to survival. Value-dominated perception Edel-
man aptly defines as the adaptive discrimination of an object or event from back-
ground 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 perceptu-
al category by conceptual means13 (Edelman 1989: 93102). 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 sys-
tems 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 exter-
nal 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 con-
sidered 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 rela-
tions (Edelman 1989: 9596).
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).
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, fail-
ure 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 catego-
ry 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 learn-
ing 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 [] en-
tities as equivalent (Edelman 1989: 49). This capacity, in turn, is based on a prop-
erty of neural groups known as degeneracy, whereby more than one combina-
tion 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 behav-
ior 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 value-
free perceptual system that categorizes the external world, the value-dominated
memory (corresponding to background knowledge), which holds complex cat-
egories, 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 ad-
vantageous in a given context, and which is not.
Appropriateness 95
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 catego-
ries with a high sensory content, based on features provided by the perceptual
system, and categories with a low sensory content (past value-dominated catego-
ries), which involve less immediate reference to objects and events. The latter,
also known as conceptual categories or concepts in the TNGS, reflect gener-
alizations arising from various mixtures of relations concerning the real world,
memories and past behavior (Edelman 1989: 143).
To account for concept formation Edelman (1989: 143145) 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 ac-
tivity in different kinds of global15 mappings must be recombined, compared and
categorized (e.g., to connect movement categories to spatial references, or distin-
guish global mappings corresponding to objects from those relating to actions),
back reentrant connections are required from the frontal cortex to other corti-
cal 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: 173181):
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).
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 re-
combination without further reference to their pre-linguistic conceptual ba-
sis (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 back-
ground 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 func-

tion 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 consti-
tutes 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: 102105) 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: 147148).
Appropriateness 97
case of complex sensory information, to topographically mapped representa-

tions in the early cortices.
When activated, dispositional representations can do one of three things:
a. fire other dispositional representations with which they have strong connec-
tions;
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 knowl-
edge 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 pro-
cedural memory). For a discussion of why this model of knowledge representa-
tion which implies a relative compartimentalization of the brain is to be pre-
ferred 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 connec-
tion 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 con-
strued?
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 topo-
graphically 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.
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, per-
taining to the same circuit. If the category of initial situations belongs, for ex-
ample, 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 vari-
ous 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 log-
ging 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 repre-
sentation 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 topograph-
ically mapped representation corresponding to a category of response options.
Alternatively, the first set of dispositional representations would cause a set of
Appropriateness 99
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 mark-
ers 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 cor-
tices). This, in turn leads to dispositional representations in higher-order cortices
to be fired. Among the various memories in which these dispositional representa-
tions are held, one must now include the Brocas and Wernickes 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 catego-
ries 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 mis-
leading, 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
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 in-
stances).
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, 176182).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 continua-
tion.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 develop-
ment 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 101
(4) This editor is meticulous.

(5) This editor is obsessive.
(Both (4) and (5) are judgments passed on the same editors attention to
details.)
The attitudinal component positive in meticulous, and negative in obses-

sive constitutes a strong constraint on what will or will not work as an appro-
priate continuation. (Strong constraints, unlike weak constraints, as in (3), cannot
be easily overruled by contextual assumptions). Thus meticulous requires con-
tinuations 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 dtre arroses, 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:
a. 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.
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 recatego-
rization 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).

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 con-
straints 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: 198199) 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 pro-
cess. In other words, the solution appears to lie in a decision-making process.
Argumentative discourse presents an obvious connection with decision mak-
ing. 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 appropri-
ate 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 Rs
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 103
4.3 Categorization of the situation of reference
Argumentation, like decision making, starts off with a categorization of the situa-
tion 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 pre-
senting 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. Searles 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-exist-
ing category of situations C1.
Now, granted that P has a categorical aspect, why should this aspect be privi-
leged 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 repre-
sented 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. Ss 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. Further-
more, 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 as-
serting 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.
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 well-
off 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 cor-
responding 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-lin-
guistic 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 cat-
egory 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 105
4.5 Attitudinal components and somatic markers
Attitudinal components present in lexical items such as meticulous and obses-

sive 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 conse-
quences of the same category of actions. Two people who have suffered the
consequences of an outstanding gas bill, and have learned from the experi-
ence, 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 con-
sequences 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
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 re-
sponse to the same situations. However, from sharing the same biological make-
up and the same environment they can certainly develop the same types of somat-
ic 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 pen-
alty 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 drivers 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 repre-
sentations of situations they associate. This, in turn, gives rise to linguistic con-
straints, 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 lan-
guage is through continued collective linkages of representations of initial situa-
tions and appropriate or inappropriate actions (see Searle 1995: 7071). 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 de-
scribe the core meaning of mais in terms of a schematized situation (or an im-
plicit argumentative sequence), as construed by Argumentation Theory (Cadiot
et al. 1985: 106124; 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 Ss 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 decision-
making 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 in-
tervenes at the stage where the SR is categorized. Consider again (6):
(6) Les plantes ont besoin dtre arroses (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 likeli-
hood 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 Rs decision-making process along is by identifying, among compet-

ing complex categories, the one that is advantageous to Ss argumentative intent.
The appropriate contribution on Rs part is one underpinned by CC2.
There remains to say a few words about how the above apparatus would ac-
count 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 vi-
ability 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: 11401146), 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 vary-
ing degrees of linguistic constraints. The overall picture that emerges is that ap-
propriateness 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-lin-
guistic conceptual (or n.l.c.) memory. This n.l.c. memory appears to have influ-
enced recategorization at the level of the linguistic conceptual (or l.c.) memories.
The constraint associated with well-off in (3), which weakly constrains prefer-
ential 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 rep-
resentations, 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. Obvi-
ously, 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 appropri-
ate 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 con-

strued from an adaptive perspective. Following this, I propose to revisit the no-
tion of context in relation to which appropriateness is usually defined.
An adaptive perspective crucially entails an internal viewpoint, that of indi-
viduals 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 trig-
gered. 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 rela-
tionship 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 un-
der 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 rel-
evance 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 decision-
making 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
Rs working memory, the expectation is that the logistics involved in context con-
struction 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 rep-
resentation, and this has the effect of triggering the corresponding complex cat-
egory in Rs 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 func-
tional view of attention (Allport 1987: 405415; and Neumann 1987: 373382)
according to which attention is adaptive in that it enables the animal to select ac-
tions appropriate to a predetermined goal. In short, linguistic representation is a
way of directing Rs attention to categories that are advantageous to Ss 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 111
ents a more complex level of interference with Rs decision-making process. Basi-

cally, this is carried out by means of two sets of constraints:
a. one concerning the relation in which Rs and Ss 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 Rs and Ss views in the schematized situation and the
conclusion.
In terms of the construction of Rs 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, argumen-
tative markers govern the type of complex categories being accessed in terms of
their relation, the way in which these competing complex categories are evalu-
ated, 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 com-
ments.
References
Allport, D.A. 1987. Selection for action: Some behavioral and neurophysiological consider-
ations of attention and action. In Perspectives on Perception and Action, H. Heuer and A.F.
Sanders (eds.), 395419. Hillsdale, NJ: Lawrence Erlbaum.
Anscombre, J.-C. (ed.). 1995. La Thorie des Topo. Paris : Kim.
Cadiot, A., Ducrot, O., Nguyen, T. and Vicher, A. 1985. Sous un mot une controverse: Les
emplois pragmatiques de toujours. Modles Linguistiques 7(2): 106124.
Damasio, A. 1994. DescartesError. 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 dtre arroses, mais
) are unambiguous as to what type of conclusion they require.
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.), 6174. Cam-
bridge MA: The MIT Pres.
Deacon, W.T. 1997. The Symbolic Species. New York: Norton.
Ducrot, O. 1988. Topo et formes topiques. Bulletin dEtude de Linguistique Franaise 22:1
14.
Ducrot, O. 1995. Topo et formes topiques. In La Thorie des Topo, J.-C. Anscombre (ed.),
8599. Paris: Kim.
Edelman, D.G. 1989. The Remembered Present. New York: Basic Books.
Fetzer, A. 2000. Negotiating validity claims in political interviews. Text 20 (4): 41559.
Fetzer, A. 2002. Micro situations: Natural language communication and context. Foundations
of Science 7: 255291.
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.), 159182. 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. Psy-
chological Review 93 (2): 136153.
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 Per-
ception and Action, H. Heuer and A.F. Sanders (eds.), 361391. Hillsdale, NJ: Lawrence
Erlbaum.
Nyan, T. 2006. Discourse markers and the decision-making process. Sprache und Datenverar-
beitung 30 (1): 109118.
Nyan, T. 2007. Common ground, categorization and decision making. In Lexical Markers of
Common Grounds, A. Fetzer and K. Fischer (eds.), 195210. Amsterdam: Elsevier.
Rosch, E. 1978. Principles of categorization. In Cognition and Categorization, E. Rosch and B.
Lloyd (eds.), 2848. 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 Habermass theory of communicative action, Gices logic and conversation
and Sbiss 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 inap-
propriate, 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 co-
participants communicative intentions and inferencing processes, a socioprag-
matic 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; Sa-
rangi & Slembrouck 1996).
Implicit in all the approaches is the assumption that natural-language com-
munication 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 commu-
nicative contributions to context and base their inferencing on context. In order
to produce and interpret communicative contributions in a rule-governed man-
ner, however, they need to be conceived of as rational agents who perform com-
municative acts intentionally and strategically. While the more traditional no-
tion 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 Levinsons interpersonal-oriented framework, rationality covers a
model persons information wants and face wants (Brown & Levinson 1987). In
Brandoms philosophy-of-language anchored work making it explicit (Brandom
1994), rationality is expressed in discourse and manifests itself in the communi-
cators acts which they perform in accordance with their conception of rules.
Habermass theory of communicative action (Habermas 1987) is based on a
sociocultural conception of rationality. In that frame of reference, rationality is
reflected in the coparticipants performance and interpretation of communicative
action in accordance with the sociocultural values and norms of a speech com-
munity. 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 con-
cept 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 Grices
logic and conversation (Grice 1975) and Habermass conception of communica-
tion as postulation and ratification of validity claims through an acceptance, re-
jection or neutral stance (Habermas 1987). However, it goes beyond Habermass
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 repre-
sented by scalar concepts, and can therefore never refer to identical individual representations.
Rather, coparticipants presuppose a commonly shared representation with fuzzy boundaries.
Indexing appropriateness 117
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 com-
municative 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 function-
ally 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 inappropri-
ate status, who refers to the inappropriateness and how the inappropriateness is
realized linguistically.
The following section, dialogue and dialogues, examines the theoretical con-
struct 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 cog-

nitive (or internal) and social2 (or external) domains of dialogue and on their im-
pact on language use. While the former tend to employ polyphonous (e.g., Bak-
thin 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 frame-
works (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 con-
straints and requirements of an external face-to-face interaction in sociocultural
contexts. In the philosophy-of-language paradigm, dialogue has been conceptual-
ized 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.
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 col-
laboration and joint production (Schegloff 1991), in the philosophy-of-language
premises of cooperation and collective intentionality (Searle 1991) and in the ac-
commodation of I-thou perspectives (Brandom 1994).
From a parts-whole perspective, dialogue has been described as a macro se-
quence 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 struc-
ture (Kamp & Reyle 1993), sentence and proposition (de Beaugrande & Dressler
1981), proposition and speech act (Sbis 2002a; Searle 1991) or utterance and
turn (Schegloff 1991). Accordingly, it is defined with regard to frame-of-investi-
gation specific constraints and requirements, and with regard to frame-of-inves-
tigation-specific purposes. Thus, a micro sequence expresses semantic content in
discourse representation theory, communicative function in conversation analy-
sis and critical discourse analysis, or both semantic content and communicative
function in speech act theory, dynamic interpretation theory and sociopragmat-
ics. 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 propo-
sitions, and they have been investigated with regard to cohesion and coherence
(Halliday & Hasan 2000). In conversation analysis, dialogue is composed of turn-
constructional units, turns and topical sections (Schegloff 1996), and in socio-
pragmatics, dialogue is composed of utterances which are employed in an inten-
tional 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 speech-
act-theoretic methodology in dialogue analysis (Sbis 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 consid-
ers communication as both process and product.
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 con-
straints (or appropriateness conditions) of genre, for instance lecture, interview
or narrative. Looked upon from a top-down perspective, a macro sequence sub-
classifies 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 lectures overall theme. It is at that stage, where the sequential
organization of the meso themes may vary. Depending on sociocultural prefer-
ences, 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 commu-
nication 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, din-
ing 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 politi-
cal 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 disagree-
ments, 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 dialogues dual status as process and product, bottom-up
and top-down perspectives are required. The former can account for the construc-
3. This holds for traditional texts without any footnote or endnotes. The more recent phe-
nomenon of hypertext exploits the constraint of strict linearity and allows for multiple text
construction.
120 Anita Fetzer
tion of micro sequences and thus for the construction of parts, such as a commu-
nicative 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 genre-
specific turn-taking system of an interview or the genre-specific discourse topics
of a political interview, job interview or medical interview. Of course, both per-
spectives 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, sociocul-
tural 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 com-
prises 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 sur-
roundings are of an extra-linguistic nature, that is non-cognitive and non-linguis-
tic material, they are called local and global social context. The former comprises
the speech situation including time, space and coparticipants, and the latter con-
sists 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 read-
ing, 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 differen-
tiation between a linear, tangible and divisible conception of time, where events
are scheduled one at a time and where this schedule takes precedence over inter-
personal 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,
culture provides us with a filter mechanism which allows us to interpret social

context in accordance with particular sociocultural-context constraints and re-
quirements.
Social, sociocultural, linguistic and cognitive context are looked upon as rela-
tional 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 construc-
tion 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. Ac-
cordingly, the acts of speaking and interpreting build contexts and at the same
time constrain the building of contexts. To employ Heritages 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 con-
textual constraints and requirements anchored to culture, institution and society
(cf. Gumperz & Levinson 1996). In a political interview, this means that an in-
terviewers 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 interviews contextual constraints and requirements and counts as an in-
stance of public and political information. For instance, the interviewers question
What did you discuss on Saturday? in the genre of a political interview does
122 Anita Fetzer
not count as a question about the interviewees 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 analysts 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, in-
terpretation 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 em-
beddedness 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 Grices cooperative principle (Grice 1975),
Habermass theory of communicative action (Habermas 1987) and Sbiss research
on speech act theory (Sbis 2002a, 2002b), which are introduced below.
2.3.1 Logic and conversation in dialogue

Grices 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 con-
versational 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 prin-
ciple, which states the following constraint: Make your conversational contribu-
tion 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 copar-
ticipants contributions are in accordance with both the cooperative principle and
the maxims of quantity (1. Make your contribution as informative as is required
(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 obscu-
rity 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 copar-
ticipants in context. The conversational implicature can be a generalized implica-
ture 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 be-
tween communicative contribution, coparticipants, and linguistic and social con-
texts. Because of the implicature, which is detachable, non-conventional, calcu-
lable and defeasible, it can also account for the dialogues 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 dialogues constitutive micro and macro components.
2.3.2 Validity claims in dialogue

An integrated approach to dialogue feeds on different, but not diverging frame-
works. In order to account for the dynamics of dialogue, the Gricean concept of
contribution is refined and dynamized through the Habermasean concept of a va-
lidity 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 hearers 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 claims embeddedness in context and for its connectedness with
context, nor for the sequential organization of dialogue and its parts-whole con-
124 Anita Fetzer
figuration. To accommodate context, sequentiality and the parts-whole configu-

ration, 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 ex-
amined in the following.
2.3.3 The dialogue act of a plus-minus-validity claim

Habermass conception of validity claim is anchored to the subjective, objective
and social worlds and its foundations of truthfulness, truth and normative right-
ness. In the dialogue act of a plus/minus-validity claim, the subjective-world con-
struct of truthfulness is seen as functionally equivalent to speech act theorys sin-
cerity 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 theorys 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 equiva-
lent 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-arch-
ing 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 tri-
partite configuration of a textual system based on the linguistic realization of a
validity claim in accordance with the Gricean maxims and implicature, an inter-
actional 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 pre-
suppositions, 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
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 de-
fined by a tripartite configuration: the former is composed of subjective, social
and objective worlds, and the latter is composed of the acts of production, inter-
pretation and ratification. But how do plus- and minus-validity claims manifest
themselves in dialogue?
Adapting Sbiss claim (2002b) that the felicity of a speech act obtains by de-
fault unless indicated otherwise to the dialogue act setting, the felicity of the dia-
logue 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 mi-
nus-validity claims. Only then, felicitous communication may proceed.
To summarize and conclude, in dialogue coparticipants negotiate the valid-
ity of their communicative contributions which comprises their truth, sincerity
and appropriateness. A contributions truth is captured by its truth conditions
anchored to the objective world, which is true in any context. A contributions
sincerity is captured by its sincerity condition anchored to the cognitive-context
concept of subjective world, and a contributions appropriateness is captured by
its appropriateness conditions anchored to the social-context concept of appro-
priateness, 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
Dijks 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 ap-
4. 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.
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 sug-
gest the following criterion of adequacy of pragmatical theories:
(CA) For every context of utterance C, every sentence S of a natural lan-
guage 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 institu-
tional 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 con-
struct 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 re-
gard to illocutionary goal and force but also with respect to the additional features
of coparticipant and their social, interactional and discursive roles, and the con-
straints 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 Linells (1998) distinction between monologue and dialogue, appropriate-
ness 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 in-
dividual 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).
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 communi-
cative act which is seen as produced in accordance with a speech communitys
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 pre-
paratory conditions (ability) with can you do X is seen as appropriate in the
Anglo-American context, while a self-reference to the negated preparatory condi-
tions (not ability) with I can not do X does not necessarily count as an appro-
priate request in similar contexts (cf. also Akman, this volume). Thus, common-
sense 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 communitys 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 con-
cept 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 strate-
gies 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 clients
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 therapists assistant,
it counts as an appropriate request to perform the action X.
Naturally, the conceptualization of appropriate coparticipant role, appropri-
ate communicative action and appropriate genre are interdependent on other
sociocultural norms, conventions and strategies, which themselves are interde-
pendent 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, chair-
person 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-
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 communica-
tive contribution. Here, the linguistic surface of the contribution is de-composed
with respect to its references to sociocultural contexts which can be realized ex-
plicitly 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 youve just said is inappropriate, and implicitly
by saying well or is that so. The latter two variants signify that the prior con-
tribution 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 categoriz-
es 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 Goffmans notions of footing and participation
framework (Goffman 1974; Levinson 1988). The interactional system is based on
the Levinsons 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-sys-
tem constraint of face-needs and face-wants is quite different in Anglo-American
and Scandinavian contexts, for instance, and validity claims referring to that sys-
tem 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.
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 prin-
ciple 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 contributions textual system which
determines the contributions formulation regarding quantity, quality, relation
and manner, it refers to the contributions 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 adja-
cency 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 omit-
ted, and the degree of positive politeness outweighs the possible intrusion an-
chored 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 inferenc-
ing would have been triggered to calculate the conversational implicature. In that
medium-presupposition scenario, the process of inferencing would have used ad-
ditional 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
Grices 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 qual-
ity, quantity, relation and manner, for face-wants and footing, and for adjacency
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 conversa-
tional 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 Grices 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 theorys felicity conditions, dialogue analysis is
anchored to appropriateness conditions. These entail a speech acts felicity condi-
tions which are adapted to the contextual constraints and requirements of a dia-
logue acts 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: 5354). Appropriateness conditions do not only accommo-
date the constitutive systems of a particular type of speech act. They also accom-
modate the systems required for its linguistic realization, viz. the syntactic, mor-
phological, 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 contri-
butions connectedness on the local level of exchange, and on the connectedness
with the macro category of genre and the genres 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 an-
other 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
conditions, which obtain in a more general sociocultural context, and particu-

larized appropriateness conditions, which obtain in a particular context, such as
the context of a genre.5 Regarding its semantics, appropriateness represents a re-
lational construct, and the relational nature is inherited to its constitutive inter-
actional, interpersonal and textual systems, which are also relational in nature.
Because of the multi-layered configuration and the interactive systems, appropri-
ateness denotes a scalar concept with fuzzy boundaries. Thus, there are commu-
nicative 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, contri-
butions can also be inappropriate, if not fully inappropriate.
In the previous section, appropriateness conditions have been assigned a pre-
suppositional status. For this reason, they are generally not made explicit. As a
consequence of that, a reference to a contributions 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 iden-

tify (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.
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 com-
municative intention, and it allows for the analysis of the macro domain of com-
munication regarding the context-sensitive specification of its constraints and re-
quirements 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 ap-
propriateness 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 genres strict regulations of the turn-tak-
ing 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 leit-
motif (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 pre-
specified 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-tak-

ing.
For a political interview to be appropriate, adjacency is particularized to
the adjacency pair of question/answer, and the turn-taking system is par-
ticularized to interviewer selection. There is the additional particulariza-
tion that the interviewer asks questions, which the interviewee answers.
The interpersonal system is based on the premises of face-wants and partici-
pation 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, rela-
tion and manner, and the implicature.
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 appropriate-

ness are found in the context of a disruptive setting which is signified by cross-
talk, follow-up questions, hesitation markers and attenuation devices. In the data,
explicitly realized references are rare. In all of the 11 instances identified, the con-
struct 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 appropriate-
ness. All references collocate with sociopolitical-context references, such as elec-
torally erm appropriate, to make what adjustments may be appropriate, the ap-
propriate 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-valid-
ity 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 con-
6. I would like to thank Peter Bull (York, UK) and Gerda Lauerbach (Frankfurt, FRG) for
sharing their data with me.
134 Anita Fetzer
text, such as we just pick up the next piece and stick it on wherever its appropri-
ate, that was an appropriate step, is the system appropriate, that the language (..)
was inappropriate, to be strong an inappropriate language and thats completely
erm inappropriate. This has also been the case in the British data. What is pe-
culiar, 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 inap-
propriate question.
In the British political interviews, explicitly realized references to appropri-
ateness 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 Its very disillusioning to be told youre going to get millions of pounds
for cleaning, and then you cant 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, Ill give you the specif the specific language theyre 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 ortho-
graphic standards. IR denotes interviewer, and IE denotes interviewee.
(10) IE ... erm and thats what erm but thats really not what the issue is here Wolf.
The issue is, is the system appropriate and proper and constitutional...
(11) IE .. that somehow thats 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 impli-
cating 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 refer-
ence to appropriateness is also a constitutive part of an argumentative sequence.
The other instances of explicitly realized references to appropriateness co-oc-
cur with attenuation devices, such as hesitation markers in (1), (4) and (11), mo-
dality 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 worlds interactional, inter-

personal 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 contributions force or its communicative status.
While explicitly realized references to appropriateness do not generally invite fol-
low-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 inves-
tigated, references to the appropriateness conditions of the interactional system
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 de-
viation at that stage of the interview with the booster thats 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 metapragmat-
ic 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 thats. Then Ill answer that question because its in terms of another
question
IE But thats very critical
IR What youre saying is, what youre 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 nei-
ther 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.
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. Interper-
sonal-system references are used by IR and IE, and the conventionalized device
Im 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 us-
ing the first name as in extracts (10) and (25) or other means of creating solidarity
as the device you can correct me if Im 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 Im
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 face-
wants by intruding into others territory through the expression of a strong dis-
agreement 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 others 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 youve 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 ...Im 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 irre-
spective of the circumstances now in my view you have to strike a balance
here
(22) IE ..... the situation that I think youre talking about the situation that was
raised in your film ...
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 youve given at the
moment is as far as I can see is that erm ....
(25) IE Ill tell you why because erm Im sorry to bring politics into this George
but I dont actually trust the labour party on defence
(26) IR ... and all they have to say to themselves you can correct me if Im
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 condi-
tions of the interpersonal system, a reference to the positive face-wants with the
first name George and a reference to the negative face-wants Im sorry signify-
ing 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 refer-
ences to the appropriateness conditions of the interpersonal systems 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 appropriate-
ness 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 con-
sideration, and the contributions 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 systems appropriateness conditions. Pro-
totypical 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,
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 systems appropriateness conditions. Analogously to references to
the interactional and interpersonal systems, hedges co-occur with contextualiza-
tion devices, such as the discourse markers well, and and now, or the interper-
sonal 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 re-
inforces the IRs 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 IRs 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 conver-
sational implicature which requests the IE to be clear and precise in his answer
thereby implicating that the IEs 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 deter-
mination to pursue our anti inflation policy and to pursue it until infla-
tion 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 Were 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 ana-
phoric function by signifying that a previous contribution has not been in ac-
140 Anita Fetzer
cordance with the textual systems generalized appropriateness conditions, more-

fuzzy hedges have a cataphoric function and indicate that a constituent, validity
claim or contribution may infringe on the textual systems appropriateness condi-
tions. 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 systems appropriateness conditions and signify a possible in-
fringement. 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 pos-
sibly inappropriate figure, that is five point seven billion in (30) and fifty percent
in (31):
(29) IE Well, I- I sometimes think were 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 dont
reject them, that something like five point seven billion of the increase
in the tax take came from the increases in indirect taxes, thats 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, theres 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, well
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
of a political interview is at stake. This is generally due to multiple infringements

on the social worlds 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 claims 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 Im 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 IRs questions is at stake. The IE has already referred to the
appropriateness conditions of the social world and its constitutive textual, inter-
personal and interactional systems. The IRs explicitly realized reference to the
genre of a political interview is a direct response to the IEs challenge that the IR
has exploited its particularized values by not having adhered to the constraint
of neutrality. This is implicated through the IEs reference to the interpersonal
systems constraint of ratified coparticipants, namely the audience (viewers), the
reference to the IRs right to ask questions only which stands in a fierce contradic-
tion to getting the last word and by the reference to the perlocutionary effects of
the IRs communicative contributions, that is leave an impression. The IEs chal-
lenge has a very strong force and is functionally equivalent to accusing the IR of
being unprofessional. This is countered by the IRs 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 appropriate-
ness conditions of the social world and to its constitutive interactional, interper-
sonal 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 ac-
cepted 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.
142 Anita Fetzer
5. Conclusions
This contribution has investigated the theory and practice of dialogue and ap-
propriateness. Dialogue has been conceived of as a context-dependent endeavour
in which coparticipants negotiate the communicative status of their contribu-
tions through the postulation, ratification and acceptance or rejection of validity
claims.
The theoretical frame of reference is an integrated one: it is anchored to
Habermass 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 prod-
uct. 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 func-
tions explicitly and implicitly realized references to the social-world construct
of appropriateness fulfil and in what contexts they occur. Explicitly and implic-
itly 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 in-
teractional systems where they signify an infringement on the constraints thus
getting in a conversational implicature about a possible inappropriateness. Refer-
ences to the appropriateness conditions of micro validity claims tend to attenu-
ate 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 appro-
priate people or an appropriate moment, and they generally co-occur with explicit
and implicit negatives, such as not appropriate, inappropriate or doubt + appropri-
ate, 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
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 sub-
jectification markers if I may say, if Im not mistaken, I think, if I may say or if I
may express my opinion.
From a social-context perspective, explicitly and implicitly realized refer-
ences 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, Context99, Trento,
Italy, September 1999, Proceedings, P. Bouquet, P. Brezillon, L. Serafini (eds.), 114. Hei-
delberg: Springer.
Atkinson, J. M. and Heritage, J. 1984. Structures of Social Action: Studies in Conversation Analy-
sis. 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 Ethnomethod-
ology 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:
Bunt, H. 1999. Contextual representation for dialogue management. In Modeling and Using
Context. 2nd International and Interdisciplinary Conference, Context99, Trento, Italy, Sep-
tember 1999, Proceedings, P. Bouquet, P. Brezillon, L. Serafini (eds.), 7790. Heidelberg:
Springer.
Clayman, S. 1992. Footing and achievement of neutrality: the case of news interview dis-
course. In Talk at Work, P. Drew and J. Heritage (eds.), 163198. Cambridge University
Press: Cambridge.
Crystal, D. 1997. Dictionary of Linguistics and Phonetics. Oxford: Blackwell.
De Beaugrande, R. and Dressler, W. 1981. Einfhrung in die Textlinguistik. Tbingen: Niemey-
er.
Ducrot, O. 1984. Le dire et le dit. Minuit.
Duranti, A. and Goodwin, C. (eds.). 1992. Rethinking Context. Language as an Interactive Phe-
nomenon. 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.
144 Anita Fetzer
Fetzer, A. 1999. Challenging the unspoken: exploiting the ideology in and of political inter-
views. In Language and Ideology: Selected Papers from the 6th International Pragmatics
Conference. Vol. 1, J. Verschueren (ed.), 98113. Antwerp: International Pragmatics As-
sociation.
Fetzer, A. 2000. Negotiating validity claims in political interviews. Text 20(4): 146.
Fetzer, A. 2002. Communicative intentions in context. In Rethinking Sequentiality: Linguistics
meets Conversational Interaction, A. Fetzer and C. Meierkord (eds.), 3769. 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.), 283301. Hildesheim: Olms.
Fetzer, A. 2006. Minister, we will see how the public judges you. Media references in political
interviews. Journal of Pragmatics 38(2): 180195.
Fetzer, A. and Meierkord, C. (eds.). 2002. Rethinking Sequentiality: Linguistics meets Conversa-
tional 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.), 4158. New York: Academic Press.
Gumperz, J. and Levinson, S. (eds.). 1996. Rethinking Linguistic Relativity. Cambridge: Cam-
bridge University Press.
Habermas, J. 1987. Theorie des kommunikativen Handelns. Frankfurt/Main: Suhrkamp.
Hall, E. and Hall, M.R. 1989. Understanding Cultural Differences. Yarmouth/Maine: Intercul-
tural Press.
Halliday, M.A.K. and Hasan, R. 2000. Cohesion in English. London: Longman.
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.), 249271. Amsterdam: John Benjamins.
Kamp, H. and Reyle, U. 1993. From Discourse to Logic: Introduction to Modeltheoretic Seman-
tics 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.),
85103. London: Routledge.
Lakoff, G. 1975. Hedges: A study in meaning criteria and the logic of fuzzy concepts. In Con-
temporary Research in Philosophical Logic and Linguistic Semantics. Hockney, D. et al.
(eds.), 221271. Dodrecht: Foris.
Levinson, S. 1979. Activity types and language. Linguistics 17: 365399.
Levinson, S. 1983. Pragmatics. Cambridge: Cambridge University Press.
Levinson, S. 1988. Putting linguistics on a proper footing: explorations in Goffmans concepts
of participation. In Erving Goffman. Exploring the Interaction Order, P. Drew and A. Woot-
ton (eds.), 161227. Cambridge University Press: Cambridge.
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, E. Goody (ed.), 175188. 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: Long-
man.
Saville-Troike, M. 1989. The Ethnography of Communication. Blackwell, Oxford.
Sbis, M. 2002a. Cognition and narrativity in speech act sequences. In Rethinking Sequential-
ity: Linguistics meets Conversational Interaction, A. Fetzer and C. Meierkord (eds.), 7197.
Amsterdam: Benjamins.
Sbis, M. 2002b. Speech acts in context. Language and Communication 22(4): 421436.
Schegloff, E.A. 1991. To Searle on conversation: a note in return. In (On) Searle on Conversa-
tion, J.R. Searle et. al. (eds.), 133128. Amsterdam: Benjamins.
Schegloff, E.A. 1996. Turn organization: one intersection of grammar and interaction. In In-
teraction and Grammar, E. Ochs, E.A. Schegloff and S.A. Thompson (eds.), 52133. Cam-
bridge: Cambridge University Press.
Searle, J.R. 1991. Collective intentions and actions. In (On) Searle on Conversation, J.R. Searle
et al. (eds.), 401415. 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: Ox-
ford.
Vanderveken, D. and Kubo, S. (eds.). 2002. Essays in Speech Act Theory. Amsterdam: Benja-
mins.
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 intersubjec-
tive stance. Text 23 (2): 259284.
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 ap-
propriateness 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 interview-
ing 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 re-
spect to the questions 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 ex-
ample 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 inter-
view between Jeremy Paxman, influential political interviewer of the BBC, also
known by the nicknames The Grand Inquisitor or Mr Kebab for his relent-
less 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 its very late in the night, but wouldnt 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 appro-
priate 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 Pax-
man 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 BBCs Head
of Political Programmes, Sue Inglish, publicly defended Paxman against his crit-
ics. All in all, both within the interview itself and in its aftermath, the appropriate-
ness of the interviewers questions in relation to the sociocultural contexts of the
election campaign in general and the BBCs 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 equat-

ed 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 syn-
onym for acceptability (Gumperz 1982a; Fairclough 1992). Occasionally, appro-
priateness has also been equated with politeness (Meier 1995), or with linguistic
etiquette (Kasper 1997) in general. However, Fetzer (2004) has suggested that ap-
propriateness 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 system-
atic use of impoliteness suggest that there are contexts and genres where face-
threatening 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-
volved here. Jucker (1986) applies Brown and Levinsons face model to broadcast
news interviews, postulating that what is primarily at issue in news interviews
is the interviewees positive face (Jucker 1986: 71). Others, such as Bull, Elliott,
Palmer and Walker (1996) and Bull and Elliott (1998) have expanded Juckers
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, inter-
actional 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), interview-
ers are entitled to ask questions that are potentially threatening to their inter-
viewees positive or negative face. Not surprisingly, they do not always succeed in
obtaining answers to their questions. For instance, several studies have demon-
strated 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 appropriate-
ness 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 dif-
ferent interview genres see Bell and van Leeuwen 1994).
Generally, the appropriateness of questions is highly dependent on their so-
ciocultural 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-cultur-
al 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 ap-
propriateness in general and the appropriateness of questions in particular is the
seemingly trivial question: Says who? Who is to judge whether a question is ap-
propriate? 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 What referred to Channel

Physical Objects, sensations Sensory
Body language Gestures, facial expressions, posture, social distance, Sensory/social
paralinguistic features
Verbal Verbal utterances Linguistic
Cultural Norms, values, expectations, assumptions of the Cultural
culture at large
Group Norms, values, expectations, assumptions of the cur- Group-dynamic
rent group
Personal Each individuals private views, attitudes, prejudices, Psychological
predilections
Imagined Playacted social scenes, frames, or roles 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 (Givn 2005: 1). This is mirrored by the fact that the recent new edi-
tion 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 Robinsons detailed survey is the institutional dimen-
sion 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 (Schegl-
off 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: 1720).

(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 funda-
mental 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 in-
variant, 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 co-
constructed 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 anothers com-
municative intentions and goals (Drew and Heritage 1992: 8). Drew and Heri-
tage (1992) see strong similarities between Gumperz linguistic concept of con-
textualization cue and Goffmans 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 (Sche-
gloff 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 inter-
preted 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 consti-
tuted by the speech event as a whole. For instance, there are many restrictions
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 institu-
tion 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 ex-
pectations are met or not met gives rise to activity type specific inferences (Levin-
son 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 ordi-
nary 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 con-
nected 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 it-
self, 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 mono-
logical concept (Fetzer 2004). The dialogic nature of appropriateness is especially
apparent in question-answer exchanges, and even more so in mediated question-
answer exchanges like television interviews. Within conversational question-an-
swer exchanges, it is the individual addressees of questions who ratify questions
as appropriate. They do so by measuring their textual, interactional, and inter-
personal properties against the appropriateness conditions of the sociocultural
context of the speech event, and design their answers accordingly.
As King and Sereno put it, (s)ituational demands may restrict or expand the
range of utterances deemed appropriate (King and Sereno 1984: 266). Addition-
ally, 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 Grices 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 ba-
sis 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 consti-
tuted and that content is interpreted against the ground of relationship. If the fig-
ure, 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 cen-
tral 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 im-
plicature and, thereby, relational change (King and Sereno 1984: 272).2
In mass media interviews, interviewees judge the appropriateness of their inter-

viewers 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 ap-
propriateness 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 poten-
tially face-threatening presuppositions which are difficult to contest.
2. Bracketing corresponds to the original.

154 Annette Becker
First Parts Second Parts

Preferred Dispreferred
Question expected answer unexpected answer or non-answer
Request acceptance refusal
Offer/Invitation acceptance refusal
Assessment agreement disagreement
Blame denial admission
Figure 2. Preference in adjacency pairs (Levinson 1983: 272)
First Parts Second Parts

Preferred Dispreferred
Question direct answer indirect answer
rejection of one or more presup-
positions 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 inter-
viewers questions as appropriate or inappropriate, and how this is connected to
the sociocultural norms of the genre interview. For this purpose, a coding prac-
tice was developed on the basis of some general assumptions and their modifica-
tions. These assumptions were developed on the basis of conversation analysis,
pragmatics, and cross-cultural media analysis, starting from Levinsons (1983)
summary of the basic preference correlations in adjacency pairs (Figure 2).
Preferred second parts are usually delivered immediately and with little com-
municative 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 trig-
ger activity-type specific inferences, as Levinson (1979) suggests. One potential
inference triggered by dispreferred formats could, of course, also be the assump-
tion 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 inter-
views, the preference framework for question-response adjacency pairs can be
tentatively modified as follows: Direct answers are preferred responses, while in-
direct 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 trig-
gered accordingly, depending on the preferred, or dispreferred format of the re-
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 interviewers 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 an-
swers, and responses that contest one or more of the presuppositions of a ques-
tion, or even its pragmatic force (for further discussion see Bull 1994). Modifica-
tions of Harris definitions for those three categories were used to distinguish the
following types of IE treatment of IR questions:
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 Whos your favorite candidate for that leadership contest?
IE I think it depends whos going to come forward. Whos prepared to put
himself or herself into the ring.
a3 question treated as inappropriate by IE (= rejection of one or more presup-

positions 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 BBCs 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 percent-
age of questions were treated either ambivalently or as inappropriate. This differs
from the interviews with experts conducted in the course of the very same elec-
tion 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 [%]

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 partys 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 in-
terviewees 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 di-
rected 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 ex-
changes were counted as interviews, with the result that single questions directed
at co-present experts did not qualify.
15 Annette Becker
80
60
40
20
0
Politicians' identities (n = 250)
Cons. 67,6
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 %]

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 propor-
tion 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 top-
ics 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 di-
rected 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).
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 %]

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 in-
terviewees 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 with-
in 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 los-
ers. 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 spec-
ulate 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 harsh-
ly? One of Jeremy Paxmans 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 Galloways elec-
tion 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
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 BBCs 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 Dimblebys (DD) question regarding the result. Moreover,
he even upgrades Dimblebys presentation:
(13) DD So, there we are, we were still waiting for the first result to come in, its
ehm three minutes if Sunderland are [going] to keep their word, I havent
heard any news yet whether theyre 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: Its an asteroid hitting the planet
and destroying practically all life on earth, (.) it is an extraordinary result
theres been nothing like it, if this is whats going to happen, since the second
World War, theres 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 youve
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 were still talking exit poll talk, we havent 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 lets 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 in-
sights as to the appropriateness of interviewer questions in political interviews in
the media. This has been demonstrated on the basis of both theoretical discussion
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 ques-
tions, because these factors may also play a role for the face-threatening poten-
tial 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 condi-
tions for questions formulated by Searle (1969: 6669), 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.), 6576.
Tbingen: 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.), 124177.
Marrickville NSW: Southwood Press.
164 Annette Becker
Boden, D. and Zimmermann, D.H. (eds.) 1991. Talk and Social Structure: Studies in Ethno-
metholology 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:
Bull, P. 1994. On identifying questions, replies, and non-replies in political interviews. Journal
of Language and Social Psychology 13: 115131.
Bull, P. and Elliott, J. 1998. Level of threat: A means of assessing interviewer toughness and
neutrality. Journal of Language and Social Psychology 17: 220244.
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: 267284.
Clark, H.H. 2006. Context and common ground. In Encyclopedia of Language and Linguistics.
Second Edition, K. Brown (ed.), 105108. Amsterdam: Elsevier.
Clayman, S. and Heritage, J. 2002. The News Interview: Journalists and Public Figures on the Air.
Crystal, D. 2003. A Dictionary of Linguistics and Phonetics. Malden, MA: Blackwell.
Culpeper, J. 1996. Towards an anatomy of impoliteness. Journal of Pragmatics 25: 349367.
Culpeper, J. 2005. Impoliteness and entertainment in the television quiz show: The Weakest
Link. Journal of Politeness Research 1: 3572.
Culpeper, J., Bousfield, D. and Wichmann, A. 2003. Impoliteness revisited: With special refer-
ence to dynamic and prosodic aspects. Journal of Pragmatics 35: 15451580.
Drew, P. and Heritage, J. (eds.). 1992. Talk at Work: Social Interaction in Institutional Settings.
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.), 233252. 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: 219236.
Givn, T. 2005. Context as Other Minds: The Pragmatics of Sociality, Cognition and Communica-
tion. Amsterdam: Benjamins.
Goffman, E. 1964. The neglected situation. American Anthropologist 66: 133136.
Goffman, E. 1974. Frame Analysis: An Essay on the Organisation of Experience. Boston: North-
eastern University Press.
Goffman, E. 1979. Footing. Semiotica 25: 129.
Grice, H.P. 1975. Logic and conversation. In Syntax and Semantics 3: Speech Acts, P. Cole and
Jerry Morgan (eds.), 4158. 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.), 130152. New York: CUP.
Gumperz, J.J. 1992a. Contextualization revisited. In The Contextualization of Language, P.
Auer and A. Di Luzio (eds.), 3953. Amsterdam, Philadelphia: Benjamins.
Gumperz, J.J. 1992b. Contextualization and understanding. In Rethinking Context: Language

as an Interactive Phenomenon, A. Duranti and C. Goodwin (eds.), 229252. New York:
Hanks, W.F. 2006. Communicative context. In Encyclopedia of Language and Linguistics. Sec-
ond Edition, K. Brown (ed.), 115128. Amsterdam: Elsevier.
Harris, S. 1991. Evasive action: How politicians respond to questions in political interviews.
In Broadcast Talk, P. Scannell (ed.), 7699. London: Sage.
Jackman, H. 2006. Contextualism in epistemology. In Encyclopedia of Language and Linguis-
tics. Second Edition, K. Brown (ed.), 128130. 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.),
374385. Oxford: Blackwell.
King, S.W. and Sereno, K. 1984. Conversational appropriateness as a conversational impera-
tive. Quarterly Journal of Speech 70: 264273.
Kotthoff, H. 1996. Impoliteness and conversational joking. Folia linguistica 30: 299325.
Lauerbach, G. 2003. Opting out of the media-politics contract: Discourse practices in con-
frontational political interviews. In Dialogue Analysis 2000: Selected Papers from the 10th
IADA Anniversary Conference, Bologna 2000, M. Bondi and S. Stati (eds.), 283294. Tbin-
gen: Niemeyer.
Lauerbach, G. 2004. Political interviews as hybrid genre. Text 24: 353398.
Levinson, S.C. 1979. Activity types and language. Linguistics 17: 365399.
Levinson, S.C. 1983. Pragmatics. Cambridge: Cambridge University Press.
Meier, A.J. 1995. Defining politeness: Universality in appropriateness. Language Sciences 17:
345356.
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,
103128. 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: 696735.
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. George-
town University Roundtable on Languages and Linguistics 1981, D. Tannen (ed.), 7193.
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.), 2852.
Schegloff, E.A. 1987. Between macro and micro: Contexts and other connections. In The
Micro-Macro Link, J.C. Alexander, B. Giesen, R. Mnch and N.J. Smelser (eds.), 20733.
Berkeley: University of California Press.
Schegloff, E.A. 1989. From interview to confrontation: Observations of the Bush/Rather en-
counter. Research on Language and Social Interaction 22: 215240.
Schegloff, E.A. and Sacks, H. 1973. Opening up closings. Semiotica 8: 289327.
Stainton, R.J. 2006. Context principle. In Encyclopedia of Language and Linguistics. Second
Edition, K. Brown (ed.), 108115. Amsterdam: Elsevier.
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.), 189226. Amsterdam: Benjamins.
Weizman, E. 2006. Roles and identities in news interviews: The Israeli context. Journal of
Pragmatics 38, 154179.
Cooperative conflict and evasive language
The case of the 911 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 911
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-in-
terrogators can best be described as evasive, representing a series of discursive
tactics which produce nonlinear patterning between adjacency pairs. Yet, de-
spite evidence that use of indirect, evasive language is not uncommon in court-
room 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 responsibil-
ity 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 in-
volved, Sbis (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 de-
picted 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 in-
teractional 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 Hallidays (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 Hallidays 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 lim-
ited, given, and objective using Sbiss (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 ger-
mane to move from a limited, one-dimensional notion of linguistic context (i.e.,
represented by constituents within a single individuals talk) to a multidimension-
al 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 Hallidays work, this level of context corre-
sponds with his notion of language as behavior wherein code [] has been de-
termined 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 Hallidays sense) and what is accomplished
(i.e., by the language or interpersonal in Hallidays sense); the distinction might
otherwise be specified as internal, speaker-oriented interactional context and ex-
ternal, hearer-oriented interactional context.
In the former, internal context, speaker-oriented choices are what partici-

pants carry with them upon entering into a communicative interaction (though
the possibility of making new choices cannot be precluded and emerges in reac-
tion to the degree of success a speaker achieves in the interaction). Internal inter-
actional 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 appar-
ent 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, Its hot in here, the
illocutionary force of the utterance may be either declarative or directive (Levin-
son 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).
Sbis (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 Austins (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 Austins performa-
tives, but also constatives, directives, etc. in that they all include some future action which can
be context-changing (cf. Levinson 1983).
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 determina-
tion of the type of speech act, then, is only interpretable through the reaction
obtained. For instance, returning to the above initiative Its hot in here, my inter-
locutor may respond by saying, Thats 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 speakers turn. Section 2.2 presents a taxonomy of
response types as a means of constructing cooperative conflict through evasion.
The final dichotomous question Sbis poses, however, must be split within
this level of context between the internal interactional context and its concomi-
tant speaker-oriented choices as subjective (and ultimately only suggestively in-
terpretable, relative to the level of explicitness), and the external interactional
context and its concomitant hearer-oriented choices as objective. Within a criti-
cal 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 Hallidays language as code and language as behavior) allows for a richer analy-
sis 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 (Malinowskis 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 situ-
ational context, Spradleys elements that define a domain correspond with the
prevalent view in many disciplines and methodologies (e.g., CDA) where a dis-
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 (Sbis 2002).
While considerations of a broader understanding of context are basic to eth-
nographic 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 sequenc-
es, and the outcomes of interactions. In the current chapter, for example, the right
of 911 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 dont ask interrogators to answer probing ques-
tions).
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-partici-
pants, the roles they assume, and their interaction. Chouliaraki and Fairclough
present four main moments of social practice for consideration at this level: ma-
terial activity (specifically non-semiotic, in that semiosis also has a material as-
pect, 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 ele-
ments 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 partici-
pants in the present study maintain multiple roles and act accordingly.
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 Kumaravadivelus (1999) 3-level
hierarchical structure of a critical classroom discourse analysis wherein he pos-
its 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 influenc-
ing and interacting upon each other simultaneously.
The first interface (cf. Janney 2002), between the (socio)linguistic and the so-
ciocultural, 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 lo-
cal 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 knowl-
edge 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, Foucaults
discourse has many similarities with Bourdieus 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 ex-
act 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 situ-
ational 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).
Cultural domains social practices and their respective discourses in ad-

dition, 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; Kumara-
vadivelu 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 (Voloinov), 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 linguis-
tic 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 conjunc-
ture. Establishing boundaries albeit fuzzy for any consideration of the extra-
situational 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
Sbis, extrasituational context presents itself as an anomaly. For the first ques-
tion 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-con-
structed 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).
1.5 Appropriateness
Appropriateness is relative to context or, more specifically, realized in the inter-

face 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: 5253).
Grices (1975) Cooperative Principle can be extremely informative for de-
limiting 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. Hes 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 co-
text 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-
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 en-
gaging in the interaction and interpreted by those outsiders who may hear, ob-
serve, and/or read the transcripts of the interaction. Thus, co-participants use
their knowledge of extrasituational context to recognize and make use of the situ-
ational 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 immedi-
ate linguistic context. Outsiders3 use their knowledge of extrasituational context,
their knowledge and/or perceptions of the situational, interactional, and linguis-
tic 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 Sbis (2002), the evaluation of what is appropriate (or inappropri-
ate) 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 com-
munication interrogators want to get answers to their questions (typically cor-
responding to the way the questions are framed) and witnesses, especially when
being cross-examined, are compelled to answer, but want to avoid self-incrimina-
tion. 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 dont necessarily hear any-
thing (e.g., Those two people look like theyre having an argument). Overhearers are defined by
Bell as fulfilling many roles of varying degrees of involvedness, such as those immediately pres-
ent 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.
(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 youve 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 911 Com-
mission Hearings (911 Commission 2004; NCTA 2004), not only do the par-
ticipants maintain their various statuses in performance of the interaction, but
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 en-
gaged 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 posi-
tion 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 be-
ing 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 911 Commission, specifically the hearing on counter-
terrorism 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 diplo-
macy. 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 911 attacks
on the United States.
2.1 Hearing structure
Since the 911 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 tran-
scripts 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.
Administration
Introduction
Instruction
Oath
Statement
Interrogation
Opening
Question {Orientation} Answer/Non-Answer
{Admonition Rebuttal}
Closing
Figure 2. 911 Commission hearing structure
interrogator and the witness also flows similarly to most courtroom interactions
(see Figure 2). Since Im primarily concerned with the structure and nature of
cooperative and co-constructed evasive talk, I focus my attention on the inter-
rogation 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 re-
spondent 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 ap-
propriate in the analysis of the practice where witnesses may choose a course of
action relative to their own needs in the interaction.
Table 1. Response tactics
RESPONSE TYPE CODE MAXIM FUNCTION

Answer
Direct Answer DIR
Disavowed Knowledge DIS Opt Out Quality
Elaboration ELA Flout Quantity/Clash Manner
Mitigation MIT Violate Manner
Non-Answer
Assertion of Platform APL Flout Quantity/Flout Manner
Reassignment of Culpability or Responsibility REA Flout Relation/Clash Quality
Restructure STR Flout Relation/Opt Out Quality
Clarification CLR Flout Relation
Partial Answer PTA Flout Quantity
Repetition REP 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 tes-
timony where the interrogator provides sufficient linguistic context for the wit-
ness 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 911 Commission Hearings (911 Commission 2004;
NCTA 2004), direct answers were most often observed in interactions between
members of the same political party (i.e., Democrat interrogator-Democrat wit-
ness 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 didnt
know, by the time we left office, you didnt know that the attack
on the Cole was the responsibility of Bin Ladin. Is that correct?
[A 4/DIR] MS. ALBRIGHT: That is correct.
(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 its because Im 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 Im 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, inter-
nally halved its internal security enforcement budget?
[A/DIR 7] MS. RICE: I was not made aware of that -- I dont 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?
[A/DIS 8] MS. RICE: I do not believe I was aware of that.

[Q 9] MR. LEHMAN: Were you aware -- to shift a little bit to Saudi
Arabia, were you aware of the program, that was well-estab-
lished, that allowed Saudi citizens to get visas without inter-
views?
[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 re-
spondent claims to be incapable of providing the requested information by negat-
ing 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 respon-
dent through the willingness of the interrogator to accept the answers. This par-
ticular 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 911
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 inter-
rogation in a way that allows Rice to give short answers which maximize the ap-
pearance 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
total number of questions: interrogators from the same political party as the re-
spondent 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 mitiga-
tion (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 Lehmans 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 informa-
tion than is necessary. The respondent also runs the risk of clashing with the max-
im 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; con-
sequently, 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 [es-
tablish or maintain] certain ideological assumptions as commonsensical (Fair-
clough 1989: 90). An example of this type of strategy is found in (5) where Rice
responds to Keans 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: Ive got a question now Id 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 intel-
ligence 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 Ive 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
based on new threat information, and I said no one could have

imagined them taking a plane, slamming it into the Pentagon --
Im 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.
[DIS 2] 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.
[MIT 2] Part of the problem is -- I think Sandy Berger made this point
when he was asked the same question --
[REA 2] 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 dis-
course 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 911
Commission, first allows Rice to go on at length, countering a particular accusa-
tion 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 911 attacks. She then proceeds with this portion of the exchange through
mitigation, suggesting that an inordinate amount of information made it impos-
sible 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 ques-
tion 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 exem-
plifying 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
the interaction itself is necessarily uncooperative or inappropriate in a theoreti-

cal 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 inter-
action.
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. Sec-
ond, 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, add-
ing the caveat that it was hard to quantify.
(6) Roemer (Democrat) Interrogation of Albright (Democrat)
[Q 2] MR. ROEMER: [] Isnt 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 -- well 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
using our military there. What percent of your time can you
best estimate that you spent on counterterrorism policy?
[N-A/APL 3] MS. ALBRIGHT: Its 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 sec-
retary 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 coun-
terterrorism issues. []
[Q 3]5 MR. ROEMER: Can you guess at all? Twenty percent? Fifty
percent?
[A 3] 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., 20012004), also re-
sponds to his interrogator without answering while going into some detail. In this
particular case, Powells aggrandizement is not self-oriented, but other-oriented
in that he depicts President Bushs 911 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 Powells response evasive and uncooperative (e.g.,
repetition of the question in Q 3 and Q 3), but he ultimately accepts that the in-
terrogation is over by deferring to the time restraints and the call by the Chairman
to conclude (Im out of time and Im going to listen to my chairman.).
(7) Ben-Veniste (Democrat) Interrogation of Powell (Republican)
[Q 2] MR. BEN-VENISTE: = Secretary Wolfowitzs 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.
time. I cant recall whether he said instead of Afghanistan.

We all knew that Afghanistan was where al Qaeda was.
[Q 3] MR. BEN-VENISTE: But was there any concrete basis upon
which that recommendation was founded in your view to
attack Iraq for 9/11?
[N-A/REA 3] 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 =.
[Q 3] MR. BEN-VENISTE: Ive asked for your view, with all due
respect, Secretary Powell.
[N-A/STR 3] SEC. POWELL: With all due respect, I dont think I should
characterize what Mr. Wolfowitzs views were.
[Q 3] MR. BEN-VENISTE: No, I asked for your view. In your
view =.
[N-A/REP 3] SEC. POWELL: My view =.
MR. BEN-VENISTE: = was there a basis?
[N-A/APL 3] SEC. POWELL: My view was that we listen to all the argu-
ments 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 didnt go after
the source of the 9/11 terrorists.
MR. BEN-VENISTE: Im out of time and Im 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. Reassign-
ment, though, tends to flout the maxim of relation while clashing with the maxim
of quality. In Albrights 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 of-
fice. 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-
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. Interest-
ingly, 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 differ-
ent 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 dont think I should characterize what Mr. Wolfowitzs 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 dont 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 its important
=.
[Q 1] MR. BEN-VENISTE: Did you tell the President? (AP-

PLAUSE).
[N-A/CLR 1] MS. RICE: Its important that I also address -- its also
important, Commissioner, that I address the other issues
that you have raised. So I will do it quickly, but if youll 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 youve asked.
First of all, yes, the August 6th PDB was in response to ques-
tions of the President.
[REA 1] In that sense he asked that this be done. It was not a par-
ticular threat report. And there was historical information
in there about various aspects of al Qaedas operations. Dick
Clarke had told me, I think in a memorandum -- I remem-
ber 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 dont 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 dont remember the al
Qaeda cells as being something that we were told we needed
to do something about.
[Q 2a/b] MR. BEN-VENISTE: Isnt 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 =.
MR. BEN-VENISTE: Thank you.
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 shar-
ing what she claims to be essential information without responding to the ques-
tion: I understand, Commissioner, but I will -- if you will just give me a moment, I
will address fully the questions that youve 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 plat-
form 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), Rices 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 in-
struct?). 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 cant pass this up. I know it will take into my
10-minute time. But as somebody who supported the war in
Iraq, Im not going to get the national security advisor 30 feet
away from me very often over the next 90 days.
(LAUGHTER).
And Ive 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; its not a war itself.
Secondly, let me say that I dont think we understand what
the -- how the Muslim world views this, and Im terribly
worried that the miliary tactics in Iraq are going to do a
number of things, and theyre all bad. One is the --
(APPLAUSE).
(to the audience) No, please dont. Please do not do that. Do
not applaud.
[]
Let me ask you, first of all, a question thats 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
[Q 1] 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?
[A/ELA 1] 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.
[Q 1] MR. KERREY: Yes, you did talk to him about terrorism?
[N-A/PTA 1] MS. RICE: We talked -- Philip and I, over a period of -- you
know, we had worked closely together as academics, of
course talked about --
[Q 2] MR. KERREY: During the transition, did you instruct him
to do anything on terrorism?
[N-A/REP 2] MS. RICE: Oh, to do anything on terrorism?
[Q 2] MR. KERREY: Yes.
[N-A/STR 2] MS. RICE: To help us think about the structure of the
terrorism -- Dick Clarkes 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 superfi-
cial analysis, responding to the concerns of various scholars to contextualize find-
ings 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
themselves, themes may be considered a unit of analysis (cf. Shuy 1982, for a dis-
cussion 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 sociopoliti-
cal 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 911
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 begin-
ning 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 worlds 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 suc-
ceeding, 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-
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 oc-
currence 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 insis-
tence 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 inher-
ited 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 Re-
publicans 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 sur-
rounding 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 911 Commission to bring all security and information agencies un-
der one office in order to assure the sharing of relevant information regarding potential threats
to the nation.
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 al-
ways due to personal volition. It may simply be that the interrogator and the wit-
ness are speaking from two different places, causing an unintentional ambiguity,
or they simply may be verbally sloppy (2001: 446). In the case of the 911 Hear-
ings, 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
Rices 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 (in-
terrogators 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 hear-
ings (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 Afghani-
stan and Iraq had already taken place. Additionally, several books had been pub-
lished and were widely known as they had condemned the Bush administrations
policies, most notably a book by Richard Clarke (2004), former counter-terror-
ism expert who had been employed since the time of the Reagan administration.
Clarkes book is also referred to numerous times in the transcripts. The combina-
tion of synchronic factors made it absolutely necessary for the Bush affiliates to
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 them-
selves. They must portray certitude when in fact they have no objective evidence
(2004: 517). In the case of the 911 Commission Hearings, however, I would argue
that it is equally important to express certitude even in the face of refuting objec-
tive evidence as this could prove disastrous for them in forthcoming elections.
4.2 Political positioning
As previously indicated, the 911 Commission was composed of both Republi-

cans and Democrats. Considering the timing shortly before the 2004 Presiden-
tial election (sociocultural influence in the situational context) it would seem
the hearings presented an ideal opportunity for the Democrats to position them-
selves 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 ma-
jority party in both houses of the Congress. Indeed, legislators who did not show
proper support for the presidents 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 enforce-
ment 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 of-
ficials 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 pri-
mary 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 Levinsons (1987) measurement of face-threatening acts,
more attention is given to the face needs of co-participants regardless of the po-
tential conflict between them when the social distance is the lowest and the is-
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 (de-
spite 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, osten-
sibly 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 coopera-
tive, 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 prob-
lematic. Politicians seem to want to hide the negative within particular formula-
tions 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, politi-
cians manipulation of or attention to linguistic forms in this case, through eva-
sive 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 re-
move 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 coop-
eration is achieved through strict adherence to the rules of the general hear-
ing 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), situ-
ational, 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.
References
911 Commission. Official website of the 911 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: 145204.
Bell, A. 1998. The discourse structure of news stories. In Approaches to Media Discourse, A.
Bell and P. Garrett (eds.), 64104. 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.), 173182. Tbingen: Max Niemeyer Verlag.
Brown, P., and Levinson, S.C. 1987. Politeness: Some Universals in Language Usage (2nd ed.).
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.),
272296. Hampshire: Palgrave Macmillan Ltd.
Chouliaraki, L., and Fairclough, N. 1999. Discourse in Late Modernity: Rethinking Critical Dis-
course Analysis. Edinburgh: Edinburgh University Press.
Clarke, R. 2004. Against All Enemies: Inside Americas War on Terror. New York: Free Press.
Coulthard, M., and Montgomery, M. (eds.). 1981. Studies in Discourse Analysis. London: Rout-
ledge & 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.), 142162. 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.), 8796. Tbingen: 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 Con-
text: Language as an Interactive Phenomenon, A. Duranti and C. Goodwin (eds.), 142.
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.
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.), 335. London and Wolfeboro, NH: Frances Pinter.
Halliday, M.A.K. 1994. Introduction to Functional Grammar (2nd ed.). London: Edward Ar-
nold.
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): 439462.
Hymes, D. 1972. Models of the interaction of language and social life. In Directions in Socio-
linguistics: The Ethnography of Communication, J.J. Gumperz and D. Hymes (eds.), 3571.
New York: Holt, Rinehart and Winston.
Janney, R.W. 2002. Cotext as context: Vague answers in court. Language & Communication
22: 457475.
Kumaravadivelu, B. 1999. Critical classroom discourse analysis. TESOL Quarterly 33 (3),
453484.
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.), 205238. 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.), 296336. 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 con-
straints. In Studies in the Organization of Conversational Interaction, J. Schenkein (ed.),
79112. New York: Academic Press.
Rudd, P.W. 2004. Weapons of mass destruction: The unshared referents of Bushs rhetoric.
Pragmatics 14 (4): 499525.
Sbis, M. 2002. Speech acts in context. Language & Communication 22: 421436.
Shuy, R.W. 1982. Topic as the unit of analysis in a criminal law case. In Analyzing Discourse:
Text and Talk, D. Tannen (ed), 113126. Washington: Georgetown University Press.
Shuy, R.W. 2001. Discourse analysis in the legal context. In The Handbook of Discourse Analy-
sis, D. Schiffrin, D. Tannen, and H.E. Hamilton (eds.), 437452. Malden, MA: Blackwell
Publishing.
Sinclair, J., and Coulthard, M. 1975. Towards an Analysis of Discourse. London: Oxford Uni-
versity Press.
Small, E. Witness for the Prosecution. Prod. by Arthur Hornblow, Dir. by Billy Wilder. 116 min.
MGM/UA, 1957. Videocassette.
Spradley, J.P. 1980. Participant Observation. New York: Holt, Rinehart and Winston.
Tannen, D. 1989. Talking Voices: Repetition, Dialogue, and Imagery in Conversational Discourse.
Urban, G. 1991. A Discourse-Centered Approach to Culture: Native South American Myths and
Rituals. Austin: University of Texas Press.
Voloinov, 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.), 3554.
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.), 398415. 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 pre-
requisites 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 negotia-
bility of the speakers acts; it preferredly encodes new information and contrib-
utes to structure the thematic progression of discourse; it tends to encode initia-
tive and/or dispreferred communicative acts. These results lead us to formulate
some hypotheses as to the condatts appropriateness with regard to global con-
text. In particular, given that the form signals negotiability and is better compat-
ible 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 acquain-
tance, 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 70s, the pragmatic turn had brought to our attention the relevance and cru-
cial significance of context, in its multifarious aspects and relationships, for lin-
guistic analysis. But the investigation of the complex relationship between gram-
mar 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 Searles theories of speech acts,
Wittgensteins language game, Grices cooperative principle), together with func-
tional linguistics, conversation/discourse analysis,2 and, more recently, cognitive
linguistics, has strongly influenced subsequent linguistic research, in the direc-
tion 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, Levinsons 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 com-
puter-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 appro-
priateness 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, dif-
ferent aspects of context constitute a complex pragmatic configuration (cf. Ba-
zzanella 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, tak-
ing 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), Givn (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 parts-
whole perspective as an entity containing sub-entities (or sub-contexts). Fetzer (this volume).
will speak of local and global aspects of context (cf. Akman & Bazzanella 2003;
Bazzanella 1998, 2004). Local features of context are features that are intrinsi-
cally 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 interac-
tion prototypically includes the linguistic co-text, the sequential development of
interaction (or dialogue common ground, cf. Fetzer 2004: 202203, 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 inter-
action, 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. Ldi 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 par-
ticipants 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-
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 Ak-
tionsart 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 mark-
ing 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 attenu-
ative 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 childrens play), and the specific relationship between co-participants.
Conversely, speakers and hearers use linguistic items in multiple ways as con-
ventionalized 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 gram-
matical 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 appro-
priateness, 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 con-
ceptualized 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 appro-
priateness 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 conver-
sational 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.
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 de-
scriptions 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 sequen-
tial 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 forms 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- mor-
pheme 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 poly-
semous 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 pro-
duction) and disambiguating (in language comprehension) role of the immediate
lexical and syntactical co-text. Section 2 will give a first impression of the forms
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, prag-
matic and sequential analyses proposed in this chapter are:
(a) A plurilingual corpus of scientific conferences and group meetings. In a rough

approximation, the French part of this corpus counts about 400000 words,
the Italian part about 15000 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://languag-
eserver.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 parents committee held in Italian, pub-
lished 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 dplacement de coordonnes, i.e. a shift of coordinates). The conditionals
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 propo-
sition 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 construc-
tion 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).
The conditionals basic meaning is to express a relationship of (temporal, causal,

cognitive or pragmatic) antecedence between a contextually given ground, con-
ceptualized as being distant from the modal-deictic centre, and an entity p con-
taining 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
conditionals modal uses instead. The conditionals modal ground may be at least
of the following types:
a condition, inferable from the context or expressed explicitly in the congiun-

tivo (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: 3135), a reading which cross-lin-
guistically is prototypical, according to Comrie 1986 (cf. also Traugott et al.
1986; Couper-Kuhlen & Kortmann 2000);
a universe of discourse (as in the forms 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 in-
terrogative 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 im-
plicit 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: 8894 the condatt signals a ddoublement 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 un-
derlying 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 presupposi-
tion 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-
ditional form does. However, in the case examined here, the idea of presupposi-
tion 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 end-
ing, the range of the conditionals possible scopes as a presupposition modal-
izer 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 conditionals 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 ei-

ther 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 par-
ticipants 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 agents 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
agents desire has been fulfilled (cf. Squartini forth. for Old Italian, which behaves differently).
in the sphere of volitional modality such as sperare/esprer. 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
utterances presuppositions is de re: the speaker expresses explicitly, in the im-
mediately subsequent co-text, a counterfactuality of q due to factors that are inde-
pendent of the on-going interaction:
(1) (lip mc 10)14
tante 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 lambulante 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 communica-
tive 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
Id 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
. falling intonation <, > increasing, decreasing volume
(:h) breathing (x) incomprehensible passages
( ), ( ; ) tentative transcription, transcr. (( )) comments
alternatives
XX anonymized names
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. 34)
1 Q va=be al secondo punto allordine del giorno c 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 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 speakers 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 (Atha-
nasiadou & Dirven 2000: 1317) or speech act conditional (Dancygier & Sweetser
2005: 113117).
2.4 The conditional with verbs expressing communicative acts
Also with a large class of verbs of saying, the conditional may be used to modal-
ize 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
acts felicity conditions, i.e. those conditions that do not depend directly on the
16. Lit. I wantcond if there are no objections to speak.

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 con-
ditional, evokes the possibility that the interlocutors might not take up this com-
municative act as a relevant and well-founded contribution to the on-going dis-
cussion:
(4) (franceschini_2)
non mi sembra inopportuno che noi,-proprio a chiusura danno adesso
it doesnt seem inappropriate to me that we, - right now at the end of the year -
mandiamo una specie di riassunto di quello che si fatto.-io personalmente
send a sort of summary of what has been done.- I personally
dato che questa bozza c - riterrei se c da limarla la si lima
given the fact that there is this outline, considercond that if its necessary to
revise it well revise it
la si completa per i mesi che non ci sono anche s un po lunga.-
well complete it for the lacking months, even if its 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 + sub-
jonctif and in Italian by che + congiuntivo presente/imperfetto or + congiuntivo
imperfetto. (5) is a typical example:
(5) (overheard)
Sarebbe meglio tenerli al guinzaglio, per. (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, per.
it would be better if you kept them on a leash, you know
However, the resulting hypothetical construction is not a prototypical condition-

al (Comrie 1986): the positive evaluation expressed by the speaker does not de-
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 conditionals
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 pro-
duce 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: 248250).
When this implicature becomes prominent, the modal ground of the con-
ditional shifts towards the pragmatic presuppositions of the utterance, making
these constructions similar to constructions with modal verbs expressing voli-
tional or deontic modality. So in example (5), the positive evaluation expresses
a wish, which, however, is not to be understood as the speakers personal desire,
but refers deontically to a specific rule of the shared community. An alternative
paraphrase is therefore (5):
(5) Bisognerebbe tenerli al guinzaglio, per.
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 anteposi-
tion 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 em-
bedded proposition. Other formal characteristics regard the syntactic coding of
the embedded proposition. So markings such as que + subjonctif or che + congiun-
tivo formally reflect the conceptualization of the embedded proposition as a pos-
sible 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 vio-
lates a sequence of tense rule,17 is evidence for the attenuating conditionals 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 imper-
fetto.
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-man-
agement (cf. Brown, Levinson 1978/1987; Roulet 1993; Kerbrat-Orecchioni 2001,
2004).
The semantic and pragmatic analysis weve 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 re-
duce persuasiveness (cf. Sbis 1999). However, it suggests that, when the presup-
positions in question are related to the on-going interaction, effects on the inter-
personal level are neither necessarily a consequence of attenuation nor necessarily
limited to face management. Wed rather insist in the conditionals expressing ne-
gotiability of the speakers 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 predi-
cates shows that the condatt can be used to index the intersubjective component of
the speakers desires, axiological judgments and communicative actions, under-
lining 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 allordine del giorno c 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 niente in contrario parlare > (xx).
to speak - if there are no objections > (xx).
5 Q vorrei ch` <vorrei >chiedere se lhanno ricevuto tutti questa lettera.

wantcond to a < I wantcond to> ask if everybody has got that letter.
6 ((4 sec. of murmur))
7 B allora il XXX se ne pu anchee andare, (in fondo).
so in fact the XXX can leave, (after all).
8 Q no. sarebbee positivoo il comitato XX rimanesse anche qui
no. it would be positive the XX committee stayed here too
9 ((4 sec. of murmur, several instances of yes are to be heard))
10 D ce lhanno=ce lhanno tutti la lettera?---All(:h)ora visto che sono
does everybody have=does everybody have the letter? --- So(:h) since
I am
11 il primo destinatario mi permetto di essere il primo a parlare,-
the primary addressee I take the liberty of talking first, -
12 allora. io direi,-- che eem- vorrei premettere una cosa. che - il
o.k. Id say, -- that erm I wantcond to say something in advance.
that -
13 signor nardini ((2 lines omitted)) mi ha dato delle prove, - veramente
Mr. Nardini ((2 lines omitted)) has given me, - really
14 particolari di <>coraggio - di <>intelligenza - di <>attaccamento - e
strong proofs of <>courage - of <>intelligence - of <>attachment
- and
15 soprattutto, - quello che mi ha fatto molto piacere di seriet. --
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 combina-
tion 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 hearers 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., 1215,
in which the speakers subjective stance is expressed very clearly. The evaluative
construction sarebbe positivo + + 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 read-
ing of the construction and the choice of the congiuntivo imperfetto suggest that
the committees further participation is a mere hypothesis. According to the anal-
ysis of the condatt proposed here, the conditionals 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,
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 dernires euhm . communications que jai trouves tout
last er . papers that I found really
3 fait in- tout fait intressantes/ .. <et importantes
in- really interesting/ .. <and important
4 ((rapid))> . pour souligner une: . distinction que nous
((rapid))> . underlining a: . distinction
5 aurions PEUT-tre intrt faire\
maybe it would be in our interest to make
((turn continues: 48 lines omitted))
6 cest cest cest c- ce ce ce mcanisme xxx faut pas le perdre
its its its th- we shouldnt loose sight of this mechanism/
7 de vue/=hein cest un mcanisme euh: xxx (constant)\
=you see thats a mechanism euh: xxx (constant)
8 BRU faut peut-tre ajouter que le grand homme cest cest
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
((laughs)) phenomena not transcribed, indicated between (( ))
comments by the transcriber N.N. name
11 BRU & sait[=euh . de qui il sagit

&knows [=who youre talking about
12 GAU [oui
[yes
13 GAU m[hm
14 BRU [pour Regulus on a galement d- des allusions qui sont
[in the case of Regulus we also have a- allusions that are
15 vraiment xx dans une SEquence xxx o on cite que son nom\
really xx in a sequence xxx where its just his name that is cited\
16 (3 s.)
17 DUM je crois qucette distinction est effectivement trs trs
I think that this distinction is very very important indeed/
18 importante/ mais . on pourrait peut-tre lenrichir aussi
but . we could maybe further enrich it taking into
19 de la considration du CONtexte dans lequel est nonc
consideration the CONtext in which
20 le[: ]le personnage ou la vertu correspondant/
the[:] the corresponding figure or vertue is mentioned/
21 GAU [hm]
((turn continues for another 10 lines))
The lexical semantics of the complex predicate avoir intrt (l. 5) allows the
speaker to express a deontic statement motivated by an evaluative judgment about
what is in the groups 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 groups readiness
to comply with the obligation expressed.
The presupposition that the agent is able and disposed to behave in accor-
dance with a given obligation imposed on him/her is in fact associated with vari-
ous 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):
(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 dont engage themselves as they should.
Well 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 presupposi-
tion, indicating that he/she cannot take the agents 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 co-
present participants (referred to by inclusive on); the choice of the condatt under-
lines 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 organi-
zation of turns and to the sequential organization of conversation.
First of all, turn-internally, the condatt is frequently combined with structur-
ing devices signaling thematic boundaries and change of perspective. More pre-
cisely, it is used to encode what follows immediately these markers, i.e. an infor-
mation 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 intona-

tion and contributes to mark the passage from preliminaries concerning the
management of the conversational floor to the first point the speaker wants to
make;
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, CON-
texte (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 develop-
ing/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 expres-
sions such as je dis / dico or je dois dire que / devo dire che / devo dire una cosa.
Consider, for instance, faut peut-tre ajouter que in example (7), l. 8, in which
the choice of the indicative contributes to mark thematic and argumentative con-
tinuity 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. 67).
The condatt at thematic and argumentative boundaries may be seen as part
of sequence-initial focusing activities (Kallmeyer 1978): it contributes to the par-
ticipants 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 in-
terpersonal functions we have underlined in the preceding section. One might
hypothesize that the condatt underlines interactive negotiability at points in dis-
course where the hearers reception and/or consensus is particularly important for
the speaker. On the other hand, the condatts 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 condatts 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-
cusing activities is up-grading (Relevanzhochstufung in Kallmeyers 1978 terms)

rather than down-grading (Relevanzrckstufung).
To understand the condatts specificity at thematic and argumentative bound-
aries, it is helpful to take into account the anaphorical dimension of presupposi-
tions. 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 in-
stance, argues convincingly that a tight relationship holds between presupposi-
tion 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 presupposi-
tion inducers can be conceived of as anaphors, only differing from normal de-
noting 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 intro-
ducing reference markers for states and events. (Van der Sandt 1989: 287)
In our view, Van der Sandts general claim may be usefully applied also to the
presupposition triggering uses of modal predicates and the pragmatically presup-
posing 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 mchanisme xxx
faut pas le perdre de vue, l. 67) 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 (mchanisme) referred to is important. Conversely, doubt-
ing 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-tre intrt faire (ex. 7, l. 5) indicates that the criteria ac-
cording 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 con-
struct a new thematic and argumentative frame, which will confer intelligibility
and plausibility to the deontic statement made.
3.3 Interactional sequences
The condatts reset function observed with regard to thematic development

within turns is relevant also for another aspect of interaction, i.e. the level of ad-
jacency 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 men-
tioned 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 ini-
tiative, 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 sec-
ond pair part, but are in fact first pair parts, thus deceiving the expectations cre-
ated by a previous speaker. Instances of this kind of dispreferred sequential pro-
gression 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 co-
texts 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 announce-
ment 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 discus-
sions advance. In the multilingual corpus examined here, they intervene, in par-
ticular, to encourage language change after long stretches of discussion in one
language (cf. Miecznikowski, Mondada, Mller & 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 specifi-
cally. In this context, his metacommunicative announcement is to be analyzed as
a reactive utterance.
As a second example, we will reconsider BRUs metacommunicative an-

nouncement 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
GAUs previous turn (l. 17). This status of BRUs turn emerges progressively
thanks to the contributions of BRUs interlocutors at l. 12, 13 and 17. In particular,
DUMs utterance in l. 17 shows that the side-sequence maintains the expectance
of an answer to GAUs turn preceding the side-sequence: The expression cette
distinction refers back to the distinction introduced by GAU at l. 45, underlin-
ing discourse cohesion by lexical repetition19 and by the choice of the determin-
er. Moreover, it retrospectively confirms the initiative character of GAUs 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 in-
dicative present is that of repeated metacommunicative announcements. An ex-
ample is mi permetto di essere il primo a parlare (ex. 6, l. 11), which echoes the
previous announcement vorrei se non c niente in contrario parlare (ex. 6, l.
34). 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 negoti-
ated 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 hear-
ers possibility of having diverging attitudes and of impeding an act intended
by the speaker;
the condatts local context changing potential as an element of sequence-initial
focusing activities (within turns and in the management of complex sequenc-
es);
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 un-
derstanding the synergies (Bazzanella, Caffi & Sbis 1991) of relevant local pa-
19. Cf. Bazzanella (1996) for an introduction to the functions of repetition in dialogue.
rameters with the uses of the condatt and for making hypotheses about the forms
solidarity with parameters of global context.
On the local level, these functions correspond directly to a prototypical or-
dering 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 infer-
ences, to contexts in which the condatts is not attested or occurs only marginally,
and is incoherent or hardly interpretable.
Prototypical local contexts are contexts in which several components com-
monly 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 in-
teraction), 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 condatts diachronic development, the condatt taking over contextually
expressed functions by ways of metonymical semantic/pragmatic change (Hop-
per & Traugott 1993).
Less prototypical contexts are those in which only some components are pres-
ent, whereas others are unmarked or clearly absent. An example is second pair
part position, a case in which the condatts non-reactive function may be re-in-
terpreted as signaling dispreference, and the function of expressing contrast and
disagreement may be reinforced.
Highly untypical contexts, finally, are those in which the condatts 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 propo-
sition 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 hypoth-
eses, 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
as the degree of acquaintance and the degree of social distance between co-par-
ticipants.
A second relevant aspect is the condatts function of signaling co-construc-
tion 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 co-
construction 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 neces-
sity of managing long turns, rather than dyadic interactions with rapid turn
taking).
Evidence in favor of the hypotheses formulated above might be gathered for ex-
ample 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 cor-
respond 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 appro-
priateness 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 hearers 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.
modeling appropriateness conditions as a second order construct (Schtz 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 interac-
tional 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 ap-
propriate. 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 marked-
ness 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 inap-
propriateness 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 sub-
sequent actions it projects) allows them to assess the necessity of evaluating that
items appropriateness in the first place. The question of an items 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 items appropriateness, especially
in the case of polysemous and polyfunctional forms like the conditional, a cru-
cial problem is disambiguation. This problem arises, of course, above all on the
hearers side: any appropriateness judgment presupposes a reduction of the to-be-
judged utterances range of possible interpretations and an at least tentative at-
tribution 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 conditionals attenuative reading listed above,
which contrast with those of the temporal, quotative, hypothetical and inferential
readings), and between a given meanings pragmatic functions and context types
(the specific functions fulfilled by the condatt tend to differ, e.g., depending on
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 regard-
ing a particular linguistic choice, knowledge about prototypical contexts plays a
role in the recipient design of the problem manifestation. Typical examples are re-
contextualizations, 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 par-
ticipants 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 items 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, mak-
ing 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 conditionals discourse func-
tions 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 forms use as an interactional resource is closely related to its semantics

as a polysemous linguistic sign, in particular to its capacity of selecting an
utterances 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 form-
context 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 com-
municative competence allowing speakers to construct relevant contexts for con-
versational events; as Fetzer (2004: 235) states, The analyses of well-formedness,
acceptability and appropriateness, require a frame of investigation which goes be-
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 Con-
text. Berlin: SpringerVerlag.
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: 321329.
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: 349371.
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 Interna-
tional Cognitive Linguistics Conference, Amsterdam, 1997, 126. Amsterdam/Philadelphia:
John Benjamins.
Auer, P. and Di Luzio, A. (eds.). 1992. The Contextualization of Language. Amsterdam/Phila-
delphia: Benjamins.
Banca dati dellitaliano 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): 237255.
Bazzanella, C. 1994, 21996, 31997. Le facce del parlare. Firenze: La Nuova Italia.
Bazzanella, C. 1996 Introduction. In Bazzanella, C. (ed.), Repetition in Dialogue, viixvii. Tbin-
gen: Niemeyer.
Bazzanella, C. 1998. On context and dialogue. In S. mejrkov, J. Hoffmannov, O. Mllerov, J.
Svtl (eds.), Dialogue in the Heart of Europe, 407416. Tbingen: Niemeyer.
Bazzanella, C. 2000. Tense and meaning. In Marconi, D. (ed.), Knowledge and Meaning Topics
in Analytic Philosophy, 177194. Vercelli: Mercurio.
Bazzanella, C. 2002. The significance of context in comprehension: The we case. Foundations
of Science 7: 239254.
Bazzanella, C. (ed.) 2002 Sul dialogo. Contesti e forme di interazione verbale. Milano: Guerini e
associati.
Bazzanella, C. 2004. Attnuation et intensification dans lItalien: dimensions et configuration
pragmatique. In Maria Helena Arajo Carreira (ed.), Plus ou Moins!? LAttnuation et
lIntensification dans les Langues Romanes (Travaux et documents 5), 173200. Paris: Uni-
versit Paris 8 Vincennes Saint-Denis.
Bazzanella, C. 2005. Linguistica e Pragmatica del Linguaggio. Roma-Bari: Laterza.
Bazzanella, C., Caffi C. and Sbis, M. 1991. Scalar dimensions of illocutionary force. In I.Z.
Zagar (ed.), Speech Acts. Fiction or Reality?, 6376. Ljubljana: IPRA distribution Center
for Jugoslavia.
Berretta, M. 1992. Sul sistema di tempo, aspetto e modalit nellitaliano contemporaneo. In
Moretti, B. et al. (eds.), Linee di Tendenza dellItaliano Contemporaneo, 135153. Roma:
Bulzoni.
Bonu, B., Mondada, L. and Relieu, M. 1994. Catgorisation: lapproche de H. Sacks. In Fra-
din, B., Qur, L. and Widmer, J. (eds.), LEnqute sur les Catgories. Raisons Pratiques, 5,
129148. Paris: Editions de lEHESS.
Bouquet, P. et al. 2003. Theories and uses of context in knowledge representation and reason-
ing. In Akman, V. and Bazzanella, C. (eds.), On Context, Journal of Pragmatics, special issue
35: 455484.
Bouquet, P., Serafini, L., Brzillon, P, Benerecetti, M. and Castellani, F. (eds.) 1999. Proceedings
of the Second International and Interdisciplinary Conference on Modelling and Using Con-
text. 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, 56248. Cam-
bridge, 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 Mo-
dality 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, 7799. Cambridge: Cambridge University
Press.
Couper-Kuhlen, E. and Kortmann, B. (eds.). 2000. Cause, Condition, Concession, Contrast. Ber-
lin: 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 dellItaliano
Parlato-LIP, Milano: EtasLibri.
Dendale, P. 2001. Les problmes linguistiques du conditionnel en franais. In Tasmowski, J.
and Dendale, P. (eds.). Le Conditionnel en Franais, 720. Metz: Universit de Metz.
Dik, S. 1989. Functional Grammar. Part I: The Structure of the Clause. Dordrecht: Foris publica-
tions.
Duranti, A. 1986. The audience as coauthor: An introduction. Text 6: 239247.
Duranti, A. and Goodwin, C. (eds.). 1992. Rethinking Context: Language as an Interactive Phe-
nomenon. Cambridge: Cambridge University Press.
Edmonds, B. and Akman, V. (eds.). 2002. Context in Context, Foundations of Science, special
issue 7: 233364.
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. 159182. Amsterdam: Elsevier.
Fetzer, A. Introduction. This volume.
Fetzer, A. and Fischer, K. (eds.). 2007. Lexical Markers of Common Grounds. Amsterdam: Else-
vier.
Franceschini, R. 1998. Riflettere sullInterazione. Milano: FrancoAngeli.
Giglioli, P.P. (ed.). 1972/1973. Language and Social Context. Harmondsworth: Penguin.
Givn, T. 1989. Mind, Code and Context. Essays in Pragmatics. Hillsdale: Lawrence Erlbaum.
Givn, T. 2005. Context as Other Minds. The Pragmatics of Sociality, Cognition and Communica-
tion. Amsterdam hiladelphia: John Benjamins.
Grice, H.P. 1989/1993. Studies in the Ways of Words. Cambridge: Cambridge University Press.
Gumperz, J. 1977. Sociocultural knowledge in conversational inference. In Saville-Troike, M.

(ed.), Linguistics and Anthropology, 191211. Washington: Georgetown University Press.
Haegeman, L. 1984. Pragmatic conditionals in English. Folia Linguistica, 18, 485502.
Haillet, P. 2002. Le Conditionnel en Franais: une Approche Polyphonique. Paris: Ophrys.
Hopper, P. and Traugott, E.C. 1993. Grammaticalization. Cambridge, Cambridge University
Press.
Kallmeyer, W. 1978. Fokuswechsel und Fokussierungen als Aktivitten der Gesprchskonsti-
tution. In Meyer-Hermann, R. (ed.), Sprechen Handeln Interaktion, 191241. Tbin-
gen: Max Niemeyer.
Kaplan, D. 1977. Demonstratives. An essay on the semantics, logic, metaphysics, and epis-
temology of demonstratives and other indexicals. In Almog J., Perry J. and Wettstein H.
(eds.), Themes from Kaplan, 481565. Oxford: Oxford University Press.
Kerbrat-Orecchioni, C. 2001. Les Actes de Langage dans le Discours. Thorie et Fonctionnement.
Paris, Nathan.
Kerbrat-Orecchioni, C. 2004. Ladjectif petit comme procd dattnuation en franais.
In Arajo Carreira, M.H. (ed.), Plus ou Moins!? LAttnuation et lIntensification dans les
Langues Romanes (Travaux et Documents 5), 153175. Paris: Universit 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 ps and qs. In Papers from the Ninth
Regional Meeting of the Chicago Linguistic Society: 292305. 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, 66100. Cambridge: Cambridge University
Press.
Light, P. and Butterworth, G. (eds.). 1993. Context and Cognition. Ways of Learning and Know-
ing. 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: 409434.
Ldi, G. 1991. Construire ensemble les mots pour le dire. A propos de lorigine discursive des
connaissances lexicales. In Dausendschn-Gay, U., Glich, E. and Krafft U. (eds.), Linguis-
tische Interaktionsanalysen, 193224. Tbingen: 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., Mller, K. & Pieth, C. 1999. Gestion des asymtries et ef-
fets de minorisation dans des discussions scientifiques plurilingues. In Actes du Colloque
Langues Minoritaires en Contexte, Coire 2123.9.98, Bulletin Suisse de Linguistique Ap-
plique, 69 (2): 167187.
Miecznikowski, J. 2005. Le Traitement de Problmes Lexicaux lors de Discussions Scientifiques
en Situation Plurilingue. Procds Interactionnels et Effets sur le Dveloppement 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.
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 presup-
posizionali. In: Atti del IX convegno SILFI, Firenze, 1517 giugno 2006.
Mondada, L. 2001. Pour une linguistique interactionnelle. Marges Linguistiques, 1: 121.
Mondada, L. 2005. Chercheurs en Interaction. Comment mergent les Savoirs. Lausanne: Presses
Polytechniques et Universitaires Romandes.
Ochs, E., Schegloff, E.A. and Thompson, S.A. 1996. Interaction and Grammar. Cambridge:
Oishi, E., 2007. Austins speech act theory and the speech situation. This volume.
Penco C. (ed.). 2002. La Svolta Contestuale. Milano: McGrawHill.
Perry, J. 1993. The Problem of the Essential Indexical and Other Essays. New York: Oxford Uni-
versity Press.
Renzi, L., Salvi, G. and Cardinaletti, A. (eds.). 1991. Grande Grammatica Italiana di Consultazi-
one, vol. 3. Bologna: Il Mulino.
Riegel, M. 1996. Grammaire Mthodique du Franais. Paris: Presses Universitaires de France.
Riehle, A. 2003. Neural correlates of contextrelated behaviour. Journal of Pragmatics 35(3):
485504.
Roulet, E. 1993. Des formes et des emplois des modalisateurs de proposition dans linteraction
verbale. In Dittmar, N. and Reich, A. (ed.), Modality in Language Acquisition, 2739. Ber-
lin: De Gruyter.
Sacks, H. 1992. Lectures on Conversation. Oxford: Basil Blackwell.
Sbis, 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, 492509. 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, 141. Cambridge: Cambridge Uni-
versity Press.
Schtz, A. 1971 [1953]. Wissenschaftliche Interpretation und Alltagsverstndnis menschli-
chen Handelns. In Gesammelte Aufstze I. Den Haag, Martinus Nijhoff.
Schwarze, Ch. 1983. Modus und Modalitt. In Ch. Schwarze (ed.), Bausteine fr eine italieni-
sche Grammatik, vol. I, 265302. Tbingen: Gunter Narr.
Selting, M. and Couper-Kuhlen, E. 2000. Argumente fr die Entwicklung einer interaktio-
nalen Linguistik. Gesprchsforschung, 1: 7695. 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): 297334.
Squartini, M. (forth.). Lespressione della modalit. In Salvi, G. and Renzi, L. (eds.), Gram-
matica dellItaliano Antico. Preprint consulted in November 2005 at http://ludens.elte.hu/
~gps/konyv/modalita.doc.
Tannen, D. 1989. Talking Voices. Repetition ,Dialogue, and Imagery in Conversational Discourse.
Tasmowski, J. and Dendale, P. (eds.). 2001. Le Conditionnel en Franais. Metz: Universit de
Metz.
Traugott, E. C., ter Meulen, A., Snitzer Reilly, J. and Ferguson, C. A. (eds.). 1986. On Condition-
als. Cambridge (UK): Cambridge University Press.
Traugott, E. and Dasher, R. 2002. Regularity in Semantic Change. Cambridge, Cambridge Uni-
versity Press.
Van der Sandt, R. 1989. Presuppositions and Discourse Structure. In Bartsch, R., van Ben-
them, J. and van Emde, B. (eds.), Semantics and Contextual Expression, 267294. Dordre-
cht: 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 Uni-
versity Press.
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 dia-
logue functions of the contrastive markers of Italian ma (but), invece (instead),
mentre (while) and per (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-partici-
pants 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 informa-
tion 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 com-
mon ground during dialogue topic management.
1. Introduction
Corpus investigation has recently evidenced that contrastive markers of Ital-

ian such as ma (but), per (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 frag-
mentarily concentrating on isolated items. For example, various roles have been
described in a fine-grained way for ma, ranging from a textual topic shifter (Ber-
retta 1984; Mortara-Garavelli 1993)1 to a turn-taking device used to interrupt the
current interlocutors 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 mul-
tifunctionality, 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 appropriate-
ness conditions of these markers, which is the focus of the present discussion.
What turns out from previous work concerned with English is that the so-
caled contrastive markers form a sub-set of the ampler functional category of
discourse connectives (Fraser 1988, 1998). Within this ampler functional cate-
gory 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 speakers 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 Mal-
amud-Makowski (1996: 866).
(1) U1. I can go at 4 p.m. tomorrow. U2. But I cant 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 wont 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 interlocutors expectation, for

instance, that being tall means being good at basketball, as is the case in the La-
koff 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 hes not good at basketball.
From an a priori unrelated perspective, the emphasis on the contribution of but

to the interlocutors optimally relevant interpretation of the speakers 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 basket-
ball.
Collaborative use of contrastive markers 237
determined the speakers 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 co-
participant seems to come from those approaches to the connectives that invoke
the notions of information structure and focus. Against this background, the but-
contrast 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 didnt 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 context-
updating potential (Kruijff-Korbayov & Webber 2001). The information struc-
tural framework offers a starting point for those analyses of the contrastive mark-
ers 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 contex-
tual 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 in-
stance, lets consider the occurrences of contrastive markers in some exploratory
examples, such as the map-task dialogue in (5), and the information-seeking dia-
logue 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.
(5) a. (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 la direzione?
Yes. But which is the direction?
c. Devi andare a sinistra.
You have to go to the left.
23 Francesca Carota
d. Per prima dello steccato io devo andare sulla collina.

But before the fence I have to reach the hill.
(6) a. La partenza quindi, c un volo di mattina
For the departure then, there is a flight in the morning.
b. Mentre con laltra compagnia c un volo alle 15.00.
While with the other company there is flight at 3 PM.
(7) a. Allora per landata 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.
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 attenu-
ate 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 per (see (5d)) opens a clarifica-
tion 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-partici-
pants negotiation of somehow competing pieces of already discussed and current
information. Such empirical evidence gives rise to the intuition that the contras-
tive markers correlate significantly with the psycho-cognitive process of ground-
ing (Clark 1996), whereby the co-participants achieve a common ground, i.e. a
cognitive-interpersonal, or social type of dialogue context, on which the informa-
tion 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 topic-
information 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 micro-
structural, 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
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 im-
port 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 theoreti-
cal 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 contras-
tive 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 dia-

logues. For the communicative interaction to fulfill effectively the current task-re-
lated 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 navi-
gation 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-ide-
ational dialogue meta-scheme, i.e. the dialogue topic.
The former coincides with the reciprocal discussion of some alternative so-
lutions for the same task-related issue (Larsson 2002), like the alternative train-
plane in example (8).
(8) A: Preferisce viaggiare in treno o in aereo?
Do you prefer to travel by train or by plane?
The latter can be configured, on the other hand, in terms of meta-negotiation

(Carota 2004). Consider example (9):
(9) B.a.: Landata va bene. Poi vorrei vedere gli orari del ritorno.
The depature is ok. Then I would like to see the return time.
B.b.: Allora, per il ritorno, vediamoci sarebbe un treno alle 8.30, troppo
presto?
So, for the return, lets seethere would be a train at 8.30 am., is it
too early?
While landata (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-ne-
gotiated as offering the ideational frame for the next co-participants contribution
in (b).
Such an interpersonal meta-communicative strategy allows the co-partici-
pants to co-opt for the ideational coordinates to which their incoming contribu-
tions are anchored and to integrate them appropriately into the cognitive-social
dimension intrinsic in the communicative context. In this line, the meta-nego-
tiation strategy contributes to the dialogue topic management while interact-
ing 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 di-
mensions 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 con-
text, 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 com-
mon ground integrates both the cognitive and the social aspects inherent in the
communicative space.
In its original formulation, the grounding process has been schematically de-
scribed as including two phases, i.e. a presentation phase, in which an ungrounded
parameter x is presented to the addressees cognitive state to form a common
ground on it, and an acceptance phase, in which the presented parameter is poten-
tially 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 con-
versation, 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 dont
know), and so on.
As for the above mentioned ungrounded parameter x, it indexes a piece of in-
formation 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 dia-
logue 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-partici-

pants 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 dis-
criminated 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 ratifica-
tion of the common ground. For instance, the information in focus dal lato sinis-
tro 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.
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-ne-
gotiated information:
(10) G067: Eh...passi dal lato...eeh SINISTRO della casa del bigne.
Ehyou pass on the left side...ehat the left of the house of
bign.
F068: Dal lato sinistro ?
On the left side ?.
G069: Si.
Yes.
F070: Si.
Yes.
G071: Lato <F072> #sinistro##
Side left side.
F072: #<G071> Si <ii>#
Yes.
G073: E arrivi fino<oo> a<aa> sopra la casa del bigne.
And you arrive until the house of bign.
F074: Sopra.
Above.
G075: #<F074> 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 cur-
rent 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 perspec-
tive 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 informa-
tive 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 substan-
tially 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
that the participants assume to be taken for granted and no subject to further dis-
cussion 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 de-
cisive 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 dis-
course-semantic content. The notion has been informally formulated in the fol-
lowing way:
A coherent aggregate of thoughts introduced by some participant in a conversa-
tion, 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 multi-
layered linguistic structure (see also Gmez-Gonzlez 2001). The semantic/prag-
matic notion of dialogue topic can then be captured through the dialogue-lin-
guistic 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 dis-
course, and new, or just introduced into the discourse), where the dialogue enti-
ties 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 de-
grees 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 phenom-
ena 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/new-
ness see Prince (1981), or on the activate, semi-active, inactive distinction see Chafe (2001).
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 struc-
ture, 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
speakers communicative goal by accomplishing the dialogue act which the utter-
ance 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 dia-
logue 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 submit-
ted to the common ground. When evoking a contextually available, closed set of
alternatives for this entity, it receives a contrastive interpretation (see also Molnr
2001). The global interplay of these co-textual parameters with the surface linguis-
tic indexes of common ground determines the dialogue structure (Carota 2004).
Globally, the dialogue structure is cognitively shaped by means of the ground-
ing 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 com-
mon 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 schemati-
cally as follows:
Co-text: Feedbacks, {information structural phenomena, dialogue entities}

Context: common ground status, Dialogue Topic status in the common ground
Thus, the dialogue structure is as an inherently dynamic, multilayered object ex-

panding 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 dia-
logue 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 pro-
file 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 ac-
counted 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 Stalnakers 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 com-
mon 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 presup-
position 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 pre-
suppositions (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 Stalnakers work, it is assumed that speakers produce a sentence in order to
update the information state roughly equivalent to Stalnakers term of context set , assuming
that the presuppositions of that sentence are satisfied in their current information state.
of common ground does not rely on the set of assumptions about the world knowl-
edge that are mutually shared by speaker and hearer, as in the philosophical view.
The illocutionary dimension and the common ground updating will be ad-
opted as being two interdependent criteria of appropriateness driven by the local
and global co(n)text. The appropriateness will be then delineated here consis-
tently 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 mark-
ers just as discourse markers are constitutive semantic-pragmatic components
of utterances and lack illocutionary autonomy, as they do not perform a prag-
matically 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 appropri-
ate 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 contex-
tual 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.
It seems worth to note additionally that the notion of appropriateness is usu-

ally seen as being dependent on the category of genre, and on the stylistic con-
notation 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 socio-
linguistic 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-seek-
ing 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), daccordo/ok (ok), like the
example (11) illustrates.6 The host utterance can be, like in the present case, a ques-
tion, 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 interlocutors current turn mentioned in Section 1.
24 Francesca Carota
B2: Il ritorno, le avevo detto domenica,

3: domenica POMERIGGIO sul presto.
As I said, the return will be on Monday, Monday early in the
afternoon.
A3: Si. ma come ORARIO?
Yes. But which is the hour?.
From the point of view of dialogue topic management, the host interrogative ut-
terance 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 con-
figured on the base of the partial acceptance, in the common ground, of the infor-
mation provided locally in B, as if it was incomplete for the co-participants com-
municative 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 com-
municative 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-negotia-
tion of the topical coordinate instantiated by the dialogue entity orario. Although
such an entity has been implicitly requested in the elliptical question A1, it can-
not 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 lho ... ce lho sulla SINISTRA rispetto alla
roulotte, hai capito?
So, me, I have the Elenas 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 Elenas furniture where do
you find them?
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 dis-
crepancy 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 givers indica-
tions). 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 1314 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: Poi sali su.
Then you go up.
F: Ma alla scritta BABBUINI te chai BANANO sotto o sopra?
But at the sign baboon do you have banana-tree below or
above?.
(14) G019: Poi sal+ sali un pochino su fino a FIUME<pb> ce lhai quella
figurina?
Then you go up a bit until the river, do have this figure?.
F020: Fiume FIUME si #G021 si#
River... river yes yes.
G021: #<F020> <eh># li, presente ?
Eh there, clear?.
F022: Si #<G023> si , ho capito#
Yes yes, I see.
G023: Poi ecco arrivi a #<F024> fiume #
Then, here, you arrive to the river.
F024: Ma a fiume e necessario usare le BARCHE ?
But at the river , do I need to use the boats?.
G025: Si, si usale usale , vacci.
Yes yes use them use them go towards them.
F026: <Mhmh>
Mhm.
G027: Vai alle BARCHE , quindi sei li, no ?
Go towards the boats, then you are there, arent you?.
F028: #<G029> Si#
Yes.
G029: Poi gira un pochino a SINISTRA, fai una specie di curva ma piccola.
Then turn on the the left, but not too much.
F030: <eh> !#<G029> Si#
Eh! Yes.
G031: Fatta la #<F032> curva ?#
Did you turn?.
F032: #<G031> Vai # #<RUMORE> vai#
Go <noise> go.
G033: E poi giri a DESTRA.
And then turn on the right.
F034: Ma sono sempre sulla BARCA, io ?
But am I always on the boat?.
G035: Si.
Yes.
F036: <Mhmh>
<Mhm>.
G037: Sempre LI giri a destra.
Always there turn on the right.
F038: Si
Yes.
G039: Ah e<ee> la valle limpida te la trovi a destra.
<Ah> the limpid valley, you will find it on the right.
F040: Si #<G041> 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 dont I dont need it
anymore.
G041: Okay, ci sei <P> no un dialettale. ti serve piu <pb> no
Ok, there you are, no, you dont 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 informa-
tion. However, the host utterance of ma aims at preventing potential misunder-
standings in a grounding phase in which the co-participant is still establishing a
common ground about the givers instructions.
(15) F084: Quindi sto risalendo e sulla mia destra trovo piazza della musica.
Im going up then and on the right I find the square of the music.
G085: Si ma non torni indietro <eh!> non torni indietro : dopo la rotonda
scendi.
Yes, but you dont go back, eh, you dont go back: after the round
square, you go down.
In the light of the previous considerations it can be stressed that, in the task-ori-
ented 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 dia-
logue 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
speakers common ground.
Generalizing on all the examples just considered, the role of ma can be char-
acterized 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 Per
A positional behavior close to the one just presented for ma pertains to per, oc-
curring 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 struc-
ture, like ma, per 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, per 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: Allora landata alle 15.30 da Romale prenoto un posto?
So, the departure at 3.30 from Rome may I reserve you a place?.
B2: Si. per vorrei prenotare anche il RITORNO.
Yes, but I would like to book the return as well.
A3: Daccordo.
Ok.
Furthermore, like in (17), per can occur -just as main a statement of clarifica-
tion that follows a confirmation (the positive feedback si in G251) about infor-
mation already co-activated in the common ground. The clarification statement
provides some additional inactive information (Sali un poco, go up slightly),
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: E arrivo al quartiere piccolo <pb> che ce lho sulla sinistra.
And I arrive at the small quarter that is on my left.
G251: Si per dopo questa piccola deviazione, SALI un poco.
Yes but after this small deviation, go up slightly.
F252: Si dopo il #<G253> quartiere#
Yes after the quarter.
G253: #<F252> Come se# ci fosse un DOSSO
As if there was a hump.
F254: Dopo il quarti+ ? <pb> <ah!> si ho capito ho capito.
Right after the quartyes Ive understood Ive understood!
The inactive information introduced into the utterance hosting per can correct
the information grounded until now or a part of it (see 18).
(18) 1 C 1: Va bene.
Ok.
2: Il rientro dovrebbe essere il MERCOLEDI.
The return should be on the Wednesday.
3: Vorrei arrivare vorrei partire scusi prima delle delle SEI di sera.
I would like to arrive, to leave sorry, before6 PM.
2 O 1: Partire prima delle sei di sera dunque vediamo subito dunque.
To leave before 6 PM, so, let see soon.
3: Le va bene partire alle DICIASSETTE e trentatre?
It is ok to leave at 5.33 PM.
3 C 1: Perfetto.
Perfect.
4 O 1: Diciassette e trentatr.
17.33.
2: In questo caso per deve CAMBIARE a Padova.
In this case, however, you have to change in Padua.
3: E l arrivo previsto a roma e` alle VENTITRE e zero cinque.
And the arrival at Rome will be at 11.05 PM.
4: Dunque, anche in questo caso, si tratta di due intercity.
So, also in this case there will be due INTERCITY trains.
Per is in accordance with a request dialogue act which opens a new unit of com-
mon 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
converge towards a shared cognitive state, the common ground. Otherwise, it cor-
relates 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 per 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-textu-
ally missing, or co(n)textually underspecified informative parameter.
The linguistic elements under investigation play the crucial function of re-
orienting the grounding process towards a new perspective to take for the achieve-
ment of a common ground. In this sense, per and ma display a distal attitude of
the speaker, whereby she takes distance from the co-participants prior contribu-
tion 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 appro-
priate 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 co-
text, 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 ques-
tion 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.
Lets have a look at the flights.
B2 a: Si.
Yes.
b: 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.
A3 a: Ah, va bene.
Ah, OK.
b: 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-participants contribution. The connective host utterance asks for new infor-
mation (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 topi-
cal 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 coordi-
nate, 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 preced-
ing 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 speakers strategy of re-orienting the ongoing negotiation, by revising a pa-
rameter 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-ori-
ents the information flow with respect to the cognitive context of the common
ground. However, in the present fragment, invece does not introduce any alterna-
tive solution to the same issue in the sense considered so far. First, the connec-
tive 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 pack-
ages the dialogue entity ritorno, which co-refers to a previously mentioned given
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: Venerd c un treno alle 8.35.
On Friday, there is a train at 8.35 AM.
2 C: Benissimo.
Perfect.
3 O: 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?.
4 C: 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 in-
tends 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 contras-
tive 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: 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.
3: In totale sono in seconda classe sessantottomila lire.
It is in total -in the second class- 68.000.
2 C 1: Ho capito.
I have understood.
3 O 1: Mentre il pendolino delle quattordici e quarantacinque sono
settantamila lire.
Whereas the train of 2.45 PM is 70.000.
2: Quindi duemila lire in piu` .
2.000 more then.
4 C 1: Si` si` non e` niente.
Yes yes its nothing.
2: E l arrivo?
And the arrival?.
5 O 1: 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.
2: 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: Per il rientro quindi mi diceva in serata tardo pomeriggio
As for the return, then, for the evening late in the afternoon.
2: 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.
4: Oppure alle diciannove e trentacinque e arrivo alle venti e
trentacinque
Otherwise at 7.35 PM and arrival at 8.35 PM.
2 C 1: Ho capito.
I understand.
3 O 1: Mentre l alpi eagles ce ne ha uno un alle diciotto e
quaranta, che arriva alle diciannove e quaranta.
Whereas lalpi 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 pa-
rameter 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 topi-
cal 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 inter-
personally 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
(meta-)negotiated information, a new unit of common ground can start to pres-

ents 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-par-
ticipants in the two domains. In the map-tasks, the co-participants social roles
are hierarchically organized, the followers moves depending on the givers in-
structions, 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 particu-
lar dialogue typology. Furthermore, an innovative criterion to determine their
appropriateness is provided with the results of the corpus analysis. The appro-
priateness 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 com-
mon 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 sus-

pended and re-opened around that parameter.
Per (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 oc-
curs within the co-negotiation of a particular topic, where a polar alternative so-
lution 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-negotia-
tion, to re-open the grounding process because of a task-related topic shift.
The interpretation of the contrastive markers used in the task-oriented dia-
logue 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 speakers 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 recip-
rocally 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 contrib-
uting 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 ground-
ing 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.
References
Bazzanella, C. 1994. Le facce del parlare. Un approccio pragmatico allitaliano 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: 223258.
Berretta, M. 1984. Connettivi testuali in italiano e pianificazione del discorso. In Linguistica
testuale, L. Coveri (ed.), 237254. Roma: Bulzoni.
Blakemore, D. 2000. Indicators and Procedures: Nevertheless and but. Journal of Linguistics
36: 463486.
Carota, F. 2004. On some effects of lexical contrast in information-seeking dialogues. In Pro-
ceedings of the Eighth Workshop on the Semantics and Pragmatics of Dialogue, Ginzburg, J.
and Enric Vallduv (eds.), 148149. Barcelona: Universitat Pompeu Fabra.
Carota, F. 2005. Fondo cognitivo comune e negoziazione dellinformazione: funzionalit dei mar-
catori discorsivi dellItaliano. PhD. Thesis. Pisa: Universit 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). Tbin-
gen: Niemeyer.
Chafe, W. 1976. Givenness, contrastiveness, definiteness, subjects, topics, and point of View.
In Subject and Topic, Charles N. Li (eds.), 2555. 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.), 673687. 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-Universit 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: 1933.
Fraser, B. and Malamud-Makowski, M. 1996. English and Spanish contrastive discourse mark-
ers. Language Sciences 18: 863881.
Fraser, B. 1998. Contrastive discourse markers in English. In Discourse Markers. Descriptions
and theory, A.H. Jucker and Y. Ziv (eds.), 113. Amsterdam/Philadelphia: John Benja-
mins.
Gmez-Gonzlez, M.A. 2001. The Theme-Topic Interface. Amsterdam/Philadelphia: John Ben-
jamins.
Heim, I. 1982. On the semantics of definite and Indefinite Noun Phrases. Ph.D. Thesis, Uni-
versity of Massachusetts at Amherst.
Kruijff-Korbayov, 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.), 6178. Helsinki: Uni-
versity of Helsinki.
Larsson, S. 2002. Issues under negotiation. In Proceedings of the Third SIGdial Workshop on
Discourse and Dialogue, K. Jokinen and McRoy (eds.), 103112. Philadephia: USA.
Lakoff, R. 1971. If s, ands and buts about conjunction. In Studies in Linguistic Semantics, Fill-
more, Langendoen (eds.), 114149. Mouton de Gruyter: Berlin.
Lombardi-Vallauri, E. 2002. La struttura informativa dell enunciato. Firenze: La Nuova Italia.
Molnr, V. 2001. Contrast from a contrastive perspective. In Proceedings of the ESSLLI01

Workshop on Information Structure, Discourse Structure and Discourse Semantics, I. Kru-
jiff-Korbayov and M. Steedman (eds.), 99114. Helsinki: University of Helsinki.
Mortara-Garavelli, B. 1993. Strutture testuali e retoriche. In Introduzione allitaliano contem-
poraneo. Le strutture, A.S. Sobrero (ed.), 371402. Roma-Bari: Laterza.
Pickering, M.J. and Garrod, S. 2004. Toward a mechanistic psychology of dialogue. Behavioral
and Brain Sciences 27: 169225.
Prince, E. 1981. Toward a taxonomy of given-new information. In Radical Pragmatics, P. Cole
(eds.), 223255. 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 Dis-
course and Dialogue, Hong-Kong, 918.
Stalnaker, R.C. 1974. Pragmatic presuppositions. In Semantics and Philosophy, Munitz, M.K.
and Unger, P.K. (eds.), 197214. New York University Press.
Steedman, M. 2000. Information structure and the syntax-phonology interface. Linguistics
Inquiry 31(4): 649689.
Traum, D. 1999. Computational models of grounding in collaborative systems. In Working
notes of AAAI Fall Symposium on Psychological Models of Communication: 124131.
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.), 175188. Helsinki: University of Helsinki.
Index
A amalgamation 59, 66, 71 C

acceptability 17, 18, 57, 148, 228 ambiguity 123, 183, 194 categorization 85, 86, 92, 93,
acceptance 62, 80, 116, 124, 125, analogy 33, 47 9597, 99, 100, 103, 108, 110,
136, 140, 142, 154, 178, 182, 188, analogical transfer 33, 53 206
227, 241, 248 anchoring 40, 116, 258 category
act/action answer alteration 84, 93, 94
collective 118, 120 direct 154, 180 complex 89, 98, 103, 104,
communicative 21, 58, 72, non- 179, 184, 185 107, 109, 110
120, 124, 127, 128, 214, 244, partial 190 conceptual 95
246 appropriateness perceptual 95, 97, 98
face-threatening 148, 195 condition 19, 49, 117, 119, value-dominated 95
illocutionary 6, 8, 65, 125, 128, 130132, 135142, Chafe 239, 243, 244
7073, 76, 77 152154, 205, 208, 227, 236, clarification
locutionary 8, 59 239, 240 request 91, 136, 190, 193,
macro speech 19, 20 pragmatic 56, 57, 239, 243 238, 247, 249253
micro speech 19 artefact Clark 20, 150, 205, 206, 238,
phatic 8, 59 linguistic 56, 59, 60, 62, 66, 239, 241
phonetic 59 67, 6971 co-construction 179, 216, 226
rhetic 8, 59 atom 43, 44 co-text 5, 151, 206, 208, 212,
social 18, 22, 115, 152, 170 attenuation 133, 135, 216, 221 216, 222, 238, 243, 244, 248,
speech 3, 8, 9, 19, 20, 22, 23, Austin 8, 57, 59, 6167, 70, 251, 253255
25, 55, 57, 61, 62, 67, 68, 115, 7377, 118, 170, 204 cognitive science 33, 48
118, 120, 122, 124, 125, 127, automatization 83, 86, 89, 99 coherence 8, 13, 118
130, 145, 170, 171, 210, 213, commitment 60, 6973, 76, 81,
216, 222, 232 102, 107
utterance 6, 207 B common ground 5, 11, 20,
activation 88, 92, 98, 99, 243, background 51, 81, 122, 206, 222, 225,
258 knowledge 14, 8486, 88, 238246, 248259
activity type 17, 1921, 117, 121, 89, 94, 96, 99, 103, 206 communicative
149, 152, 207, 226 Barwise 32, 36, 37 genre 17, 1921, 118
adjacency Bateson 3, 9, 10 move 60, 65, 67, 69, 71, 72
pair 4, 129, 132, 135 Bell 149, 153, 176, 194 project 17, 1921
position 129 bottom-up 3, 22, 119, 121 concept
relation 5, 129 Brandom 116, 118, 126 formation 83, 85, 93, 95,
Akman 8, 121, 206 Brown and Levinson 116, 149, 96, 100
alternatives 88, 89, 244, 254, 195 condition
255, 258 essential 9
262 Index
felicity 6, 9, 19, 56, 6167, social 12, 14, 16, 18, 32, 57, marker 107, 108, 136, 137,
69, 70, 76, 130, 148, 213 58, 121123, 138, 247 139, 169, 207, 222, 225, 246
macro 132 sociocultural 16, 18, 20, 117, structure 197
micro 132 127, 128, 130, 148 dispreference 154, 225
preparatory 9, 127 subjective 12 Dummett 58, 59
propositional content 9 type 13, 225, 227 dyad 20, 21, 126
sincerity 9, 124, 125 unmarked 16
conditional contextualization E
attenuating 208, 215, 216, cue 17, 151, 206, 207, 226 Edelman 82, 85, 86, 90, 9296,
221 device 7, 136, 137, 139 99, 100, 108
form 205, 209, 210 recontextualization 227, 228 elaboration 176, 183, 185, 190
configuration contrast 34, 35, 81, 95, 236239, environment
common 15 255 social 12, 104, 109
parts-whole 123, 124 conventionality 62, 70, 72 ethnography 15, 16, 148
tripartite 124, 125, 128, 130 cooperation 118, 122, 178, 179, ethnomethodology 4, 13
connective 236, 246248, 254, 182, 184, 187, 216, 226 evasive
255, 257 cooperative principle 122, 124, language 178, 194, 196
consciousness 129, 204 talk 179
higher-order 82, 83 coordination 205, 240, 258
context counterfactual 45, 46, 48 F
cognitive 4, 5, 9, 11, 12, 14, cross-examination 180, 191 Fairclough 58, 115, 117, 148, 149,
16, 18, 116, 120, 121, 238, culture 12, 14, 15, 18, 57, 120, 168, 171, 172, 173, 174, 183
240, 242, 243, 246, 251, 121, 128, 149, 150, 173, 226 felicity
253255 condition 6, 9, 19, 55, 56, 61,
construct of 59, 61, 70 D 62, 6467, 69, 70, 76, 130,
default 14, 16 Damasio 82, 8486, 8894, 96, 148, 213
external 16 97, 102 Ferrara 19, 25
extrasituational 168, 174, Deacon 81, 82 Fetzer 13, 18, 20, 32, 57, 80, 81,
176 default 14, 16, 80, 117, 125, 130, 101, 118, 120, 122, 124, 127, 129,
generalized 16 178, 227 132, 139, 140, 147, 148, 152, 155,
global 122, 227 Devlin 32, 3740, 49 176, 205207, 222, 225, 226,
individual 12 dialogue 228, 247
institutional 14 act 118, 122, 124, 125, 128, flout 177, 187
interactional 147, 168172, 130, 132, 133, 241, 244, 246, footing 13, 26, 128, 129, 144, 151
176, 184, 194 247, 253 force 7, 32, 44, 55, 57, 58, 67,
internal 55, 6165, 67, 69, entity 244, 248, 253255, 257 68, 7274, 88, 126, 135, 141,
70, 170 structure 235, 238, 244, 245, 154156, 167, 170, 183, 216,
linguistic 5, 6, 9, 12, 14, 16, 248, 251 245, 246
120, 121, 168172, 176, 180, task-oriented 258 frame 4, 810, 13, 14, 17, 1922,
185, 186, 188, 190, 192 topic 238241, 243, 244, 44, 71, 116, 118, 121, 122, 126,
local 121, 152, 206208, 220, 246, 248, 254 151, 194, 215, 222, 228, 240
224, 225, 227, 228 unit 238, 244
marked 16 disambiguation 227, 237 G
organizational 14 discourse game 6, 21, 38, 42, 43, 49, 75,
particularized 16 analysis 143, 148, 204 117, 204
prototypical 225, 227, 228 argumentative 82, 102104, Garfinkel 4, 115, 144
sequential 208 107, 110 Goodman 4345
situational 125, 168, 171173, conditional 213 Goodwin and Duranti 4, 167,
176, 182, 184, 194, 195 institutional 19, 150 169, 173
Index 263
Goody 20 internalization 69, 70, 87 of quantity 183, 185

Grammar interrogation 179, 182, 185188, of relation 123, 187, 188
functional 6, 7, 11, 215 190 memory
grammaticality 17, 147 interview declarative 91, 97, 98, 109
Grice 69, 115118, 122124, institutional 4 explicit 108
128130, 144, 153, 175, 179, media 147150, 153 implicit 91
197, 204 political 13, 132134, 141, procedural 84, 97, 98, 109
ground/grounding 148, 149, 153, 155 recategorical 84, 95
common 5, 11, 20, 51, 81, with experts 147, 156, 157 value-dominated 93, 94
122, 203, 206, 222, 225, 235, interviewee 122, 132, 134136, mentre 238, 247, 254257
238, 239246, 248257 149, 157, 162 Mey 56, 57, 118, 171
Gumperz 4, 7, 1517, 121, 144, interviewer 121, 132, 134, 147, Minsky 31, 32
148, 151, 206 148, 155, 159, 162 misapplication 63
invece 238, 247, 254257 misexecution 64
H misfire 63
Habermas 115, 116, 122124 J misinvocation 62
habitus 14, 173 Janney 5, 168, 170, 173, 176, mitigation 68, 183, 184
Halliday 7, 118, 167169, 171 177, 179 modality 98, 135, 212, 215
hedge multifunctionality 208
less-fuzzy 139, 140 K
more-fuzzy 138140 know-how 87, 91, 94 N
Heritage 4, 14, 115, 117, 121, negotiation
151, 152 L co- 239, 258
Lakoff 138, 216, 236 normative rightness 116, 124
I language
imperfetto 207, 208, 214, 215, game 117, 126, 170, 204 O
217, 229 processing 11, 12, 23, 82, 84 object
implicature use 3, 9, 12, 56, 58, 115, 117, vague 4143
conversational 122, 123, 129, 132, 135, 169, 204, 228 opt out 180
130, 133, 139 learning 21, 58, 82, 85, 91, 92,
generalized 123 94, 104, 109 P
particularized 123 Levinson 5, 13, 17, 5658, 62, part
inference/inferencing 116, 118, 121, 128, 148, 149, 152, dispreferred second pair
conversational 1517, 231 154, 157, 170, 195, 204, 206, 216 223
process 11, 12, 115 Lewis 4549 first pair 223
infon 39, 40, 50 Linell 17, 117, 126, 205 second pair 223, 225
information participation 13, 15, 124, 128,
flow 40, 239, 254 M 129, 132, 137, 151, 217, 226
seeking 237, 247, 257 ma 24, 212, 235, 238, 247253, parts-whole 410, 19, 118, 123,
structure 237, 238, 243, 244, 258 124, 205
246, 253 Malinowski 169, 171, 198 performative 8, 56, 58, 64, 65,
intention marker 7274, 124, 170, 213, 221, 222
communicative 8, 14, 18, 67, argumentative 100, 110 performativity 62, 70, 72
68, 115, 124, 132, 151 contrastive 238, 239 per 214, 215, 235, 251253
informative 18 discourse 107, 136, 222 Perry 32, 36, 37, 40, 52, 205
intentionality 18, 87, 115, 118, maxim personification 62, 69, 70, 72
194 of manner 123, 139, 183, 185 politeness 129, 137, 148, 207, 216
collective 118 of quality 123, 129, 139, polysemy 207, 208, 227, 228
182, 187
264 Index
practice S event 14, 15, 149, 151, 152

interviewing 147, 163 salience 34 Sperber and Wilson 4, 11, 18,
social 127, 172, 174, 204, 228 Saville-Troike 15, 16, 130 68, 121, 206
predicate Sbis 9, 67, 68, 115, 118, 122, 125, substantiation 65, 66
evaluative 215, 216 167, 168, 170172, 174, 176, system
preference 90, 154, 223 216, 224 interactional 124, 128130,
presupposition Schegloff 4, 14, 118, 150152, 132, 135, 136
downgrading 210 204 interpersonal 124, 128130,
maximal 122 Searle 6, 8, 9, 49, 62, 67, 68, 80, 132, 136138, 141
medium 123 83, 8691, 103, 106, 115, 118, textual 124, 128130, 132,
minimal 122, 123 124, 126, 148, 170, 204 133, 138140
processing sequence
efficiency 83, 88 argumentative 107, 109, 135 T
language 11, 12, 82, 84 macro 118, 119, 121 theme 48, 119, 144, 192, 193,
prolixity 123, 183, 185, 190 micro 118120 244, 248, 249, 254, 255
prominence 50, 244 negotiation-of-meaning 143 token 11, 209, 223
negotiation-of-validity 122, top-down 116, 119, 121
Q 141 topic management 238, 240,
question question-answer 136, 178 246, 248
question-response pair 154, sequentiality 22, 24, 124 truth
179 Shuy 192, 194 condition 8, 125
Quine 31, 52 similarity turn
comparative 46, 49 constructional unit 120
R structural 43, 47 taking 226
ratification 116, 122125, 132, sincerity 9, 124, 125 Tversky 32, 34, 35
156, 241 situation type 12, 40, 57, 207
rationality 115, 116, 122 abstract 39
reality 21, 59, 60, 112 fictional 36 U
reasoning future 36 uptake 55, 60, 71, 72, 86
practical 11, 32, 57 game 38, 42, 43, 49 utterance
Recanati 12, 44, 49 historic 55, 5762, 6466, non-performative 56, 73
reentry 85, 86, 93, 96, 99 7072, 75 performative 56, 58, 72,
reinforcement 68, 109 initial 89, 91, 93, 9799, 73, 75
relevance 105108
conditional 4, 128, 129 of reference 98, 103, 107 V
theory 11 past 36 vagueness
repetition 169, 186, 224 schematized 107, 109110 ontic 41
representation similar 31, 32 validity claim
dispositional 92, 9699 theory 31, 32, 35, 3941, macro 20, 140141
resemblance 47, 48, 88, 94, 210 49, 51 micro 142
response speech minus 122, 124, 125, 128, 130,
option 89, 93, 9799, 107, act 3, 8, 9, 19, 20, 25, 55, 57, 132, 133
108 61, 62, 67, 68, 115, 118, 120, plus 122, 124, 125, 128, 130,
tactic 179, 196 122, 124, 125, 127, 130, 170, 132, 133
type 171, 191 171, 210, 213, 216, 222 value
rheme 244, 255 act theory 3, 19, 55, 67, 68, adaptive 90, 95
rule 115, 118, 122, 124, 130 van Dijk 12, 13, 18, 19, 115, 117,
constitutive 6, 22, 62 community 15, 16, 56, 116, 125, 126, 148
regulative 6 127
Index 265
W world
verb
well-formedness 17, 19, 57, 228 objective 116, 124, 125
cognitive 182
Widdowson 167, 175 possible 45, 49, 245
modal 211, 215, 221
Wittgenstein 46, 71, 88, 170, social 116, 124, 128132, 135,
of saying 221
204 141, 204
perceptual 182
subjective 116, 124, 125
reported 182
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 HBler (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, agns 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 GrammarPragmatics Interface. Essays in honor of
Jeanette K. Gundel. 2007. viii, 345 pp.
154 HBler, 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 FlttuM, 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 BHrIg, 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., Mara de los ngeles gMez-goNzlez and susana M. doVal-surez
(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 Mller, 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.

Anita Fetzer - Context and Appropriateness PDF

Caricato da

Informazioni sul documento

Titolo originale

Copyright

Formati disponibili

Condividi questo documento

Condividi o incorpora il documento

Opzioni di condivisione

Hai trovato utile questo documento?

Questo contenuto è inappropriato?

Copyright:

Formati disponibili

Anita Fetzer - Context and Appropriateness PDF

Caricato da

Copyright:

Formati disponibili

Context and Appropriateness

Pragmatics & Beyond New Series (P&BNS)

John Benjamins Publishing Company

American National Standard for Information Sciences Permanence of

Library of Congress Cataloging-in-Publication Data

Context and appropriateness : micro meets macro / edited by Anita Fetzer.

2007 John Benjamins B.V.

Context, contexts and appropriateness 3

Part I. Bridging problems between context and appropriateness

Party II. Bridging problems between communicative action

If I may say so: Indexing appropriateness in dialogue 115

Part III. Bridging problems between micro and macro

The attenuating conditional: Context, appropriateness and interaction 203

Communication is both context-creating and context-dependent

Context has become a major field of research in information technology, engi-

1.1 Linguistic context

a whole) is looked upon analogously to Searles conceptions of regulative rules

an offer. Again, a context-anchored parts-whole perspective to the investigation

1.2 Cognitive context

Cognitive context is not only of relevance to cognitive linguistics and cognitive

A similar type-token differentiation is manifest in Pencos distinction be-

1.3 Social context

Social context is often considered to comprise the context of a communicative ex-

The categories of speaker, hearer and audience are no longer conceived of

1.4 Sociocultural context

Social context is frequently used synonymously with extra-linguistic context

interpreted in a particular mode. For instance, the distinction between mono-

2. Culture is used synonymously to Hymes conception of speech community. It may com-

as a social activity. Saville-Troike summarizes his contribution to the field of eth-

In the following, the process of contextualization is examined, which provides

1.5 Context and contextualization

Contexts in context has examined linguistic, cognitive, social and sociocultural

In the previous section, the research paradigm of interactional sociolinguistics

Appropriateness has frequently been compared and contrasted with acceptability,

language-oriented evaluation of a sentence, acceptability and appropriateness are

2.1 Appropriateness conditions

In a social- and socioculturally oriented examination of speech act theory, felic-

whose definition needs to be based a larger frame of reference, such as an activity

2.2 Appropriateness and the micro-macro interface

Appropriateness has been examined as the product of a process of evaluating a

Macrostructures are conceived in analogy to a common language that represents

Part I, Bridging problems between context and appropriateness, comprises three

Part II, Bridging problems between communicative action and appropriateness,

Bridging problems between context

How do we ever understand anything? Almost always, I think, by using one or

is it appropriate to think that two situations (faced by an intelligent agent) are

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

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

The features shared by and , i.e. (set intersection)

In what is commonly called a contrast model, the similarity of to is given as a

Situations usually have individuals standing in relation at various (spatiotem-

. Direct perception of a situation is not as unproblematic as it sounds, due to ever-present,

Think of the situation of someone watching a live football match on television. In

It is probably redundant to remind that when an agent individuates a situation, he

While we are in general unable to trim down situations to a complex of more

One of the attractive features of situation theory is its insistence on an infor-

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

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

5. Russell and Goodman on similarity

. Direct perception of a situation is not as unproblematic as it sounds, due to ever-present,